Bichic SME eo WAR RN

Mh RAR BOAT dete ML

rs , Rete asl Refe hers
ree ke = Cees th 8.
r a > a IRN

Tha WANs TG gaat tiene ve

EGE Tak eR
Da eihaida Poehite 5s Ns aR ce

Ar Anda a rare tee
WIS AAPG S Sam A,

ae ReMcdadaeRad
MSS R A ig

Pe Rete Aa AMO

2 Ra BI 4s SBP idiot iat te
5 cs yeaa RINSING NRCG ae Rs Rese
"4 oo. Ran: = r ea ree To tite. Rey) howe 42 ay

RRNA
Bes re Se te Inti,
BER LAG Me SRC A WR AE RRS CREO TS
aE ta Asie MstqedeR he te sugay ores

EMAAR re DS Rates Raise aati.
> oes RTRs Re ET ndlen bases a

SW a Ree ete ae at ate Ary, «netniae
Seo Arne cdo hs hat aeen

amie de Aish Ne Nem tee mamalinns er!
SR ac

HPS

mW ae eee
OR iar ected fe

ren fi oo Stet areas Y
a Tain Fore m, aga

Neto aRE NNO Tere As La ee
tee ecg TERRE OER

Y SE REIN. ike, Ue has ¥ BoM AsR TT
SX=a
te BIS Ne
“Wefiee ROXAS RA

J bee Nee

sens
Sa ICR iy

cneaii. BEES eerie

ae aeeehe io
weer
VRE AAR

fer
Snel veces

ay icin a
ssiyer

FL PE Helier MacaCoys Aare
Sabi gL phaidiae ts tk eae fe ian
FREI HRT

co Seelgs puPr rete
Soe HE ply FNM EE SI

re:

I ase
asses wit sb 28S"
oh etetatie capioid zx

arteries ! 3
ate Uy fe Et Rreng zeae
5 tS inte pe rela
age’
isin Poets big

seetaeets
sa Genie

e vet hiee ei
die. oftine igi MTR AL) scene te
: 2 Tene

panes of each
AOD re eR HK OR SSB.

As ge eee aN vesien Gait

Bs ays bAiattiss) 24
Sues 2 Oe we Seveape ESN!
Bebgip las ure ete cecerine ged

ee ie Ly as
Apt Seer’
an iss

SUS Get 24 BEAR BSR Sie
ee

cl
Serie Raia

sa eeGe eae
oe
pee Reece eine >
Hinson =a

- ol od > Pig _ "See > Or a | ee, anemeae Pr > a Ee oF Se
a a |
1 INSTITUTION NOILNLILSNI NVINOSHLINS S3!YVYHSIT LIBRARIES SMITHSONIAN
ie = ws ‘> = ars = 1 ae
Je 5 oi : 2) BB 5 ’ B
F m0 = = &§ YH >
> > XY . y >
2D é Ps) SAAS a os a Uf ee,
rea - ie NN oD a ee =
m ps m SASS m ” m
W z w mM z oy = on
> SAFIYVYSIT_LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IuvUug
w” ; = ve w” a ” Ed ” ;
= < < <5 < eo, 2 ns =
r 6 cr 5 Mm FZ Vy, 5 z
” i) eee ” n . ” <ie 7p) 7)
Oo = Oo pe ew PSY O oe O
2 cs Z E WY Z = “2
i => > ’ — ¥
2 Eee hae BS ee 5 :

INSTITUTION NOILNLILSNI_ NVINOSHLINS. S31NYVUAIT- LIBRARIES SMITHSONIAN. INSTITUTIO

Ym,

thy

NVINOSHLINS SAIYu¥Vugle

NOILNLILSNI
LIBRARIES
LIBRARIES
NOILNLILSNI
NOILNLILSNI

SalYVYdIt LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI

INSTITUTION = NOILALILSNI

Sa'tuyvug!y

INSTITUTION NOILALILSNI NVINOSHLIWS

INSTITUTION
INSTITUTION
INSTITUTION

SAIYVYUGIT_ LIBRARIES SMITHSONIAN INSTITUTIO

NSS

NVINOSHLINS S31uVvuait LIBRARI ES

NVINOSHLINS S3IYNVYSII

NVINOSHLIWS
SMITHSONIAN
SMITHSONIAN
Ge
SMITHSONIAN

*

SMITHSONIAN INSTITUTION NOILOLILSNI

NVINOSHLINS S3IYVuar

NOILALILSNI
LIBRARIES
LIBRARIES

INSTITUTION NOILALILSNI_NVINOSHLINS S3i¥vyugity LIBRARIES SMITHSONIAN INSTITUTIOS
\ 4

VUdIT LIBRARIES SMITHSONIAN
17

“ss

LIBRARI ES SMITHSONIAN

SSIYVYRITLIBRARIES
INSTITUTION NOILNLILSNI
INSTITUTION NOILNLILSNI

INSTITUTION
Saey dei

s’salyuyvyg

INSTITUTION NOILOLILSNI NVINOSHLIWS SaIuvugiy

Ws -931¢

SAtYVy 9!

SMITHSONIAN

NYINOSHLIWS

NVINOSHI!
SMITHSONIAN
NVINOSHLIWS
SMITHSONIAN
NWINOSHLIW

INSTITUTION NOILOLILSNI_ NVINOSHLINS. S31YVUGIT- LIBRARIES SMITHSONIAN. INSTITUTION

NOILIILILONI
LIBRARIES
NOILNLILSNI
NOILNLILSNI
NOILNLILSNI

S3tYVUEll LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31uVvudl

INSTITUTION, NOILNLILSNI NVINOSHLIWS

Saiuvadl

Wy,
e va fol
INSTITUTION

INSTITUTION

INSTITUTION
aS
VS
NC
SS

INSTITUTION

% |
SAIYVYRIT_ LIBRARIES SMITHSONIAN INSTITUTION

LINS SZINYVUSIT LIBRARIES
INS SA1YVYdIT LIBRARIES

Z as Zz Z SO Z < = NS
ww « anon t= : “. X BS wa
= Mo = , AS (SIS yl i Se Ne Kost Fis

+,

a ee

— ————_

—————————_

eee 2 gt ~ Zz pte Z on ag
LSNI_NVINOSHLINS _S31NVYdI7 LIBRAR| ES_ SMITHSONIAN _ INSTITUTION NOILALILSNI_NVINOS!
‘“ - rd NS
wer ‘ oO Sr © Sinem oO = :
w =e oO — bs oO = ow >
a 5 a S id 0 ss 0 \
a & > K Up fof. ? —— Bae ON
rae ra a fe Ma Go 8 “a a AK
rae wY ra on” NEN at = F ee x
a) 2 ae aie 7p) = w z= w Be
-1ES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS LIBRARIES SMITHS
” a ia : ” Zz: w er ”
= et = = = eal =
= 2 4 Fe = z =4
. a5 Oo 2S pa IG) als Oo =
\ 2 ” ae ap) ” 7) ”
VE z : : g c z
ys = a = >" = mS >
2 “” a4 7) = z ” z=
LSNI_NVINOSHLIWS SMITHSONIAN INSTITUTION NOILNLILSNI_ NVINOS
ae g 3 is Sse fi a
$e 7) ; ” en
x = a = Yim, ce Shae
5 oe 4 < ff 7s < =
= mm ea oO - A = oO = SS
x Oo S a Oo ae Oo am Oo ae
mie rae ay z coy 2 a} =
21ES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYEIT LIBRARIES SMITH
Be 2 3 = £ z Se =
aE. ra ; a
ry = = > | = > Ps
Us os . ms ee = = i
= ae ue a zZ oe z
ILSNI NVINOSHLINS S3ZIYVYSIT LIBRARIES INSTITUTION NOILOALILSNI NVINOS
: ral ” Zz we ” z Ae ” =
= = ee = <x NS = ae
: ad AH VA SS yr Ay . — > =4 =~
, ; <1. ete Aig, Oo < Re x
7’ BGM INC 3 2 ae :
y ie 2 Gy = = E a EB
= > ie = > = > =
is ieee = ”n Ti eS ” ee : Yn
2 | ES SMITHSONIAN _INSTITUTION NOILNLILSNI_NVINOSHLIWS S31yYvVuYsiT_ LIBRAR! ES SMITHS
aah = ve) z SX a a WwW
rag 4 a, a Bes 4 We Tae tee
ws _ ox s ac = «x. vy
aes z < = < 2 Kf 3
a 5 = = a 5 0 YG
ey z oo Fas ee ; z 3
SJINVUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILOLILSNI_NVINOS
ia Ss ia z Ee z Gs
o = w a GL, = = = \
Aaa = a) > Visi 7 > YN
2 = 2 = Lio bi is |
a = a = GY fii > = oN
mm 2s = Daa ese hw == w = WN
is Hi = w” = ” z w ;
RIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31¥vyugIT_LIBRARIES SMITH:
2 Za ” z ” ) Zp ”
= =a ee = WL < z abe z
4 Fs = et hy = = a =
: mt Mi yy. a &
\ 2 2 & GFF? 2 2 SN 2
VE 2 8 GG F z 2 WN 3
Ww 2 = Ze Gis Z EF Nor Zz
S = = = = = = ee
vy) Zz YW) v Py
ILSNI_NVINOSHLINS LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_ NVINO
Z Zz ae : Zz z
nl = n baat YY, oD Y ” rN
= \ = = S pga < cs
S Se w eS Be Wehr C a =4 WS
— “~ oa) a ” ys PS = wo =a SS
oO Sy S rE oO mM. ro) ical ©
. ~ = ce
ae _ z ily Fak _
RIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYUdIT LIBRARIES SMITH
= sg Zz - z e =
| YA a i = XN \N = aa = Gyys
me ‘ ZY st = X\ ee ae Z. Z
Sa (ep) = ap) a, = op) _
HLSNI_NVINOSHHWS SAIYVUEIT LIBRARIES SMITHSONIAN INSTITUTION |, Beg
“ft KES SS PR gen ce Pee. eee SO RS Tei pte ie

International Journal

PHYSIOLOGY e aN
CELL and MOLECULAR BIOLOGY ?
GENETICS

IMMUNOLOGY

BIOCHEMISTRY —

DEVELOPMENTAL BIOLOGY
REPRODUCTIVE BIOLOGY
ENDOCRINOLOGY |

BEHAVIOR BIOLOGY

ENVIRONMENTAL BIOLOGY and ECOLOGY
SYSTEMATICS and TAXONOMY

published by Zoological Society of Japan

Business Center for Academic Societies Japan

distributed by VSP, Zeist, The Netherlands

ISSN 0289-0003

ZOOLOGICAL SCIENCE

The Official Journal of the Zoological Society of Japan

Editors-in-Chiet: . The Zoological Society of Japan:
Seiichiro Kawashima (Tokyo) Toshin-building, Hongo 2-27-2, Bunkyo-ku,

Hideshi Kobayashi (Tokyo) Tokyo 113, Japan. Tel. (03) 814-5675
Managing Editor: Officers:

Chitaru Oguro (Toyama)

President: Hiromichi Morita (Fukuoka)
Assistant Editors:

an Secretary: Hideo Namiki (Tokyo)
Yuichi Sasayama (Toyama) Treasurer: Tadakazu Ohoka (Tokyo)

Hitoshi Michibata (Toyama) Librarian: Masatsune Takeda (Tokyo)
Miéko Komatsu (Toyama) Sac

Editorial Board:

Howard A. Bern (Berkeley) Walter Bock (New York) Aubrey Gorbman (Seattle)
Horst Grunz (Essen) Robert B. Hill (Kingston) Yukio Hiramoto (Chiba)
Susumu Ishii (Tokyo) Yukiaki Kuroda (Mishima) John M. Lawrence (Tampa)
Koscak Maruyama (Chiba) Roger Milkman (Iowa) Kazuo Moriwaki (Mishima)
Richard S. Nishioka (Berkeley) Tokindo S. Okada (Okazaki) Andreas Oksche (Giessen)
Hidemi Sato (Nagano) Hiroshi Watanabe (Tokyo) Mayumi Yamada (Sapporo)

Ryuzo Yanagimachi (Honolulu)

ZOOLOGICAL SCIENCE is devoted to publication of original articles, reviews and communications
in the broad field of Zoology. The journal was founded in 1984 as a result of unification of Zoological
Magazine (1888-1983) and Annotationes Zoologicae Japonenses (1897-1983), the former official
journals of the Zoological Society of Japan. ZOOLOGICAL SCIENCE appears bimonthly. An annual
volume consists of six numbers of more than 1200 pages including an issue containing abstracts of papers
presented at the annual meeting of the Zoological Society of Japan.

MANUSCRIPTS OFFERED FOR CONSIDERATION AND CORRESPONDENCE CONCERN-
ING EDITORIAL MATTERS should be sent to:
Dr. Chitaru Oguro, Managing Editor, Zoological Science, Department of Biology, Faculty of Science,
Toyama University, Toyama 930, Japan, in accordance with the instructions to authors which appear in
the first issue of each volume. Copies of instructions to authors will be sent upon request.

SUBSCRIPTIONS. ZOOLOGICAL SCIENCE is distributed free of charge to the members, both
domestic and foreign, of the Zoological Society of Japan. To non-member subscribers within Japan, it is
distributed by Business Center for Academic Societies Japan, 6—16—3 Hongo, Bunkyo-ku, Tokyo 113.
Subscriptions outside Japan should be ordered from the sole agent, VSP, Utrechtseweg 62, 3704 HE
Zeist (postal address: P.O. Box 346, 3700 AH Zeist), The Netherlands. Subscription rates will be
provided on request to these agents. New subscriptions and renewals begin with the first issue of the
current volume.

All rights reserved. No part of this publication may be reproduced or stored in a retrieval system in any
form or by any means, without permission in writing from the copyright holder.
© Copyright 1990, The Zoological Society of Japan

Publication of Zoological Science has been supported in part by a Grant-in-Aid for
Publication of Scientific Research Results from the Ministry of Education, Science
and Culture, Japan.

ZOOLOGICAL SCIENCE 7: 563-580 (1990)

REVIEW X

ECDYSTEROIDS

JAN KOOLMAN

© 1990, Zoological Society of Japan

Institute of Physiological Chemistry, University of Marburg

D-3550 Marburg, Germany

INTRODUCTION

In 1954 Adolf Butenandt and Peter Karlson
published a report concerning the successful isola-
tion and crystallisation of the insect hormone,
ecdysone [1]. It was the first invertebrate hormone
available in pure form. A decade passed until the
chemical structure of ecdysone was elucidated [2,
3]. Soon more ecdysone-like substances were
found and collectively called “ecdysteroids”. In
recent years a considerable amount of biochemical
and physiological information about ecdysteroids
has been assembled which is reflected in recent
articles and books [4-15]. An attempt will be
made here to give an overview of the wealth of
information about ecdysteroids. In the introduc-
tory part some basic principles are given. The
second part will provide more details. Also,
exceptions from the reported principles will be
mentioned. In the third part some trends in the
research of ecdysteroids will be described.

PRINCIPLES

Steroid structure

Ecdysone is a classical steroid hormone (Fig.
la). It has the usual four ring nucleus of steroids
and the full side-chain of cholesterol. Characteris-
tic features of ecdysone are a keto group, found in
ring B conjugated to a double bond, as well as five
hydroxyl groups. In the three-dimensional repre-
sentation of the molecule (Fig. 1b) an additional
characteristic of ecdysone is visualized: the plane

Received May 1, 1990

of ring A deviates from the plane formed by the
rings B, C and D. This bending between rings A
and B is due to a cis-fused ring junction (hydrogen

OH

Fic. 1.

Structure of ecdysone.
a) Classical representation
b) Stick-and-ball model

c) Van der Waals’ model

564 J. KOOLMAN

at C5 in P position) by which ecdysteroids differ
markedly from all vertebrate-type steroids. The
structure of the steroid nucleus is rather rigid. The
side-chain may acquire various conformations de-
pending on the immediate environment. Rigidity
of the nucleus as well as flexibility of the side-chain
are relevant to the interaction of ecdysteroids with
their receptors and other macromolecules. The
recognition of ecdysteroids by proteins is also
influenced by the polarity of ecdysteroids, which is
seen from the three dimensional model: ecdysone
has a polar face, the a-side (looking downwards in
Fig. 1b, side-view) caused by hydroxyl groups and
an apolar face, the @-side, (looking upwards), with
aliphatic groups. The most realistic view of the
molecule is shown by its van der Waals’ repre-
sentation (Fig. ic).

Moulting hormone

Ecdysteroids are the moulting hormones of
arthropods. The hormonal control includes the
regulation of many physiological and biochemical
processes related to the moulting process. In
insect larvae, for example, ecdysteroids control
preparatory steps of moulting, such as termination
of feeding, emptying of gut and spinning of a
cocoon. Ecdysteroids initiate biochemical changes
in the old cuticie, e.g. tanning, as well as processes
directly involved in moulting, such as apolysis and
proliferation of epidermal cells, secretion of layers
of new cuticle, digestion of the old cuticle, and
sclerotization of the new one [16-18]. Control of
each moult includes both stimulation and inhibi-
tion of defined processes by ecdysteroids: apolysis,
cell division and the onset of cuticle synthesis are
stimulated; later processes of moults such as pro-
duction of pigment, digestion of the old cuticle and
ecdysis are all inhibited [19]. Ecdysteroids exert
their control of the moulting process in coopera-
tion with another family of morphogenetic hor-
mones, the juvenile hormones (JHs). In the
presence of JHs ecdysteroids initiate conservative,
larval moults. Progressive moults, e.g. from larva
to pupa, occur when JHs are diminished or even
absent (Fig. 2).

A comparison of ecdysone with the vertebrate-
type steroid hormones not only indicates similar
structural features with regard to the carbon

Brain
4h
Av

i iran

‘,. Neurosecretory cells

Prothoracicotropic <2 Prothoracicotropic
hormone | hormone
Allatotropin
i

Corpora allata

Prothoracic gland Prothoracic gland

\ fea hormone

onan

fi Fat body

20- /

lao Wy

Imaginal moult

Ecdysene

Fat body x

20-Hydroxyecdysone

/

Larval moult

STR = FR

Pupal moult

Fic. 2. Ecdysteroids initiate moults of arthropods. The
example shows the coordinate control by ecdyster-
oids and juvenile hormones. The larva of a
holometabolous insect moults to a larva of the next
stage, then to a pupa, and eventually to an imago
(adult) (adapted from [199]).

skeleton, hydroxyl and carbonyl groups. As will
be detailed below, the homology applies also to
the biosynthesis, transport, metabolism, and mode
of action. However, a striking difference remains:
vertebrates have several classes of steroid hor-
mones, that govern reproduction, and carbohy-
drate or mineral metabolism, etc. Insects and
other arthropods apparently have only one class of
steroid hormone, ecdysteroids. No explanation for
this difference has so far been found.

Hormone system

Ecdysone is part of a hormone hierarchy (Fig.
2). In this system the brain controls the activity of
peripheral glands by neurohormones [20]. Effects
of the brain hormones are either tropic (ie.,
stimulatory) or inhibitory. In response to these
positive or negative stimuli, the ecdysteroid synth-
esizing gland will, or will not, secrete a hormone
that in turn acts on its target organs. The tropic
hormone of insects, that is secreted by certain
neurosecretory cells of the brain, is a peptide
termed prothoracicotropic hormone (PTTH; old
terms; brain hormone, activation hormone) [21-
23]. It stimulates prothoracic glands to synthesize
ecdysone which in turn controls the function of
epidermal cells. A closer view as to the identity of

Ecdysteroids 565

the moulting hormone has shown that ecdysone
itself is not the active form. It is converted to
20-hydroxyecdysone (other terms: ecdysterone, /-
ecdysone, crustecdysone). This ecdysteroid repre-
sents the active form of the moulting hormone in
larval stages of arthropods.

DETAILS

Zooecdysteroids

So far, only two ecdysteroids have been men-
tioned, ecdysone and 20-hydroxyecdysone. In
fact, many more ecdysone-like compounds, for
which the generic term “ecdysteroids” was coined
(Table 1) [24, 25], have been isolated or chemical-
ly synthesized. A recent review by Hew Rees [26]
quotes 61 different ecdysteroids found in animals.
According to their chemical structure these
“zooecdysteroids” [24] may be classified into
several groups (Table 2). It is thus clear that the
term “ecdysteroids” is a chemical definition.

Although the term “ecdysteroid” overlaps with
the biologically defined term “moulting hormone”
these two expressions should not be used synony-
mously, because not all 61 ecdysteroids are hormo-
nally active. Most of them are biosynthetic in-
termediates or metabolites of ecdysone. Neverthe-
less, more than a dozen ecdysteroids were found to
be biologically active when tested in various bioas-
says [10, 27-30]. Even these ecdysteroids cannot
all be regarded as hormones. In some cases, the
explanation for their bioactivity is found in their
bioconversion to 20-hydroxyecdysone during the
assay. In other cases, the compound shares a

TABLE 1. Definition of ecdysteroids

Goodwin et al., 1978 [25]:

Ecdysteroids are all compounds structurally re-
lated to ecdysone.

Lafont and Horn, 1989 [38]:

Ecdysteroids are steroids with a nucleus which
bears a cis-fused A/B ring junction, a 7-ene-6-one
chromophore and a 14a-OH, irrespective of their
activity in a moulting hormone assay.

Ecdysteroid-related compounds do not fulfill all
the above criteria; either they lack the 14a-OH, or
the 7-ene, or they present one additional 4- or
14-ene double bond, etc.

TABLE 2. Naturally occurring ecdysteroids,
grouped according to their structures (after [49])

20-hydroxyecdysone and less hydroxylated com-
pounds
26-hydroxyecdysteroids
ecdysonoic acids
3-dehydroecdysteroids
3-epiecdysteroids
C28-ecdysteroids
side-chain cleavage products
ecdysteroid conjugates
phosphate esters
nucleotides (artefacts ?)
glucosides
acyl esters
acetophosphates

significant structural similarity with 20-
hydroxyecdysone, allowing it to mimic its effects.
There is sufficient evidence, however, that several
structurally different ecdysteroids which are active
forms of the hormone exist in arthropods.

Active form of the hormone

Endocrinologists will admit that the definition of
a hormone is not very strict [31]. One of the
difficulties is that either the product secreted by
the endocrine gland or the compound bound by
the hormone receptor in the target organ may be
regarded as a hormone. Any difference between
these two is due to metabolism of the hormone
between its site of secretion and action.

The active form of the moulting hormone bound
by ecdysteroid receptors of insects, spiders and
crustaceans is 20-hydroxyecdysone. However, ex-
ceptions to this rule have been found and are
increasing in number. In certain bugs (Hemiptera)
and bees (Hymenoptera), and possibly also in
some flies (Diptera), 20-hydroxyecdysone coexists
with, or is even replaced by another active form of
the hormone, makisterone A [32]. In crab species,
20-hydroxyecdysone coexists with ponasterone A,
a 25-deoxy derivative of 20-hydroxyecdysone, that
is more active as a hormone [33, 34]. In insect
oocytes and embryos ecdysone, 2-deoxyecdysone
and 26-hydroxyecdysone may have hormonal func-
tions of their own [35]. Structures of these hor-
mone candidates are shown in Figure 3. In the

566 J. KOOLMAN

future yet other ecdysteroids may turn out to have
hormonal functions. A proof of their hormonal
role would include the demonstration of their
occurrence, bioactivity and interaction with a hor-
mone receptor under physiological conditions.
The existence of ecdysteroid receptor(s) able to
bind the compound in question has been shown so
far only for 20-hydroxyecdysone and ponasterone
A.

qe Ho 9H
San ~.
HO HO
OH OH
HO HO
H H

Ponasterone A

2-Deoxyecdysone Makisterone A

Fic. 3. Ecdysteroids with a likely hormonal function in
arthropods (after [34, 35]).

Phytoecdysteroids

Before aspects of hormone biosynthesis, meta-
bolism and action are discussed, it should be
pointed out that ecdysteroids occur not only in
animals but are also present in plants [36, 37].
Accordingly, the plant ecdysteroids were called
“phytoecdysteroids”. The number of known phy-
toecdysteroids increases steadily; currently it is
twice that of zooecdysteroids. A recent review by
René Lafont and Dennis Horn quotes 118 diffe-
rent phytoecdysteroids [38]. Quantitatively most
prominent among the phytoecdysteroids are ecdy-
sone, 20-hydroxyecdysone and ponasterone A, all
of which are biologically active in arthropods.

Plant species that contain ecdysteroids are
numerous [38]. Ecdysteroids have been found in
embryophytes, including ferns, gymnosperms and
angiosperms. In contrast, their occurrence in
thallophytes has not yet been reported. The
concentrations in plants may be very high; in some

cases they make up more than 1% of dry weight of
the plant. Concentrations of plant ecdysteroids
may exceed those found in arthropods by several
orders of magnitude. Thus, the majority of ecdys-
teroids (in terms of their mass) is found in plants
and not in arthropods. Furthermore, the wide
occurrence of ecdysteroids makes them by far the
most important steroid hormones in the biosphere
both in terms of quantity and diversity.

The significance of phytoecdysteroids for plants
is unknown [38]. Ecdysteroids might be simply
end products of plant metabolism. Another,
perhaps more likely, explanation is that ecdyster-
oids fulfil the role of allelochemicals protecting
plants against phytophagous insects [37]. Evidence
exists that dietary ecdysteroids affect development
in some insect species [39]. As some plant-feeding
insects have developed efficient detoxification
mechanisms, this chemical defence of plants may
be overcome [40]. These insect species employ
intestinal enzymes to convert exogenous ecdyster-
oids to inactive metabolites.

Plant ecdysteroids show a large variety of struc-
tures. This diversity led René Lafont and Dennis
Horn to define the term “ecdysteroid” more pre-
cisely (Table 1) [38]. They introduced a distinction
between “true ecdysteroids” and “ecdysteroid-like
compounds”. The occurrence of ecdysteroid-like
compound in plants [38] is compatible with the
assumption that they have endogenous functions.
This has been clearly shown for brassinosteroids
which serve as growth-promoting plant hormones
[41, 42]. The relationship between the two steroid
hormone systems, ecdysteroids in arthropods and
brassinosteroids in plants, is a fruitful research
area both for practical (plant protection) as well as
theoretical aspects (hormone evolution).

We will now return to arthropods and discuss
different aspects of the hormone system of ecdys-
teroids, namely biosynthesis, transport, metabol-
ism, excretion, and action of ecdysteroids (Fig. 4).

Biosynthesis

The biosynthesis of ecdysone begins with
cholesterol [43], which is the major sterol in anim-
als. As arthropods cannot synthesize cholesterol
de novo [44], they derive it from dietary sources.
This poses no problem for zoophagous arthro-

Ecdysteroids 567

precursor

oe
If

A
enzyme

Z

enzyme

i, Ge

hormone metabolite H

H H

4 —enzyme
non-target

hormone

cell

hormone
metabolite

target cell

“yi

hormone H ;
metabolite excreting
cell
PPLE ey € 20
excreta

Fic. 4. Generalized scheme of a hormone system. A hormone system is characterized by several elements that are
either involved in the formation and elimination of the hormonal signal or in the perception of and the reaction to
the signal. These elements include: cells that synthesize and secrete the hormone into the blood (gland cells);
possible plasma proteins able to bind the steroid hormone (hormone carriers); cells that take up the hormone,
metabolize it and return the metabolites to the circulatory system; cells that are able to remove the hormone and
its metabolites from circulation by excretion; and, cells of the target organs which take up the hormone and react
to it with a specific response. These elements are linked by the free (unbound) steroid hormone in the circulation

(adapted from [88}).

pods. In contrast, phytophagous arthropods have
the difficulty that cholesterol does not occur in
their diet because phytosterols differ slightly from
zoosterols in their chemical structure. Typical
plant sterols such as sitosterol, stigmasterol and
campesterol, are homologues of cholesterol with
additional methyl and ethyl groups in their side-
chain. By dealkylation most phytophagous arthro-
pods can convert these phytosterols to cholesterol
[45, 46]. But some phytophagous insects are
unable to perform this reaction sequence. They
have solved the problem by modification of their
hormone system (see below) [47].

The pathway of ecdysone biosynthesis is not
completely known (Fig. 5) [48, 49]. Extensive
modification of the nucleus of cholesterol occurs
before hydroxylation of the side-chain. The first
step is the introduction of a second double bond
into ring B: cholesterol is converted to 7-

dehydrocholesterol. With the following steps,
characteristic structural features of the ecdyster-
oids are introduced: the A/B cis ring junction, the
keto-ene structure of ring B, and the hydroxyl at
C14. Neither the sequence nor details of these
reactions are fully understood. 3f-Hydroxy-5a-
cholestan-6-one may be an intermediate of ecdy-
sone biosynthesis in larvae of the silkworm [50]. A
steroid called “5-ketodiol” (2, 22, 25-trideo-
xyecdysone) is a likely intermediate in the
biosynthesis of ecdysone in locusts, as was shown
by Jules Hoffmann’s group [51, 52]. Three con-
secutive hydroxylations convert 5f-ketodiol to
ecdysone. The most important sequence in hy-
droxylation is at C25, at C22, and finally at C2 [53].
Other sequences appear to exist, suggesting multi-
ple pathways of ecdysone biosynthesis which may
differ between arthropod species.

Some arthropod species have adapted their hor-

568 J. KOOLMAN

| 7-Dehydrocholesterol

Fic. 5.

mone system to ecdysteroids that have an addition-
al alkyl group, the moulting hormone 20-
hydroxyecdysone being replaced by makisterone
A which has an additional methyl group at C24
(see above; Fig. 3). The change of moulting hor-
mone is due to the inability of these species to
dealkylate their dietary sterols to cholesterol. In

58-Ketodiol

OH

Cholesterol

20-Hydroxyecdysone

Pathways of ecdysone biosynthesis (after [49]).

these species the enzymes involved in the biosynth-
esis of the moulting hormone have been adjusted
according to the modified substrate [54]. This
modification of the biosynthetic pathway of the
moulting hormone was discovered in some hyme-
nopteran and heteropteran species. It may also
occur in a few species belonging to other orders of

Ecdysteroids

arthropods. The modification is a remarkable
example of the evolution of steroid hormone sys-
tems.

Ecdysiosynthetic tissues

Ecdysteroids are found at all developmental
stages of arthropods: in embryos, larvae, pupae
(holometabolous insects), and adults [26]. Their
occurrence throughout development suggests that
they have different, stage-specific functions (see
below) and that they are synthesized in different
tissues (Table 3) [55]. The classical moulting
glands of insects are the prothoracic glands (ven-
tral glands, ring gland) [56], which synthesize and
secrete ecdysone [57, 58]. Ecdysone is released
from the moulting glands in a protein bound form
by exocytosis [59]. There is no storage of hormone
within the glands [57].

Only recently has it become apparent that the
moulting glands of a number of arthropod species,
notably lepidopteran and crustacean species, sec-
rete not only ecdysone but also 3-dehydro-
ecdysone, which after its secretion is rapidly con-
verted to ecdysone [60-62].

The moulting glands develop and become func-
tional during late embryogenesis, are active in
larval stages and degenerate in many insect species
after the last larval moult. In adult insects, which
in contrast to adult crustaceans do not moult, an
important source of ecdysteroids is the gonads [55,
63]. Ovaries [64-66] of all arthropod species and

569

testes [67] of some insect species are able to
synthesize ecdysteroids. The source of ecdyster-
oids in the Crustacea are the Y-organs [68, 69].
Their secretory products are ecdysone, 3-
dehydroecdysone, and 25-deoxyecdysone, the im-
mediate precursor of ponasterone A [34, 70]. The
moulting glands of spiders have not yet been
identified. Moreover, along with the classical
glands of insects (Table 3) other sources of hor-
mone must exist [71] and remain to be discovered,
e.g., the hormone source in pupae of beetles such
as Tenebrio molitor [72].

Oocytes are also a source of ecdysteroids [63].
The compound is originally synthesized in the
ovary of adult females, being transferred to the
oocytes which convert the ecdysteroid immediately
to 22-conjugates. Chemically these ecdysteroids
are esters with phosphoric acid or long-chain fatty
acids. Physiologically the ecdysteroid conjugates
are a maternal gift to the oocytes where they
accumulate and form a hormone reserve. Recent
work of Mary Bownes and her group indicates that
these ecdysteroid conjugates are tightly bound to
vitellins [73]. Upon breakdown of vitellins, ecdys-
teroid conjugates are released and cleaved by
esterase to form hormonally active ecdysteroids.
Although some aspects of this scheme are still
hypothetical, we may assume that the released
ecdysteroid then fulfils its role as hormone: in
oocytes ecdysone breaks meiotic arrest; in young
embryos it initiates cuticle formation (see below).

TABLE 3. Ecdysiosynthetic tissues and controlling neuropeptides in arthropods
gland occurrence stage neuropeptide
prothoracic gland insects mature embryos Proto TAC Ou ite
and larvae ormone (PTTH) [23]
Y-organ crustaceans larvae and el Sen A
adults hormone (MIH) [158]
ovary insects adults follicle cell
crustaceans ? stimulating hormone
(FCSH) [196]
egg development
neurosecretory hormone
(EDNH) [197]
testes many insects adults testis ecdysiotropin
(TE) [198]
additional unknown insects pupae z

sources (e.g. beetles)

? chelicerates

570 J. KOOLMAN

Hormone titre

The hormone titre in the haemolymph is largely
determined by the rate of hormone secretion from
moulting glands. Concentrations of hormonally
active ecdysteroids range between 10° and 10°
M and show characteristic cyclic profiles [74, 75].
Generally, stage-dependent cycles of hormone
titre are observed with one or several large peaks
per instar. The most prominent peak in insect
larvae, the “moulting peak”, is found at the end of
each stage shortly before the moult. In many
holometabolous insects a small peak of hormone
concentration preceding the large one is found in
the middle of the last instar. Jim Truman and
Lynn Riddiford introduced the name “commit-
ment peak” (or “wandering peak”) as it initiates
specialized behaviours and switches the genetic
programme from the larva to the pupa [19, 76, 77].
Although we are far from understanding the de-
tails of the titre variations, it is evident that a
precise sequence of highs and lows of hormone
titre are necessary for a correct control of moult
and development [19, 78, 79].

Transport

Ecdysteroids circulate in two forms in the
haemolymph of insects: unbound and bound to
protein [80]. The unbound ecdysteroid fraction
represents the active form of the hormone. The
fraction bound to carrier protein(s) is inactive both
hormonally and metabolically, and may function
as a hormone reserve. While no ecdysteroid-
binding proteins have been found in the Crustacea
as yet, in insects ecdysteroid-specific [81] as well as
nonspecific carrier proteins [82] have been
observed. There is an apparent species-dependent
heterogeneity of ecdysteroid transport.

Plasma membrane transport

How do ecdysteroids enter cells? The work of
Klaus-Dieter Spindler and his group suggests that
components of the cell membrane are involved in
the uptake of ecdysteroids. Free ecdysteroids from
the haemolymph penetrate the cell membrane
apparently by two different mechanisms: by a
specific, carrier-mediated uptake and by diffusion
[83-86]. The specific mechanism operates at low

concentrations of ecdysteroid, is energy depen-
dent, and can be inhibited by SH-reagents. The
relative importance and physiological role of the
two transport processes is at present not clear [87].

Ecdysteroid metabolism

In general, steroids are metabolized intracellu-
larly. This also applies to the metabolism of
ecdysteroids in arthropods [88], with one possible
exception: the reduction of 3-dehydroecdysone to
ecdysone in the haemolymph of lepidopteran spe-
cies [61, 62]. Many insect tissues are able to
metabolize ecdysteroids to a certain extent, most
effective being the fat body and Malpighian
tubules [89]. These tissues convert ecdysone to
20-hydroxyecdysone and to a variety of hormonal-
ly inactive metabolites. The gut also contributes
significantly to the metabolism of ecdysteroids. It
is especially active on exogenous ecdysteroids re-
sorbed from the food which are mostly inactivated
[40, 90-92].

Individual biochemical reactions of ecdysone
metabolism show a species- and stage-dependent
variation. Six different types of enzymatic reaction
can take place: hydroxylation, oxidation, reduc-
tion, side-chain cleavage, conjugate formation and
conjugate hydrolysis. These reactions act at va-
rious sites of the molecule and in different com-
binations (Table 4) [88, 93-99].

The formation of ecdysteroid conjugates is espe-
cially heterogeneous. It may lead either to highly
polar ecdysteroid—mostly phosphate esters, or in
a few cases glucosides—or to apolar ecdysteroids,
which have been identified as esters of acetic acid
or long-chain fatty acids. Mixed ecdysteroid con-
jugates were also observed [26]. Interestingly,
conjugate formation is reversible, in principle.

TABLE 4. Reactions of ecdysone metabolism

reaction site at C-

hydroxylation 20, 26
oxidation DD, 3, 26
reduction 3

side-chain cleavage 20/22

conjugate formation De 3 22, 255 26
conjugate hydrolysis 22, 26

Major reactions are underlined (after [98]).

Ecdysteroids 571

CONJUGATION

FATTY ACIDS

PHOSPHATE

ACETIC ACID

GLUCOSE OXIDATION TO

GLYCOLIC ACID ECDYSONOIC
ACID

CLEAVAGE

HYDROXYLATION ;
! le | 122

CONJUGATION:
PHOSPHATE

: CONJUGATION :
HYDROXYLATION PHOSPHATE

¥

CONJUGATION :

Sr ¢— DEHYDROXYLATION GLUCOSE

: REDUCTION

REVERSIBLE
OXIDATION
EPIMERIZATION
CONJUGATION:
PHOSPHATE

ACETIC ACID (m): widespread reaction

(—): reaction found in a few species
only

(--->): tentative pathway not directly
demonstrated

Fic. 6. Reactions of ecdysone metabolism (after [88,
97, 98]).

Steroid conjugates can be hydrolyzed. In fact, 22-
and 26-conjugates appear to be storage forms from
which active ecdysteroids can be released (see
above). 3-Conjugates, on the other hand, are true
inactivation products of the hormone.

What are the functions of these many reactions?
Hydroxylation of ecdysteroids at C20 activates the
molecule, at least in larvae. In insect embryos
hydroxylation at C26 may have the same role,
whereas this reaction inactivates the hormone in
other developmental stages. Hydrolysis of ecdys-
teroid 22- or 26-conjugates in oocytes and embryos
may also lead to active ecdysteroids. The reduc-
tion of 3-oxoecdysteroids to 38-hydroxyl ecdyster-
oids also activates. However all other reactions
inactivate the hormone: they are tailoring ecdys-
teroids for storage or excretion by the introduction
of characteristic functional groups.

As seen in Figure 4, excretion of ecdysteroids is
an integral part of the hormone system. In insects
either the gut or Malpighian tubules are the site of
hormone excretion [40, 100].

Biosynthesis, metabolism, and excretion toge-
ther control the hormone concentration in the
circulatory system [100]. Rates of all three proces-
ses vary with development in a coordinated
fashion. During increased biosynthesis of ecdys-
teroids catabolic reactions and excretion of hor-
mone are often reduced. In the interplay of these

processes the metabolism of ecdysteroids fulfils a
specific role. Metabolism is assumed to provide a
delicate balance between individual ecdysteroids,
e.g. between ecdysone and 20-hydroxyecdysone, a
balance that is required for the control of biologic-
al effects. Quantative analysis of the ecdysteroid
profile in an insect is available, e.g., for the
butterfly, Pieris brassicae [101] and the fruitfly,
Drosophila melanogaster [102]. Careful analysis of
the haemolymph ecdysteroids in Manduca sexta
revealed that ecdysone and 20-hydroxyecdysone
and other free ecdysteroids are present as well as
ecdysteroid conjugates [103].

Biological activity

Three major processes are under control of the
ecdysteroids: moulting, growth, and gametogene-
sis. The moult and physiological events associated
with moulting are controlled by ecdysteroids dur-
ing both embryogenesis and larval development of
arthropods. The role of ecdysteroids as moulting
hormone has been discussed above, already [18,
63, 104]. Clearly, the term “moulting hormone”
covers ecdysteroid functions only partly. Rather
ecdysteroids can be regarded as a multifunctional
hormone system acting as a sort of “Taktgeber”
(pacemaker). The many different targets respond
to the hormonal signal with their preprogrammed
(i.e. target- and time-specific) answer.

A growth and differentiation promoting activity
of ecdysteroids has been proved, e.g. for imaginal
discs of holometabolous insects [79, 105, 106], for
differentiation of gonads in insects [107, 108] and
for regeneration of lost limbs in the Crustacea and
Chelicerata [109, 110]. While ecdysteroids can
promote growth in one type of target tissue they
can cause death in others. Death of proleg muscles
during the larval-pupal transition caused by ecdys-
teroids was the first example found [111]. Of
special interest is the hormonal control of
neurogenesis and neuronal differentiation, and of
neuronal cell death in insects as detailed in a very
readable review by Jim Truman [19]. During
pupal development of holometabolous insects
ecdysteroids cause regression of certain neurites
and outgrowth of others. Ecdysteroids also trigger
cell death of immature and larval neurons in a
precisely time sequence [112, 113].

572 J. KOOLMAN

Gonadal ecdysteroids in adult arthropods con-
trol gametogenesis and related functions [114,
115]. In females of dipteran species, these ecdys-
teroids induce vitellogenin synthesis by the fat
body and ovary [116-118]. In other arthropod
species, however, vitellogenesis is contolled by
juvenile hormones. In those species in which
ecdysteroids also occur in adult males, spermato-
cyte maturation as well as functions of accessory
glands are probably under the control of ecdyster-
oids.

The ability of ecdysone to reinitiate meiosis in
oocyte maturation, which was observed by Jules
Hoffmann’s group, may be a general feature in all
arthropods [63, 119, 120]. Control of gastrulation
could be another feature of embryonic ecdyster-
oids [63]. Among the many other functions of
ecdysteroids, only the control of a temporal halt of
development (diapause) will be mentioned here
[18].

Mode of action

Steroid hormones act by modification of trans-
cription of specific genes [121]. This concept has
the property of a dogma and applies to all steroid
hormones. It is based on experiments done with
ecdysone by Ulrich Clever and Peter Karlson in
1960. The injection of ecdysone into larvae of
Chironomus tentans caused formation of puffs in
the chromosomes of salivary glands [122, 123].
The current paradigm of ecdysteroid action is the
following: the ecdysteroid enters the target cell
and migrates via an unknown route to the nucleus
where it binds to its receptor molecule. The
ecdysteroid receptor complex then interacts with
DNA to induce new RNA transcripts [124-126].

Ecdysteroid receptors are the key to an under-
standing of the mode of action of the hormone
[125, 127]. The receptors are site-specific, DNA-
binding proteins of approximately 100 kDa. They
occur in target cells at an extremely low concentra-
tion of about 1,000 molecules per cell. Evidence
for a nuclear localization has been shown [128].
Like receptors of vertebrate-type steroids the
ecdysteroid receptor has probably three domains:
a steroid-binding domain, a DNA-binding domain
and a regulatory domain. Ecdysteroid binding is
characterized by high specificity and high affinity.

The dissociation constant Kp of the blowfly recep-
tor complex with 20-hydroxyecdysone is around 30
nM and correlates well with the physiological
hormone concentration [129-132].

On the basis of bioactivity measurements, Ber-
gamasco and Horn have defined the structural
elements required for hormone activity. Three
contact sites for interaction with the ecdysteroid
receptor were postulated (Fig. 7) [27]. Interaction
of the ecdysteroid receptor with DNA is less well
understood. There is evidence that the receptor
has a general affinity for DNA [133]. A high
affinity for specific nucleotide sequences called
“consensus sequences” in the vicinity of ecdyster-
oid-controlled genes is likely. Binding to these
sequences allows the receptor to interact with the
transcription machinery [134].

The receptor concentration apparently is de-
velopmentally regulated. Peaks of receptor con-
centration follow shortly after peaks of hormone
concentration. It is likely that 20-hydroxyecdysone
itself leads to an increase of its receptor [135].

Fic. 7. Model of the interaction sites of ecdysteroids
with their hormone receptor. 20-Hydroxyecdysone,
the major hormone, is shown to interact with the
receptor at three different regions, X: f-side of ring
A and B, Y: a-side of ring C and D, Z: the
side-chain (adapted from [27]).

TRENDS

A number of new techniques and compounds
will add impetus to and stimulate ecdysteroid
research. Among them are: modern immunoas-
says for the detection and quantitation of ecdyster-
oids [136, 137]; improved chromatographic
methods for their separation [138]; sophisticated
physico-chemical methods for structural analysis
[139-141]; new radiolabelled tracers for work on

Ecdysteroids 573

hormone receptors [131, 142, 143]; and, various
types of agonists and antagonists interfering with
ecdysteroid systems (see below).

Occurrence of ecdysteroids

Sufficient evidence has accumulated to indicate
that ecdysteroids occur in many protostomian spe-
cies within the animal kingdom, including mem-
bers of the phyla of arthropods, platyhelminthes,
nemathelminthes, annelids, molluscs, and echi-
noderms [144]. Apparently ecdysteroids do not
occur in deuterostomian species including, among
others, the vertebrates. Whether ecdysteroids
function as hormones in protostomian species out-
side the phylum of arthropods is an unsolved
question. To demonstrate this role all major
aspects of the hormone system have to be ex-
amined: ecdysteroid biosynthesis, metabolism, ex-
cretion, hormonal function and mode of action
(e.g., interaction with receptors) [145-147].

Among protostomian species in question are the
helminths. These are mainly parasitic organisms
and belong to the taxa of Nematoda, Trematoda,
and Cestoda. Many major infections of man are
caused by helminths. Their list includes: schistoso-
miasis, echinococcosis, onchocercosis, river blind-
ness, and taeniasis, to mention only a few of the
diseases which affect several hundred million hu-
mans. Proof of a hormonal function of ecdyster-
oids in helminth species would allow the develop-
ment of suitable therapeutic methods which could
disturb the hormone system of the parasites with-
out interfering with the steroid systems of the host.
Unfortunately, we are a long way from this impor-
tant goal [148, 149].

Regulatory mechanisms

Although knowledge of the hormone system of
ecdysteroids is still incomplete, current research
activities have turned to the analysis of regulatory
mechanisms controlling these processes.

Biosynthesis of ecdysteroids in insect larvae is
under the control of PTTH, a tropic hormone,
which itself is regulated by neuronal activities in
the central nervous system (CNS) [21, 150, 151].
Several levels of feedback appear to exist [152-
155]. The presence of ecdysteroid receptors in all
tissues involved suggests that the concentration of

Proprioreceptive
mechanoreceptive
photoperiodic
temperature
hormonal

stimuli

Neurosecretory Site of synthesis

eeeedbacies cells of the brain of ecdysiotropin
endocrine

endocrine / Corpora cardiaca Siteiofeecretion

paracrine BH of ecdysiotropin

corpora allata

Ecdysiotropin

Prothoracic glands

Ecdysteroid

Target tissue

Fic. 8. Postulated feedback control of ecdysteroid
biosynthesis in insect larvae.

Site of sysnthesis
and secretion of
ecdysteroid

autocrine

circulating ecdysteroid is measured at three levels:
at the site of PTTH synthesis in or near the
neurosecretory cells of the CNS, at the site of
PTTH secretion in the neurohaemal organ, and at
the site of ecdysone synthesis itself (Fig. 8) [128].
The effect of ecdysteroid feedback may be stimula-
tory or inhibitory depending on the physiological
stage and the hormonal environment [70, 155].

In common with other neuropeptides, PTTH
exhibits considerable species specificity and heter-
ogeneity. The tropin of lepidopteran species exists
in at least two size classes, a large form of ca. 22
kDa and a small form of ca. 5 kDa. Within each
size class molecular variants occur. Surprisingly
the tertiary structure of the small form of PTTH
resembles insulin [22, 23].

Each ecdysiosynthetic tissue appears to have its
specific tropic hormone (Table 3). Collectively
these peptides are called “ecdysiotropins” [21, 22].
The crustacean neurohormone differs from insect
ecdysiotropins significantly, in that, it inhibits
rather than stimulates ecdysone secretion. This

574 J. KOOLMAN

difference is reflected in its name “moult inhibiting
hormone” (MIH) [156-158].

The ecdysteroidogenic tissues respond :to bind-
ing of the regulatory peptide by an intracellular
increase in cAMP level. Calcium also plays a role
in stimulating steroidogenesis. These second mes-
sengers transmit their signal via calmodulin and a
protein kinase to a 34 kDa protein that is ultimate-
ly phosphorylated [61, 70, 159]. While the exact
control point in the pathway from cholesterol to
ecdysone is not known, evidence exists that MIH
inhibits the conversion of ketodiol to ecdysone [70,
160]. While ecdysiotropins are the primary regula-
tors of ecdysteroid biosynthesis, other hormones
are also involved, most notably the juvenile hor-
mones [161, 162]. Clearly, there is an interendoc-
rine network of hormones that control moults and
other effects of ecdysteroids [61, 70, 163, 164].

The metabolism of ecdysteroids varies during
the development of arthropods. As outlined above
these variations affect the titre of active hormone,
and they are obviously physiologically regulated.
Individual reactions are switched on and off due to
enzyme induction (and repression?), interconver-
sion of enzyme [165], feedback inhibition and
control by superimposed hormone systems [166,
167].

Ecdysteroid receptor

Analysis of the mode of action of ecdysteroids
requires knowledge of their receptor’s structure,
function, and occurrence. Major goals are the
isolation of the receptor gene and production of
receptor-specific antisera. It is likely that ecdyster-
oid receptors belong to the steroid receptor super-
family containing several metal binding domains
called “zinc fingers” which allow specific contacts
to the DNA [168]. Insect systems, in particular

Drosophila, can provide effective models to study _

the role of zinc finger proteins due to the availabil-
ity of polytene chromosomes and a number of
mutants (ecdysteroid-deficient mutants: [102,
169]). Of special interest is receptor function
during embryogenesis and its relationship to other
members of the superfamily of zinc finger proteins
involved in embryonic development [170, 171].

Ecdysteroid agonists and antagonists

A breakthrough has been made in the search for
ecdysteroid agonists and antagonists [172]. Keith
Wing, a research biologist with Rohm and Haas,
has recently found that certain substituted hydra-
zines can act as non-steroidal ecdysteroid agonists
[173, 174]. RH 5849, a representative of this
group, is shown in Figure 9. The nonsteroidal
agonists bind to ecdysteroid receptors with moder-
ate affinity and elicit all known biological effects of
ecdysteroids. They have a potential as pesticides,
as the substituted hydrazines were more active
than ecdysteroids in bioassays due to a slower
catabolism [173].

Azadirachtin KK 42

Fic. 9. Structures of substances with effects on the

hormone system of ecdysteroids. RH 5849 is a
non-steroidal ecdysteroid agonist. Homobrassino-
lide is an ecdysteroid antagonist. Azadirachtin

inhibits the activation of ecdysteroids (hydroxylation
at C-20). KK-42 interrupts ecdysone biosynthesis.

The search for ecdysteroid antagonists has also
been successful. Among brassinosteroids, a group
of plant steroid hormones with growth-promoting
activity, some efficient inhibitors of insect moult-
ing were found [175, 176]. Binding assays with
ecdysteroid receptors revealed that certain brassi-
nosteroids are true ecdysteroid antagonists: they
bind competitively to the hormone (ecdysteroid)
receptor and thereby block their action [177]. The
structure of an active ecdysteroid antagonist, 22S,
23S-homobrassinolide, is shown in Figure 9.

Ecdysteroids SS

The most efficient natural substance with moult-
inhibiting activity is azadirachtin, a_ tetranor-
triterpenoid with ecdysteroid-like structure (Fig.
9) [178, 179]. The plant-derived compound has a
multitude of adverse effects on many insect spe-
cies. Azadirachtin is active as an antifeedant and,
when ingested, injected, or topically applied to
insects, inhibits moulting, growth, and reproduc-
tion [180]. Some of the effects seem to be related
to the perturbance of ecdysteroid titre [181]. Pre-
sumably azadirachtin interferes with the control of
ecdysone biosynthesis. An inhibitory effect on the
metabolism of ecdysteroids, especially on the 20-
monooxygenation of ecdysone has been shown
[182-184]. No direct effects of azadirachtin on
ecdysteroid receptors were found, and other
effects on the hormone system of ecdysteroids may
be explained by inhibition of neurosecretory pro-
cesses [183, 185, 186]. Azadirachtin is but one
representative among a steadily growing number
of plant-derived substances, which due to a per-
turbance of the ecdysteroid system, have potential
as insect growth regulators [187].

Viruses are also able to interfere with the hor-
mone system of ecdysteroids. A gene which was
recently isolated from an insect baculovirus was
shown to code for an ecdysteroid UDP-glucosyl
transferase. This enzyme inactivates ecdysteroids
and allows the virus to block normal development
of its insect host [188].

Synthetic compounds interfering specifically
with the synthesis of ecdysteroids were also found.
The imidazole compound KK-42 (Fig. 9) acts on
prothoracic glands directly and at very low levels
[189]. Allenic cholesterol derivatives which inhibit
ecdysteroid hydroxylation at C-20 and C-22 [190],
and fluorinated cholesterol derivatives [191]
obviously act as suicide-substrates. Azasteroids as
well as certain nonsteroidal secondary and tertiary
amines can block the conversion of phytosterols to
cholesterol and thereby may inhibit ecdysone
biosynthesis [192, 193]. The modification of diet-
ary sterols to block ecdysone synthesis is an in-
teresting approach which is presently under inves-
tigation [194].

More than two decades ago Carol Williams
suggested that insect hormones may provide a
basis for control of insect pests. He pointed out

that juvenile hormones would be good candidates
for these “third-generation pesticides” [195]. The
trends described above indicate that ecdysteroids
may also contribute to realize this goal.

ACKNOWLEDGMENTS

Critical reading of the manuscript by Drs. Gerd Kaus-
er, Del Bradbrook, and Roger Dennis is gratefully
acknowledged.

REFERENCES

1 Butenandt, A. and Karlson, P. (1954) Z. Natur-
forsch., 9b: 389-391.

2 Karlson, P., Hoffmeister, H., Hummel, H.,
Hocks, P. and Spiteller, G. (1965) Chem. Ber., 98:
2394-2402.

3 Huber, R. and Hoppe, W. (1965) Chem. Ber., 98:
2403-2424.

4 Hoffmann, J. (editor) (1980) Progress in Ecdysone
Research. Elsevier/North Holland, Amsterdam.

5 Sehnal, F., Zabza, A., Menn, J.J. and Cymbor-
owski, B. (editors) (1981) Regulation of Insect
Development and Behaviour. Vol. 1 and 2, Tech-
nical University of Wroclaw, Poland.

6 Downer, R. G. H. and Laufer, H. (editors) (1983)
Endocrinology of Insects. Alan Liss, New York.

7 Hoffmann, J. and Porchet, M. (editors) (1984)
Biosynthesis, Metabolism and Mode of Action of
Invertebrate Hormones. Springer-Verlag, Berlin/
Heidelberg/New York/Tokyo.

8 Kerkut, G. A. and Gilbert, L.I. (editors) (1985)
Comprehensive Insect Physiology, Biochemistry
and Pharmacology, Vol. 7 and 8, Pergamon Press,
Oxford/New York.

9 Bownes, M. (editor) (1986) Ecdysone: From
Metabolism to Regulation of Gene Expression.
Insect Biochem., 16 (special issue).

10 Wilson, I. D. (1987) In “Handbook of Natural
Pesticides” Vol. III Insect Growth Regulators, Part
A. Ed. by E. D. Morgan and N. B. Mandava, CRC
Press, Boca Raton, Florida, pp. 15-86.

11 Koolman, J. (editor) (1988) Ecdysone: From
Biosynthesis to Mode of Action. J. Insect Physiol.,
34 (special issue).

12 Zabza, A., Sehnal, F. and Denlinger, D. L. (edi-
tors) (1988) Endocrinological Frontiers in Phy-
siological Insect Ecology. Vol. 1 and 2, Technical
Univ. of Wroclaw, Poland.

13 Koolman, J. (editor) (1989) Ecdysone, Georg-
Thieme Verlag, Stuttgart/New York.

14 Lafont, R. (1990) Proceeding of the 9th Ecdysone
Workshop, Paris. (In press)

576

15

16
7

18

19

20

21

jij)

23

24

25

26

27

28

29

30

31

J. KOOLMAN

Gupta, A. P. (editor) (1990) Morphogenetic Hor-
mones of Arthropods. Rutgers Univ. Press, New
Brunswick/London.

Richards, G. (1981) Biol. Rev., 56: 501-549.
Bownes, M. and Redfern, P.F. (1985) In “Meta-
morphosis” Ed. by M. Balls and M. Bownes,
Clarendon Press, Oxford, pp. 117-126.

Sehnal, F. (1989) In “Ecdysone”. Ed. by J. Kool-
man, Georg Thieme Verlag, Stuttgart/New York,
pp. 271-278.

Truman, J. W. (1988) In “Advances in Insect
Physiology”. Vol. 21., Ed. by P. D. Evans and V.
B. Wigglesworth, Academic Press, London, pp. 1-
34.

Richter, K., Kauser, G. and Bidmon, H. J. (1989)
In “Ecdysone”. Ed. by J. Koolman, Georg Thieme
Verlag, Stuttgart/New York, pp. 319-326.
Richards, G. and Hoffmann, J. A. (1987) In
“Handbook of Natural Pesticides”. Vol. III, Insect
Growth Regulators. Part A. Ed. by E. D. Morgan
and N. B. Mandava, CRC Press, Boca Raton,
Florida, pp. 1-14.

Mordue, W. and Morgan, P. J. (1987) in “Hand-
book of Natural Pesticides”. Vol. III, Insect
Growth Regulators, Part A. Ed. by E. D. Morgan
and N. B. Mandava, CRC Press, Boca Raton,
Florida, pp. 153-184.

Ishizaki, H. (1989) In “Ecdysone”. Ed. by J.
Kooiman, Georg Thieme Verlag, Stuttgart/New
York, pp. 204-210.

Karlson, P. and Koolman, J. (1976) In “Colloque
internationaux C.N.R.S., No 251—<Actualités sur
les Hormones d’Invertebres”. Ed. by M. M. Dur-
chon, C.N.R.S., Paris, pp. 325-341.

Goodwin, T. W., Horn, D. H. S., Karlson, P.,
Koolman, J., Nakanishi, K., Robbins, W. E.,
Siddall, J. B. and Takemoto, T. (1978) Nature,
2a [e28 WD,

Reese ee Hen (i989) aang scdysone.. Eds by. Jr
Koolman, Georg Thieme Verlag, Stuttgart/New
York, pp. 152-160.

Bergamasco, R. and Horn, D. H. S. (1980) In
“Progress in Ecdysone Research”. Ed. by J. Hoff-
mann, Elsevier/North Holland, Amsterdam, pp.
299-324.

Cherbas, L., Yonge, C. D., Cherbas, P. and Wil-
hams, C. M. (1980) Wilhelm Roux’ Arch., 189: 1-
ID);

Cymborowski, B. (1989) In “Ecdysone”. Ed. by J.
Koolman, Georg Thieme Verlag, Stuttgart/New
York, pp. 144-149.

Dinan, L. (1989) In “Ecdysone”. Ed. by J. Kool-
man, Georg Thieme Verlag, Stuttgart/New York,
pp. 345-354.

Karlson, P. (1982) Naturwissenschaften, 69: 3-14.

a2

33

34

35

36

Sy)

38

39

40

41

42

43

44
45

46

47

48
49

50

Sl

52

Feldlaufer, M. and Svoboda, J. A. (1986) Insect
Biochem., 16: 45-48.

McCarthy, J. P. (1979) Steroids, 34: 799-806.
Lachaise, F. (1989) In “Ecdysone”. Ed. by J.
Koolman, Georg Thieme Verlag, Stuttgart/New
York, pp. 313-318.

Feldlaufer, M. (1989) In “Ecdysone”. Ed. by J.
Koolman, Georg Thieme Verlag, Stuttgart/New
York, pp. 308-312.

Nakanishi, K., Koreeda, M., Sasaki, S, Chang, M.
L. and Hsu, H. Y. (1966) Chem. Commun.,
915-917.

Galbraith, M. N. and Horn, D. H. S. (1966) Chem.
Commun., : 905-906.

Lafont, R. and Horn, D. H. S. (1989) In “Ecdy-
sone”. Ed. by J. Koolman, Georg Thieme Verlag,
Stuttgart/New York, pp. 39-64.

Singh, P. and Russel, G. B. (1980) J. Insect
Physiol., 26: 139-142.

Hikino, H., Ohizumi, Y. and Takemoto, T. (1975)
J. Insect Physiol., 21: 1953-1963.

Adam, G. and Marquardt, V. (1986) Phytochemis-
try, 25: 1787-1799.

Mandava, N. B. (1988) Ann. Rev. Plant Physiol.
Plant Mol. Biol., 39: 23-52.

Karlson, P. and Hoffmeister, H. (1963) Z. Physiol.
Chem., 331: 298-300.

Clayton, R. B. (1964) J. Lipid Res., 5: 3-19.
Clark, A. I. and Bloch, K. (1959) J. biol. Chem.,
234: 2589-2594.

Svoboda, J. A. and Thompson, M. J. (1985) In
“Comprehensive Insect Physiology, Biochemistry,
and Pharmacology”. Vol. 10, Ed. by G. A. Kerkut
and L. I. Gilbert, Pergamon Press, Oxford, pp.
137-175.

Ikekawa, N. (1985) In “Sterols and Bile Acids”.
Ed. by H. Danielson and J. Sjovall, Elsevier,
Amsterdam, pp. 199-230.

Ikekawa, N. (1983) Experientia, 39: 466-472.
Rees, H. H. (1989) In “Ecdysone”. Ed. by J.
Koolman, Georg Thieme Verlag, Stuttgart/New
York, pp. 28-38.

Sakurai, S., Ikekawa, N., Ohtaki, T. and Chino,
H. (1977) Science, 198: 627-629.

Hetru, C., Kappler, C., Hoffmann, J. A., Nearn,
R., Luu, B. and Horn, D. H. S. (1982) Mol. Cell.
Endocrinol., 26: 51-81.

Meister, M. F., Dimarcq, J. L., Kappler, C.,
Hetru, C., Lagueux, M., Lanot, R., Luu, B. and
Hoffmann, J. A. (1985) Mol. Cell. Endocrinol.,
41: 27-44.

Kappler, C., Hetru, C., Durst, F. and Hoffmann,
J. A. (1989) In “Ecdysone”. Ed. by J. Koolman,
Georg Thieme Verlag, Stuttgart/New York, pp.
161-166.

54

55

56

S/

58

59
60

61

62

63

64

65

66

67

68

69

70

71

72

Ecdysteroids

Feldlaufer, M. F. and Svoboda, J. A. (1986) Insect
Biochem., 16: 45-48.

Redfern, C. P. F. (1989) In “Ecdysone”. Ed. by J.
Koolman, Georg Thieme Verlag, Stuttgart/New
York, pp. 182-187.

Fukuda, S. (1940) Proc. Imperial. Acad. Tokyo,
16: 417-420.

Chino, H., Sakurai, S., Ohtaki, T., [Ikekawa, N.,
Miyazaki, H., Ishibashi, M. and Abuki, H. (1974)
Science, 183: 529-530.

King, D. S., Bollenbacher, W. E., Borst, D. W.,
Vedeckis, W. V., O’Connor, J. D., Ittycheriah, P.
I. and Gilbert, L. I. (1974) Proc. Natl. Acad. Sci.
U.S.A., 71: 793-796.

Birkenbeil, H. (1990) Zool. Jb. (In press)
Warren, J. T., Sakurai, S., Rountree, D. B. and
Gilbert, L. I. (1988) J. Insect Physiol., 34: 571-
576:

Gilbert3> ie (1989) In “Ecdysone”. Ed. “by: J:
Koolman, Georg Thieme Verlag, Stuttgart/New
York, pp. 448-471.

Kiriishi, S., Rountree, D. B., Sakurai, S. and
Gilbert, L. I. (1990) Experientia (In press)
Lanot, R., Dorn, A., Gitinster, B., Thiebold, J.,
Lagueux, M. and Hoffmann, J. A. (1989) In
“Ecdysone”. Ed. by J. Koolman, Georg Thieme
Verlag, Stuttgart/New York, pp. 262-270.
Hagedorn, H. H., O’Connor, J. D., Fuchs, M. S.,
Sage, B., Schlaeger, D. A. and Bohm. M. K.
(1975) Proc. Natl. Acad. Sci. U.S.A., 72: 3255-
S259!

Ohnishi, E., Mizuno, T., Chatani, F., Ikekawa, N.
and Sakurai, S. (1977) Science, 197: 66-67.
Goltzené, F., Lagueux, M., Charlet, M. and Hoff-
mann, J. A. (1978) Hoppe-Seylers Z. Physiol.
Chem., 359: 1427-1434.

Loeb, M. J., Woods, C. W., Brandt, E. P. and
Borkovec, A. B. (1982) Experientia, 218: 896-
898.

Chang, E. S. and O’Connor, J. D. (1977) Proc.
Natl. Acad. Sci. USA, 74: 615-618.

Spindler, K.-D., Keller, R. and O’Connor J. D.
(1981) In “Progress in Ecdysone Research”. Ed.
by J. Hoffmann, Elsevier/North Holland, Amster-
dam, pp. 247-280.

Watson, R. D., Spaziani, E. and Bollenbacher, W.
E. (1989) In “Ecdysone”. Ed. by J. Koolman,
Georg Thieme Verlag, Stuttgart/New York, pp.
188-203.

Nakanishi, K., Moriyama, H., Okauchi, T., Fu-
jioka, S. and Koreeda, M. (1972) Science, 176: 51-
323

Delbecque, J. P., Delachambre, J., Hirn, M. and
De Reggi, M. (1978) Gen. Comp. Endocrinol., 35:
436-444.

73

74

75

76

Ty

78
79

80

81

82

83

84

85

86

87

88

89

90

Oi

92

aii)

Bownes, M., Shirras, A., Blair, M., Collins, J. and
Coulson, A. (1988) Proc. Natl. Acad. Sci. USA,
85: 1554-1557.

Smith, S. L. (1985) In “Comprehensive Insect
Physiology, Biochemistry, and Pharmacology”
Vol. 10. Ed. by G. A. Kerkut and L. I. Gilbert,
Pergamon Press, Oxford, pp. 295-342.

Steel, C. G. H. and Vafopoulou, X. (1989) In
“Ecdysone”. Ed. by J. Koolman, Georg Thieme
Verlag, Stuttgart/New York, pp. 221-231.
Truman, J. W. and Riddiford, L. M. (1974) J.
Exp. Biol., 60: 371-382.

Riddiford, L. M. (1981) Amer. Zool., 21: 751-
762.

Richards, G. (1976) Develop. Biol., 54: 256-263.
Fristrom, J. W., Natzle, J., Doctor, J. and Fris-
trom, D. (1985) In “Metamorphosis”. Ed. by M.
Ball and M. Bownes, Clarendon Press, Oxford, pp.
162-180.

Whitehead, D. L. (1989) In “Ecdysone”. Ed. by J.
Koolman, Georg Thieme Verlag, Stuttgart/New
York, pp. 232-244.

Feyereisen, R. (1985) In “Methods in Enzymolo-
gy”. Vol. 111, Ed. by J. H. Law and H. C. Rilling,
Academic Press, New York, pp. 442-453.
Koolman, J. (1985) In “Methods in Enzymology”.
Vol. 111, Ed. by J. H. Law and H. C. Rilling,
Academic Press, New York, pp. 429-437.

Natori, S. and Ohtaki, T. (1976) J. Insect Physiol.,
22: 227-234.

Daig, K. and Spindler, K.-D. (1983) Mol. Cell
Endocrinol., 31: 93-104.

Daig, K. and Spindler, K.-D. (1983) Mol. Cell
Endocrinol., 32: 73-79.

Dinan, L. and Spindler, K.-D. (1986)
Biochem., 16: 135-141.

Spindler, K.-D. and Spindler-Barth, M. (1989) In
“Ecdysone”. Ed. by J. Koolman, Georg Thieme
Verlag, Stuttgart/New York, pp. 245-249.
Koolman, J. (1982) Insect Biochem., 12: 225-250.
King, D. S. (1972) Gen. Comp. Endocrinol., 3:
221-227.

Feyereisen, R., Lagueux, M. and Hoffmann, J. A.
(1976) Gen. Comp. Endocrinol., 29: 319-327.
Modde, J. F., Lafont, R. and Hoffmann, J. A.
(1984) Int. J. Invertebr. Reprod. Dev., 7: 161-
183.

Connat, J.-L., Dotson, E. M. and Diehl, P. A.
(1988) Arch. Ins. Biochem. Physiol., 9: 221-235.
Koolman, J. (1978) In “Comparative Endocrinolo-
gy”. Ed. by P. J. Gaillard, and H. H. Boer,
Elsevier/North Holland, Amsterdam, pp. 495-498.
Ohtaki, T. (1981) Amer. Zool., 21: 727-731.
Lafont, R. and Koolman J. (1984) In “Biosynth-
esis, Metabolism and Mode of Action of Inverte-

Insect

578

96

97

98

99

100

101

102

103

104

105

106
107

108

109

110

111

112

13)

114

brate Hormones”. Ed. by J. Hoffmann and M.
Porchet, Springer-Verlag, Berlin/New York, pp.
196-226.

Koolman, J. and Karlson, P. (1985) In “Compre-
hensive Insect Physiology, Biochemistry, and Phar-
macology” Vol. 10, Ed. by G. A. Kerkut and L. I.
Gilbert, Pergamon Press, Oxford, pp. 343-361.
Lafont, R., Beydon, P., Blais, C., Garcia, M.,
Lachaise, F., Riera, F., Somme, G. and Girault, J.
P. (1986) Insect Biochem., 16: 11-16.

Lafont, R. and Connat, J. L. (1989) In “Ecdy-
sone”. Ed. by J. Koolman, Georg Thieme Verlag,
Stuttgart/New York, pp. 167-173.

Thompson, M. J., Weirich, G. F. and Svoboda, J.
A. (1990) In “Morphogenetic Hormones of
Arthropods”. Ed. by A. P. Gupta, Rutgers Univ.
Press, New Brunswick/London, pp. 325-360.
Koolman, J. and Walter, J. (1985) In “Meta-
morphosis”. Ed. by M. Bownes and M. Balls,
Clarendon Press, Oxford, pp. 198-220.

Beydon, P. and Lafont, R. (1987) Arch. Ins.
Biochem. Biophys., 5: 139-154.

Somme-Martin, G., Colardeau, J. and Lafont, R.
(1988) Insect Biochem., 18: 735-742.

Warren, J. T. and Gilbert, L. I. (1986) Insect
Biochem., 16: 65-82.

Riddiford, L. M. (1985) In “Comprehensive Insect
Physiology, Biochemistry and Pharmacology”.
Vol. 8, Ed. by G. A. Kerkut and L. I. Gilbert,
Pergamon Press, Oxford, pp. 37-84.

Fristrom, J. W., Raikow, R., Petri, W. and Ste-
wart, D. (1969) In “RNA in Development” Ed. by
E. W. Hanly, Univ. of Utah Press, Salt Lake City,
pp. 381-401.

Mandaron, P. (1970) Develop. Biol., 22: 298-320.
Dumser, G. B. (1980) Ann. Rev. Entomol., 25:
341-369.

Raabe, M. (1986) Advance. Insect Physiol., 19: 29-
154.

Spindler, K.-D. (1989) In “Ecdysone”. Ed. by J.
Koolman, Georg Thieme Verlag, Stuttgart/New
York, pp. 290-293.

Lachaise, F. (1990) In “Morphogenetic Hormones
of Arthropods”. Ed. by A. P. Gupta, Rutgers
Univ. Press, New Brunswick/London, pp. 276-
324.

Runion, H. I. and Pipa, R. L. (1970) J. exp. Biol.,
53: 9-24.

Schwartz, L. M. and Truman, J. W. (1984) J. exp.
Biol., 111: 13-30.

Booker, R. and Truman, J. W. (1987) J. Neuros-
ci., 7: 4107-4114.

Hagedorn, H. H. (1985) In “Comprehensive Insect
Physiology, Biochemistry and Pharmacology”.
Vol. 8, Ed. by G. A. Kerkut and L. I. Gilbert,

J. KOOLMAN

US

116

iLiL7/

118

I)

120

121

122

123

124

125

126

ZT)

128

129

130

131

132

133

134

Pergamon Press, Oxford, pp. 205-262.

Hagedorn, H. H. (1989) In “Ecdysone”. Ed. by J.
Koolman, Georg Thieme Verlag, Stuttgart/New
York, pp. 279-289.

Briers, T. and Huybrechts, R. (1984)
Biochem., 14: 121-126.

Bownes, M. (1989) In “Ecdysone”. Ed. by J.
Koolman, Georg Thieme Verlag, Stuttgart/New
York, pp. 414-420.

Schwartz, M. B., Kelly, T. J., Imberski, R. B. and
Rubenstein, E. C. (1985) J. Insect Physiol., 31:
947-957.

Lanot, R., Thiebold, J., Lagueux, M., Goltzené F.
and Hoffman, J.A. (1987) Dev. Biol., 121: 174-
181.

Lanot, R., Thiebold, J., Costet-Corio, M. F.,
Benveniste, P. and Hoffmann, J. A. (1988) De-
velop. Biol., 126: 212-214.

Karlson, P. (1963) Perspect. Biol. Med., 6: 203-
214.

Clever, U. and Karlson, P. (1960) Exp. Cell. Res.,
20: 623-627.

Lezzi, M. and Richards, G. (1989) In “Ecdysone”.
Ed. by J. Koolman, Georg Thieme Verlag, Stutt-
gart/New York, pp. 393-406.

Koolman, J. and Spindler, K.-D. (1983) In “En-
docrinology of Insects”. Ed. by R. G. H. Downer
and H. Laufer, Alan Liss, New York, pp. 179-201.
Pongs, O. (1989) In “Ecdysone”. Ed. by J. Kool-
man, Georg Thieme Verlag, Stuttgart/New York,
pp. 338-344.

Kauser, G., Koolman, J. and Karlson, P. (1990) In
“Morphogenetic Hormones of Arthropods”. Ed.
by A. P. Gupta, Rudgers Univ. Press, New Bruns-
wick/London, pp. 361-388.

Yund, M. A. and Osterbur, D. L. (1985) In
“Comprehensive Insect Physiology, Biochemistry,
and Pharmacology”. Vol. 8, Ed. by G. A. Kerkut
and L. I. Gilbert, Pergamon Press, Oxford, pp.
473-490.

Bidmon, H.-J. and Koolman, J. (1989) Experien-
tia, 45: 106-109.

Maroy, P., Dennis, R., Beckers, C., Sage, B. A.
and O’Connor, J. D. (1978) Proc. Natl. Acad. Sci.
USA, 75: 6035-6038.

Yund, M. A., King, D. S. and Fristrom, J. W.
(1978) Proc. Natl. Acd. Sci. USA, 75: 6039-6043.
Sage, B. A., Horn, D. H. S., Landon, T. M. and
O’Connor, J. D. (1986) Arch. Ins. Biochem.
Physiol. Suppl., 1: 25-33.

Lehmann, M. and Koolman, M. (1988) Mol. cell.
Endocrinol., 57: 239-249.

Lehmann, M. and Koolman, J. (1989) Eur. J.
Biochem., 181: 577-582.

Berg, J. M. (1989) Cell, 57: 1065-1068.

Insect

[S5

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

Syl

S52

£53

154

Ecdysteroids

Deak, P., Zavorsky, P. and Maroy, P. (1988)
Insect Biochem., 18: 847-852.

Porcheron, P., Moriniere, M., Grassi, J. and
Pradelles, P. (1989) Insect Biochem., 19: 117-122.
Reum, L. and Koolman, J. (1989) In “Ecdysone”.
Ed. by J. Koolman, Georg Thieme Verlag, Stutt-
gart/New York, pp. 131-143.

Morgan, E. D. and Wilson, I. D. (1989) In
“Ecdysone”. Ed. by J. Koolman, Georg Thieme
Verlag, Stuttgart/New York, pp. 114-130.
Girault, J. P. and Lafont, R. (1988) J. Insect
Physiol., 34: 701-706.

Thompson, M. J. and Lusby, W. R. (1989) In
“Ecdysone”. Ed. by J. Koolman, Georg Thieme
Verlag, Stuttgart/New York, pp. 65-73.

Lafont, R. and Beydon, P. (1990) In “Mor-
phogenetic Hormones of Arthropods”. Ed. by A.
P. Gupta, Rudgers Univ. Press, New Brunswick/
London, pp. 455-512.

Luu, B. and Hetru, C. (1989) In “Ecdysone”. Ed.
by J. Koolman, Georg Thieme Verlag, Stuttgart/
New York, pp. 106-112.

Cherbas, P., Cherbas, L., Lee, S.-S. and Naka-
nishi, K. (1988) Proc. Natl. Acad. Sci. USA, 85:
2096-2100.

Kauser, G. (1989) In “Ecdysone”. Ed. by J.
Koolman, Georg Thieme Verlag, Stuttgart/New
York, pp. 327-336.

Rees, H. H. and Mercer, J. G. (1986) In “Adv-
ances in Invertebrate Reproduction”, Vol. 4, Ed.
by M. Porchet, J.-C. Andries and A. Dhainaut,
Elsevier, Amsterdam, pp. 173-186.

Spindler, K.-D. (1988) In “Parasitology on Focus,
Facts and Trends” Ed. by H. Mehlhorn, Springer-
Verlag, Berlin, pp. 465-476.

Franke, S. and Kauser, G. (1989) In “Ecdysone”.
Ed. by J. Koolman, Georg Thieme Verlag, Stutt-
gart/New York, pp. 296-307.

Simon, P. and Koolman, J. (1989) In “Ecdysone”.
Ed. by J. Koolman, Georg Thieme Verlag, Stutt-
gart/New York, pp. 254-259.

Koolman, J. (1990) In “Progress in Comparative
Endocrinology.” Ed. by A. Epple, C. G. Stanes
and M. H. Stetson, John Wiley, New York, pp.
704-709.

Kauser, G., Brandtner, H., Bidmon, H.-J. and
Koolman, J. (1988) J. Insect Physiol. , 34: 563-570.
Richter, K., Kauser, G. and Bidmon, H.-J. (1989)
In “Ecdysone”. Ed. by J. Koolman, Georg Thieme
Verlag, Stuttgart/New York, pp. 319-326.
Williams, C. M. (1952) Biol. Bull., 103: 120-138.
Beydon, P. and Lafont, R. (1983) J. Insect Phy-
siol., 29: 529-533.

Bodnaryk, R. P. (1986) Arch. Ins. Biochem.
Physiol., 3: 53-60.

155

156

357/

158

159

160

161

162

163

164

165

166

167

168

169

170

iL 7/A

172

173

174
7S)

176

iT)

178

579

Sakurai, S. and Williams, C. M. (1989) Gen.
Comp. Endocrinol., 75: 204-216.

Keller, R. and O’Connor, J. D. (1982) Gen.
Comp. Endocrinol., 46: 384.

Huberman, A. and Aguilar, M. B. (1989) Comp.
Biochem. Physiol., 93B: 299-303.

Webster, S. G. and Keller, R. (1989) In “Ecdy-
sone”. Ed. by J. Koolman, Georg Thieme Verlag,
Stuttgart/New York, pp. 211-216.

Gilbert, L. I., Combest, W. L., Smith, W. A.,
Meller, V. H. and Rountree, D. B. (1988) BioEs-
says, 8: 153-157.

Lachaise, F., Hubert, M., Webster, S. G. and
Lafont, R. (1988) J. Insect Physiol., 34: 557-562.
Gilbert, L. I. and Schneiderman, H. (1959) Na-
ture, 184: 171-173.

Sakurai, S., Okuda, M. and Ohtaki, T. (1989)
Gen. Comp. Endocrinol., 75: 222-230.
Riddiford, L. M. (1980) In “Progress in Ecdysone
Research”. Ed. by J. A. Hoffmann, Elsevier/North
Holland, Amsterdam, pp. 409-430.

Watson, R. D., Whisenton, L. R., Bollenbacher,
W.E. and Granger, N. A. (1986) Insect Biochem..,
16: 149-155.

Hoggard, N. and Rees, H. H. (1988) J. Insect
Physiol., 34: 647-654.

Lehmann, M. and Koolman, J. (1986) Biol. Chem.
Hoppe-Seyler, 367: 387-393.

Lehmann, M. and Koolman, J. (1989) In “Ecdy-
sone”. Ed. by J. Koolman, Georg Thieme Verlag,
Stuttgart/New York, pp. 217-220.

Beato, M. (1989) Cell, 56: 335-344.

Garen, A., Kauvar, L. and Lepesant, J.-A. (1977)
Proc. Natl. Acad. Sci. USA, 79: 5099-5113.
Nauber, U., Pankratz, M.J., Kienlin, A., Seifert,
E., Klemm, U. and Jaeckle, H. (1989) Nature,
336: 489-492.

Grau, V. and Gutzeit, H. O. (1990) Roux’s Arch.
Dev. Biol., 198: 295-302.

Spindler-Barth, M. and Spindler, K.-D. (1987) In
“Receptor Mediated Antisteroid Action”. Ed. by
M. K. Agarwal, De Gruyter, Berlin/New York,
pp. 497-511.

Wing, K. D., Slawecki, R. A. and Carlson, G. R.
(1988) Science, 241: 470-472.

Wing, K. D. (1988) Science, 241: 467-469.
Hetru, C., Roussel, J.-P., Mori, K. and Nakatani,
Y. (1986) C. R. Acad. Sci. Paris, 302: 417-420.
Richter, K., Adam, G. and Vorbrodt, H. M.
(1987) J. Appl. Entomol., 103: 532-544.
Lehmann, M. Vorbrodt, H.M., Adam, G. and
Koolman, J. (1988) Experientia, 44: 355-356.
Butterworth, J. H. and Morgan, E. D. (1968) J.
chem. Soc., Chem. Commun., — : 23-24.

580

179

180

181

182

183

184

185

186

187
188

189

Turner, C. J., Tempesta, M. S., Taylor, R. B.,
Zagorski, M. G., Termini, J. S., Schroeder, D. R.
and Nakanishi, K. (1987) Tetrahedron, 12: 2789-
2803.

Schmutterer, H. (1988) J. Insect Physiol., 34: 713-
720:

Sieber, K.-P. and Rembold, H. (1983) J. Insect
Physiol., 29: 523-527.

Smith, S. L. and Mitchell, M. J. (1988) Biochem.
Biophys. Res. Commun., 154: 559-563.
Koolman, J., Bidmon, H.-J., Lehmann, M. and
Kauser, G. (1988) In “Endocrinological Frontiers
in Physiological Insect Ecology”. Vol. 1, Ed. by F.
Sehnal, A. Zabza, and D. L. Denlinger, Wroclaw
Technical Univ. Press, pp. 55-68.
Subrahmanyam, B. and Rembold, H. (1989) Cell
Tissue Res., 256: 513-517.

Bidmon, H.-J., Kauser, G., Modbus, P. and Kool-
man, J. (1987) In “Natural Pesticides from the
Neem Tree (Azadirachta indica A. Juss) and Other
Tropical Plants”, Ed. by H. Schmutterer, and K.
R. S. Ascher, GTZ, Eschborn, pp. 253-272.
Pener, M. P., Rountree, D. B., Bishoff, S. Y. and
Gilbert, L. I. (1988) In “Endocrinological Fron-
tiers in Physiological Insect Ecology”. Vol. 1, Ed.
by F. Sehnal, A. Zabza and D. L. Denlinger,
Wroclaw Technical Univ. Press, pp. 41-54.
Richter, K. (1989) Biol. Rundsch., 27: 73-87.
O’Reilly, D.R. and Miller, L.K. (1989) Science,
245: 1110-1112.

Kadono-Okuda, K., Kuwano, E., Eto, M. and

J. KOOLMAN

190

194

195

196

197

198

199

Yamashita, O. (1987) Develop. Growth & Differ.,
29: 527-533.

Burger, A., Roussel, J.-P., Hetru, C., Hoffmann,
J. A. and Luu, B. (1989) Tetrahedron, 45: 155-
164.

Prestwich, G. D., Hong Ming S. and Apurba, K.
G. (1983) Steroids, 41: 79-94.

Svoboda, J. A., Thompson, M. J. and Robbins, W.
E. (1972) Lipids, 7: 553-556.

Robbins, W. E., Thompson, M. J., Svoboda, J.
A., Shortino, T. J., Cohen, C. F., Dutky, S. R. and
Duncan, O. J. (1975) Lipids, 10: 353-359.
Corio-Costet, M. F., Charlet, M., Benveniste, P.
and Hoffmann, J. A. (1989) Arch. Ins. Biochem.
Physiol., 11: 47-62.

Williams, C. M. (1967) Sci. Amer., 217: 13-17.
Hoffmann, J. A., Lagueux, M., Hetru, C., Char-
let, M. and Goltzene, F. (1980) In “Progress in
Ecdysone Research”. Ed. by J. A. Hoffmann,
Elesvier North/Holland, Amsterdam, pp. 431-465.
Hanaoka, K. and Hagedorn, H. H. (1980) In
“Progress in Ecdysone Research”. Ed. by J. A.
Hoffmann. Elesvier North/Holland, Amsterdam,
pp. 467-480.

Loeb, M. J., Brandt, E. P., Woods, C. W. and
Borkovec, A. B. (1987) J. exp. Zool., 243: 275-
282.

Karlson, P., Doenecke, D., Fuchs, G., Koolman,
J. and Schafer, G. (1988) Biochemie. Georg-
Thieme Verlag, Stuttgart/New York, 13th ed.

ZOOLOGICAL SCIENCE 7: 581-591 (1990)

REVIEW

Functions and Properties of Animal Lectins

Norio YOSHIZAKI

Department of Biology, Faculty of General Education,
Gifu University, Gifu 501-11, Japan

ABSTRACT—A variety of animal lectins have been isolated and characterized by extensive studies
during a century after the discovery of a lectin in 1888. Animal lectins distribute in a variety of
locations of the body and perform a variety of roles. However, it is still difficult to determine
endogenous functions of lectins since their structures both of the primary amino acid sequence and of
the tertiary structure detected immunologically were alike each other among several lectins, and
endogenous ligands were obscure. This review describes animal lectins whose functions are relatively
apparent. Agglutination of erythrocytes and/or glycoproteins has been an important factor for the
definition of a lectin so far, representing the existence of two sugar-binding sites. Since the discovery
of sugar-binding proteins which have a single binding site, it is claimed to build a new definition to

© 1990 Zoological Society of Japan

reconcile all sugar-binding proteins under the term lectin.

INTRODUCTION

Since Hermann Stillmark discovered a hemag-
glutinin in extracts of castor beans in 1888, many
blood cell-binding proteins have been isolated
from many plant seeds. The term lectin, which is
from Latin legere, was first used by Boyd and
Shapleigh [1] in 1954 to describe blood group-
specific plant agglutinins. As lectins were known
to be present universally in virus and in animal
cells and/or tissues as well as in those of plants, in
1980 Goldstein et al. [2] proposed the following
definition, with comments, to distinguish aggluti-
nins specific to glycoconjugates from the others: A
lectin is a sugar-binding protein of nonimmune
origin that agglutinates cells and/or precipitates
glycoconjugates. The comments were:

1. Lectins contain at least two sugar-binding
sites; sugar-binding proteins with a single site are
not classified as lectins.

2. The specificity of a lectin is usually defined
in terms of the monosaccharide(s) or oligosacchar-
ides that inhibit lectin-induced agglutination or
precipitation reactions.

Received April 26, 1990

3. Lectins occur in many types of organism;
they may be soluble in biological fluids or mem-
brane-bound.

4. Sugar-specific enzymes, transport proteins,
hormones and toxins may qualify as lectins if they
have multiple sugar-binding sites.

Although Barondes [3] proposed to change the
definition, as will be mentioned later, this review
will employ the definition heretofore in use.

In the field of zoology, plant lectins have widely
been used mostly as experimental tools, such as in
purification of polysaccharides and glycoproteins,
and in a variety of biological applications including
separation of cells, induction of mitogenesis in
lymphocytes, subcellular localization of glyco-
conjugates and targeting of hormones and toxins
to cells. Although a large number of lectins were
isolated from many species of animals and from a
variety of tissues in the 1970’s, it was not evident
what role the endogenous lectins have in situ.
Functions of plant lectins were not clear either [4],
nor are they yet. The main reason for ill success in
finding functions is inability to determine endoge-
nous ligands. Thus reviews dealing with lectin
function are few. A review by Barondes [5] in 1981
was still limited to a description of functions of

582 N. YOSHIZAKI

hepatic and embryonic lectins. During the last ten
years, papers dealing with endogenous functions of
lectins or with new lectins have increased. This
review will deal with functions of lectins which
have been relatively clarified. The review will also
be made on the definition of lectins which is
claimed to be changed after findings of monovalent
sugar-binding proteins which have highly homolo-
gous, amino acid sequences to those of known
lectins.

FUNCTIONS OF ANIMAL LECTINS

(a) Primitive immune surveillance

It is a well-known fact that invertebrates lack
immunoglobulins. However, lectins are conjec-
tured to behave like immunoglobulins by binding
to non-self materials such as bacteria invading a
host. Lectins present in a coelomic fluid or
hemolymph are apparently different from anti-
bodies in that they recognize differences only in
polysaccharides [6]. Hemolymph lectins which

TABLE 1.

Animal Sugar specificity
clam GalNAc**
snail NANA
coastal crab NANA
horseshoe crab NANA
giant stick insect Gal
beetle Gal
beet armyworm Gal
flesh fly Gal
giant silk moth Gal/GalNAc
oak silk moth Gal/Glu
grasshopper Gal/Glu
Bombyx Glucuronic acid
cockroach GalN Ac/Fuc
sea urchin NANA-Gal-NANA-

GalNAc
ascidian Gal
amphibian Gal
chicken Man/GluNAc
rat Man/GluNAc
human Glu-Gal

have been isolated and characterized [7-21] are
listed in Table 1. Hemolymph lectins of inverte-
brates are produced and secreted by coelomocytes
or hemocytes [6] and thus are present also on the
cell surface [22, 23]. Parasites and/or bacteria
which invade host animals may be bound by lectins
on hemocytes and engulfed directly by the hemo-
cytes or they may be coated with hemolymph
lectins and trapped by the hemocytes after the
opsonic effect of the lectins [6]. Opsonizing prop-
erties of lectins were clarified by an experiment
demonstrating the removal by hemocytes of yeast
injected in Mytilus [22]. On the other hand, it is
known that in holometabolous insects lectin activi-
ties increase transiently just before pupation [14,
18]. Thus the lectin may also play a role in
surveillance of decomposed larval tissues by acti-
vating hemocytes during metamorphosis.
Primitive immune system mediated by serum
lectins (Table 1) is also found in vertebrates [24—
28]. Lectins which bind specifically to Man and
N-GluNAc are generically named as mannan-
binding proteins (MBP) in mammals. The role of

Hemolymph or serum lectins of animals

Molecular weights* Rfs
470 K (10, 20, 40 K) [7]
242 K (15 K) [8]
? 36K 9
Sue 1
72-716 K Ce K) it
38 K (17.5, 20 K) 12
39 K (19.2 K)

P(r, SILLS) 13
ISOIK (C0; 32 Ry 14
160K (38, 41K 1S)
380 K (38 K) 16
600-700 K (28, 40K) Wy)
260 K 18
1500 K (30 K) 19
300 K (13 K) [20]
56.6 K (13-14K ) [21]
57.5K (14-15 K

69K [24]
169 K (27, 74K) [25, 26]
650 K (31 K) [27]
240 K (29.5 K) [28]

* Numbers in parenthesis are molecular weights of subunits.
** Abbreviations: Fuc, fucose; Gal, galactose; GalNAc; N-acetyl galactosamine; Glu, glucose; GluNAc,
N-acetyl glucosamine; Man, mannose; NANA, N-acetyl neuraminic acid.

Functions of Animal Lectins 583

TABLE 2. Macrophage lectins of mammals

Source of lectins

Sugar specificity

body cavity macrophage

rat Man/Fuc/GluNAc

mouse Gal/GalNAc

lung alveolar macrophage

rat Man/Fuc/GluNAc
Man/Fuc/GluNAc

rabbit

Molecular weights Rfs
180K [30]
45-60 K [31]
180K [32533]
175K [34]

MBP as a bactericidal factor was proved in rat [27].
Sheep erythrocytes which were coated with yeast
mannan showed hemolysis under coexistence with
rat MBP and a guinea pig serum (complement
source). This fact suggested that MBP activates
the classical pathway of complement system and
thus the MBP itself behaves the same as Clq does.
High homology in the overall organization of the
amino acid sequence in MBP and Clq was also
observed [29].

Mammalian macrophage lectins present on the
cell surface [30-34] are listed in Table 2. Although
the role of macrophage lectins is hypothesized to
endocytose invaded bacteria [35], there is little
evidence. However, it was confirmed that some
glycoproteins with terminal Man or Glu specificai-
ly bound to the surface of rat alveolar mac-
rophages [36].

(6) Serum surveillance by the liver

Extensive investigations were performed in the
1970’s on sugar receptors bound on liver cells in
order to elucidate the regulation of serum gly-
coprotein homeostasis. The history of and indi-
vidual studies in this field had been reviewed by
Ashwell and Harford [37]. The terminal sugar
sequence of glycoproteins generally present in
living organisms is NANA-Gal-GluNAc-(Man)-
[38]. Partially degraded circulating glycoproteins
may be produced by the aging of molecules them-

selves and/or by the metabolism performed in |

macrophages of non-self or unnecessary self mate-
rials. In aserum surveillance mechanism, the liver
engulfs and removes those molecules which pos-
sess an unusual terminal sugar sequence, especially
those lacking NANA. As a matter of fact, an
increase in individual NANA-deficient glycopro-

teins appears in various liver disease patients [37].
At the beginning of these sugar-receptor studies,
the glycoproteins used were those which naturally
possess terminal sugars other than NANA, such as
ceruloplasmin, orosomucoid, IgG, RNase B, fe-
tuin, etc. [39], or those which were treated with
neuraminidase, (-D-galactosidase or $-N-acetyl-
glucosaminidase to remove the corresponding
sugars from the sugar chain terminals [40, 41].
Recently, “neoglycoproteins” have been synthe-
sized by binding specific sugars to BSA or ferritin
by amidination [39]. Definition of the structural
parameters of the glycose determinants capable of
influencing binding has been achieved using the
neoglycoprotein approach [42-44].

When '”I-labeled glycoproteins with unusual
terminal sugar residues were injected intravenous-
ly, they were promptly removed from the blood
plasma and transferred to the liver [40, 44-47].
The transfer of labeled glycoproteins was also
confirmed histochemically [42, 44]. That sugar
residues of glycoproteins participate in this trans-
fer phenomenon was demonstrated by proving the
inability of the liver to take up glycoproteins
modified by peroxidase oxidation [46]. Involve-
ment of a lectin as sugar receptor was confirmed in
an experiment showing that after perfusion of the
liver by an antibody against the lectin, the influx
rate of injected asialoglycoprotein decreased 80%,
whereas no change occurred in the influx rate of
biliverdin [48].

When Gal-specific lectins on rat liver cell mem-
branes bound with ligand molecules of asialogly-
coproteins, they formed microaggregations [49]
and were then involved morphologically in the
formation of coated pits and vesicles under the
influence of intracellular clathrin [50]. After 2 min

584 N. YOSHIZAKI

of glycoprotein administration, the receptor-ligand
complexes were transferred to larger smooth-
surfaced vesicles and/or tubular structures [44]. It
is remarkable that no appreciable aryl sulfatase
activity was found in the large vesicles, and the
particulate tracer labeling the ligand was found
scattered throughout the lumen of the vesicles,
possibly indicating dissociation of the ligand from
its receptor. After 5 min, the particulate tracer
entered the Golgi-lyscsome regions which con-
tained the aryl sulfatase activities [43, 44]. On the
other hand, Tanabe et al. [51] found that the
receptor site of the lectin lies at the cytosolic
(outer) surface of lysosomal membranes and
metioned that the receptor (lectin) is spared des-
truction but the ligand is destroyed; they suggested
the receptor is recycled. As a matter of fact,
turnover time of receptors on the cell membrane
was fast (3 min) in the presence of asialoglycopro-
teins and two times more ligands than were pro-
jected were metabolized in the presence of cyc-
loheximide [52]. These results support the recycl-
ing receptor hypothesis.

Three groups of receptors on the liver cell
surface are known: Gal-binding receptors, Fuc-
binding receptors on hepatocytes and Man- and
GluNAc-binding receptors (MBP) on Kupffer and
endothelial cells [37, 40, 42]. These lectins, iso-
lated from solubilized membranes [47, 53-60], are
listed in Table 3.

TABLE 3.

Lectin species Animal

Gal-binding lectin
rabbit
rabbit
rat
rat

human

MBP
chicken
chicken
rat

Fuc-binding lectin
rat

The amino acid sequence of lectins was able to
be deduced recently from cDNA prepared by
techniques of molecular genetics. The protein
structure of rat MBP thus deduced was 148-150
amino acids in carbohydrate-binding domain at the
COOH-terminal, 18-20 repeats of the sequence
Gly-X-Y (collagen-like domain) at the middle and
18-19 amino acids with cysteine-rich segments at
the NH>-terminal [29, 61]. Human liver MBP is
likewise composed of 148 amino acids at the
COOH-terminal (carbohydrate-binding domain),
19 repeats of Gly-X-Y and a cysteine-rich NH>-
terminal [62], and a liver Gal-binding lectin with a
Gal-binding domain at the COOH-terminal and a
hydrophobic segment at the NH>-terminal [56].

Cystein-rich NH)-terminals would mediate the
polymerization of proteins liberated from mem-
branes in aqueous solution. If polymerization
occurs at the NH>-terminal, receptors which have
a single binding site, though they should not be
classified as lectins according to the definition, may
behave like lectins. As the number of binding sites
present in the carbohydrate-binding domain is not
clear for the lectins listed in Table 3, receptors
with a single binding site may be accepted as
lectins. The NH>-terminal is an anchor site at the
membrane and probably spans the membrane.
Transbilayer disposition of the receptor was shown
by staining with anti-lectin antiserum [63].

There are lectins of another type which are

Lectins present on the surface of mammalian liver cells

Molecular weights* Rfs

500 K (40, 48 K) [53, 54]
109 K [55]
104K [55]

2? (41.5, 49, 54K) [56]

2 (EIR) [S7]
210 K (26 K) [58]
640 K (41 K) [59]

2 (@iIS) [47]

2 Gl BIS) [60]

* Numbers in parenthesis are molecular weights of subunits.

Functions of Animal Lectins 585

bound to intracellular membranes in the liver [59,
64, 65]. They are considered to participate in
transportation of phosphorylated enzymes from
microsomes to Golgi complexes and to lysosomes.

(c) Morphogenesis

Embryonic chick pectoral or thigh muscle pro-
duced a 30 K lectin whose hemagglutinating activ-
ity was suppressed by thiodigalactoside or lactose
[66, 67]. The lectin activity, distributing both on
the cellular surface and in intracellular cytoplasm,
increased transiently at 8-16 days of development
[66]. Increase in the lectin activity was also
observed when multinucleated myotubes were
formed by fusion in cultures of rat L6 myoblasts
[68], the fusion of which was suppressed by 15 mM
thiodigalactoside [69]. Thus participation of lec-
tins in morphogenesis during development has
been advocated.

Although there have been studies with results
contradictory to lectin participation in the fusion of
myoblasts [67, 70], it was histochemically apparent
that lectins distributed intracellularly in myoblasts
and extracellularly in myotubes, and were present
only in trace amounts in adult muscles [71]. Sever-
al lectins were also obtained from developing brain
of rat and mouse: Gal-specific lectin whose activity
changed during development [72], 30-32 K Gal-
specific lectins which seemed to be involved in cell
ageregation [73], and a 13K Man-specific lectin
which participated in synapse formation [74].

The importance of the extracellular matrix in the
regulation of morphogenesis is common know-
ledge. Recent studies revealed the participation of
lectins in organization of the extracellular matrix.
Mammalian neonatal lungs have no alveoli; the
newborn embryo breathes with shallow saccules.
During the first 21 days of neonatal life, the lungs
are alveolarized and become mature respiratory
organs. The activity of 14K f-Gal-binding lectin
obtained from rat lung [75] was negligible at birth,
was apparent at about day 6, was maximal between
days 10 and 13 and thereafter declined 8-10 fold to
reach adult values [76]. The peak of lung-lectin
activity was coincident with the beginning of rapid
cross-linking of elastin which is one of the consti-
tuents of the extracellular matrix and has been
implicated as critical to lung restructuring [76]. In

bovine lung, localization of the same lectin in the
extracellular elastic fibers of pulmonary parenchy-
ma and blood vessels was demonstrated by im-
munofluorescent staining with antiserum against
the lectin [77].

A 67 K Gal-binding receptor which reacted to
the antiserum against the 14 K lectin but was not
accepted as a lectin was obtained from the surface
of bovine fibroblasts [78] and chondroblasts as well
as from the lung [79] by treating these cells with
lactose. That these 14 K lectins and 67 K receptors
are implicated in elastic fiber organization of cul-
tured chondroblasts was suggested by the fact that
addition of 5mM lactose to the culture medium
blocked the formation of elastic fibers in the
extracellular matrix but did not alter the ex-
tracellular distribution of fibronectin or microfib-
rillar-associated glycoprotein [79]. Probably bind-
ing of the elastins (or elastin fibers) to the cell
membrane is mediated by the lectins whereas the
extracellular matrix promotes cellular attache-
ment, migration, growth and differentiation. A
development-dependent Gal-binding lectin was
also obtained from chicken-skin extracellular mat-
rix [80].

Several kinds of lectins have been isolated from
amphibian eggs and embryos [81-89]. Some lec-
tins were exposed at the embryonic cell surface
[88] and are speculated to mediate the mor-
phogenetic movements of gastrulation and
neurulation [86, 87]. However, there is little
evidence to determine the role of these lectins.

(d) Polyspermy block

Most animal eggs possess mechanisms which
prevent polyspermy at fertilization. One such
polyspermy block mechanism is the so-called zona
reaction which blocks next sperm by physical bar-
riers formed through interaction between cortical
granule exudates and egg envelopes. In Xenopus
laevis, lectins were among the constituents of the
cortical granule exudates [90-93]. The cortical
granule lectins are the first whose function was
clearly shown because it is evident where the
lectins are compartmentalized, what and where the
ligands are, and finally where lectin-ligand com-
plexes are formed. The cortical granule lectins
consist of two molecules of 539 K and 655 K [93].

586 N. YOSHIZAKI

Fics. 1-2. Electron micrographs showing cross sections of egg envelopes of Xenopus uterine egg (Fig. 1) and an egg
30 min after insemination (Fig. 2). Note in Fig. 1 the existence of the prefertilization layer (PF) between the
vitelline coat (VC) and the jelly layer (J), showing a distribution pattern similar to the fertilization (F) layer in Fig.

Zs

Fics. 3-4. Thin sections of Xenopus unfertilized (Fig. 3) and fertilized eggs (Fig. 4) after incubation with antiserum
against cortical granule lectin and then treatment with protein A-gold solution. Gold particles are localized
exclusively in cortical granules (CG) in unfertilized egg but are present at approximately even densities over the F
layer, the VC and the perivitelline space (*) in fertilized egg. Over the F layer, however, particles are seen only
over the condensed (Fc) region immediately adjacent to the outer margin of the VC.

The two lectins are composed of identical subunit
proteins but are different in number of subunits,
both Gal-specific, contain the same amount of
Ca** , which is indispensable for hemagglutinating
activity, and share antigenicity [91]. Localization

of the lectins was shown immunohistochemically in
the cortical granules of unfertilized eggs [94, 95]
(Figs. 1, 3). The ligands of the cortical granule
lectins were glycoproteins in the prefertilization
layer. The prefertilization layer was apparently

Functions of Animal Lectins 587

different from the jelly layer and was located
between the vitelline coat and the jelly layer [96,
97]. Materials of the prefertilization layer were
produced in secretory cells of the pars recta of
oviducts and supplied to the eggs during passage
through this portion [98].

When cortical granule dehiscence occurred upon
fertilization or by artificial stimulation, the lectins
passed through interstices among fibrillar networks
of the vitelline coat and bound to the prefertiliza-
tion layer materials. After lectin-ligand interac-
tions, the prefertilization layer became electron-
dense and was then called the fertilization layer
[99-101] (Fig. 2). The electron-dense layer could
be formed in vitro by treating with isolated lectins
oviducal eggs which had been taken from pars
recta portions of the oviduct and covered with the
prefertilization layer [90]. Lectin-ligand interac-
tions could be visualized by a precipitin line forma-
tion in agar diffusion tests betwene two materials.
The interactions were inhibited by lactose, D-Gal
and methyl-a-D-Galactoside or EDTA which had
been impregnated in agarose plates [90]. About
5% of the secreted lectins were included in the
fertilization layer, whereas remaining lectins re-
mained enclosed in the perivitelline space [102].
Gold particles labeling the cortical granule lectins
were present at approximately even densities over
the fertilization layer, the vitelline coat and the
perivitelline space of fertilized eggs when observed
with immunoelectron microscopy (Fig. 4). Close
observation of the fertilization layer, however,
showed that the gold labeling was limited to its
condensed region and was poor in its dispersed
region. The fertilization layer was refractory to
various chemical and biochemical treatment,
whereas the prefertilization layer was easily dis-
solved [99, 103]. Resistence of the fertilization
layer to reagents might mirror resistence of ferti-
lization coats against invading sperm [104].

BIOCHEMICAL PROPERTIES OF LECTIN

Animal lectins generally recognize terminal
sugars of glycoproteins and bind them, although
exceptionally there are core-specific lectins which
bind to hydrophobic sites of sugar residues [105].
The sugars recognized by lectins are mainly Gal,

GluNac and Man. This implies such great biologic-
al significance that naturally produced glycopro-
teins usually possess NANA as a terminal sugar
but lectins ignore them. However, as listed in
Table 1, NANA-binding lectins are present in
hemolymph of lower invertebrates. NANA-
binding lectins would be expected in these animals
since these protostomes cannot synthesize their
own Sialic acids, although sialic acids taken in a
food are present in digestive organs [106].

The sugar-binding specificity of lectins is fre-
quently expressed with monosaccharides. This
may give the impression that the specificity of
lectins is low. However, if data concerning sugar
specificity of lectins is carefully inspected, it will be
appreciated that disaccharides—Gal-Glu [12, 90],
Fuc-GluNAc [46], Man-GluNAc [105] (and prob-
ably all MBPs) and Gal-Gal [66, 107] bind more
strongly than respective monosaccharides; on
some occasions, oligosaccharides can bind but
monosaccharides absolutely not [20, 28, 108]. The
binding specificity of lectins may have to be reex-
amined by using neoglycoproteins on which a
variety of oligosaccharides are coupled.

Animal lectins usually need Ca ions for activa-
tion. This characteristic has been used to dissoci-
ate lectins from immobilized ligand when an affini-
ty column was used as a purification tool. Howev-
er, there are lectins which show activity without
metals, for example, chick thigh muscle Gal-lectin
[70], chick serum MBP [25], bovine Man-6-P re-
ceptor [65] and rat lung Gal-binding protein [109].

Usually lectin molecules are polymers of
homogeneous or heterogeneous subunit glycopro-
teins as shown in Tables1 and 3. Post-
transcriptional regulation was suggested in flesh-fly
lectins for the production of heterogeneous subun-
its from a compact single copy gene [110].

Several lectins obtained from different sources
share antigenicity. Mucus-secreting goblet cells in
adult intestine, tubule cells in the kidney and
extracellular matrix in the pancreas of chickens
were ail stained by immunofluorescent techniques
with an antibody against lactose-binding lectin of
embryonic muscle [111]. By the same techniques
with an antibody against bovine heart muscle
Gal-binding lectin, all these organs—the brain,
spinal cord, small intestine, stomach, spleen and

588 N. YOSHIZAKI

lung were stained [112]. Recent analyses of amino
acid sequences of lectins and the inferred sequ-
ences of lectin cDNA have given a basis for
understanding common antigenicity among lectins.
High homology of amino acid sequences was
observed in a group of lectins which possess 13-14
K subunits and D-Gal-binding specificity. Murine
3T3 fibroblast lectin [113] had 94% homology with
rat lung lectin [109], 89% with human placenta
lectin [114], 86% with bovine heart muscle lectin
[115] and 46% with chicken skin lectin [80]. Elec-
tric-eel D-Gal-binding lectin of 12-16 K also had
high homology with this group of lectins [107]. In
another group, human liver (or serum) MBP [62]
had 48-51% homology of amino acid sequences
with rat liver MBPs [29, 61]. Thus lectins belong-
ing to each group might have a common ancestor
gene.

REVIEW OF DEFINITION OF LECTIN

Since the proposal of definition by Goldstein et
al [2] in 1980, the term lectin has been used to
denote those glycoproteins which have at least two
sugar-binding sites and agglutinate erythrocytes
but are not of immune origin. However, sugar-
binding proteins which do not conform with the
definition in the number of sugar-binding sites
have been isolated. For example, (1) a 35K
Gal-specific carbohydrate-binding protein
(CBP35) was obtained fom mouse 3T3 fibroblasts
[116]. It appeared in nuclei of multiplying cells,
reached its maximum amount at the beginning of
the S-phase [117], disappeared from the nuclei
after treatment with RNase [118], and had high
homology in amino acid sequence with hnRNP
[119]. CBP35 might be a component of hnRNP
and be implicated in DNA duplication and/or
transcriptional and post-transcriptional processes
[120]. Thus CBP 35 has two different binding sites
for sugar and RNA or protein, respectively. (2) A
surfactant-associated protein, SP28-36, was
obtained from dog and human lung [121]. The
surfactant is responsible for the maintenance of
alveolar stability by lowering the surface tension at
the air-liquid interface. SP28-36 has two different
binding sites, one for sugars and the other for
phospholipids [122]. (3) A 67K receptor isolated

from fibroblasts and chondroblasts, as mentioned
above [78, 79], has at least two binding sites, one
for D-Gal and the other for elastin (which consists
of pure proteins). (4) A lymphocyte homing
receptor that mediates selective attachment of
lymphocytes to specialized high endothelial
venules within secondary lymphoid organs was
obtained from mouse lymphocytes [123]. Analyses
of a cDNA sequence have revealed that this recep-
tor contains a series of protein motifs, that is,
sugar-binding domain, epidermal growth factor
domain, complement binding domain and trans-
membrane domain [124, 125]. As shown in several
papers [119, 121, 124, 125], inferred amino acid
sequences of the sugar-binding motif are highly
homologous to those of several known lectins.

Of hitherto-mentioned lectins, the only lectins
soluble in biological fluid are cortical granule and
serum lectins. Most of the remainder are consti-
tuents of membranes or extracellular matrices and
were isolated after solubilization by some reagent.
It is unthinkable for such artificially solubilized
lectins to sustain their natural structures in
aqueous solutions. Membrane-associated lectins .
may be polymerized at their hydrophobic segment
[56] after solubilization. More precisely, some
serum lectins are thought to have been released
from membranes of hemocytes or liver cells [22,
23, 62]. That most hemagglutinating lectins re-
ported are complexes of subunit proteins may
represent such an artificial or natural secondary
polymerization. At present, no report convincing-
ly demonstrates more than two carbohydrate-
binding sites in a monomeric polypeptide of lec-
tins. Hemagglutination by monovalent carbohy-
drate-binding polypeptides [understood by 89,
126] might be possible if they aggregated in high
concentration. Thus, some monovalent carbohy-
drate-binding polypeptides on the membranes or
in the extracellular matrices might show hemagglu-
tinating activities after artificial secondary
polymerization, and be accepted as lectins. Then it
is questionable how many “natural” proteins are
Classified as lectins under the application of the
Goldstein definition.

Considering the appearance of new _ sugar-
binding proteins, Barondes [3] advocated that con-
fining the term lectin to bivalent proteins because

Functions of Animal Lectins 589

of their historical identification as agglutinins
seems to have lost its usefulness, and he proposed
changing the definition of a lectin to the less
restrictive “a carbohydrate-binding protein other
than an enzyme or an antibody”. Barondes’ prop-
osal has the merit of including all the sugar-binding
proteins in a single category of monomer
polypeptides. Namely, a lectin at a monomer level
has at least two binding sites: one is for binding
with sugars and the other is for binding with
sugars, proteins, RNA or phospholipids, or it is
merely a transmembrane segment. This new de-
finition does not contradict the origin of the term
lectin, that is, the Latin /egere meaning to pick up
or choose [1], and it may be useful in totally
understanding the molecular evolution of carbohy-
drate-binding proteins.

REFERENCES

1 Boyd, W. C. and Shapleigh, E. (1954) Science,
119: 419.

2 Goldstein, I. R., Hughes, R. C., Monsigny, M.,
Osawa, T. and Sharon, N. (1980) Nature, 285: 66.

3 Barondes, S. H. (1988) Tren. Biochem. Sci., 13:

480-482.

4 Lis, H. and Sharon, N.

Biochem., 42: 541-574.

5 Barondes, S. H. (1981) Ann. Rev. Biochem., 50:

207-231.

6 Ratcliffe, N. A., Rowley, A. F., Fitzgerald, S. W.
and Rhodes, C. P. (1985) Intern. Rev. Cytol., 97:
183-350.

Balds, B. A., Sawyer, W. H., Stick, R. V. and
Uhlenbruck, G. (1978) Biochem. J., 175: 467-477.

Basu, S., Sarkar, M. and Mandal, C. (1986)
Molec. Cell Biochem., 71: 149-157.

Ravindranath, M. M., Higa, H. H., Cooper, E. L.
and Paulson, J. C. (1985) J. Biol. Chem., 260:
8850-8856.

10 Roche, A. C., Schauer, R. and Monsigny, M.
(1975) FEBS Letters, 57: 245-249.

11 Richards, E. H., Ratcliffe, N. A. and Renwrantz,
L. (1988) Insect Biochem., 18: 691-700.

12 Umetsu, K., Kosaka, S. and Suzuki, T. (1984) J.
Biochem., 95: 239-245.

13. Pendland, J. C. and Doucias, D. G. (1986) De-
velop. Comp. Immu., 10: 477-487.

14 Komano, H., Mizuno, D. and Natori, S. (1980) J.
Biol. Chem., 255: 2919-2924.

15 Castro, V. M., Boman, H. G. and Hammarstrém,
S. (1987) Insect Biochem., 17: 513-523.

(1973) Annes Rev:

~—

oOo

Ne)

16 Qu, X. M., Zhang, C. F., Komano, H. and Natori,
S. (1987) J. Biochem., 101: 545-551.

17 Stebbins, M. R. and Hapner, K. D. (1985) Insect
Biochem., 15: 451-462.

18 Suzuki, T. and Natori, S. (1983) J. Biochem., 93:
583-590.

19 Kubo, T. and Natori, S. (1987) Eur. J. Biochem.,
168: 75-82.

20 Giga, Y., Satoh, K. and Ikai, A. (1985) Bioche-
mistry, 24: 4461-4467.

21 Vasta, G. R., Hunt, J. C., Marchalouis, J. J. and
Fish, W. W. (1986) J. Biol. Chem., 261: 9174-9181.

22 Renwrantz, L. and Stahmer, A. (1983) J. Comp.
Physiol. B., 149: 535-546.

23 Coombe, D. R., Ey, P. L. and Jenkin, C. R. (1984)
J. Comp. Physiol. B., 154: 509-521.

24 Roberson, M. M., Wolffe, A. P., Tata, J. R. and
Barondes, S. H. (1985) J. Biol. Chem., 260: 11027-
11032.

25 Wang, K. Y., Kuhlenschimdt, T. B. and Lee, Y. C.
(1985) Biochemistry, 24: 5932-5938.

26 Wang, K. Y., Hoppe, C. A., Dotta, P. K., Fogel-
strom, A. and Lee, Y. C. (1986) Proc. Natl. Acad.
Sci. USA, 83: 9670-9674.

27 Ikeda, K., Sannoh, T, Kawasaki, N., Kawasaki, T.
and Yamashina, I. (1987) J. Biol. Chem., 262:
7451-7454.

28 Hamazaki, H. (1986) J. Biol. Chem., 261: 5455-
5459.

29 Drickamer, K., Dordal, M. S. and Reynolds, L.
(1986) J. Biol. Chem., 261: 6878-6887.

30 Ti, M., Wada, M., Kawasaki, T. and Yamashina, I.
(1988) J. Biochem., 104: 587-590.

31 Oda, S., Sato, M., Toyoshima, S. and Osawa, T.
(1988) J. Biochem., 104: 600-605.

32 Haltiwanger, R. S. and Hill, R. L. (1986) J. Biol.
Chem., 261: 7440-7444.

33 Haltiwanger, R. S. and Hill, R. L. (1986) J. Biol.
Chem., 261: 15696-15702.

34 Lennartz, M. R., Wileman, T. E. and Stahl, P. D.
(1987) Biochem. J., 245: 705-711.

35 Sharon, N. (1984) Immunol. Today, 5: 143-147.

36 Stahl, P. D., Rodman, J. S., Miller, M. J. and

Schlesinger, P. H. (1978) Proc. Natl. Acad. Sci.

USA, 75: 1399-1403.

Ashwell, G. and Harford, J. (1982) Ann. Rev.

Biochem., 51: 531-554.

Spiro, G. (1973) Adv. Protein Chem., 27: 349-

467.

Stowell, C.P. and Lee, Y. C. (1980) Adv. Car-

bohy. Chem. Biochem., 37: 225-281.

40 Steer, C. P. and Clarenburg, R. (1979) J. Biol.

Chem., 254: 4457-4461.

Winkelhake, J. L. and Nicolson, G. L. (1976) J.

Biol. Chem., 251: 1074-1080.

3

~

3

(oe)

3

\o

4

—

590

42

43

A

45

46

47

48

49

50

51

52

33)

54

55

56

Sf

58

59

60

61

62

63

64

65

Hubbard, A. L., Wilson, G., Ashwell, G. and
Stukenbrok, H. (1979) J. Cell Biol., 83: 47-64.

Haimes, H. B., Stockert, R. T., Morell, A. G. and
Novikoff, A. B. (1981) Proc. Natl. Acad. Sci. USA,
78: 6936-6939.

Wall, D. A., Wilson, G. and Hubbard, A. L.
(1980) Cell, 21: 79-93.

Stockert, R. T., Morell, A. G. and Scheinberg, I.
H. (1976) Biochem., Biophys. Res. Comm., 68:
988-993.

Prieels, J. P., Pizzo, S. V., Glasgow, L. R., Paul-
son, J. C. and Hill, R. T. (1978) Proc. Natl. Acad.
Sci. USA, 75: 2215-2219.

Kawasaki, T., Etoh, R. and Yamashina, I. (1978)
Biochem. Biophys. Res. Comm., 81: 1018-1024.

Stockert, R. J., Gartner, U., Morell, A. G. and
Wolkoff, A. W. (1980) J. Biol. Chem., 255: 3830-
3831.

Kolb-Bachofen, V. (1981)
Acta, 645: 293-299.

Heuser, J. E. and Anderson, R. G. W. (1989) J.
Cell Biol., 108: 389-400.

Tanabe, T., Pricer, W. E. Jr. and Ashwell, G.
(1979) J. Biol. Chem., 254: 1038-1043.

Steer, C. J. and Ashwell, G. (1980) J. Biol.
Chem., 255: 3008-3013.

Kawasaki, T. and Ashwell, G. (1976) J. Biol.
Chem., 251: 1296-1302.

Kawasaki, T. and Ashwell, G. (1976) J. Biol.
Chem., 251: 5292-5299.

Steer, C. J., Kempner, E. S. and Ashwell, G.
(1981) J. Biol. Chem., 256: 5851-5856.

Halberg; D: F., Wager, Ro ES Farrell} D: €.,
Hildreth, J. W., Quesenberry, M. S., Loeb, J. A.,
Holland, E. C. and Drickamer, K. (1987) J. Biol.
Chem., 262: 9828-9838.

Baenziger, J. U. and Maynard, Y. (1980) J. Biol.
Chem., 255: 4607-4613.

Kawasaki, T. and Ashwell, G. (1977) J. Biol.
Chem., 252: 6536-6543.

Oka, S., Kawasaki, T. and Yamashina, I. (1985)
Archiv. Biochem. Biophys., 241: 95-105.

Lehrman, M. A. and Hill, R. L. (1986) J. Biol.
Chem., 261: 7419-7425.

Oka, S., Itoh, N., Kawasaki, T. and Yamashina, I.
(1987) J. Biochem., 101: 135-144.

Ezekowitz, R. A. B., Day, L. E. and Herman, G.
A. (1988) J. Exp. Med., 167: 1034-1046.

Harford, J. and Ashwell, G. G. (1981) Proc. Natl.
Acad. Sci. USA, 78: 1557-1561.

Maynard, Y. and Baenziger, J. U. (1982) J. Biol.
Chem., 257: 3788-3794.

Lobel, P., Dahms, N. M., Breitmeyer, J., Chirg-
win, J. M. and Kornfeld, S. (1987) Proc. Natl.
Acad. Sci. USA, 84: 2233-2237.

Biochim. Biophys.

N. YOSHIZAKI

66

67

68

69

70

et

72

73

74

is

76

77

78

79

80

81

82

83

89

90

Nowak, T. P., Kobiler, D., Roel, L. E. and
Barondes, S. H. (1977) J. Biol. Chem., 252: 6026-
6030.

Den, H., Malinzak, D. A. and Rosenberg, A.
(1976) Biochem. Biophys. Res. Comm., 69: 621-
627. |

Nowak, T. P., Haywood, P. L. and Barondes, S.
H. (1976) Biochem. Biophys. Res. Comm., 68:
650-657.

Gartner, K. and Podleski, T. R. (1975) Biochem.
Biophys. Res. Comm., 67: 972-978.

Den, H. and Malinzak, D. A. (1977) J. Biol.
Chem., 252: 5444-5448.

Barondes, S. H. and Haywood-Reid, P. L. (1981)
J. Cell Biol., 91: 568-572.

Joubert, R., Caron, M. and Bladier, D. (1988)
Cell. Molec. Biol., 34: 79-87.

Caron, M., Joubert, R. and Bladier, D. (1987)
Biochim. Biophys. Acta, 925: 290-296.

Zanetta, J. P., Dontenwill, M., Meyer, A. and
Roussel, G. (1985) Develop. Brain Res., 17: 233-
243.

Powell, J. T. (1980) Biochem. J., 187: 123-129.

Powell, J. T. and Whitney, P. L. (1980) Biochem.
J., 188: 1-8.

Cerra, R. F., Haywood-Reid, P. L. and Barondes,
S. H. (1984) J. Cell Biol., 98: 1580-1589.

Wrenn, D. S., Hinek, A. and Mecham, R. P.
(1988) J. Biol. Chem., 263: 2280-2284.

Hinek, A., Wrenn, D. S., Mecham. R. P. and
Barondes, S. H. (1988) Science, 239: 1539-1541.

Ohyama, Y., Hirabayashi, J., Oda, Y., Ohno, S.,
Kawasaki, H., Suzuki, K. and Kasai, K. (1986)
Biochem. Biophys. Res. Comm., 134: 51-56.

Kawauchi, H., Sakakibara, F. and Watanabe, K.
(1975) Experientia, 31: 364-365.

Sakakibara, F., Takayanagi, G., Kawauchi, H.,
Watanabe, K. and Hakomori, S. (1976) Biochem.
Biophys. Acta, 444: 386-395.

Sakakibara, F., Kawauchi, H., Takayanagi G. and
Ise, H. (1979) Cancer Res., 39: 1347-1352.

Roberson, M. M. and Armstrong, P. B. (1980)
Proc. Natl. Acad. Sci. USA, 77: 3460-3463.

Roberson, M. M. and Barondes, S. H. (1982) J.
Biol. Chem., 257: 7520-7524. |

Roberson, M. M. and Barondes, S. H. (1983) J.
Cell Biol., 97: 1875-1881.

Harris, H. and Zalik, S. E. (1985) J. Cell Sci., 79:
105-117.

Outenreath, R. L., Roberson, M. M. and
Barondes, S. H. (1988) Develop. Biol., 125: 187-
194.

Yoshizaki, N. (1990) Develop. Growth Differ., 32:
343-352.

Yoshizaki, N. (1986) Develop. Growth Differ., 28:

ot
o2

93

94

95

96

97

98
99

100

101

102

103

104

105

106

107

108

109

Functions of Animal Lectins

275-283.

Yoshizaki, N. (1989) Zool. Sci., 6: 507-514.

Chamow, S. M. and Hedrick, J. L. (1986) FEBS
Letters, 206: 353-357.

Nishihara, T., Wyrick, R. E., Working, P. K.,
Chen, Y. H. and Hedrick, J. L. (1986) Biochemis-
try, 25: 6013-6020.

Greve, L. C. and Hedrick, J. L. (1978) Gamete
Res., 1: 13-18.

Yoshizaki, N. (1989) Develop. Growth Differ. , 31:
325-330.

Yoshizaki, N. and Katagiri, C. (1984) Zool. Sci.,
1: 255-264.

Yoshizaki, N. (1984) Develop. Growth Differ., 26:
191-195.

Yoshizaki, N. (1985) J. Morph., 184: 155-169.

Grey, R. D., Wolf, D. P. and Hedrick, J. L. (1974)
Develop. Biol., 36: 44-61.

Wyrick, R. E., Nishihara, T. and Hedrick, J. L.
(1974) Proc. Natl. Acad. Sci. USA, 71: 2067-2071.

Wolf, D. P. (1974) Develop. Biol., 36: 62-71.

Wolf, D. P. (1974) Develop. Biol., 38: 14-29.

Wolf, D. P., Nishihara, T., West, D. M., Wyrick,
R. E. and Hedrick, J. L. (1976) Biochemistry, 15:
3671-3678.

Grey, R. D., Working, P. K. and Hedrick, J. L.
(1976) Develop. Biol., 54: 52-60.

Colley, K. J., Beranek, M. C. and Baenziger, J. U.
(1988) Biochem. J., 256: 61-68.

Segler, K., Rahmann, H. and Rosner, H. (1978)
Biochem. System. Eco., 6: 87-93.

Paroutaud, P., Levi, G., Teichberg, V. I. and
Strosberg, A. D. (1987) Proc. Natl. Acad. Sci.
USA, 84: 6345-6348.

Teichberg, V. I., Shilman, I., Beitsch, D. D. and
Resheff, G. (1975) Proc. Natl. Acad. Sci. USA, 72:
1383-1387.

Clerch, L. B., Whitney, P., Hass, M., Brew, K.,
Miller, T., Werner, R. and Massaro, D. (1988)
Biochemistry, 27: 692-699.

110 Takahashi, H., Komano, H., Kawaguchi, N., Kita-

111

112

Js)

114

115

116

117

118

119

120

121

122

123

124

125

126

591

mura, N., Nakanishi, S. and Natori, S. (1985) J.
Biol. Chem., 260: 12228-12233.

Beyer, E. C. and Barondes, S. H. (1980) J.
Supramol. Struct., 13: 219-227.

Carding, S. R., Childs, R. A., Thorpe, R., Spitz,
M. and Feizi, T. (1985) Biochem. J., 228: 147-153.

Wilson, T. J. G., Firth, M. N., Powell, J. T. and
Harrison, F. L. (1989) Biochem. J., 261: 847-852.

Gitt, M. A. and Barondes, S. H. (1986) Proc. Natl.
Acad. Sci. USA, 83: 7603-7607.

Southan, C., Aitken, A., Childs, R. A., Abbott,
W. A. and Feizi, T. (1987) FEBS Letters, 214: 301-
304.

Roff, C. F. and Wang, J. L. (1983) J. Biol. Chem.,
258: 10657-10663.

Moutsatsos, I. K., Wade, M., Schindler, M. and
Wang, J. L. (1987) Proc. Natl. Acad. Sci. USA, 84:
6452-6456.

Laing, J. G. and Wang, J. L. (1988) Biochemistry,
27: 5329-5334.

Jia, S. and Wang, J. L. (1988) J. Biol. Chem., 263:
6009-6011.

Hubert, J., Seve, A. P., Facy, P. and Mousigny,
M. (1989) Cell Differ. Develop., 27: 69-81.

Benson, B., Hawgood, S., Schilling, J., Clements,
J., Damm, D., Cordell, B. and White, R. T. (1985)
Proc. Natl. Acad. Sci. USA, 82: 6379-6383.

Haagsman, H. P., Hawgood, S., Sargeant, T.,
Buckley, D., White, R. T., Drickamer, K. and
Benson, B. J. (1987) J. Biol. Chem., 262: 13877-
13880.

Gallatin, W. M., Weissman, I. L. and Butcher, E.
C. (1983) Nature, 304: 30-34.

Siegelman, M. H., Van de Rijn, M. and Weissman,
I. L. (1989) Science, 243: 1165-1172.

Lasky, L. A., Singer, M. S., Yednock, T. A.,
Dowbenko, D., Fennie, C. F., Rodriguez, H.,
Nguyen, T., Stachel, S. and Rosen, S. D. (1989)
Cell, 56:1045-1055.

Beyer, E. C., Zweig, S. E. and Barondes, S. H.
(1980) J. Biol. Chem., 255: 4236-4239.

an Mt a eas i filet eRait
fie cred OOM NEL LPs
oe qhtre a aii wees RRR Dag ote ee
, rasan hi faa
Aes : Na dee £2, BEE: Eiht ke: §
‘ anes ahh ws 4 aii ; gh (aan, panei RN,.

ae =i a. pe MS ibe condotaag MILs Hy oa
Me eke ae ee ee or

pes ie 2) $5 a se ae ; : ;
= vay iol ACRES. eg fred Ose ge
by, Sh sc HEN ve! aa ie ie pet — hia sere Dede lS

en ye tobi MM dati.
Mc Get se a vr

2
#2

: #
x
i leg seh seat’ (OR SS
rr hh ae ~ aE: 5 si
nis"
x pina i ahs : wt
‘ : a
Be (OMe b) in
: i
a bey
ae = | =
at a
%
pi
2S
A

ZOOLOGICAL SCIENCE 7: 593-603 (1990)

Putative Neurotransmitter in the Ocellar Neuropil
of American Cockroaches

Jin Tun Lin!, YosHruiro ToH, Makoto MIZUNAMI

and HIDEKI TATEDA

Department of Biology, Faculty of Science, Kyushu University,
Fukuoka 812, Japan

ABSTRACT — It has been reported in the cockroach Periplaneta americana that there are at least two
types of synapses upon large afferent neurons (L-neurons) within the dorsal ocellus; photoexcitation of
retinular axons hyperploarize L-neurons, whereas efferent spikes depolarize them. In the present study,
effects of some chemicals, widely known as neurotransmitters upon L-neurons, have been examined to
identify the neurotransmitter in the two types of synapses. Among various chemicals tested, histamine,
applied both by perfusion and by ionophoresis through a micropipette, hyperpolarized L-neurons.
Several histamine antagonists depressed photoresonses of L-neurons. Persistence of histamine-induced
hyperpolarization during blockade of synaptic transmission was suggestive of histamine receptors on the
L-neuron. Gamma-aminobutyric acid (GABA) depolarized L-neurons by ionophoretical application.
Depolarization of the L-neuron caused by efferent spikes was depressed by perfusion of picrotoxin.
Depolarization evoked by GABA application even after blockade of synaptic transmission was
suggestive of GABA receptors on the L-neurons. These data suggest that histamine and GABA may be
released from retinular axons and efferent neurons, respectively, at their synapses upon L-neurons.

© 1990 Zoological Society of Japan

INTRODUCTION

The neural mechanism of sensory organs has
been studied using two major tenchiques: elec-
trophysiology and morphology. The machinery of
the neural circuitry obtained by these tranditional
studies has been further verified by pharmacologic-
al studies aimed at identifying the neurotransmit-
ter substances working in the circuitry. The trend
of such a neurobiological strategy can be clearly
shown when one refers to the research history of
the vertebrate retina [1].

The dorsal ocellus of the insect is one of the
photoreceptor organs whose morphology and phy-
siology are well documented in a variety of insects
[2]. However, there have been few pharmacolo-
gical and histochemical accounts about the ocellar
neuropil (see DISCUSSION). Recently, some
evidence that supports histamine as a possible

Accepted October 24, 1989
Received July 5, 1989

" On leaving Department of Biology, National Taiwan
Normal University, Taipei 118, Taiwan, The Republic
of China

neurotransmitter from the retinular axon to thick
second order neurons (L-neurons) has been pre-
sented on the ocellar neuropils of the locust [3] and
the fly [4]. A GABAergic nature and/or cho-
linergic nature has been also reported in the ocel-
lar neuropils of the dragonfly [5, 6], the honeybee
[7] and the locust [3, 8]. Although different
synapses occur among more than two types of
interneurons and retinular axons in the ocellar
neuropil, which synapses involve the proposed
transmitter has not always been identified in these
previous study. For further understanding of the
neural mechanism of the ocellar neuropil, a trans-
mitter substance should be identified for each
synapse type.

In the ocellar neuropil of the American cock-
roach Periplaneta americana, afferent synapses
and efferent synapses upon L-neurons can be
idenpendently driven by different sensory stimuli
[9]. It was assumed in our preliminary report that
histamine and y-aminobutyric acid (GABA) are
involved in neurotransmission at afferent and effe-
rent synapses upon L-neurons in the cockroach
ocellus [10]. This assumption has been further

594 J. T. Lin, Y. Tou et al.

examined using more pharmacological techniques
in the present study, and the data obtained will be
discussed in comparison with those obtained in
other insects.

MATERIALS AND METHODS

Preparation

Adult male cockroaches Periplaneta americana
reared in the laboratory were used. The animal
was mounted, dorsal side up, in a lucite chamber
and its head was fixed with bees wax. The dorsal
surface of the head capsule was partially removed.
The compound eyes and esophagus were excised.
The ocellar nerve was exposed and immersed in
1% protease dissolved in a physiological saline [11]
for 1 min to soften connective tissues around the
ocellus and the ocellar nerve. The chamber was
about 0.4 ml in volume, and the saline solution was
continously perfused at a rate of 1 ml per min.

Recording

Responses of L-neurons were intracellularly re-
corded by glass capillary microelectrodes filled
with 2M potassium acetate solution with resist-
ances of 50-70 MQ. Electrical signals picked up by
the electrode were DC-amplified, displayed on an
oscilloscope and recorded on a pen recorder. A
small piece of platinum was bathed in the saline
solution as an indifferent electrode. Stable in-
tracellular recordings usually continued for more
than 1 hr. Spike discharges of ocellar efferent
neurons were extracellularly monitored by suction
electrodes attached to the ocellar nerve.

Stimulations

For ocellar illumination a light guide of 0.5 mm
was attached to the ocellar cornea, and a light
beam derived from a tungsten lamp was delivered
to the ocelluls through the light guide. The light
intensity was controlled by calibrated neutral-
density filters.

To drive efferent spikes, cercal mechanorecep-
tors were stimulated by air puffs. An air derived
from a compressor was delivered through silicon
tubes toward the cerci. An outlet of the tube (2
mm in diameter) was set 10mm apart from the

cerci. The flow rate of the puff was 4 m/sec.

Perfusion

Effects of the following chemicals (Sigma, unless
noted otherwise) were tested by perfusion experi-
ments: histamine di-hydrochloride, acetylcholine,
y-aminobutyric acid, cimetidine hydrochloride
(Aldrich), cyclizine, d-chlorpheniramine, mepyr-
amine maleate (May & Baker), promethazine hyd-
rochloride, atropine sulphate, picrotoxin, bicucul-
line, a-bungarotoxin, bicuculline methiodide, car-
bachol, eserine, neostigmine, curare, ranitidine
(Glaxo), dimaprit (Fujisawa), burimamide (SK &
F), 2-pyridylethylamine dihydrochloride (SK & F)
and impromidine (SK & F). Each test chemical
was dissolved in the physiological saline (test solu-
tion).

The inlet of the specimen chamber was con-
nected to three reservior bottles; one for the
normal saline (control solution) and others for test
solutions. The specimen could be alternately per-
fused by the control solution and the test solution
by manipulating a flow exchanger. After the
impalement of a neuron and the establishment ofa .
steady recording in the control solution, the ex-
changer was switched to introduce a test solution
into the chamber. After testing the effects of a
chemical, the control solution was again perfused
to test the reversibility of the chemical effects.

In order to see the effects of synaptic blockade
upon sensory stimulation or application of chemic-
als, Ca**-free and Co*t-containing saline were
perfused either alone or together with test chemic-
als. Intermittent flashes (0.2—0.4 Hz) were usually
superimposed on the perfusion experiments to
monitor photoresponses of L-neurons. All test
solutions were made up from stock solutions on
the day of experiments. Experiments were carried
out at room temperature (20-22°C).

Ionophoresis

Some chemicals were applied ionophoretically
through a triple-barrelled micropipette. Of the
three barrels, two were filled with test chemicals.
The test chemicals (conc. in M, and pH) were
GABA (1.0, 3.5), acetylcholine (1.0, 5.0), hista-
mine (0.5, 5.0), L-sodium glutamate (0.2, 8.0) and
L-sodium aspartate (0.2, 8.0). The third was filled

Neurotransmitter in Cockroach Ocellus 595

with 0.2 M NaCl and used for active current com-
pensation or control injections. The triple-
barrelled micropipette was positioned within the
ocellar cup. A pulse of drug was ejected by an
ionophoretic current. Control ionophoresis of 0.2
M NaCl or saline solution did not elicit any re-
sponses.

RESULTS

An outline of neural connections manipulated in the
present study in the cockroach ocellus

The cockroach ocellus contains more than
10,000 retinular cells, whose axons synaptically
converge upon three to five (usually four) thick
second order neurons (L-neurons) in the posterior
region of the ocellus [11-13]. Afferent synapses
between retinular axons and L-neurons are ribbon

C

Fia. 1.

a a

synapses [14]. Morphologically identified efferent
neurons also synapse with L-neurons, and the
synapses are conventional ones lacking a synaptic
ribbon [14].

Depolarization of the retinular cells by ocellar
illumination hyperpolarized L-neurons. Each
hyperpolarization often accompanied with a few
spikes at the cessation of the illumination (Fig.
1A) [15]. Spike discharges of physiologically iden-
tified efferent neurons, referred to as “SM-
neurons”, depolarized L-neurons (Fig. 6) [9]. The
SM-neurons received various sensory inputs, of
which cercal mechanoreceptors most effectively
elicited a train of spikes in L-neurons. Recently,
intracellular staining by Lucifer yellow suggested
that the SM-neuron is identical to morphologically
identified efferent neurons (Ohyama and Toh,
unpubl. data). Since afferent synapses and effe-
rent synapses hyperpolarized and depolarized L-

Saw eis

Effects of histamine upon membrane potentials and photoresponses of L-neurons. A: Histamine (0.1 mM) is

perfused in the period indicated by a horizontal bar (upper trace). In the lower traces, photoresponses before (a)
and during (b, c) perfusion of histamine, and during washing (d-f) are sequentially shown at an enlarged time
scale. Calibrations: 300 msec for the lower trace and 20 mV for both traces. B: Responses of an L-neuron to
ionophoretically applied histamine by 10 nA for 9 sec (horizontal bar). Calibrations: 5sec and 5mV. In all
figures except Fig. 6, rapid downward deflections are hyperpolarization by intermittent flashes (300 msec in

duration and 0.2—0.4 Hz) to the ocellus.

596

neurons, respectively, different transmitter subst-
ances must be implicated in the two synapses.
Effects of several chemicals, widely known as
possible neurotransmitter substances in various
nervous systems, and their agonists and antagon-
ists upon L-neurons are compared with respones of
L-neurons to afferent and efferent stimulations.

Effects of chemicals, applied by perfusion or
ionophoresis upon L-neurons, are summarized in
Table 1. Since, of the chemicals tested, histamine
and aspartate hyperpolarized L-neurons as the
ocellar illumination did, they might be involved in
neurotransmission at afferent synapses on the L-
neurons. Since acetylcholine and y-aminobutyric
acid (GABA) depolarized L-neurons as the effe-
rent activity did, they might be involved in neurot-
ransmission at the efferent synapses. Thus, effects
of these chemicals were examined in detail.

Hyperpolarizing chemicals: histamine, aspartate
and their agonists and antagonists

Responses of L-neurons to histamine and ocellar
illumination Perfusion through the ocellar
neuropil by saline containing 0.1 mM _ histamine

A

qi" Metra

Jere Dn ¥ <Touvenal:

hyperpolarized the L-neuron. Amplitude of the
hyperpolarization reached about 30mV. The
hyperpolarization gradually reduced when the his-
tamine-containing perfusate was replaced by the
normal saline (Fig. 1A).

When ocellar illumination was presented during
the hyperpolarizing process by perfusion of hista-
mine, photoresponses of L-neurons reduced de-
pending upon the level of hyperpolarization by
histamine. When the histamine-induced hyperpo-
larization was large, the amplitude of the subse-
quent photoresponse was small. Moreover, mem-
brane noises of the L-neurons were reduced by
histamine application. Almost identical results
were obtained by ionophoretic application of hista-
mine toward the ocellar neuropil (Fig. 1B).

Effects of histamine antagonists upon photore-
sponses of L-neurons Since histamine mimicked
photoresponses of L-eurons, it was assumed that
histamine might be a putative neurotransmitter of
retinular axons. In order to test this assumption,
effects of several histamine antagonists upon neu-
ral transmission at afferent synapses were ex-
mained. Histamine receptors are classed into Hy,

i | |

aes Te TE ee i
25 min

TH

0 5 10

We

20 min

Fic. 2. Inhibitory effects of cimetidine on responses of L-neurons to light-induced responses (A) and histamine-
induced responses (B). A: Perfusion of 0.5mM cimetidine (thick horizontal bar) reversibly suppresses
photoresponses. B: The L-neuron is hyperpolarized by perfusion of 0.1 mM histamine (thin horizontal bar), and
additional perfusion of 0.5 mM cimetidine (thick horizontal bar) suppresses the effects of histamine. Calibration:
20 mV.

Neurotransmitter in Cockroach Ocellus

H, and H; in vertebrate nervous tissues [16, 17].
Four histamine antagonists for H, receptors (d-
chlorpheniramine, cyclizine, promethazine and
mepyramine), two for H, receptors (cimetidine
and ranitidine) and one for H3 (burimamide) were
used. Effects of these antagonists are summarized
in Table 1. Of the seven antagonists prometha-
zine, mepyramine, cimetidine and ranitidine re-
duced and abolished both photoresponses and
histamine-induced hyperpolarization of L-
neurons. Their inhibitory effects were recovered
after washing with the normal saline. Figure 2
shows such inhibitory effects of cimetidine upon
photoresponses (A) and histamine-induced hyper-
polarization (B) of L-neurons. These chemicals
influenced little the resting membrane potential of
L-neurons. D-chlorpheniramine and burimamide
had no effects upon photoresponses of the L-
neurons. Cyclizine behaved like an agonist in the
present study: 0.5mM cyclizine in the normal
saline hyperpolarized L-neurons and reduced
photoresponses of L-neurons as histamine did.
Effects of histamine agonist upon L-neuron 2-
pyridylethylamine dihydrochloride, impromidine

A

apni

oo

97,

and dimaprit are known as histamine agonists in
vertebrate nervous tissues [16, 17]. Perfusion of
0.1mM_ 2-pyridylethylamine dihydrochloride
hyperpolarized L-neurons and reduced superim-
posed photoresponses as histamine did (Table 1).
However, perfusion of 0.5 mM dimaprit and 0.1
mM impromidine had no effects upon the resting
membrane potential and the photoresponses of
L-neurons.
Direct
rons

effects of histamine upon L-neu-
Data presented hitherto favors the assump-
tion that histamine may be a putative neurotrans-
mitter from retinular axons upon L-neurons,
though a few pharmacological data do not agree
with it. However, there is a doubt that histamine
does not directly affect L-neurons, but does so
indirectly by the mediation of histamine sensitive
interneurons between retinular axons and L-
neurons. To answer this question direct effects of
histamine upon L-neurons were examined in pre-
paration where synaptic transmission was blocked.
Preparations were perfused by Ca**-free and 0.1
mM CoCl,-containing saline. Photoresponses of
the L-neuron gradually reduced in this saline. The

al f

10

Fic. 3.

15 20 min

age

Effects of blockade of synaptic transmission upon responses of L-neurons to ocellar flashes and application of
histamine. A: Photoresponses of an L-neuron to intermittent flashes reduce in the Co**

-containing and Ca*t

-free saline (thick horizontal bar), whereas additional perfusion of 0.1 mM histamine (thin horizontal bar)
hyperpolatizes the L-neurons. Calibration: 20 mV. B: Responses of an L-neuron to ionophoretic application of

histamine in normal saline (left) and in Co?*
mV.

-containing and Ca*t

-free saline (right). Calibrations: 5 sec and 5

598 S22 sbin; YY: Tontietxal.

10 15.10

;
ugg ni ' |

Fic. 4.

Inhibitory effects of curare upon photoresponses and histamine-induced responses of L-neurons. A:

Photoresponses decrease and are abolished by perfusion of 0.05 mM curare (thick horizontal bar). Additional °
perfusion of histamine (thin horizontal bar) in the curare-containing saline does not produce hyperpolarization
(upper trace). In lower trace, curare (thick horizontal bar) reversibly suppresses the response of L-neuron to
perfusion of histamine (thin horizontal bar) (lower trace). Calibration: 20mV. B: Hyperpolarization of an
L-neuron to light and histamine (horizontal bars) in the normal saline (left) and after 20 min perfusion by 0.05
mM curare-containing saline (right). Applied currents are 5 nA, durations are 1 sec and 5 sec. Calibrations: 10

sec and 5mV.

addition of histamine to this perfusate hyperpola-
rized L-neurons (Fig. 3A). Similar results were
obtained by ionophoretic application of histamine
(Fig. 3B). These results suggest that the L-neuron
has histamine receptors.

Effects of cholinergic inhibitors upon photore-
sponses and histamine-induced responses of L-
neurons Hyperpolarizations of L-neurons caused
either by ocellar illumination or by application of
histamine were suppressed by 0.05mM curare
(Fig. 4). Atropine, a muscarinic antagonist, also
inhibited photoresponses of the L-neurons, but
a-bungarotoxin, a strong antagonist of acetylcho-
line, had no effects upon them (Table 1).

Responses of L-neurons to aspartate Ion-
ophoretic application of aspartate towards the
ocellar neuropil hyperpolarized L-neurons. The

hyperpolarization was small being less than 3 mV.
Amplitudes of hyperpolarizations to histamine and
aspartate were compared by alternate ejection of
the two chemicals through a triple-barrelled micro-
pipette toward the ocellar neuropil. Responses of
L-neurons to aspartate were always less than those
to histamine (Fig. 5).

acid

Depolarizing chemicals: y-aminobutyric

(GABA) and acetylcholine

Responses of L-neurons to cercal stimulation and
GABA _ The L-neuron depolarized in response to
cercal stimulation (Fig. 6). The amplitude of the
depolarization was about 5 mV. Ionophoretic ap-
plication of GABA to the ocellar neuropil also
depolarized L-neurons (Fig. 7a, c). The amplitude
of the depolarization by GABA was about 2-5

Neurotransmitter in Cockroach Ocellus

TABLE 1. Summary of effects of several transmitter candidates and their agonists and antagonists upon
L-neurons by perfusion or ionophoresis

599

Chemicals Specificity (oN Bee r cea a
Histamine 0.01 hyper(*) reduced(*)
Aspartate hyper* reduced*
GABA depol* reduced*
Acetylcholine depol* reduced*
Glutamate —* —*
eis lethlamine H1+ 0.1 hyper reduced
ihydrochloride
Impromidine H2+ 0.5 — —
Dimaprit H2+ 0.5 — --
d-Chlorpheniramine H1— 0.1 — —
Cyclizine jell 0.5 hyper reduced
Promethazine Hi— 0.5 — inhibited
Mepyramine H1— 0.1 — inhibited
Cimetidine H2— 0.5 — inhibited
Ranitidine H2— 0.5 — inhibited
Burimamide H3— 0.1 — -
Picrotoxin GABA— 1.0 — —
Bicuculline GABA— 0.1 — ——
Bicuculline GABA— 0.1 — —_—
methiodide
Eserine AChE— 0.1 — —
Neostigmine AChE— 1.0 — —
Carbachol AChM/N + 1.0 — —
a-Bungarotoxin AChN— 0.1 — —
Curare AChN— 0.05 — inhibited
Atropine AChM— 1.0 — inhibited
M: muscarinic; N: nicotinic; H1, H2, H3: vertebrate histamine classes. +: agonist; —: antagonist; —:

ineffective; *: by ionophoresis (*): by both perfusion and ionophoresis; AChE: acetylcholinesterase.

mV.

Effects of GABA antagonists upon L-neuron
responses to cercal stimulation Effects of picroto-

a

a

Ht 4

xin,

a well known GABA antagonist, upon L-

neuron responses to ocellar illumination and cercal
stimulation were tested. Photoresponses of L-

neurons were not affected by perfusion of 0.1 mM
picrotoxin-containing saline (Fig. 8A). Depolar-
ization of L-neurons caused by cercal stimulation
was suppressed by perfusion of 0.1 mM picrotoxin,

efferent spikes were little affected (Fig. 6).

5. Responses of L-neurons to hyperpolarizing che-
micals. Histamine (a) and aspartate (b, c) are
alternately ejected from two of a triple-barrelled
micropipette for comparison of their hyperpolariz-
ing effects. Injected currents (horizontal bars) are 5
nA in a and b and 10nA inc. Calibrations: 5 sec
and 5 mV.

600 J.T. Lin, Y. Tou et al.

Fic. 6. Simultaneous recordings of intracellular responses of an L-neuron (upper trace) and spike discharges of
ocellar efferent neurons by a suction electrode (lower trace) to cercal stimulation by an air puff (horizontal bars)
in the normal saline (A) and in picrotoxin-containing saline (B). The air puff was 4 m/sec in velocity and 0.5 sec in
duration. Calibrations: 1 sec and 5 mV for upper records and 100 «V for lower records.

a b

d

Fic. 7. Responses of L-neurons to depolarizing chemicals. GABA (a) and acetylcholine (b) are alternately ejected

from two of a triple-barrelled micropipette for comparison of their depolarizing effects.

Responses of an

L-neuron to ionophoretic application of GABA in normal saline (c) and in Co’ *-containing and Ca’*-free saline
(d). Injected currents (horizontal bars) are 5SnA. Calibrations: 5 sec and 5 mV.

Another GABA antagonist, bicuculline or bicucul-
line methiodide, had no effects upon L-neuron
responses to cercal stimulation, but observed
effects of picrotoxin suggest that the efferent
neuron may release GABA and the L-neuron may
possess GABA receptors. Validity of this assump-
tion was further tested.

Direct effects of GABA upon L-neurons To
test the GABA sensitivity of L-neurons, synaptic
transmission was blocked by perfusion of Ca?*
-free and 0.1mM CoCl,-containing saline, and

responses of L-neurons to ionophoretic application
of GABA were examined. Depolarizing responses
to GABA were unchanged in the saline, where
photoresponses were blocked (Fig. 7c, d). The
result suggests that the L-neuron has GABA re-
ceptors.

Responses of L-neurons to acetylcholine Ion-
ophoretic application of the acetylcholine into the
ocellar neuropil depolarized L-neurons (Fig. 7b).
The depolarization was less than 2 mV, and could
be seen in two of eight preparations.

Neurotransmitter in Cockroach Ocellus 601

A

in

| \
npn! Wamaninns

re ere

B

==

Fic. 8. Effects of picrotoxin (A) and curare (B) upon responses of L-neurons to intermittent flashes and an air puff to
cerci (horizontal bars). The upper traces are responses in the normal saline. The lower traces are responses to
perfusion of 0.1 mM picrotoxin for 40 minutes (A), and to perfusion of 0.05 mM curare for 15 minutes (B). Light
intensity is lowered by 6 log units in these recordings. Calibrations: 10 sec and 5 mV.

Effects of acetylcholine antagonists upon re-
sponses of L-neurons to cercal stimulation If
acetylcholine worked in the efferent synapses,
depolarization of the L-neuron by cercal stimula-
tion would be depressed by acetylcholine antagon-
ists. However, depolarization of L-neurons caused
by cercal stimulation was not influenced by perfu-
sion of curare containing saline (Fig. 8B).

DISCUSSION

Three substances, acetylcholine, GABA and
histamine, have been so far reported as possible
neurotransmitters involved in dorsal ocelli of in-
sects. It was supposed that afferent synapses from
retinular cells to L-neurons might be cholinergic,
and feedback synapses from L-neurons to retinular
cells might be GABAergic in the dragonfly ocellus
[S, 6]. In the honeybee ocellus, acetylcholinester-
ase was histochemically detected in retinular cells
and glial cells [7]. Different data were presented
for the locust ocelli. Because GABA activities
were immunohistochemically detected in the small
ocellar interneurons (S-neurons), but not in L-
neurons, GABA was considered as a transmitter
candidate from S-neurons to retinular axons [8]. A
recent electrophosiological study in the locust ocel-

lus strongly suggested that histamine might be a
putative neurotransmitter substance released from
retinular axons [3].

Although histamine and aspartate hyperpola-
rized L-neurons in the present study, histamine
appears to be a major neurotransmitter substance
of retinular axons in the cockroach ocellus, be-
cause of the following facts: 1) Both ionophoretic
application and superfusion of histamine produced
graded hyperpolarization in the L-enuron, ampli-
tudes of which are as large as that caused by ocellar
illumination. 2) Most of histamine antagonists
effectively depressed or inhibited light-induced
hyperpolarizations of L-neurons, and some hista-
mine agonists hyperpolarized L-neurons. 3)
Hyperpolarizing responses to histamine after the
blockade of synaptic transmissions suggests that
histamine receptors occur in the L-neuron. Based
upon these data, histamine is assumed to be a
putative neurotransmitter substance of retinular
axons in the cockroach ocellus.

This assumption is also supported by the follow-
ing accounts: 1) Almost identical data, histamine
as a neurotransmitter of retinular axons, have been
reported for the locust ocellus [3] as well as in the
dipteran ocellus [4] and compound eye [4, 18]. 2)
It has been reported that the cockroach ocellar

602 J. Bsbin;) Y¥: Towverial.

retina contained histamine at a high level [19].
However, some dat in the present study do not
agree with this. Curare, known as an antagonist of
acetylcholine, behaved like an antagonist of hista-
mine in the present study as it did in the locust
ocellus [3] and other nervous tissues [20-22]. Such
results may be due to a less specific action of
curare. Some histamine antagonists, d-chlorpheni-
ramine and burimamide, had no effects upon the
photoresponse of L-neurons, whereas one hista-
mine antagonist, cyclizine, behaved like an agon-
ist, and some histamine agonists, dimaprit and
impromidine, did not produce any responses in
L-neurons. Such ambiguity may be due to some
structural difference of histamine receptors be-
tween vertebrate nervous tissues and insect ner-
vous tissues.

Aspartate also hyperpolarized L-neurons in the
present study. But it is unlikely to be a major
neurotransmitter substance released from retinular
axons, because hyperpolarization caused by aspar-
tate was samll as compared with that caused by
histamine. Functional aspects of aspartate in the
ocellar neuropil will be discussed later.

In the present study, GABA and acetylcholine
depolarized L-neurons. It is assumed, based upon
the following data, that GABA may be released
from efferent ocellar neurons to L-reurons. 1)
Effects of GABA on the L-neuron were stable as
compared with those of acetylcholine. 2) Depolar-
izations of L-neurons caused by ionophoretic ap-
plication of GABA were comparable in amplitude
to thsoe caused by cercal stimulation, which is the
most powerful sensory input to the efferent
neurons. 3) Perfusion of picrotoxin thorugh the
ecellar neuropil blocked depolarization of L-
neurons by cercal stimulation, but that of curare
did not. 4) During blockade of synaptic transmis-
sion, depolarization of the L-neuron to GABA
suggests that GABA receptors occur in the L-
neuron. GABA receptors of vertebrate nervous
tissue are classified into two sub-types: GABA,
and GABAg receptors [23]. Among GABA,
antagonists tested in the present study, picrotoxin
inhibited the depolarizing responses of L-neurons
caused by efferent spikes, but bicuculline did not.
Similar results have been reported for GABA
receptors of motor neurons of the cockroach:

GABA-induced responses of motor neurons of the
cockroach were blocked by picrotoxin but not by
bicuculline [24]. Fine structural differences in
GABA receptors between vertebrate nervous tis-
sues and insect nervous tissues may be also prop-
osed.

In conclusion, the present results suggest that
the ribbon synapses from retinular axons to L-
neurons are mediated by histamine, and conven-
tional synapses from efferent neurons to L-
neurons are mediated by GABA.

Additional synapses, for example, feedback
synapses from L-neurons or other unidentified
neurons upon retinular axons and reciprocal
synapses between L-neurons, also occur in the
cockroach ocellus [14], though their occurring fre-
quencies are extremely low. The membrane
potential of the L-neuron may be affected directly
or indirectly through such minor synapses. If one
supposes that depolarization of a given L-neuron
depolarizes the adjacent L-neuron through recip-
rocal synapses in the cockroach ocellus as in the
locust ocellus [25], the neurotransmitter substance
involved in such synapses would depolarizes L- -
neurons. Acetylcholine may be included in such
reciprocal synapses. Both aspartate and acetylcho-
line are supposed to be implicated in such minor or
unknown synapses. The morpholigical and phy-
siological identification of other synapses together
with pharmacological identification of neurotrans-
mitter substances included there, must be the next
investigation for further understanding of the neu-
ral mechanism in the ocellar system.

ACKNOWLEDGMENTS

This research was supported by grants from the
National Science Council of The Republic of China and

the Ministry of Education, Science and Culture of Japan.
Gifts of drugs from Smith Kline & French, Glaxo and
Fujisawa Pharmaceutical Co., Ltd are gratefully acknow-
ledged.

REFERENCES

1 Dowling, J. E. (1979) Information processing by
local circuits: In “The Neurosciences”. Ed. by F. O.
Schmitt and F. G. Worden, The MIT Press, Cam-
bridge/London, pp. 163-181.

10

11

12

13

14

Neurotransmitter in Cockroach Ocellus

Goodman, L. J. (1981) Organisation and physiology
of the insect dorsal ocellar system. In “Vision in
Invertebrates”. Handbook of Sensory Physiology,
Vol. VII/6C, Ed. by H. Autrum, Springer, Berlin/
Heidelberg/New York, pp. 201-286.

Simmons, P. J. and Hardie, R. C. (1988) Evidence
that histamine is a neurotransmitter of photorecep-
tors in the locust ocellus. J. Exp. Biol., 138: 205-
ZA9:

Nassel, D. R., Holmqvist, M. H., Hardie, R. C.,
Hakanson, R. and Sundler, F. (1988) Histamine-
like immunoreactivity in photoreceptors of the com-
pound eyes and ocelli of the flies Calliphora
erythrocephala and Musca domestica. Cell Tissue
Res., 253: 639-646.

Klingman, A. and Chappell, R. L. (1978) Feedback
synaptic interaction in the dragonfly ocellar retina.
J. Gen. Physiol., 71: 157-175.

Stone, S. L. and Chappell, R. L. (1981) Synaptic
feedback onto photoreceptors in the ocellar retina.
Brain Res., 221: 374-381.

Kral, K. (1980) Acetylcholinesterase in the ocellus
of Apis mellifica. J. Insect Physiol., 26: 807-809.
Ammermuller, J. and Weiler, R. (1985) S-neurons
and not L-neurons are the source of GABAergic
action in the ocellar retina. J. Comp. Physiol. A,
157: 779-788.

Ohyama, T. and Toh, Y. (1986) Multimodality of
ocellar interneurones of the American cockroach. J.
Exp. Biol., 125: 405-409.

Lin, J. T., Mizunami, M., Toh, Y. and Tateda, H.
(1988) Histamine as a photoreceptor transmitter in
the cockroach (Periplaneta americana) ocellus.
Zool. Sci., 5: 1206. (Abstract)

Yamasaki, T. and Narahashi, T. (1959) The effects
of potassium and sodium ions on the resting and
action potentials of the cockroach giant axon. J.
Insect Physiol., 3: 146-158.

Cooter, R. (1975) Ocellus and ocellar nerves of
Periplaneta americana (Orthoptera: Dictyoptera).
Int. J. Insect Morphol. Embyol., 4: 273-288.
Weber, G. and Renner, M. (1976) The ocellus of
the cockroach Periplaneta americana (Blattariae).
Receptory area. Cell Tissue Res., 168: 209-222.

Toh, Y. and Sagara, H. (1984) Dorsal ocellar system

15

16

7)

18

19

20

21

22

23

24

US

603

of the American cockroach. I. Structure of the
ocellus and ocellar nerve. J. Ultrastruct. Res., 86:
119-134.

Mizunami, M., Yamashita, S. and Tateda, H. (1982)
Intracellular stainings of the large ocellar second
order neurons in the cockroach. J. Comp. Physiol.
A, 149: 215-219.

Prell, G. D. and Green, J. P. (1986) Histamine as a
neuroregulator. Annu. Rev. Neurosci., 9: 209-254.
Schwartz, J. C., Arrang, J. M., Garbarg, M. and
Korner, M. (1986) Properties and roles of the three
subclasses of histamine receptors in brain. J. Exp.
Biol., 124: 203-224.

Hardie, R. C. (1987) Is histamine a neurotransmit-
ter in insect photoreceptors? J. Comp. Physiol. A,
161: 201-213.

Elias, M. S. and Evans, P. D. (1983) Histamine in
the insect nervous system: distribution, synthesis
and metabolism. J. Neurochem., 41: 562-568.
Carpenter, D. O., Swann, J. W. and Yarowsky, P.
J. (1977) Effect of curare on responses to different
putative neurotransmitters in Aplysia neurons. J.
Neurobiol., 8: 119-132.

Claiborne, B. J. and Selverston, A. I. (1984) Hista-
mine as a neurotransmitter in the stomatogastric
nervous system of the spiny lobster. J. Neurosci., 4:
708-721.

Hardie, R. C. (1988) Effects of antagonists on
putative histamine receptors in the first visual neuro-
pile of the housefly Musca domestica. J. Exp. Biol.,
138: 221-241.

Bowery, N. G., Price, G. W., Hudson, A. L., Hill,
D. R., Wilkin, G. P. and Turnbull, M. J. (1984)
GABA receptor multiplicity. Visualization of diffe-
rent receptor types in the mammalian CNS.
Neuropharmacol., 23: 219-231.

Sattelle, D. B., Pinnock, R. D., Wafford, K. A. and

David, J. A. (1988) GABA receptors on the cell-
body membrane of an identified insect motor
neuron. Proc. R. Soc. Lond. B, 232: 443-456.
Simmons, P. J. (1982) Transmission mediated with
and without spikes at connexions between large
second-order neurons of locust ocelli. J. Comp.
Physiol. A, 147: 401-414.

tat ; ° veuiss ii y a,

he ns
a

) a ‘ee "ie wt
) MS Bee 2 a1 As hes pe:

i

aon = ns i jalan ua “ri Fo ' pee
an ’ os Geta =a lt tied

: ee
gon ¢-cet ty } ath cathe, at: a Ne i ne
2 [ hay Sash. aia : :
ites et fin . i aT oie “
ah cee at ‘ ected ; t ze
: ‘ eats se 1h tii ie ee .
Min May IS see A !
* Fepetale a8
: - {cepa ryatte ig sito

€ ye ats teas ‘trie

c e ;
: bet) ~ wey tp Na
= J 4 ‘ oy . j es ee be | hy
£: ae ‘ ; ¥ ba oa \
2 : | | |
i ey ‘ * ‘ ice athe yee ta
| 6
; 4
= ; A |
7 | |
; oa
; ¢
es \
A i
> i

ZOOLOGICAL SCIENCE 7: 605-612 (1990)

Aminergic, Cholinergic and Peptidergic Innervation of
Hepatic Portal Vein in the Anuran Amphibians

KIYOSHI OKAMOTO and TAKASUKE TAGAWA

Department of Anatomy, Fukuoka University School of Medicine,
Jonan-ku, Fukuoka 814-01, Japan

ABSTRACT—The innervation of the hepatic portal vein in four species of anuran amphibians, Rana
catesbeiana, Xenopus laevis, Rana nigromaculata, Bufo japonicus, was studied using catecholamine
fluorescence, acetylcholinesterase and immunohistochemical methods.

The hepatic portal vein of the anuran amphibians was well supplied with aminergic nerve fibers
(Amn) and they appeared equally developed in all regions of the hepatic portal vein in all of the animals
examined. Abundant Amn were observed in the gastrointestinal vein of X. /aevis. In contrast to Amn,
the hepatic portal vein of the amphibians had a sparse perivascular plexus of cholinergic nerve fibers
(Chn) and fewer Chn were present in the pre-abdominal vein than in the gastrointestinal vein. The
gastrointestinal vein in R. catesbeiana was innervated moderately by Chn. And no acetylcholinesterase
activity positive fiber was found in the pre-abdominal vein in X. /aevis and B. japonicus.

Substance P (SP)-, neurokinin A (NKA)- and calcitonin gene-related peptide (CGRP)-like im-
munoreactive (LI) fibers were observed in the hepatic portal vein in all of the animals examined. The
gastrointestinal vein in R. catesbeiana and X. laevis was especially richly innervated. Only a very few
vasoactive intestinal polypeptide (VIP)-LI fibers associated with the gastrointestinal vein were found in
R. catesbeiana. However, in other species, almost no VIP-LI fibers were observed. In the
gastrointestinal veins in all of the amphibians examined, large bundles of nerves, often parallel to the

© 1990 Zoological Society of Japan

course of the vein, were observed together with networks of thin nerve fibers.

INTRODUCTION

It has been thought that blood vessels in several
peripheral organs of mammals are innervated only
by sympathetic, noradrenaline-containing fibers
(Amn) and_ parasympathetic, acetylcholine-
containing fibers (Chn). In recent years, it has
become apparent that perivascular nerves in mam-
mals and submammals contain a variety of biologi-
cally active neuropeptides, as well as Amn and
Chn, which can be demonstrated by immunohis-
tochemical procedures [1-3]. The presence of
Amn, Chn and peptide-like immunoreactive (LI)
fibers in the hepatic portal vein of mammals has
also been reported [4-7]. However, no systematic
study of the peptide-LI fibers associated with the
submammalian hepatic portal vein is available.
The angio-architecture of the amphibian hepatic
portal vein is considerably different from that of

Accepted October 11, 1989
Received April 20, 1989

mammals or birds, that is, the pre-abdominal vein
is a tributary of the hepatic portal system. The
gastrointestinal vein mainly drains the stomach,
pancreas, intestine and spleen, while the pre-
abdominal vein mainly drains the hind limbs [8].

The purpose of the present study was to examine
the difference of the distribution of these Amn,
Chn and peptide-LI fibers between the gastrointes-
tinal vein and the pre-abdominal vein.

MATERIALS AND METHODS

Thirty Rana catesbeiana, 18 Xenopus laevis, 20
Rana nigromaculata and 30 Bufo japonicus were
used in this study.

For immunohistochemistry, the animals were
anesthetized with ethyl ether and perfused through
the pre-abdominal vein with ice-cold Ringer’s solu-
tion, followed by Zamboni’s fixative [9] at 4°C.
The hepatic portal veins were rapidly removed
from the abdominal cavity and post-fixed in the
same fixative for 24 hr at 4°C. Each of them were

606 K. OKAMOTO AND T. TAGAWA

TABLE 1.

Details of antibodies used

nnn nn ee ee UE UEyE ESSE SSE SUIS EE

First antibody Antibody code Host
SP FP03 Rabbit
NKA FK03 Rabbit
CGRP R.842701 Rabbit
VIP R.785202 Rabbit

Dilution Source Reference

1: 5000 Dr. H. Kamiya Takano et al. [11]
Fukuoka Univ.

1: 5000 Dr. H. Kamiya Takano et al. [11]
Fukuoka Univ.

1: 1200 Milab Uddman et al. [29]
Malmo, Sweden

1: 2560 Milab Matsuyama et al. [34]

Malmo, Sweden

then washed throughly for 30 min in three changes
of cold 0.1 M phosphate buffered saline (PBS, pH
7.2), and stored in PBS for 24 hr at 4°C. Immunos-
taining was performed according to the perox-
idase-antiperoxidase (PAP) technique [10]. The
hepatic portal veins were incubated in PBS con-
taining 1% HO» for 1 hr at room temperature,
washed in cold PBS containing 0.3% Triton X-100
(PBST) and transferred to the first antibodies
which were raised against SP, NKA, CGRP and
VIP (Details of the antibodies used in this study
are given in Table 1). The specificity of these
antibodies were checked by the absorption test.
Antibody to SP and NKA were previously char-
acterized by Kamiya’s group [11]. The materials
were incubated for 2 days at 4°C in antisera,
followed by washing in PBST. They were incu-
bated in goat anti-rabbit IgG (Cappel, PA) at a
dilution of 1: 200 for 2 hr at 4°C, washed in PBST,
and incubated in rabbit peroxidase-antiperoxidase
complex (Cappel, PA) at the same dilution for 2 hr
at 4°C (Table 1).

The materials were washed with PBS, stained
with 3,3’-diaminobenzidine (DAB, Sigma) for 10
min, and then stretched on glass slides. The
stretched tissues were dehydrated through graded
alcohols, cleaned with xylene, and mounted in
balsam beneath a coverslip.

For demonstration of the aminergic nerve fibers,
the formaldehyde fluorescence technique [12] were
adopted. For cholinergic nerve fibers, the whole
mount preparations fix in folmaldehyde were
maintained in Karnovsky’s medium without acety-
Ithiocholine iodide for 30 min at 4°C and then
incubated in the complete medium containing 2 x
10-*M iso-OMPA (tetraisopropylpyrophosphor-
amide) as a nonspecific cholinesterase inhibitor,

for 1 to Shr at 20°C [13]. Detailed procedures

have previously been presented by Tagawa et al.
[14].

RESULTS

Schematic illustrations of the angio-architecture
of the hepatic portal system in anuran amphibians
are presented in Figure 1. In the four amphibians
used in this study, the hepatic portal system con-
sisted of the pre-abdominal vein and gastrointes-
tinal vein. The gastrointestinal vein mainly
drained the stomach, pancreas, intestine and .
spleen. The pre-abdominal vein, which mainly
drained the hind limbs, also was a tributary of the
hepatic portal system. The intestinal and gastric
veins jointed with the pre-abdominal vein to form
a hepatic portal vein. The pre-abdominal vein
bifurcated into the right and left branches, and
entered into the hilus of liver. In X. laevis,
however, the intestinal vein entered into the right
branch and gastric vein the left branch because the
right and left hiluses of the liver were located apart
from each other [15].

The relative density of Amn, Chn and pep-
tidergic nerve fibers in the amphibian hepatic
portal vein is presented in Table 2. The hepatic
portal vein of the anuran amphibians was well
supplied with Amn, and fine networks of Amn
were equally developed in all regions of the hepa-
tic portal vein in all of the animals examined.
Especially, abundant Amn were observed around
the gastrointestinal vein in X. laevis (Fig. 2a, b).
In contrast to Amn, the hepatic portal vein of the
amphibians had a sparse perivascular plexus of
Chn. Fewer Chn were present in the pre-
abdominal vein than in the gastrointestinal vein.

Innervation of Amphibian Hepatic Vein

Bullfrog Clawed toad

Fic. 1.

607

Japanese toad

Leopard frog

Ventral view of the hepatic portal vein in four anuran amphibians. A: Pre-abdominal vein B: Gastric vein C:

Intestinal vein 1: Right lobe of liver 2: Left lobe of liver 3: Gallbladder 4: Spleen 5: Stomach 6: Small intestine 7:
Rectum In X. laevis, the right and left hiluses of the liver are located apart from each other.

TABLE 2. Density of aminergic, cholinergic and peptidergic nerve fibers of the portal vein in the four

amphibians
ps ; SP- NKA- CGRP- VIP-

amphibians ems aingIe ne Wage containing containing containing containing
Rana Ee toate aE =F al a ale

catesbeiana gastrointestinal eat almiate ae ae staal stale oF
Xenopus pre-abdominal alpate = Als AF ae aE

laevis '—gastrointestinal Seatac =P sect: sl =F SE
Rana -—pre-abdominal =P ar a6 = aa ot. =

nigromaculata ‘—gastrointestinal sleet oF at =P ate SE
Bufo -—pre-abdominal slaat = SE SE aE SE

Japonicus gastrointestinal aiate =f ar ala ate ar

The relative number of nerves was graded arbitrarily; — no nerve was observed, + few nerves, + moderate

number of nerves, ++ rich supply of nerves, +++ very rich supply of nerves.

The gastrointestinal vein in R. catesbeiana was
innervated moderately by Chn. No Chn were
observed in the pre-abdominal vein in X. Jaevis
and B. japonicus (Fig. 3a, b). Improvement of the
fluorescence by longer paraformaldehyde gas
treatment suggested that the predominant catecho-
lamine in the hepatic portal vein of the examined
anuran amphibians was also considered to be
adrenaline [16].

SP-LI fibers were observed in the hepatic portal
vein of all of the animals examined. The gastroin-

testinal vein in R. catesbeiana and X. laevis was
especially richly innervated. In R. catesbeiana, X.
laevis and B. japonicus, the gastrointestinal vein
had the innervation of SP-LI fibers denser than
that of the pre-abdominal vein (Fig. 4a, b). SP-,
NKA- and CGRP-LI fibers show a similar distribu-
tion pattern (Figs. 5a, 5b, 6a, 6b).

Innervation of fewer VIP-LI fibers in the hepatic
portal vein of the anuran amphibians was sparser
than that of SP-, NKA- and CGRP-LI fibers.
Moderate VIP-LI fibers associated with the gas-

608 K. OKAMOTO AND T. TAGAWA

Nill a
ast

: us of axsons

showing catecholamine fluorescence associated with the gastrointestinal vein. Bundles of fibers can also be seen

running along the vessels (arrow). (X66).

Fic. 3. Whole mount preparations for cholinesterase staining of B. japonicus hepatic portal vein. a: Pre-abdominal
vein Note absence of staining. Arrow head, melanocyte (x26) b: Gastrointestinal vein (35) Innervation of

Chn is fewer than that of Amn.

oor
ty

Fic. 4. Whole mount preparations of R. catesbeiana hepatic portal vein exposed to SP antiserum. a: Pre-abdominal
vein (111) b: Gastrointestinal vein SP-LI perivascular nerve fibers are seen in the vessel walls. Note bundles of

axsons (arrow) (X83).

trointestinal vein were found in R. catesbeiana
(Fig. 7a, b). In the cases of other species, few
VIP-LI fibers were observed.

In Amn of the gastrointestinal vein in X. laevis,
in SP-LI fibers of the gastrointestinal vein in R.
catesbeiana, in NKA-LI fibers of the gastrointes-

tinal vein in R. catesbeiana and B. japonicus, and
in CGRP-LI fibers of the gastrointestinal vein in R.
catesbeiana, R. nigromaculata and B. japonicus,
large bundles of nerves, often parallel to the
course of vein, were observed together with net-
works of thin nerve fibers (Fig. 8).

Innervation of Amphibian Hepatic Vein 609

ene eS pon Xr 2. Say
= : 3 as : Wate 2 <4 3
i 5 > * (ren
oot ONG ae EN SO
ate E 4) Boe Ce ee
=o : Ne ) Ney SEN *
= 2 SN < y Dray ;
Ga Fee 2S i Dae S
2 ee
xe, = ieee oe
prea “ § ~ ; oe les Ree £ .
= SS s oe c :
at Se : 3 ‘ : :
z Say on Sipe SS - b Qe aN i
- : Bee ACen ENS
a ee Ponce = ~
= : ioe oes > 4%. S
2 Ni esr ? cA
2 iN GS aN
T Pea y ; R bs *
2 { ; e a ee
<} St 2 SEED
Foe mises PAS 4 es
Coe 3 ‘i s So RG $ dip
£ ie Dae} Ses
ML NS a a bien ck ?
= An 3 - Se Gee ee. eae
z RS ci Rae eo yee =
A oS ae ee \
& y MoS % S g ay * é
Pee Le 3 . oe 7 en

Fic. 5.

Immunoreactive NKA forms a network of perivascular nerve fibers in the wall of R. catesbeiana hepatic

portal vein. a: Pre-abdominal vein (66) b: Gastrointestinal vein (X62).

Poe

pocars te
: eet é
5 Ee - > >
<<
‘ A x ee erect
“ards ios si =
Me =
t ae
: Ps
oe 4 2
= = ; Z
aa
; < : 5
a a Be
wd = < =
= ae
: ,
~
i

.

Fic. 6. CGRP-LI fibers in R. catesbeiana hepatic portal vein a: Pre-abdominal vein ( x 140) b: Gastrointestinal vein (
x38) SP-, NKA- and CGRP-LI nerve fibers show a distribution pattern similar to each other in R. catesbeiana

hepatic portal vein.

oe ae
ee :
eee —
Ss
z
ate
~~ &
~ .
ten:
ss
Je a
Z 35S
\
~ =
‘ he)

Fic. 7. VIP-LI nerve fibers in R. catesbeiana hepatic
portal vein a: Pre-abdominal vein (x 106) b: Gas-
trointestinal vein (X81) Innervation of VIP-LI
nerve fibers (arrow) is fewer than that of SP-, NKA-
and CGRP-LI nerve fibers.

DISCUSSION

The hepatic portal vein of the anuran amphi-
bians is well supplied with Amn. The Amn is
equally developed in all regions of the hepatic

Fic. 8. Nerve fibers displaying CGRP immunoreactiv-
ity in the X. /aevis gastrointestinal vein Nerve fibers
are organized in a plexus composed of large nerve
bundles (thick arrow), and of thin nerve fibers (thin

arrow) (X85).

portal vein in all of the animals examined. The
hepatic portal veins of the rat and the guinea-pig
also received a rich supply of Amn [5, 17]. The
predominant catecholamine in the urinary tracts of
anuran amphibians was adrenaline [18, 19]. And

610 K. OKAMOTO AND T. TAGAWA

the predominant catecholamine in the examined
anuran amphibians was also adrenaline.

Only a very few Chn were observed in the
hepatic portal vein of the anuran amphibians. X.
laevis and B. japonicus lacked Chn on the wall of
the pre-abdominal vein. Nerve fibers in the hepa-
tic portal vein of the guinea-pig and the rabbit did
not exhibit acetylcholinesterase activity [20].
However, a dense Chn innervation was demons-
trated in hepatic portal vein of the rat and the
human [7, 21], and Tagawa et al. [22] observed
acetylcholinesterse positive fibers in the cerebral
carotid artery in three species of amphibians. It
may be, therefore, noted that there are distinct
species differences in cholinergic innervation of
the hepatic portal vein.

In the cerebral arteries of mammals the cho-
linergic nerve plexuses showed approximately the
same density as the adrenergic ones [1]. In the
cerebral vessels of R. catesbeiana, however, we did
not observe any Chn, although we observed a rich
innervation of Amn [14]. In the hepatic portal vein
of the anuran amphibians, Chn were also fewer
than Amn.

In the hepatic portal vein of anuran amphibians,
very sparse VIP-LI fibers were observed in this
study, as compared with SP-, NKA- and CGRP-LI
fibers. In the cat, rat and guinea-pig, VIP-LI fibers
were also sparse around veins in the wall of the
small intestine [23]. Kobayashi et al. [24] observed
that VIP coexisted within Chn in the cerebral
arteries of the rat. We also observed VIP-LI fibers
and Chn, which showed the same distribution
pattern in the cerebral vessels of the hamster [25].
In this study, the possibility could not be excluded
that VIP coexists with acetylcholine in nerve fibers
around the hepatic portal vein in anuran amphi-
bians.

Barja et al. [6] observed a rich innervation of
SP-LI fibers in the hepatic portal vein of the rat.
These SP-LI fibers were almost certainly sensory in
nature at least in some mammalian species, since
1) they were substantially reduced in number
following treatment with capsaicin [26], 2) the
major location of peripheral SP-LI perikarya were
the dorsal root ganglia [6] and 3) after lesion of the
trigeminal ganglion, SP-LI fibers around appropri-
ate blood vessels disappeared [27]. The hepatic

portal vein of the anuran amphibians was provided
with SP-LI fibers. Morris et al. [3] reported that in
submammals (Bufo marinus), as in mammals, the
perivascular SP-LI fibers were sensory in nature.
The SP-LI fibers around hepatic portal vein in
anuran amphibians may be sensory.

In the hepatic portal vein of these anuran amphi-
bians, SP-LI fibers, CGRP-LI fibers and NKA-LI
fibers showed a similar distribution pattern. NKA
and SP belong to the tachykinin family and it was
concluded that immunoreactivity of NKA and SP
coexisted in the primary sensory neurons of rat
[28], and also SP coexisted with CGRP both in the
trigeminal ganglion and the nerve fibers around
cerebral blood vessels of the cat [29]. In the South
American toad (Bufo marinus), most of the im-
munoreactive cell bodies of SP in the dorsal root
ganglia contained CGRP [3]. These facts may
propose some interrelationship among these pep-
tides in the hepatic portal vein of anuran amphi-
bians.

The hepatic portal vein of anuran amphibians
was surrounded by networks of thin Amn, Chn
and peptidergic fibers. Besides these thin fibers, .
large nerve bundles were observed in the gastroin-
testinal vein, often running parallel to the course
of the vein. Nerve fibers from the vagus and from
the coeliac plexus reached the liver along the
hepatic artery and the hepatic portal vein in man
[4, 30]. The existence of the large nerve bundles in
the wall of the hepatic portal vein in anuran
amphibians may support this fact. The gastrointes-
tinal vein possessed innervation of Chn and pep-
tide-LI fibers somewhat more dense than the pre-
abdominal vein. This difference may reflect the
functional difference that exists between the two
veins.

Amn contracted the hepatic portal vein, and
Chn presumably counterbalanced the effect of
catecholamine [4, 31]. SP and CGRP, present in
the trigeminal cerebrovascular system of the rat,
may interact at the neuroeffector junction [29]. SP
contracted the hepatic portal vein of the rat, and
VIP caused a reduction of the amplitude of spon-
taneous contraction of the hepatic portal vein of
the rat [32]. NKA and SP possessed biological
activities characteristic of the tachykinin family
such as smooth muscle contraction. NKA may

Innervation of Amphibian Hepatic Vein

have a physiological role which is related to that of
SP in mammalian organisms [33]. It should be
suggested that nerves which innervate the hepatic
portal vein of the anuran amphibians may be local
regulators of the hepatic portal circulation.

10

11

i

13

14

REFERENCES

Edvinsson, L. (1985) Functional role of perivascular
peptides in the control of cerebral circulation. TINS,
8: 126-131.

Bowers, C. W., Jan, L. Y. and Jan, Y. N. (1986) A
substance P-like peptide in bullfrog autonomic nerve
terminals: anatomy biochemistry and physiology.
Neuroscience, 19: 343-356.

Morris, J. L., Gibbins, I. L., Campbell, G., Mur-
phy, R., Furness, J. B. and Costa, M. (1986)
Innervation of the large arteries and heart of the
toad (Bufo marinus) by adrenergic and peptide-
containing neurons. Cell Tissue Res., 243: 171-184.
De Luca, C., Cantagalli, A., De Angelis, E. and
Amenta, F. (1982) Cholinergic nerves in the rat
portal vein. Experientia, 38: 397-398.

Barja, F. and Mathison, R. (1982) Adrenergic and
peptidergic (substance P and vasoactive intestinal
polypeptide) innervation of the rat portal vein.
Blood Vessels, 19: 263-272.

Barja, F. and Mathison, R. (1984) Sensory innerva-
tion of the rat portal vein and the hepatic artery. J.
Autonom. Nerv. Syst., 10: 117-125.

Cantagalli, A., De Luka, C., Angelis, E. and
Amenta, F. (1983) Autonomic innervation of rat
portal vein. Ant. Anz. Jena., 154: 55-59.

Kent, G. C. (1983) Comparative Anatomy of the
Vertebrates. 5th ed, C. V. Mosby Company, St.
Louis.

Zamboni, L. and De Martino, C. (1967) Buffered
picric acid-formaldehyde: a new, rapid fixative for
electon microscopy. J. Cell Biol., 35: 148A.
Sternberger, L. A. (1986) Immunocytochemistry.
3rd ed, John Wiley & Sons, New York.

Takano, Y., Nagashima, A., Masui, H., Kuromizu,
K. and Kamiya, H. (1986) Distribution of substance
K (neurokinin A) in the brain and peripheral tissues
of rats. Brain Res., 369: 400-404.

Falck, B. (1962) Observation on the possibilities of
the cellular localization of monoamines by a
fluorescence method. Acta Physiol Scand., suppl,
197, 56: 1-25.

Karnovsky, M. J. and Roots, L. (1964) A “direct-
coloring” thiocholine method for cholinesterases. J.
Histochem. Cytochem., 12: 219-221.

Tagawa, T., Ando, K., Wasano, T. and Iijima, T.
(1979) A Histochemical study on the innervation of
cerebral blood vessels in the bullfrog. J. Comp.

15

16

17!

18

19

20

ZAM

dips

23

24

5)

26

DE

28

611

Neurol., 183: 25-32.

Deuchar, E. M. (1975) Xenopus: The South Afri-
can Clawed Frog. John Wiley & Sons, London.
Wasano, T. (1980) Morphology and its implication
of the vasomotor control of the cerebral vessels -
From the viewpoint of the comparative histochemic-
al study. Med. Bull. Fukuoka Univ., 7: 139-158.
Reilly, F. D., McCuskey, P. A. and McCuskey, R.
S. (1978) Intrahepatic distribution of nerves in the
rat. Anat. Rec., 191: 55-68.

Mc Lean, J. R. and Burnstock, G. (1966) Histoche-
mical localization of catecholamines in the urinary
bladder of the toad (Bufo marinus). J. Histochem.
Cytochem., 14: 538-548.

Unsicker, K., Axelsson, S., Owman, C. and Svens-
son, K. G. (1975) Innervation of the male genital
tract and kidney in the amphibia, Xenopus laevis

Daudin, Rana temporaria L, and Bufo bufo L. Cell
Tissue Res., 160: 453-484.

Burnstock, G., Crowe, R. and Wong, H. K. (1979)
Comparative pharmacological and histochemical
evidence for purinergic inhibitory innervation of the
portal vein of the rabbit, but not guinea-pig. Br. J.
Pharmac., 65: 377-388.

Amenta, F., Cavallotti, C., Ferrante, F. and Tonel-
li, F. (1981) Cholinergic nerves in the human liver.
Histochem. J., 13: 419-424.

Tagawa, T., Ishikawa, A., Su, W. C. and Okamoto,
K. (1989) Autonomic innervation of intra- and
extra-cranial arteries in the amphibia. Zool. Sci., 6:
233-239.

Feher, E. and burnstock, G. (1986) Ultrastructural
localization of substance P, vasoactive intestinal
polypeptide, somatostatin and neuropeptide Y im-
munoreactivity in perivascular nerve plexuses of the
gut. Blood Vessels, 23: 125-136.

Kobayashi, S., Kyoshima, K. Olschowka, J. A. and
Jacobowitz, D. M. (1983) Vasoactive intestinal
polypeptide immunoreactive and cholinergic nerves
in the whole mount preparation of the major cere-
bral arteries of the rat. Histochemistry, 79: 377-381.
Ishikawa, A. and Tagawa, T. (1983) The distribu-
tion of anti-VIP immunoreactive fibres of cerebral
blood vessels in rodent. Acta Anat. Nippon, 58: 305.
Barja, F., Mathison, R. and Huggel, H. (1983)
Substance P-containing nerve fibres in _ large
peripheral blood vessels of the rat. Cell Tissue Res.,
229: 411-422.

Liu-chen, L., Mayberg, M. R. and Moskowitz, M.
A. (1983) Immunohistochemical evidence for a
substance P-containing trigeminovascular pathway
to pial arteries in cats. Brain Res., 268: 162-166.
Dalsgaard, C. J., Haegerstrand, A., Theodorsson-
Norheim, E., Brodin, E. and Hokfelt, T. (1985)
Neurokinin A-like immunoreactivity in rat primary
sensory neurons coexistence with substance P. His-

29

30

Sil

612

tochemistry, 83: 37-39. |

Uddman, R., Edvinsson, L., Ekman, R., Kingman,
T. and McCulloch, J. (1985) Innervation of the
feline cerebral vasculature by nerve fibers contianing
calcitonin gene-releated peptide: trigeminal origin
and co-existence with substance P. Neurosci. Lett.,
62: 131-136.

Nobin, A., Baumgarten, H. G., Falck, B., Ing-
emansson, S., Moghimzadeh, E. and Rosengren, E.
(1978) Organization of the sympathetic innervation
in liver tissue from monkey and man. Cell Tissue
Res., 195: 371-380.

Mastrangelo, D. and Mathison, R. (1983) Everted
portal vein: a sensitive model for studies of vasoac-
tive compounds. J. Cardiovasc. Pharmacol., 5: 98-
101.

32

3h)

34

K. OKAMOTO AND T. TAGAWA

Hellstrand, P. and Jarhult, J. (1980) Effects of nine
different gastrointestinal polypeptides on vascular
smooth muscle in vitro. Acta Physiol Scand., 110:
89-94.

Nawa, H., Doteuchi, M., Igano, K. Inoue, K. and
Nakanishi, S. (1984) Substance K: a novel mamma-
lian tachykinin that differs from substance P in its
pharmacological profile. Life Sci., 34: 1153-1160.
Matsuyama, T., Shiosaka, S., Matsumoto, M.,
Yoneda, S., Kimura, K., Abe, H., Hayakawa, T.,
Inoue, H. and Tohyama, M. (1983) Overall dis-
tribution of vasoactive intestinal polypeptide-
containing nerves on the wall of cerebral arteries: an
immunohistochemical study using whole-mounts.
Neuroscience, 10: 89-96.

ZOOLOGICAL SCIENCE 7: 613-621 (1990)

Cell Culture of the Antennal Imaginal Disc of the
Silkworm, Bombyx mori L. and Differentiation
of Neurons from the Culture

YOSHIO WAKU, MITSUHIRO KOIKE and NAOYUKI YOSHIDA

Department of Applied Biology, Kyoto Institute of Technology,
Sakyo-ku, Kyoto 606, Japan

ABSTRACT—Antennal imaginal disc of the silkworm larvae is an epidermal organ composed of
undifferentiated cells, while during metamorphosis a rapid and remarkable neurogenesis of adult
olfactory neurons takes place from the disc. The discs of the 5th instar larvae were excised and disrupted
enzymatically or mechanically and then cultured in vitro using some insect culture media. Cells
proliferated in these media in various ways, i. e. formation of flat cell sheet, network of cytoplasmic
elongations, hollow monolayered vesicles and cell dispersions. Presently attempts to establish the cell
lines from the third one, vesicles, are being in progress but still wait for the success. In the last one, cell
dispersions, several cell types differentiated from the proliferated cells. Among them, two types refered
to ‘small and medium-sized round cells’ seemed to be neuroblasts. Cytoplasmic processes elongated
from those cells forming synapse, varicosities and filopodia which are characteristic to neurons. Such
morphological characteristics and the origin of those cells, antennal imaginal disc, strongly suggested

© 1990 Zoological Society of Japan

that they were differentiated neurons in vitro.
discussed.

INTRODUCTION

The antenna of the male silkmoth is well-known
as trapping site of the sex pheromone molecule,
Bombykol, and its ultrastructure and function
have been studied extensively by many authors. It
is equipped with thousands of olfactory sensilla
through its entire length, and each sensillum con-
tains usually two primary neurons which are re-
sponsible for trapping of the pheromone molecules
[1]. The antennal imaginal disc, which resides in
the larval head in connection with the larval anten-
na, is fundamentally a simple and small pseudo-
stratified epidermal tissue consisted of uniform and
undifferentiated cells and contains no neurons in it
(Fig. 1). The neurons begin to differentiate im-
mediately after pupation from the imaginal disc
tissue which now becomes a large sac-like mono-
layered epidermis by eversion, and by the third
day of the pupal life the differentiation of numer-
Ous neurons is mostly completed (Waku, unpubl.

Accepted September 28, 1989
Received July 21, 1989

The importance of this in vitro neurogenesis was

observation). A rapid and remarkable neurogene-
sis occurs during this short period in vivo, and
therefore success of in vitro culture of the imaginal
disc itself or the cell lines derived from it will offer
a great advantage for the study of insect
neurogenesis which still deserves more extensive
works. We have been engaged in the establish-
ment of cell lines derived from the antennal im-
aginal disc, and in the present paper we would like
to report the in vitro differentiation of the neurons
from the cultured imaginal disc cells.

MATERIALS AND METHODS

The silkworm larvae were raised with artificial
diet under 25°C and 12L-12D light condition.
They took 12 days for the 5th instar development
and then pupated. The antennal imaginal discs of
the male 5th instar larvae (5 days old), together
with their ‘peripodial membranes’ which are thin
epidermal tissue surrounding the discs (Fig. 1),
were excised aseptically from the larval heads and
rinsed in the insect physiological saline. The discs
were either teared to three fragments by the

614 Y. Waku, M. KoIKE AND N. YOSHIDA

forceps or moved intact to Ca-Mg-free saline. In
the latter case, the discs were rinsed several times
in the same saline, immersed in the 0.025% pan-
creatin solution at 25°C for 1 hr to dissociate the
tissue, pipetted vigorously and then were centri-
fuged at 1,000 rpm for 1 min. The sedimentation
contained numerous small cell clusters consisted of
some hundred cells (Fig. 2).

These preparations, both tissue fragments and
cell clusters, contained no live neurons since in the
larva the presence of live and functional neurons is
restricted in the larval antenna which is remote
from the dsics and can be removed easily at the
excision of the discs.

The preparations were cultured at 25°C in the
double-stoppered small culture bottles or culture
tubes. Each bottle or tube contained disc frag-
ments or cell clusters derived from 20-50 discs.
The used culture media were as follows: (1)
Grace’s medium (Gibco, without bovine plasma
albumen) + 10% fetal bovine serum (FBS, Gibco);
(2) Grace’s medium+10% Bombyx hemolymph
plasma; (3) Grace’s medium+10% FBS+10%
Bombyx plasma; (4) MGM 443 medium after
Mitsuhashi [2]; (5) MGM 448 medium [2]; and (6)
MGM 448 medium 1 part+ conditioned medium 1
part, obtained from the MGM448- culture of a
silkworm cell line (SES-BoMo _ ) which was kindly
supplied by Dr. S. Imanishi of the National Insti-
tute of Sericultural and Entomological Science. A
half volume of each medium was discarded and
replenished at least once a week through the
course of culture. Generally the last two media
gave better results.

Apart from above culture procedures, the discs
of the 5th instar larvae (5 days old) were excised,
fixed with Bouin’s fixative and embedded in pa-
raffin. Sections (6 ~m) were stained with toluidine

blue to elucidate the state of cell proliferation and
differentiation in the disc tissue at the culture
initiation. To detect the proliferating S-phase
cells, immunocytochemical method by use of 5-
bromo-2 deoxyuridine as labelling substance [3]
was applied (cell proliferation kit, Amersham).

RESULTS

Histological observation of the imaginal discs at the
culture initiation

Figure 1 shows the cross-sectional view of the
imaginal disc at the culture initiation stage. The
disc was composed of uniform cells which were
proliferating actively as indicated by uptake of
5-bromo-2’-deoxyuridine (blackened nuclei), but
figures of cellular differentiation including
neurogenesis were not found at all. Also the cell
clusters derived from dissociation of the discs by
enzymatic treatment did not contain any differenti-
ated neurons or other specialized cells (Fig. 2).

Proliferation types in the cell culture

Various types of cell proliferation were observed
during the course of cell culture which have been
continueing for more than one year.

First, proliferating cells immigrated from the
disc fragments formed thin cell sheets which
adhere to the original fragments (Figs. 3 and 4).
The margin of the cell sheet was well-defined, and
detachment of the cells from the sheet did not
occur. Cells constituting the sheet were round in
shape and variously sized (5-50 ~m in diameter)
and often contained lipid droplets in the cyto-
plasm. This type of cell proliferation was found in
the disc fragments cultured in a bottle containing
Grace’s medium+10% Bombyx plasma. Forma-

Fic. 1. Cross section view of a part of antennal imaginal disc (Id) and its peripodial membrane (Pm). Note apparent
uniformity of the disc cells. Blackened nuclei (arrowheads) are those of S-phase cells uptaking 5-bromo-2’-

deoxyuridine. 400.

Fic. 2. Cell clusters resulted from the imaginal disc dissociation with enzymatic treatment. 300.

Fic. 3.

Cell sheet formed by the immigration and proliferation of cells from the disc fragment. 106 days after culture

initiation. Cultured in Grace+10% Bombyx plasma. x 100.

Fic. 4. Ditto, 122 days after culture initiation.
proliferation. 100.
Fic. 5.

culture initiation. 600.

Note apparent enlargement of the cell sheet caused by cell

Network of cytoplasmic elongations from the cell clusters cultured in MGM 448 medium. 90 days after

4X

15

6

1SC

1D

Ina

Cell Culture of Antennal Imag

KOIKE AND N. YOSHIDA

M

’

Y. WAKU

616

Cell Culture of Antennal Imaginal Disc 617

tion and enlargement of the sheets by cell prolif-
eration (compare Fig. 3 to Fig. 4) began about 3
months after the culture initiation, then continued
for about 7 months and finally the cells degener-
ated.

The second type, observed in the disc fragments
and cell clusters cultured in MGM 448 medium,
was also a kind of cell sheet but the immigrated
and proliferated cells formed a network of broad
and flat cytoplasmic elongations (Fig. 5). This type
cells have survived for more than half a year until
now, but vigorous proliferation has not yet
occured.

The third type of cell proliferation was observed
mainly in the disc fragments and cell clusters
cultured in Grace+10% Bombyx plasma and
MGM 448 medium, respectively. The proliferated
cells formed many hollow vesicles of monolayered
tissue with various sizes (Figs. 6-8). The vesicles

TABLE 1.
the culture bottles containing various media.

began to appear about 1 month after the initiation
of culture. Appearance rate of such vesicles in
these cultures was considerably high, ranging 60-
75% of the culture bottles. The diameter of a
vesicle shown in Figure 6 increased about twice
within 20 days of continued culture (see Fig. 8). If
the size of each cell remained constant, this in-
crease may mean that the doubling time of cells in
this vesicle was about 10 days. Attempts to obtain
serial subcultures of these vesicles by speeding up
of this rather slow proliferation are in progress, but
the establishment of cell lines is yet to be expected
in the future.

The final and fourth type appeared in all bottles
containing disc fragments cultured in MGM 448
and MGM 448+ conditioned medium. Contrally
to above three proliferation types which formed
more or less organized ‘tissues’, the fourth
appeared as migrated and freely dispersed cells in

Summarized results showing occurence of four proliferation types and neuron differentiation in
For further explanations, see the text

State of . No. of Cell Neuron

cultured discs Culture media bottles Cell sheet Network Vesicles dispersion differentiation
pues ne 4 (100) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
Grace + 10% 5 (100) 1 (20) 0 (0) 3 (60) 0 (0) 0 (0)
Bombyx plasma

Disc fragments Grace+FBS+ 4 (100) 0 (0) 0 (0) 3) (75) 0 (0) 0 (0)
Bombtx plasma
MGM 443 5 (100) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
MGM 448 17 (100) 0 (0) 3 (18) 11 (65) 17 (100) 15 (88)
MGM 488+ 6 (100) 0 (0) 0 (0) 4 (67) 6 (100) 6 (100)
conditioned
medium (1:1)

Cell clusters Grace + 10% 1 (100) 0 (0) 0 (0) 1 (100) 0 (0) 0 (0)
Bombyx plasma
MGM 488 1 (100) 0 (0) 1 (100) 0 (0) 0 (0) 1 (100)

Numbers in parenthesis denote the percentages.

Fic. 6. A vesicle (V) derived from the cell clusters cultured in Grace+ 10% Bombyx plasma. 80 days after culture

initiation. 100.

Fic. 7. Ditto, 90 days after culture initiation.
porliferation. x 100.

Fic. 8.

The diameter of vesicle has apparently increased due to cell

Ditto, 100 days after culture initiation. Vesicle size increased almost twice of that in Fig. 6. 100.

Fic. 9. Dispersed cells migrated from the disc fragments cultured in MGM 448. 7 days after culture initiation.

x 300.

Fic. 10. Ditto, 21 days after culture initiation. Note vigorous cell proliferation. 300.

Fic. 11.

al

Ditto, 28 days after culture initiation. White arrows indicate small and medium-sized round cells. 300.

618 Y. Waku, M. KolIkE AND N. YOSHIDA

nn
oo —

Cell Culture of Antennal Imaginal Disc 619

the bottle as early as in the first week of culture
(Fig. 9). Apparent cell.proliferation occured dur-
ing the third and fourth weeks (Fig. 10), and
during and after these weeks cell differentiation
took place (Fig. 11). Judging from the size and
shape of cells at least six differentiated cell cate-
gories were recognized, and two of them desig-
nated as ‘small and medium-sized round cells’ are
of special interest as will be described later (white
arrows in Fig. 11). These differentiated cells sur-
vived for 5—8 months and then degenerated.

Although the proliferation types were classified
to four as above described, it should be noticed
that they often coexisted even in the same culture
bottle. Especially the third and fourth types were
often found simultaneously in the bottles with
MGM 448 or MGM 448+ conditioned medium.

Cultures performed using the media other than
above cited ones generally did not give good
results, yielding only small numbers of migrated
cells which died soon. Table 1 summarizes the
results described above.

Neurogenesis in vitro

Among the dispersed cells shown in Figure 11,
some ‘small and meidum-sized round cells’ (dia-
meter 10-15 ~m and 20-25 um, respectively)
often began to extend short but apparent processes
from their cytoplasm (Fig. 12). As the culture
proceeds, such processes elongated further: some
of them were monopolar (Fig. 13) while others
were bipolar (Fig. 14). Cells with multipolar pro-
cesses also appeared, often forming synapse-like
connections with other cells (Fig. 15). Some pro-
cesses were equipped with periodical dilations
which remind varicosities in the neuron (Fig. 16).
Not only these morphological characteristics be-

longing to neurons, but also the nature and origin
of the culture (antennal imaginal discs which are
destined to produce numerous neurons) strongly
suggest that these cells should be regarded as true
neurons. Thus the ‘small and medium-sized round
cells’ should be neuroblasts. The differentiation of
neurons continued rather sporadically during the
third week and fourth month of culture, and each
bottle contained about 30 neurons in it. Neurons
survived for 10-60 days often changing actively the
shape of their processes.

Apart from the neurons differntiated from the
disc fragments, a few cells with long axon-like
processes were found in a cell cluster culture in
MGM 448 (Fig. 17). The terminal of the process
was equipped with a growth cone and neumerous
filopodia, which could be again the morphological
evidence of the neuron (Fig. 18). These results are
shown in Table 1.

DISCUSSION

The imaginal discs of the holometaborous in-
sects are composed of undifferentiated epidermal
cells, while they undergo rapid and conspicuous
differentiation during early half of adult develop-
ment and finally establish each particular adult
organ. Several attempts have been done aiming
the cell culture of the dissociated imaginal discs
either enzymatically or mechanically, and final
establishment of cell lines obtained from such
cultures. Kurtti and Brooks [4] were the first who
carried out such study using the wing-discs of some
lepidopterous larvae, and obtained fibroblast-like
and epitheliocyte-like cells which survived in vitro
for 4 weeks. However, continued cell lines were
not obtained from their cell culture. Schneider [5]

Fic. 12. Two cells extending processes from their cytoplasm. Cultured in MGM 448. 28 days after culture initiation.

x 300.
Fig. 13.
Fic. 14.
Fic. 15.
448. 35 days after culture initiation. 150.
Fic. 16.
days after culture initiation. 150.
Fic: 17:

A cell with monopolar process. Cultured in MGM 448. 35 days after culture initiation. 300.
A cell with bipolar process. Cultured in MGM 448. 35 days after culture initiation. 300.
A neuron with multipolar processes and forming a synapse (arrow) with another one. Cultured in MGM

A cluster of neurons with varicosity-like dilations (arrows) in their processes. Cultured in MGM 448. 35

A neuron with long axon (N), appeared in cell cluster culture with MGM 448. Cp: cells with flat and broad

cytoplasmic processes. 30 days after culture initiation. 150.
Fic. 18. Enlarged view of the axon terminal of a neuron shown in Fig. 17. Note the presence of growth cone (arrow)

and filopodia (arrowheads). 400.

620 Y. Waku, M. KolIkKE AND N. YOSHIDA

obtained 3 cell lines from trypsinized embryonal
fragments of Drosophila. When these cell aggre-
gates were implanted into the adult abdomen, they
yielded several adult organs such as head, antenna,
sclerite and sternite. Thus the original cell aggre-
gates from the embryos were reasonablly assumed
that they contained some imaginal disc-derived
cells with unspecified nature. Works along this line
were carried out by several authors since then [6-
8]. Specified adult organs or cell lines, however,
were difficult to obtain by these works since the
whole embryos of Drosophila were composed of
many heterogenous cell categories with unknown
nature.

First cell lines from a specified imaginal disc,
wing discs of three lepidopterans, were established
by Lynn et al. [9] and Lynn and Oberlander [10].
These cell lines showed similar patterns of protein
synthesis with the cells in the intact wing discs.

Ui et al. [11] succeeded to establish 10 cell lines
from speficied imaginal discs of Drosophila such as
wing, antenna and mixed discs. Each cell line
consisted of 3 morphologically different cell types:
round, bipolar and polygonal cells. Heterogeneie-
ty of cells comprising the original discs might be
the cause of such results, and the nature of these
cells were not refered to in this work.

Our present work, aiming the establishment of
cell lines from a speficied disc, antennal disc of
Bombyx, is yet in progress presently. As the
establishment of insect cell lines often requires
long period more than one year, the success of this
purpose seems to be expected in the future. Pre-
sently we would like to point out only the fact that
in all cases of succeeded cell culture of the imaginal
discs [4-11], proliferated cells took more or less
vesicular shapes and our present results also
observe this general rule.

Several works on the in vitro culture of insect
neurons and neuroblasts have been done since
Shields and Sang [12]. In most works, however,
neurons were obtained from already differentiated
and established central nervous system of late
embryos, young larvae or pupae [13-18]. Apart
from these works, Keil [19] cultured for a short
while the already differentiated and isolated anten-
nal sensilla and their neurons of the silkmoth
pupae, Antheraea polyphemus and A. pernyi.

Our present work showed in vitro differentiation
of the neurons from undifferentiated antennal disc
cells of Bombyx, and is comparable to the previous
works dealt with the neuronal differentiation from
cultured embryonal cells of Drosophila, which
were actually the mixture of various cell types with
unspecified nature [12, 20-25]. Advantage of our
work is derived from the choice of the specified
antennal imaginal disc, which should contain
numerous neuroblasts or their precursors, as the
material for investigation of in vitro neurogenesis.
Further improvement of this culture system will
give much information on insect neurogenesis.

As to the cytoplasmic processes which characte-
rize the neurons, it should be pointed out that
every cell with cytoplasmic processes is not neces-
sarily a neuron. Glia cells, fibroblasts and myob-
lasts are also instances of such cells with prominent
processes. However, the possibility confusing
these cells with our ‘neurons’ is unlikely since (1)
the morphological characteristics of the processes
in our neurons are quite different from those in the
glia cells or fibroblasts, as shown in Figures 12-18,
and (2) the antennal imaginal disc is of ectodermal
origin and therefore the mixing of myoblasts into
our cell culture can be ruled out. The broad and
flat cytoplasmic processes shown in Figures 5 and
17, at least partly, could be those of glia cells, since
Lees and Beadle [17] found the resembling flat
processes in their glia cell culture.

Theoretically, proliferation of undifferentiated
cells which eventually leads to the establishment of
cell lines should be another subject than in vitro
differentiation of specialized cells such as neurons,
and may require different culture conditions than
the latter. We found, however, both the vesicles
composed of undifferentiated cells and dif-
ferentiating neurons were coexisting together in
the MGM 448 medium. Since this medium did not
contain Bombyx plasma, differentiation of
neurons cannot be attributd to insect hormones
which might be present in the hemolymph in small
quantity. Segregation and establishment of
appropriate culture conditions for each of these
two subjects will require further studies in the
future.

Cell Culture of Antennal Imaginal Disc

ACKNOWLEDGMENTS

We are indebted to Drs. Y. Kuroda (National Institute

of Genetics), J. Mitsuhashi (Tokyo University of Agri-
culture and Technology), K. Ui and T. Miyake (Mitsu-
bishi-Kasei Institute of Life Sciences) and S. Imanishi
(National Institute of Sericultural and Entomological
Science) for their valuable discussions and suggestions.
This work was supported by a Grant-in-Aid for scientific
research from the Ministry of Education, Science and
Culture of Japan (No. 6348008).

10

11

2

REFERENCES

Schneider, D. und Kaissling, K-E. (1957) Der Bau
der Antenne des Seidenspinners Bombyx mori L. I.
Sensillen, Cuticulare Bioldungen und innerer Bau.
Zool. Jahrb. Abt. Ant., 76: 223-250.

Mitsuhashi, J. (1984) Isolation of a continuous cell
line from larval fat bodies of an arctiid moth,
Spiloarctia seriatopunctata (Insecta, Lepidoptera,
Arctiidae). Zool. Sci., 1: 415-419.

Gratzner, H. G., Lief, R. C., Ingram, D. J. and
Castro, A. (1975) The use of antibody specific for
bromodeoxyuridine for the immunofluorescent de-
termination of DNA replication in single cells and
chromosomes. Exp. Cell Res., 95: 88-94.

Kurtti, T. J. and Brooks, M. A. (1970) Growth of
lepidopteran epithelial cells and hemocytes in prim-
ary cultures. J. Invert. Pathol., 15: 341-350.
Schneider, I. (1972) Cell lines derived from late
embryonic stages of Drosophila melanogaster. J.
Embryol. exp. Morph., 27: 353-365.

Shields, G., Dtirbendorfer, A. and Sang, J. H.
(1975) Differentiation in vitro of larval cell types
from early embryonic cells of Drosophila melano-
gaster. J. Embrycl. exp. Morph., 33: 159-175.
Dutrbendorfer, A., Shields, G. and J. H. Sang
(1975) Development and differentiation in vitro of
Drosophila imaginal disc cells from dissociated early
embryo. J. Embryol. exp. Morph., 33: 487-489.
Kambysellis, M. P. and Schneider, I. (1975) In vitro
development of insect cells. III. Effects of ecdysone
on neonatal cells. Dev. Biol., 44: 198-203.

Lynn, D. E., Miller, S. G. and Oberlander, H.
(1982) Establishment of a cell line from lepidopter-
an wing discs: induction of newly synthesized pro-
teins by 20-hydroxyecdysone. Proc. Natl. Acd. Sci.
USA, 79: 2589-2593.

Lynn, D. E. and Oberlander, H. (1983) The estab-
lishment of cell lines from imaginal wing discs of
Spodoptera frugiperda and Plodia interpunctella. J.
Insect Physiol., 29: 591-596.

Ui, K., Ueda, R. and Miyake, T. (1987) Cell lines
from imaginal discs of Drosophila melanogaster. In
Vitro Cell. Dev. Biol., 23: 707-711.

Shields, G. and Sang, J. H. (1970) Characteristics of
five cell types appearing during in vitro culture of

13

14

5

16

Uy

18

19

20

a)

jij

23

24

25

621

embryonic material from Drosophila melanogaster.
J. Embryol. exp. Morph., 23: 53-69.

Chen, J. S. and Levi-Montalcini, R. (1970) Axonal
growth from insect neurons in glia-free culture.
Proc. Nati. Acad. Sci. USA, 66: 32-39.

Hicks, D., Beadle, D. J., Giles, D. P. and Usher-
wood, P. N. R. (1980) Ultrastructure of dissociated
nerve cells of Periplaneta americana (L) (Dictyop-
tera:Blattidae) growing in culture. Int. J. Insect
Morph. and Embryol., 10: 225-233.

Hicks, D. and Beadle, D. J. (1981) Ultrastructure
of dissociated neurons from pupae of Spodopera
litoralis Boiduval (Lepidopera:Noctuidae) growing
in culture. Int. J. Insect Morph. and Embryol. 10:
399-347.

Wu, C-F., Suzuki, S., and Poo, M-M. (1983) Dis-
sociated neurons from normal and mutant Dro-
sophila larval central nervous system in cell culture.
J. Neurosci., 3: 1888-1899.

Lees, G. and Beadle, D. G. (1988) Cockroach
neuronal cultures as models for investigating the
neuropharmacology of insect central nervous sys-
tem. In “Cell Culture Approaches to Invertebrate
Neuroscience”. Ed. by D. J. Beadle, G. Lees and S.
B. Kater, Academic Press, London, pp. 109-148.
Wu, C-F. (1988) Neurogenetic studies of Drosophi-
la central nervous system neurons in culture. In
“Cell Culture Approaches to Invertebrate Neurosci-
ence”. Ed. by D. J. Beadle, G. Lees and S. B.
Kater, Academic Press, London, pp. 149-188.
Keil, T. A. (1987) Structure of isolated sensilla and
sensory neurons in vitro: Observations on develop-
ing silkmoth antennae. Cell Tissue Res., 520: 543-
549.

Seecof, R. L., Alleaume, N., Teplitz, R. L. and
Gerson, I. (1971) Differentiation of neurons and
myocytes in cell cultures made from Drosophila
gastrulae. Exp. Cell Res., 69: 161-173.

SECCOMmR ome aD Onadyara0 Je anda aieplitzs IRase-
(1973) Differentiation of Drosophila neuroblasts to
form ganglion-like clusters of neurons in vitro. Cell
Differentiation, 2: 143-149.

Kuroda, Y. (1974) Jn vitro activity of cells from
genetically lethal embryos of Drosophila. Nature,
252: 40-41.

Kuroda, Y. (1982) Drosophila tissue culture: retros-
pect and prospect. In “Invertebrate Cell Culture
Applications”. Ed. by K. Maramorosch and J. Mit-
suhashi, Academic Press, London, pp. 53-104.
Furst, A. and Mahowald, A. P. (1985) Differentia-
tion of primary embryonic neuroblasts in purified
neural cell cultures from Drosophila. Devel. Biol.,
109: 184-192.

Furst, A. and Mahowald, A. P. (1985) Cell division
cycle of cultured neural precursors cells from Dro-
sophila. Devel. Biol., 112: 467-476.

=

i

a
ae ut y
=

-acigbep: ener nisutt ep oe eArciosibele
ie Boorse wich pea map ym

, bist erie rt NAS
j yeh Lia angi ade: a ha pe
‘ ‘ Setonqatlepitl,
5 ime th a Jonumint
e

ZOOLOGICAL SCIENCE 7: 623-630 (1990)

Structure of the Internal Gills in Tadpoles of the
Crab-Eating Frog, Rana cancrivora

Minoru UcHIYAMA, HIDEKI YOSHIZAWA‘, CHIKASHI WAKASUGI

and CHITARU OGURO?

Department of Oral Physiology and *Department of Oral Histology, Shool of
Dentistry at Niigata, The Nippon Dental University, Niigata 951,
‘Department of Oral Histology, Matsumoto Dental College,

Shiojiri 399-07, and *Department of Biology, Faculty
of Science, Toyama University, Toyama 930, Japan

ABSTRACT—The morphology of internal gills was studied by light and electron microscopy in
crab-eating frog tadpoles (T-K stages XIII-X VII) raised in tap-water. The internal gills consist of four
pairs of gill arches, on which gill tufts are arranged linearly. Each gill tuft is composed of a stem and
numerous ramifications. Histologically, a few PAS-positive cells are evident in the stem of the gill tuft.
In fine structure, bi- or multilayered epithelia of internal gills are composed of pavement cells,
mitochondria-rich (MR) cells and basal cells. The cuboidal pavement cells are abundant in the ventral
epithelium of the gill arches and the epithelium of the stem of the gill tufts. On the other hand, the
simple epithelium in the apical portion of the gill tufts consists of squamous pavement cells. The
cuboidal and squamous pavement cells are identical in ultrastructure and contain many apical vacuoles
and short microvilli covered with a fuzzy coat at the free surface. The MR cell is pleomorphic and
characterized by numerous metochondria in the supranuclear area. This type of cell has a round, ovoid
or pear-shaped profile in sections perpendicular to the surface. The matrix of the cytoplasm of the MR
cells is electron-dense. Simple basal squamous cells located under the pavement or the MR cells are
interconnected with the covering cells by desmosomes. Thus, four types of cell, pavement, MR, basal
Squamous and mucous, are distinguishable in the internal gill of the tadpole of the crab-eating frog.

© 1990 Zoological Society of Japan

INTRODUCTION

In anuran larvae, the internal gills are composed
of well branched short tufts, which bulge from the
ventral walls of the gill arches and function until
metamorphosis [1]. In fish, the gills serve not only
as an organ for respiration, but also as the main
site of ion exchange and nitrogenous waste excre-
tion [see 2]. In contrast, only one report has been
published on the ultrastructure of the internal gills
in anuran tadpoles [3]. Based on electron micros-
copy, Hourdry [3] suggested that the internal gills
of tadpoles may play an osmoregulatory as well as
a respiratory function. In fact, physiological stud-
ies have shown that the gill chamber is the major
site of ion exchange [4], although no morphologi-

Accepted October 30, 1989
Received October 14, 1989

cal evidence was presented.

The tadpole of the crab-eating frog, Rana can-
crivora, is the only amphibian larva able to inhabit
a brackish-water environment [5, 6]. Gordon and
Tucker [6] suggested that the tadpole of the crab-
eating frog is a good osmoregulator and that the
gills may be important for ion exchange in euryha-
line environments, although they did not show any
cytological or physiological evidence. The present
study was therefore undertaken to clarify the histo-
logical and cytological structure of the internal gills
of the tadpole of the crab-eating frog. One pur-
pose of this study was to provide basic cytological
knowledge on the role of the internal gills in
osmotic regulation of this euryhaline tadpole.

MATERIALS AND METHODS

Adult males and females of the crab-eating frog,

624 M. UcuiyAMA, H. YOSHIZAWA et al.

Rana cancrivora, were captured in mangrove
swamps (salinity 31%c) at Ang-Sila near Bangkok,
Thailand, in April 1987. Tadpoles for the present
experiments were obtained from these adults and
raised in the laboratory at Niigata. Details of
fertilization and development of tadpoles were

described previously by Uchiyama et al. [7]. The
embryonic stages were judged according to the
developmental stages for R. pipiens by Taylor and
Kollros [8]. Tadpoles (T-K stages XIIJ-XVII)
were reared in tap-water at 24.5-26°C.

The internal gills were dissected out quickly and

Fic. 1.

Light micrograph of a cross-section at the ear level in a tadpole (stage XIII). Gill rakers and gills tufts are

seen in the dorsal and ventral sides of ceratobranchialia, respectively. AC, auditory cavity; Cl-C4, ceratobran-
chialia 1 to 4; GC, gill chamber; GT, gill tuft; PH, pharynx. Hematoxylin and eosin. X40.

Fic. 2. Light micrograph of gill tufts of the 2nd and 3rd gill arches (stage XVI). B, branchial muscle; C2 and C3,
ceratobranchialia 2 and 3; GR, gill raker; GT, gill tufts. Arrows indicate the location of eosinophilic cells.

Hematoxylin and eosin. 160.

Fig. 3.

pavement cell. Toluidine blue. 630.

Light micrograph of semi-thin section of a gill tuft (stage XV). C, capillary; M, mitochondria-rich cell; P,

Gills in Tadpoles of Crab-Eating Frog 625

treated as follows. For light microscopy, they were
fixed with Bouin’s solution, dehydrated and
embedded in paraffin. They were then sectioned
serially at 6 or 8ym by the routine paraffin
method. The stains used were: 1) Mayer’s hema-
toxylin and eosin, 2) Heidenhain’s azan, or 3)
periodic acid-Schiff reagent (PAS) and hematoxy-
lin. Some semi-thin (1 zm) sections were stained
with toluidine blue.

For electron microscopy, the internal gills were
excised rapidly and immediately bathed with a
solution of 2.5% glutaraldehyde—2% paraformal-
dehyde in 0.1M sodium cacodylate buffer (pH
7.2), then cut into small pieces and placed into
fresh fixative for 6 hr. Tissue pieces were washed
twice with 0.1.M sodium cacodylate buffer, then
post-fixed with 1% osmium tetroxide for 1 hr.
They were washed in distilled water, dehydrated in

a
al

= Eee a i a

an ethanol series and embedded in epoxy resin.
Thin sections were stained with methanolic uranyl

acetate and alkaline lead citrate, and examined
with a JEOL JEM-100B electron microscope.

RESULTS

Light microscopy

There are four pairs of gill arches, which are
supported by cartilaginous skeletons, the cerato-
branchialia 1-4. Ventral and dorsal terminals of
the ceratobranchialia are joined with each other,
thus forming a gill basket. On the inside of this
structure, the dorsal parts of the gill arches are
covered with the gill rakers, which are composed
of a bilayer of cells which are columnar and
Squamous.

Fics. 4-9. Electron micrographs of tadpole gill tufts (stage XV and XVII).
Fic. 4. Basal parts of ramifications of a gill tuft (stage XVII). C, capillary; M, mitochondria-rich cell; P, pavement

cell. 2,700.

4

M. UcutvAMa, H. YosHIzawa et all.

tochondria-rich (M),

ing mi

ll arch (stage XVII) consist of a bilayer of cells, includ

f the gi

1um O

thel

Ventral epi
pavement (P) and basal squamous (B) cells. 6,900.

Fic.

lli covered with a

icrOV1

Apical site of pavement cell (stage XVII). Note many apical vacuoles and short m

fuzzy coat at the free surface. 13,200.

Fic. 6.

4

627

Gills in Tadpoles of Crab-Eating Frog

AAG
g 8
ee
oOo wv
Es
>
aS
(al, (ay
CHG
Con
i?)
(oy
Qo
aay
oes
oo
@ 5
2
=e)
ae
oe)
Q, G
(S)
Qe)
>
oO
a
ae)
7 2
SES
—
iS

capi

’

E

18

f the apical portion of a gill tuft (stage XV)
hboring cells.

1 surfaces of ne

Cross-section o
he apica

covering t
x 3,300.

Fic. 7.

Progressive stage of degeneration of a mitochondria-rich cell, showing condensation of the cytoplasm, and

condensation and indention of the nucleus (stage XVII). 9,300.

Fic. 8.

628 M. UcuiyaMA, H. YOSHIZAWA et al.

The internal gills consist of soft gill tufts situated
linearly on the ventral and lateral parts of the gill
arches. The gill tuft is composed of a.stem and
numerous ramifications and is well vascularized.
The epithelium of the stem of the gill tuft consists
of bi- or multilayers of cells which are similar to
those of the ventral epithelium of the gill arches.
In the 2nd and 3rd gill arches the gill tufts are more
numerous than in the other gill arches. Figure 1
shows the gross morphology.

Large round or ovoid cells characterized by
eosinophilic cytoplasm are scattered in the ventral
and lateral epithelium of the gill arches and the
stem of the gill tufts, but not in the dorsal epithe-
lium of the gill arches (Fig. 2). The cytoplasm of
this cell type is stained densely by toluidine blue in
semi-thin sections (Fig. 3). A few PAS-positive
cells are evident in the gill tufts. The columnar
cells of the gill rakers are also PAS-positive,
showing the presence of mucopolysaccharides.

Electron microscopy

Bi- or multilayered epithelia composed of

cuboidal and squamous cells are evident in the
ventral parts of the gill arches and the stems of the
gill tufts. These cells are connected with each
other by tight junctions at their apical portion and
bear a few irregular, lateral and basal infoldings
that interdigitate with each other. Broad inter-
cellular spaces are sometimes observed at the
lateral and basal portion (Fig. 4).

At the ventral and lateral epithelia of the gill
arches and the basal parts of the gill tufts, simple
Squamous cells are situated under the cuboidal
cells (Fig. 5). In these squamous cells the cytoplas-
mic organelles are sparse and the nucleus-
cytoplasm volume ratio is high. The basal
Squamous cells are interconnected with the cover-
ing cells by desmosomes (Fig. 5).

In the ventral and lateral epithelia of the gill
arches and the gill tufts, two types of cuboidal cell
are distinguishable. One, the clear cell, has elec-
tron-lucent cytoplasm and the other, the dark cell,
has electron-dense cytoplasm. The clear cells,
which are pavement cells, sometimes take a
squamous profile and are abundant in the ventral

Fic.9. Note many clear or dense vesicles and simple tubular structures in the supranuclear area of a mature

mitochondria-rich cell (stage XV). 13,000.

Gills in Tadpoles of Crab-Eating Frog 629

and lateral epithelia and the gill tufts. In the
pavement cells, there are many apical vacuoles and
short microvilli which are covered with a fuzzy coat
at the free surface (Fig. 6). The pavement cells
contain a Golgi apparatus with numerous associ-
ated vesicles, such as rough endoplasmic reticulum
and free ribosomes. The apical part of the gill tufts
consists of a monolayer of the squamous pavement
cells, which may function in respiration (Fig. 7).
Electron-lucent cytoplasm resulting from poor de-
velopment of cellular organelles and fewer micro-
villi are observed in some squamous pavement
cells in the apical parts of the gill tuft. Between the
Squamous pavement cells and the basal layers
extracellular spaces are observed.

The dark cells are characterized by electron-
dense cytoplasm and numerous mitochondria in
the supranuclear area. These cells are termed
mitochondria-rich cells (MR cells). The MR cells
are pleomorphic and often show a round, ovoid or
pear-shaped profile in sections perpendicular to
the surface. In this type of cell, the apical surface
is furnished with microvilli and the nucleus is
usually present at the basal portion (Figs. 4, 8 and
9). The MR cells are fewer in the apical portion
and numerous in the basal parts of the gill tufts and
the ventral epithelium of the gill arches. The MR
cells seem to be absent in the gill rakers. The
apical surfaces of MR cells are often covered by
cytoplasmic extensions of neighboring pavement
cells. Degenerating MR cells are frequently
observed, being characterized by cellular shrink-
age, condensation of cytoplasm, condensation and
indentation of the nucleus and the presence of
many lysosomes (Fig. 8). In contrast, many clear
or dense apical vesicles, tubular structures and
Golgi apparatus are present in the mature MR
cells (Fig. 9).

DISCUSSION

The present observations using light microscopy
make it clear that the basic structure of the gill of
the present tadpoles is similar to that of R. cates-
beiana [9]. Hourdry [3] reported from ultra-
structural studies that the respiratory squamous
cell and two types of cuboidal cell can be recog-
nized in the gill tufts of tadpoles of Discoglossus

pictus. Greven [10] reported that the epithelium of
the external gills of larval intrauterine Salamandra
salamandra is composed of pavement cells, basal
cells, pea-shaped cells and ciliated cells. In tele-
osts, the gill is composed of five types of cell:
Squamous pavement, chloride, mucous goblet,
neuroepithelial, non-differentiated [see 2, 11].
The epithelia of gill arches and the gill tufts of the
tadpoles of R. cancrivora consist of four types of
cell: pavement, MR, basal squamous, mucous.
The function of the basal squamous cell is un-
known from the present study. Only a few mucous
cells, which are PAS positive, are present in the
gill tufts, although mucous cells are numerous in
the gill rakers, which are considered to function as
a filter [9].

Pavement cells

The pavement cells form a sheet covering the
epithelia of the gill arches and the gill tufts. In the
present study, two types of pavement cell, the
cuboidal and the squamous, are observed at the
stem and the apical parts of the gill tuft, respec-
tively. Ultrastructural features of the cuboidal
pavement cells and the squamous pavement cells
are identical. The feature of these cells is also
similar to that of the squamous pavement cells
which have been reported in filament epithelium
and secondary lamellae of teleosts [see 2, 12].
Columnar pavement cells have also been reported
in the filament epithelium of teleost [see 2].
Although the exact function of vesicles concent-
rated beneath the apical membrane is unknown,
they may be involved in regulation of the per-
meability of the gill surface as well as the fuzzy
coat of the apical membrane.

Mitochondria-rich cells

Hourdry [3] first reported that the MR cell is
present in the gill tufts of tadpoles of D. pictus.
The presence of MR cells has also been observed
in the integuments of anuran tadpoles and the
integuments and gill epithelium in the larval
urodele [10, 13]. This type of cell is considered to
have a function in transport of ions [3, 13]. In the
present study, eosinophilic cells which seem to be
equivalent to MR cells were observed in the
epithelia of gill arch and gill tufts of tap-water-

630

adapted tadpoles of R. cancrivora. In contrast,
Gordon and Tucker [6] reported no signs of aci-
dophilic cells similar to those found in the gills of
teleosts in the gills of tap-water-, 60% seawater-
and 80% seawater-adapted R. cancrivora tadpoles.
The chloride cells are present in both euryhaline
and stenohaline species of fresh- and seawater
teleosts, and function specifically in relation to
environmental salinity. It has been reported that
the chloride cells of teleosts take up small quanti-
ties of ions in freshwater [14] and secrete large
amounts of salts in seawater [15, 16]. Ultrastruc-
tural characteristics of the chloride cells have been
elucidated in both fresh- and seawater teleosts [see
12]. It has also been reported that freshwater
larval lampreys possess chloride cells which are
rich in apical vesicles and mitochondria without a
distinct tubular system or apical pits [17, 18]. This
type of chloride cell is characteristic of freshwater
teleosts and has been suggested to participate in
the active uptake of ions [17, 18]. Hourdry [3] also
indicated a similar type of MR cell in the gill tufts
of D. pictus tadpoles.

In the present study the features of the MR cell
(Fig. 9) are consistent with those of the MR cells
reported previously in ammocoetes and tadpole
gills [3, 17, 18]. It is suggested, therefore, that the
gills of R. cancrivora tadpoles may play a role in the
osmoregulation of their body fluid by means of
mechanisms similar to those of teleosts in freshwa-
ter. On the other hand, the mechanisms of salt
water adaptation remain to be elucidated in the
tadpole of the present species.

ACKNOWLEDGMENTS

This study was supported in part by a Grant-in-Aid for
Overseas Scientific Research (No. 62041035) from the
Ministry of Education, Science and Culture of Japan.
The authors thank Drs. Akira Chiba and Sumio Yoshie
of The Nippon Dental University at Niigata for their
valuable suggestions.

REFERENCES

1 Witschi, E. (1956) Development of Vertebrates. W.
B. Saunders Co., Philadelphia.

2 Laurent, P. (1984) Gill internal morphology. In
“Fish Physiology”. vol. 10a, Ed. by W. S. Hoar and
D. J. Randall, Academic Press, New York, pp. 73-
183.

3. Hourdry, J. (1974) Etude des branchies <Internes)

10

11

12

13

14

15

16

17

18

M. UcuiyaAMA, H. YOSHIZAWA et al.

puis de leur régression au moment de la métamor-
phose, chez la larve de Discoglossus pictus (Otth),
Amphibien Anoure. J. Microscopie, 20: 165-182.
Dietz, T. H. and Alvarado, R. H. (1974) Na and Cl
transport across gill chamber epithelium of Rana
catesbeiana tadpoles. Am. J. Physiol. , 226: 764-770.
Alcara, A. C. (1962) Breeding behavior and early
development of frogs of Negros, Philippine Islands.
Copeia, 4: 679-726.

Gordon, M. S. and Tucker, V. A. (1965) Osmotic
regulation in the tadpoles of the crab-eating frog
(Rana cancrivora). J. Exp. Biol., 42: 437-445.
Uchiyama, M., Murakami, T. and Yoshizawa, H.
(1990) Notes on the development of the crab-eating
frog, Rana cancrivora. Zool Sci., 7: 73-78.

Taylor, A. C. and Kollros, J. J. (1946) Stages in the
normal development of Rana pipiens larva. Anat.
Rec., 94: 7-23.

Gradwell, N. (1972) Gill irrigation in Rana cates-
beiana. Part I. On the anatomical basis. Can. J.
Zool., 50: 481-499.

Greven, H. (1980) Ultrastructural investigations of
the epidermis and the gill epithelium in the in-
trauterine larvae of Salamandra salamandra (L.)
(Amphibia, Urodela). Z. Mikrosk. Anat. Forsch.,
94: 196-208. |
Karnaky, K. J., Jr. (1980) Ion-secreting epithelia:
chloride cells in the head region of Fundulus hetero- -
clitus. Am. J. Physiol., 238: R185—R198.

Laurent, P. and Dunel, S. (1980) Morphology of gill
epithelia in fish. Am. J. Physiol., 238: R147—R159.
Fox, H. (1986) The skin of Amphibia. In “Biology
of the Integument”. vol. 2 Vertebrates, Ed. by J.
Bereiter-Hahn, A. G. Matoltsy and K. S. Richards,
Springer-Verlag, Berlin/Heidelberg/New York/
Tokyo, pp. 78-135.

Gardaire, E., Avella, M., Isaia, J., Bornancin, M.
and Mayer-Gostan, N. (1985) Estimation of sodium
uptake through the gill of rainbow trout Salmo
gairdneri. Exp. Biol., 44: 181-189.

Foskett, J. K. and Scheffey, C. (1982) The chloride
cell: Definitive identification as the salt-secretory
cell in Teleosts. Science, 215: 164-166.

Karnaky, K. J., Jr., Degnan, K. J., Garretson, L.
and Zadunaisky, J. A. (1984) Identification and
quantification of mitochondria-rich cells in trans-
porting epithelia. Am. J. Physiol., 246: R770-—R775.
Morris, R. and Pickering, A. D. (1975) Ultrastruc-
ture of the presumed ion-transporting cells in the
gills of ammocoete lampreys, Lampetra fluviatilis
(L.) and Lampetra planeri (Bloch). Cell. Tiss. Res.,
163: 327-341.

Youson, J. H. and Freeman, P. A. (1976) Morphol-
ogy of the gills of larval and parasitic adult sea
lamprey, Petromyzon marinus L. J. Morphol., 149:
73-104.

ZOOLOGICAL SCIENCE 7: 631-637 (1990)

© 1990 Zoological Society of Japan

Estradiol-Induced Lymphopenia in the Lizard,
Chalcides ocellatus

ABDEL HAKIM SAAD

Zoology Department, Faculty of Science, Cairo University,
Cairo 12613, Egypt

ABSTRACT—AA single dose (0.025—0.1 mg/g body wt) of estradiol valerianate (E,) was administered to
the female lizard, Chalcides ocellatus, a viviparous skink. Lymphoic cell populations were depleted in
the thymus, peripheral blood, and spleen at about 14 days, but in the bone marrow counts were nearly

normal.

Histologically, the thymic cortex was destroyed, but the medulla was less affected. In

periarteriolar lymphocyte sheath, areas where T lymphocytes predominate in the spleen, the hormone
caused significant depletion. Lymphocyte redistribution may occur between various compartments of
the body as the main effect of hormone induced lymphocyte changes, although the mechanism in lizards

and mammals is not understood.

INTRODUCTION

The regulatory influence of sex steroids, particu-
larly estrogen, on the immune system is not well
recognized [1, 2]. From a number of studies
apparently pharmacologic and physiologic levels of
steroidal and non-steroidal estrogens regulate the
quality and/or the levels of lymphocytes and mono-
cytes, resulting in altered immune function [3, 4].
The immunologic effects associated with the admi-
nistration to rodents of natural estrogens such as
estradiol or synthetic estrogenic compounds such
as diethylestilbestrol (DES) include thymic atro-
phy, suppression of cell-mediated immunity and
depression of natural killer cell activity [5].

Physiologic levels of estrogens provide some
degree of immunoregulation, as most clearly evi-
denced by estrogen treatment which results in the
pathogenesis of autoimmunity [6]. There is also a
correlation between increased serum estradiol
levels during pregnancy and depression of cell-
mediated immunity and lymphopenia [7, 8].
However, the effects of sex steroids on the im-
mune systems of lower vertebrates have hardly
been investigated. In fisth, thymic involution and
decrease in splenic lymphocytes appeared to be

Accepted December 6, 1989
Received Octoboer 2, 1989

related to sexual maturation and gestation [9, 10].
Decreased responses of thymic lymphocytes to
mitogen have been documented in Rana pipiens as
being coincident with increasing ovarian develop-
ment [11]. Moreover, adult orchidectomy in R.
perezi during mating also results in hyperatrophy
of the thymic cortex [12]. Finally seasonal involu-
tion in thymus and spleen is known to coincide
with increasing levels of testosterone in the turtle,
Mauremys caspica [13].

In the present study the author analyzed the
effects of a single dose of estradiol valerianate on
the lymphoid organs of the lizard, Chalcides ocella-
tus, a viviparous skink found in most parts of
Egypt. Recently increased emphasis has been
placed on relationships between the neuroendoc-
rine and immune systems, and has confirmed that
many reptilian species are valid models for such
analysis [12-16].

MATERIALS AND METHODS

Lizards

Sexually mature female lizards, Chalcides ocella-
tus (Scincidae, viviparous, non-hibernator), 30-40
g in weight, were collected from gardens, fields
and arid regions in the vicinity of Cairo (Egypt).
Lizards were maintained in a large terrarium with

632 A. H. SAAD

30 cm depth of sand in a sunny animal room as
described previously [14, 15] and given wood lice
and water ad libitum. The present investigation is
performed in July through September (tempera-
ture ranges of 30-38°C).

Preparation of cell suspensions

Single cell suspensions from thymus and spleen
of adult C. ocellatus were prepared as described in
detail earlier [14, 15]. To obtain peripheral blood
lymphocytes (PBL), lizards were decapitated and
whole blood was collected in heparin solution (100
IU/ml; Sigma, St. Louis, MO, USA). Heparinized
blood was layered over Ficoll-Hypaque (Pharma-
cia Fine Chemicals; density, 1.077; Uppsala,
Sweden). Cell bands at the interface were isolated
and washed twice with cold RPMI-1640 medium
(BIBCO, NY, USA) as described previously [14,
15]. Adult bone marrow cells (BM) were obtained
by aspirating the marrow from femurs and then
preparing the cells by Ficoll-Hypaque as described
above. Cell suspensions contained 3% erythro-
cytes, 77% mononuclear cells and 20% granulo-
cytes as revealed by May-Grunwald-Giemsa stain-
ing [14]. In each experiment, pooled viable cells
(trypan-blue exclusion) were counted by standard
haemocytometer methods. The total number was
divided by the number of lizards in each pool to
determine an average number of cells per lizard.

Sex steroid treatment

Estradiol valerianate (E2; primogyn-depot; Che-
mical Indust. Develop., Cairo, Egypt) was use. A
total of 160 lizards were injected in the summer
(Juny/July), with a single intraperitoneal (i.p.)
dose of E, ranging from 0.025 to 0.1 mg/g body
weight and observed over a 14 day period for body
weight, organ involution and survival times.
Another group of lizards, the controls, received
0.3 ml intraperitoneally of sterile limpid oil. The
reasons for using intact lizards in these short term
studies are as follows: First, there are no unifying
results on the effect of gonadectomy and replace-
ment with physiologic doses of sex hormones [17].
Second, there may be immunologic consequences
resulting from surgical stress [18].

At 3, 7, 10 and 14 days after E> injection, 5-8
E,-injected lizards and 5 controls were decapitated

for isolation of thymic lobes, spleens, PBL and
BM. Lymphocytes from each test organ pool were
separated and assayed for viable cell numbers as
described above. It was necessary to pool cells
from 5-8 lizards, because sex steroid treatment,
particularly at high doses, resulted in thymic atro-
phy and also yielded few cells in other lymphoid
organs compared with those obtained from con-
trols.

Histopathology

Following E, treatment, thymus and spleen
were removed and fixed in phosphate buffered
formalin. After fixation, they were trimmed of
adherent tissue, weighed individually, and embed-
ded in paraffin. Tissues were sectioned at 6 um,
mounted and stained with Harris’ haematoxylin
and eosin [19]. Thymus and spleen were examined
for the degree of differentiation following E> treat-
ment.

RESULTS

Survival of experimental animals

Survival of lizards given 0.025, 0.05 or 0.1 mg/g
body weight E> is shown in Figure 1. All animals
(n=34) given 0.1 mg/g body weight died within 21
days except three, which survived for 28 days. At
0.05 mg/g body weight 8/30 animals died on days 3,
7 and 14, respectively, after starting E> treatments,
whereas all lizards given 0.025 mg E,/g body

E> :0.025 mg/g body weight (n=32)

100
scl Eo :0.05mg/g body weight (n=30)
3 60]
>
5
ne
xs 40
20! E>:0.lmg/g body weight (n=34)
1 ee | l ie
3 7 10 14 21 28
Time (days)
Fic. 1. Survival of lizards given various concentration
of E,.. There was no mortality in control lizards

which received placebo treatment.

Lymphopenia in Lizard 633

weight survived for the full 4 weeks. There was no
mortality in control lizards which received placebo
treatment.

Lymphocyte count

As shown in Figure 2, in the thymus, PBL and
spleen, the greatest reduction in cell number
occurred during the first week of E, administra-
tion. The thymus was the most severely affected
organ in lizards given 0.025 or 0.05 mg E,/g body
weight. Mean cell number in the thymus dropped
to 50% of control values after 3 day of treatment
with 0.025 mg E>/g body weight (p<0.01), while
cell counts in the thymus of lizards treated with
0.05 mg E,/g body weight dropped to 25% of
control values after 3 day (p<0.01). Cell number

PBL

O 3) if lO

14 day

Number of viable lymphocytes (x lO’)
fo)
Nw

PTI PELL TTSETLLLLLLL2

Bone Marrow

O 3 iG ie)

14 day

Fic. 2. Effect of E, on viable lymphocytes numbers of
thymus, peripheral blood (PB), spleen and femoral
bone marrow (BM) in lizards given E> 0.025 (©) or
0.05 (@) mg/g body weight. Hatched areas indicated
corresponding values for controls given i.p. injection
of limpid oil. The sample number of each case is 5-8
(mean+SEM).

remained depleted and reached its maximum
values at 14 day.

There was a similar progressive decrease in
mean number of PBL, although this was not as
marked as for the thymus. Mean cell counts
dropped to 40% of control values after 3 day of
treatment with 0.025 mg/g body weight (p<
0.001), while cell counts for lizards treated with
0.05 mg/g body weight dropped to 30% after 3 day
(p<0.01). This decrease in PBL continued for 14

Number of viable lymphocytes (x 10°)

Lees eee Ol are
10 4 day

Spleen

Number of viable lymphocytes (x 10°)

O 3 L 0) I4 day

634 A. H. SAAD

days after treatment, with no sign of returning to
normal values.

The effect of E, on the number of viable cells in
the spleen was not as marked as in thymus and
peripheral blood (PB); there was a statistically
significant difference between the control group
and those given 0.025 mg E,/g body weight after 3
day of administration, but not thereafter. Admi-
nistration of 0.05 mg E>/g body weight resulted in
a significant reduction in cell number after 3 day (p
<0.01). This decrease in spleen continued for 14
day of treatment with no sign of normalization.

The total number of viable lymphocytes in BM
was not affected by E, treatment. Cell counts
remained constant in all cases and there was little
variation between experimental and _ control
lizards.

Histological changes

Thymus Light microscopic examination of the
thymus of C. ocellatus has been described pre-
viously in details and does not deserve special
attention [19]. Briefly, the thymus of this lizard
showed a clear demarcation between cortex and
medulla. The thymic cortex was composed of
ramified epithelial-reticular cells where lymphoid
cells were housed. In addition to epithelial-
reticular cells and lymphocytes, the thymic medul-
la contained myoid cells, granulocytes and epithe-
lial cysts. The thymus of intact lizards given
placebo oil solution appeared normal over the
whole 14 day period of the experiment (Fig. 3A).
This general histological pattern underwent impor-
tant changes throughout E, administration. The
thymus of lizards given 0.05 mg E>/g body weight
was so atrophied that usually no thymic tissue
could. be harvested for subsequent histology.
Those examined showed severe hypocellularity of
both cortex and medulla, total disintegration of
normal architecture, lack of cortico-medullary dif-
ferentiation and numerous pyknotic cells (data not
Shown). In those given 0.025mg E,/g body
weight, the appearence of the thymus was vari-
able, ranging from virtually normal at 3 day to
severely atrophic at 14 day. Generally, however,
the cortex was markedly reduced and pale, and
cortico-medullary differentiation was indistinct.
The medulla was less affected and contained small

clusters of pyknotic lymphocytes (Fig. 3B). There
was no replacement by adipose tissue. The epithe-
lial structures of the thymus were not affected by
this treatment (Fig. 3C).

oo Cc

Fic. 3. Morphological changes in the thymus of Chal-
cides ocellatus after single i.p. injection of E, (0.025
mg/g body weight). (A) Thymus lobes of control
lizards given placebo injections. Note the various
aspects of the thymic cortex and medulla. (B)
Thymus of lizard 3 day after treatment with E).
Histological evidence of thymic involution appears
in the form of a sharp decrease in the number of
thymocytes from both cortex and medulla. (C)
Thymus of lizard 14 day after E, treatment. The
maximum destructive process can be observed in a
massive degeneration of thymocytes. X 180.

Lymphopenia in Lizard 635

Spleen The structure of the splenic parenchy-
ma is composed of two clearly distinguishable
regions, the red and the white pulp. The white
pulp consists of lymphoid tissue which surrounds
central arterioles, forming the so-called periarter-
iolar lymphocyte sheath (PALS). The PALS con-
tains a regular reticular network densely packed

Slee
=<
Sleees: ee eat eke
:

with lymphocytes and macrophages. Delimiting
the lymphoid follicles of the white pulp, the red
pulp is composed of a system of venous sinuses and
cell cords. No abnormalities were evident in
spleens of lizards given placebo injection (Fig.
4A). Like the thymus, the histological organiza-
tion of the spleen underwent a great modification
during E, treatment. In lizards given 0.05 mg E,/g
body weight and to a lesser extent in those given
0.025 mg E,/g body weight there was little notice-
able differences between spleens. As illustrated in
Figure 4B, spleen involution was maximum at 3
day post E, treatment. The splenic white pulp
consisted of a small collection of loosely packed
lymphocytes. PALS had many areas devoid of
lymphocytes, which then appeared pale. Spleens
exhibited approximately normal cellular architec-
ture on 14 day (Fig. 4C).

DISCUSSION

Most successful experiments demonstrating the
effect of sex steroid hormones have been con-
ducted in vivo despite the fact that most in vitro
experiments use doses much higher than those
used in living animals [5]. In the current examina-
tion of interactions between immune cells and
gonadal hormones in the present lizard, we de-
signed a protocol that avoided excessive doses of
steroids. Because serum hormone levels in E>
injected lizards were similar to the physiological
levels in intact sexually mature lizards during mat-
ing season (Saad and Badir, unpubl. data), it was
anticipated that our results would reflect the
effects of physiologic dose/level of steroids. By

Fic. 4. Morphological changes in the spleen of Chal-
cides ocellatus after single i.p. injection of E> (0.025
mg/g body weight). (A) Spleen of control lizards
given placebo injections. White pulp is extensively
developed with densely packed lymphocytes aggre-
gated around central arteriole. The red pulp is well
represented by cords of reticular cells enclosing
vascular sinuses. (B) Spleen of lizard 3 day after E,
treatment. Density of lymphocytes is clearly dimi-
nished. (C) Spleen of lizard 14 day after E, treat-
ment. Splenic lymphoid aggregates increase in size
and lymphocyte density, tending to become con-
fluent with each other and obscuring the red pulp.
x 180.

4

636 A. H. SAAD

using intact lizards with naturally low levels of
steroid during June-August (a time of the year
when the immune system is maximally developed),
we hoped to eliminate the major source of gonadal
hormones.

Adminstration of E>, for even a short duration,
had significant effects on the immune system of C.
ocellatus as evidenced by quantitative and qualita-
tive changes in various lymphoid organs. The most
interesting feature in these results on in vivo action
on lymphoid elements in C. ocellatus is one of
different depletions observed between the thymus,
spleen and PBL. The thymus in this lizard is the
most sensitive organ and is almost destroyed by E,
treatment. The present histological examination
clearly showed that cortical lymphocytes were
more susceptible to the destructive influence of E>
than medullary thymocytes. Sex steroids reduce
mitotic activity in the thymic cortex in a similar
manner as corticosteroids [20], and thus suppress
the proliferation of the most actively dividing short
lived lymphocytes.

There is no report as to whether reptilian cortic-
al and medullary thymocytes differ in terms of
maturation and differentiation. In fact, recent
studies in C. ocellatus demonstrated atrophy fol-
lowing in vivo corticosteroid injection caused a
remarkable and almost total depletion of medul-
lary cells which accompanies destruction of the
cortex [14]. From this information which was
confirmed by cell surface marker analyses [14],
lizard medullary thymocytes are apparently not
resistant to in vivo corticosteroid treatment. Dur-
ing treatment with E>, the medulla became nearly
depleted of lymphocytes, too, resulting in a pro-
nounced atrophy. Similar results have been
observed in the thymus of gonadectomized rats
[21], in the developing fetal thymus of guinea pigs
[22], and in mice [23] treated with estrogen. These
results could be attributed to the high circulating
levels of estrogen which occur after E> injection.
Histologically, this involution is characterized by a
decreased corticomedullary ratio, and thymocyte
pyknosis [23]. Murine thymic involution is reversi-
ble and normal architecture is re-established after
withdrawal of estrogen [22]. The mechanism of
this bi-phasic pattern of thymus repopulation re-
mains speculative.

Of the other organs investigated, I note that the
spleen, which seems less sensitive to E> than the
thymus in different mammalian species, is affected
to a remarkable degree. All authors agree in
pointing out that the E> effects are generally less
severe on peripheral lymphoid organs, especially
the spleen [23]. Splenic weight was reduced in
female mice injected with E, or DES for the first
days after birth without causing striking histologic-
al changes [24]. Contrary to this, DES treatment
of adult animals increases 3H-thymidine incor-
poration in the spleen, with most labeled cells
localized in the subcapsular region as well as in the
red pulp and marginal zones [27]. Spleen enlarge-
ment also occurred in adult mice following E>
implants or injections with DES (2 or 8 mg/kg
body weight) for 5 consecutive days [24]. The
decrease in total number of splenic and PB lym-
phocytes is another of the numerous actions of E>
on leukocytes.

The present results confirm previous ones in
mice and rats which demonstrated that E> de-
creases the number of splenic and circulating lym-
phocytes [23, 24]. However, Thompson et al. [25] °
found no effect on numbers of blood lymphocytes,
while a four-fold increase in circulating lympho-
cytes was observed by others [26, 27]. The slow
rate of recovery is more consistent with regenera-
tion than with reappearance after migration. The
fastest rate of recovery was in the spleen, but was
probably due to considerable proliferation in situ
as histological evidence indicated (see Figs. 2 and
4). Therefore, I suggest the possibility of lympho-
cyte redistribution among various compartments
of the body as the main cause of E, induced
lymphoid tissue involution in C. ocellatus. Never-
theless, the exact mecahnisms of action of sex
hormones are not well established even in the
mammalian system which has been more widely
studied.

REFERENCES

1 Thompson, J. S., Crawford, M. K., Reilly, R. W.
and Severson, C. D. (1967) The effects of estrogenic
hormones on immune responses in normal and
irradiated mice. J. Immunol., 98: 331-335.

2 Ablin, R. J. (1981) Modulatory effects of estrogen
on immunologic responsiveness. Am. J. Rep. Im-

10

il

12

13

14

Lymphopenia in Lizard

munol., 1: 206-216.

Talal, N. and Roubinian, J. R. (1978) Sex hormones
and autoimmunity. In “Genetic Control of Autoim-
mune Disease”. Ed. by N. R. Rose, P. E. Bigazzi
and N. L. Warner, Elsevier/North Holland, New
York, pp. 163-185.

Bick» 2 H.. Turker, A. N., White; K. L. and
Hosapple, M. P. (1984) Effects of subchronic expo-
sure to diethylstilbestrol on humoral immune func-
tion in adult female (C3B6) F1 mice. Immunophar-
macology, 7: 27-35.

Grossman, C. J. (1984) Regulation of the immune
system by sex steroids. Endocrin. Rev., 5: 435-455.
Ansar Ahmed S., Penhole, W. J. and Talal, N.
(1985) Sex hormones, immune responses, and auto-
immune diseases. Mechanism of sex hormone ac-
tion. Am. J. Pathol., 121: 531-554.

Hamilton, M. S. and Hellstrom, I. (1977) Altered
immune responses in pregnant mice. Transplanta-
tion, 23: 423-430.

Mathur, S., Mathur, R. S., Goust, J. M., William-
son, H. O. and Fudenburg, H. H. (1979) Cyclic
variations in white cell subpopulations in the human
menstrual cycle: correlations with progesterone and
estradiol. Clin. Immunol. Immunopathol., 13: 256-
262.

Honma, Y. and Tamura, E. (1984) Histological
changes in the limphoid system of fish with respect
to age, seasonal and endocrine changes. Dev.
Comp. Immunol., Suppl. 3: 239-244.

Honma, Y., Tamura, E. and Chiba, A. (1985)
Seasonal thymic activity in relation to the reproduc-
tive endocrine activity of the viviparous surfperch,
Ditrema temminicki. In “Current Trends in Compa-
rative Endocrinology”. Ed. by B. Lofts and W. N.
Holmes, Hong Kong, pp. 223-224.

Wright, R. K. and Cooper, E. L. (1980) Tempera-
ture and immunological memory in anuran amphi-
bians. In “Phylogeny of Immunological Memory”.
Ed. by M. J. Manning, Elsevier/North-Holland
Biomedical Press, Amsterdam, pp. 155-163.
Zapata, A., Garrido, E., Leceta, J. and Gomariz,
R. P. (1983) Relationships between neuroendocrine
and immune systems in amphibians and reptiles.
Dev. Comp. Immunol., 7: 771-774.

Leceta, J. and Zapata, A. (1985) Seasonal changes
in the thymus and spleen of the turtle Mauremys
caspica. A morphometrical light microscopical
study. Dev. Comp. Immunol., 9: 653-668.

Saad, A-H., El Ridi, R., El Deeb, S. and Soliman,
A. W. (1987) Corticosteroids and immune system in

15

16

17

18

19

20

21

vps

23

24

25

26

Dy)

637

the lizard, Chalcides ocellatus. In “Developmental
and Comparative Immunology”. Ed by E. L. Coop-
er, C. Langlet, and J. Bierne, Alan R. Liss, New
York, pp. 141-151.

Saad, A-H. and El Ridi, R. (1988) Endogenous
corticosteroids mediate seasonal cyclic changes in
immunity of lizards. Immunobiology, 177: 390-403.

Saad, A-H and Shoukrey, N. (1988) Sexual
dimoxphism on the immune responses of the snake,
Psammophis sibilans. Immnobiology, 177: 404-419.
Ansar Ahmed, S., Dauphinee, M. J. and Talal, N.
(1985) Effects of short-term administration of sex
hormones on normal and autoimmune mice. J.
Immunol., 134: 204-209.

Kinnaert, P., Mahieu, A. and Van Geetruyoden, N.
(1978) Stimulation of antibody synthesis induced by
Surgical trauma in rats. Clin. Exp. Immunol., 32:
243-247.

Hussein, M. F., Badir, N., El Ridi, R. and Akef, M.
(1978) Differential effect of seasonal variations on
the lymphoid tissue of the lizard, Chalcides ocella-
tus. Dev. Comp. Immnol., 2: 297-310.

Dougherty, F. F. (1952) Effects of hormones on
lymphatic tissue. Physiol. Rev., 32: 379-401.
Sobhon, P. and Jirasathan, C. (1974) Effect of sex
hormones on the thymus and lymphoid tissue of
ovariectomized rats. Acta Anat., 89: 211-225.
Gulino, A., Screpanti, I. and Pasqualini, J. R.
(1983) Estrogen and antiestrogen effect on different
lymphoid cell populations in the developing fetal
thymus of guinea pig. Endocrinology, 113: 1754-
1760.

Thompson, E. A. (1981) The effects of estradiol
upon the thymus of the sexually immature female
mouse. J. Steroid Biochem., 14: 164-174.

Kalland, T., Forsberg, T.M. and Foresberg, A. G.
(1978) Effect of estrogen and corticosterone on the
lymphoid system in neonatal mice. Exp. Mol.
Pathol., 28: 76-95.

Thompson, J. S., Severson, C. D. and Reilly, R. W.
(1969) Autoradiographic study of the effect of
estradiol and irradiation on nucleic acid metabolism
of the thymus and lymph node of mice. Radiat.
Res., 40: 46-62.

Fox, L. E. (1961) Actions of estrogens on blood and
bone marrow. J. Pharmacol. Sci., 50: 436-438.
Boorman, G. A., Luster, M. I., Dean, J. H. and
Wilson, R. E. (1980) The effect of adult exposure to
diethylstilbesterol in the mouse on macrophage
functions and numbers. J. Reticuloendothel. Soc.,
28: 547-560.

eink

ee Mebamela a Sa ew
i Sera sdhiva tarcianien atetion ablayasteonhee
‘oo ae wae ciklasn cuter eam a eR

Sie ape moenet): Lit er ay eae aC

A So" ysl et hx itaakasseay nk comeseguret ite muidaupeilrs
aR ely RT Ee inicdeveeraactl Ania Mig aater a a it
oe veut bas, .b, Ne ssoalgund) ne shorste,, wendy.)
; . . re : ett! ,
' ; r: ~ at) meh at idsherytus ayes woe 40 a. vents x i (eae
; fast oY] b> a] vai Daren weiss yr? ehy Nit ae Laat
abi Oe raat, ALOR. HOS Do Mice cet ns
; yy es Mize egies ad ‘Fiat Xe i bok +2 a bitte Ty ak fseneA BE
e il whe smoiaicniacviseecr: (heii te jeretet) hecaerard eons . 3
ed = : ines : is r res t. - > ey
a ws “ 4 ME hae oh, hee 45 WEY An ithe H ut tad ne lh 4 ny si8 sy his
erie aj eeamal Sali bart oy
= is ital Lae VE alge hs = ef "RG eo uae
, 7 ree) oP x + ie c 7 Ce he oe - ene tk iad ) aU)
ACLANCHIRY Theale hoe, 2 =
4 Oi, een Atris ee) beet ae
Ty f 4 ‘s - Fl a vi weer ¥ : a) j ,
"ga neeMEied tee see OE) 8 Meoaeeiat be KI one pip ier lacomdia te Th
‘7 ¢ Sw ‘va ' é § Pa ’ +4 iy ak te e) Ry \. ee ‘
(MARTE OE Aa ids. ong orem VASA et A LUNs (00,
2at0 335 tid fe\Ul) SS eRe bare: dodged US Sanisnogut
4 as ae si ar Broce Wh Wel a
: : ino ks,’ & bide BETTER aeila: “ERD oe OEARATEY CURES F ~~. :
ccc rrs ee ink er hearse ie ye ns enclose ee
ki sah Ee, S45 if 7 7 ’
pete ates: “a ae ve ge ees i. fae ss) rye + << a4 Hen inv! et he fhietin:
ns ey id oe ia | ba oveairsiine a had iy). yee) eat te Ae na tral ot eboney
> ‘ bs te 1e4 ; fe r 1231 Be he E, . tla >
é x fy ye i ‘ re oF a
I # 4 $ : * : i j ‘
i ~ f J Chee
es ’ ‘ : {REIN
e Bits . Bi . ” fe vey
‘4 ‘ if ghee

ae

ZOOLOGICAL SCIENCE 7: 639-648 (1990)

Changes in the Rigidity of the Hamster Egg during
Meiotic Maturation and after Fertilization

CHENG-HSIUNG YANG and Ryuzo YANAGIMACHI

Department of Anatomy and Reproductive Biology, University of
Hawaii School of Medicine, Honolulu, Hawaii, U.S.A.

ABSTRACT—When mature unfertilized hamster eggs were freed from their zonae pellucidae and
centrifuged at 9,500 x g, each egg elongated and then separated into light and heavy halves in about 10
min. The rigidity of eggs at various stages of meiotic maturation and preimplantation development was
examined by determining the minimum time needed for complete separation of all eggs in a group into
halves or fragments.

Eggs in the germinal vesicle stage were very rigid. They did not separate into halves even after 40 min
of centrifugation at 16,000xg. This rigidity appeared to be a property of the egg cortex rather than the
endoplasm. The rigidity of the egg diminished sharply with the progression of meiotic maturation,
reaching a minimum at metaphase of second meiosis, then increasing progressively after fertilization.
Cytochalasin D reduced the rigidity of the egg regardless of the stage of meiotic maturation and
development, indicating that the actin-based cytoskeleton in the cortical region is largely responsible for

© 1990 Zoological Society of Japan

the rigidity of the egg.

INTRODUCTION

When sea urchin and starfish eggs with very thin
vitelline envelopes are centrifuged in a sucrose
solution (or sucrose-seawater mixture) of approx-
imately same density as the eggs, each egg elon-
gates and then separates into a light and a heavy
halves [1, 2]. Mammalian eggs with thick though
envelopes (zonae pellucidae) do not separate into
halves under the same conditions unless zonae
have been removed prior to centrifugation [3, 4].
We have found that zona-free hamster eggs col-
lected from the oviduct shortly after ovulation
separate by centrifugation faster than those iso-
lated from the ovary immediately before ovulation
[3]. This seems to suggest that the former is less
“rigid” than the latter. Thus, a reduction in the
rigidity of the egg could be a feature of meiotic
maturation. This prompted us to investigate how
the “rigidity” of hamster egg changes during meio-
tic maturation, fertilization and preimplantation
development.

Accepted September 30, 1989
Received August 28, 1989

MATERIALS AND METHODS

Medium and reagents

The medium M2 [5] containing 0.4% bovine
serum albumin fraction V (BSA) and 50 g/ml
gentamicin sulfate was used throughout this study.
All inorganic salts were purchased from either
Mallinckrodt Chemical Co. (St. Louis, MO) or
Sigma Chemical Co. (St. Louis, MO). The sources
of the other compounds were as follows: D-
glucose, Na-pyruvate, Na-lactate, HEPES, genta-
micin sulfate, cytochalasin D (CD), dimethy] sul-
foxide (DMSO), Percoll, and human chorionic
gonadotropin (HCG) (Sigma); BSA and pregnant
mare serum gonadotropin (PMSG) and colcemid
(Calbiochem, La Jolla, CA); bovine pancreatic
trypsin and bovine testicualr hyaluronidase (ICN
Biochemicals, Cleveland, OH); _lysolecithin
(Avanti Polar Lipids, Birmingham, AL); N-(7-
nitrobenz-2-oxa-1, 3-diazol-4-yl) phallacidin or
NBD-phallacidin (Molecular Probes Inc., Eugene,
OR). Stock solutions of cytochalasin D (1 mM),
colcemid (0.27 mM) and NBD-Ph (100 units/ml)
were prepared with 100% DMSO, 0.9% NaCl and
100% methanol, respectively.

640 C.-H. YANG AND R. YANAGIMACHI

Collection of eggs at various stages of maturation
and preimplantation development

Golden hamsters were raised and maintained in
an air-conditioned room with controlled light con-
ditions (light, 05:00-19:00hr; dark, 19:00-
05:00hr). Under these conditions, mature
females came into behavioral estrus around 18 : 00
hr of every fourth day (day 4) and ovulated early
the next morning (01 : 00-02 : 00 hr of day 1). The
age of females used in this study was 2-4 months
and that of males was 4-6 months.

Females were each injected intramuscularly with
30 IU PMSG in the morning of day 1 followed by
injection of 30 IU HCG in the afternoon or even-
ing of day 3. Ovulation took place between 13 and
14 hr after HCG injection. The number of eggs
ovulated was about 50 per female. Ovarian eggs
with “intact” germinal vesicles were collected 1 hr
after HCG injection by puncturing enlarged folli-
cles. Ovarian eggs in the process of germinal
vesicle breakdown and those at metaphase I,
anaphase to telophase I and metaphase II of
meiotic division were collected in the same manner
at 3.5, 7, 10 and 13hr after HCG injection,
respectively. Ovulated eggs were collected 17 hr
after HCG injection by flushing them out of the
oviduct. To obtain fertilized eggs at various stages
of development, estrous females which did not
receive any hormones were allowed to mate in the
evening of day 4. Fertilized eggs, at pronuclear
stage of development, were flushed from the ovi-
duct around 10: 00 and 16: 00 hr of next day (day 1
of pregnancy). Eggs undergoing the first cleavage
were obtained in the eveining (18:00 hr) of day 1
of pregnancy. Eggs 2-, 4- and 8-cell stages were
collected 10:00 hr of day 2, 23: 00 hr of day 2 and
09 : 00 hr of day 3, respectively.

Prior to centrifugation, all eggs were freed from
both cumulus oophorus and zona pellucida. All
Ovarian eggs, except those at metaphase II, were
first treated for 5 min with 0.2% hyaluronidase in
M2 to “loosen” their cumuli and then drawn in and
out of a small bore pipette to disperse the remain-
ing cumulus cells. No pipetting was necessary for
mature ovarian and oviductal eggs which had
reached metaphase II. A 5 min treatment with
0.1% hyaluronidase in M2 dispersed the cumulus

cells from these eggs. Most pronuclear eggs and all
developing eggs were already free of cumulus cells.
All of cumulus-free eggs thus obtained were tre-
ated for 1-2 min with 0.1% trypsin in M2 to
dissolve their zonae pellucida. Zona-free eggs
were rinsed thoroughly and kept in M2 for no
longer than 20 min before they were centrifuged.

Examination of egg “rigidity” by centrifugation

The rigidity of an egg was assessed by determin-
ing the minimum time needed for complete separa-
tion of all eggs or blastomeres into heavy and light
halves or fragments. Centrifugation was carried
out as described by Yang and Yanagimachi [3].
Briefly, zona-free eggs (usually, 10 to 15) were
loaded onto the top layer of a Percoll gradient
(7.5%, 30% and 45% Percoll in M2) and centri-
fuged at 9,500 xg (in a few instances, at 16,000 x
g). During centrifugation, the eggs came to rest in
the 30% Percoll layer. Each egg become elon-
gated and then, in time, separated into a heavy and
a light half. The time needed for complete separa-
tion of all the eggs in a particular group into halves
or fragements was determined by varying the
duration of centrifugation. First, several experi-
ments were carried out to determine the approxi-
mate time needed for complete separation of all
eggs, then the minimum time was determined
more acculately by increasing or decreasing the
duration of centrifugation 0.25 to 1 min every
time. At least 6 series of experiments were per-
formed to determine mean+S.D. of the minimum
time necessary for separation of all eggs. In some
experiments, eggs were treated for 30 min at 25°C
with 10—40 uM cytochalasin D, 1-4% DMSO, or
25 uM colcemid (all in M2) prior to centrifugation.

Egg rigidity was also examined by drawing a
small region of the cortex into a micropipette
attached to a micromanipulator. The diameter of
the egg was approximately 80 “m and the aperture
of the micropipette was about 5 um.

Distrubution of F-actin in eggs

The distribution of F-actin was examined using
an F-actin-specific fluorescent probe, NBD-
phallacidin (abbr. NBD-ph), according to the pro-
cedure described by Battaglia and Gaddum-Rosse
[6]. Briefly, eggs were fixed 10-15 min with 2%

Rigidity of Hamster Egg 641

Fic. 1. Separation of an unfertilized oviductal egg into halves by centrifugation (9,500<g). A: before centrifuga-
tion. B, B’: light and heavy halves, after 10 min of centrifugation (an arrow in Fig. 1B indicates the direction of
centrifugation). C: light (1) and heavy (h) halves which have “rounded-up” after being removed from centrifugal
force. Magnification, 270.

Fic. 2. Animmature egg at the germinal vesicle stage, centrifuged for 40 min at 9,500 x g; the egg elongated, but did
not separate into halves. Before (A) and after (B) compression between a slide and coverslip. The germinal
vesicle (gv) is clearly visible at the border between clear and granular regions of the compressed egg, B. An arrow
in 2A indicates the direction of centrifugation. Magnification, x 380.

a

642 C.-H. YANG AND R. YANAGIMACHI

paraformaldehyde in Dulbecco’s phosphate buf-
fered saline (PBS), rinsed with PBS, and stained
for 15 min with NBD-ph (5 units/ml) in PBS con-
taining 0.1 mg/ml lysolecithin. After rinsing in
PBS, eggs were mounted between a slide and
coverslip and examined with a fluorescence micro-
scope.

RESULTS

Rigidity of eggs at various stages of maturation and
development

When unfertilized oviductal eggs were centri-
fuged at 9,500 Xg, each egg separated into havles
by 10 min after the start of centrifugation (Fig.
1A-C) [3, 4]. Ovarian eggs at the germinal vesicle
(GV) stage did not separate, even after 40 min of
centrifugation at 9,500 xg (Fig. 2A-B). An even
stronger centrifugation (16,000xg for 40 min)
could not separate these eggs, indicating that ova-
rian eggs at GV stage were very rigid. Soon after
GV breakdown, eggs were separated by 32 min of
centrifugation at 9,500 xg. Apparently, the rigid-
ity of the egg diminishes during or after GV
breakdown. Changes in the rigidity of the egg
during maturation, fertilization and development,
as determined by the minimum time needed for
separation of all eggs or blastomeres into halves or
fragments, are shown in Figure 3. It can be seen
that egg rigidity diminishes during or after GV
breakdown and increases after the first cleavage.
It should be noted that the mature, unfertilized
eggs collected from the oviduct are the least rigid.
Figures 4 and 5 show fertilized eggs at the pronuc-
lear and 2-cell stage which were elongated and
separated by centrifugation at 9,500 xg.

The rigidity of eggs or blastomeres was not
affected by 30 min pretreatment with 25-40 uM
colcemid or 1-4% DMSO. On the contrary,
cytochalasin D (CD) diminished egg and blasto-
mere rigidity significantly. Regardless of the stage
of meiotic maturation and development, all the
eggs or blastomeres treated with 10-40 uM CD for
30 min separated into halves within 3-12 min of
centrifugation at 9,500xg (Fig. 3). It should be
noted that the reduction in the rigidity by CD is
more marked in GV eggs than in developing eggs.

40 pM Colcemid

iL
a men, ; o
Ve cD
Ms
< 20
E
-
es. x ale ie I
a 10yM CD ee Saal
/
J i
40 pM CD Je Ny /
A ee ee ]
= Ey BS

asad
ry
uelieao zw
“JONPIAO ZW
Ajsea -uol,
938] UOld.
“ACB|D ISL
ek
Thre) 2
1998

Fic. 3. Changes in the egg rigidity during meiotic
maturation and after fertilization as revealed by
centrifugation. Abscissa: stages of egg maturation
and development. Ordinate: minimum time (min)
needed for separating all eggs (or blastomeres) in a
group into halves by centrifugation at 9,500xg,
mean+S.D. It can be seen that the rigidity of intact
egg (—@—) diminishes with the progression of
meiotic maturation, reaches a minimum with the
completion of maturation, and then increases after
fertilization. 10-40 «~M CD (—©O—) reduces egg
rigidity regardless of the stages of egg maturation
and development, whereas 10-40 uM colcemid (*)
and 1-4% DMSO (() do not.

GV, germinal vesicle stage; GVBD, during and
immediately after germinal vesicle breakdown; M,,
T,, M2, metaphase and telophase of the Ist and 2nd
meiosis; Ovarian and oviduct., mature ovarian and
oviducal eggs at metaphase of the 2nd meiosis; pron,
pronuclear stage. /st cleav., during the Ist cleavage.

Figure 6 shows GV eggs separated by centrifuga-
tion after CD-treatment.

Results of our micromanipulation experiments
suggested that the cortex of the GV egg is much
more “rigid” than that of the mature egg. When a
GV egg was held by a micropipette and sucked by
applying a steadily increasing negative pressure,
the egg resisted the pressure until the tip of

Rigidity of Hamster Egg 643

Fic. 4. Successive stages of separation of a pronuclear egg into halves. A: before centrifugation. B, C: 7 and 11 min
after start of centrifugation at 9,500 x g (an arrow in Fig. 4B indicates the direction of centrifugation). D, E: light
and heavy halves; in some instances, an egg can separate into three or more fragments. Magnification, x 330
except for the main figure of C which is X65.

Fic. 5. Successive stages of separation of the blastomeres of a 2-cell egg into light and heavy halves. A: before
centrifugation. B, D: 5-20 min after the start of centrifugation at 9,500 x g; only one of the blastomeres is shown
in Fig. 5C, D. E-F: light and heavy halves before “rounding-up”. Magnification, x 200.

644 C.-H. YANG AND R. YANAGIMACHI

Fic. 6. Separation of a cytochalasin-treated germinal
vesicle egg into halves. A: before and after separa-
tion into two, an arrow indicates the direction of
centrifugation. B, C: separated into light (B) and
heavy (C) halves. The cord-like portion of the light
half may or may not be integrated into the main
body of the light half during “rounding-up” process.
gv, germinal vesicle. Magnification, x 360.

evagination broke abruptly (Fig. 7A-D). The con-
tents of the egg (endoplasm) did not appear to be
very viscous; it mingled with the medium instan-
taneously (Fig. 7E). In contrast, mature oviducal
egg hardly resisted the negative pressure. It was
easily drawn into the micropipette even by a very
weak negative pressure (Fig. 8A-D). The entire
egg could be drawn into the tube, without being
ruptured. When the portion of an egg which had
been drawn into the micropipette (Fig. 8E) was

then pushed out, it resembled a soft sausage (Fig.
8 F-I). GV eggs treated with CD behaved like
mature oviductal eggs.

Distribution of F-actin in eggs at various stages of
meiotic maturation and development

When unfertilized oviductal eggs were stained
with NBD-ph and then examined with a fluoresc-
ence microscope, the entire cortex exhibited
strong fluorescence, indicating the presence of
F-actin in the egg cortex. The fluorescence was
particularly strong in the area of cortex above the
metaphase II chromosomes (Fig. 9A), supporting
previous reports that this area of cortex is rich in
F-actin [7-9]. There were differences in the
fluorescence pattern between CD-treated and un-
treated (control) eggs. The fluorescence of CD-
treated eggs was strong during the first 5-10 sec of
exposure to UV light (Fig. 9B), but it faded quick-
ly thereafter. The fluorescence of control eggs
faded with time, but not as fast as that of CD-
treated eggs. The cortical fluorescence of control
eggs was evenly distributed, whereas that of tre-
ated egg was patchy (Fig. 9C).

When a GV egg and a mature unfertilized egg
were stained with NBD-ph simultaneously in the
same dish and examined side-by-side, the former
always displayed stronger fluorescence (Fig. 9D).
A 2-cell egg and a mature egg exhibited similar
cortical fluorescence (Fig. 9E). Perinuclear fluo-
rescence was evident in the 2-cell egg.

DISCUSSION

In this study we estimated the rigidity of hamster
eggs by determining the minimum time needed for
complete separation of all eggs in a group into
halves or fragments. The rigidity we measured is
the sum of various mechanical properties (e.g., the
yeild point, elasticity and viscosity) of the egg
cortex and endoplasm. The results of our experi-

Fic. 7. Reaction of a germinal vesicle egg to suction. Arrows in the micropipette indicate directions of fluid flow.
The egg resists to suction force (A-C) until the tip of evagination breaks abruptly and the broken egg is drawn into
the pipette; the contents of broken egg are pushed out of the pipette as granular and amorphous materials (E).

Fic. 8. Reaction of mature oviducal eggs to suction. Arrows in micropipette indicate directions of fluid flow. A-D:
an egg is easily drawn into the pipette (the entire egg can be drawn into the pipette without being ruptured). E-I:
when the portion of an egg, which has been drawn into the pipette, is pushed out, it resembles a soft sausage.
Two eggs are shown here, one of which (b) being drawn into the pipette deeper than the other (a).

Rigidity of Hamster Egg

645

646 C.-H. YANG AND R. YANAGIMACHI

Fic. 9. NBD-pallacidine fluorescence of eggs. A: a mature oviducal egg, showing strong fluorescence of the cortex,
particularly the area where metaphase chromosomes are located, the top of the figure. B-C, side and surface view
of a mature oviducal egg treated with cytochalasin D; note the endoplasmic fluorescence (B) and spotty —
fluorescence of the egg surface (C). D: mature oviducal egg (m) and germinal vesicle egg (gv) side-by-side; note
that cortical fluorescence is stronger in the germinal vesicle egg. E: mature oviducal egg (m) and 2-cell egg (t)
side-by-side; in this micrograph, intensity of cortical fluorescence appears same in (m) and (t), but in most cases,
(t) showed stronger fluorescence than (m) within 5-10 sec of exposure to UV light. Magnification, x 270.

ments indicate that hamster eggs at the GV stage
are much more rigid than those undergoing meio-
tic maturation. In this respect, hamster eggs seem
to be similar to starfish eggs. Nakamura and
Hiramoto [10] and Nemoto et al. [11] measured the
rigidity of starfish egg by sucking a small portion of
the egg into a micropipette or compressing the egg
under a coverslip. They found that the rigidity of
the egg diminishes progressively during and after
GV breakdown. In both the starfish [11] and the
hamster (this study), the rigidity of eggs is reduced
significantly by cytochalasin, suggesting that these
properties of both starfish and hamster eggs are
largely dependent on the actin-based cytoskeleton.

According to Kaplan et al. [12], cytoskeletal
proteins, including actin, are synthesized and
accumulated during the growth of the egg. Actin
synthesis declines during meiosis and is insigni-
ficant in mature unfertilized eggs as well as recent-
ly fertilized eggs [13]. Our study, using NBD-ph as

a F-actin probe, has revealed that the density of
F-actin in the cortex and endoplasm of hamster egg
changes, but not very dramatically during meiotic
maturation and after fertilization. It may be the
manner of interaction of F-actin with other cytos-
keletal proteins (e.g., filamin and alpha-actinin),
not the concentration of actin per se, that changes
significantly during meiotic maturation and after
fertilization. According to Wassarman [14], the
distribution of synthesis (percentage of total)
among alpha, beta and gamma actin in growing
mouse oocyte is approximately 5, 80 and 15%,
respectively, whereas in the mature unfertilized
egg the distribution changes dramatically to 20, 20
and 60% for alpha, beta and gamma actin, respec-
tively. Thus, changes in the type of actin synthe-
tized during meiotic maturation and after fertiliza-
tion may also be responsible for the changes in the
rigidity of the egg cortex.

In the sea urchin, Temmnopleurus toreumaticus,

Rigidity of Hamster Egg 647

the elasticity of the egg cortex increases and
reaches a maximum about 1.5 min after fertiliza-
tion. It then decreases to a minimum about 4 min
after fertilization, follwed by a gradual increase
until the onset of first cleavage [15]. We were
unable to detect a rise and fall of the rigidity in the
hamster egg during fertilization and first cleavage.
Only a slow, gradual increase in the rigidity was
noted (Fig. 3). It is possible that we did not detect
short-term changes in the rigidity because we did
not examined eggs continuously during the 14 hr
between fertilization and first cleavage.

In the mouse, actin synthesis is low in both
unfertilized and 2-cell eggs. It becomes higher
after the egg reaches the 4- to 8-cell stage [16-19].
In this respect, it is somewhat surprising that the
2-cell hamster egg is considerably more rigid than
the unfertilized egg (Fig. 3). The density of cortic-
al actin in the 2-cell egg appeared to be equal to or
only slightly higher than that in the unfertilized egg
(Fig. 9E). The striking difference between unfer-
tilized and 2-cell eggs with respect to the rigidity
may not be explained in term of the density and
distribution of actin alone. As stated previously,
the rigidity of an egg (and blastomere) may be
influenced greatly by other cytoskeletal proteins.
The manner of interaction of F-actin with other
cytoskeleton proteins could be a prime factor
controlling the rigidity of an egg (or blastomere).
One may argue that the rigidity of the blastomere
remains unchanged during development (cleav-
ages) of the egg. The increasing resistance to the
centirifugal force may be due to the reduction in
size of the blastomere with the advancement of
cleavage. As the cell becomes smaller, it may
become increasingly resistant to centrifugal force
due to the so-called “scale effect”. If this is the
case, then smaller blastomeres of 4- and 8-cell eggs
should be more resistant to centrifugal force than
larger blastomeres of 2-cell egg even if they are
treated with CD. Apparently, this is not the case
(cf. Fig. 3). Therefore, a steady increase in the
rigidity of the blastomeres during development
must be largely due to the changes in the cytos-
keletal system in the blastomere.

ACKNOWLEDGMENTS

This study was supported by a grant from the U.S.
Public Health Service (HD-03402). We thank Mr. T.
Timothy Smith and Mrs. Hiroko Yanagimachi for their
assistance in the preparation of the manuscript.

REFERENCES

1 Harvey, E. B. (1936) Parthenogenetic merogony or
cleavage without nuclei of Arbacia punctulata. Biol.
Bull., 71: 101-121.

2 Maruyama, Y. K., Yamamoto, K. and Mita-
Moyazaki, I. (1986) Mainpulation methods for
analyzing embryogenesis. In “Methods in Cell Biol-
ogy. Vol. 27, Echinoderm Gametes and Embryos”.
Ed. by T. E. Schroeder, Academic Press, Orlando,
pp. 325-344.

3 Yang, C. H. and Yanagimachi, R. (1989) Differ-
ences between mature ovarian and _ oviductal
oocytes: A study using the golden hamster. Human
Reprod., 4: 63-71.

4 Yang, C. H., Yanagimachi, R. and Yanagimachi,
H. (1989) Morphology and Fertilizability of zona-
free hamster oocytes separated into halves and
quarters by centrifugation. Biol. Reprod., (In press)

5 Quinn, P., Barros, C. and Whittingham, D. G.
(1982) Preservation of hamster oocytes to assay the
fertilizing capacity of human spermatozoa. J. Fert.
Steril., 66: 161-168.

6 Battaglia, D. E. and Gaddum-Rosse, P. (1986) The
distribution of polymerized actin in the rat egg and
its sensitivity to cytochalasin B during fertilization.
J. Exp. Zool., 237: 97-105.

7 Maro, B., Johnson, M. H., Pickering, S. J. and
Flach, G. (1984) Changes in actin distribution
during fertilization of the mouse egg. J. Embryo.
Exp. Morph., 81: 211-237.

8 Maro, B., Howlett, S. K. and Houlinston, E. (1986)
Cytoskeletal dynamics in the mouse egg. J. Cell Sci.,
Suppl. 5: 343-359.

9 Longo, F. J. (1987) Actin-plasma membrane asso-
ciation in mouse eggs and oocytes. J. Exp. Zool.,
243: 299-309.

10 Nakamura, S. and Hiramoto, Y. (1978) Mechanical
properties of the cell surface in starfish eggs. De-
velop. Growth Diff., 20: 317-327.

11 Nemoto, S., Yoneda, M. and Uemura, I. (1980)
Marked decrease in the rigidity of starfish oocytes
induced by 1-methyladenine. Develop. Growth
Diti 222315 —325:

12 Kaplan, G., Abreu, S. L. and Bachvarova, R.
(1982) rRNA accumulation and protein synthetic
pattern in growing mouse oocytes. J. Exp. Zool.,
220: 361-370.

13

14

15

16

648

Bachvaroova, R., De Leon, V., Johnson, A., Ka-
plan, G. and Paynton, B. V. (1985) Changes in total
RNA, polyadenylate RNA, and actin mRNA during
meiotic maturation of mouse oocytes. Dev. Biol.,
108: 325-331.

Wassarman, P. M. (1983) Oogenesis: Synthetic
events in the developing mammalian egg. In
“Mechanism and Control of Animal Fertilization”.
Ed. by J. F. Hartmann, Academic Press, New York,
pp. 1-54.

Hiramoto, Y. (1974) Mechanical properties of the
surface of the sea urchin egg at fertilization and
during cleavage. Exp. Cell Res., 89: 320-326.

Van Blerkom, J. and Brockway, G. (1975) Qualita-

1!

18

19

C.-H. YANG AND R. YANAGIMACHI

tive patterns of protein synthesis in the preimplanta-
tion mouse embryo. I. Normal pregnancy. Dev.
Biol., 44: 148-157.

Howe, C. C. and Solter, D. (1979) Cytoplasmic and
nuclear protein synthesis in preimplantation mouse
embryo. J. Embryol. Exp. Morph., 52: 209-225.
Abreu, S. L. and Brinster, R. L. (1978) Synthesis of
tubulin and actin during the preimplantation de-
velopment of the mouse. Exp. Cell Res., 114: 135-
141.

Giebelhaus, D. H., Hekkila, J. J. and Schuetz, G. S.
(1983) Changes in the quantity of histone and actin
messenger RNA during the development of preim-
plantation embryos. Dev. Biol., 98: 148-154.

ZOOLOGICAL SCIENCE 7: 649-655 (1990)

Role of Symbiotic Algae in Hatching of Gemmules of the
Freshwater Sponge, Radiospongilla cerebellata

MUNETOSHI KANAYAMA and YOSHIHISA KAMISHIMA!

Department of Biology, Faculty of Science, Okayama University,
Okayama 700, Japan

ABSTRACT—The role of symbiotic algae in the gemmule hatching of the freshwater green sponge,
Radiospongilla cerebellata, was investigated. Under optimal conditions, gemmules of this sponge
hatched almost 100% in 10 days. Dormant gemmules of the sponge colonized chlorellae in the
thesocytes. These symbiotic algae were in two states: those with a high chlorophyll content (active
algae) and those with little or no chlorophyll content (inactive algae). As incubation time proceeded,
the ratio of thesocytes which contained active algae and the number of the active algae in a single
thesocyte increased. Uner darkness, gemmules of this sponge were unable to hatch. The minimum light
intensity required to effect 100% hatching was 160 Ix in a 12L-12D photoperiod, and the minimum
potoperiod at 1300 Ix was 30 min per day. A photosynthetic inhibitor, Diuron, suppressed the gemmule
hatching, and concurrently suppressed the increase of thesocytes contained the acitive algae. These
results suggested that the hatching of the gemmule of this sponge is promoted by photosynthetic activity
in the symbiotic algae colonized in the thesocytes. This may be an example of plant/animal obligatory
endosymbiosis, in which the symbiotic algae in gemmular cells are essential for the sponge to survive

© 1990 Zoological Society of Japan

through the unfavorable winter season.

INTRODUCTION

Algal endosymbionts have been the subject of
study of many animal phyla, in which symbiotic
algae influence the host cell physiology, so that
they modify the lives of the animals [1-3]. Biolo-
gical significance of the symbiotic algae has inten-
sively studied, especially in protozoa [4] and lower
invertebrates such as cnidarians [5—7] and molluscs
[8-10]. In relations between animal and algae, one
of the most interesting cases is that the animals
totally depend on these intracellular plants for
their fundamental way of life, and the symbriosis is
obligatory.

In Polifera, some freshwater green sponges in
Spongillidae colonize zoochlorellae as symbionts
in archeocytes and choanocytes in adult forms [11-
14]. Most of these sponges produce gemmules, by
means of which they survive the unfavorable win-
ter season. Symbiotic chlorellae are also found in

Accepted September 30, 1989
Received June 6, 1989
" To whom all correspondence should be addressed.

gemmular cells. These symbiotic algae have been
shown to produce various effects on the lives of
these sponges [15, 16]. Several studies have sug-
gested that symbionts affect the gemmule germina-
tion or hatching [15, 17, 18]. However, conclu-
sions have varied depending on species studied or
by research methods. Illumination of gemmules
that colonize symbionts has been shown to acceler-
ate the rate of hatching only in limited cases [18,
19]. It is, further, not clear whether light affects
germination directly or through the symbionts,
although there is a report suggesting that the effect
was direct [20].

Present study focused on the biological signi-
ficance of algal symbionts in dormant gemmules of
the freshwater sponge. We study dormant release
in two species of freshwater Spongillid sponge,
Radiospongilla cellebelata and Ephydatia flurivati-
lis. These sponges live in similar habitats in
southern area of Japan [12, 14] and colonize
symbiotic chlorellae in the gemmule. However,
only in the former species, light affected gemmule
hatching consistently, so that the present experi-
ment carried out exclusively on this species. This

650 M. KANAYAMA AND Y. KAMISHIMA

report, to our knowledge, is the first study of a
freshwater sponge in which light has been shown to
decisively affect gemmule hatching, due to photo-
synthesis of symbiotic algae.

MATERIALS AND METHODS

Gemmules of Radiospongilla cerebellata Bower-
bank were collected from fields in August and
September of 1986 and 1987. Sponges were
marked and then checked frequently to identify
gemmule sources and determine the level of gem-
mule maturation. Gemmules collected from the
same sponge at one time were designated as a
batch. Collected gemmules were kept in the M
medium [21] in a refrigerator throughout their
dormancy. Gemmules were not hatched prior to
three to four months’ dormancy, but those from
any batches showed 85% to 100% hatching after
130 days’ dormancy. No differences were
observed in the hatching rate between gemmules
from the same batch.

For control experiments, gemmules were incu-
bated in the M medium under the photoperiod of
12 hr light and 12 hr dark (12L-12D) at 1,300 Ix at
24°C for 21 days. The plant light Homolux F1 20S
PG (Matushita Electric Co.) was used as the
source of illumination. When other kinds of light
sources such as white fluorescent lamp or incandes-
cent lamp were used, no significant differences
were observed in the final hatchign rate. Under
these optimal conditions, most of gemmule batch-
es reached their final hatching rate within 10 days
and no further improvement in the rate was
observed in the 21-day’s experiment. For incuba-
tion 30 gemmules were placed in a plastic petri dish
(30mm in diameter) containing 5ml of M
medium. At least five dishes were used for each
experiment. During the illuminating phase, the
dishes were inspected daily for hatching rate or
abnormalities such as fungal growth. Gemmules
treated in darkness were inspected under dim light
(20 lx) in 10 to 20sec. Hatched gemmules were
removed from the dishes to prevent possible en-
vironmental changes caused by the growing
sponges. Any dishes showing abnormal signs were
discarded.

In inhibition experiments for the photosyn-

thesis, gemmules were pretreated in an inhibiting
medium in a refrigerator for one week, and then
incubated in the same inhibiting medium under the
optimal conditions described previously. The in-
hibiting medium contained 3-(3, 4-dichlorophe-
nyl)-1, 1-dimethylurea (Diuron, Sigma Chem.
Co.) in M medium. In order to check cytotoxic
effect of the agent on the animal cells, the pretre-
ated gemmules were rinsed with fresh M medium
and incubated under conditions optimal for
hatching.

RESULTS

Gemmules of the freshwater green sponge,
Radiospongilla cerebellata showed very high hatch-
ing rates, ranging from 85% to 100% under condi-
tions optimal for hatching, that is, 12L-12D photo-
period at 1,300 Ix at 24°C. The final hatching rate
produced under these conditions varied in the
different batches (Figs. 1 and 2). Daily illumina-
tion for more than 30 min brought all batches to
the maximum hatching rate under these condi-
tions. Daily illumination of less than 10 min:

12hr L./ day

©O—© 30min L./ day
©—® 10min L./ day
O—O
o—e
@—@ 05min L./ day

2min L./ day

Imin L./ day

@—®@ Continuous darkness

Hatching (% cumulative)

D © © £) Oo O @

Period of incubation

Fic. 1. Hatching rate in gemmules of a batch during
various photoperiods illuminated at 1300 Ix. Daily
illumination of more than 30min brought 100%
hatching within a week in this batch. Illumination of
less than 10 min per day produced a hatching rate
roughly proportional to the length of illunination.
In order to simplify of this figure, lines showing 1, 2,
5, and 8hr illumination, each of which showed a
100% final hatching rate, were omitted.

Sponge Gemmules and Symbitotic Algae 651

12hr L./ day
O—-O-O-O-O-O-O-O-O-O-O
80 ‘eo O
70 e
7) 2min L./ day
Z 69 6 DO-©-9©-©-©-©-©-©-©-9-9©-©
m
E 15min L./ day
2 at 5 Ath ESEEEEES
< © D 10min L./ day
= 40 5d
2 30 pO
imin L./ day
aa DOO © © © © © @@,
e 0.5min L./ day
10 iy
j, 7 Continuous darkness
0 'Q-0-0-0-F@-0-9-@-0-@-0-@-©-9 0 © 0-0-0 04
o 7 14 21
(days)

Period of incubation

Fic. 2. Hatching rate in gemmules of another batch.
This batch of gemmules also showed a hatching rate
proportional to the length of light administered
daily. Gemmules of this batch hatched under as
little as 0.5 min daily illumination, but under optimal
light conditions the hatching rate reached only 85%.

brought roughly proportional hatching rate to the
length of the photoperiod in each batch. However,
the same length of daily illumination produced
varing results in the the different batches. For
example, in one batch, 10 min, or even 2 min,
illumination produced a final hatching rate as high
as 95% (Fig. 1), while in other batches the same
photoperiod brought only a 50% or 60% rate.
(Fig. 2). In the latter, however, 30 sec daily
illumination produced a 23% hatching rate (Fig.
2), but it did not produced better results than the
former under the optimal conditions previously
described (compare Figs. 1 and 2). Continuous
darkness did not cause hatching at all any of the
batches over a period of 21 days.

Light intensity affected the final hatching rate of
the gemmules. When gemmules were illuminated
at more than 160 1x for a 12L-12D photoperiod,
they reached 100% hatching in 10 days (Fig. 3)
When the gemmules were incubated at 40 or 80 Ix,
the hatching rates were 90% and 92%, respective-
ly, significantly lower than those of batches incu-
bated at 160 1x or more. Even when gemmules
were illuminated at 20 1x, 25% of the gemmules
had hatched by 13th day, but the batch show no
furhter increase of the rate during the experiment.

100

Oe teeter oo C-©-6-6-6-©

S100 lux
1300 lux
160 lux

80 lux

Hatching (% cumulative)

DoD of of 2 2 © 2 @

One® aCe nf we One 2@ 0-098 AS Sawer
0 4] 14 21

Period of incubation (days)

Fic. 3. Gemmule hatching under various light intensi-
ties at a 12L-12D photoperiod. Hatching was
affected by the light intensity. Illumination of more
than 160 lx brought a 100% hatching rate, and 20 Ix
a 25% hatching rate. Illumination of 40 or 80 Ix
brought an approximately 90% final hatching rate.

Untreated: 12hr L./day

Diuron 10 4M ; rinsed
Diuron 10°°M
Diuron 104M

Untreated; continuous darkness

Hatching (% cumulative)

DD © © © © © © ©

aLALA2A2.4 4 2.
0 7 14 21
(Days)

Period of incubation

Fic. 4. Effects of the photosynthetic inhibitor Diuron
on gemmule hatching. Gemmules treated with 10~*
M Diuron for one week prior to incubation showed a
markedly low hatching rate of 15.6%. When the
Diuron pretreated gemmules were rinsed and incu-
bated in the normal M medium, no effect of the
agent was observed on the hatching rate. Pretreat-
ment with 10~° M Diuron had no effect on hatching.

When gemmules were pretreated with a photo-
synthtic inhibitor Diuron (10° * M) for one week in
a refrigerator and then brought to the optimal

652 M. KANAYAMA AND Y. KAMISHIMA

conditions (12L-12D, 24°C), the final hatching rate
was reduced to 15.6% (Fig. 4). When the inhibit-
ing media containing Diuron were replaced with
the normal M media during the incubation, the
hatching rate returned to the control level of 100%
in 10 days. When concentration of the agent were
lowered to 10-° M, no inhibition for the hatching
was observed.

Observed under a fluorescent microscope, sym-
biotic algae in thesocytes of dormant gemmules
appeared in two states. Algae in one state showed
bright red fluorescence, but those in the other
showed only very pale yellow fluorescence. In

100 —
60 8
;
ol 50
i 2
¥ gs
2 4)
Oo
E 40 =
>
a o 2
loa) fo.
(=
< %
a aD
5 so WD th
ce) a)
=
a x)
= oO
ie EB 3
2 a
a 20 van
£ :
S B 2
é =
o~ 2
10
1
0) 0

0) Wy 12 OB WA) 0 264 On 2 ee SON 20)

Period of incubation (hrs) Period of incubation ‘hrs)
Fig. 5. Fig. 6.
Fic. 5. Changes in the rate of thesocytes that contained

symbiotic algae under conditions of light (open
symbol) and dark (filled symbol). Thesocytes con-
taining red fluorescent (symbiotic) algae increased
over time as illumination continued (open circle).
The total number of thesocytes containing either
type of algae stayed constant (open square). Over
72 hr of illuminated incubation all algae contained in
thesocytes became the red fluorescent type. In
continuous darkness, no changes were noted in the
algal features.

Fic. 6. Changes in the number of symbiotic algae in a
single thesocyte during incubation. The red fluores-
cent algae in the thesocyte increased over time as
illumination proceeded (open circle), while total
number of algae remained almost unchanged during
the incubation (open square). No changes were
observed in the red fluorescent algae (filled circle) or
the total algae (filled square) in conditions of dark-
ness. The dotted line shows the average number of
algae in an adult cell.

dormant gemmules, about 50% of the thesocytes
colonized symbiotic algae of either one or both
types. Only 6% of these contained the red fluores-
cent algae (Fig. 5). The number of symbiotic algae
of both states in a single thesocyte in dormant
gemmules averaged 2.44, of which the number of
the red fluorescent ones was only 1.42 (Fig. 6).
The number of red fluorescent algae increased as
incubation time was increased. By 96 hr, the time
just prior to hatching, the ratio of thesocytes
containing red fluorescent algae rose to 55.6%
(Fig. 5); the number of red fluorescent algae in a
single thesocyte increased to 3.07 (Fig. 6); and the
total number of thesocytes containing algae as well
as the total number of algae in a single thesocyte
remained practically unchanged throughout in-
cubation. On the other hand, no such increase of
the red fluorescent algae was observed in gem-
mules incubated under the darkness.

BSG

60

wi
Oo

i
oOo

%o0
%o0 SZ
%o8 “IL

%o0

is)
oO

=
oO

% of thesocytes containing symbiotic algae
Ww
oOo

Apuewog
Aep/] JUZL
Aep/] ulw Z
WO) UOINIG q
2 ISSES
SsauyJep
Snonui4uo} LK \ NW

Aep/'7 uw 0
INGNGS
Uo! FEUIWN}]! XN} OZ
MOVE

Fic. 7. Comparison of thesocytes that contained sym-
biotic algae in gemmules incubated for 96 hr under
various conditions. Open column shows the retio of
thesocytes that contained either type of algae, and
the barred column shows the retio of thesocytes that
contained red fluorescent type. Final hatching rate
is shown on each of the columns. The percentage of
thesocytes that contained red fluorescent algae is
higher in gemmules with high hatching rates. Ver-
tical line on each column indicates the range of the
standard error.

Sponge Gemmules and Symbitotic Algae 653

Number of symbiotic algae per thesocyte

Aep/*] JY ZL
Aep/-] ulw 2
Aep/ ulwig-9
Wa OL uouniq

UOIJEUIWN))! XN) OZ

Fic. 8. Comparison of the number of red fluorescent
algae in a single thesocyte incubated for 96 hr under
various conditions. The open column shows the
total number of algae and the barred one shows the
number of red fluorescent algae. The hatching rate
is shown on each of the columns. Thesocytes in
gemmules with higher hatching rate contained larger
number of red fluorescent algae than those in gem-
mules with low hatching rates. The vertical line on
each column indicates the range of the standard
error.

Algal features were compared for gemmules
incubated under various conditions that produced
different hatching rates after 96 hr incubation.
Both the number of the thesocytes that contained
red fluorescent algae and the number of red
fluorescent algae in a single thesocyte were larger
in the batches with higher hatching rates than in
the batches with lower hatching rates (Figs. 7 and
8). However, no significant difference was ob-
served in these algal features between batches with
low hatching rates and those with no hatching.

DISCUSSION

The hatching of gemmules of R. cerebellata was
enhanced by light. When gemmules are illumin-
ated for a 12L-12D photoperiod, the hatching rate
was directly proportional to the light intensity

under the saturation point at 160 Ix. At 1,300 Ix,
daily illumination for more than 30 min brought all
batches to the control hatching rate of 85% to
100% at a 12L-12D photoperiod over 7 to 10 days.
(Figs. 1 and 2). In one cese, daily illumination for
only 2 min effected nearly full hatchig (Fig. 1).
Total darkness, however, failed to induce hatching
in any batches of gemmules. These observations
suggest that light is essential to gemmule hatching,
and that even every low total doses, such as 6,500
Ix-hr (1,300 Ix x 0.5 hr x 10 days) will suffice.

Two algal states observed under fluorescent
microscopy showed differences in chlorophyll con-
tent of the algae in dormancy. Chlorellae without
red fluorescence, indicating little or no chlorophyll
content, were present in larger numbers in dor-
mant gemmules. This observation confirms mor-
phological reports which have shown _ loosey
arranged thylakoid membrane of symbiotic algae
in dormant gemmules [13, 14]. In dormancy, 50%
of thesocytes in a gemmule contained symbiotic
algae (Fig. 5) and about 50% of the algae in a
single thesocyte was of the red fluorescent type
(Fig. 6). Under the optimal condition, all the
algae became the red fluorescent type (Fig. 6), so
that all the thesocytes colonizing algae became
those with red fluorescent algae (Fig.5). The
conditions that induced low hatching rates pro-
duced slighter increase of the red fluorescent
algae, so that the percentage of thesocytes contain-
ing red fluorescent algae was equivalently low
(Figs. 7 and 8). Thesocytes are known to divide
and differentiate into histoblasts during the in-
cubation [11, 15]. Thus, present resutls that the
total number of algae per cell stays nearly constant
(Fig.6, open square) will indicate concomitant
multiplication of symbiotic algae with the host cell
division. Since no increase in either algal feature
was observed in darkness (Figs.5 and 6), red
fluorescence of the algae seemed to be induced by
illumination.

An inhibitor of photosynthesis, Diuron, at a
concentration of 10~* M, suppressed the hatching
rate to 15% (Fig. 4). Increase both in the number
of thesocytes containing red fluorescent algae and
in the number of the red fluorescent algae in a
single thesocyte were also suppressed by Diuron
treatment (Figs. 7 and 8). However, when the

654 M. KANAYAMA AND Y. KAMISHIMA

agent was removed by rinsing the pretreated gem-
mules with the fresh M medium just before the
incubation, the gemmules resumed the normal
hatching rate (Fig. 4). Longer pretreatments with
Diuron for 2 or 3 weeks brought further suppres-
sion on the hatching rate to 10% or 8%, respec-
tively, while treatment only for the incubation
period caused lesser suppression to 70% (data not
shown). Lower concentration of the agent (107°
M) did not affect the hatching rate. All these
observations show that photosynthetic activity of
the symbionts in a majority of thesocytes in a
gemmule seems to promote the dormancy release
in this sponge.

It remains to learn how photosynthesis in the
algal cell promotes cell division or cytodifferentia-
tion of the thesocytes in this sponge. It has been
reported that symbiotic chlorellae promote bud-
ding of hydra through photosynthesis by transmit-
ting photosynthetic products ‘to the host animal
[22]. Considering the extremely low dose of light
needed to cause hatching and the small number of
chlorophyll containing symbionts in a single theso-
cytes, it is difficult to imagine that enough products
are synthesized and transfered to the host cell to
support energy requirements of hatching. On the
other hand, since thesocytes are rich in vitelline
platelets which are used to supply energy for
germination [11], photosynthetic products by sym-
bionts are not necessarily used as the source of
energy. Indeed, it has been demonstrated that
energy production through respiration during the
hatching is independent from the illumination [17],
and of the symbionts colonized [20].

On the other hand, photosynthesis in the sym-
bionts may supply oxygen to and remove carbon
dioxide from the surrounding cytoplasm of dor-
mant thesocytes. It may also cause ionic changes
in the cytoplasm. These microenvironmental
changes in thesocytes may trigger a certain
mechanism that counteracts dormancy and makes
the cell start to divide and differentiate. Oxygen
supplied by the symbiotic chlorellae to the de-
veloping cell has been reported in studies of the
development of a certain salamander egg [23].
Direct effects of light on gemmule hatching is
known in spongillidae [20]. But this may not be
the case in the present work, since Diuron, a

photosynthetic inhibitor, clearly affects hatching.

It has been reported that certain substances are
involved in gemmule hatching. Germination is
inhibited by gemmulostasin, a substance released
from young sponges in Ephydatia flurivatilis (24,
25]. Dormancy is maintained by the presence of
cyclic AMP in Spongilla lacustris [26, 27|. These
substances are thought to suppress cell division
and cytodifferentiation. Decrease in cyclic AMP
was observed at the beginning of incubation in the
hatching of gemmules [27]. In the present experi-
ment, however, hatched gemmules were removed
from the dishes to prevent any effects of subst-
ances released from young sponges, although no
such substances have ever been reported in R.
cellebelata.

There have been several reports that light is not
essential for gemmule hatching in Spongillidae [12,
17-19]. Since the total amount of light required
for hatching in R. cerebellata is very low, care must
be taken not to expose gemmules to light before
the experiment is begun, especially during the later
phase of storage.

ACKNOWLEDGMENTS

We are very much grateful to Emeritus Professor Siro
Kawaguti of Okayama University for his valuable advice

and discussion in the course of this study.

REFERENCES

1 Mascatine, L. and Pool, R. R. (1979) Regulation of
numbers of intracellular algae. Proc. Roy. Soc.
Lond. B., 204: 131-139.

2 Hinde, R. (1983) Host release factors in symbioses
between algae and invertebrates. Endocytobiol., 2:
709-726.

3 Harris, L. G. (1971) Nudibranch association as
Symbioses In “Aspects of the biology of Symbiosis”.
Ed. by T. C. Theng, Univ. Park Press, Baltimore,
pp. 77-90. |

4 Lee, J. J., Lee; M. J. andi Weiss DENS ;3(1985)
Possible adaptive value of endosymbionts to their
protozoan host. J. Protozool., 32: 380-382.

5 Kawaguti, S. (1964) An electron microscopic proof
for a path of nutritive substances from zooxanthellae
to the reef coral tissue. Proc. Japan Acad., 40: 832-
835.

6 Douglas, A. E. and Smith, D. C. (1983) The cost of
symbionts to their host in green hydra. Endocyto-
biology, 2: 631-647.

10

11

it

13

14

15

16

17

Sponge Gemmules and Symbitotic Algae

Rahat, M. and Reich, V. (1986) Algal endosym-
biosis in brown hydra: Host/symbiont specificity. J.
Cell Biol., 86: 237-286.

Muller-Parker, G. and Pardy, R. L. (1987) The
green hydra symbiosis: Analysis of a field popula-
tion. Biol. Bull., 173: 367-376.

Kawaguti, S.,. Yamamoto, M. and Kamishima, Y.
(1965) Electron microscopy on the symbiosis be-
tween blue-green algae and opisthobranch, Placob-
ranchus. Proc. Japan Acad., 41: 614-617.
Kawaguti, S. and Yamasu, T. (1965) Electron
microscopy on the symbiosis between an elysioid
gastropod and chloroplasts of a green alga. Biol. J.
Okayama Univ., 11: 57-65.

Simpson, T. L. (1984) The Cell Biology of Sponges.
Springer Verlag. New York.

Mukai, H. (1980) Notes on the freshwater sponges
of Tatara-numa (Gumma-ken, Central Japan), with
review of literature on some developmental aspects
of gemmules. Sci. Rep. Facult. Educ. Gunma Univ.,
29: 35-71.

Williamson, C. E. (1979) An ultrastructural inves-
tigation of algal symbiosis in white and green Spon-
gilla lacustris (L.) (Porifera: Spongillidae). Trans.
Amer. Micros. Soc., 98: 59-77.

Masuda, Y. (1985) Electron microscopic study on
the zoochlorellae of adult green sponges and gem-
mules of Radiospongilla cerebellata (BOWER-
BANK) (Porifera: Spongillidae) Kawasaki Igak-
kaishi, 11: 63-68. (In Japanese)

Wilkinson, C. R. (1980) Nutrient translocation from
green algal symbionts to the freshwater sponge,
Epydatia fluviatilis. Hydrobiologia, 75: 241-250.
Rozefeld, F., Masson, H. and Rasmont, R. (1979)
Analyse statistique du mouvement des cellules ami-
boides au cours de la gemmulation d’une éponge
d’eau douce. In “Biologie des Spongiaires. Colloq.
Interant”. Ed. by C. Levi and N. Boury-Esnault.,
CNERS. Pans, 291: 31-37.

Brondsted, A. and Brondsted, H. V. (1953) The
effect of symbiotic zoochlorellae on the germination
rate of gemmules of Spongillia lacustris (L.) Vid.

18

19

20

Jas

22

23

24

25

26

Da

655

Medd. Dansk. Nat. Foren., 115: 133-144.

Benfey, T. J. and Reiswig, H. M. (1982) Tempera-
ture, pH, and photoperiod effects upon gemmule
hatiching in freshwater sponge, Ephydatia mulleri
(Porifera, Spongillidae). J. Exp. Zool., 221: 13-21.
Strekal, T. A. and McDiffett, W. F. (1974) Factors
affecting germination, growth, and distribution of
the freshwater sponge, Spongilla fragilis LEIDY
(Porifeta). Biol. Bull., 146: 267-278.

Rasmont, R. and Schmidt, I. (1967) Mise en evi-
dence du caractere photo-sensible de la respiration
des gemmules de spongillidae (Porifera). Comp.
Biochem. Physiol., 23: 959-967.

Rasmont, R. (1961) Une technique de culture des
éponges d’eau douce en milieu controlé. Ann. Soc.
Roy. Zool. Belg., 91: 147-156.

Rahat, M. and Reich, V. (1983) A comparative
study of tentacle regeneration and number in sym-
biotic and aposymbiotic Hydra viridis: effect of
zoo-chlorellae. J. Exp. Zool., 227: 63-68.
Bachmann, M. D., Carton, R. G., Buckholden, J.
M. and Wetzel, R. G. (1985) Symbiosis between
salamander eggs and green algae: microelectrode
measurements inside eggs demonstrate effect of
photosynthesis on oxygen concentration. Can. J.
Zool., 64: 1586-1588.

Rasmont, R. (1965) Existence d’une régulation

biochimique de l’eclosion des gemmules chez less
spongillides. C. R. Acad. Sci., Paris, 261: 845-847.
Rozenfeld, F. (1974) Biochemical control of fresh
water sponge development: effect on DNA, RNA
and protein synthesis of an inhibitor secreted by the
sponge J. Embryol. Exp. Morphol., 32: 287-295.
Simpson, T. L. and Rodan, G. A. (1976) Role of
cAMP in the release from dormancy of freshwater
sponge gemmules. Dev. Biol., 49: 544-547.
Harrison, F. W., Rosenberg, E. M., Davis, D. A.
and Simpson, T. L. (1981) Correlation of cyclic
GMP and cyclic AMP immunofluorescence with
cytochemical patterns during dormancy releas and
development from gemmules in Spongilla lacustris
L. (Porifera: Spongillidae) J. Morphol., 167: 53-63.

ONS. puahiged
0 : ns th he kets ; th a zy — te 7) De = ron
a. eimeunt: ois pip cone wine: aft Appears; 3° a ‘hed “toditie BES
er tae fat alae pe Paci Atle: aD. safe everene ere et yie
} : Pirie. 1 as) 5 d iz J aA 35 _ ¥ ab. a d 13.
Res : at 7) ; seb yea Me BS als SH. ee dane, it pony
7 ghee a: i Tae it Cu: sone
s by SiH ie ay abipiatis Shee asthe © k ae ae : en es
| sora aE Bebe Re 2 haat sivierds ot al ie s en ibe andi ont
ae ou, 7 Mee cham ait Ae, ght loos. 408 > eS Chea
emer ete . 4 oer fh portehs BS. nemgn oar acta
| i Sutae ie sotaa 2 ; edt Pause “ | Wee dense I ey wey

=)

1"
ns
o

rae

: ; =)
: et,
a f z
re a
i =
: o
rolyg ae cet gee
‘ {abi inoa te
‘
3 See a Ot) Te 8 af
i r i mM
. ‘ A
| sect re geinieyit)
Pe ; ,
f p 4) Gad * f, ay & te ,
rE r : ’ € - Sy ye
on" ; . ; So : ‘hi
i : gee hase fat ee F WeaR Pe ie ae c) eae Peat hs ie nee
Ls es z cS, SOP ‘AAPA 2 mre i PIS ad , ~
: 7 ne af
Lt 2 ‘5 = " ‘ ft hens ; _ wy i ~~ i‘ Ct Ae oy 7 r Pa soni
Favre Gyre Goew Woe OR RE em? ees CE PERCE) aS Aes. fae
} ; ; i <
~ ny : ¥ "4 Pad i. ? 4 ”
Th eed ‘ : ; ehevar? tbth x : ab
, Pa a 3 van iy er, ¥ : E 3 tes } ~<
“ ee t -
\ 3 ; “a
i i x ¢ n
J vot rt Fs : ba
Fe Sf b + e : es {
\ a a “a ment {
ix a meee ’ ne
5 , " a) ‘
7 hs i x Py a . eg y N " i
ig 4 2 i ie ; yes eas.
a eg a 4 % y thé
E i :
z ,
td a ‘
MY :
» ab i
rab le Py 5
4 } © TT. f
zie! < rp e i
% d +) > a 4 i
one . )
ae 7
j —
a) r

ZOOLOGICAL SCIENCE 7: 657-670 (1990)

© 1990 Zoological Society of Japan

Effect of Cytochalasin B, Nocodazol and Procaine
on Binding and Fate of Concanavalin A in
Competent Ectoderm of Xenopus laevis

LorHar TAcCKE! and Horst GRuNz

Department of Zoophysiologie, FB 9 (Biologie), University
GHS Essen, Universitadtsstr. 5,4300 Essen 1, FRG

ABSTRACT—The microfilament-inhibitor [Cytochalasin B] or microtubule-inhibitors (nocodazol and
procaine) cause similar morphological changes in isolated gastrula ectoderm of Xenopus laevis even at
the ultrastructural level i.e. a change of cell shape and an increase of the intercellular spaces. The
inhibitors used in this study do not influence the binding of Con A-gold to Con A-binding sites (CABS)

on the plasma membranes of ectoderm cells.

However, in contrast to Cytochalasin B-treated cells

patching and internalization of CABS take place after microtubule-inhibitor treatment. Due to the
enlarged intercellular spaces apparently a larger number of ectodermal cells is susceptible to the

inducing stimulus.

INTRODUCTION

Isolated competent amphibian ectoderm differ-
entiates into neural (archencephalic) structures,
when treated with the plant lectin Concanavalin A
(Con A) [1, 2]. This neural inducing effect can be
enhanced by the incubation in Cytochalasin B
(CB) (10 g/ml for Smin) prior to Con A-
treatment [3]. CB is known to interfere with the
cytoskeleton system, especially with intermediate
filaments and microfilaments [4, 5]. The earlier
results did not give any information about the
amount of Con A bound to different cells of the
ectodermal explant after the incubation with CB
and whether microtubule-inhibitors enhance the
neuralizing effect of Con A, like CB does. There-
fore we used Con A labelled with colloidal gold,
which is an excellent marker for transmission
electron microscopy (TEM). For further elucida-
tion of the correlation between CABS and cyto-
Skeleton structures (microtubules) we tested noco-
dazol and procaine, which reversibly act on micro-
tubules.

Accepted October 31, 1989
Received September 27, 1988
" To whom all correspondence should be addressed.

This explains the described increase of neural inducing activity of Con A in
combination with Cytochalasin B or with nocodazol.

MATERIALS AND METHODS

Embryological and histological methods

Eggs of Xenopus laevis were collected after
injection of males and females with choriongona-
tropin (SO0IE units per animal; Schering AG,
Berlin). After removal of the jelly coat and
vitelline envelope, pieces of competent ectoderm
were isolated from early gastrulae (stage 10, [6]),
using fine glass needles as described elsewhere [7].
The influence of the microtubule-inhibitor nocoda-
zol on the inducing activity of Con A was tested by
incubation of the deep layers of competent
ectodermal tissue. These deep ectoderm layers
turned out to be more reactive to the inducing
stimuli of Con A than the whole animal cap
(including superficial and deep layers). The deep
layers were separated by using fine glass needles
from the superficial layers [8]. Isolated deep layers
were incubated for 1 hr in Con A-solution (50 1
/ml and 100 g/ml). In another series deep layers
were pretreated for 5 min with nocodazol (1 pg/
ml) followed by an incubation in Con A-solution
(50 4g/ml). After washing pieces of tissue were
cultured for 5 days at 19°C in Holtfreter’s solution.
As a control series pieces of inner ectoderm were

658 L. TACKE AND H. GRUNzZ

treated with nocodazol (1 g/ml) for 5 min before
culturing in Holtfreter’s solution. For histological
examination cultured pieces of tissue were fixed in
Bouin’s solution, gross stained in borax carmine
and dehydrated in an ethanol series. Paraffin
sections (6 ~m) were stained with aniline blue/
orange G.

Con A-gold preparation and incubation of the
ectoderm in Con A-gold

A colloidal gold sol containing particles of an
average diameter of 17 nm was used [9]. The gold
was coupled with Con A (Medac Hamburg, E-Y-
lab) at pH7.2 in the presence of polyethylene
glycol (PEG) [10]. After centrifugation (twice at
27,000 x g for 45 min, Beckman Ti 50.2) the pellets
were collected in 0.5M Tris/HCl buffer (final
protein concentration 125 4g/ml).

In the first series, pieces of ectoderm were
incubated in Con A-gold for 30min at room
temperature prior to the fixation in glutaraldehy-
de; others were kept in CB (Sigma, Miinchen) for
5 min, in nocodazol (Aldrich Chem., Milw., USA)
for 10 min or in procaine hydrochloride (Merck,
Darmstadt) for lhr, respectively. The reversible
effect of each drug was proofed by dilution series.
CB and nocodazol were dissolved in dimethyl-
sulfoxide (Merck, Darmstadt) with a final con-
centation of 0.1%. These treatments were fol-
lowed by an incubation in Con A-gold for 30 min,
1 or 2 hr prior to the fixation, respectively. In the
control series explants were rinsed in 0.2M
methyl-a-D-mannopyranoside, which was added
to the Con A-gold medium.

Light and transmission electron microscopy

All explants were fixed in 1.5% glutaraldehyde
buffered with 0.1 M sodium cacodylate (pH 7.4).
After rinsing the material was postfixed with 1%
OsQO, in cacodylate buffer (pH 7.4), dehydrated in
an ethanol series, transferred to propylene oxide,
and embedded in Durcupan (Fluka, Neu-Ulm).
Semithin and ultrathin sections were cut on a
Dupon ultra-microtome (MT 5000), stained with
methylene blue and azur A for light microscopy
and with uranyl acetate and lead citrate for TEM
[11]. Ultrathin sections were studied in Zeiss EM
9-S 2 electron microscope.

RESULTS

Morphological changes of ectoderm cells after CB-,
nocodazol-, and procaine-treatment

Ectoderm of early gastrulae (animal caps) tre-
ated with cytoskeleton inhibitors showed signi-
ficant differences in comparison to untreated con-
trols.

a. Changes of the cell shape

Xenopus ectoderm consisted of several cell
layers, which could be easily distinguished by their
amount and distribution of pigment granules.
Cells of the outer ectoderm layer (0) contained
much more pigment granules than those of the
inner ectoderm (i) consisting of several layers. The
pigment granules are arranged in cells of the outer
layer like strings of pearls just beneath the plasma
membrane (Fig. 1), in contrast to the cells of the
deep ectoderm layers, where they were irregulary
distributed in the cytoplasm.

After incubation in CB the cells changed their
shape. Within 5 min of CB-treatment (5 g/ml)
especially cells of the inner ectoderm layer were -
affected. These cells began to round up and
showed a spherical shape (Fig. 2a). Those cells
never possessed any cytoplasmatic protrusions like
filo- or lobopodia, which could often be observed
on untreated cells. This dramatical effect could
not be seen on cells of the outer ectoderm layer.
Corresponding observations could be made after
30 min in procaine (Fig. 2b) and after 10 min in
nocodazol (Fig. 2c).

b. Changes of cell contacts

Cells of the outer ectoderm layer had a more
intimate contact to each other than those of the
inner layer. The outer cells were connected quite
tightly preferentially at their apical portions,
where tight junctions (zonula occludens) and des-
mosomes could be observed. The diameter of
intercellular spaces (IS) between cells of untreated
outer ectoderm was 0-20 um in the apical part
(side a, Fig. 1) and 500-1,600 ~m in the distal
portions of the cells (side b, Fig. 1). The gaps
between cells in the outer layer were apparently
unchanged after treatment with all three cyto-
skeleton-inhibitors (Table 1).

Cytoskeleton-Inhibitors, Xenopus Ectoderm 659

#@ConA

(control series)

+ cytoskeleton- inhibitors

*# ConA

former blastocoelic side

Fic. 1.

Schematic presentation of morphological changes in isolated competent ectoderm after treatment with

cytoskeleton-inhibitors and Con A. Within pieces of tissue outer (0) and inner (i) ectoderm layers can be
distinguished. The four sides of the tissue are called a-d. Side C’ is exposed in explants only, which are treated
with cytoskeleton-inhibitors. The dots in the figure indicate distribution of pigment granules.

o=outer ectoderm layer, i=inner ectoderm layer, side a=apical side, side b=distal side of the outer ectoderm
layer, side c=apical parts of the outer cells of inner ectoderm layers, side c’ =distal parts of the outer cells of inner
ectoderm layers, side d=distal parts of the inner cells of inner ectoderm layers.

TaBLE 1. Diameter of intercellular spaces (um)
side of gee asker lon mutters
ectoderm +Con A (CB, nocodazol,
tissue procaine)
+Con A
a 0- 20 0- 20
b 500-1600 600-1800
Cc 60— 450 60-— 700
d 0- 90 1400-— c

The results of each row are based on the calculation
of about 20 cells per series (for explanation of sides
of ectodermal tissue a-d see Fig. 1).

On the other hand untreated cells of the inner
layer were less tightly connected (IS: 0-450 um,
side c+d, Table 1). The diameter of intercellular
spaces between those cells increased significantly
after inhibitor-treatment; in some cases even the
spaces between cells of the inner and outer
ectoderm layer were enlarged (side c, Table 1).
The sign co (Table 1) means that the cells were
almost completely separated from the remaining
piece of tissue (Fig. 2c).

c. Changes of cytoskeleton system

The control cells often contained microtubules
and microfilaments. Bundles of microfilaments
were found just beneath the plasma membrane
(Fig. 3a, b). Microtubules were present in cells of

both ectoderm layers (Fig. 4a, b). Another phe-
nomenon, the receptor-mediated endocytosis by
the formation of coated pits, which depends on the
intact cytoskeleton system, could also be observed
in untreated pieces of ectoderm. The plasma
membrane of cells of both layers were able to form
coated pits (Fig. 5) with exception of the super-
ficial area of outer ectoderm cells (side a, Fig. 1),
where never any coated pit-formation could be
observed. After incubation with inhibitors the
cytoskeleton of outer ectoderm cells appeared to
be intact and not be impaired. On the other hand
the cells of the inner layer, especially those with
changed shape and nearly lost contact to their
neighbours, showed distinct reactions to the dif-
ferent inhibitors. In CB-affected cells no orga-
nized filaments could be observed and the forma-
tion of coated pits seemed to be inhibited. On the
other hand cells treated with nocodazol or pro-
caine contained quite normal microfilaments and
showed coated pit formation. As expected micro-
tubules were missing (Fig. 6).

In summary the change in cell morphology is a
good indication, whether a cell within the piece of
tissue was affected, less affected or not affected by
the inhibitor-treatment.

660 L. TACKE AND H. GruNZ

Binding of Con-A-gold on ectoderm cells treated
with cytoskeleton-inhibitors

After treatment with inhibitors of cytoskeleton
the explanted pieces of ectoderm were incubated
in Con A-gold (125 g/ml) for 30 min. Ectoderm
of the control series, which was incubated simul-
taneously with mthyl-a-D-mannopyranoside
bound negligible amounts of Con A-gold particles.
This clearly shows the specificity of the Con A-gold
labelling.

a. Binding of Con A-gold on plasma membranes
of cells which were less and not impaired by cytos-
keleton-inhibitors

According to the features described above there

remained a lot of cells within the studied pieces of
tissue, which should be called “less” or “not
affected” by the inhibitor-treatment. This holds
for all cells of the outer ectoderm layer and some
closely associated cells of the inner layer. The
highest number of Con A-gold particles was found
on the cells of the inner layer (side c’, Fig. 1).
Only few particles were present on the superficial
plasma membranes of the outer layer (side a, Fig.
1; Table 2). After 30 min of incubation in Con
A-gold most of the receptors on plasma mem-
branes of inner layer cells were distributed at the
intercellular spaces (patching) and could be
observed on the apical parts of the cells (Fig. 7).
Due to the increasing space of intercellular gaps
after the treatment with inhibitors (Table 1) the

TABLE 2. Number of gold particles per 100 “4m length of plasma membrane (Time of Con A-incubation: 1/2

hr)
Side of + Mf-inhibitor + Mt-inhibitors
RS RE A +Con A Coa ae or B) noc daed e
a Sees) S80) Bees)
b 0 30+ 10 30+10
c (c’) 0 30+10 (180+30) 30+10 (160+30)
d 200 + 40 *30+10 180+ 30

Number of gold particles left on 100 ~m length of plasma membrane (Time of Con A-incubation: 2 hr)

d 200 + 40

The sides a-d were determined by low magnification studies.
whole sectioned surfaces of about 20 cells per series at higher magnifications.

30+ 10 20+4

Afterwards for each side counting was made on
Because it was difficult to

determine side d of cells, which became round by inhibitor-treatment, we chose cells, which had still some

connections to neighbouring cells.

* The comparable small amount of gold particles on the side is due to the distribution fo Con A-receptors.

The

gold particles are randomly distributed (no patching) on the whole plasma membrane of the cells (not only on side
d). They could be observed even on the distal zones of the cells, which in controls could not be reached by the

Con A-gold solution.

Fic. 2a-c.

Inner ectoderm layers (side d=inner surface of the inner layers. Changes of cell shape and enlarged

intercellular spaces. a: after cytochalasin B-treatment (5 min, 5 4g/ml), b: after procaine-treatment (30 min, 2.7
mg/ml), c: after nocodazol-treatment (10 min, 1 g/ml). Bar=5 um.

Fic. 3a, b. Control series: outer ectoderm layer (side a=outer surface of the outer layer). Bundles of microfilaments
(mf) appear within the cytoplasm. No gold particles could be observed. Bar=0.5 ~m. Control series: inner
ectoderm layer (side d=inner surface of the inner layer). Microfiaments (mf) just beneath the plasma membrane,
where patches of Con A-binding sites (CABS) can be observed (au=Con A-gold). Bar=0.1 um.

Fic. 4a, b. Control series: inner ectoderm layer (side d=inner surface of the inner layer).

Microtubules (mt)

underneath the plasma membrane are often located adjacent to CABS (au=Con A-gold) and coated vesicles

(cv). Bar=0.1 um.
Fic. 5.

Control series: inner ectoderm layer (side d=inner surface of the inner layer). CABS (au=Con A-gold) are

internalized via coated pits (cp). (li=lipid droplet, fi=filopodium). Bar=0.1 um.

Fic. 6.

Inner ectoderm layer (side d=inner surface of the inner layer) after nocodazol-treatment. Microfilaments

(mf) appear to be intact and can be observed close to patches of CABS (au=Con A-gold). Bar=0.1 um.

661

Xenopus Ectoderm

itors,

i

-Inh

Cytoskeleton

662 L. TACKE AND H. GRUNZ

See,

ve ee

Cytoskeleton-Inhibitors, Xenopus Ectoderm 663

664 L. TACKE AND H. GRUNZ

Con A-incubation medium could interact with
much more cells than in the control series. This
explains why Con A-gold particles could be
observed on plasma membranes of cells of the
inner layer (side c, Fig. 1) and even on membranes
of outer ectoderm cells (side b, Fig. 1).

Each of these cells is able to form coated pits,
zones of the plasma membrane, where Con A-
receptors and their ligands (Con A-gold) will be
internalized. The formation of coated pits includ-
ing the internalization of Con A-gold could even
be observed on the distal parts of the outer
ectoderm cells (Fig. 8).

b. Binding of Con A-gold on plasma membranes
of inhibitor-affected cells

Cells, which showed clear morphological
changes, are called “affected”. Their origin was
the inner layer (side d, Fig. 1) exclusively. These

spherical cells apparently without any connection
to their neighbouring cells were able to bind Con
A-gold. On CB-treated cells the gold particles
were randomly distributed on the plasma mem-
brane and could be observed as small accumula-
tions at many sites of the membrane (Fig. 9)
Grains were also found on distal zones of the cells,
which in controls could not be reached by the Con
A-gold solution. Because of the inhibition of
coated pit formation on these cells after CB-
treatment never any internalized Con A-gold
particles could be observed. Cells treated with
nocodazol or procaine, however, are covered with
Con A-gold particles in large patches (Fig. 10).
On those spherical cells of the innermost areas of
ectoderm (side d, Fig. 1) they can be observed
even as one big cap. Neither nocodazol nor
procaine inhibited the internalization of Con A-
binding sites (CABS) via coated pits (Fig. 11).

TABLE 3. Neural inducing effect of Con A on deep layers of early gastrula ectoderm with or without

nocodazol treatment

Size of induction

Series Number of TAR ion : nocodazol- Sk ea small medium large
cases (g/ml) treatment Snictiiee S M i:

I 17 50 — 14 (82) 6 (42) 4 (29) 4 (29)

II 15) 100 _ 12 (80) 5) (25) 2-Cim) oF (58)

Il 28 50 + 22 (79) 6 (27) 4 (18) 12765)
IV 10 0 == 0 (0) — — —

Figures in parenthesis indicate percentage of cases.

The size of induction is determined by correlating the volume of neural tissue with the volume of the whole

explant.

Size of induction is called “medium” when the estimated volume of neural structures reached about

50% of the total tissue volume; above 50% size of induction is called “large” and below 50% it is called “small”.

Fic. 7.

Cells of the inner ectoderm layer after cytochalasin B-treatment (side c’/side c)=distal/apical parts of the

outer cells of the inner ectoderm layers). In this case these cells are less or not affected by CB so that CABS (au=
Con A-gold) can be observed within intercellular spaces (IS) and even endocytosed within coated vesicles (cv).

Bar=0.1 um.
Fic. 8.

Inner and outer ectoderm layer after CB-treatment (side b=distal parts of cells of the outer layer, side c=

apical parts of outer cells of the inner layer). Afte CB-treatment intercellular spaces (IS) are enlarged so that Con

A-gold can reach even plasma membrane-receptors of the outer ectoderm layer (side b).

Also coated

pit-formation (cp) can be observed in those cells (aua=Con A-gold). Bar=0.1 ~m.

Fic. 9.

Inner and outer ectoderm layer after CB-treatment (side d=inner surface of the inner layer). Part of a

spherical, CB-affected cell with a small accumulation of CABS (au=Con A-gold) on the plasma membrane. Bar

=()),5 arn.
Fic. 10.

Inner ectoderm layer after nocodazol-treatment (side d=inner surface of the inner layer).

Part of a

spherical, nocodazol-affected cell with a cap of CABS (au=Con A-gold) on plasma membrane. On each cell only
one cap with an extraordinarily high density of grains can be observed. The remaining parts of cell surface appear
free of gold-grains. Some endocytosed grains (arrow) can be observed. Bar=0.5 um.

Rice

Endocytosed gold particles (au) accumulate within multivesicular bodies (mv). Bar=0.5 um.

665

Cytoskeleton-Inhibitors, Xenopus Ectoderm

L. TACKE AND H. GRUNZ

666

Cytoskeleton-Inhibitors, Xenopus Ectoderm 667

Prolonged incubation times with Con A-gold re-
sulted in increasing amounts of internalized gold
grains per cell, which could be observed within
endosomes. In contrast to the control series the
amount of gold grains on the plasma membranes of
labelled cells decreased during Con A-gold incuba-
tion (Table 2).

The influence of nocodazol pretreatment on the
neural inducing activity of Con A

In two series pieces of competent ectoderm
(only deep layers, Fig. 1, 1) were treated with Con
A-solution (series I=50 g/ml; series I= 100 ug/
ml). Histological analysis of cultured explants
showed the differentiation of well defined neural
structures with an induction rate of 80% in series I
and 82% in series II (Table 3). The rate of neural
induction was not influenced by the alteration of
the Con A-concentration of series I and series II,
but the size of the induced neural structures was
significantly different within the two series. The
size of induction is determined by correlating the
volume of neural tissue with the volume of the
whole explant. Size of induction is called
“medium” when the estimated volume of neural
structures reached about 50% of the total tissue
volume; above 50% size of induction is called
“large” and below 50% it is called “small”. Ex-
plants with large amounts of neural structures
could be analyzed in 58% of the cases of series II
but only in 29% of the cases of series I (Table 3).
In series III explants were pretreated with the
microtubule-inhibitor nocodazol (1 ug/ml for 5
min) followed by a Con A-incubation (50 “g/ml
for hr). In 79% of the cases neural structures
could be defined. From these positive cases 55%
contained large amounts of induced neural struc-
tures (Table 3). The enhancement of Con A-
concentration or the nocodazol-pretreatment did
not influence the rate of neural induction. The rate
of induction ranged from 82% in series I and 80%
in series II to 79% in series III. In comparison with
series I with 29% of induced cases with large sized
neural structures, the results of series II and series
III showed both an increase of the degree of large
sized neural structures. Large neural structures
could be observed in 58% of induced cases (series
IT) and in 55% of induced cases in series III (Table

3). The explants of the control series (series IV)
developed into atypical epidermis. Thus a neura-
lizing effect of nocodazol could be ruled out.

DISCUSSION

In addition to many other effects CB acts revers-
ibly on the cytoskeleton system of eucaryote cells
by disarranging the intermediate filament and mic-
rofilament components [12-16], whereas nocoda-
zol and procaine are known to interfere reversibly
with microtubules [17-21]. In Xenopus ectoderm
used in this study cytoskeleton-inhibitors cause the
change of cell shape especially of those cells of the
inner ectoderm layer (side d, Fig. 1) (for reivew
see [22]). In our approach cells are no longer able
to establish filo- or lobopodia, features, which are
involved in active cell movements [13]. As a
consequence of these events the diameter of the
intercellular spaces between inner ectoderm cells
increases significantly (Table 1). According to
these features cells of the outer ectoderm seem to
be less or not affected by the drug. This may be
due to different plasma membrane properties [12,
23] or probably to tight junctions and desmo-
somes, which are established between cells of the
outer ectoderm, and which are not affected by the
inhibitors under our experimental conditions [5,
24].

The binding of Con A-gold to its plasma mem-
brane receptors is not inhibited by CB, nocodazol
or procaine. As we have shown in our study on
Con A-binding sites (CABS) on the four different
sides (Fig. 1) of the two layered ectoderm tissue,
the amount of Con A-gold particles on plasma
membranes of inner ectoderm cells (side d, Fig. 1)
is much higher than on cell surfaces of the outer
ectoderm layer (side a, Fig. 1) [25]. The same
holds true for ectoderm explants, which have been
treated with cytoskeleton-inhibitors prior to the
Con A-gold incubation (Table 2). These results in
addition to the TEM-observations show that the
cytoskeleton system in these cells, which is in-
volved in endocytotic processes [26, 27], is intact.
Therefore it is likely that CB has only little or no
effect on these cells. This is probably due to the
length of incubation. Under our experimental
conditions the drug could not reach and affect

668 L. TACKE AND H. GRUNzZ

these cells of the inner ectoderm. The inner cells
of the deep layers, which we called ’affected’,
because they show all of the described morpholo-
gical changes, bind Con A-gold, too. However,
concerning the fate of CABS there exist striking
differences between cells treated either with mic-
rofilament- or microtubule-inhibitors. After CB-
treatment CABS show no patching and could be
found in small numbers at several sites of the
plasma membrane. The relatively small amount of
gold particles per 100 ~m of membrane length (30
+10) cannot necessarily be explained by a de-
creased number of CABS on these cells. The
CABS situated on the distal parts of the cells may
shift over the whole cell surface after CB-
treatment, when these cells round up and finally
loose almost completely the contact to their neigh-
bouring cells. This translocation of Con A-
receptors on the plasma membranes may result in
the relatively small number of Con A-gold parti-
cles per unit of membrane length (Table 2). On
cells treated with microtubule-inhibitors, however,
patching and even capping of CABS is apparently
not inhibited, resulting in such high numbers of
gold particles (180+30, Table 2). This is in agree-
ment with the studies of Roberson and
Oppenheimer [28], who showed an increase of
receptor-mobility after colchicin-treatment on sea
urchin cells. Since even internalization of CABS
takes place in these cells the authors assume that
microtubules are not necessarily involved in this
process. This is also known from other systems
[29]. After 2 hr of Con A-gold incubation follow-
ing the treatment with microtubule-inhibitors the
amount of gold on the plasma membranes is
significantly lower compared with the number of
particles after 1/2 hr of incubation. Apparently the
number of receptors decreases by a continuous
internalization of receptor-ligand complexes. It is
likely that no new receptors are incorporated into
the plasma membrane, which is probably due to a
disturbed vesicle-transport. This view is supported
by the studies of Monk and Hedrick [20], who
showed that procaine blocks exocytosis in Xenopus
eggs. In CB-affected cells never coated pit forma-
tion and internalization of Con A-gold occurred.
The inhibition of Con A internalization after CB-
treatment has been described by Bliokh et al. [26]

on mouse fibroblasts. It is highly probable that
also the cells of our system are able to internalize
the bound Con A-gold after rinsing and reincuba-
tion in the absence of the drug, because of the
reappearance of organized filaments as reported
on chicken oviduct by Wrenn and Wessels [4].

Concerning the neural inducing effect of Con A
this aspect may not be very important since inter-
nalization of CABS bound to their ligand is prob-
ably not necessary for neural induction. This view
is supported by our results and the studies of Born
et al. [30], Takata et al. [1] and Takata [31].
According to these results, there is an additional
number of target cells within the nocodazol- and
CB-treated pieces of tissue, which can be triggered
by the Con A-stimulus. This is thought to be
responsible for the described increase of the neural
inducing effect of Con A after CB-treatment [3]
and also after nocodazol-treatment. The effect of
the cytoskeleton-inhibitors can be quantitatively
measured by differentiation of large sized neural
structures, but not by an enhancement of induction
rate.

In conclusion the results of the present study -
show that inhibitors of both microfilaments and
microtubules cause similar morphological changes,
i.e. cell shape and diameter of intercellular spaces.
Both types of inhibitors do neither prevent the
binding of Con A-gold to the plasma membranes
of ectoderm cells nor do they probably alter the
number of CABS per single cell, although their
specific sites of action are quite different. Howev-
er, their influence on the fate of CABS (patching,
internalization) is quite dramatical. Even the cells
of the outer ectoderm layer were found to be
labelled with Con A-gold at their distal side,
although these cells indeed do not differentiate
into neural structures at all after incubation with
Con A as described elsewhere [3].

The enhancement of the neural inducing effect
of Con A after CB-treatment [3] and also after
nocodazol-treatment seems to be correlated to an
increased number of target cells, which can be
triggered by inducing stimuli. Furthermore the
disintegration of ectoderm into single cells can
Support neuralization as could be shown by
another approach [32].

Cytoskeleton-Inhibitors, Xenopus Ectoderm 669

ACKNOWLEDGMENTS

This investigation was supported by the Dutsche Fors-

chungsgemeinschaft and in part by a grant of the Uni-
versity GHS Essen.

10

11

12

13

14

REFERENCES

Takata, K., Yamamoto, Y. and Ozawa, R. (1981)
Use of lectins as probes for analyzing embryonic
induction. Roux’ Arch., 190: 92-96.

Grunz, H. (1984) Early embryonic induction. The
ectodermal target cells. In “The Role of Cell In-
teractions in Early Neurogenesis”. Ed. by A. M.
Duprat, A. C. Kato, M. Weber, Plenum Press, pp.
21-38.

Grunz, H. (1985) Effect of concanavalin A and
vegetalizing factor on the outer and inner ectoderm
layers of early gastrulae of Xenopus laevis after
treatment with cytochalasin B. Cell Diff., 16: 83-92.
Wrenn, J. T. and Wessels, N. K. (1970) Cytochala-
sin B: effects upon microfilaments involved in mor-
phogenesis of estrogen-induced glands of oviduct.
Proc. Natl. Acad. Sci. USA, 66: 904-908.
Spooner, B. S. and Wessels, N. K. (1970) Effects of
cytochalasin B upon microfilaments involved in mor-
phogenesis of salivary epithelium. Proc. Natl. Acad.
Sci. USA, 66: 360-364.

Nieuwkoop, P. D. and Faber, J. (1956) Normal
Table of Xenopus laevis (Daudin). North Holland,
Amsterdam.

Grunz, H. (1985) Information transfer during
embryonic induction in amphibians. J. Embryol.
Exp. Morph., 89 Suppl.: 349-364.

Asashima, M. and Grunz, H. (1983) Effects of
inducers on inner and outer gastrula ectoderm layers
of Xenopus laevis. Differentiation, 23: 206-212.
Frens, G. (1973) Controlled nucleation for the
regulation of the particle size in monodisperse gold
suspensions. Nature Phys. Sci., 241: 20-22.
Horisberger, M. and Rosset, J. (1977) Colloidal
gold, a useful marker for transmission and scanning
electron microscopy. J. Histochem. Cytochem., 25:
295-305.

Venable, J. H. and Coggeshall, R. (1965) A sim-
plified lead citrate stain for use in electron micros-
copy. J.Cell Biol., 25: 407-408.

Luchtel, D., Bluemink, J. G. and Laat, S. W. (1976)
The effect of injected cytochalasin B on filament
organization in the cleaving egg of Xenopus laevis. J.
Ultrastruct. Res., 54: 406-419.

Perry, M. M. (1975) Microfilaments in the external
surface layer of the early amphibian embryo. J.
Embryol. Exp. Morph., 33: 127-146.

Lin, S., Lin, D. C. and Flanagan, M. D. (1978)

15

16

17

18

19

20

21

22

Ji)

24

ju)

26

Specificity of the effect of cytochalasin B on trans-
port and motile processes. Proc. Natl. Acad. Sci.
USA, 75: 329-333.

Britch, M. and Allen, T. D. (1981) The effects of
cytochalasin B on the cytoplasmic contractile net-
work revealed by whole-cell transmission electron
microscopy. Exp. Cell Res., 131: 161-172.
Nickola, I. and Frimmer, M. (1986) Effects of
phalloidin and cytochalasin B on cytoskeletal struc-
tures in cultured rat hepatocytes. Cell Tissue Res.,
245: 635-641.

De Brabander, M. J., Van de Veire, R. M., Aerts,
F., Burgess, M. and Janssen, P. A. (1976) The
effect of methyl-5-(2-thienyl-carbonyl)-1H-ben-
zimidazol-2ylcarbamate (R 17934), a new synthetic
antitumoral drug interfereing with mircotubules, in
mammalian cultured cells in vitro. Cancer Res., 36:
905-915.

Scharf, S. R., Lieberman, M. B. and Candle, W. Z.
(1986) Determination of dorsoventral polarity in the
Xenopus egg requires microtubules. In “Progress in
Developmental Biology”., Ed. by H. Slavkin, Alan
R. Liss. New York, Part B pp. 345-348.

Snyder, J. A., Hamilton, B. T. and Mullins, J. M.
(1982) Loss of mitotic centrosomal microtubule
initiation capacity at the metaphase-anaphase transi-
tion. Eur. J. Cell Biol., 27: 191-199.

Monk, B. C. and Hedrick, J. L. (1986) The cortical
reactions in Xenopus laevis eggs: Cortical granule
lectin release as determined by radioimmunoassay.
Zool. Sci., 3: 459-466.

Raymond, M. N., Fourcault, G., Coffe, G. and
Pudles, J. (1986) DNA-synthesis and microtubule
assembly-related events in fertilized Paracentrotus
lividus eggs: reversible inhibition by 10mM pro-
caine. Eur. J. Cell Biol., 40: 296-303.

Vasiliev, J. M. and Gelfand, I. M. (1988) Mem-
brane-induced segregation of motile and stable do-
mains in cytoplasm: possible role in morphological
differentiation of tissue cells. Cell Diff., 24: 75-82.
Selman, G. G., Jacob, J. and Perry, M. M. (1976)
The permeability to cytochalasin B of the unpig-
mented surface in the first cleavage furrow of the
newt’s egg. J. Embryol. Exp. Morph., 36: 321-342.
Meller, K. and El-Gammal, S. (1983) A quantita-
tive freeze-etching study on the effects of cytochala-
sin B, colchicine and concanavalin A on the forma-
tion of intercellular contacts in vitro. Exp. Cell Res.,
146: 361-369.

Tacke, L. and Grunz, H. (1986) Electron micro-
scope study of the binding of Con A-gold to super-
ficial and inner ectoderm layers of Xenopus laevis
and its relation to the neural-inducing activity of this
lectin. Roux’ Arch. Dev. Biol., 195: 158-167.
Bliokh, Z. L. Domnina, L. V., Ivanova, O. Yu.,
Pletjushkina, O. Yu., Svitkina, T. M., Smolyani-

i. el

27

28

29

670

nov, V. A., Vasiliev, J. M. and Gelfand, I. M.
(1980) Spreading of fibroblasts in medium contain-
ing cytochalasin B: Formation of lamellar cytoplasm
as a combination of several functionally different
processes. Proc. Nat. Acad. Sci. USA, 77: 5919-
5922.

Wagner, D. D. and Hynes, R. O. (1982) Fib-
ronectin-coated beads are endocytosed by cells and
align with microfilaments bundles. Exp. Cell Res.,
140: 373-381.

Roberson, M. and Oppenheimer, S. B. (1975)
Quantitative agglutination of specific populations of
sea urchin embryo cells with concanavalian A. Exp.
Cell Res., 91: 263-268.

Sager, P. R., Syversen, T. L. M., Clarkson, T. W.,
Cavangh, J. B., Elgsaeter, A., Guldber, H. C., Lee,
S. D., Lichtman, M. A., Mottet, N. K., and Olm-

L. TACKE AND H.

30

31

32

GRUNZ

sted, J. B. (1986) Structure and function of the
cytoskeleton. The cytoskeleton: a target for toxic
agents. Ed. by T. W. Clarkson, P. R. Sager, and T.
L. M. Syversen, Plenum Press, New York, pp. 3-21.
Born, J., Grunz, H., Tiedemann, H. and Tiede-
mann, H. (1980) Biological activity of the vegetaliz-
ing factor: Decrease after coupling to polysaccaride
matrix and enzymatic recovery of active factor.
Roux’ Arch., 189: 47-56.

Takata, K. (1985) A molecular profile of the recep-
tor responsive to the neural inducing signals in the
amphibian competent ectoderm. Zool. Sci., 2: 443-
453.

Grunz, H. Tacke, L. (1989) Neural differentiation
of Xenopus laevis ectoderm takes place after dis-
aggregation and delayed reaggregation without in-
ducer. Cell Differentiation, 28: 211-218.

ZOOLOGICAL SCIENCE 7: 671-679 (1990)

Instability of the Number of Segments of Unoperated and
Regenerated Maxillary Palpi in the Maxillary-Palp-
Elongate (mpe) German Cockroach Mutant

AKIRA TANAKA and Mary H. Ross!

Department of Biology, Nara Women’s University, Nara 630, Japan, and
"Department of Entomology, Virginia Polytechnic Institute and
State University, Blacksburg, Virginia 24061, USA

ABSTRACT— Maxillary-palp-elongate (mpe) is a mutant modifying the terminal (Sth) segment of the
wild-type palpus to a variable number of divisions, increasing the total number of segments up to 9. The
supernumerary segments sometimes fuse during postembryonic development. The mutant palpi of both
sides of the mouth are often asymmetric in length and in the number of segments. The mutant palpus
occasionally carries a prominent bristle, never found in wild type. It resembles bristles on legs and
occurs exclusively on the 4th segment, a homologous segment to the tibia of legs. The percentage of
bristle-bearing palpi increases during nymphal development, and exceeds 50% in the last instar. A few
mutant palpi bear more than one bristle on the 4th segment. Amputation experiments were carried out
at various levels on mpe maxillary palpi. The right palpus was operated in the 2nd instar, and the
regenerate was examined in the 4th and 6th instars. Irrespective of amputated levels from the Ist-2nd to
the 6th-7th segment, the total number of segments of a regenerated palpus was irrelevant to that before
operation. This suggests that the stump does not regenerate new segments according to a “memory” of
the original number of segments. The number of segments of a regenerate was also irrelevant to that of
the contralateral palpus, which suggests that the stump does not use the contralateral palpus as a
“reference” for the number of segments to be regenerated. As shown in unoperated palpi, regenerated

© 1990 Zoological Society of Japan

segments sometimes fused during subsequent molts. The bristle was occasionally found in the

regenerated 4th segment.

INTRODUCTION

Maxillary-palp-elongate (mpe) is a simple auto-
somal recessive in the German cockroach, Blattella
germanica. The locus lies within a group of closely
linked loci on chromosome 9 [1]. This mutant
modifies the terminal (Sth) segment of the wild-
type palpus to a variable number of divisions,
increasing the total number of segments up to 9.
The mutant palpus often carries a prominent bris-
tle exclusively on the 4th segment, the segment
homologous to the tibia of legs. These facts
support homology of the maxillary palpus with the
leg, and suggest that some distal segments of the
mutant palpi correspond to the tarsal segments of
legs.

Regeneration experiments with insect appen-

Accepted January 10, 1990
Received September 21, 1989

dages have been mainly on the thoracic legs [2]. It
is puzzling that regenerated cockroach legs always
have tetramerous tarsi and never pentamerous as
in the normal legs [3-7], although exceptionally
pentamerous after amputation distal to the 2nd
tarsomere [8-10]. In this situation, it was of
interest to carry out regeneration experiments with
the mpe mutant, since maxillary palpi are homolo-
gous to legs and mpe maxillary palpi show a
variable number of segments [1]. Here we report
(i) further description of mpe maxillary palpi fo-
cused on the fusion of segments during nymphal
development, bilateral differences in the number
of segments, and the percentage of bristle-bearing
segments, and (ii) regeneration experiments with
amputation at various levels on mpe maxillary

palpi.

672 A. TANAKA AND M. H. Ross

MATERIALS AND METHODS

Observations on mpe maxillary palpi

Maxillary-palp-elongate (mpe) is a simple auto-
somal recessive with full penetrance and viability
[1]. Both right and left maxillary palpi were
observed in all nymphal instars, from the Ist to the
6th (last) instar. The total number of palpi
observed was 1386, thus the number in each instar
averaged 231. Two small basal segments were
designated as the Ist and the 2nd, and two suc-
ceeding longer segments as the 3rd and the 4th
segments. The number of segments distal to the
4th varied from 1+ to 4+. “+” denotes that a
palpus has an elongated or incompletely-divided
segment. The total number of segments hence
ranged from 5+ to 8+. In accordance with the
total number, the palpi were classified into follow-
ing seven categories: 5+,6,6+,7,7+,8and8+.
Maxillary palpi with 9 or more segments were not
found. No significant sex difference was detected
in the number of segments so that the data of both
sexes were combined.

Regeneration experiments

Each nymph was raised in a small plastic con-
tainer (61 X43 X17 mm high, Sanplatec, No. 2) as
previously described [7]. Since the 1st instar was
too small to be precisely amputated on the maxil-
lary palpus, the 2nd instar was operated. Prior to
amputation, carbon dioxide was introduced into

the container for anesthetizing the animal. Right
maxillary palpi were amputated with fine scissors
at various levels, i.e., between the Ist and 2nd
segment, the 2nd and 3rd, the 3rd and 4th, the 4th
and Sth, and if available, between the 5th and 6th,
and the 6th and 7th segment. These cutting levels
were abbreviated to 1-2, 2-3, 3-4, 4-5, 5-6, and
6-7, respectively, in this paper. For the controls,
the number of segments were recorded in both
befpre-operation right palpi and contralateral left
palpi.

Some operated maxillary palpi regenerated their
distal segment(s) in the 3rd instar. The regenerat-
ing segment(s), however, were rather small and
often not so clearly segmented. Then we examined
the regenerates in the 4th instar when the seg-
mentation of regenerates became fully recogniz-
able. In order to examine further growth of
regenerates, final observation was made in the 6th
instar when regeneration was almost completed.
Since no significant sex difference was detected in
regeneration experiments, the data were again
combined.

RESULTS

Fusion of segments during nymphal development

Supernumerary segments found in this mutant
were first thought to be retained throughout nym-
phal development [1]. It has been revealed in this
study, however, that the additional segments

TABLE 1. Percentages of segmentation types in mpe maxillary palpi

Instar

I II Ill IV Vv VI
Sie 6.3 EES 126 Tied 19.4 3332
6 DAS 16.8 ei 18.6 25.4 Dee
6+ 28-1 21.6 28.9 36.2 SET, 28.3
Segmentation type 7 Des 27.4 23.6 12.9 JIL8) 8.0
We 9.4 113 8.9 10.0 8.7 ml
8 8.1 14.4 6.9 ee 2.8 1.8

8+ 4.4 1.0
Av. No. of segments 21/5 6.83 6.59 6.48 62577 6.20
No. of palpi examined 160 292 246 210 Usp 226

+ denotes that a palpus has an elongated or incompletely-divided segment and that the value of 0.5 was given to

calculate the average number of segments.

Maxillary Palpi in mpe German Cockroach 673

sometimes fuse during postembryonic develop-
ment. Table 1 shows percentages of segmentation
types, 5+ to 8+, in each nymphal instar. In the
1st and 2nd instar the percentages of palpi with 7
or more segments were relatively high. After the
2nd instar, the percentages tended to decrease as
development advanced. On the other hand, seg-
mentateion type 5+ tended to increase. The
average number of segments decreased by about
0.6 during nymphal development. The average
reduction was caused by the fusion of segments in
some, but not all, palpi during postembryonic
development. Fusion took place between the
segments distal to the 4th segment, and never
between the proximal four segments.

Figure 1 shows various types of mpe maxillary
palpi (b-h) together with a typical wild-type palpus
(a). All specimens are of the 6th (last) instar. The

mutant palpi were always longer than wild type,
showing typical supernumerary segments (b-d) and
various degrees of fusion (e-h). An elongated Sth
segment (Fig. le) consisted of three distinct distal
segments in the 2nd instar. Figure 1f shows an
elongated part lying proximal to the last segment.
The elongated part consisted of three distinct
segments in earlier development, which shows that
fusion can take place between intermediate seg-
ments. An incompletely-fused distalmost segment
in Fig. 1g was two separate segments in the 2nd
instar. Another example of incomplete fusion is
presented in Figure 1h.

Asymmetry in the number of segments

A particularly interesting characteristic of this
mutant is an unequality in the length and the
number of segments of the right and left palpi.

\
y

Fic. 1. Various types of maxillary palpi (6th instar). (a) a wild-type palpus showing five typical segments, two small
basal and three succeeding longer segments. (b-h) mpe palpi: (b) 6-segmented, (c) 7-segmented, (d) 8-segmented
with an incipiently-divided 8th-segment, (e) with an elongated Sth segment resulted from complete fusion of three
distalmost segments, (f) an originally 8th segment fusing incompletely to an elongated 5th segment, (g)
incomplete fusion of two distal segments, and (h) incomplete fusion of three distal segments. (18)

674 A. TANAKA AND M. H. Ross

The palpi are not always symmetrical, differing
rather frequently in length and number of seg-
ments. Figure 2 shows a correlation of the number
of segments between right and left palpi in the 2nd
instar. Correlations of the other five instars
showed similar tendencies. The correlation was
low although significant (r=0.41, p<0.0001).
Only about 22% (32/146) of the nymphs had right
and left palpi with the same number of segments.
Average difference was 0.65, including nymphs
with the same number of segments.

Left

Right

Fic. 2. Correlation in the number of segments between
right and left mpe maxillary palpi (2nd instar). In
this and subsequent figures, “+” denotes that a
palpus has an elongated or incompletely-divided
segment and that the value of 0.5 was given when
correlation coefficient was calculated.

Bristles on maxillary palpi

Another point of interest is the existence of
bristles on mpe maxillary palpi, since no bristles
are found on wild-type palpi. Bristles were found
exclusively on the 4th segment of mpe palpi. The
bristle most frequently appeared as a single bristle
as shown in Figure 3a. Double bristles were rarely
encountered, either along the same generatrix
(Fig. 3b) or around the same level (Fig. 3c) of the
4th segment.

Table 2 shows percentages of bristle-bearing
mpe palpi in each nymphal instar. No significant

Fic. 3. Bristles found on the 4th-segment of mpe maxil-
lary palpi (6th instar). (a) a single bristle, (b) double
bristles along the same generatrix, and (c) double
bristles across the same level of the segment. (x 40)

difference in the percentage was detected between
segmentation types. No bristles were found in
new-born nymphs. Bristles first appeared in the
2nd instar, but were very rare. Then the percent-
age of bristle-bearing palpi gradually increased to
become the frequency exceeding 50% in the 6th
instar. Palpi with more than one bristle were
found only in the 6th instar, i.e., two palpi with
double bristles (Fig. 3b, c), and only one palpus

with triple brisles.

Regeneration experiments

Regeneration experiments started with 316
nymphs. Out of 316 nymphs, 238 survived until
the 4th instar, and 207 until the 6th instar. The
number of segments was counted in the 4th and 6th
instar and compared with that before operation.
Observations in the 4th instar are summarized in
Table 3. Changes in the number of segments from
the 2nd instar (before operation) to the 4th instar
(regenerated) ranged from —3.0 to +2.5. Irre-
spective of amputated levels, the number of seg-
ments in palpi with fewer segments at the time of
amputation tended to increase. Conversely, the
number tended to decrease in palpi with larger
numbers at amputation.

Correlation was analyzed in the number of seg-
ments between the 2nd instar palpi before amputa-
tion and the 4th instar palpi with regenerated
segments (Fig. 4). The data from 1-2 to 4-5
amputation levels were combined in Figure 4 (156
surviving nymphs in total). The data of 5—6 and 6-
7 amputation levels were excluded because of
lacking 5+, 5+ to 6+ segmentation types at the

Maxillary Palpi in mpe German Cockroach 675

TABLE 2. Percentages of bristles on mpe maxillary palpi

ik iol EP as, By ye

Ath

In 2 13 @ il © i Bw Op

Instar
I II III IV Vv VI

5+ 0.0 0.0 3302 19.4 24:5 52.0

6 0.0 2AY) Deal 12.8 35-9 61.2
6+ 0.0 0.0 2.8 lial 3.5) Sel

Segmentation type 7 0.0 1.3) 3), 14.8 Mya) DOM

War 0.0 0.0 0.0 14.3 36.4 56.3

8 0.0 0.0 5.9 9.1 28.65 S303
8+ 0.08 0.08

AV. 0.0 OF 3.3 IS.7 SLO 5)

(0/160) (2/292) (8/246) (33/210) _ (78/252) (125/226)

Bristle-bearing palpi usually had only one bristle on the 4th segment. In the instar VI exceptionally found were
* two palpi with double bristles and ** one palpus with triple bristles on the 4th segment. * denotes that the
number of palpi examined was less than 10. In parentheses are shown the numbers of bristle-bearing palpi/total

palpi.

TABLE 3. Effects of regeneration on the number of segments of 4th-instar mpe maxillary palpi

Amputated Nomot No. of segments before amputation

level nymphs Sap 6 6+ 7 Usk 8 8+
1-2 29 +0.25 aU) 75) + 0.06 S119) — 1-00 113)

2-3 37 +0.38 aU) Wl +0.17 =()), SW =O —1.00 —— AU
3-4 40 +1200" +0.67 +0.07 WAZ —0.63 0192

4_5 50 taal +0.95 +0.75 +0.17 —0.08 =()275)

5-6 50 +0.55 —0.14 —0.71 lO) =il,{31l S00
6-7 32 —0.71 — 1:25 1.32

Average changes in the number of segments are shown. 1-2 indicates amputation between the Ist and the 2nd
segment, and so on. * Only one nymph was examined.

time of amputation, respectively. The correlation
was very low although significant (r=0.24, p<
0.01). Only 37 nymphs out of 156 retained the
same number of segments through regeneration.
The average difference was 0.69, including nymphs
that regenerated the same number of segments.
Another type of analysis was made in Figure 5.
The number of segments of regenerated palpi was
compared with that of the contralateral unoper-
ated palpi. The data of different amputation levels
from 1-2 to 4-5 were again combined. There was
no correlation (r=0.11, p>0.1). Only 29 out of
156 nymphs had the same number of segments
between regenerates and contralateral palpi. The

Fic. 4. Correlation in the number of segments between
the 2nd instar (before amputation) and the 4th instar
(regenerated). For the symbol “+”, see Fig. 2.

676

average difference was 0.77, including nymphs that
had the same number of segments in both palpi.

The final examination was carried out on 207
nymphs that survived until the 6th instar (Table 4).
Irrespective of amputated levels, the number of
segments was mostly unchanged or decreased by
0.5, and less frequently decreased by 1.0 or 1.5
from the 4th to the 6th instar. The average
decrease in the number of segments during the two
instars was about 0.3, which was close to that of
unoperated palpi (Table 1).

Figure 6 shows examples of some right regener-
ated maxillary palpi together with the left unoper-
ated controls in the 6th instar. Regenerated seg-
ments have grown and are almost indistinguishable
from unoperated contralateral segments. The
right palpus shown in Figure 6a was originally
7-segmented in the 2nd instar. After amputation
between the 2nd and 3rd segment (arrow), a
8-segmented palpus was observed in the 4th instar,
and a bilobed segment has finally formed in the 6th
instar. The right palpus in Figure 6b was 7-
segmented and regenerated the same number of
segments after operation. The contralateral pal-
pus, also 7-segmented in the 2nd instar, has fused
into 6-segments. Figure 6c shows a right palpus,
8-segmented in the 2nd instar, in which the original
number of segments was restored after operation
and retained thereafter. The left unoperated pal-
pus was also 8-segmented, but three distal seg-
ments had fused by the 6th instar. The right palpus
in Figure 6d was originally 8-segmented, and did
not restore the number until the final instar. In
Figure 6e, the right palpus was also originally
8-segmented. After operation the two distalmost

A. TANAKA AND M. H. Ross

segments regenerated in the 4th instar, and were
retained throughout following instars. The left
unoperated palpus, 8-segmented with a subtle
tendency of fusion in the 2nd instar, showed more
advanced fusion in the 6th instar.

The percentages of bristles on regenerated max-
illary palpi are shown in Table 5. Bristle-bearing
palpi usually had only one bristle on the 4th
segment. In experiments with amputation levels
1-2, 2-3 and 3-4, the 4th segment was ablated by
amputation and later regenerated. In this case, no
bristle appeared until the 4th instar. Some palpi
later developed a bristle, but the percentage was
lower than that of the unoperated palpi even in the

u

@

»

a 8

G

2 .

o

» .

@

© ©5P

co)

Cc

o

no)

cD)

om

5+

5+ 6 ie aan 7 7+ 8 8 +
GomuErPra lateral. Amc ams @ir

Fic. 5. Comparison of the number of segments between

contralateral unoperated palpi (2nd instar) and re-
generated palpi (4th instar). For the symbol “+”,
see Fig. 2.

TABLE 4. Decrease in the number of segments in regenerated mpe maxillary palpi from the 4th

to 6th instar

Decrease in No. of segments

Av. No. of segments

Amputated No. of Instar

level nymphs 0 Or) 10m PS AeSoe Ay. IV VI

1-2 21 10 8 3 0 —0.33 6.50 6.17
2-3 32 14 14 4 0 —0.34 6.56 6.22
3-4 36 22 11 2 1 —0.25 6.75 6.50
4-5 41 19 18 4 0 —0.32 7207 6.76
5-6 48 22 20 4 2 —0.35 6.32 S87)
6-7 29 16 12 1 0 —0.24 6.40 6.16

Maxillary Palpi in mpe German Cockroach 677

Fic. 6. Examples of regenerated mpe maxillary palpi (6th instar). Right maxillary palpi were amputated in the 2nd
instar at the indicated levels (arrows): between the 2nd and 3rd segment (a), 3-4 (b), 4-5 (c), 5-6 (d), and 6-7
(e). Left maxillary palpi were unoperated controls. One or two labial palpi are also seen. (a andc, X16; bande,

x12; d, x14)
TABLE5. Percentages of bristles on regenerated mpe maxillary palpi
Amputated Instar

level IV VI

4th segment 1-2 0.0 (0/29) 28.6 (6/21)

Ie gone 23 0.0 (0/37) 28.1 (9/32)
3-4 0.0 (0/40) 41.7 (15/36)

4th segment 4-5 10.0 (5/50) 51.2 (21/41)

SO pee 5-6 16.0 (8/50) 50.0 (24/48)*
6-7 18.8 (6/32) 51860) (IZ)

* In the instar VI exceptionally found were a few palpi with double bristles (two

in 5—6, one in 6—7 amputation) on the unoperated 4th-segment.

In parentheses

are shown the numbers of bristle-bearing palpi/total palpi.

6th instar. On the other hand, the 4th segment was
not eliminated in experiments with amputation
levels 4-5, 5-6 and 6-7. In this case the percent-
age of bristles was similar to that of unoperated
palpi both in the 4th and in the 6th instar.

DISCUSSION

Maxillary palpi clearly evolved through the
modification of leg-like appendanges [11, 12].
However, legs of wild-type German cockroaches
have nine segments whereas maxillary palpi nor-
mally have five segments. Prominent bristles occur

on the tibia but never on wild-type maxillary palpi.
The mpe mutant increases the resemblance of
maxillary palpi to legs by variable increases in the
number of segments up to nine and the presence of
a bristle on the 4th segments, a segment presum-
ably homologous to the tibia [1, 11]. The discovery
here of double and triple bristles supports homolo-
gy of the 4th maxillary-palp segment to the tibia.
Supernumerary segments are apparently a variable
expression of tarsal segments.

The increased resemblance to legs suggests that
mpe expresses a primitive developmental pathway.
It belongs to a group of mutants on chromosome 9

678 A. TANAKA AND M. H. Ross

that apparently express primitive traits [1, 13],
including prowing (Pw) [14, 15], notched sternite
(st) [16], and stumpy (sty) [17]. Although 8+
segments in mpe palpi suggested a rare expression
of five tarsal segments, four tarsal segments is
probably a more primitive trait than five segments.
A leg mutant, fused tarsi, has 4-segmented tarsi,
i.e., 8-segmented legs [18]. Regenerated cock-
roach legs also have 4-segmented tarsi [3-7]. It
was suggested that archetypical appendages were
midleg-like nonspecific forms with 8 segments [6].

Maxillary-palp-elongate (mpe) is of interest in
that the mutation affects the bilateral symmetry of
body plan. The external morphology of cock-
roaches shows complete bilateral symmetry. The
exact likeness between the opposite sides of the
body has been verified by high correlation coef-
ficients calculated in the length of appendages of
the German cockroach [19]. For example, the
correlation coefficients between right and left
hindleg ranged 0.86-0.98 in the femur and 0.90-
0.99 in the tibia during nymphal instars and adult
stage. These very high correlations suggest the
existence of a regulatory mechanism to make right
and left structures the same in size. The correla-
tion in numbers of segments on right and left mpe
palpi was exceptionally low (Fig.2, r=0.41).
Furthermore, the correlation was even lower when
regenerated palpi were compared with the same
palpi before operation (r=0.24, Fig. 4) or with
contralateral unoperated controls (r=0.11, Fig.
Sy))s

Why is the number of segments frequently differ-
ent between right and left in mpe maxillary palpi?
The fluctuation in the number of segments suggests
a deficiency of regulation for bilateral symmetry.
We suggest that the increase and instability of the
number of segments of mpe maxilary palpi is
caused by a deficiency of function of the mpe+
gene. If so, mpe+ not only suppresses the ex-
pression of leg-like characteristics of maxillary
palpi but also controls the bilateral symmetry in
length and segment number of the palpi.

During postembryonic development of the Ger-
man cockroach, the number of segments increases
in antennae and cerci [20], but is unchanged in legs
[21] and wild-type maxillary palpi. mpe maxillary
palpi are the first appendages found where some

segments fuse with each other to decrease the
number of segments during postembryonic de-
velopment.

In regeneration experiments of wild-type maxil-
lary palpi, the number of segments remained five
after regeneration in Blabera [22], and Blattella
(Tanaka, unpubl.]. This was true with amputation
at any levels from 1-2 to 4-5 in Blattella. In mpe
maxillary palpi regeneration, however, the num-
ber of segments often changed within the range
from 5+ to 8+, including both increases and
decreases. Palpi with fewer segments tended to
increase the number of segments, and vice versa
(Table 3). It was first thought that there might be a
mechanism averaging the number of segments of
regenerated mpe palpi. This idea, however, is
probably wrong, since not only the correlation
between the number of segments before amputa-
tion and the number in regenerated 4th instar was
very low but also the variation of the latter was as
large as that of the former (Fig. 4).

The number of segments of a regenerate was
irrelevant to that of the original unoperated pal-
pus. This fact shows that an amputated stump does -
not regenerate new segments accoring to a “mem-
ory” of the original number of segments before
amputation. The number of the segments of a
regenerate was also irrelevant to that of the contra-
lateral palpus. This likewise shows that the stump
does not use the number of segments of contra-
lateral palpus as a “reference” for the number of
segments to be regenerated.

Both the increase in the number of segments and
the presence of bristles in mpe maxillary palpi are
characteristics that increase their resembrance
with legs. Hence we wondered if there might be a
positive correlation between the number of seg-
ments and the frequency of bristles. However, we
failed to detect such a relationship between the
two. Maxillary palpi with more segments did not
always show higher frequency of bristles (Table 2).
Moreover, the frequency of bristles increased dur-
ing development whereas the number of segments
sometimes decreased. There seemed to be an
increase in leg-like characteristics in the segment
corresponding to the tibia (4th segment) but a
decrease in those corresponding to tarsal segments
(tarsomeres). mpe is similar to homeotics and, like

Maxillary Palpi in mpe German Cockroach 679

homeotics, probably controls a battery of genes.
Possibly one that controls bristle formation is not
“turned off” during post embryonic development,
but another controlling segment number is subject
to modifiers that cause fusion or less subject to
control by mpe in later development. In spite of
variability of expression, supernumerary segments
and bristles are evident during post embryonic
development in both unaltered mpe and mpe re-
generates that provide new insight into the charac-
teristics of this mutant and its wild-type allele.

ACKNOWLEDGMENTS

We thank Mrs. Nancy Boles for technical assistance.

REFERENCES

1 Ross, M. H. and Tanaka, A. (1988) Genetic
variability in the German cockroach. XII: A third
mutant that suggests chromosome 9 carries a highly
conserved group of closely linked genes. J. Hered.,
79: 439-443.

2 Bulliére, D. and Bulliére, F. (1985) Regeneration.
In “Comprehensive Insect Physiology, Biochemistry
and Pharmacology, Vol. 2 Postembryonic Develop-
ment”. Ed by G. A. Kerkut and L. I. Gilbert,
Pergamon Press, Oxford, pp. 371-424.

3. Brindley, H. H. (1897) On the regeneration of the
legs in the Blattidae. Proc. Zool. Soc. Lond., 60:
903-916.

4 O*Farrell, A. F. and Stock, A. (1953) Regeneration
and the moulting cycle in Blattella germanica L. 1.
Single regeneration initiated during the first instar.
Aust. J. Biol. Sci., 6: 485-500.

5 Roberts, B. (1973) The relative growth of normal
and regenerated legs of the cockroach Blattella
germanica (L.). Trans. R. ent. Soc. Lond., 126:
447-461.

6 Tanaka, A., Ohtake-Hashiguchi, M. and Ogawa, E.
(1987) Repeated regeneration of the German cock-
roach legs. Growth, 51: 282-300.

7 Tanaka, A. and Ross, M. H. (1989) Tibia to femur
ratios of unaltered and regenerated legs of the
stumpy mutant of the German cockroach. Zool.
Sci., 6: 927-933.

8 Penzlin, H. (1963) Uber die Regeneration bei
Schaben (Blattaria) I. Das Regenerationsvermégen
und die Genese des Regenerats. Wilhelm Roux’
Arch., 154: 434-465.

9

ih

12

13

14

15

16

17

18

19

20

21

22

Bohn, H. (1965) Analyse der Regenerationsfahig-
keit der Insektenextremitat durch Amputations und
Transplantationsversuche an Larven der Afrikanis-
chen Schabe Leucophaea mederae Fabr. (Blattaria)
I. Regenerationspotenzen. Wilhelm Roux’ Arch.,
156: 49-74.

Bulliére, D. (1967) Etude de la régénération chez
un insecte Blattoptéroide Blabera craniifer Burm.
(Dictyopteére). I. Influence du niveau de la section
sur la régénération de la patte métathoracique. Bull.
Soc. Zool. Fr., 92: 523-536.

Snodgrass, R. E. (1935) Principles of Insect Mor-
phology. McGraw-Hill, New York. pp. 130-156.
Matsuda, R. (1965) Morphology and Evolution of
the Insect Head. Amer. Entomol. Inst., Ann Arbor.
pp. 1-334.

Ross, M. H. and Keil, C. B. (1978) Genetic
variability in the German cockroach XI. Does
chromosome 9 carry remnants of a primitive gene
system? J. Hered., 69: 337-340.

Ross, M. H. (1964) Pronotal wings in Blattella
germanica (L.) and their possible evolutionary sig-
nificance. Amer. Mid. Natur., 71: 161-180.

Ross, M. H. and Cochran, D. G. (1971) Cytology
and genetics of a pronotal-wing trait in the German
cockroach. Can. J. Genet. Cytol., 13: 522-535.
Ross, M. H. (1966) Notched sternite: A mutant of
Blattella germanica, with possible implications for
the homology and evolution of ventral abdominal
structures. Ann. ent. Soc. Amer., 59: 473-484.
Ross, M. H. (1975) Genetic variability in the
German cockroach X. Genetics of pale purple,
pearl, and stumpy. J. Hered., 66: 155-159.
Cochran, D. G. and Ross, M. H. (1967) Cockroach
genetics. In “Genetics of Insect Vectors of Disease”.
Ed. by J. W. Wright and R. Pal, Elsevier, Amster-
dam, pp. 403-415.

Tanaka, A. (1984) Relative growth between femur
and tibia in the German cockroach, Blattella germa-
nica. Growth, 48: 278-296.

Tanaka, A. and Hasegawa, A. (1979) Nymphal
development of the German cockroach, Blattella
germanica Linné (Blattaria: Blattellidae), with spe-
cial reference to instar determination and intra-
instar staging. Kontya (Jpn. J. Ent.), 47: 225-238.
Tanaka, A., Ogawa, E. and Ohtake, M. (1986)
Relative growth of leg segments of the German
cockroach, Blattella germanica. Growth, 50: 273-
286.

Urvoy, J. et Le Bris, R. (1968) Etude de la
régénération des cerques, des palpes labiaux et des
palpes maxillaires chez Blabera craniifer [Orth. Blat-
tidae]. Ann. Soc. ent. Fr. (N. S.), 4: 371-383.

The

ee ME el Gee wae DS “ er nial

| x i inti Baba woah Bie lig sie
fal ae sree: “nett sshacaaesant ss 4 ; foe isos
vee ' veep ciel 4" ' pee ee ev? yi %
Ria Ma” MEG Op teeny e saa ao

ZOOLOGICAL SCIENCE 7: 681-690 (1990)

Changes in Protein Secretory Patterns during
the Development of the Rat Epididymis

HIROSHI UEDA, TAKESHI HIRANO and SUNAO FUJIMOTO

Department of Anatomy, University of Occupational and Environmental Health,
School of Medicine, Kitakyushu 807, Japan

ABSTRACT—The developmental pattern of [*°S]methionine-labeled proteins secreted by the rat
epididymis was studied by two-dimensional polyacrylamide gel electrophoresis and autoradiography.
From fetal to sexually immature rats, epididymal tissues secreted polypeptides very similar to those of
fetal and neonatal rat testes; polypeptide T35 and F37 were especially prominent. Although both T35
and F37 became indistinct in the testis of 10-day-old rat, they were detected in the epididymal three
segments (caput, corpus and cauda) up to 50-day-old rat. In the sexually mature rat epididymis
(100-day-old), neither T35 nor F37 was observed in the three segments, but the appearance of several
newly synthesized polypeptides and the enhancement of radioactivity in pre-existent polypeptides were
noted particularly in the caput segment. Moreover, polypeptide EH52 appeared only in the caput
segment. Indirect immunofluorescence and immunogold electron microscopy provided additional data
that a polyclonal antiserum against Sertoli cell specific polypeptide S35 recognized the presence of
immunoreactive multivesicular bodies in the epididymal principal cells of sexually immature rats but not
of sexually mature ones. The timing of disappearance of T35 and F37 indicates that both polypeptides
may have functional significance at the onset of spermatogenesis and sperm maturation in the testis and

© 1990 Zoological Society of Japan

epididymis, respectively.

INTRODUCTION

It is well known that post-testicular maturation
of spermatozoa requires the specialized epididy-
mal environment which is regulated by selective
secretory and absorptive functions of this organ.
The mechanism by which the epididymis induces
the functional maturation of spermatozoa has fo-
cused mainly on the interactions beween sperm
and proteins produced by the epididymis [1, 2]. A
number of electrophoretic studies have investi-
gated specific proteins released from the rat epi-
didymis [3-9]. However, to date, no studies have
been concerned with protein secretory patterns
during the development of the rat epididymis.

Shabanowitz et al. [10] reported both age- and
spermatogenic stage-dependent secretory patterns
of [*S]methionine-labeled proteins of rat fetal
testes and seminiferous tubules during testicular
development. We also found several significant

Accepted November 21, 1989
Received October 13, 1989

similarities of protein secretory patterns between
testis and epididymis in the fetal rat [11].

In the present study, the time dependent chase
[S]methionine labeling of secretory proteins
accumulated in incubation medium of the rat epi-
didymis of various developmental stages was
analyzed by two-dimensional polyacrylamide gel
electrophoresis and autoradiography. The protein
secretory patterns were compared between testis
and epididymis during fetal and postnatal develop-
ment as well as between sexually immature (50-
day-old) and mature (100-day-old) rat epididymal
three segments (caput, corpus and cauda). In
addition, immunocytochemical studies were car-
ried out to examine whether a polyclonal anti-
serum against rat Sertoli cell specific polypeptide
S35, which immunoreacted with the acrosome and
tail of rat spermatids [12] and immunoprecipitated
a polypeptide of about Mr 35,000 of ovarian
sample [11], could detect antigenic sites in the rat
epididymis.

682 H. Uepa, T. HIRANO AND S. Fusimoto

MATERIALS AND METHODS

Isolation and [*°S|methionine labeling of tissues

Twenty white rats (Wistar) were purchased from
Seiwa Experimental Animals (Fukuoka, Japan).
Epididymides and testes were isolated from fetal
(18- and 20-days of gestation) and postnatal (0-,
10- and 20-day-old) rats, and placed in Hanks’
balanced salt solution (HBSS; Sigma, St. Louis,
USA). Epididymides from 50- and 100-day-old
rats were dissected into three segments (caput,
corpus and cauda), and placed in HBSS. Each
tissue was cut into small pieces (about 2 mm”) with
a razor blade and transferred to tissue culture wells
(17x 16 mm; Corning, New York, USA) contain-
ing 1 ml of Eagle’s minimum essential medium
with 10% of the normal concentration of
methionine (10% Met-MEM; Sigma). This
medium was supplemented with non-essential ami-
no acids (0.1 mM), antibiotics (penicillin 100 U/ml
and streptomycin 100 ug/ml) and 100 «zci/ml of
[*°S]methionine (New England Nuclear, Boston,
USA, Specific activity: 1100 Ci/mM). For a timed
series of chase labeling, epididymal tissues from
20-day-old rats were radiolabeled for 4, 8, 12 and
16 hr at 32°C in humidified atmosphere of 5% CO,
in air. The other tissue specimens were labeled for
16-18 hr in the same condition. After incubation,
the medium was removed and centrifuged at
15,600 xg for 5 min to remove cell breakdowns
and debrises. The supernatant fluid was removed,
frozen at —70°C and lyophilized. Samples were
reconstituted in 100 yl lysis buffer [13]. Trichlor-
oacetic acid-precipitable radioactivity was deter-
mined as previously reported [14].

Two-dimensional polyacrylamide gel electrophore-
sis (2D-PAGE) and autoradiography

Aliquots of samples containing 1,000,000 cpm
were fractionated by 2D-PAGE according to the
methods of O’Farrell [13] and Kierszenbaum et al.
[15] using a Protein II 2D system (Bio-Rad, South
Richmond, USA) except for a timed series of
chase labeling experiment in which the same
volume (40 wl) of radiolabeled proteins was ap-
plied. Isoelectric focusing gels contained ampho-
lytes (pH 5-8: 4%, pH3-10: 1.5%, pH2.5-5:

1%, pH6.5-9: 1%, Pharmacia Fine Chemicals,
Uppsala, Sweden). Samples were focused in 3x
130 mm tube gels at 400 constant volts for 16 hr
followed by 1 hr at 800 volts (7,200 volt-hr). After
20 min incubation in equilibration buffer, the
second dimension was run in a slab gel consisting
of 5-15% logarithmic gradient of polyacrylamide
and 5% stacking gel. Molecular weight markers
(Pharmacia) were run in a separate lane during the
second dimension of the electrophoretic proce-
dure. Standards of known molecular weight (Mr)
and isoelectric point (pl) were resolved by 2D-
PAGE in the same gels containing radiolabeled
samples. Coomassie blue-stained gels containing
these markers were used for determination of both
Mr and pl.

After Coomassie blue staining, 2D-PAGE gels
were destained, rinsed in distilled water, soaked in
Amplify (Amersham, Arlington Heights, USA)
for 30 min, and dried on filter paper using a gel
dryer. Dried gels were exposed to Kodak X-O mat
AR film (Eastman Kodak, Rochester, USA) for
1-2 weeks at —70°C and developed in Kodak
GBX X-ray film developer.

The method for determination of comparative
polypeptide radioactivity was previously reported
[14]. :

Indirect immunofluorescence and immunoelectron
microscopy

For immunofluorescence, epididymal tissues
were fixed in 4% paraformaldehyde in phosphate-
buffered saline (PBS) for 2 hr and embedded in
Historesin (LKB, Bromma, Sweden). Approx-
imately one 4m sections were etched with xylene,
treated with 2.5% trypsin in PBS, and im-
munoreacted with a polyclonal antiserum to S35
(working dilution 1:50) for 24hr at 4°C. The
production and characterization of anti-S35 serum
were previously described [16]. The sections were
rinsed with PBS, and then reacted with goat anti-
rabbit IgG conjugated with rhodamine (1: 100,
Cappel, Malvern, USA) for 1 hr at room tempera-
ture. Sections were examined on a Zeiss Axioskop
fluorescent microscope and images were recorded
on Kodak Tri-X film.

For immunoelectron microscopy, epididymides
were fixed in 2% paraformaldehyde and 1% glu-

Protein Secretion

taraldehyde in 0.1 M phosphate buffer, embedded
in Lowicryl K4M (Polaron Equipment, Watford,
England), and polymerized in an_ ultraviolet
polymerizer (Dosaka EM, Kyoto, Japan) at 4°C.
Ultrathin sections of gold interface color were
collected on nickel grids and placed on a drop of
0.1% bovine serum albumin (BSA)-PBS. Sections
were immunoreacted with anti-S35 serum diluted
at 1:500 in 0.1% BSA-PBS for 2hr at room
temperature. After rinsing in PBS, the grids were
reacted with goat anti-rabbit IgG-coated colloidal
gold of 15 nm (Janssen Life Sciences, Olen, Bel-
gium) diluted at 1: 100 in 0.1% BSA-PBS for 1 hr
at room temperature. After rinsing in PBS and
distilled water, the grids were stained with 5%
uranyl acetate in distilled water for 5 min, and
examined on a JEM 100 CX electron microscope.
The specificity of the immunocytochemical label-

47 53

94= ~ ieee
67" = - :
43= oe fsa. -

=. -
30= : A —
20=

12 hr

eccrine

of Rat Epididymis 683

ings was confirmed by substitution of normal rab-
bit serum or PBS for anti-S35 serum and increasing
dilutions of either first or second antibody.

RESULTS

Time-dependent chase labeling

Epididymal tissues from 20-day-old rats were
labeled for 4, 8, 12 and 16 hr with [*°S]methionine,
and secretory proteins accumulated in the culture
medium were resolved by 2D-PAGE and auto-
radiography. Figure 1 shows [*°S]methionine-
labeled proteins at four labeling times as a repre-
sentation of their synthetic and secretory time
sequence. Two polypeptides, designated as T35
(Mr 35,000/pI 4.67) and F37 (Mr 37,500/pI 4.63)
were detected at all labeling times, and their

45 463 47 #53
1 a t ] |
o —
@ vs
oo
a 8hr b

ee

lee

C 16hr

Fic. 1. Autoradiograms of time-dependent chase [*°S]methionine-labeled proteins (40 yl of samples) accumulated in
the medium of incubated 20-day-old rat epididymis for 4 hr (a), 8 hr (b), 12 hr (c) and 16 hr (d). Polypeptides F37
(open arrowhead), T35 (closed arrowhead) and SPI (circle) are observed in all autoradiograms. Numbers on the

left indicate molecular weight (Mr) standards x 10°

and those on the top do isoelectric point (pI).

684 H. UEDA, T. HIRANO AND S. Fusimoto

labeling intensity seemed to be in proportion to the
labeling time. Table 1 provides comparative cpm
values that confirm subjective estimates of [*°S]-
methionine-labeling intensity of these poly-
peptides as well as SPI (Mr 42,000/pI 4.50-4.55),
which was also found in the culture medium of rat
Sertoli cells [15], detected at four labeling times.

TABLE 1. Comparative radioactivity of [*°S]methi-
onine-labeled polypeptides

Time (hr) T35 F37 SPI
115 176 129

8 312 442 230

2 Si) 805 452

16 300 741 409

Cpm values for equivalent surface area of
radiolabeled spots (T35, F37 and SPI) were deter-
mined by paris of electrophoretic protein maps.

+
i
| a :
94= :
67= ce
43- ae
@ 7
i en
30 i.
20=
| C
Epididymis

Fic. 2.

per eenanborer anes

These findings indicated that the synthesis and
secretion of T35 and F37 progressively increase in
degree until 12 hr and remain stable at 16 hr.

Comparison of protein secretory patterns be-
tween testis and epididymis during fetal and
postnatal development

Fetal rats of 18- and 20-days gestation [°?S]-
Methionine-labeled protein accumulated in the
incubation medium of fetal testes and epididymis
generated a nearly identical pattern (Fig. 2).
Although the number of synthesized secretory
proteins in 20-days fetal tissues decreased as com-
pared to those in 18-days fetal tissues, the relative
labeling intensities of T35 and F37 became greater
in the late gestation than in the early gestation.

Postnatal sexually immature rats of 0-, 10- and 20-
day-old [°°S]Methionine-labeled protein secre-

Testis

Autoradiograms showing radiolabeled proteins (1,000,000 cpm) accumulated in the medium of incubated

fetal epididymides (a, c) and tested (b, d) from 18- (a, b) and 20-days (c, d) gestational rat fetuses. All
autoradiograms display polypeptides F37 (open arrowhead), T35 (closed arrowhead) and SPI (circle).

Protein Secretion

pl
45463 47 53
Mir | I |
94- q
67= =
43- ee -
| 2S ee
30- ee =
A
20-
a
94=
67= ——
i ass
30= A
20=
C
94-= : i
67= tt
43= @ en
30 ey - z
20= : |
eC
Epididymis

of Rat Epididymis 685

45463 47 53
ee i i

ag OF

Testis

Fic. 3. Autoradiograms illustrating radiolabeled proteins (1,000,000 cpm) accumulated in the medium of incubated
neonatal epididymides (a, c, e) and testes (b, d, f) from 0- (a, b), 10- (c, d) and 20-day-old (e, f) rats. The open
arrowhead and the closed arrowhead indicate the expected position of F37 and T35, respectively. The circle
indicates SPI. The brackets indicate Sertoli cell specific secretory proteins, S70, S45, and S35.

tory patterns in postnatal epididymal tissues were
essentially similar to those in fetal epididymis (Fig.
3a, c,e). During this time period, several proteins
in testicular tissues changed in their [°°S]methio-
nine-labeling characteristics: T35 decreased grad-

ually its radiolabeling intensity at day 10 (Fig. 3d)
and day 20 (Fig. 3f). The identification of F37 was
difficult because $35, one of the polypeptide com-
ponents of S45-S35 heterodimeric protein, over-
lapped with this polypeptide (Fig. 3d). S45 and

686 H. Uepa, T. HIRANO AND S. FUJIMOTO

S70 were also visible during this period (Fig. 3d, f). The protein secretory patterns of three segments
in the epididymal tissues from 50-day-old rat were
compared with those from 100-day-old rat. At day
50, few spermatoza were present in the lumen of
the epididymis. Both T35 and F37 were still

Comparison of protein secretory patterns of epi-
didymal three segments (caput, corpus and
cauda) between 50- and 100-day-old rats

i

45 463 47 53
Ly L i ee

[=]
67= OIE Bs ‘
ee O Saas _ Corpus
43-= oe. — ~** y=
30= A : : fi |
20- : : -
C d
94= =]
67= — er ae < |
43 = eee | | 0 & 4 | Cauda
Bei. vie | a a
30= ei : aay ;
20= :
e f
50d 100d

Fic. 4. Autoradiograms depicting radiolabeled proteins (1,000,000 cpm) accumulated in the medium of incubated

epididymal three segments, caput (a, b), corpus (c, d) and cauda (e, f) from 50- (a, ¢, e) and 100-day-old (b, d, f)
rats. The open arrowhead and the closed arrowhead identify the expected position of F37 and T35, respectively.
Boxes indicate pre-existed polypeptide spots that increase their labeling intensity. Arrow, EHS52; Circle, SPI;
Rhombus, E50; Vectrial bracket, a group of high Mr protein.

Protein Secretion of Rat Epididymis 687

detected, and no distinct differences were
observed among these three segments (Fig. 4a, c,
e). At day 100, the lumen of epididymis was filled
with spermatoza, and neither T35 nor F37 was
detected in the three segments (Fig. 4b, d, f). E50
(Mr 50,000/pP 4.56) and a group of high molecular
weight proteins (above Mr 94,000) were observed
in the three segments (Fig. 4b, d, f). The enhance-
ment of radioactivity was prominent in E16 (Mr
16,000/pI 4.46), E20 (Mr 20,000/pI 4.55), E 25 (Mr
25,000/pI 4.54), E68 (Mr 68,000/pI 4.62) and E70
(Mr 70,000/pI 4.63) particularly in the caput seg-
ment (Fig. 4b).Furthermore, EH 52 (Mr 52,000/pI
4.52) was detected only in the caput segment at
day 100 (Fig. 4b).

Immunocytochemical localizations of anti-S35
serum in the epididymis

Antiserum against polypeptide S35 secreted by
rat Sertoli cells was used for detection of possible

wal ee
Re eee
Pv

2 . <
i 4 i > \ xy j
4 a ee td

a
am ¢ J

To ae

antigenic determinants in the epididymis. Figure 5
shows phase contrast (a, b) and immunofluores-
cent localizations of antigenic sites (c, d) in epi-
didymis from the caput segment at day 20 (a, c)
and day 100 (b, d). Immunoreactive sites were
observed on the straight lined portion of stereocilia
and spot-like immunoreactivities were seen in the
cytoplasm of the epididymal epithelial cells at day
20 (Fig. 5c). At day 100, spot-like immunoreactive
sites were very few in number, but the acrosome
and the tail of spermatozoa in the lumen were
strongly immunoreactive (Fig. 5d).

Immunogold electron microscopy confirmed the
present immunofluorescent results. The im-

munoreactive gold particles to S35 were localized
on the multivesicular bodies contained amorphous
materials of 300-700 nm in diameter in the prin-
cipal cells of the epididymis in the caput segment at
day 20 (Fig. 6a) but not at day 100 (Fig. 6b).

Fic. 5.

Indirect immunofluorescence of epididymis in the caput segment at 20- (a, c) and 100-day-old (b, d) rats.

Phase contrast micrographs (a, b) correspond to fluorescence micrographs (c, d). White arrows indicate spot-like

immunoreactive sites. 600.

688 H. UepA, T. HIRANO AND S. FusIMoto

Fic. 6.

DISCUSSION

The present study involved a timed series of
chase radiolabelings of secretory proteins of the
sexually immature rat epididymis and describes the
protein secretory patterns durings its develop-
ment. The results indicate that the synthesis and
secretion of proteins specific to the epididymal
tissues can be accomplished very rapidly and that
they then remain stable at 16hr incubation. A
similar progressive incorporation of [*°S]methio-
nine-labeled proteins was found in the rat cultured
Sertoli cells [15].

From fetal to sexually immature rats, the epi-
didymis displayed a uniform protein secretory pat-
tern: T35 and F37 were always the predominant
secretory products. In contrast, the secretion of
both polypeptides from the rat testis was limited
only in fetal and newborn stages. The gradual
disappearance of radiolabeling intensity of T35
and the enhancement of that of $35, which caused
resulting indistinctness of F37, coincided with the
onset of spermatogenesis [10, 17]. T35 and F37
also disappeared in the three segments of the

Immunogold electron micrographs of multivesicular bodies from 20- (a) and 100-day-old (b) rat epididymis in
the caput segment. 37,000.

sexually mature rat epididymis. This finding is
interesting if we take into consideration the cor-
relation between the disappearance of T35 and F37
and the occurrence of sperm maturation in the
epididymis. No studies have pointed out what kind
of specific proteins disappear at the course of
sperm maturation.

It has been suspected that specific proteins pro-
duced by the epididymis of sexually mature rat
associate with spermatozoa and play a functional
role in sperm maturation [18-20]. There have
been many studies that acidic epididymal glycopro-
tein (Mr 32,000) [4, 6, 7] and 22 KDa protein (Mr
22,000) [3, 5, 8] have a functional role in the
development of fertilizing ability by the spermato-
zoa. Rifkin and Olson [21] have shown that
HIS-50 (Mr 50,000) binds to the sperm surface
during maturation. Some newly synthesized and
secreted proteins, such as E50 and a group of high
molecular weight proteins in the three segments as
well as EH52 only in the caput segment, were
detected in the present sexually mature rat epi-
didymis. The enhancement of radioactivity was
observed in several pre-existent polypeptide spots.

Protein Secretion of Rat Epididymis 689

However, the functional significance of these
polypeptides to the sperm maturation remains to
be elucidated.

Immunocytochemical studies revealed that mul-
tivesicular bodies found in the epididymal princi-
pal cells contained a polypeptide that crossreacts
with anti-S35 serum. This antiserum was raised
against rat Sertoli cell specific polypeptide S35,
and had possible antigenic homology in a
polypeptide of about Mr 35,000 in incubated ova-
rian samples [11]. The present immunocytoche-
mical data confirmed the developmental protein
secretory patterns of the rat epididymis. The
disappearance of T35 and F37 was coincident with
the non-crossreaction of multivesicular bodies with
anti-S35 serum in the 100-day-old rat epididymis.
The multivesicular bodies have been considered to
be the storage site of androgen-binding protein by
means of immunocytochemical [22, 23] and auto-
radiographic techniques [24]. This component was
also believed to be involved in an active uptake of
proteins [25]. The interpretation as to the positive
immunolocalization of multivesicular bodies lead
us to two possibilities; (1) this component may
absorb proteins released from testicular fluid, (2)
this component may secrete specific proteins which
crossreact with anti-S35 serum. The latter possibil-
ity argues for the present data in the sexually
mature epididymis that neither T35 nor F37 was
present and that the multivesicular body did not
immunoreact with anti-S35 serum. We have found
that this antiserum immunoreacts with the acro-
some [12] and keratin-like components of sperm
tail [16]. In the present sexually mature epididy-
mis, the acrosome and tail of spermatozoa showed
immunoreactions with anti-S35 serum.

In summary, the present study indicates that the
protein secretory pattern of the rat testis and
epididymis is changed at the onset of sperma-
togenesis and sperm maturation, respectively. It is
worthwhile speculating that the shift in protein
synthesis and secretory mechanism from F37 and
T35 to other proteins is involved to some extent in
the initiation of spermatogenesis and sperm
maturation. The functional significance of T35 and
F37 is under investigation by several molecular
biological approaches.

ACKNOWLEDGMENTS

We thank Professor Abraham L. Kierszenbaum for
his generous supply of the antisera. We also thank Mr.
Tomoko Nishino for her technical assitance, and Ms.
Toyono Nobukuni for typing the manuscript. This study
was supported in part by a Grant-in-Aid for scientific
research from the Ministry of Education, Science and
Culture of Japan (No. 63770037).

REFERENCES

1 Bedford, J. M. (1975) Maturation, transport and
fate of spermatozoa in the epididymis. In “Hand-
book of Physiology, Sec. 7: Endocrinology, Vol 5,
Male Reproductive System”. Ed. by D. W. Hamil-
ton and R. O. Greep, American Physiological Socie-
ty, Washington D.C., pp. 303-317.

2 Austin, C. R. (1985) Sperm maturation in the male
and female genital tracts. In “Biology of Fertiliza-
tion”. Ed by C. B. Metz and A. Monroy, Academic
Press, New York, pp. 121-155.

3 Cameo, M. S. and Blaquier, J. A. (1976) Andro-
gen-controlled specific proteins in rat epididymis. J.
Endocrinol., 69: 47-55.

4 Lea, O. A., Petrusz, P. and French, F. S. (1978)
Purification and localization of acidic epididymal
glycoprotein (AEC): a sperm coating protein se-
creted by the rat epididymis. Int. J. Androl., 2: 592-
607.

5 Brooks, D. E. and Higgins, S. J. (1980) Character-
ization and androgen-dependence of proteins associ-
ated with luminal fluid and spermatozoa in the rat
epididymis. J. Reprod. Fertil., 59: 363-375.

6 Faye, J. C., Duguet, K., Mazzuca, M. and Bayard,
F. (1980) Purification, radioimmunoassay, and im-
munocytochemical localization of a glycoprotein
produced by the rat epididymis. Biol., Reprod., 23:
423-432.

7 Wong, P. Y. D. and Tsang, A. Y. F. (1982) Studies
on the binding of a 32 K rat epididymal protein to rat
epididymal spermatozoa. Biol. Reprod., 27: 1239-
1246.

8 Jones, R. and Brown, C. R. (1982) Association of
epididymal secretory proteins showing a-
lactalbumin-like activity with the plasma membrane
of rat spermatozoa. Biochem. J., 206: 161-164.

9 Shabanowitz, R. B. and Killian, G. J. (1987) Two-
dimensional electrophoresis of proteins in principal
cells, spermatozoa, and fluid associated with the rat
epididymis. Biol. Reprod., 36: 753-768.

10 Shabanowitz, R. B., DePhilip, R. M., Crowell, J.
A., Tress, L. L. and Kierszenbaum, A. L. (1986)
Temporal appearance and cyclic behavior of Sertoli
cell-specific secretory proteins during the develop-
ment of the rat seminiferous tubule. Biol. Reprod.,

11

12

13

14

15)

16

7/

18

690

35: 745-760.

Ueda, H. and Fujimoto, S. (1987) Protein secretory
patterns during the development of the rat ovary. In
“Proc. of Ist Cong. AOSCE”. Ed by E. Ohnishi, Y.
Nagahama and H. Ishizaki, Nagoya Univ Corp,
Nagoya, pp. 169-170.

Ueda, H., Abdullah, M., Tres, L. L. and Kierszen-
baum, A. L. (1987) Immunocytochemical and im-
munogold electron microscopic localization of two
Sertoli cell secretory proteins in developing sperma-
tids. J. Cell Biol., 105: 167a.

O'Farrell, P. H. (1975) High resolution two-
dimensional gel electrophoresis of proteins. J. Biol.
Chem., 250: 4007-4021.

Ueda, H., Tres, L. L. and Kierszenbaum, A. L.
(1988) Culture patterns and sorting of rat Sertoli cell
secretory proteins. J. Cell. Sci., 89: 175-188.
Kierszenbaum, A. L., Crowell, J. A., Shabanowitz,
R. B., DePhilip, R. M. and Tres, L. L. (1986)
Protein secretory patterns of rat Sertoli and peri-
tubular cells are influenced by culture conditions.
Biol. Reprod., 35: 239-251.

Abdullah, M., Tres, L. L., Ueda. H., Hu, P. C. and
Kierszenbaum, A. L. (1988) Antigenic homology
between rat sperm tail polypeptides and Sertoli cell
secretory proteins. Mol. Cell. Biochem., 81: 165-
176.

Kierszenbaum, A. L., Abdullah, M., Ueda, H. and
Tress, L. L. (1987) Spermatogenesis in vitro: Sear-
ching for in vivo correlates. In “Regulation of Ova-
rian and Testicular Function”. Ed. by V. B. Mah-
esh, D. S. Dhindsa, E. Anderson and S. P. Kalra,
Prenum Press, New York, pp. 535-560.

Brooks, D. E. (1985) Characterization of a 22 KDa
protein with widespread tissue distribution but

Wy)

20

Pa)

Vip)

753)

24

25

H. UEDA, T. HIRANO AND S. Fusimoro

which is uniquely present in secretions of the testis
and epididymis and on the surface of spermatozoa.
Biochim. Biophys. Acta, 841: 59-70.

Klinefelter, G. R. and Hamilton, D. W. (1985)
Synthesis and secretion of proteins by perifused
caput epididymal tubules, and association of se-
creted proteins with spermatozoa. Biol. Reprod.,
33: 1017-1027.

Olson, G. E. and Hinton, B. T. (1985) Regional
differences in luminal fluid polypeptides of the rat
testis and epididymis revealed by two-dimensional
electrophoresis. J. Androl., 19: 20-35.

Rifkin, J. M. and Olson, G. E. (1985) Characteriza-
tion of maturation-dependent extrinsic proteins of
the rat sperm surface. J. Cell Biol., 100: 1582-1591.
Feldman, M., Lea, O. A., Petrusz, P., Tres, L. L.,
Kierszenbaum, A. L. and French, F. S. (1981)
Androgen-binding protein. Purification from rat epi-
didymis, characterization and immunocytochemical
localization. J. Biol. Chem., 256: 5170-5175.
Pelliniemi, L. J.. Dym, M., Gunsalus, G. L., Musto,
N. A., Bardin, C. W. and Fawcett, D. W. (1981)
Immunocytochemical localization of androgen-
binding protein in the male rat reproductive tract.
Endocrinology, 108: 925-931.

Gerard, A., Khanfri, J., Gueant, J. L., Fremont, S.,
Nicolas, J. P., Grignon, G. and Gerard, H. (1988)
Electron microscope radioautographic evidence of -
in vivo androgen-binding protein internalization in
the rat epididymal principal cells. Endocrinology,
122: 1297-1307.

Cooper, T. G., Yeung, C. H. and Bergmann, M.
(1987) Protein transport to the epididymal lumen.
Cell Tissue Res., 248: 527-530.

ZOOLOGICAL SCIENCE 7: 691-696 (1990)

Protein Secretory Patterns during the Development
of the Rat Ovary

HirRosHI UEDA, YOSHIO FUKUI, HIDETOSHI ARAKI

and SUNAO FUJIMOTO

Department of Anatomy, University of Occupational and
Environmental Health, School of Medicine,
Kitakyushu 807, Japan

ABSTRACT—Two dimensional polyacrylamide gel electrophoresis and autoradiography were used to
study the time course of [*°S]methionine-labeled proteins accumulated in the incubation medium of rat
ovaries during their development. In 18-days gestation rat fetuses, the ovary secreted polypeptide
components very similar to those of the testis of male fetal rats, except that Sertoli cell secretory protein
S70 was detected only in testicular samples. During postnatal development, polypeptides T35 (Mr
35,000/pI 4.67) and F37 (Mr 37,500/pI 4.63) continued to be expressed in the ovary throughout
development. Antiserum against polypeptide S35, a component of Sertoli cell secretory S45-S35
heterodimeric protein, recognized immunoreactive sites in zona pellucida detected by indirect im-
munofluorescence. Results from this study demonstrate a contrast in the temporal expression of
testicular and ovarian secretory polypeptides T35 and F37, and a possible antigenic homology between

© 1990 Zoological Society of Japan

zona pellucida material and polypeptide S35.

INTRODUCTION

One of the major differences between mamma-
lian female and male gametogenesis is the timing
of proliferation and differentiation of germ cells.
While female gametogenesis starts during fetal life
[1-3], the onset of male gametogenesis takes place
at puberty [4, 5]. The molecular and biochemical
signals responsible for this gonadal dimorphic pat-
tern have not been identified. The identification of
bioregulatory molecules which may coordinate
events during the initiation and progression of
oogenesis and spermatogenesis is of considerable
interest.

Shabanowitz et al. [6] have reported that tem-
poral appearance and cyclic behavior of
[*°S]methionine-labeled proteins during the de-
velopment of the rat testes and seminiferous
tubules. To compare male and female gonadal
dimorphism, the developmental expression of ova-
rian secretory proteins is highly interested.
However, there have been no studies concerning

Accepted February 8, 1990
Received December 11, 1989

protein secretory patterns during the development
of the rat ovary.

In this paper, the pattern of [*S]methionine-
labeled secretory proteins of the developing rat
ovary was analyzed by two-dimensional polyacryl-
amide gel electrophoresis and autoradiography. In
addition, immunological studies were carried out
to obtain the evidence that a polyclonal antiserum
against rat Sertoli cell specific polypeptide S35
could recognize antigenic sites in the ovary.

MATERIALS AND METHODS

Isolation and [*°S |\methionine labeling of tissues

Pregnant rats were purchased from Holtzman
Co. (Madison, USA) and Seiwa Experimental
Animals (Fukuoka, Japan). Ovaries and testes
were isolated from 18-days gestation rat fetuses
and placed in phenol red-free Hanks’ balanced salt
solution (HBSS; Sigma, St. Louis, USA). Each
organ was cut into small pieces with a razor blade
and transferred to tissue culture wells (7 x 10 mm)
containing 300 #1 of Eagle’s minimum essential
medium with 10% of the normal concentration of

692 H. Uepa, Y. Fuxut et al.

methionine (10% Met-MEM) and supplemented
with glutamine (4 mM), sodium pyruvate (1 mM),
non-essential amino acids (0.1mM), antibiotics
(penicillin 100 U/ml and streptomycin, 100 g/ml),
and 200 ~Ci/ml of [*°S]methionine (New England
Nuclear, Boston, USA, Specific activity: 1,120
Ci/mM). Ovaries were isolated from postnatal (0-
and 5-day-old), sexually-immature (10-, 15- and
25-day-old), and mature (90-day-old) rats, placed
in HBSS, cut into small pieces (about 1-2 mm),
and transferred to 10% Met-MEM for radiolabel-
ing. For a timed series of chase labeling, ovaries
from 20-day-old rats were radiolabeled for 4, 8, 12
and 16hr at 37°C in a humidified atmosphere of
5% CQO) in air. The other tissue specimens were
labeled for 16-18 hr in the same condition. After
incubation, the medium was removed and centri-
fuged at 15,600 X g for 5 min to remove cell break-
downs and debrises. The resulting supernatant
was frozen at —70°C and lyophilized. Samples
were reconstituted in 100 sl lysis buffer [7]. Trichlo-
roacetic acid-precipitable radioactivity was deter-
mined as previously reported [8].

Two-dimensional polyacrylamide gel electrophore-
sis (2D-PAGE) and autoradiography

Fither the same volume of radiolabeled samples
(30 wl) in a timed series of chase labeling experi-
ment or aliquots of radiolabeled samples
(1,000,000 cpm) in the other experiment were frac-
tionated by 2D-PAGE. The 2D-PAGE procedure
has been described in detail [9].

Indirect immunofluorescence

For immunofluorescence, ovarian samples were
fixed in 3.7% paraformaldehyde in phosphate buf-
fered saline (PBS) for 30 min and embedded in
Historesin (LKB, Bronma, Sweden). Plastic sec-
tions were prepared and etched as reported [9].
About one um sections were exposed to polyclonal
antisera against S35 and S45 (working dilution
1:50) for 24 hr at 4°C. Both the preparation and
the characterization of anti-S35 and anti-S45 sera
have been reported [10]. The sections were rinsed
with PBS, reacted with goat anti-rabbit IgG conju-
gated with fluorescein isothiocyanate (working
dilution 1:25; Cappel, Malvern, USA) for 1 hr at
room temperature, rinsed with PBS, and mounted

with Elvanol. The specificity of the immunocy-
tochemical labelings was confirmed by increasing
dilutions of either first or second antibodies, and
omission of first antibody. Sections were observed
on a Leitz Ortholux II photomicroscope and im-
ages were recorded on Kodak Tri-X film (Eastman
Kodak, Rochester, USA).

RESULTS

Timed series of chase labeling

Ovarian tissues from 20-day-old rats were
radiolabeled for 4, 8, 12 and 16 hr, and secretory
proteins accumulated in the culture medium were
analyzed by 2D-PAGE and autoradiography.
[*°S]Methionine-labeled protein secretory patterns
in 20-day-old rat ovary were nearly identical with
those in 15-day-old rat ovary (Fig. 2-15). Table 1
shows cpm values that provide subjective estimate
of [°°S]methionine-labeling intensity of T35 (Mr
35,000/pI 4.67) and F37 (Mr 37,500/pI 4.63).
These data demonstrated the progressive increase
of radioactivities of T35 and F37 until 16 hr.

TaBLE 1. Radioactivity of [*°S]methionine-labeled
polypeptides in a timed series of chase labeling
experiment
Time (hr) 135 F37

4 104 141
8 262 289
12 481 Sy)
16 764 643

Cpm values for equivalent surface area of
radiolabeled spots (T35 and F37) were determined
by pairs of electrophoretic protein maps.

Protein secretory patterns during the development
of the rat ovary

A 2D-PAGE analysis of [*S]methionine-
labeled proteins accumulated in the incubation
medium of ovary was compared with those of testis
from fetuses at the same age (18-days gestation),
and demonstrated almost similar polypeptide pat-
terns (Fig. 1). An exception was protein S70 which
was observed in testicular samples. Polypeptides

Protein Secretion of Rat Ovary

pl
463 —
Mir ae ! Ga
94- = :
= iS
67- 7? :

45 463

ovary

693

47 5,357

testis

Fic. 1. Autoradiograms of [*°S]methionine-labeled proteins accumulated in the medium of ovary and testis of
18-days gestational rat fetuses. All autoradiograms (1,000,000 cpm, 2 weeks exposure) display polypeptides SP1
(circle), T35 (closed arrowhead), and F37 (open arrowhead). The box identifies protein S70 observed in testicular
samples. Numbers at the left are Mr standards x 10° and those on the top represent pl.

94—

67—

D5

30-

20—*

45 463 4.7 5.3 57
!

=
= te Rae
4 <
neg Mee q :
ae Re ~ 2
: noe se
©. a3
et a

5
6
- eas
cit... fo.
+. /~ “ Vv : : ae
= A +o
25

eS

a = ‘° ee

10
iad > +te]ee
“s -
Gaye |
oe
90

Fic. 2. Developmental pattern of [°°S]methionine-labeled proteins accumulated in the medium of incubated ovaries
of 0-, 5-, 10-. 15-, 25- and 90-day-old rats. Polypeptides SP1 (circle), T35 (closed arrowhead), and F37 (open
arrowhead) can be observed in all autoradiograms (1,000,000 cpm; 2 weeks exposure). The vertical brackets
indicate a group of less acidic polypeptide becoming prominent at day 5.

694 H. Uepa, Y. Fuku! et al.

Fic. 3.

Immunocytochemical localization of zona pellucida antigenic sites using antisera against Sertoli cel secretory

protein S45-S35 heterodimeric protein. The arrow in B indicates the immunoreactive zona pellucida (depicted in
A by phase contrast microscopy). Anti-S45 serum fails to detect zona pellucida immunoreactivity.

SP1 (Mr 42,000/pI 4.50—4.55) as well as T35 and
F37 were identified in the autoradiograms of ovary
and testis.

During postnatal ovarian development, the pro-
tein secretory patterns of the ovary were essential-

ly similar to those of the fetal ovary (Fig. 2). The
appearance of a group of less acidic polypeptides
(Mr 43,000-30,000/pI 4.7-5.5) from postnatal day
5 on, and a decrease in the intensity of
polypeptides SP1, F37 and T35 at day 25 were two

Protein Secretion of Rat Ovary 695

exceptions.

Immunofluorescent localization of anti-S35 serum
in the ovary

Antisera raised against polypeptide components
of S45-S35 heterodimeric protein secreted by rat
Sertoli cells were used to determine possible anti-
genic sites in the ovary. An interesting finding was
the immunolocalization of antigenic sites in zona
pellucida with anti-S35 serum but not with anti-S45
serum (Fig. 3).

DISCUSSION

The present study indicated that minces of ova-
ries from fetal to adult rat incubated with
[S]methionine yielded a rather uniform protein
secretory pattern with minor age-dependent varia-
tions. This observation was a striking contrast with
the developing rat testis [6, 9, 11]. While in the
developing testis, large populations of spermato-
gonia, spermatocytes, and spermatids coexist in
seminiferous tubules, the developing ovary con-
tains a pool of non-growing oocytes and a number
of oocytes entering the phase stages of follicu-
logenesis. Although the growth of the oocyte and
its surrounding follicle is progressive and coordi-
nated, the population of germ cells is restricted to
late meiotic prophase oocytes. It is possible that
these developmental characteristics of the ovary
account for limited variations observed in the
protein synthetic patterns. Our studies have not
determined whether a particular polypeptide or
polypeptide patterns could be related to the fol-
licular growth. Of particular interest is the elec-
trophoretic relationship between polypeptide S35
of rat testis and polypeptide F37 of rat ovary. A
polypeptide with similar Mr/pI to F37 was
observed in incubated testes of fetal to neonatal
rats [6, 9]. During later testicular development,
the more prominent polypeptide S35 obscured
polypeptide F37, whose fate was difficult to deter-
mine using 2D-PAGE and autoradiography.

Indirect immunofluorescent microscopy showed
that zona pellucida contains a polypeptide that
crossreacts with anti-S35 serum. Abdullah et al.
[10] have reported that, in protein samples sec-
reted by rat Sertoli cells, anti-S35 serum does not

crossreact with polypeptide S45 and anti-S45
serum does not crossreact with polypeptide S35.
This lack of reciprocal antigenic homology be-
tween the two polypeptide components of $45-S35
heterodimeric protein suggests that existence of
different antigenic domains in the precursor of
S45-S35 heterodimeric protein. This possibility is
supported in the ovary by our finding that anti-S45
serum failed to immunoreact with zona pellucida
whereas immunoreactivity was detected with anti-
S35 serum. Whether the same precursor gives rise
to polypeptide F37, an antigen homologous to
polypeptide S35, is under investigation using
molecular biological techniques.

ACKNOWLEDGMENTS

We thank Professor Abraham L. Kierszenbaum for his
interest and support in this study. We also thank Ms.
Tomoko Nishino for her technical assistance, and Ms.
Toyono Nobukuni for typing the manuscript. This work
was supported in part by a Fogarty International Fel-
lowship to H.U. (1F05 TW0381) and a Grant-in-Aid for
Scientific Research from the Ministry of Education,
Science and Culture of Japan.

REFERENCES

1 Borum, K. (1961) Oogenesis in the mouse. A study
of meiotic prophase. Exp. Cell Res., 24: 495-507.

2 Ohno, S. and Smith, J. B. (1964) Role of fetal
follicular cells in meiosis of mammalian oocytes.
Cytogenesis, 3: 324-333.

3 Peters, H. (1978) Folliculogenesis in mammals. In
“The Vertebrate Ovary”. Ed. by R. E. Jones,
Plenum Press, New York, pp. 121-144.

4 Franchi, L. L. and Mandl, A. M. (1964) The
ultrastructure of germ cells in foetal and neonatal
male rats. J. Embryol. Exp. Morph., 12: 289-308.

5 Clermont, Y. (1972) Kinetics of spermatogenesis in
mammals: Seminiferous epithelium cycle and sper-
matogonial renewal. Physiol. Rev., 52: 198-235.

6 Shabanowitz, R. B., DePhilip, R. M., Crowell, J.
A., Tres, L. L. and Kierszenbaum, A. L. (1986)
Temporal appearance and cyclic behavior of Sertoli
cell-specific secretory proteins during the develop-
ment of the rat seminiferous tubule. Biol. Reprod.,
35: 745-760.

7 O’Farrell, P. H. (1975) High resolution two-
dimensional gel electrophoresis of proteins. J. Biol.
Chem., 250: 4007-4021.

8 Ueda, H., Tres, L. L. and Kierszenbaum, A. L.
(1988) Culture patterns and sorting of rat Sertoli cell

10

696

secretory proteins. J. Cell Sci., 89: 175-188.

Ueda, H. Hirano, T. and Fujimoto, S. (1990)
Changes in protein secretory patterns during the
development of the rat epididymis. Zool. Sci., 7:
679-688.

Abdullah, M., Tres, L. L., Ueda, H., Hu, P. D. and
Kierszenbaum, A. L. (1988) Antigenic homology
between rat sperm tail polypeptide and Sertoli cell

11

H. Uepba, Y. Fukui et al.

secretory proteins. Moll. Cell. Biochem., 81: 165-
176.

Kierszenbaum, A. L., Abdullah, M., Ueda, H. and
Tress, L. L. (1987) Spermatogenesis in vitro: Sear-
ching for in vivo correlates. In “Regulation of Ova-
rian and Testicular Function”. Ed. by V. B. Mah-
esh, D. S. Dhindsa, E. Anderson and S. P. Kalra,
Prenum Press, New York, pp. 535-560.

ZOOLOGICAL SCIENCE 7: 697-704 (1990)

Stage-Dependent Changes in the Activity of the
Prothoracicotropic Hormone (PTTH) in the
Brain of the Asian Comma Butterfly,
Polygonia c-aureum L.

KATSUHIKO ENDO, YASUHIRO FUJIMOTO, TAKAKATSU MASAKI

and Kann Kumacar’

Environmental Biology Laboratory, Institute of Biology, Faculty of
Sciences, and ‘Biological Institute, Faculty of Liberal Arts,
Yamaguchi University, Yamaguchi 753, Japan

ABSTRACT— In addition to a neuroendocrine factor producing summer morphs (SMPH), ecdysteroids
in the pupal hemolymph play a significant role in summer morph development of Polygonia c-aureum L.
The hemolymph ecdysteroids appearing in the early pupal stage and forcing the pupae to undergo
metamorphosis are secreted by the prothoracic glands. The ecdysteriod secretion is governed by one (or
two) neuroendocrine factor, named (small and big) prothoracicotropic hormones (PTTHs) in insects.

Activity of the PTTH in the brain extracts of P. c-aureum as well as the SMPH activity fluctuated
stage-dependently during the larval and pupal stages. The fluctuating pattern of small PTTH activity
was also different from that of SMPH activity in LD-insects.

In P. c-aureum, production of the small PTTH as well as its secretion seems to be enhanced by long
days existing during the larval stages as in the case of the SMPH of this butterfly. But, a factor in the
brain extracts of B. mori may produce autumn-morph-producer-originated summer morphs of P.
c-aureum without the assistance of the B. mori PTTHs, since they did not show any tropic action on the

© 1990 Zoological Society of Japan

glands from P. c-aureum 5th instar larvae.

INTRODUCTION

The Asian comma butterfly, Polygonia c-
aureum L., exhibits seasonal dimorphism, i.e.,
summer and autumn morphs. Development of the
seasonal morphs is governed by photoperiod and
temperature during the larval stages [1]. The
physiological mechanism underlying the photo-
periodic control of seasonal morph development
involves a neuroendocrine factor producing sum-
mer morphs (SMPH), which was produced in the
medial neurosecretory cells of the brain and re-
leased into hemolymph in the early pupal stage [1-
Si.

Brain extracts of Bombyx mori moths showed
strong SMPH activity in P. c-aureum as the brain
extracts of P. c-aureum short-day pupae. The

Accepted October 30, 1989
Received February 25, 1989

factor showing SMPH activity in the brain extracts
of Bombyx mori moths was almost the same in
nature as the SMPH of P. c-aureum as far as the
physico-chemical characteristics to which we con-
cerned [4-6].

Hemolymph ecdysteroids appearing in the early
pupal stage were thought to play a significant role
in the summer morph development of P. c-aureum
[4] as in the case of several other species of
butterflies, i.e. L. phlaeas daimio [7|, Arashnia
burejana [8] and A. levana [9]. The hemolymph
ecdysteroids are secreted by the prothoracic
glands, whose secreting activity is governed by one
or two neuroendocrine factors, named prothoraci-
cotropic hormones (PTTHs) [10, 11]. One of the
PTTHs being larger than the other in the molecu-
lar size (big PTTH) was shown to be produced in
the lateral neurosecretory cells of the brain of the
brain of Manduca sexta [12]. In contrast, the
neurosecretory cells responsible for the other,

698 K. Enpbo, Y. Fusimoto et al.

named 4K-PTTH, small PTTH or bombyxin in B.
mori [14], were medial neurosecretory cells localiz-
ing in the pars intercerebralis of the brains of B.
mori [11, 15]. The cells producing the small PTTH
in B. mori may correspond to the same cells as the
SMPH-producing cells of P. c-aureum [3].

The present study was made first to show
whether or not the two PTTHs varying in the
molecular size exist in the brain extracts of P.
c-aureum pupae. Then, the study was extended to
know the stage-dependent fluctuations of the small
PTTH and total PTTH activities of the brain, on
which the role of the PITHs played in summer
morph development of P. c-aureum was discussed.

MATERIALS AND METHODS

Animals

P. c-aureum and Papilio xuthus collected in
Yamaguchi City were used. Eggs and larvae of P.
c-aureum as well as those of P. xuthus were
grouped (n=400-500), put in a rearing container
of transparent plastic (13x206cm°*) and sub-
jected either to long-day conditions, consisting of
alternating 16 hr light and 8 hr dark periods (16L-
8D), or to short-day conditions of 8L-16D. The
rearing containers were placed in cabinets (62 x 39
110 cm*) with constant temperature (20°C or
25°C) and illuminated by two 20 W white fluores-
cent tubes, which were controlled by a 24hr
time-switch. In the light period, light intensity was
about 500 lux on the rearing containers. Larvae of
P. c-aureum were fed on leaves of Humulus japo-
nicus and those of P. xuthus were grown on leaves
of Sagara ailanthoides. Fresh leves were provided
daily in the light period. Under long-day condi-
tions (16L-8D) at 20°C and 25°C, larvae and pupae
of P. c-aureum all developed into summer morphs,
whereas, under short-day conditions (8L-16D) at
20°C, they all developed into autumn morphs.
Pupae which were reared under 8L-16D at 25°C
developed into butterflies of summer (90-95%)
and autumn morphs (5-10%).

Larvae and pupae raised from the egg stage
under 16L-8D at 25°C were referred to as LD-
larvae and LD-pupae, whereas those reared under
8L-16D at 20°C were referred to as SD-larvae and

SD-pupae, respectively.

Extraction of SMPH and PTTHs

Four hundred brains were obtained from pupae
of P. c-aureum and from moths of B. mori by
dissection in saline (0.9% NaCl) and were stored
at 85°C. They were homogenized with a Teflon
homogenizer in ice-cold acetone (a total volume of
ca. 4ml), washed 3 times in 80% ethanol (a total
volume of ca. 0.8 ml) and extracted 3 times with
either 0.6 ml of 2% NaCl (for Sephadex G-50 gel
filtration) or Grace’s medium of the same volume
as above (for in vitro PTTH-assay). At each step,
insoluble materials were removed off by a centri-
fugation at 12,000xg for 20min at 4°C. The
brain-extracts (2% NaCl or Grace’s medium) were
heated in a billing-water bath for 3 min, rapidly
colled and centrifuged at 12,000Xg for 20 min.
The supernatant (2% NaCl) was added with solid
ammonium sulfate up to 0.8 saturation to precipi-
tate the hormones. The precipitates dissolved in
0.1M ammonium acetate (crude extracts) as well
as the brain extracts of Grace’s medium were
stored at —85°C.

Sephadex G-50 gel filtration

Crude extracts derived from either 800 brains of
P. c-aureum SD-pupae (0 day-old) or 800 brains of
B. mori moths were applied to a column (12 x 912
mm) of Sephadsx G-50 (Pharmacia, superfine).
The Sephadex G-50 column was eluted with 0.1 M
ammonium acetate (7.3 ml/hr) and sample frac-
tions consisting of a hundred drops each (3.9 ml)
were collected for about 15 hr. Then, they were
lyphilized and stored at —85°C.

In vitro assay of PTTH

PTTH activity of each sample fraction was quan-
tified by an in vitro assay for PTTH [12]. Pairs of
the prothoracic glands (PGs) were obtained from
either Sth instar LD-larvae of P. c-aureum (day 1)
or those of P. xuthus (day 2) by dissection in saline
[16] and were rinsed in standing drops of Grace’s
medium (50 1) for 30-60 min. One of the glands
from an individual was transferred to a standing
drop of Grace’s medium alone (50 wl) and incu-
bated for 2hr at 25°C. The contralateral gland
serving as experimental was incubated for 2 hr ina

Prothoracicotropic Hormone 699

standing drop of Grace’s medium (50 wl) with an
assaying sample. Following the incubation, the
glands were removed and the amount of ecdyster-
oids in each incubation medium (5 pl or 20 1) was
quantified by radioimmunoassay (RIA). Then,
activation ratio (Ar) representing the PTTH activ-
ity of the assaying sample was obtained by the
incubations of the glands of 5 to 6 pairs. The Ar
score shows the amount of ecdysteroids synthe-
sized by the experimental gland divided by that
synthesized by the control gland [12]. When
showed a larger Ar-score than 2 (p<(0.01: t-test),
the sample was judged as being PTTH-active.
Dose-dependencies of the reactivity of the glands
of P. c-aureum and P. xuthus Sth instar larvae were
examined using the brain extracts of P. c-aureum
pupae and B. mori moths, whose doses were
varied from 0.25 to 2 brain equivalent.

Radioimmunoassay (RIA) of ecdysteroids

A sample fraction of 5 yl or 20 «l was collected
from each incubation medium of the gland (50 yl)
and titer of ecdysteriods in each sample was quan-
tified against ecdysone obtained from Sigma Che-
mical Co. (St. Louis, USA), using the RIA
method [17]. Anti-serum raised against 20-hydro-
xyecdysone from Rhoto Pharmaceutical (Osaka,
Japan) was obtained from Meguro Institute (Osa-
ka, Japan). This antiserum exhibits almost 5-fold
stronger cross reactivity against 20-hydro-
xyecdysone than that against ecdysone. The
radioactive ecdysone, (23, 24-3H(N))-ecdysone
(80 Ci/mmol), was obtained from New England
Nuclear (Boston, USA). Lower limit of the ecdys-
teroid concentration detected by this assay was 2.5
pg (ecdysone equivalent)/20 ul incubation me-
dium.

Bioassay of SMPH

Five yl of saline (0.9% NaCl) with or without a
gel filtrated fraction through Sephadex G-50 (50
brain equivalent) was injected into each the
abdomed of 5-6 female Polygonia SD-pupae (day
0; 4 to 12hr after larval-pupal ecdysis). The
injection was made through the ventro-lateral in-
tersegmental region between the 6th and 7th abdo-
minal segments and the recipient pupae were
allowed to develop at 25°C.

On the day of eclosion, the female butterflies
developed from the recipient pupae were ex-
amined for the characteristics of summer morphs
and were classified into one of the grades 0-4. An
average grade score for summer morphs (AGS)
was obtained in the same manner as was described
in the previous paper [4].

RESULTS

The reactivity of the prothoracic glands derived
from Sth instar larvae of P. c-aureum to PTTH

To know how the reactivity of PITH of the
prothoracic glands of P. c-aureum Sth instar larvae
changes with the age of donors, pairs of the glands
obtained from 5th instar larvae of different ages
(day 0-4) were incubated in Grace’s medium with
or without brain extracts of P. c-aureum pupae (2
brain equivalent). In addition, dose dependencies
of the reactivity against the brain extracts of P.
c-aureum and B. mori were examined using the
glands from Sth instar larvae (day 1) of P. c-
aureum and P. xuthus, respectively.

The glands derived from 5th-instar larvae (day
0) of P. c-aureum secreted ecdysteroids at a low
rate (5.1+1.1 pg/gland/hr) in Grace’s medium
alone. But, the rate of ecdysteroids secretion rose
gradually as the days went on in the 5th instar
larval stage and reached 999.27 + 203.0 pg/gland/
hr on day 3 (Fig.1). The secretion rate also
increased in all the glands examined by adding the
brain extracts of P. c-aureum pupae (2 brain
equivalent) into incubation medium (Ar 2) (Fig.
1). A maximum Ar of 6.1+1.1 was rocorded by
the glands from 5th instar larvae of day 1. The
response became stronger as the dose of brain
extracts added into incubation medium was in-
creased (Fig. 2). A half maximum response of the
glands of P. c-aureum Sth instar larvae (day 1) was
obtained at the dose of 1.5 brain equivalent.

A far stronger response (Ar 29.8) than in the
larval glands of P. c-aureum was obrained when
the glands of 5th instar larvae (day 3) of P. xuthus
were incubated in Grace’s medium containing
brain extracts of B. mori, (5 brain equivalent).
But, this 5 brain equivalent seemed not to be still
sufficient for inducing the maximum response of

700 K. ENpo, Y. Fuyimoto et al.

Ecdysone equivalents (ng)
>

0

Days after the 4th larval ecdysis

Fic. 1. Stage-dependent changes in reactivity of the
prothoracic glands of P. c-aureum Sth instar larvae
against the brain-extracts of P. c-aureum pupae (day
0). Open and solid circles show the amounts of
ecdysteroids (ng/gland/2 hr) which the glands sec-
reted by an incubation in Grace’s medium with
(control) and without the extracts of 2 brain equiva-
lent (experimental), respectively. Vertical thin lines
show the standard errors of the 10-12 glands. Solid
triangles connected with broken line show the scores
of Ar representing the amounts of ecdysteroids
secreted by the experimental glands divided by those
secreted by the control glands.

0 0.25 0.5 1 2 4
Brain equivalents

Fic. 2. A dose-dependent response curve in the
prothoracic glands of P. c-aureum 5th instar larvae
(day 1) obtained using the brain-extracts of P.
c-aureum pupae (day 0). Open circles show the
scores of Ar with the standard erros of 10-12 gland
pairs shown by vertical lines.

the glands of P. xuthus larvae.

The results indicate that the glands from 5th
instar LD-larvae of P. c-aureum as well as those
from 5th instar larvae of P. xuthus show a response
to the brain extracts added into incubation
medium and increase the rate of ecdysteroid secre-
tion in vitro. The secretion rate becomes higher
when the dose of brain extracts added into incuba-
tion medium is increased.

Distribution of PTTH and SMPH after Sephadex
G-50 gel filtration of the brain extract of P.
coaureum

A crude extract made from 800 brains of SD-
pupae (day 0) was applied onto a column of
Sephadex G-50. Each fraction through the column
was divided into two parts and lyophilized. One
was dissolved in Grace’s medium (50 brain equiva-
lent/50 #1) and used for assaying PTTH activity
using the glands of P. c-aureum 5th instar larvae
(day 1), whereas the other was dissolved in the
saline (50 brain equivalent/5 ~l) and used for
assaying SMPH activity. In addition, some of the
fractions (no. 11-12 and 16-17) were dissolved in
Grace’s medium and assayed for PTTH activity
using the glands from P. xuthus 5th instar larvae
(day 2).

The PTTH activity was detected in the fractions
10-14 and 16-17 forming two broad peaks by the
P. c-aureum assay (Fig. 3). They are tentatively
referred to big and small PTTHs, respectively. By
the P. xuthus assay, the fractions 16-17 were
judged as showing the PTTH activity (Ar 11.4 and —
6.2), which was far stronger than that of the
fractions 11-12 (Ar 1.5 and 2.8). In addition, the
SMPH activity was detected in only 3 tubes (no.
17-19).

The results indicated that the tropic action of the
P. c-aureum small PTTH on the glands of P.
xuthus 5th instar larvae seemed to be far stronger

Cytochrome c

Bacitracin
Vo (MW. 12,384) (MW. 1,450)
y y y
Absorbance
Zs
15 7 at 280 nm i
aa

Ar ib
a
et
Le

Se
og Ne
we

AGS

\ f L,
ale Se oa j |

10 15 20

Fraction no

Fic. 3. Sephadex G-S0 gel filtration of the brain extract
of P. c-aureum SD-pupae showing the distribution
of PTTH and SMPH. Open and solid circles show
SMPH activities (AGS) and the PTTH (Ar), respec-
tively. Vertical lines show the standard errors of 10-
12 pairs (Ar) or those of the 20 individuals (AGS).

Prothoracicotropic Hormone 701

that the action of the big PTTH of this butterfly.
The recoveries in the gel filtration through
Sephadex G-50 were estimated to be 50-60% in
PTTH and 40-50% in SMPH activities.

Sephadex G-50 gel-filtration of the brain extract of
B. mori

A crude extract made from 800 brains of B. mori
was applied onto a column of Sephadex G-50.
Each eluent was divided into two parts, lyophilized
and dissolved in Grace’s medium (50 brain equiva-
lent/50 ul). One was assayed for PTTH using the
glands of P. xuthus Sth instar larvae (day 2),
whereas the other was assayed for PITH using the
glands of P. c-aureum 5th instar larvae (day 1) in
the same manner as above.

By the P. xuthus assay, the PTTH activity was
detected from the fractions no. 14-17. These
fractions most probably contained bombyxin
which was detected in Bombyx brain extracts by
Samia assay by Ishizaki et al. [18]. In contrast, no
PTTH activity was detected by the assay using the
glands of P. c-aureum (Fig. 4).

The results indicate that the PITHs of B. mori
show no prothoracicotropic action on the glands
from P. c-aureum Sth instar larvae (day 1).

Vo Cytochrome c Bacitracin
(MW.12,384) (MW. \450)

i v y

B25
<

S
Absorbance at 280nm

AN
a eohi7 sepa

eS RS

oe

0

10 15 19

Fraction no.

Fic. 4. Sephadex G-50 gel filtration of the brain extract
of B. mori moths showing the distribution of small
PTTH. Open circles show the small PTTH (bomby-
xin) activity detected by the glands from P. xuthus
5th instar larvae (day 2), whereas solid circles show
the PTTH activity detected by the glands from P.
c-aureum 5th instar larvae (day 1). Vertical thin
lines show the standard errors (6-8 pairs), respec-
tively. A curved thin line shown absorbance at 280
nm.

Stage-dependent changes in the small PTTH and
total PTTH activities of the brain-extracts of P.
c-aureum

Extracts were made from 200 brains of 5th instar
LD-larvae (day 0, 2 and 3) and LD-pupae (day 0,
1, 2 and 4) of P. c-aureum with Grace’s medium,
and were divided into two halves. One was used
for assaying the total PTTH activity using the
glands from P. c-aureum Sth instar larvae (day 1),
whereas the other was used for assaying small
PTTH activity using the glands of P. xuthus Sth
instar larvae (day 2) whose half maximum re-
sponse (Ar 12) to the brain extracts of P. c-aureum
SD-pupae (day 0) was obtained by the dose of 2
brain equivalent. Brain extracts were also made
from 5th instar SD-larvae and SD-pupae of P.
c-aureum of different stages, and the total PITH
and small PTTH activities were assayed in the
Same manner as above.

In LD-insects, the total PITH activity of the
brain extracts was low through the early and
middle 5th instar larval stages (Fig. 5). The activ-
ity rose abruptly on day 3 of the 5th instar (3-6 hr
after hanging) to reach a constant level which
lasted throughout the early pupal stage. SD-
insects showed a similar pattern to LD-insects in

= 20s reat
o-/_ pase
pea ; fe)
pee
@)
0 2 PP O | 2 4
5th instar larva Pupa

Days

Fic. 5. Stage-dependent changes in total PTTH activity
of the brain extracts of P. c-aureum. Ooen and solid
circles show the scores of Ar obtained using brain
extracts of LD- and SD-insects of different stages (2
brain equivalent/50 ul), respectively. Vertical thin
lines show the standard errors (8-10 pairs). The
glands for the assay were obtained from Sth instar
larvae of P. c-aureum (day 1). “PP” shows the stage
of pharate pupae 2-6 hr after hanging (day 3 of the
Sth instar).

702 K. Enpo, Y. Fusmmoro et al.

30 4
SD

ye
-O-
oe
Bo
AGS

15 :
$
uss
= 995-977
0 0
30 4
LD
7 aie
heey diel
fe Tee )
EalD oe NaG us é 2 ©
¢
Oe we - ° 1
~-¢
0 0
0 2 PP 0 1 2 4
5th instar larva Pupa

Days

Fic. 6. Stage-dependent changes in small PTTH and
SMPH activities of the brain-extracts of SD- (upper
panel) and LD-insects (lower panel). Solid circles
with thick lines show the scores of Ar obtained using
P. c-aureum brain extracts of different larval and
pupal stages (2 brain equivalent/SO wl). The glands
served for the assay were obtained from P. xuthus
Sth instar larvae (day 2). Open circles show the
scores of AGS detected in the brain extracts by
Polygonia pupal assay. Vertical thin lines show the
standard errors (Ar: 6-8 pairs, AGS: 8-10 indi-
viduals), respectively.

the fluctuation of the total PTTH activity of brain
extract. But, in SD-insects, the rising occurred one
day later (on the day of larval-pupal ecdysis) than
in the case of LD-insects.

The small PTTH activity in the brain extracts of
LD- and SD-insects also fluctuated stage-
dependently. From the differential Ar’s obtained
before and after larval-pupal ecdysis, the patterns
were judged to be different from those of the total
PTTH activity (Fig. 6).

The SMPH activity in the brain extracts of LD-
and SD-insects fluctuated with the stages of de-
velopment (Fig. 6). The fluctuating patterns were
almost the same as the ones reported in P. c-
aureum by Masaki et al. [6]. The pattern of
fluctuations in the SMPH activity of the brain
extracts of LD- and SD-insects were also different
from the patterns of the small PTTH activity.

It may be that production of the small PTTH in
the brain of P. c-aureum as well as its secretion is
affected by long days in which they are held during
the larval stage.

DISCUSSION

Autumn morph producers (SD-pupae) of P.
c-aureum developed into butterflies having the
characteristics of summer morphs when they were
injected with either pupal brain extracts of P.
c-aureum or 20-hydroxyecdysone on the day of
larval-pupal ecdysis [3-5]. The reactivity of SD-
pupae to the SMPH as well as the reactivity to
20-hydroxyecdysone disappeared on the next day
of larval-pupal ecdysis, when ecdysteroids forcing
the SD-pupae to undergo pupal-adult metamorph-
Osis appeared in the hemolymph [5]. The
hemolymph ecdysteroids were secreted by the
prothorecid glands, whose secreting activity was
shown to be governed by one or two prothoraci-
cotropic hormones (PTTHs) of different molecular
sizes as have been shown in several other lepidop-
teran insects [18].

P. c-aureum had two PTTHs of different molec-
ular sizes, both of which showed a tropic action
(increasing the rate of ecdysteroids secretion) on
the glands from P. c-aureum Sth instar larvae (Fig.
3). The total activity of the PTTHs in the larval
and pupal brain extracts of P. c-aureum as well as
its small PTTH activity fluctuated with the stages
of development (Figs. 5 and 6). The fluctuating
patterns in the total PTTH and small PTTH activ-
ity were different between LD- and SD-insects.
The patterns were also different from the patterns
of fluctuations in the SMPH activity of LD- and
SD-insects. In summer morph producers (LD-
insects), the total PITH activity representing the
sum of the action of the PTTHs of the brain
extracts to the glands of P. c-aureum 5th instar
larvae reached a high pupal level one day before
larval-pupal ecdysis (the pharate pupal stage),
which was 1 day earlier than in the case of autumn
morph producers (SD-insects).

A. burejana and A. levana exhibit spring and
summer morphs. With respect to the determina-
tion of seasonal morph development of these but-
terflies, an endocrine principle was already prop-

Prothoracicotropic Hormone 703

osed [8-9]. That is, pupae developed into summer
morphs when they were exposed to hemolymph
ecdysteroids within a few days after larval-pupal
ecdysis. Secretion of the ecdysteroids into
hemolymph is governed by the PTTH(s) in these
butteriflies. But, the principle proposed in relation
to the seasonal morph determination of B. bure-
jana and A. levana seemed not to be applicable to
the case of P. c-aureum seasonal morph develop-
ment since the titer of hemolymph ecdysteroids did
not vary between LD- and SD-insects on the day
and on the next day of larval-pupal ecdysis [6]. It is
also supported by another evidence that one (out
of 4) Sephadex G-50 gel filtrated/SMPH-active
fractions of P. c-aureum brain extract showed no
tropic action on the glands from P. c-aureum 5th
instar larvae (tube no. 19 in Fig. 3). Furthermore,
3 gel filtrated fractions of the brain extract of B.
mori moths (tube no. 14-16 in Fig. 4) showed
strong SMPH activity as shown in our previous
study [5], but they did not show any tropic action
on the glands of P. c-aureum 5th instar larvae.

In P. c-aureum, both SMPH and hemolymph
ecdysteroids seem to play significant roles in the
determination of summer morph development.
The ecdysteroids may share the roel(s) with the
SMPH in the determination of seasonal morphs
development of this butterfly. It may become clear
on a course of further study on the physiological
mechanism underlying the photoperiodic control
of the determination of seasonal morph develop-
ment of P. c-aureum.

ACKNOWLEDGMENTS

The authors with to express their sincere gratitude to
Professor A. Okajima and to Professor S. Y. Takahashi
of Yamaguchi University for advice and valuable sugges-
tions during the course of this work. This work was
supported in part by a grant from the Ministry of
Education, Science and Culture of Japan (No. 6154052).

REFERENCES

1 Fukuda, S. and Endo, K. (1966) Hormonal control
of the development of seasonal morphs in the but-
teffly, Polygonia c-aureum L. Proc. Japan Acad.,
42: 1082-1087.

2 Endo, K. (1972) Activation of corpora allata in
relation to ovarian maturation in the seasonal forms

10

11

12

13

of the butterfly, Polygonia c-aureum L. Develop.
Growth Differ., 14: 263-274.

Endo, K. (1984) Neuroenmdocrine regulation of
the development of seasonal forms of the Asian
comma butterfly, Polygonia c-aureum L. Develop.
Growth Differ., 26: 217-222.

Endo, K., Masaki, T. and Kumagai, K. (1988)
Neuroendocrine regulation of the development of
seasonal morphs in the Asian comma butterfly,
Polygonia c-aureum L.: Difference in activity of
summer-morph-producing hormone from_ brain-
extracts of the long-day and short-day pupae. Zool.
Sci., 5: 145-152.

Masaki, T., Endo, K. and Kumagai, K. (1988)
Neuroendocrine regulation of the development of
seasonal morphs in the Asian comma butteffly,
Polygonia c-aureum L.: Is the factor producing
summer morphs (SMPH) identical to the small
prothoracicotropic hormone (4K-PTTH)? Zool.
Sci., 5: 1051-1057.

Masaki, T. Endo, K. and Kumagai, K. (1989)
Neuroendocrine regulation of the development of
seasonal morphs in the Asian comma butterfly,
Polygonia c-aureum L.: Stage-dependent changes in
activity of summer-morph-producing hormone of
the brain-extracts. Zool. Sci., 6, 113-119.

Endo, K. and Kamata, Y. (1985) Hormonal control
of seasonal-morph determination in the small cop-
per butterfly, Lycaena phlaeas daimio Seitz. J. In-
sect Physiol., 31: 701-706.

Keino, H. and Endo, K. (1973) Studies on the
determination of seasonal forms in the butterfly,
Araschnia burejana Nrtmrt. Zool. Mag., 82: 48-52.
(In Japanese with English summary)

Koch, P. B. and Bikmann, D. (1988) Hormonal
contrl of seasonal morphs by the timing of ecdyster-
oid release in Insect Physiol., 33: 823-829 .
Williams, M. C. (1947) Physiology of insect di-
apause. II. Interaction between the pup0al brain and
prothoracic glands in the metamorphosis of the glant
silkworm, Platysamia cecropia. Biol. Bull., 93: 89-
98.

Agui, N., Granger, N. A., Gilbert, L. I. and Bollen-
bracher, W. E. (1979) Cellular localization of the
insect prothoracicotropic hormone: /n vitro assay of
a single neurosecretory cells. Proc. Natl. Acad. Sci.
USA, 76: 5694-5698.

Mizoguchi, A., Ishizaki, H., Nagasawa, H.,
Kataoka, H., Isogai, A., Tamura, S. Susuki, A.,
Fujino, M. and Kitada, C. (1987) A monoclonal
antibody against a synthetic fragment of bombyxin
(4K-prothoracicotropic hormone) from the slik-
moth, Bombyx mori: characterization and immuno-
histocemistry. Mol. Cell. Endocrinol., 51: 227-135.
Bollenbacher, W. E., Augi, N., Granger, N. A. and
Gilbert, L. I. (1980) Jn vitro activation of insect

Ng

— 2) sm

14

15

16

17

704

prothoracic glands by the prothoracicotropic hor-
mone. Proc. Natl. Acad Sci., 76: 5148-5142.
Mizoguchi, A., Ishizaki, H., Nagasawa, H. and
Suzuki, A. (1987) Change in bombysin contents in
the brain of Bombyx mori. Zool. Sci., 4: 1091.
Ishizaki, H. (1986) A PTTH pilgrimage from
Leuhdorfia to Bombyx. Zool. Sci., 3: 921-929.
Wyatt, G. R. (1961) The biochemistry of insect
homolymph. Ann. Rev. Entomol., 6: 75-102.
Borst, D. W. and O’Conner J. D. (1972) Arthropod

18

K. ENpbo, Y. Fusimoto et al.

molting hormone: radioimmune assay. Science, 178:
418-419.

Ishizaki, H., Mizoguchi, A., Fujishita, M. Suzuki,
A., Horiya, I., O’oka, H., Kataoka, H., Isogai, A.,
Nagasawa H., Tamura, S. and Suzuki, A. (1983)
Species specificity of the insect prothoracicotropic
hormons (PTTH): the presence of Bombyx- and
Samia-Specific PTTHs in the brain of Bombyx mori.
Develop. Growth Differ., 25: 593-600.

ZOOLOGICAL SCIENCE 7: 705-713 (1990)

Effects of Gonadotropin-Specific Antibodies on the Interaction
of Follicle-Stimulating Hormone and Luteinizing Hormone
with Testicular Receptors in the Bullfrog,

Rana catesbeiana

HIROSHI YAMANOUCHI and SUSUMU ISHII

Department of Biology, School of Education, Waseda University,
Shinjuku-ku, Tokyo 169, Japan

ABSTRACT—To study the properties of the receptor-binding site in molecules of bullfrog gonadotro-
pins, we treated follicle-stimulating hormone (fFSH) and luteinizing hormone (fLH) from the bullfrog
with antibodies raised against the whole molecules or subunits of these hormones, and then examined
the binding of the treated hormones to bullfrog testicular receptors. When low concentrations of
fFSH-specific and fLH-specific antisera were used, increased concentrations of the antisera led to
increased inhibition of binding of both hormones. At higher concentrations, however, the extent of
binding was elevated by the antisera in both cases. These stimulatory effects were abolished by
treatment of antisera with papain, while the inhibitory effects persisted. Binding of fFSH to its
receptors was inhibited by a monoclonal antibody against [{FSH- and by one against fFSH-a but not by
a monoclonal antibody against f[LH-8. The inhibitory potency of the antibody against fFSH-a was
stronger than that of the antibody against fFSH-?. Binding of fLH was inhibited by the monoclonal
antibody against f{LH-? more strongly than by one against fFSH-a. Binding of {LH was not inhibited or
slightly inhibited by the monoclonal antibody against fFSH-@. These results suggest that both the a and
8 subunits contribute to the binding of both hormones to their receptors. Although fFSH and fLH bind
to a common receptor in the bullfrog testis, the chemical structures of the receptor-binding sites in these
hormones are assumed not to be identical. The importance of the a subunit in the binding of fFSH to

© 1990 Zoological Society of Japan

receptors is also suggested.

INTRODUCTION

The pituitary gland of anuran amphibians, as
well as that of higher vertebrates, secretes two
species of gonadotropins, i.e. follicle-stimulating
hormone (FSH) and luteinizing hormone (LH).
These hormones have been reported to exert diffe-
rent effects in the frog testis [1-4]. However,
Takada et al. [5] found that specific binding of
'°°1-fFSH was completely inhibited by both fFSH
and f{LH. Recently, we further showed that spe-
cific binding of '*°I-fLH was similarly inhibited by
both {FSH and fLH, although the inhibition curves
were not parallel for these two hormones [6].
These results suggest the close similarity or identi-
ty of the chemical structures of the receptor-

Accepted January 18, 1990
Received October 12, 1989

binding sites in the fFSH and fLH molecules.

Antibodies, raised against intact molecules or
subunits of gonadotropins, have been employed as
useful tools for the analysis of the interactions
between gonadotropins and their receptors in
mammals [7-17]. Using antibodies raised against
fFSH, fLH and their subunits, we examined the
effects of the antibodies on binding of fFSH and
{LH to testicular receptors of the bullfrog, in order
to analyze the interactions between bullfrog go-
nadotropins and their receptors, with special em-
phasis on the loss or incompleteness of FSH/LH
specificity of the receptors.

MATERIALS AND METHODS

Tissues

Fresh testes of adult bullfrogs, Rana catesbeiana,

706 H. YAMANOUCHI AND S. IsHl

were collected at the time of meat processing at the
shop of a vendor of frog meat. The testes were
immediately put in a chilled solution of 0.68%
NaCl and were transported under chilled condi-
tions to the laboratory.

Hormones, antisera and monoclonal antibodies

Highly purified fFSH and fLH, designated
FF1341B and FL461B by Takada and Ishii [4],
respectively, were used.

Two types of antiserum were used: one was an
antiserum raised against f{[LH, designated No. 1 by
Takada and Ishii (unpublished) and the other was
an antiserum against [FSH designated No. 3.

Three types (BF3A20, BF3B25, and BL4B11) of
monoclonal antibody against subunits of gonado-
tropin were used. They were raised against the
fFSH-a, fFSH-8 and fLH-f subunits, respectively.
The first two were prepared by Tanaka et al. [18]
and the last by Park et al. [19].

In addition, a monoclonal antibody against
mammalian luteinizing hormone-releasing hor-
mone (LHRH), HAC-MM02-MSM6§84 of Park and
Wakabayashi [20], was used. All these monoclon-
al antibodies were supplied by Prof. K. Waka-
bayashi of Gunma University.

Goat anti-rabbit immunoglobulin was supplied
by Dr. T. Yoshida of Tsukuba Primate Center,
and goat anti-mouse immunoglobulin (IgA+IgG
+IgM) was purchased from Cappel (West Ches-
ter, PA., USA). They were used as second
antisera to precipitate antibody-bound hormones.

Chemicals

Insoluble papain was obtained from Sigma (St.
Louis, MO., USA; lot 24F-9620; 850 units/g).
Na!**I was obtained from ICN _ biochemicals
(Irvine, CA., USA). All other reagents were
purchased from standard commercial sources.

Radioiodination

Bullfrog FSH and LH were radioiodinated with
251 by the lactoperoxidase method reported by
Takada et al. [5]. Specific radioactivities, esti-
mated by the method of Kubokawa and Ishii [21],
of '*I-FSH and '°I-LH were about 85,000 and
90,000 cpm/ng (64 and 68 “Ci/ug), respectively.

Digestion of antisera

Digestion of antisera with papain was performed
by the method of Dias et al. [10], after appropriate
dilution of the antisera. A suspension of enzyme
was prepared first by incubating 12 mg of insolubi-
lized papain with 1.14 ml of an activating solution
(a mixture of deionized water, 0.01M EDTA,
0.006 M mercaptoethanol, and 0.05 M cysteine-
HCl in a ratio of 70:10:1:10) for 30 min. To 1.5
ml of each solution of antiserum, 125 yl of the
enzyme suspension were added. Digestion was
allowed to proceed for 30 min at 37°C in a shaking
bath. After the incubation, 2041 of 0.1M
dithiothreitol were added. The mixture was then
centrifuged to remove insoluble papain. The su-
pernatant was dialyzed overnight against 40 mM
Tris buffer. The dialyzate was used as papain
digested monovalent antiserum in binding experi-
ments.

Preparation of receptors

A crude fraction of plasma membranes was
prepared from bullfrog testis by the method pre-
viously described [6], and it was used as the
preparation of receptors for binding experiments.
An aliquot of 0.1 ml of the final suspension of
membranes contained membrane particles derived
from 50 to 100 mg of testicular tissue.

Binding experiments

Various dilutions of each antibody or buffer
alone (50 1) were preincubated with 50 yl of
51-fFSH or '*I-fLH (about 0.2 to 0.4ng or
18,000 to 32,000 cpm) in plastic tubes at 20°C for
3.5 hr. To each tube, 100 ul of the preparation of
receptors were added after the preincubation.
Then tubes were incubated at 20°C for 3 hr with
shaking. At the end of the incubation, 1 ml of cold
buffer was added to each tube, and the tubes were
centrifuged at 10,000g for 3min at 4°C. The
supernatant was removed by aspiration, and the
pellet was resuspended in buffer and centrifuged
again. Radioactivity of the resulting pellet was
measured in an automatic gamma counter. The
hormone-binding level was expressed as a per-
centage of the radioactivity bound in the absence
of antibody. The level of nonspecific binding

Effects of Antibodies on GTH Binding 707

estimated in the absence of the antibody but in the
presence of excess nonradioiodinated gonadotro-
pins was 9 and 19% of the total binding for {FSH
and fLH, respectively.

In order to study immunological cross-reactivity
and potency of antibodies, we used the following
procedures. An antibody and labeled hormone
were incubated under conditions similar to those in
the preincubation described for receptor-binding
experiments, and then 200 yl of appropriately di-
luted second antiserum was added. Reaction tubes
were then incubated at 20°C for 5hr. After
addition of 500 ul of 1% BSA-PBS, the tubes were
centrifuged at 3000 xg for 40 min, and radioactiv-
ity in the resulting pellets was measured.

200

160

120

Binding of '*SI-FSH to the receptors [Z)

40

800
Antibody dilution

S200 Za00 = 51700

RESULTS

Effects of polyclonal antisera against fFSH and
against fLH on binding of radioiodinated fFSH
and fLH to testicular membrane receptors

The antiserum against {FSH inhibited the bind-
ing of '”I-fFSH to the receptors in proportion to
the concentration of the antiserum at low concen-
trations of the antiserum (Fig. 1a). However, at
higher concentrations, the extent of the inhibition
became smaller as the concentration of the anti-
serum became higher, and augmentation of the
binding of '*°I-fFSH above the 100% level was
observed at the highest concentration tested.

The antiserum against fLH also inhibited the
binding of '*I-fFSH but, in this case, no aug-
mentation above the 100% level was observed

Binding of '?°I-H to the receptors (/)

800
Antibody dilution

SA) ae) Silao 8

Fic. 1. Effects of a polyclonal antiserum against {FSH (@, solid line), a polyclonal antiserum against fLH (m, solid
line) and normal rabbit serum (x, dotted line) on binding of '*°I-fFSH (a) and !*°I-fLH (b) to a preparation of
bullfrog testicular receptors at different dilutions. The ordinate shows the binding of radioiodinated hormone as a
percentage of radioactivity bound in the absence of antiserum. Each point and each vertical bar show the mean
and the range of duplicate determinations, respectively. See text for details.

708 H. YAMANOUCHI AND S. IsHi

although the extent of inhibition became smaller at
the highest concentration of antiserum tested in
this antigen excess condition.

The antiserum against {LH strongly inhibited
the binding of '°I-fLH, with the extent of inihbi-
tion depending on the concentration of antiserum
at lower concentrations of antiserum (Fig. 1b). At
higher concentration, the inhibitory effect of the
antiserum became smaller as the concentration
was increased, and the binding of Ie fleHisiexe
ceeded the 100% level at the three highest concen-
trations tested. Antiserum against {FSH inhibited
the binding of '*I-fLH, with the extent of inhibi-
tion depending on the concentration over the
whole range of concentrations tested. However,
the inhibitory potency of the antiserum against
fFSH was lower than that of the antiserum against
fLH. The normal rabbit serum (NRS) showed no
effect on binding of either 1-fF SH or '”I-fLH at

120

100

40

Binding of '°I-FSH to the receptors (7)

12800 51200
Antibody dilution

800 3200

any concentration of the serum tested.

Effects of papain-digested antisera on binding of
FSH and fLH to testicular receptors

After digestion with papain, both the antiserum
against {FSH and that against {LH inhibited bind-
ing of '*I-fFSH to the receptors, with inhibition
depending on the concentration over almost the
entire range of concentrations tested (Fig. 2a). No
augmentation was observed at all with either tre-
ated antiserum. The potency of the inhibition of
binding of the digested antiserum against {FSH
was 1.7 times higher than that of the digested
antiserum against {LH at the 50% level of inhibi-
tion.

Binding of '*I-fLH was also inhibited by both
treated antisera in a concentration-dependent
manner, and absolutely no augmentation was
observed (Fig. 2b). The inhibitory potency of the

Binding of '°I-+H to the receptors (/)

12800 51200
Antibody dilution

800 3200

Fic. 2. Effects of a papain-digested antiserum against fFSH (@, solid line), a papain-digested antiserum against [LH
(m, solid line) and papain-digested normal rabbit serum (x, dotted line) on the binding of '”°I-fFSH (a) and
125T-fLH (b) to the preparation of bullfrog testicular receptors at different dilutions. See the legend to Figure 1.

and text for details.

Effects of Antibodies on GTH Binding 709

pD

oe

100
<
=
Ss 80
5
cab)
==
re) BU
cH
at
Sig
“Se
2
=
(ag
aay 1 AL Anti LH

Pitt FSH
NRS
0 a

400 1600 6400 25600
Antibody dilution

Binding of '°I-LH to the antibody (7)

400 1600 6400 29600
Antibody dilution

Fic. 3. Binding of '*°I-fFSH (a) and '*°I-fLH (b) to different dilutions of a papain-digested antiserum against fFSH
(@, solid line), papain-digested antiserum against fLH (@, solid line) and papain-digested normal rabbit serum (x
, dotted line). The ordinate shows the proportion of antibody that bound radioiodinated hormone as a percentage
of the amount of added hormone. Each point is the mean and each vertical bar shows the range of duplicate

determinations. See text for details.

digested antiserum against fLH was far higher
(12.6 times higher at the 50% level of inhibition)
than that of the digested antiserum against fFSH.

The FSH/LH specificity and immunoreactive
potency of the antiserum against [FSH and fLH
were studied with the papain-digested antisera
(Fig. 3a, b). Proportions of '°I-fFSH bound to
papain-digested antiserum against {FSH and to
antiserum against {LH were similar, being about
40% at a dilution of 1:400. In contrast, the
antiserum against fLH bound '*I-fLH more
strongly (about 70% binding at a dilution of 1:
400) than the antiserum against fFSH (20% bind-
ing at the same dilution).

Effects of monoclonal antibodies on binding of
fFSH and fLH to testicular receptors

Binding of '*°I-fFSH to the receptors was inhi-

bited by both the monoclonal antibodies raised
against {FSH-f and against fFSH-a, but not by that
against {[LH-8 (Fig. 4a). The monoclonal antibody
against [{FSH-a was more potent than that against
fFSH-8 in inhibition of the binding of 'I-fFSH to
the receptors. The ratio of the inhibitory poten-
cies, calculated with 100- and 400-time dilution
data, was about 2.8: 1.

Binding of '*I-fLH to the receptors was also
inhbited by both the monoclonal antibody against
f{LH-@ and that against fFSH-a. The inhibitory
potency of the former antibody was far higher than
that of the latter. The potency of the former
exceeded that of the latter 60 fold. No significant
inhibition of binding of '*I-fLH to its receptors by
the monoclonal antibody against fFSH-§ was
observed (Fig. 4b).

A monoclonal antibody against mammalian

710 H. YAMANOUCHI AND S. IsHi

120

100

Binding of °I-FSH to the receptors (/)

1600
Antibody dilution

Fic. 4. Effects of monoclonal antibodies against f{FSH-a (x, dotted line), against fFSH- (@, solid line) and against
fLH-f (@,, solid line) on binding of '*°I-fFSH (a) and '?°I-fLH (b) to a preparation of bullfrog testicular eos

100 400 6400

at different dilutions.

LHRH did not affect the binding of either '*I-
fFSH or '”I-fLH to the receptors (data not
shown).

The FSH/LH specificity and immunological
potencies of the monoclonal antibodies are shown
in FiguresSa and b. The extent of binding of
'°1-fFSH to the monoclonal antibody against
fFSH-@ was about 32% at a dilution of the anti-
body of 1:100, while the extent of binding of
°1-fFSH to the monoclonal antibody against
fFSH-a was only 12% at same dilution of anti-
bodies.

The extent of binding of 'I-fLH to the mono-
clonal antibody against {LH-£ reached about 62%
at a dilution of the antibody of 1: 100 and that of
'°1-fLH to the monoclonal antibody against
fFSH-a was only about 10%. No apparent binding
of '°I-fFSH to the monoclonal antibody against
fLH-f or of '*I-fLH to the monoclonal antibody
against fFSH-£ was observed.

Binding of °I+H to the receptors (7)

6400

100 400 1600
Antibody dilution

See the legend to Fig. 1 and text for details.

DISCUSSION

We showed in the present study that both of the
polyclonal antisera against fFSH and fLH inhi-
bited the binding of both fFSH and fLH to their
receptors at low concentrations of antiserum.
However, at higher concentrations, the antisera
became less effective in inhibiting the binding or
even enhanced the binding to receptors. Dias et al.
[10] also reported that antisera against subunits of
human FSH enhanced binding of human FSH to its
receptor, and they suggested that the enhancement
of the binding might be caused by formation of
aggregates of antibody and hormone molecules
due to bivalency of the antibody molecule. They
were able to eliminate the enhancement of binding
by pretreating the antisera with papain, which
digests bivalent molecules of the antibody and
produces monovalent molecules. In the present
investigation, the pretreatment of antisera against

Effects of Antibodies on GTH Binding 711

D

40

Pati FO+-B

Binding of '°I-SH to the antibody (7)

PatlL Fo4+-@
ee
Pt LH-B >

100 400 1600 6400
Antibody dilution

Binding of 'SI4H to the antibody (7%)

100 400 1600 6400
Antibody dilution

Fic. 5. Binding of '*°I-fFSH (a) and '*°I-fLH (b) to different dilutions of monoclonal antibodies against fFSH-a (x,
dotted line), against fFSH-f (@, solid line) and against f{LH-f (™,, solid line). See the legend to Fig. 3 and text for

details.

fFSH and fLH with papain completely eliminated
the enhancement of the binding to receptors
observed at high concentrations of the antisera.
Consequently, the enhancement of the binding to
receptors by the antisera, observed in the present
study, is also considered to be caused by the
formation of aggregates that results from the
bivalency of the antibodies.

The antiserum against [FSH was more potent
than the antiserum against fLH in inhibiting the
binding of '*°I-fFSH, while the former was less
potent than the latter in terms of immunological
potency for binding fFSH. The ratio of the inhibi-
tory potencies in binding to receptors was 1.7: 1
and the ratio of the immunological potencies was
1:1.2. The antiserum against {LH was more
potent than the antiserum against fFSH in terms of
both inhibition of the binding of '*°I-fLH to recep-
tors and in terms of the immunological capacity to
bind fLH, although the potency ratios differed

significantly between these two parameters (13:1
in the inhibition of receptor-binding and 44:1 in
the immunological binding). These results imply
that some components of these antisera inhibit the
binding to receptors by binding to the hormone
molecule, but the other components do not, and
also that the proportion of these two types of
component differs between antiserum against
fFSH and that against fLH.

When monoclonal antibodies were used, that
against f{LH-f inhibited the binding to receptors of
only fLH and not fFSH, and that against FSH-@
inhibited binding of only fFSH and not fLH.
These results seem to conflict with the fact that
both fFSH and fLH bind to a common receptor
site in the bullfrog testis, since specific antibodies
to a unique portion of each hormone molecule
inhibit the binding of the hormones to the common
receptor site. However, we can explain these
results by assuming that the sites which are related

‘
th J

TAZ H. YAMANOUCHI AND S. IsHII

to the receptor binding in the fFSH and fLH
molecules have at least somewhat different chem-
ical structures, and our monoclonal antibodies
against fFSH-8 and f{LH-f bound to these specific
or unique parts. In the case of mammalian LH and
hCG, a similar result was reported by Schwarz et
al. [16]. They found that monoclonal antibodies
specific to human LH and to hCG inhibited bind-
ing of these hormones to the common testicular
receptor in the rat. Hattori et al. [17] reported that
the conformation of the £ subunit of deglycosy-
lated hCG was altered by the binding of antibody
against hCG-f. It is possible that the binding of
antibody causes the conformational change of re-
ceptor binding site in hormone molecule.

Unlike the antibodies against {LH-f and fFSH-
8, the antibody against fFSH-a inhibited the bind-
ing to receptors of both hormones, although the
binding of {FSH was more strongly inhibited than
that of f{LH. This result suggests that masking of
the a subunit with antibodies influences the hor-
mone-receptor interaction, and it seems to conflict
with the idea that the a subunit is involved in the
action of the hormone while the f subunit is
involved in the specificity of the hormone [22-25].
However, a contribution of not only the § subunit
but also of the a subunit to the direct interaction of
the hormone and receptor has been reported by
several investigators: Milium and Midgley [11] in
the case of binding of hCG to the ovarian receptor
of the rat; Dias et al. [10] in the case of binding of
human FSH to the testicular receptor of the calf;
Moyle et al. [8] in the case of binding of hCG to the
Leydig cell receptor of the rat; and Schwarz et al.
[16] in the case of binding of hCG and human LH
to the testicular receptor of the rat.

In the inhibition of binding of f{LH, the antibody
against f[FSH-a was less potent than the antibody
against {LH-8. However, in inhibition of binding
of fFSH, the relationship was reversed. The
potency of antibody against the fFSH-a subunit
was higher than that of antibody against fFSH-~,
although their immunological potencies were re-
versed. This result suggests the importance of the
a subunit in the binding of fFSH to receptors, in
particular, in the bullfrog.

As mentioned in the “Introduction”, the same
gonadotropin receptor in the bullfrog testis recog-

nizes both {FSH and fLH, although in mammals
the two hormones have separate receptors. Ac-
cordingly, the importance of the a subunit in the
hormone-receptor interactions of both fFSH and
fLH in the bullfrog, as revealed in the present
study, can be related to the absence of FSH/LH
specificity in the bullfrog testicular gonadotropin
receptor. However, this phenomenon is not u-
nique to the bullfrog. Dias et al. [10] reported that
the binding of human FSH (hFSH) to the calf
testicular receptor was inhibited by an antiserum
against hFSH-a more strongly than by an anti-
serum against hFSH-8. Therefore, we can not
simply relate our result on the inhibitory effect of
the antibody against fFSH-a on the binding to
receptors to lack of FSH/LH specificity in the
bullfrog gonadotropin receptor.

This is the first report of a study of the interac-
tion of subunits of FSH and LH with their receptor
using monoclonal antibodies in nonmammalian
vertebrates. Even in the mammal, there is only
one report to our knowledge, on the inhibition of
the binding of FSH to receptors by a monoclonal
antibody [13]. Further studies using various types’
of monoclonal antibody are needed to elucidate
the consequences of the lack of FSH/LH specificity
in the binding to receptors in amphibians.

ACKNOWLEDGMENTS

We are grateful to Professor K. Wakabayashi and Dr.
S. Tanaka of Gunma University for the supply of monoc-
lonal antibodies. Thanks are also due to Dr. T. Yoshida
for the supply of goat anti-rabbit immunoglobulin. This
study was supported by a grant from the Ministry of
Education, Science and Culture, Japan, and also a grant
from Waseda University to S. I.

REFERENCES

1 Licht, P. and Papkoff, H. (1974) Separation of two
distinct gonadotropins from the pituitary gland of
the bullfrog Rana catesbeiana. Endocrinology, 94:
1587-1594.

2 Muller, C. H. (1977) In vitro stimulation of 5a-
dihydrotestosterone and testosterone secretion from
bullfrog testis by nonmammalian and mammalian
gonadotropins. Gen. Comp. Endocrinol., 33: 109-
120

3 Muller, C. H. (1977) Plasma 5a-dihydrotestosterone
and testosterone in the bullfrog, Rana catesbeiana:

10

11

2

13

14

Effects of Antibodies on

stimulation by bullfrog LH. Gen. Comp. Endocri-
noles3; 122-132.

Takada, K. and Ishii, S. (1984) Purification of
bullfrog gonadotropins: presence of new subspecies
of luteinizing hormone with high isoelectric points.
Zool. Sci., 1: 617-629.

Takada, K., Kubokawa, K. and Ishii, S. (1986)
Specific gonadotropin binding sites in the bullfrog
testis. Gen. Comp. Endocrinol., 61: 302-312.
Yamanouchi, H. and Ishii, S. (1987) FSH and LH of
the bullfrog bind to common receptors. In “Proceed-
ings of the First Congress of the Asia and Oceania
Society for Comparative Endocrinology (AOSCE)”.
Ed. by E. Ohnishi, Y. Nagayama and H. Ishizaki,
Nagoya, pp. 306-307.

Pierce, J. G., Bloomfield, G. A. and Parsons, T. F.
(1979) Purification and receptor binding properties
of complexes between lutropin and monovalent anti-
bodies against its a subunit. Int. J. Peptide Protein
Res., 13: 54-61.

Moyle, W. R., Ehrlich, P. H. and Canfield, R. E.
(1982) Use of monoclonal antibodies to subunits of
human chorionic gonadotropin to examine the
orientation of the hormone in its complex with
receptor. Proc. Natl. Acad. Sci., USA, 79: 2245-
2249.

Milius, R. P., Midgley, A. R. Jr. and Birken, S.
(1983) Preferential masking by the receptor of
immunoreactive sites on the a subunit of human
choriogonadotropin. Proc. Natl. Acad. Sci., USA,
80: 7375-7379.

Dias eon Eluston, J. S. and Reichert, L- E. Jr.
(1984) Differential effects of monovalent and biva-
lent antisera on the interaction of follicle-stimulating
hormone with its receptor. Endocrinology, 114:
1259-1265.

Milius, R. P. and Midgley, A. R. Jr. (1984) Inhibi-
tion of receptor binding to hCG by subunit specific
antibodies. Biol. Reprod., 30: Suppl. 1, 131.
Moyle, W. R., Anderson, D. M., Guan, T.,
Ehrlich, P. H. and Macdonald, G. J. (1984) Mono-
clonal antibodies distinguish the formation of oLH-
and hCG-receptor complex in homogenates of lutei-
nized rat ovaries. Biol. Reprod., 30: Suppl. 1, 132.
Thakur, A. N., O’Shea, A., Sesay, A., Ekins, R. P.
and Dias, J. A. (1986) Effect of monoclonal anti-
bodies to human follicle stimulating hormone
(hFSH) on the interaction of FSH with its receptor.
J. Endocrinol., 108: Suppl. 35.

Moyle, W. R., Pressey, A., Dean-Emig, D., Ander-
son, D. M. and Demeter, M., (1987) Detection of

1)

16

17

18

19

20

all

Ups

Jip)

24

U5)

GTH Binding 713
comformational changes in human chorionic gonad-
otropin upon binding to rat gonadal receptors. J.
Biol. Chem., 262: 16920-16926.

Schneyer, A. L., Sluss, P. M., Huston, J. M.,
Ridge, R. J. and Reichert, L. E. Jr. (1988) Iden-
tification of a receptor binding region on the
subunit of human _ follicle-stimulating hormone.
Biochemistry, 27: 666-671.

Schwarz, S., Berger, P., Nelboeck, E., Khashabi,
D., Panmoung, W., Kleiber, R. and Wick, G.
(1988) Probing the receptor-interaction of glyco-
protein hormones with monoclonal antibodies. J.
Rec. Res., 8: 437-453.

Hattori, M., Hachisu, T., Shimohigashi, Y. and
Wakabayashi, K. (1988) Conformation of the £
subunit of deglycosylated human chorionic gonado-
tropin in the interaction at receptor sites. Mol. Cell.
Endocrinol., 57: 17-23.

Tanaka, S., Park, M. K., Hayashi, H. Hanaoka, Y..,
Wakabayashi, K. and Kurosumi, K. (1986) Mono-
clonal antibodies against bullfrog follicle-stimulating
hormone and luteinizing hormone and their im-
munocytochemical application. Proc. Japan Soc.
Comp. Endocrinol., 1: 42.

Park, M. K., Tanaka, S., Hayashi, H., Hanaoka,
Y., Wakabayashi, K. and Kurosumi, K. (1987)
Production and characterization of monoclonal anti-
body against the @-subunit of bullfrog lutropin. Gen.
Comp. Endocrinol., 68: 82-90.

Park, M. K. and Wakabayashi, K. (1986) Prepara-
tion of a monoclonal antibody to common amino
acid sequence of LHRH and its application. Endoc-
rinol. Japon., 33: 257-272.

Kubokawa, K. and Ishii, S. (1984) Adaptation of
testicular follicle-stimulating hormone receptors to
ambient temperatures in vertebrates: equilibrium
analysis. Gen. Comp. Endocrinol., 54: 277-282.
Muralidhar, K. and Moudgal, N. R. (1976) Studies
on rat ovarian receptors for lutropin (luteinizing
hormone). Biochem. J., 160: 615-619.

Dighe, R. R., Muralidhar, K. and Moudgal, N. R.
(1979) Ability of human chorionic gonadotropin
f-subunit to inhibit the steroidogenic response to
lutropin. Biochem. J., 180: 573-578.

Licht, P., Farmer, S. W. and Papkoff, H. (1978)
Biological activity of hybrid combinations of ovine
and sea turtle LH subunits. Gen. Comp. Endocri-
nol., 35: 289-294.

Strickland, P. W. and Puett, D. (1981) Contribution
of subunits to the function of luteinizing hormone/
human chorionic gonadotropin receptors. Endocri-
nology, 109: 1933-1942.

ae

ec

‘ iba z oat q :
way. Po ens She

oe

San a Pik =
t sy a toouyge oy pohopnor Arayony |

' ! ; x <eH thi, * i : )
eR Ps ee i Allasio (AGO
ae ai bieeedides oiniacese ies aaah? ide aa

: aisahe
is SOC
+
< {
sae % ahs & hy’ #4: Sock sold idaeila ri
“ts yee Om te hal oa
ik Tite SH “_ be haol $f" #*u Fs tit cht
Lecrerredputwoinnit :
reer a ne SK
x S

ZOOLOGICAL SCIENCE 7: 715-721 (1990)

© 1990 Zoological Society of J apan

Effects of Ovariectomy and Steroid hormone
Implantation on Serum Gonadotropin
Levels in Female Goldfish

Maxito Kospayasut! and Norman E. STACEY

Department of Zoology, University of Alberta,
Edmonton, Alberta, Canada T6G 2E9

ABSTRACT— Effects of ovariectomy and steroid hormone implantation on blood gonadotropin (GtH)
levels were examined in sexually mature female goldfish. Ovariectomized fish had significantly higher
levels of GtH than intact and sham-operated fish from two to ten weeks post-surgery. Ovariectomized
fish implanted with testosterone or estradiol capsules which provided physiological blood steroid levels
had lower GtH levels than fish implanted with empty capsules. Pituitary GtH content was not affected
by ovariectomy and steroid implantation. This and an earlier study show that female goldfish have both
positive and negative steroid feedback systems for GtH release which are considered to regulate cyclic
(ovulatory surge) and tonic (basal secretion) GtH secretion, respectively.

INTRODUCTION

Production and secretion of gonadotropin
(GtH) in the pituitary gland of teleosts is control-
led by various factors. The release of GtH appears
to be directly regulated by gonadotropin-releasing
hormone (GnRH) and dopamine which acts as a
gonadotropin release-inhibiting factor [1]. Gonad-
al steroids also modulate GtH production and
secretion in teleosts, as in other vertebrates. Three
modes of stimulatory effect (positive feedback) of
sex steroid have been shown in teleosts by adminis-
tration of sex steroids: (1) stimulation by testoster-
one of the ovulatory GtH surge in female goldfish
Carassius auratus [2], a role played by estradiol
(Ez) in mammals [3]; (2) accumulation of pituitary
GtH by estrogens or aromatizagle androgens in
rainbow trout Salmo gairdneri [4], catfish Clarias
gariepinus [5], and eel Anguilla anguilla [6]; (3)
increase of pituitary responsiveness to GnRH by
testosterone or E, in goldfish [7], as shown in
mammals [3]. On the other hand, an inhibitory

Accepted January 27, 1990
Received November 29, 1989

" Present address and to whom all correspondence
should be addressed: Department of Fisheries,
Faculty of Agriculture, University of Tokyo, Bunkyo-
ku, Tokyo 113, Japan.

effect (negative feedback) on GtH secretion has
also been shown in common carp Cyprinus carpio
[8], rainbow trout [9-11], and catfish [5]. These
studies showed an elevation of blood GtH levels
after an abrupt removal of gonadal steroids, using
techniques such as gonadectomy and administra-
tion of antisteroids. Some of these studies also
showed a depression of the elevated GtH levels by
steroid replacement.

In the goldfish, negative steroid feedback on
GtH secretion is suggested by histological changes
in GtH cells of the pituitary gland in ovariecto-
mized and estrogen treated fish [12, 13]. Billard
and Peter [14] showed an estrogenic inhibition on
GtH secretion by implantation of antiestrogens
into the brain and pituitary of goldfish; however,
the pituitary implantation of antiestrogen pellets
may have damaged the pituitary stalk, thus in-
creasing GtH by blocking the tonic dopaminergic
inhibition on GtH release (R. E. Peter, personal
communication). The present study sought more
direct evidence of negative feedback of sex ster-
oids in female goldfish by using techniques of
ovariectomy and steroid replacement therapy.

716 M. KOBAYASHI AND N. E. STACEY

MATERIALS AND METHODS

Goldfish

Sexually mature goldfish were obtained from
Ozark Fisheries Co., Stoutland, Missouri in March
and held in 1000 liter stock tanks at 15°C under a
16L-8D photoperiod (lights on at 0800 hr) to main-
tain ovaries in a vitellogenic state. Fish were fed
with commercial trout pellets ad libitum at least
once daily. In September, the water temperature
was gradually raised to 20°C in an attempt to
promote and maintain vitellogenesis [15].

Experiment 1

On October 24, 45 females weighing 20-87 g
were selected from stock tanks and divided into
three groups. Sixteen fish were totally ovariecto-
mized (OVX) as described by Kobayashi et al. [2].
Fifteen fish were sham operated (SHAM), leaving
the ovaries intact. The remaining 14 fish (IN-
TACT) received no surgery. Ovarian develop-
ment of operated fish was variable, ranging from
regressed (non-vitellogenic) to post-vitellogenic.
Although it is difficult to calculate gonadosomatic
index (GSI) of OVX fish, as the ovary is aspirated
and thus fragmented during removal, fish from the
same stock which were not used for this experi-
ment had GSI values of 8.4-14.5 and 1.2-1.5 in
mature and regressed fish, respectively. After the
operation, all the experimental fish were held
overnight in 80 liter aquaria (20°C, 16L-8D) con-
taining 0.5% NaCl and 25 ppm tetracycline (Sig-
ma). Two days after the operation, the tetracy-
cline treatment was repeated. One week after the
operation, the sutures were removed, and fish
were transferred to two 90 liter experimental tanks
(20°C, 16L-8D) until the end of the experiment.
To prevent the occurrence of repeated spon-
taneous ovulation in INTACT and SHAM fish,
which might have influenced basal blood GtH
levels [15], fish were held in flowing water with no
aquatic vegetation [16]. Fish were fed daily with
trout pellets.

At 2-week intervals from two to ten weeks after
the operations, blood samples (300-400 1) were
taken from the caudal vasculature after anesthe-
tization with MS222. Since blood sampling at the

time of operation caused high mortality in pre-
liminary studies (Kobayashi, unpubl. data), blood
sampling was started one week after the removal
of the sutures. Blood samples were allowed to clot
on chipped ice and centrifuged (10,000 x g, 5 min).
Serum samples were stored at —20°C until
assayed. After the last blood sampling, fish were
dissected to check the completeness of the ovariec-
tomy in OVX fish, and to determine GSI in
INTACT and SHAM fish. In addition, pituitary
glands were collected, sonicated with GtH assay
buffer, centrifuged (10,000 g, 20 min), and the
supernatant frozen until GtH radioimmunoassay
(RIA).

Experiment 2

On November 7, 39 fish weighing 20-70 g were
selected from the same stock as Experiment 1, and
divided into three groups consisting of 13 fish. All
fish were ovariectomized and empty Silastic cap-
sules (CONTROL) or capsules containing either
testosterone (T) or E, (E) were implanted in the
body cavity before the incision for ovariectomy
was sutured. After the implantation, fish were ~
subjected to the same schedule as in Experiment 1
except tht the experimental period was reduced (8
weeks), and the water temperature was lowered to
16°C one week after implantation to prevent the
occurrence of steroid-induced GtH surges which
occur in T and E fish at 20°C [2].

Steroid capsules were prepared as decribed pre-
viously [2]. Silastic tubing (Dow Corning, 1.98
mm i.d., 3.18 mm o.d.) was cut into 25 mm length
for testosterone capsule and 30 mm length for E,
and empty capsules. Each testosterone capsule
contained approximately 10 mg of crystaline tes-
tosterone, and each E, capsule approximately 30
mg of E,. CONTROL, T, and E fish received one
empty capsule, one testosterone capsule, and two
E, capsules, respectively.

Hormone Assays

Serum GtH levels and pituitary GtH content
were measured by RIA as described by Peter et al.
[17]. Serum steroid levels were measured by RIA
as described previously [18, 19]. Minimum detect-
able level of the steroids in this study was 0.3
ng/ml.

GtH and Steroids in OVX Goldfish WAY)

Statistics

Data were analyzed by ANOVA and least
square means test after log-transformation.

RESULTS

Experiment 1

Only one fish (OVX) died during the ex-
perimental period. At the end of the experiment,
most INTACT and SHAM fish had regressed
ovaries despite having been kept in 20°C water

30

20

INTACT

Gonadotropin (ng/ml)

2 4 6 8 10

Weeks after operation

Fic. 1. Effect of ovariectomy on blood GtH levels in
female goldfish. Each point shows the mean and
SEM (INTACT, N=14; SHAM, sham-operated, N
=15; OVX, ovariectomized, N=15). Small @,
significance compared to INTACT; small a, signi-
ficance compared to SHAM. Levels of significance:
one symbol, p<0.05; two symbols, p< 0.01; three
symbols, p< 0.001.

(see [15]). GSI values of INTACT and SHAM
groups were 2.7+1.2 and 1.6+0.4 (mean and
SEM), respectively. At autopsy, eight of the OVX
fish appeared to have been completely ovariecto-
mized, whereas, one to several nonvitellogenic
oocytes were found in the other seven OVX fish.
Since there was no difference in blood levels of
steroids and GtH between the OVX fish with
oocytes and those without oocytes, the data of all
OVX fish were combined.

GtH levels in OVX fish were significantly higher
than those of INTACT and SHAM fish throughout

Testosterone

INTACT

(ng/ml)

Estradiol
& INTACT
SHAM
i= °
~~
2 e
aaie”4
OVX
i xian om ea
2 4 6
Weeks after operation
Fic. 2. Changes in blood testosterone (upper) and

estradiol (lower) levels after operation. Abbrevia-
tions and symbols as in Fig. 1.

718 M. KOBAYASHI AND N. E. STACEY

TABLE 1. Effects of ovariectomy and steroid hormone im-

plantation in female goldfish

Experiment 1

Treatment* INTACT
Mean+SEM ao aeor!,
N 10
Experiment 2
Treatment* CONTROL
Mean +SEM 67.4+3.3
N tl

@ See text for details.

20

CONTROL

Gonadotropin (ng/ml)

2 4 6 8

Weeks after implantation

Fic. 3. Effect of steroid hormone implantation in ovar-
iectomized goldfish. Each opint shows the mean and
SEM (CONTROL, empty capsule implanted, N=
12; T, testosterone capsule implanted, N=12; E,
estradiol capsule implanted, N=8). Small O, signi-
ficance compared to CONTROL. Levels of signi-
ficance as in Fig. 1.

Pituitary GtH content (g/pituitary)

SHAM OVX
Se) ae Sec 68.2+21.9
10 8
v E
Sk ae 1/4) 56.2+7.6
7 4

30 Testosterone

20 yada

10

(ng/ml)

CONTROL & E

(ng/ml)

CONTROL

2 6

Weeks after implantation

Fic. 4. Changes in blood testosterone (upper) and

estradiol (lower) levels after steroid hormone im-
plantation. Abbreviations and symbols as in Fig. 3.
Small ©, significance compared to CONTROL;
small @, significance compared to T. Levels of
significance as in Fig.1. Testosterone levels of
CONTROL and E were nondetectable.

GtH and Steroids in OVX Goldfish 719

the experiment (p<0.01-0.001, Fig.1). GtH
levels in INTACT and SHAM fish showed a
decrease (p<0.01 and p<0.05, respectively) at
four weeks. Testosterone levels in OVX fish were
lower than those in INTACT and SHAM fish at
two weeks (p<0.001 and p<0.05, respectively,
Fig. 2), and nondetectable at 6 and 10 weeks. E,
levels of OVX fish were measurable throughout
the experiment and lower than those of INTACT
fish at 2 and 4 weeks (p<0.05) but not different
from those of SHAM fish at any time. At two
weeks after ovariectomy, E> levels in some OVX
fish were similar to those of INTACT fish, but
there was no correlation between E> levels and
completeness of ovariectomy. SHAM fish had
levels of testosterone intermediate between those
of INTACT and OVX fish, and a change in E>
similar to OVX fish.

There were no significant differences in pituitary
GtH content among groups (Table 1).

Experiment 2

Twelve of 13 CONTROL and T fish, and eight
of 13 E fish survived. One to several oocytes were
found in four CONTROL and one E fish. As
completeness of ovariectomy was not correlated
with blood hormone levels, data of completely and
incompletely ovariectomized fish were combined
as in Experiment 1.

Serum GtH levels of steroid implanted fish were
significantly lower than those of CONTROL fish
(T, 4 to 8 weeks, p<0.05-0.01; E, 2 to 8 weeks,
p<0.01-0.001; Fig. 3). In the steroid implanted
fish, serum levels of implanted steroids were high-
er than levels of the same steroid in CONTROL
fish (Fig. 4).

Pituitary GtH content did not differ among
groups (Table 1).

DISCUSSION

The present study demonstrated the existence of
negative feedback of gonadal steroids on basal
GtH secretion in female goldfish. Experiment 1
showed that ovariectomy is associated with ele-
vated blood GtH levels, whereas Experiment 2
showed that these elevated levels were depressed
by testosterone or E, replacement. Male goldfish

also seem to have a negative feedback system since
elevated GtH levels were observed three weeks
after castration (Kobayashi, unpubl. data).
Although the results of the present study are
consistent with those obtained in other studies of
gonadectomized teleosts [5, 9-11], it also is clear
that the relationship between steroid and GtH
levels is not a simple one.

The elevation of GtH levels after ovariectomy in
Experiment 1 suggests that pituitary GtH cells
receive inhibitory effects from the ovary. Sex
steroids are expected to be one of the ovarian
inhibitory factors. However, although testoster-
one and E, levels in INTACT and SHAM fish
declined dramatically during Experiment 1, GtH
levels did not increase, as would be predicted if
GtH were simply under negative feedback control
by ovarian steroids, but declined. The reason for
this GtH decline is not known.

Although testosterone and E>, implants reduced
GtH levels in Experiment 2, the decrease in testos-
terone and E, in Experiment 1 are not likely to be
the only factors responsible for elevation of GtH
levels, because reduced levels of these two steroids
in SHAM fish at week 2 were not associated with
increased levels of GtH. It is possible that the
goldfish ovary produced some other GtH release-
inhibiting factors, such as ‘inhibin’ identified in
some mammals [20].

The results of Experiment 2 are more straight-
forward than those of Experiment 1. Here, testos-
terone and E, implants, which produced blood
steroid levels typical of mature female goldfish,
significantly reduced blood GtH. These results
indicate that sex steroids are one of the ovarian
factors which inhibit GtH secretion.

The high mortality of E, implanted fish, which
occurred in this and a previous study [2], is be-
lieved to be caused by high concentration of blood
vitellogenin. Vitellogenin levels in E fish in Ex-
periment 2 (30 mg/ml) were much higher than
those of normal mature females (2.0-10 mg/ml)
(Kobayashi and Nunez-Rodriguez, unpubl. data),
presumably because vitellogenin could not be re-
moved from the blood in ovariectomized fish. A
new method for longterm E, administration need
to be developed.

In the present study, both testosterone and E,

720 M. KoBAYASHI AND N. E. STACEY

were effective in inhibiting GtH secretion. In male
catfish, where the mode of action of steroid nega-
tive feedback has been intensively studied, it has
been proposed [5, 21-24] that catechol estrogens
derived from estrogen or aromatizable androgens
compete with dopamine for methylation by
catechol-O-methyltransferase. The presence of
catechol estrogens thus inhibits the inactivation
(methylation) of dopamine, with the result that
more dopamine is available to inhibit GtH release.
Since goldfish have dopaminergic inhibition of
GtH release, the catfish model may be applicable
to the goldfish negative feedback system.

A stimulatory effect of sex steroids on pituitary
GtH accumulation has been shown in some juve-
nile fish [4-6, 25, 26] and adult catfish [5]. In
juvenile salmonids [4, 22], this mode of positive
feedback may be important for induction of puber-
ty. Although we failed to find a stimulation by sex
steroids of pituitary GtH content of adult female
goldfish, such an effect on juvenile goldfish cannot
be ruled out.

To our knowledge, the female goldfish is the
only teleost in which both positive and negative
feedback effects of gonadal steroids on GtH re-
lease have been demonstrated. During ovulatory
cycles in the spawning period, basal circulating
GtH in female goldfish is much lower than ovula-
tory surge levels [19, 27] but higher than in regres-
sed fish (Kobayashi in preparation) or immature
fish [28]. This intermediate GtH level (tonic
secretion) of vitellogenic fish is considered to be
maintained mainly by a negative feedback of ova-
rian steroids and to be an optimum level for
ovarian development. It is well known in aquacul-
ture that repeated injections of high doses of GtH
do not always advance the maturity of vitellogenic
ovaries although the same dose is effective in
inducing ovulation in ovaries which have com-
pleted vitellogenesis [29]. The ovulatory GtH
surge, which occurs at the end of each cycle (cyclic
secretion), is considered to be regulated by a
positive feedback of testosterone [2]. Trudeau et
al. [7] have shown that testosterone and E, po-
tentiate GnRH-induced GtH release, and sug-
gested that these steroids in vitellogenic fish may
facilitate the occurrence of the ovulatory GtH
surge.

ACKNOWLEDGMENTS

We thank Dr. Jesus Nunez-Rodriguez for measuring
serum vitellogenin. This study was supported by the
National Science and Engeneering Research Council of
Canada (Grant A2903 to N.E.S.) and the Alberta Herit-
age Foundation for Medical Research (Postdoctoral Fel-
lowship to M.K.).

REFERENCES

1° Peter, R. E., Chang,’ J. P.)” NahorakeaC-s.,
Omeljaniuk, R. J., Sokolowska, M., Shih, S. H. and
Billard, R. (1986) Interactions of catecholamines
and GnRH in regulation of gonadotropin secretion
in teleost fish. In “Recent Progress in Hormone
Research, Vol. 42”. Ed. by R. O. Greep, Academic
Press, Orlando, pp. 513-548.

2 Kobayashi, M., Aida, K. and Hanyu, I. (1989)
Induction of gonadotropin surge by steroid hormone
implantation in ovariectominzed and sexually re-
gressed female goldfish. Gen. Comp. Endocrinol.,
73: 469-476.

3. Flink, G. (1988) Gonadotropin secretion and its
control. In “The Physiology of Reproduction”. Ed.
by E. Knobil and J. Neil, Raven Press, Ltd., New
York, pp. 1349-1377.

4 Crim, L. W., Peter, R. E. and Billard, R. (1981) ©
Onset of gonadotropic hormone accumulation in the
immature trout pituitary gland in response to
estrogen or aromatizable androgen steroid hor-
mones. Gen. Comp. Endocrinol., 44: 374-381.

5 De Leew, R., Wurth, Y. A., Zandbergen, M. A.,
Peute, J. and Goos, H. J. Th. (1986) The effects of
aromatizable androgens, non-aromatizable andro-
gens, and estrogens on gonadotropin release in
castrated African catfish, Clarias gariepinus (Bur-
chell): A physiological and ultrastructural study.
Cell. Tissue Res., 243: 587-594.

6 Dufour, S., Delerue-Le Belle, N. and Fontaine, Y.
A. (1983) Effects of steroid hormones on pituitary
immunoreactive gonadotropin in european eel,
Anguilla anguilla L. Gen. Comp. Endocrinol., 52:
190-197.

7 Trudeau, V., Somoza, G. and Peter, R. E. (1989)
Estradiol administration potentiates the gonadotro-
pin response to gonadotropin-releasing hormone in
the female goldfish. In “Abstracts of XIth Intl.
Symp. on Comp. Endocrinol.”, Malaga, Spain, p-
340.

8 Breton, B., Jalabert, B., Fostier, A. and Reinaud,
P. (1975) Induction de decharges gonadotropes
hypophysaires chez la carp (Cyprinus carpio L.) a
Vaide du citrate de cisclomiphene. Gen. Comp.
Endocrinol., 25: 400-404.

9 Billard, R., Richard, M. and Breton, B. (1977)

10

11

12

13

14

15

16

17

18

19

GtH and Steroids in OVX Goldfish TPA

Stimulation of gonadotropin secretion after castra-
tion in rainbow trout. Gen. Comp. Endocrinol., 33:
163-165.

Bommelaer, M. C., Billard, R. and Breton, B.
(1981) Changes in plasma goandotropin after ovar-
iectomy and estradiol supplementation at different
stages at the end of the reproductive cycle in the
rainbow trout (Salmo gairdneri R.). Reprod. Nutr.
Dévelop., 21: 989-997.

Van Putten L. J. A., Peute, J., Van Oordt, P. G. W.
J., Goos, H. J. Th. and Breton, B. (1981) Glycopro-
tein gonadotropin in the plasma and its cellular
origin in the adenohypophysis of sham-operated and
ovariectomized rainbow trout, Salmo gairdneri. Cell
Tissue Res., 218: 439-448.

Nagahama, Y. and Yamamoto, K. (1969) Basophils
in the adenohypophysis of the goldfish (Carassius
auratus). Gunma Symp. Endocrinol., 6: 39-55.
Nagahama, Y. (1973) Histo-physiological studies on
the pituitary gland of some teleost fishes, with
special reference to the classification of hormone-
producing cells in the adenohypophysis. Mem. Fac.
Fish. Hokkaido Univ., 21: 1-63.

Billard, R. and Peter, R. E. (1977) Gonadotropin
release after implantation of anti-estrogens in the
pituitary and hypothalamus of goldfish Carassius
auratus. Gen. Comp. Endocrinol., 32: 213-220.
Kobayashi, M., Aida, K., and Hanyu, I. (1988)
Hormone changes during ovulatory cycle in goldfish.
Gen. Comp. Endocrinol., 69: 301-307.

Stacey, N. E., Cook, A. F. and Peter, R. E. (1979)
Spontaneous and gonadotropin-induced ovulation in
the goldfish, Carassius auratus L: Effects of external
factors. J. Fish Biol., 15: 305-316.

Peter, R. E., Nahorniak, C. S., Chang, J. P. and
Crim, L. W. (1984) Gonadotropin release from the
pars distalis of goldfish, Carassius auratus, trans-
planted beside the brain or into the brain ventricles:
Additional evidence for gonadotropin-release in-
hibiting factor. Gen. Comp. Endocrinol., 55: 337-
346.

Aida, K., Kato, T. and Awaji, M. (1984) Effect of
castration on the smoltification of precocious male
masu salmon Oncorhynchus masou. Bull. Japan.
Soc. Sci. Fish., 50: 565-571.

Kobayashi, M., Aida, K. and Hanyu, I. (1987)
Hormone changes during ovulation and effects of

20

21

Vij

23

24

75)

26

i

28

uy)

steroid hormones on plasma gonadotropin levels
and ovulation in golsfish. Gen. Comp. Endocrinol.,
67: 24-32.

De Jong, F. H. and Robertson, D. M. (1985)
Inhibin: 1985 update on action and purification.
Mole. Cell. Endocrinol., 42: 92-103.

Lambert, J. G. D., Goos, H. J. Th. and Van Oordt,
P. G. W. J. (1984) Aromatase activity in the brain
of the rainbow trout, Salmo gairdneri. Gen. Comp.
Endocrinol., 53: 459.

Goos, H. J. Th. (1987) Steroid feedback on pituit-
ary gonadotropin secretion. In “Reproductive Phy-
siology of Fish 1987”. Ed. by D. R. Idler, L. W.
Crim. and J. M. Walsh, Marine Science Research
Laboratory, Memorial University of Newfoundland,
St. John’s, pp. 16-20.

Timmers, R. J. M., Granneman, J. C. M., Lambert,
jenGeaeDasand= Vane Oordt Ps (Gra We Je (1988)
Estrogen-2-hydroxylase in the brain of the male
African catfish, Clarias gariepinus. Gen. Comp.
Endocrinol., 72: 190-203.

Timmers, R. J. M., and Lambert, J. G. D. (1989)
Catechol-O-methyltransferase in the brain of the
male African catfish, Clarias gariepinus; distribution
and significance for the metabolism of
catecholestrogens and dopamine. Fish Physiol.
Biochem., 7: 201-210.

Gielen, J. Th., Goos, H. J. Th., Peute, J. Van Den
Bosch, R. A. and Van Oordt, P. G. W. J. (1982)
The brain-pituitary-gonadal axis in rainbow trout,
Salmo gairdneri: Gonadal hormones and_ the
maturation of gonadotropic cells. Cell Tissue Res.,
225: 45-56.

Ikuta, K., Aida, K., Okumoto, N. and Hanyu, I.
(1987) Effects of sex steroids on the smoltification of
masu salmon, Oncorhynchus masou. Gen. Comp.
Endocrinol., 65: 99-110.

Stacey, N. E., Cook, A. F. and Peter, R. E. (1979)
Ovulatory surge of gonadotropin in the goldfish,
Carassius auratus. Gen. Comp. Endocrinol., 37:
246-249.

Kobayashi, M., Aida, K. and Hanyu, I. (1985)
Radioimmunoassay for silver carp gonadotropin.
Bull. Japan. Soc. Sci. Fish., 51: 1085-1091.
Billard, R. (1989) Endocrinology and fish culture.
Fish Physiol. Biochem., 7: 49-58.

ad
"
<
\
za
i :
L i 5
3 =
'
. a
3

i oon eseonm on re aeny uf

| {ane whet are ee ; = eT ee Hess
' bys sven a Abe.

vo

_ set eg > ‘
, 7 i ‘ = \d 2 ot
nee-in ue at cdeemiinathS eon fia auphianttiog
tees : 3 '
/ Nv e“e - & _. 5 aa Ae Sine BA ae R : PAs. als =
A dt | 2" ht a ae a ak, rea see STL tae i NP i mt * ad ad es i
cn ees : :
te ee ae i SesaetennvivcismAad joo > | eee & ;
ae PS, , cae There? ., 7al pa pt aro aphe Apia if ' Leo ent HS dss Sent z
7 " ° ; * - ‘ v y a ad :
: beth i
as. Bis Dee pe: oe
i Tenor fer Hu BEET ‘i
hag iat cs ones ; ; - = a i
ieee Were. sete Lieve at oi hseew Venpe
Ml Pan en ee ru oe Mi t & 435 -
= Q cd pe
~’ aD 5 : ¥ = aa é “Tee ; he oar * -_ iv 4 of . Peewee 2
Nat) ABs . nts OF Gat S St 7 oe Poth ie - ENE hg serie §
j fs . rf 4
“ , ote - A a x ‘ ty = 7 Pies ee f
(neds » Re eet Fe tide, 2 EY) a wth. eae 1 it Tia ig st
an s7| =i 2 ae : sie Sy
oh Be f ‘ 4 )
a G + r ma a -
Sti te 1} ee um coe 4 < "
Se) Mae | : 7" } Pe Pe) " +
i : tas tt Bes RAT
tare f AAS 2 ap tare, 2
fey prrejatey AE cote ot biaxy Shae ns
Milt} out | potas
ty ag o t
: a '
E r A
, Je h
>
++ i .
E
Li J |
= ,
x
a Fi “se = ¥
& + Fy 5
‘
1
a 3 >
}
a ‘ 4 =
} = T=
*
\
)

RENOVA | Be cates

; ey a q. “p10
eh se: ous sae wee a a, ror &, ee eet)

PS SRS: ET pet * re e on if aa cs on
‘aabaeuasie nl yor i bape cette ee
Cal sae is: 4teta violas RMD AR EMETC ‘sd | a eyatiane

ic EAL 8» Sarye 2 al sapped iapet ny a

2s neeenen, nt isos Sen

ago,

has

ty anal Ye

r Senne bee:

Stead pers ‘aber HAL WSs itive papas tg : ‘Yeolenbe

ZOOLOGICAL SCIENCE 7: 723-730 (1990)

Stress-induced Adrenal Steroidogenesis in Neonate
Mallard Ducklings and Domestic Chickens

WILLIAM N. Homes, JAMES CRONSHAW and José L. REDONDo!

Department of Biological Sciences, University of California,
Santa Barbara, CA93106, USA

ABSTRACT —Stress-induced changes in adrenal steroidogenesis were compared in Mallard ducklings
and domestic white leghorn chickens on emergence at the end of embryogenesis. The effects of
restraint, and restraint preceded by blood sampling, on the plasma corticosterone, aldosterone and
deoxycorticosterone concentrations in each species were used to assess and compare these responses. A
hypophysiotropic reflex capable of stimulating a responsive adrenal steroidogenic tissue is fully
developed in newly-hatched ducklings. Restraint, and not hemorrhagic shock associated with taking a
blood sample, is the major factor responsible for evoking the stress-induced increases in corticosterone,
aldosterone and deoxycorticosterone release. The responses of the newly-hatched chicken to restraint,
and to restraint preceded by bleeding, were qualitatively similar to those of the ducklings. In contrast to
the duckling, however, hemorrhagic shock was more effective than restraint in causing an increase in
plasma corticosteroid concentration. These data indicate that at least some of the sensory pathways
responsible for the activation of the hypophysiotropic reflex in the chicken are fully functional at the
time of hatching; this observation is contrary to that of other investigators who have concluded that the
hypothalamo-hypophyseal system is incapable of stimulating adrenal steroidogenesis during the first few

© 1990 Zoological Society of Japan

days of post-natal life.

INTRODUCTION

The hypophysiotrophic reflex in the Mallard
duck appears to have developed to a fully function-
al state by end of embryogenesis, and exposure of
newly-hatched birds to stress causes a rapid and
profound increase in their plasma corticosterone
concentrations [1]. In contrast, the results of
studies on the chicken suggest that this reflex may
be non-functional in the neonate, and may remain
so for several days after hatching [2-5]. It is
possible, however, that this discrepancy is related
to the slightly different experimental protocols that
were used in the investigations on the Mallard
duckling and the chicken.

In each of the studies on the chicken, the mean
plasma corticosterone concentration measured in a
group of stressed birds was compared to the mean

Accepted February 8, 1990
Received December 11, 1989

' Present address: Centro de Investigacion sobre Ing-
enieria Genetica y Biotecnologia, AP 510-3, Colonia
Miraval 62270, Cuernavaca, Morelos, Mexico.

concentration measured in a group of unstressed
birds [2-4]. On the other hand, in the experiments
on the neonatal ducklings the plasma hormone
concentrations were measured in individual birds
before and after their exposure to stress [1]. Each
of these protocols has its advantages and weak-
nesses. For example, a comparison of mean con-
centrations derived from highly variable groups of
stressed and unstressed birds may not reveal a
significant induced response, but if the variability
is low there will be no doubt regarding the cause of
an induced response. In contrast, an induced
response is more easily recognized if the unstress-
ed and stressed states are compared in each
individual; this procedure, however, renders it
impossible to determine if the response was due to
hemorrhagic shock associated with taking the first
blood sample, the period of exposure to the stress-
or, or the combined effects of these both factors.

In the present study we have used both of these
experimental protocols in order to compare the
effect of restraint alone with that of a similar
period of restraint preceded by blood sampling on

724 W.N. HotmeEs, J. CRONSHAW AND J. L. REDONDO

the circulating concentrations of corticosteroids in
neonatal Mallard duckling. Also, using the same
experimental plan we have compared the re-
sponses of the ducklings with those of similarly
treated neonatal domestic chickens.

MATERIALS AND METHODS

Fertilized Mallard duck eggs were purchased
from a commercial breeder (Whistling Wings,
Hanover, IL) and fertilized White Leghorn chick-
en eggs were donated by Rosemary Farm (Santa
Maria, CA). The eggs were artificially incubated
at 37.4°C and 85% relative humidity, and were
rotated through 90° every 2 hr until the embryos
showed the first signs of starting to break through
the shell (the “pipping” stage). At this time the
eggs were transferred to a stationary tray on the
floor of the incubator where the hatching process
was completed; upon emergence, the neonates
were given an ad libitum supply of drinking water.
In a preliminary study, the changes in egg-weight,
the periods of incubation up to the “pipping” stage
of development, the hatching times, and the neon-
ate body weights at emergence were recorded for
fertilized Mallard duck and chicken eggs incubated
under these conditions.

Protocol

The birds were used experimentally on the
morning of the day following their emergence, at
which time they were completely dry and ambula-
tory. Two series of experiments were performed
on the neonates of each species; in each series the
experiments were performed at between 09.00 and
10.00 hr and blood samples were taken by heart
puncture, using heparinized syringes. In the first
series, each bird was gently restrained by wrapping
it in muslin gauze immediately upon removal from
the incubator and a blood sample was taken 30 min
later; the responses of these birds were compared
with those of a separate group of unstressed birds
from which a blood sample was taken immediately
upon their removal from the incubator. In the
second series, a blood sample (0.3 ml) represent-
ing approximately 1% (w/v) of the body weight
was taken within 1 min of removing each bird from
the incubator; the bird was then restrained like the

birds in the first series of experiments and a second
blood sample was taken 30 min later. Since it was
not possible to obtain a sufficiently large sample of
blood from each individual bird, samples taken
from three birds were pooled; the pooled samples
were then centrifuged for 5 min at 12,300 g and
5°C and the plasma was stored at —20°C prior to
analysis.

Sample purification

Plasma samples were thawed at room tempera-
ture and a 2.0 ml aliquot of each sample was trans-
ferred to a new scintillation vial containing 8,000
dpm of chromatographically pure [1, 2-°H]-corti-
costerone, [1, 2-°H] aldosterone and [1, 2-H]-
deoxycorticosterone, (Amersham Corporation,
Arlington Heights, IL). The samples were allowed
to equilibrate for 1 hr at room temperature before
extraction with 10 volumes cold anhydrous ethyl
ether. The ether phase was separated, dried under
nitrogen, redissolved in 0.5 ml benzene : cyclohex-
ane: methanol: water (25:12.5:4:1) and _ puri-
fied by gel filtration chromatography using
Sephadex LH-20 and the same solvent system.
The fractions containing labelled corticosterone,
aldosterone and deoxycorticosterone were col-
lected and pooled separately, evaporated to dry-
ness under nitrogen and redissolved in 1.0 ml of
buffer containing 0.1 M tricine, 0.18 M NaCl and
0.015 M NaN3 at pH8. The radioactivity con-
tained in 0.2 ml of this extract was used to deter-
mine the recovery of each steroid and the remain-
der of sample was stored at —20°C.

Hormone analysis

The concentrations of corticosterone, aldoster-
one and deoxycorticosterone were measured
by radioimmunoassay [6-8]. The corticosterone
antiserum was produced in rabbits immunized
with corticosterone-2i-hemisuccinyl thyroglobulin
(ICN Biomedicals Inc., Costa Mesa, CA). Sensi-
tivity for this assay was 15 pg corticosterone per
assay tube, intra-assay variation was 5.0% and
interassay variation was 12.8%. Aldosterone anti-
serum was raised in sheep with an aldosterone-3-
oxine-bovine serum albumin conjugate (Arnel
Products Co., New York, NY); sensitivity was 10
pg of aldosterone per assay tube and intra- and

Stress-Induced Responses in Avian Neonate 725

interassay variation was 3.0% and 10.3% respec-
tively. Deoxycorticosterone antiserum was pro-
duced in rabbits using a deoxycorticosterone-3-
oxine-bovine serum albumin conjugate (Endocrine
Sciences, Tarzana, CA); sensitivity was 4 pg deox-
ycorticosterone per assay tube and intra- and in-
terassay variation was 8.3% and 6.8% respec-
tively.

Statistics

All values are expressed as means+standard
error of the mean (SEM). In the first experimental
series, the mean stress-induced changes were esti-
mated from the differences between the paired
values measured before and after exposure to
stress. In the second experimental series the SEM
of the difference between the mean hormone
concentration in the unstressed birds and the cor-
responding mean value measured in the stressed
bird were estimated from the pooled variances
according to the method described by Sokal and
Rohlf [9]. Mean values and mean differences were
compared by either the one- or the two-tailed
Student t-test [9]. Standard curves for the radioim-
munoassays were fitted by the method of least
squares.

RESULTS

The initial weights of the fertilized Mallard duck
eggs and their weights at the pipping stage of
development did not differ significantly from the
corresponding values recorded for the domestic
chicken; both of these mean values for the domes-

tic chicken, however, were more variable than
those for the Mallard duck eggs (Table 1).

The period of incubation up to pipping and the
interval between the occurrence of pipping and the
ultimate emergence of the neonate were each
significantly longer for the Mallard duck than for
the domestic chicken (Table 1); the mean hatch-
ing-time for the Mallard ducklings being 25.4+3.7
hr while the corresponding value for the chicken
was 17.1+1.1 hr. The mean body weights of the
neonate ducklings and chickens were the same but
the absolute and relative (mg per 100g body
weight) weights of the adrenal glands in the neon-
ate ducklings were approximately 30% heavier
than those in the chickens (Table 1).

The mean plasma concentrations of corticoster-
one, aldosterone and deoxycorticosterone in the
group of ducklings that had been restrained for 30
min prior to blood sampling were significantly
higher than the corresponding concentrations re-
corded in the group of unstressed ducklings (Table
2). In the group of newly hatched chickens that
had been treated in exactly the same way the mean
plasma concentrations of corticosterone and the
deoxycorticosterone, but not aldosterone, were
significantly higher than the corresponding concen-
trations measured in the unstressed birds (Table
2). The differences between the mean concentra-
tions of corticosterone and deoxycorticosterone
recorded in the groups of unstressed and stressed
chickens, however, were significantly lower than
the corresponding differences recorded in the
ducklings (Table 2).

Significant stress-induced increases in the plas-

TABLE 1. The incubation periods of fertilized Mallard duck eggs (N=15) and White Leghorn domestic
chicken eggs (N=15) incubated in the laboratory at 37.5°C and 85% relative humidity. The body
weights of the neonates derived from these eggs were recorded at the time they had completely emerged
from the shell and were free-standing. The eggs were rotated through 90 degrees every two hours up to
the pipping stage of development. All means are expressed +SEM

Egg weight (g) Incubation period (hr) Neonate
Initial Pipping To pipping To emergence Body wt (g) Adrenal wt (mg)
Duck 63.2 54.8 625.1 650.5 40.9 7.86
st OE S2 +0.28 se +0.88 +0.84 an) Sl
Chicken Olmigs 54.5"° SOS as SO 40.8 So
+1.08 +1.47 tall 2 +1.4 +0.68 +0.30

** p<0.01, * p<0.05, and ns=not significantly different with respect to the corresponding value recorded for

the incubated Mallard duck eggs.

——————

726 W.N. Hotmgs, J. CRONSHAW AND J. L. REDONDO

TABLE 2. The plasma corticosteroid hormone concentrations in unstressed
neonatal Mallard ducklings and domestic chickens, and the correspond-
ing concentrations in separate groups of birds after a 30-min period of
gentle immobilization stress

Corticosteroid concentration (ng per ml plasma)

Corticosterone Aldosterone DOC
Duckling
Unstressed 14.7 0.451 0.444
se WellS +0.163 +0.053
Stressed 96.7** 1.931** 239474
+ 6.69 +0.200 +0.291
Difference 82.0 0.980 1.950
se 7/54! +0.265 +0.320
Chicken
Unstressed 10.2 0.300 0.124
qe 41338) +0.116 +0.018
Stressed 29.4** ORS) 0.283*
+ 2.85 +0.190 +0.068
Difference 19D 0.431"° 0.159t +
de 5). Il} +0.244 +0.071

* p<0.05, ** p<.001 and ™=not significant with respect to the corresponding
value recorded in the group of unstressed neonates. ** p<0.01 with respect to
the corresponding difference recorded in the duckling neonates.

TABLE 3. The plasma corticosteroid hormone concentrations (mean +SEM)
in neonatal Mallard duckling and domestic chickens before and after the
exposure of each individual to a 30 min period of gentle immobilization

stress
Corticosteroid concentration (ng per ml plasma)
Corticosterone Aldosterone DOC

Duckling

Prestress IBs7/ 0.326 0.542
+ 0.71 +0.089 +0.085

Stressed 65.2** Lat 178525
se 344 +0.178 +0.195

Increment SL 55) 1.388 1.243
fe Sell +0.118 +0.266

Chicken

Prestress 20.1 0.763 0.226
OS +0.168 +0.052

Stressed 63.255 1.461* 0.539*
ae 7 sell +0.174 +0.115

Increment Agile 0.697* 0.313
+ 6.86 +0.125 +0.082

* p<0.05, ** p<.001 with respect to the corresponding value recorded before
exposure to stress. ** p<0.01, * p<0.05 and ns=not significant with respect
to the corresponding increment recorded in the duckling neonates.

Stress-Induced Responses in Avian Neonate 727

ma concentrations of all three corticosteroids also
occurred in the neonates from which blood sam-
ples were taken before and after their exposure to
stress (Table 3). The induced increment in the
plasma corticosterone of the ducklings, however,
was significantly less (p<0.01 with 10 degrees of
freedom) than the difference recorded in the
groups of stressed and unstressed ducklings (Table
2). In contrast, the induced increment in corticos-
terone measured in the chickens was significantly
greater (p<0.05 with 10 degress of freedom) than
the difference recorded in the groups of stressed
and unstressed chickens (Table 2). The stress-
induced increases in plasma aldosterone and deox-
ycorticosterone concentrations in both the duck-
lings and the chickens were similar to the corres-
ponding differences measured in the groups of
stressed and unstressed birds (cf. Tables 2 and 3).

DISCUSSION

The results of these experiments on newly-
hatched ducklings confirm our earlier observation
that a hypophysiotropic reflex capable of stimulat-
ing a responsive adrenal steroidogenic tissue is
fully developed at the end of the hatching process
[1]. The present data, however, establish that
restraint, and not hemorrhagic shock associated
with taking the first blood sample, was the major
factor in eliciting the response.

Our observations on the neonatal chicken,
however, are quite contrary to those reported by
other investigators [2-3]. These workers con-
cluded that the hypothalamo-hypophyseal system
in the chicken, like that in some mammals [10-14],
is incapable of stimulating adrenal steroidogenesis
during the first few days of post-natal life. The
discrepency between the results of our studies and
those of Wise and Frye [2] and Freeman [3],
however, may reflect the different conditions im-
posed upon the birds to test their responses.
Freeman [3] reduced the environmental tempera-
ture from 30 to 20°C, placed the birds in separate
cages to prevent huddling, and collected only one
blood sample from each bird to minimize the
effects of handling. In contrast, Wise and Frye [2]
fractured a leg bone 20 min before taking a blood
sample. It is tempting to conclude that fracturing a

bone may constitute a more severe stress, and may
be more likely to stimulate the hypophysiotropic
reflex, than either reducing the environmental
temperature or gently restraining the birds. The
situation is not quite so simple, however, for the
responses we have observed in the newly-hatched
chicken clearly indicate that at least some of the
sensory pathways leading to the activation this
reflex are functional at the time of hatching.

In birds, as in mammals, stimulation of the
sympatho-chromaffin axis, as well as the hypotha-
lamo-pituitary-adrenal axis, is characteristic of
their short-term responses to stress [15-23]. In
addition to their well-known vasopressor and gly-
colytic effects, there is reason to believe that
catecholamines released under conditions of stress
may also act to facilitate, and perhaps enhance, the
responses of the adrenal steroidogenic tissue to
corticotropin [19, 24, 25]. The mechanism by
which this may occur, however, is not clear. One
possibility is that a direct aminergic stimulation of
the hypothalamus further increases the discharge
of corticotropin-releasing hormone (CRH) and
thus contributes to the overall stress-induced en-
hancement of corticosteroid hormone synthesis
[19, 22, 26]; the avian hypothalamus does contain a
CRH-like peptide that is capable of triggering the
release of ACTH from the pituitary [27]. Another
possibility is that circulating catecholamines enter-
ing the hypothalamo-hypophyseal portal system
may stimulate a release of ACTH by acting direct-
ly on the pituitary corticotrophs [28, 29]. A third
possibility is that catecholamines may bind to beta
receptors in the plasma membrane of the adrenal
steroidogenic cells and stimulate hormone synth-
esis directly; the evidence in support of this
hypothesis is scant and equivocal [30, 31].

Circulating catecholamines may also influence
adrenal steroidogenesis indirectly by causing
adrenal blood flow to increase [32] and so facilitate
the diffusion of stored hormone from the ster-
oidogenic cells; this type of effect may account for
the quick increases in plasma corticosterone con-
centrations that occur immediately following expo-
sure of adult birds to stress [33, 34]. In the
neonatal duckling, however, corticosterone is dis-
tributed in a relatively large volume [35] and the
adrenal glands contain only a small amount of this

=

728

hormone [36]. It is doubtful, therefore, whether
even an immediate stress-induced release of all of
the stored hormone from the neonate adrenal
glands would cause a significant rise in plasma
corticosterone concentration. Our preliminary
studies indicate that the same may also be true for
the neonatal chicken.

The stress-induced increases in aldosterone con-
centration that occurred in both the ducklings and
the chickens may have reflected in part an activa-
tion of the renin-angiotensin II system following
adrenergic stimulation of B; receptors in the jux-
taglomerular apparatus. Such a system has been
described in birds [37, 38] but little is known
regarding its role as a specific regulator of aldoster-
one synthesis.

It has also been suggested that direct cholinergic
stimulation of the adrenal steroidogenic cells may
be another factor involved in the stress-induced
response of the adrenal gland [31]. This hypothesis
is based on a series of unrelated observations in
several vertebrate species. Cholinergic nerves
terminals have been reported in the vicinity of the
adrenal steroidogenic cells in the sheep [39], pig,
rat, hamster [40, 41], five species of birds [42] and
two species of lizard [43]. Binding studies have
also indicated the presence of muscarinic receptors
in bovine zona fasciculate cells [44] and cholinergic
agonists have been shown to stimulate ster-
oidogenesis in vitro in amphibian and bovine
adrenal tissue [31, 45-48]. The possible roles of
catecholamines, and other putative neurogenic
secretogogues, as regulators of adrenal ster-
oidogenic function is one of the topics of our
continuing research.

ACKNOWLEDGMENTS

The work was supported by grants to J. Cronshaw and
W.N. Holmes from the University of California, Com-
mittee on Research and the National Science Foundation
(DIR-8820923), Washington, DC, USA.

REFERENCES

1 Holmes, W. N., Redondo, J. L. and Cronshaw, J.
(1968) Changes in the adrenal steroidogenic respon-
siveness of the Mallard duck (Anas platyrhynchos)
during early post-natal development. Comp.

10

11

12

13

14

15)

W.N. Homes, J. CRONSHAW AND J. L. REDONDO

Biochem. Physiol., 92A: 403-408.

Wise, P. M. and Frye, B. E. (1973) Functional
development of the hypothalamo-hypophyseal-
adrenal axis in the chick embryo. J. Exp. Zool., 185:
277-292.

Freeman, B. M. (1982) Stress non-responsiveness in
the newly-hatched fowl. Comp. Biochem. Physiol.,
72A: 251-253.

Freeman, B. M. and Flack, I. H. (1981) The
sensitivity of the newly-hatched fowl to corticotro-
pin. Comp. Biochem. Physiol., 70A: 257-259.
Freeman, B. M., Flack, I. H. and Manning, A. C.
C. (1980) The sensitivity of the domestic fowl to
corticotrophin. Comp. Biochem. Physiol., 67A:
561-567.

Nishida, S., Matsumura, S., Horino, M., Oyama,
H. and Tenku, A. (1976) A radioimmunoassay for
human plasma _ corticosterone. Endocrinology
(Japan), 23: 465-469.

Mayes, D. M., Furuyama, S., Kem., D. C. and
Nugent, C. A. (1970) Radioimmunoassay of plasma
aldosterone. J. Clin. Endocrinol., 30: 682-685.
Arnold, M. M. L. and James, V. H. T. (1971)
Determination of deoxycorticosterone in plasma;
double isotope and immunoassay methods. Steroids,
18: 789-891.

Sokal, R. R. and Rohlf, F. J. (1969) Biometry.
Freeman, San Francisco.

Jailer, J. W. (1950) Maturation of the pituitary-
adrenal axis in the newborn rat. Endocrinology, 46:
420-425. |

Levine, S. D., Glick, D. and Nakane, P. (1967)
Adrenal and plasma corticosterone and vitamin A in
rat adrenals during post-natal development. Endoc-
rinology, 80: 910-914.

Milkovic, S. (1969) Responsiveness of the pituitary-
adrenocortical system during embryonic and early
post-natal periods of life. In “Physiology and Pathol-
ogy of adaptation Mechanisms, Neural, Neuroen-
docrine, Humoral”. Ed. by E. Bajusz, Pergamon
Press, Oxford/New York, pp. 28-47.

Walker, C-D., Perrin, M., Vale, W. and Rivier, C.
(1986) Ontogeny of the stress response in the rat:
role of the pituitary and the hypothalamus. Endocri-
nology, 118: 1445-1451.

Walker, C-D., Sisonenko, P. C. and Aubert, M. L.
(1989) Modulation of the neonatal pituitary and
adrenocortical responses to stress by thyroid hor-
mones in the rat: effects of hypothyroidism and
hyperthyroidism. Neuroendocrinology, 50: 265-
Diey

Davison, T. F. (1975) The effects of multiple
sampling by cardiac puncture and diurnal rhythm on
plasma glucose and hepatic glycogen of the imma-
ture chicken. Comp. Biochem. Physiol., 50A: 569-
573:

16

17

18

19

20

Z|

22

23

24

25

26

Dah

28

29

30

Stress-Induced Responses in Avian Neonate 729

Edens, F. W. and Siegel, H. S. (1975) Adrenal

responses in high and low ACTH response lines of
chickens during acute heat stress. Gen. Comp. En-
docrinol., 25: 64-73.

El-Halawani, M. E., Waibel, P. E., Appel, J. R.
and Good, A. L. (1974) Effects of temperature
stress on catecholamines and corticosterone of male
turkeys. Am. J. Physiol., 224: 384-388.

Freeman, B. M. (1983) The adrenal glands. In
“Physiology and Biochemistry of the Domestic
Fowl”. Ed. by B. M. Freeman, Academic Press,
New York, pp. 191-209.

Harvey, S., Phillips, J. G., Rees, A. and Hall, T. R.
(1984) Stress and adrenal function. J. Exp. Zool.,
232: 633-645.

Howard, B. R. (1974) Assessment of stress in
poultry. Br. Vet. J., 130: 88-90.

Jurani, M., Vyboh, P., Lamosova, D. and Nvota, J.
(1978) Effect of restraint upon hypothalamic and
adrenal catecholamines in Japanese quail. Br. Poult.
Sci., 19: 321-325.

Rees, A., Harvey, S. and Phillips, J. G. (1985)
Adrenergic stimulation of adrenocortical secretion
in immature fowl. Comp. Biochem. Physiol., 81C:
387-389.

Zachariasian, R. D. and Newcomer, W. S. (1975)
Influence of corticosterone on the stress-induced
elevation of phenylethanolamine-N-methyltransfe-
rase. Gen. Comp. Endocrinol., 25: 332-328.
Freeman, B. M. and Manning, A. C. C. (1979) The
effects of repeated injections of adrenaline on the
response of the fowl to further alarm stimulation.
Res. Vet. Sci., 27: 76-81.

Hissa, R., George, J. C., John, T. M. and Etches,
R. J. (1982) Propranolol induced changes in plasma
catecholamines, corticosterone, T,, T3 and prolactin
levels in the pigeon. Horm. Metab. Res., 14: 606-
610.

Axelrod, J. and Reisine, T. D. (1984) Stress hor-
mones: their interaction and regulation. Science,
224: 452-459.

Stainer, I. M. and Holmes, W. N. (1969) Some
evidence for the presence of a corticotropin releas-
ing factor (CRF) in the duck (Anas platyrhynchos).
Gen. Comp. Endocrinol., 12: 350-359.

Rivier, C. and Vale, W. (1983) Modulation of
stress-induced ACTH release by corticotropin re-
leasing factor, catecholamines and vasopressin. Na-
ture, 345: 325-327.

Vale, W., Vaughan, J., Smith, M., Yamamoto, G.,
Rivier, J. and Rivier, C. (1983) Effects of synthetic
ovine GRF, _ glucocorticoids, catecholamines,
neurohypophysial peptides and other substances on
cultured corticotropic cells. Endocrinology, 113:
1121-1131.

Delean, A., Racz, K., McNicoll, N. and De-

31

32

38)

34

35)

36

37)

38

39

40

4]

42

43

44

sroisiers, M. L. (1984) Direct B-adrenergic stimula-
tion of aldosterone secretion in cultured bovine
adrenal subcapsular cells. Endocrinology, 115: 485-
492.

Benyamina, M., Leboulenger, F., Lirhamm, I.,
Delarue, C., Feuilloley, M. and Vaudry, H. (1987)
Acetylcholine stimulates steroidogenesis in isolated
frog adrenal gland through muscarinic receptors:
evidence for a desensitization mechanism. J. Endoc-
rinol., 133: 339-348.

Harrison, R. G. and Hoey, M. J. (1960) The
adrenal circulation, Blackwell, Oxford.

Beuving, G. and Vonder, G. M. A. (1978) Effect of
stressing factors on corticosterone levels in the plas-
ma of laying hens. Gen. Comp. Endocrinol., 36:
1153-1159.

Harvey, S., Merry, B. J. and Phillips, J. G. (1980)
Influence of stress on the secretion of corticosterone
in the duck (Anas platyrhynchos). J. Endocrinol.,
87: 161-171.

Holmes, W. N. and Kelly, M. E. (1976) The
turnover and distribution of labelled corticosterone
during post-natal development of the duckling
(Anas platyrhychos). Pflugers Archiv., 365: 145-
150.

Cronshaw, J., Holmes, W. N., Ely, J. A. and
Redondo, J. L. (1989) Pre-natal development of the
adrenal steroidogenic tissue in the Mallard duck
(Anas platyrhynchos). Cell. Tiss. Res., 258: 593-
601.

Chan, M. Y. and Holmes, W. N. (1971) Studies on
a “renin-angiotensin” system in the normal and
hypophysectomized pigeon (Columba livia). Gen.
Comp. Endocrinol., 16: 304-311.

Nishimura, H. (1980) Comparative endocrinology
of renin and angiotensin. In “The Renin-
Angiotensin System” Ed. J. A. Johnson and R. R.
Anderson, Plenum, New York, pp. 29-77.
Robinson, P. M., Perry, R. A., Hardy, K. J.,
Coghlan, J. P. and Scoggins, B. A. (1977) Innerva-
tion of the adrenal cortex in the sheep, Ovis ovis. J.
Anat., 124: 117-129.

Unsicker, K. (1969) Zur innervation der neben-
nierenrinde vom goldhamster. Z. Zellforsch., 95:
608-619.

Unsicker, K. (1971) On the innervation of the rat
and pig adrenal cortex. Z. Zellforsch., 116: 151-
156.

Unsicker, K. (1973) Fine structure and innervation
of the avian adrenal gland. V. Innervation of inter-
renal cells. Zeit. Zellforsch., 146: 403-416.
Unsicker, K. (1974) Innervation of adrenal cells in
the lizards Lacerta dugesi and Lacerta pityusensis.
Gen. Comp. Endocrinol., 24: 409-412.

Hadjian, A. J., Ventre, R. and Chambaz, E. M.
(1981) Cholinergic muscarinic receptors in bovine

45

46

730

adrenal cortex. Biochem. Biophys. Res. Comm., 98:
892-900.

Hadjian, A. M., Guidicelli, C. and Chambaz, E. M.
(1982) Cholinergic muscarinic stimulation of ster-
oidogenesis in bovine adrenal cortex fasciculata cell
suspensions. Biochim. Biophys. Acta, 714: 157-163.
Hadjian, A. J., Cultey, M. and Chambaz, E. M.
(1984) Stimulation of phosphatidylinositol turnover
by acetylcholine, angiotensin II and ACTH in
bovine adrenal fasciculata cells. Biochim. Biophys.

47

48

W.N. HoLMgEs, J. CRONSHAW AND J. L. REDONDO

Acta, 804: 427-433.

Kawamura, M., Yonezawa, Y., Tanaka, Y., Imaga-
wa, N., Tomita, C. and Matsuba, M. (1985) Corti-
cogenic effect of acetylcholine in bovine adrenocor-
tical cells. Endocrinol. Japon., 32: 17-20.

Kojima, I., Kojima, K., Shibata, H. and Ogata, E.
(1986) Mechanism of cholinergic stimulation of
aldosterone secretion in bovine zona glomerulosa
cells. Endocrinology, 119: 284-291.

ZOOLOGICAL SCIENCE 7: 731-738 (1990)

© 1990 Zoological Society of Japan

Effects of Androgen on Mouse Seminal Vesicle Epithelial
Cells in Serum-Free Culture

YASUHIRO ToMOOKA!’” Marc Epery>*, KAREN T. MILLs°,

Howarp A. BERN? and JoHN A. McLacuHLan!

‘Developmental Endocrinology and Pharmacology Section, Laboratory of Reproductive
and Developmental Toxicology, National Institute of Environmental Health Sciences,
National Institutes of Health, P. O. Box 12233 Research Triangle Park, North
Carolina 27709, U.S.A. and *RIKEN Tsukuba Life Science Center,
Laboratory of Cell Biology, Tsukuba, Ibaraki 305, Japan,
and *Department of Integrative Biology and Cancer
Research Laboratory, University of Californi,

Berkeley, California 94720, U.S.A

ABSTRACT—The effects of androgen were studied on mouse seminal vesicle epithelial cells in
serum-free primary collagen gel culture. Androgen had no growth-promoting effect on the epithelial

cells when tested in “optimal” serum-free medium.

However, androgen stimulated growth in

serum-free medium supplemented with epidermal growth factor and insulin at lower concentrations
(“suboptimal” medium). Both testosterone and 5a-dihydrotestosterone (DHT) were effective, but
progesterone and estrogens had no significant effect on growth. The effect of DHT (10°-°M) on
androgen receptor content was studied in epithelial cells grown in both the optimal and the suboptimal
serum-free media. The hormone reduced the number of receptors in the cytosol and increased the

number in the nucleus in both media.

INTRODUCTION

Androgen is a major hormone regulating growth
of the male reproductive tract. Organ culture has
been used to study androgen’s effect on these
target tissues in vitro [1-3]; however, the organ
culture system cannot distinguish direct effects on
tissues from total effects including interaction be-
tween epithelial and stromal tissues.

Recently androgen’s growth effect has been
directly assessed in cell culture. Androgen is
reported to stimulate growth of “androgen-
dependent” tumor cell lines [4]. However, the
hormone is not growth-promoting when tested on
normal prostatic epithelial cells in primary cultures
[5, T. Turner et al, personal communication).

Accepted February 8, 1990
Received December 11, 1989

* To whom all correspondence should be addressed.

* Present address: Unité d’Endocrinologie Molécu-
laire, I.N.R.A., 78350 Jouy-en-Josas, France.

These conflicting reports have encouraged us to
reexamine androgen’s mitogenicity on different
target tissues. We have established a serum-free
collagen gel culture system for mouse seminal
vesicle epithelial cells [6]. In the present study we
report the growth effect of androgen and also the
response of the androgen receptor to the hormone.

MATERIALS AND METHODS

Cell separation

We have improved our previous isolation
method of mouse seminal vesicle epithelial cells
[6]. Seminal vesicles were dissected from CD-1
adult male mice (2-4 months old, Charles River),
freed from surrounding tissues and squeezed with
forceps to remove fluid. In order to expose the
lumen to the following enzyme solution, a single
seminal vesicle was cut into 4 pieces by bisection in
both the transverse and longitudinal planes with a

4

ee - =

732 Y. Tomooka, M. Epery et al.

single-edge blade under a dissecting microscope.
In this cutting process, remaining fluid was re-
moved by the blade as much as possible. The
tissues were transferred to a 50-ml test tube and
washed with a culture medium or Hanks’ balanced
salt solution.

The tissues were treated with 0.5% trypsin
(Cooper Biomedical, Malvern, PA) in phosphate-
buffered solution without Cat * and Mg*~ at pH
7.4, and incubated for 1 hr at 4°C. The volume of
the enzyme solution was 10-15 ml for 40-50
glands (enough to immerse tissues). The test tube
was transferred to a water bath at 37°C and
incubated for 45-60 min. After the final incuba-
tion, the test tube was agitated gently by Vortex
for 20sec. Culture medium was added to the
tissues, and the supernatant containing mainly
secretory materials was removed and discarded
after shaking the tube. To remove epithelium, the
tube was again agitated by Vortex at the highest
speed for approximately 20sec. Medium was
added to the tube. After big tissue clumps had
settled, the supernatant, containing mainly small
clumps of epithelial cells, was collected and centri-
fuged. This process of vortexing at the highest
speed and then collecting was repeated 2-3 times.
The resultant epithelial cell pellets were resus-
pended in medium and kept on ice.

The remaining large tissue fragments were
placed in a 100-mm plastic dish and minced with a
single-edge blade. The minced tissues were again
transferred to the tube and agitated by vortexing at
the highest speed for 20 sec. Medium was added to
the test tube, and the supernatant was collected.
The supernatant at this time contained many
stromal fragments in addition to epithelial clumps
(the non-epithelial components were completely
separated in the subsequent gradient centrifuga-
tion). The process (see; lines 15-16) was repeated
2-3 times. All resuspensions of pellets after centri-
fugation of supernatants were combined and
loaded on pre-run Percoll gradient centrifugation
tubes [7]. The epithelial fraction was collected and
centrifuged as described earlier [7].

Cell culture

For growth studies, epithelial cells were cultured
in collagen gel in a 24-multiwell dish as previously

described [6]. Culture medium was a 1:1 mixture
of Ham’s F12 and Dulbecco’s modified Eagle’s
medium supplemented with antibiotics [6]. The
“optimal” serum-free medium contains epidermal
growth factor (EGF, 10 ng/ml. Collaborative Re-
search), transferrin (10 “g/ml, Sigma), insulin (10
pg/ml, Sigma), cholera toxin (10 ng/ml, Sigma)
and cortisol (0.1 “g/ml, Sigma). The “suboptimal”
serum-free medium contains EGF (i ng/ml), in-
sulin (1 «g/ml) and all other factors at the same
concentration. Sex hormones (Sigma) were dis-
solved in absolute ethanol. Cell number was
determined as previously described [7].

For androgen receptor study, isolated epithelial
cells (1.4-2.4 x 10°/dish) were embedded in 10 ml
collagen solution on a 100-mm plastic dish pre-
viously coated with collagen (5 ml). Cells were fed
with 10ml optimal or suboptimal serum-free
medium with or without 5a-dihydrotestosterone
(DHT) (10°°M). Medium was changed on the
3rd day. On the 7th day, medium was removed
and cells were recovered by digestion of the col-
lagen with collagenase (0.025%, wt/vol) (Cooper
Biomedical, Malvern, PA).

Cytosol receptor assays

(CH) 17-hydroxy-17-a-methyl-estra-4,9,11-tri-
en-3-one [(7H) R1881 (methyltrienolone) AS87
ci/mmol] and radio-inert R1881 were obtained
from New England Nuclear (Boston, MA).

All procedures were performed at 4°C as pre-
viously described [8, 9]. Freshly-isolated cells were
homogenized with five 5-sec bursts of a polytron
PT-10-ST (Brinkman Instruments, Westbury,
NY), in 4 vol (vol/wt) buffer A (25 mM Tris-HCl;
1.5mM EDTA, 10 mM thioglycerol, 10% glycerol
and 10mM sodium molybdate, pH7.4). The
homogenate was centrifuged at 13,000Xg for 15
min and the resulting supernatant was again centri-
fuged at 105,000xg for 90min to obtain the
cytosol (supernatant fraction). The nuclear pellet
was assayed for DNA according to the method of
Burton [10].

For the determination of receptor levels in cul-
tured cells [8, 9] cultured cells were first freed from
collagen gels by collagenase (0.025%, wt/vol) dis-
sociation at 30°C for 30min. The cells were
washed twice with Medium 199 and once with

Androgen Effects on Mouse Seminal Vesicle 188

buffer A and then homogenized in | ml buffer A
using three 5-sec bursts of the polytron. This
procedure resulted in the disruption of more than
90% of the cells as observed by phase microscopy.
An aliquot of the homogenate was removed for
DNA assay and the homogenate was centrifuged
as described above to obtain the cytosol.

The labeled ligand for the determination of
androgen receptors was R1881 (see above). In a
standard assay, 0.2-ml aliquots of cytosol (pretre-
ated for 10 min at 0°C with dextrancoated charcoal
in order to remove any free steroid) were incu-
bated with 0.1 ml (°H) R1881 in buffer B (10 mM
Tris-HCl, 1.5 mM EDTA, 10 mM thioglycerol and
10mM sodium molybdate, pH7.4). The final
concentrations of (7H) R1881 were 7 mM [11, 12]
for freshly-isolated cells and 2mM for cultured
cells, respectively. The final saturating concentra-
tions of steroid for cultured cells were determined
by saturation analysis in the presence or absence of
a 200-fold molar excess of unlabeled steroid.
Scatchard analysis of cultured epithelial cells
showed a single class of receptor with a Kd of 6.7+
3.1x107'°M (n=5; data not shown), which is
similar to the Kd of whole tissues [3, 4] and in
agreement with previous data for steroid receptor
assay of cultured cells [8, 9]. Incubations were for
20 hr at 10°C. Time-course studies have shown
that these incubation times represent equilibrium
or steady-state conditions and allow optimal ex-
change of bound hormone in mouse tissues [11,
12]. An excess of triamcinolone acetonide and
diethylstibestrol was always added to the (H)
R1881 solution to saturate progesterone and
estrogen receptors, respectively. Androgen recep-
tors were determined by a dextran-coated charcoal
assay. Unbound steroid was removed by adsorb-
tion by charcoal. Tubes were incubated for 40 min
at 0-4°C with 0.3 ml 0.5% (wt/vol) Norit A-0.05%
(wt/vol) Dextran T-70 (DCC) in buffer B and
centrifuged at 2000 x g for 10 min at 4°C. Aliquots
of the supernatant were taken for scintillation
counting in 8ml Aquaso. Specific binding was
obtained by subtracting nonspecific from total
bound radioactivity.

Nuclear androgen receptor assay

For nuclear receptor determination, cultured

cells were first freed from collagen gels by collage-
nase (0.025%, wt/vol) dissociation at 30°C for 30
min. The cells were washed twice with Medium
199 and once with buffer A then homogenized in 1
ml buffer A using a polytron as already described
[9]. The crude nuclear fraction was prepared by
centrifuging the homogenate at 800 x g for 20 min.
The nuclear pellet was then washed three times in
buffer A containing 1.4 M KCI shaking for 60 min
and centrifuged at 105,000 x g for 30 min to obtain
the supernatant (nuclear extract).

The procedure for the nuclear receptor assay has
been previously described by Brinkmann eft al.
[12]. The nuclear extract was incubated with 2 mM
(7H) R1881 in the presence or absence of a 200-
fold molar excess of R1881 for 20hr at 10°C.
Bound and free hormones were separated by the
dextran-coated charcoal technique described be-
fore.

RESULTS

Cell number was increased 6-8 fold within an 8-
to 12-day culture period in the optimal serum-free
medium. Addition of DHT to the optimal medium
did not result in further increase in cell number
(Fig. 1). However, both DHT and testosterone
(10° to 10 *M) stimulated growth in the sub-
optimal serum-free medium containing EGF and
insulin at lower concentration (Fig. 2). Other sex
hormones were tested in the suboptimal medium.
Estradiol-178 and diethylstilbestrol did not stimu-
late growth when tested in the range 10° '° to 10 *
M (Fig. 3). Progesterone also had no growth-
promoting effect. Estradiol-178, diethylstibestrol
and progesterone inhibited growth at higher con-
centrations.

Next we studied the effects of androgen on its
receptor. The results are summarized in Table 1.
Freshly-isolated epithelial cells had specific bind-
ing sites in the cytosol (775 +106 fmol/mg DNA)
and in the nucleus (38+10 fmol/mg DNA). Cul-
tured epithelial cells in the optimal serum-free
medium had a slightly changed number of specific
binding sites in both cytosol and nucleus. Addition
of DHT at 10-°M to the optimal serum-free
medium caused significant changes in binding sites;
the hormone reduced by nearly one-half the num-

734 Y. Tomooka, M. Epery et al.

90

80

(X 10%)

70

60

20
10 3
-10 me) a 7 = = res
To 9 10 10 10° 10; iliOn ty Gallons 10.

CONCENTRATION (M)

CELL NUMBER

Fic. 1. Cells cultured in optimal serum-free medium (see Materials and Methods) with DHT. To stands for cell ©

number seeded.

Taye ae © i nn ed dtm ° ial ree 0) ema |) mh 1)
CONCENTRATION (M)

CELL NUMBER (x1074)

Fic. 2. Cells cultured in suboptimal serum-free medium (see Materials and Methods) with DHT (@) or testosterone
(x). To stands for cell number seeded.

Androgen Effects on Mouse Seminal Vesicle 135

Ce AO
y
fan)
-
30
oc /\
= ERS
= 20 a7
< 5 eae!
=|
rT
1
O : 6
: .
To 0 OW Cie ecOMceant One 10°" 10° 107° 10"

CONCENTRATION (M)

Fic. 3. Cells cultured in suboptimal serum-free medium with 17-estradiol (@), diethylstibestrol 4 or progesterone
(x). To stands for cell number seeded.

TABLE 1. Androgen receptor levels in freshly-isolated and cultured epithelial cells

Specific Binding Sites (fmol/mg DNA)

Culture Condition

Cytosol ‘by DEL 2 Nucleus ‘by DEL ©
Optimal Medium* 641+ 102 100 71+14 100
Optimal Medium+ DHT (107*® M) 342+ 62 —§3 IW3ae25) +172
Suboptimal Medium? 457+ 48 100 83+11 100
Suboptimal Medium+DHT (107* M) 216+ 29 —47 144+ 16 +73
Epithelial Cells from Intact Mice 775 + 106 — 38+ 10 —

* Serum-free medium supplemented with EGF (10 ng/ml), insulin (10 g/ml), transferring (10 g/ml),
cortisol (0.1 ~g/ml) and cholera toxin (10 ng/ml).
> Serum-free medium supplemented with EGF (1ng/ml), insulin (1 g/ml), transferring (10 g/ml),
cortisol (0.1 ~g/ml) and cholera toxin (10 ng/ml).

—————————————————— a ee

ber of specific binding sites in the cytosol, and it
increased by more than 100% the number of
binding sites in the nucleus. Similar chnges by
DHT at 10~°M were observed in epithelial cells
cultured in the suboptimal serum-free medium;
cytosol binding sites were reduced by nearly one-
half and nucleus binding sites were almost dou-
bled. The amount of DNA was not significantly
changed by DHT in the optimized serum-free
medium, but it was almost doubled by DHT in the
suboptimal medium (data not shown).

DISCUSSION

Castration of male mice causes regression of

seminal vesicles and other reproductive organs,
while administration of androgen to the castrated
mice induces cell proliferation and restores the
organs [13]. Many investigators have attempted to
demonstrate this androgen dependency in tissue
culture. At present, the reported results are
conflicting. In serum-supplemented media, growth
stimulation by androgens has been reported for
human benign prostate hyperplastic cells [14, 15],
rat ventral prostatic epithelial cells [16] and a cell
line of hamster ductus deferens [4], but no or weak
effects have been reported for canine prostatic
epithelial cells [17] and a human prostatic carcino-
ma cell line [18]. Horoswewicz et al. [19] observed
both a stimulating effect in stripped serum-

736

containing medium and no effect in intact serum-
containing medium on a human prostatic carcino-
ma cell line. Most recently androgens have been
shown to be active in the presence of neurotrans-
mitters in growth of rat seminal vesicle epithelial
cells [20]. These inconsistent observations on
androgen’s effect may stem from differences in the
quality and quantity of supplemented serum in
culture media, in addition to other factors.

In serum-free culture media, on the other hand,
a growth-stimulating effect of androgen has not
been observed [5, 17, 18]. To explain the discre-
pancy of androgen’s mitogenicity between in vivo
and in vitro, and the conflicting results observed,
new growth factor(s) or hormone(s) have been
proposed as possible mediators of androgen’s
effect on growth [5, 7]. This argument is similar to
that of estrogen’s mitogenicity: estrogen’s growth
stimulation is shown mainly on cell lines in serum-
containing media [21-23], but not observed on
primary culture in serum-free media [24-26]. The
intervention of ther factors [27] or another
mechanism [28] has been proposed to explain the
conflicting results on estrogen’s action on growth
[see 29].

In order to reexamine androgen’s effect on
growth, we have established a method for separat-
ing seminal vesicle epithelial cells and have
adapted the serum-free collagen gel system [30] to
the culture of these epithelial cells [6]. In our
optimal serum-free medium, addition of androgen
did not result in further increase in cell number.
However, androgens stimulated growth when the
concentrations of EGF and insulin were reduced.’
This growth-promoting effect is androgen-specific;
17f-estradiol, diethylstilbestrol and progesterone
have no stimulating effect. Seminal vesicle epithe-
lial cells in the optimal medium may not be grow-
ing at the maximal rate, because further growth
can be achieved by addition of nutritional factors
such as bovine serum albumin [6]. These facts
indicate that the mitogenic message(s) evoked by
androgen and EGF-insulin might be related and
may be working at associated intracellular sites to
trigger DNA synthesis. Another explanation of
this inability of androgen to stimulate further
growth with higher concentrations of EGF and/or
insulin is that androgen stimulates production of a

Y. Tomooka, M. Epery et al.

growth factor (such as EGF or insulin-like growth
factor) which ultimately stimulates growth of the
seminal vesicle epithelial cells. The existence of
such mechanisms in epithelial cells of sex hormone
target tissues is supported by several recent stu-
dies. Estrogen- or androgen-dependent tumors
secrete insulin-like growth factor-1 [31, 32]. Mam-
mary epithelial cells contain mRNA for EGF [33].
Prolactin and progesterone can substitute for EGF
which is a strong mitogen on mammary epithelial
cells [25]. Mouse uterine epithelial cells contain
immunoreactive EGF-like materials [34]. All
these studies raise the likelihood of an association
between androgen or estrogen dependency and the
production of EGF and other growth factors in the
target tissues. However, the existence of such an
autocrine or paracrine system in the seminal vesi-
cle is not yet established.

It is believed that androgen’s action is mediated
by its receptor [35]. However, the involvement of
receptors in growth stimulation by androgen is not
well studied, especially in vitro. Cultivation of the
epithelial cells in our optimal medium slightly
reduced the number of receptors in the cytosol and °
in the nucleus. Addition of DHT to both the
optimal and the suboptimal media reduced cytoso-
lic receptors (53% and 47%, respectively), and
increased nuclear receptors (172% and 73%, re-
spectively). This indicates the functional integrity
of the androgen receptor in both optimal and
suboptimal conditions; however, it is consistent
with the biological effect of DHT in culture only in
suboptimal medium. We have shown that estrogen
receptors are maintained in mammary [8, 9] or
vaginal epithelial cells [36] grown inside collagen
gels and that these receptors are functional:
estrogens are able to induce nuclear accumulation
of the receptor and synthesis of progesterone
receptors, although they are not mitogenically
active in this in vitro system. The presence of the
receptor for androgen may not be required for
androgen to exert its growth effect. In any case,
the absence of an effect cannot be attributed to the
absence of receptors or to inability of receptor
function in this culture system. Cytosolic depletion
and nuclear accumulation of androgen receptors in
the ventral prostate occurs in 1-day-castrated male
rats after testosterone administration [37]. In

Androgen Effects on Mouse Seminal Vesicle 737

serum-free medium, androgen receptors were
maintained and reached levels only slightly lower
than those observed in 7-day-castrated animals
[38] where the highest levels of androgen receptor
are found, probably due to the clearance of en-
dogenous androgen after castration. The mainte-
nance of high levels of receptors in culture may
indicate either that the receptor is a constitutive
protein of the seminal vesicle, which does not
require the presence of androgen to be synthesized
(although the carry-over of hormones in culture
cannot be eliminated since noncastrated animals
were used) or that hormones present in the
medium (EGF, insulin) support the synthesis of
androgen receptors.

REFERENCES

1 Buchanan, L. J. and Riches, A.C. (1986) Prolifera-
tive responses of rat ventral prostate: effects of
variations in organ culture media and methodology.
Prostate, 8: 63-74.

2 Lasnitzki, I. (1974) The prostate gland in organ .

culture. In “Male Accessory Sex Organs”. Ed. by D.
Brandes, Academic Press, New York, pp. 348-382.

3. Franks, L. M. (1980) Primary cultures of human
prostate. Methods in Cell Biol., 21B: 153-169.

4 Syms, A. J., Norris, J. S. and Smith, R. G. (1983)
Proliferation of a highly androgen-sensitive ductus
deferens cell line (DDT1MF-2) is regulated by glu-
cocorticoids and modulated by growth on collagen.
In Vitro, 19: 929-936.

5 McKeehan, W. L., Adams, P. S. and Rosser, M. P.
(1984) Direct mitogenic effects of insulin, epidermal
growth factor, glucocorticoid, cholera toxin, un-
known pituitary factors and possibly prolactin, but
not androgen, on normal rat prostate epithelial cells
in serum-free, primary cell culture. Cancer Res., 44:
1998-2010.

6 Tomooka, Y., Harris, S. E. and McLachlan, J. A.
(1985) Growth of seminal vesicle epithelial cells in
serum-free collagen gel culture. In Vitro, 21: 237-
244.

7 Imagawa, W., Tomooka, Y., Yang, J., Guzman,
R., Richards, J. and Nandi, S. (1984) Isolation and
serum-free cultivation of mammary epithelial cells
within a collagen gel matrix. In “Cell Culture
Methods for Molecular and Cell Biology”. Vol. 2,
Ed. by D. W. Barnes, D. A. Sirbasku and G. H.
Sato, Alan R Liss, New York, pp. 127-141.

8 Edery, M., McGrath, M., Larson, L. and Nandi, S.
(1984) Correlation between in vitro growth and
regulation of estrogen and progesterone receptors in

10

il

12

13

14

15

16

17

18

19

20

rat mammary epithelial cells. Endocrinology, 115:
1691-1697.

Edery, M., Imagawa, W., Larson, L. and Nandi, S.
(1985) Regulation of estrogen and progesterone
receptor levels in mouse mammary epithelial cells
grown in serum-free collagen gel cultures. Endocri-
nology, 116: 105-112.

Burton, K. (1956) A study of the conditions and
mechanism of the diphenylamine resction for the
colorimetic estimation of deoxyribonucleic acid.
Biodhem J., 62: 315-323.

Schein, L., Donovan, M. P. and Thomas, J. A.
(1978) Characterization of cytoplasmic binding of
dihydrotestosterone by the prostate gland, seminal
vesicles, kidney and liver of the mouse. Toxicol.
appl. Pharmacol., 44: 147-158.

Brinkmann, A. O., Lindh, L. M., Breedveld, D. I.,
Mulder, E. and Van der Holen, H. J. (1983) Cyp-
rosterone acetate prevents translocation of the
androgen receptor in the rat prostate. Mol. Cell.
endocrinol., 32: 117-129.

Tuohimaa, P. and Niemi, M. (1974) Cell renewal
and mitogenic activity of testosterone in male sex
accessory glands. In “Male Accessory Sex Organs.”
Ed. by D. Brandes, Academic Press, New York, pp.
329-343.

Syms, A. J., Harper, M. E., Battersby, S. and
Griffiths, K. (1982) Proliferation of human prostatic
epithelial cells in culture: aspects of identification. J.
Urol., 127: 561-567.

Syms, A. J., Harper, M. E. and Griffiths, K. (1985)
The effect of prolactin on human BPH epithelial cell
proliferation. Prostate, 6: 145-153.

O’Connor, T. and Sinha, D. K. (1985) Character-
ization of rat ventral prostatic epithelial cells in
collagen gel culture. Prostate, 7: 305-319.
Chevalier, S., Bleau, G., Roberts, K. D. and Cha-
polelaine, A. (1984) Nonsteroidal serum factors
involved in the regulation of the proliferation of
canine prostatic epithelial cells in culture. Prostate,
5: 503-512.

Hasenson, M., Hartley-Asp, B., Kihlfors, C., Lun-
din, A., Gustafsson, J-A. and Pousette, A. (1985)
Effect of hormones on growth and ATP content of a
human prostatic carcinoma cell line, LNCaP-r. Pros-
tate, 7: 183-194.

Horoszewicz, J. S., Leong, S. S., Kawinski, E.,
Karr, J. P., Rosenthal, H., Chu, T. M., Mirand, E.
A. and Murphy, G. P. (1983) LNCaP model of
human prostatic carcinoma. Cancer Res., 43: 1809-
1817.

Kinghorn, E. M., Bate, A. S. and Higgins, S. J.
(1987) Growth of rat seminal vesicle epithelial cells
in culture: neurotransmitters are required for
androgen-regulated synthesis of tissue-specific secre-
tory proteins. Endocrinology, 121: 1678-1688.

21

22

23

24

25

26

Z)}

28

29

30

738

Leung, B. S., Qureshi, S. and Leung, J. S. (1982)
Response to estrogen by the human mammay carci-
noma cell line CAMA-1. Cancer Res., 42: 5060-
5066.

Chalbos, D., Vignon, F., Keydar, I. and Rochefort,
H. (1982) Estrogens stimulate cell proliferation ana
induce secretory proteins in a human brease cancer
cell line (T47D). Endocrinology, 55: 276-283.
Aitken, S. C. and Lippman, M. E. (1980) Hormon-
al regulation of net DNA synthesis in human breast
cancer cells in tissue culture. In “Control Mechan-
isms in Animal Cells.” Ed. by L. Jimenes de Asus,
R. Levi-Montalcini, R. Shields and S. I[acoballi,
Raven Press, New York, pp. 133-156.

Yang, J., Larson, L., Flynn, D., Elias, J. and Nandi,
S. (1982) Serum-free primary culture of human
normal mammary epithelial cells in collagen gel
matrix. Cell Biol. Int. Reprots, 6: 969-975.
Imagawa, W., Tomooka, Y., Hamamoto, S. and
Nandi, S. (1985) Stimulation of mammary epithelial
cell growth in vitro: interaction of epidermal growth
factor and mammogenic hormones. Endocrinology,
116: 1514-1524.

Iguchi, T., Uchima, F.-D. A., Ostrander, P. L. and
Bern, H. A. (1983) Growth of normal mouse
vaginal epithelial cells in and on collagen gels. Proc.
Natl. Acad. Sci. USA, 80: 3743-3747.

Sirbasku, D. A. and Benson, R. H. (1979)
Estrogen-inducible growth factors that may act at
mediators (estromedins) of estrogen-promoted
tumor cell growth. In “Hormones and Cell Culture”.
Ed. by G. H. Sato and R. Ross, Cold Spring Harbor
Laboratories, New York, pp. 477-497.
Sonnenshein, C. and Soto, A. M. (1980) The
mechanism of estrogen action: the old and new
paradigm. In “Estrogens in the Environment”. Ed.
by J. A. McLachlan Elsevier, New York, pp. 169-
195%

Bern, H. A. (1990) Is estrogen a cellular signal for
femal genital tract epithelia? In “Cellular Signals for
Uterine Function”. Ed. by L. Lawia, Alan Liss,
New York. (In press)

Imagawa, W., Tomooka, Y. and Nandi, S. (1982)

31

32

33

34

S)

36

37)

38

Y. Tomooka, M. EpeEry et al.

Serum-free growth of normal and tumor mouse
mammary epithelial cells in primary culture. Proc.
Natl. Acad. Sci. USA, 79: 4047-4077.

Huff, K. K., Kaufman, D., Gabbay, K. H.,
Spencer, E. M., Lippman, M. E. and Dickson, R.
B. (1986) Secretion of an insulin-like growth factor-
1-related protein by human breast cancer cells.
Cancer Res., 46: 4613-4619.

Dikson, R. B., Huff, K. K., Spencer, E. M. and
Lippman, M. E. (1985) Induction of epidermal
growth factor-related polypeptides by 17/-estradiol
in MCF-7 human breast cancer cells. Endocrinolo-
gy, 118: 138-142.

Rall, L. B., Scott, J. and Bell, G. I. (1985) Mouse
prepro-epidermal growth factor synthesis by the
kidney and other tissues. Nature, 313: 228-231.
DiAugustine, R. P., Petrusz, P., Bell, G. I., Brown,
C. F., Korach, K. S., McLachlan, J. A.and Teng, C.
T. (1988) Influence of estrogens on mouse uterine
epidermal growth factor precursor protein and mes-
senger ribonucleic acid. Endcrinology, 122: 2355-
2363.

Liao, S., Fang, S., Tymoczko, J. L. and Liang, T.
(1974) Androgen receptors, antiandrogens, and
uptake and retention of androgens in male sex
accessory organs. In “Male Accessory Sex Organs”.
Ed. by D. Brandes, Academic Press, New York, pp.
238-262.

Uchima, F.-D. A., Edery, M., Iguchi, T., Larson,
L. and Bern, H. A. (1987) Growth of mouse vaginal
epithelial cells in culture: functional integrity of the
estrogen receptor system and failure of estrogen to
induce proliferation. Cancer Letters, 35: 227-235.
Steinsapir, J., Evans, A. C. Jr., Bryhan, M. and
Muldoon, T. G. (1985) Androgen receptor dyna-
mics in the rat ventral prostate. Biochim. Biophys.
Acta, 842: 1-11.

Turner, T., Edery, M., Mills, K. T. and Bern, H. A.
(1989) Influence of neonatal diethylstibestrol treat-
ment on androgen and estrogen receptor levels in
the mouse anterior prostate, ventral prostate and
seminal vesicle. J. Steroid Biochem., 32: 559-564.

ZOOLOGICAL SCIENCE 7: 739-746 (1990)

© 1990 Zoological Society of Japan

Inner Structures of the Cerebral Vesicle in the Ascidian
Larva, Styela plicata: A SEM Study

HISASHI OHTSUKI

Biological Institute, Faculty of Education,
Oita University, Oita 870-11, Japan

ABSTRACT — The inner structures of the cerebral vesicle in larvae of the ascidian, Styela plicata, were
examined by scanning electron microscopy. The ocellus of this species is situated in the postero-dorsal
wall of the cerebral vesicle. The ocellus is as degenerate as that of the other species of Styela which
consists of only a single cell containing a small pigment body. The cell body of the statocyte was found
to be attached to the ventral floor of the cerebral vesicle by two junctions. The cell body is hollowed
between the junctions, forming an inverted pigment cup. Four types of protuberance were recognized
in the cerebral vesicle. Three of these project from the ventral floor under the cell body of the statocyte.
The distribution of these three types is restricted to the floor on the right side of the junctions of the

statocyte.

The fourth type of protuberance was always found as a pair on the left part of the

postero-dorsal wall of the cerebral vesicle. Probable functions of these protuberances are discussed.

INTRODUCTION

Three types of sensory receptor (statocyte, ocel-
lus and hydrostatic pressure receptor) have been
reported in the cerebral vesicle of ascidian larvae
after observations by transmission electron micros-
copy (TEM) [1-12].

The larva Styela plicata has a large statocyte,
which protrudes from the mid-ventral wall of the
cerebral vesicle and a considerably degenerate,
small ocellus, which is situated in the postero-
dorsal wall of the vesicle [13] like that of the other
species of Styela [14-19]. Torrence [’80, Am.
Zool., 20, 886. Abstract] has reported that the
statocyte of S. plicata is a photolith, an organ
sensitive to both light and gravity, similar to that
found in the larvae of species belonging to the
Botryllinae (colonial species of Styelidae). He
reported that the statocyte in S. plicata forms an
inverted pigment cup, which is invaded by globular
protuberances arising from the ventral wall of the
cerebral vesicle. The present author has confirmed
his observation by light microscopy with serial
paraffin sections [20]. Torrence [11] has also

Accepted October 2, 1989
Received August 18, 1989

reported that in the cerebral vesicle of S. plicata
two sensory endings, from which many narrow
processes extend against the surface of the cell
body of the statocyte, protrode from the left floor
to the junction of the statocyte. He suggested that
the sensory endings function to detect the move-
ment of the statocyte. According to Reverberi, the
hydrostatic pressure organs in Styela sp. are
rudimentary [9] or absent [10].

Scanning electron microscopy (SEM) is an ex-
cellent means of studying three-dimensional fine
structure. However, there has apparently been no
SEM study on the structures of the sensory organs
in the cerebral vesicle of ascidian larvae. The
present paper describes the structures within the
cerebral vesicle of S. plicata shown by SEM and
their probable functions are discussed.

MATERIALS AND METHODS

Fertilized eggs of Styela plicata (Lesueur) were
obtained by natural shedding of gametes from
mature adults (4-7 individuals) in a plastic dish
containing aerated natural seawater. The eggs
were washed once with filtered seawater by settling
and decantation. Hatching occurred about 12 hr
after shedding at 20°C. Free-swimming larvae

740 H. OHTSUKI

were fixed in 2% glutaraldehyde in 77% seawater
[21] for 20-24 hr. The fixed larvae were soaked
briefly in distrilled water, dehydrated in a graded
ethanol series and stored in 80% ethanol below
4°C until used.

The stored materials were dehydrated and the
ethanol-infiltrated larvae were cryofractured at
liquid nitrogen temperature. The larvae, which
were cut through their cerebral vesicle, were col-
lected under a stereoscopic microscope and rinsed
with M/30 phosphate buffer (pH 7.4) to remove
precipitated debris in the cerebral vesicle with
ethanol. The rinsed larvae were dehydrated again
and dried with a critical-point dryer (Hitachi HPC-
2) using CO, with isoamyl acetate as the transition
fluid. The dried materials were mounted on elec-
tron-conductive tape (Sumitomo-3M) attached to
an aluminium stub. They were coated by plati-
num-palladium usting an ion sputterer (Hitachi, E
102). All dried and coated materials were ex-
amined with a scanning electron microscope
(Hitachi S-800).

RESULTS

The cerebral vesicle in the swimming larvae of
Styela plicata was situated in the right side of the
dorsal half of the trunk. Figures 1 and 2 show the
structures in the right, posterior half of the cere-
bral vesicle exposed by a sagittal cut through the
right side to the median plane of the larval trunk of
S. plicata. An ocellar pigment body was seen in
the postero-dorsal wall of the cerebral vesicle (Fig.
2). It consisted of a mass of small grains. Neither
lens cells nor retinal cells were found around the
pigment body.

The cell body of the statocyte was attached to
the ventral wall of the cerebral vesicle, where it
occupied a large space in the lumen (Figs. 1-3).
The cell body of the statocyte was joined to the
floor of the cerebral vesicle by two junctions (a and
b in Figs. 3-6). When the cell body of the stato-
cyte was accidentally lost during the preparation
procedure, two scars were made by the cell body
being torn off at its junctions (a and b in Figs. 4,
5). The two junctions of the statocyte were
confirmed by an oblique section, which fortunately
cut the statocyte near the junctions (Fig. 6). The

anterior junction looked like a conical stalk (2 ~m
height, 2 um base diameter), which extended some
fibrous projections from its base to the surface of
the floor (a in Figs. 4-6 and 9). The posterior
junction (0.7 ~m height) was shorter than the
anterior one (b in Fig. 6).

Figure 6 also shows that the cell body of the
statocyte developed an invagination to form an
inverted pigment cup at its ventral surface between
the two junctions.

Many protuberances projected from the ventral
floor under the cell body of the statocyte (Figs. 2,
3). The protuberances were distributed on the
ventral floor in the right-hand region of the junc-
tions of the statocyte (Figs. 4, 5). These protuber-
ances could be classified into three types (1-3 in
Figs. 4 and 5); 1: globular (0.7—2.0 ~m diameter),
2: tapered band (0.15-0.4 ~m proximal width,
0.6-1.7 um length), and 3: small tubular (0.13 ~m
diameter, 0.1-0.5 um length).

The protuberances of type 1 were arranged in a
cluster at the anterior position right of the anterior
junction of the statocyte. Those of type 2 were
distributed in a crescent-shaped area just right of .
the cluster of type 1. The type 3 protuberances
were mostly limited in their distribution around
the anterior and posterior junction of the stato-
cyte. A network structure was seen beneath the
ventral floor from which the protuberances ex-
tended (Figs. 2, 3).

The fourth type of protuberance (2 um dia-
meter) was always present in pairs on the left part
of the postero-dorsal wall of the cerebral vesicle (4
in Figs. 6-8). They extended some fibrous projec-
tions to the inner surface of the cerebral wall and
were loosely surrounded by small protuberances
(Fig. 8) like those of type 3 on the ventral floor.

DISCUSSION

It has been generally accepted that the statocyte
of ascidian larvae is unicellular and that the cell
body of the statocyte is joined by its foot-piece in
the ventral wall of the cerebral vesicle by only a
single narrow junction [22]. This paper is the first
report of a statocyte having two junctions. In the
developing embryos of Styela plicata, the pigment
mass of the statocyte was seen to consist of two

Cerebral Vesicle of Ascidian Larva 741

Fic. 1.
its median plane. Scale: 25 um.

Scanning electron micrograph of the larval trunk of Styela plicata, cut longitudinally through the right side to

Fic. 2. Enlargement of Fig. 1, showing inner structures of the cerebral vesicle viewed from the left, slightly
postero-dersal direction. op: ocellar pigment, pt: protuberances on the ventral floor, st: statocyte. Scale: 5 um.

pigment blocks joined together from an early
period of their pigmentation. The pigment blocks
retain their conection during their growth period in
subsequent development up to the swimming tad-
pole stage [20]. These observations suggest the
possibility that the statocyte of the species consists
of two cells. Another possibility is that one of the

two junctions may be a sensory ending as reported
by Torrence, since sensors for perception of stato-
cyte movement [11] cannot be discarded. Howev-
er, unlike his report, no narrow prosesses ex-
tended from the surface of both junctions, though
many protuberances on the floor (type 3) encircled
the base of each junction (Figs. 4, 5).

742 H. OnTsukI

ca

PG

Fic. 3. Scanning electron micrograph of the cerebral vesicle, its left, dorsal wall having been cut off. Arrows (a) and
(b) show the anterior and posterior junctions of the cell body of the statocyte, respectively. Scale: 5 um.

Fic. 4. Scanning electron micrograph of the cerebral vesicle, its dorsal hemisphere having been cut off. Arrows (a)
and (b) show scars of the anterior and posterior junctions of the statocyte at which its cell body was torn off. 1-3:
type 1- type 3 protuberances on the ventral floor of the cerebral vesicle. Scale: 5 um.

Cerebral Vesicle of Ascidian Larva 743

show scars of anterior (a) and posterior (b) junctions at which the cell body of the statocyte was torn off. Three
types of protuberance (1-3) kare shown to the right of the junctions of the statocyte. Scale: 5 um.

Fic. 6. Scanning electron micrograph of the cerebral vesicle, its right anterior half having been cut off near the
anterior junction (a) and through the poster ior one (b). A pair of protuberances (4) are present on the left part of
the postero-dorsal wall of the vesicle. Scale: 5 um.

744 H. OnTSUKI

Fic. 7. Scanning electron micrograph of the cerebral vesicle, its right half having been cut off. a: anterior junction at

which the cell body of the statocyte was torn off. 4: a pair of protuberances on the left part of the postero-dorsal
wall of the vesicle. Scale: 5 «7m.

Fic. 8. Enlargement of 4) in Fig. 7. Scale: 1 um.
Fic. 9. Enlargement of a) in Fig. 7. Scale: 1 um.

Cerebral Vesicle of Ascidian Larva 745

It was confirmed in this work that the ocellus of
S. plicata is degenerate and consists of only a cell
containing a tiny pigment body [13, 20] similar to
that of the other species of the same genus [14-19].
In some other genera of Styelidae (Botryllinae),
the larvae completely lose their ocellus. In those
species, the statocyte changes to a “photolith”, an
organ sensitive to both light and gravity [14, 22—
26]. The statocyte of S. plicata forms a hollow cup
like the photolith (Fig. 6). In the photolith and the
typical ocellus (consisting of lens cells, retinal cells
and a pigment cell), light-sensitive cells project
their fibrous endings into the pigment cup.

Globular protuberances extending from the ven-
tral floor of the cerebral vesicle into the pigment
cup of the statocyte have been shown in serial
paraffin sections of the larvae of S. plicata [20].
The globular-type protuberances (type 1) in S.
plicata are different from the sensory endings of
the photoreceptive cells in either the photolith [12]
or the typical ocellus [1, 2, 4, 6, 7] in that the type 1
protuberances lack fine processes at their apical
surface. In the photolith of Stolonica socialis [14]
and Botryllus schlosseri [25], photosensitive pro-
cesses arise from the posterior wall of the cerebral
vesicle where the retinal cells in the ocellus of
Enterogona and Pyuridae are located. In contrast,
the type 1 protuberances in S. plicata project from
slightly anterior to the middle position of the
ventral wall of the cerebral vesicle (Figs. 2-5). It
remains unclear if the type 1 protuberances are
really photosensitive.

The other types of protuberance (type 2-4) are
newly found structures in the cerebral vesicle of S.
plicata. The type 2 protuberances are arranged
like many rows of fences making a right angle with
the stream of fluid which may occur when the
statocyte is bent at the two junctions by the
rotation of the larva. It is possible that they play a
role in the perception of the stream on the ventral
floor of the cerebral vesicle.

The type 3 protuberances encircle the bases of
both junctions of the statocyte. They may be
identical with the fine processes protruding from
dendrites which, as Torrence suggested, are re-
sponsible for sensing movements of the statocyte
[11]. The present author agrees with him on the
function of the type 3 protuberances, although

unlike his report, they extend directly from the
ventral floor of the cerebral vesicle.

The fourth type of protuberance (type 4) pro-
jects from the left, postero-dorsal wall of the
cerebral vesicle in pairs (4 in Figs. 6-8). At an
identical position in the cerebral vesicle, the larvae
of Ciona intestinalis have a cluster of globular
structures. These were first reported by Dilly [5] as
a second type of photoreceptor in ascidian larvae.
Eakin and Kuda [6] and Reverberi [9, 10] consi-
dered them to be sensitive to hydrostatic pressure,
because their structures closely resemble those of
the globular body of the cornet cells in other
Chordates. According to Svane [8] and Reverberi
[9], such globular structures are found at the dorsal
wall of the cerebral vesicle in all species of Phle-
bobranchia (Enterogona). The latter author also
found the globular structures at an auxiliary brain
vesicle in two species of Pyuridae (Stolidobran-
chia, Pleurogona), Pyura tessellata and Boltenia
echinata. He recognized the structural resembl-
ance between the globular structures in the cere-
bral vesicle of the ascidian larvae and those of
coronet cells, but he doubted their function as a
pressure receptor.

The protuberances (type 4) in S. plicata (Styeli-
dae, Stolidobranchia) and the globular structures
in the other ascidian larvae are almost identical in
size (2 um in diameter), while the former is not so
globular and does not have a stalk like the latter.
The type 4 protuberances may thus be rudimentary
structures of the globular bodies found in other
species [9].

ACKNOWLEDGMENTS

The author is grateful to Dr. M. Yoshida, Professor of
Oita University, for his kind advice and encouragement
throughout the course of this work.

REFERENCES

1 Barnes, S. N. (1971) Fine structure of the photore-
ceptor and cerebral ganglion of the tadpole larva of
Amaroucium constellatum (Verrill). (Subphylum:
Urochordata; Class: Ascidiacea). Z. Zellforsch..,
117: 1-16.

2 Dzilly, N. (1961) Electron microscope observations
of the receptors in the sensory vesicle of the ascidian
tadpole. Nature, 191: 786-787.

10

11

12

13

14

iS

746

Dilly, N. (1962) Studies on the receptors in the
cerebral vesicle of the ascidian tadpole. 1. The
otolith. Quart. J. Micr. Sci., 103: 393-398.

Dilly, N. (1964) Studies on the receptors in the
cerebral vesicle of the ascidian tadpole. 2. The
ocellus. Quart. J. Micr. Sci., 105: 13-20.

Dilly, N. (1969) Studies on the receptors in Ciona
intestinalis. III. A second type of photoreceptor in
the tadpole larva of Ciona intestinalis. Z. Zell-
forsch., 96: 63 —65.

Eakin, R. M. and Kuda, A. (1971) Ultrastructure of
sensory receptors in ascidian tadpoles. Z. Zell-
forsch., 112: 287-312.

Gorman, A. L. F., McReynolds, J. S. and Barnes,
S. N. (1971) Photoreceptors in primitive chordates:
Fine structure, hyperpolarizing receptor potentials,
and evolution. Science, 172: 1052-1054.

Svane, I. (1982) Possible ascidian counterpart to the
vertebrate saccus vasculosus with reference to Pyura
tessellata (Forbes) and Boltenia echinata (L.). Acta
Zoo. Stockh., 63: 85-89.

Reverberi, G. (1975) On the “third” sensory recep-
tor of the ascidian tadpoles. Rend. Acc. Naz. Lincei,
58: 945-948.

Reverberi, G. (1979) On the third sensorial organ of
ascidian tadpoles. Acta Embryol. Exp., 1: 91-99.
Torrence, S. A. (1986) Sensory endings of the
ascidian static organ (Chordata, Ascidiacea).
Zoomorph., 106: 61-66.

Vorontsova, M. N. and Malakhov, V. V. (1984)
Anatomy and fine structure of larvae in Cnemido-
carpa finmarkiensis (Ascidia). 2. Sense organs.
Zool. Zh., 63: 1036-1045.

Hirai, E. (1958) Morphological observations of
ascidian larvae. 4. Styela plicata (Lesueur). Bull.
Mar. Biol. St. Asamushi, 9: 21-22.

Berrill, N. J. (1948) The nature of the ascidian
tadpole, with reference to Boltenia echinata. J.
Morphol., 82: 269-285.

Grave, C. (1944) The larva of Styela (Cynthia)

16

17

18

19

20

21

Maps

73)

24

25

26

H. OntTsukI

partita: Structure, activities and duration of life. J.
Morphol., 75: 173-191.

Wallace, H. (1961) The breeding and development
of Styela mammiculata Carlisle. J. Mar. Biol. Assoc.
U.K., 41: 187-190.

Whittaker, J. R. (1966) An analysis of melanogene-
sis in differentiating pigment cells of ascidian
embryos. Develop. Biol., 14: 1-39.

Whittaker, J. R. (1973) Tyrosinase in the presump-
tive pigment cells of ascidian embryos: Tyrosine
accessibility may initiate melanin synthesis. De-
velop. Biol., 30: 441-454.

Litzen, J. (1960) The reproductive cycle and larval
anatomy of the ascidian Styela rustica (L.). Vidensk.
Medd. Dansk. Naturh. Foren. Kbb., 123: 227-235.
Ohtsuki, H. (1990) Statocyte and ocellar pigment
cell in embryos and larvae of the ascidian, Styela
plicata (Lesueur). Develop. Growth Differ. 32: 85-
90.

Tegner, M. J. and Epel, D. (1976) Scanning elec-
tron microscope studies of sea urchin fertilization. I.
Eggs with vitelline layers. J. Exp. Zool., 197: 31-58.
Berrill, N. J. (1975) Chordata: Tunicata. In “Repro-
duction of Marine Invertebrate II”. Ed. by A. C.
Giest and J. S. Pease, Academic Press, New York,
pp. 241-282. .
Berrill, N. J. (1949) The gonads, larvae, and bud-
ding of the polystyelid ascidians Stolonica and Dis-
tomus. J. Mar. Biol. Assoc. U.K., 28: 633-650.
Abbott, D. P. (1955) Larval structure and activity in
the ascidian Metandrocarpa taylori. J. Morphol., 97:
569-593. |

Grave, C. and Woodbridge, H. (1924) Botryllus
schlosseri (Pallas): The behavior and morphology of
the free-swimming larva. J. Molphol. Physiol., 39:
207-247.

Grave, C. and Riley, G. (1935) Development of the
sense organs of the larva of Botryllus schlosseri. J.
Morphol., 57: 185-211.

ZOOLOGICAL SCIENCE 7: 747-753 (1990)

© 1990 Zoological Society of Japan

Immunohistochemical Demonstration of S-100 Protein
in the Chick Non-Nervous Tissues

YaAsuRO ATOoJI, KouyI TAKAYANAGI, YOSHITAKA SUZUKI

and MaKotTo SuGIMURA!

Department of Veterinary Anatomy, Faculty of Agriculture, Gifu University,
Gifu 501-11, and ‘Department of Veterinary Anatomy, Faculty of
Veterinary Medicine, Hokkaido University,

Sapporo 060, Japan

ABSTRACT— Distribution of S-100 protein in the chick non-nervous tissues was investigated by
immunohistochemical method using anti-bovine S-100 protein serum. S-100 protein immunoreactivity
was detected in stellate cells of the pituitary gland, insulin cells and somatostatin cells of the pancreatic
islet, epithelial cells of the proventriculus, and epithelial cells of the distal and the collecting tubules in
the kidney. The Sertoli cells and oocytes also contained S-100 protein. These findings indicate that the
occurrence and distribution of S-100 protein immunoreactive cells of the chick is less numerous than that

of mammals.

INTRODUCTION

S-100 protein was first extracted from vertebrate
brains and peripheral nervous system and was
initially considered to be a unique Ca**-binding
protein in the nervous system [1, 2]. In mammals,
S-100 protein is contained in glial cells of the
central nervous system [3] and the Schwann cells of
the peripheral nervous system [4]. On the other
hand, in invertebrates, it occurrs both in neuronal
and glial cells [5, 6]. However, immunohistoche-
mical studies have recently reveales S-100 protein
in many cell types on non-nervous tissues of va-
rious mammalian species [7-9]. In avian species,
there are some reporte of S-100 protein im-
munoreactive cells in the nervous system [10, 11]
and in the non-nervous organs: ovary and hypoph-
ysis of the quail [12, 13]. The exact function of
S-100 protein in neuronal and non-neuronal cells
still remains obscure, although some roles have
been suggested [7]. Immunohistochemistry of S-
100 protein has also been applied to pathological
analysis [14, 15]. In the present study, we describe
the presence and distribution of S-100 protein

Accepted October 18, 1989
Received June 6, 1989

immunoreactive cells in the chick non-nervous
tissues.

MATERIALS AND METHODS

Six adult chicks (four White Leghorn and two
Rhode Island Red) and two chickens (White
Leghorn, 70—80 days old) were used in this study.
After anesthesia with sodium pentobarbital (25
mg/kg), various organs were removed: comb, ton-
gue, eye, eyelid, lacrimal gland, skin, esophagus,
crop, proventriculus, gizzard, duodenum, ju-
junum, ileum, Meckel’s diverticulum, caecum, col-
on, cloaca, liver, gall bladder, bile duct, pancreas,
pancreatic duct, kidney, ureter, testis, epididymis,
ductus deferens, ovary, oviduct, trachea, bonchus,
lung, heart, pituitary gland, pineal gland, thyroid
gland, parathyroid gland, ultimobrachial body,
adrenal gland, thymus, spleen, bursa of Fabricius,
and-uropygial gland. Blocks were fixed in Bouin’s
solution, embedded in paraffin and cut serially at
3-5 um in thickness.

Sections were stained with hematoxylin and
eosin, and were immunostained with anti-S-100
protein serum. Immunohistochemical procedure
of S-100 protein was carried out by the peroxidase-
antiperoxidase (PAP) method [16]. Sections were

748 Y. Aton, K. TAKAYANAGI et al.

preincubated with 1% normal goat serum and then
incubated with polyclonal anti-bovine S-100 pro-
tein rabbit serum (Dakopatts and Advance,
1:500) overnight at 4°C. After rinsing in 0.01 M
phosphate-buffered saline (PBS), they were tre-
ated with anti-rabbit lgG goat serum (Tago Inc.,
1: 100) followed by incubation with PAP complex
(Cappel Inc., 1: 500) for 30 min at room tempera-
ture. Reaction product was visualized by incuba-
tion with PBS containing 3,3’-diaminobenzidine
(25 mg/100 ml) and 0.003% H2O> for 10 min. Sec-
tions were slightly stained with hematoxylin. Con-
trol sections were incubated with non-immunized
rabbit serum (Tago Inc., 1: 2,000) substituting for
the primary antiserum or preabsorbed anti-S-100
protein serum by excessive concentration (0.1 mg/
ml) of bovine S-100 protein (Sigma) in advance.
To detect cell types of S-100 protein im-
munoreactive cells in the pancreatic islet, serial
sections were immunostained with polyclonal anti-
glucagon rabbit serum (Polysciences Inc., PAP
method, 1:10,000), monoclonal anti-insulin
mouse serum (Sekisui Chemical Ltd., indirect
method, 1: 500), polyclonal anti-somatostatin rab-
bit serum (Polysciences Inc., PAP method, 1:
4,000) and polyclonal anti-pancreatic polypeptide
(Cambridge Medical Diagnostics, PAP, 1: 4,000).

RESULTS

S-100 protein immunoreactive cells were distri-
buted in the endocrine, urogenital and digestive
organs. Schwann cells found in various tissues also
showed intense immunoreactivity. No specific
staining was found in control sections, nor breed
difference between White Leghorn and Rhode
Island Red was seen.

Endocrine system

In the hypophysis, two types of cells showed
S-100 protein immunoreactivity. One was stellate
cells (Fig. 1) and distributed throughout the ade-
nohypophysis. The other was round or ovoid cells
which were the largest among other endocrine cells
(Fig. 2). They showed eosinophilic cytoplasm and
were located in the anterior lobe. In the pancreatic
islet, S-100 protein immunoreactive cells were
located in both a and f islets. Many cells of the a

islet stained with anti-somatostatin serum showed
S-100 protein immunoreactivity (Fig. 3a, b), and
most cells of the # islet cells stained with anti-
insulin serum revealed S-100 protein positive (Fig.
4a, b). No S-100 protein immunoreactive cell was
found in the thyroid gland, parathyroid gland,
ultimobranchial gland and adrenal gland.

Urogenital system

The Sertoli cells were S-100 protein im-
munoreactive (Fig. 5), and oocytes, 1.8—2.2 mm in
diameter, contained S-100 protein immunoreactive
granules in their peripheral ooplasm (Fig. 6). In
the kidney, luminal surface or supranuclear re-
gions of epithelial cells of the distal tubules showed
positive reaction to S-100 protein antiserum (Fig.
7). The epithelial cells of collecting tubules were
also S-100 protein immunoreactive (Fig. 8).

Digestive system

S-100 protein immunoreactive cells in the prove-
ntriculus were round or pyramidal in shape and
scattered among the surface epithelial ells (Fig. 9).
In the intestinal glands, S-100 protein positive cells .
were rather sparse and recognized as spindle-
shaped form (Fig. 10).

DISCUSSION

The present study demonstrated the occurrence
and distribution of S-100 protein immunoreactive
cells in the chick non-nervous tissues. Immunohis-
tochemical localization of S-100 protein in the
stellate cells of the hypophysis has been reported
in human, monkey, goat, rat and quail [13, 17-20].
In the goat, S-100 protein immunoreactive cells
corresponded to growth hormone cells [18], but in
the human and monkey, they were not related with
growth hormone, prolactin, gonadotropin and
adrenocorticotropin cells [19, 20]. S-100 protein
immunoreactive cells in the quail adenohypophysis
were round or oval [13], while they were stellate
cells and round cells in the present study. It is
unknown whether S-100 protein immunoreactive
cells in the chick correspond to any types of
endocrine cells. In the chick pancereatic islet,
S-100 protein immunoreactive cells corresponded
to insulin cells and somatostatin cells. However,

S-100 Immunoreactive Cells in Chick 749

Fic. 1. Stellate cells show S-100 protein positive in the hypophysis. 550.

Fic. 2. Round or oval cells reveal S-100 protein in the hypophysis. 550.

Fic. 3. Serial sections stained with S-100 protein (a) and somatostatin (b) in @ islet. Most S-100 protein positive cells
correspond to somatostatin positive cells. 470.

Fic. 4. Serial sections stained with S-100 protein (a) and insulin (b) in f islet. S-100 protein cells correspond to
insulin positive cells. 330.

750 Y. Atost, K. TAKAYANAGI et al.

Fic. 5. Intense S-100 protein immunostaining is found in the Sertoli cells. 550.
Fic. 6. Granules (arrows) in an oocyte show weak S-100 protein positive. 280.

Fic. 7. Distal tubules of the kidney show S-100 protein immunoreactivity in the supranuclear portion of epithelial
cells. P: proximal tubule. 550.

Fic. 8. Many cells are S-100 protein positive in a collecting tubule of the kidney. 420.

Fic. 9. S-100 protein positive cells (arrows) of the glandular stomach. They are located above the basal membrane.
x 550.

Fic. 10. Two S-100 protein positive cells are seen in an intestinal gland of the jejunum. 600.

S-100 Immunoreactive Cells in Chick 751

S-100 protein immunoreactive cells of the monkey
pancreatic islet seem to be a new type different
from well-known endocrine cells such as glucagon,
insulin, somatostatin and pancreatic polypeptide
cells [21]. Uchida and Endo [22] reported that
secretory granules of insulin cells of the guinea pig
contained S-100 b protein. Leduque et al. [23]
observed the co-existence of S-100 protein and
insulin in cultured rat pancreatic islet cells and
speculated that S-100 protein might be associated
with differentiation of insulin cells or release of
hormone.

In the mammalian testis, S-100 protein im-
munoreactivity was demonstrated in the Leydig
cells of the human and rat, and in the Sertoli cells
of the monkey [24, 25]. In the chick, the Sertoli
cells were positive to anti-S-100 protein serum. In
the human and bovine ovaries, S-100 protein is
located in oocytes of normal or atretic follicles [9,
26]. S-100 protein of the quail ovary occurs in the
Golgi zone situated at a center of the oocyte; on
the other hand, it is observed in peripheral gra-
nules of the chick oocyte at certain developmental
stages. The variation in S-100 protein immunos-
taining of testis and ovary among these animals
appears to be a species difference.

The presence of S-100 protein in distal tubules of
the chick kidney is consistent with that of the
human [8, 9], but there is an apparent difference of
S-100 protein localization. Reaction product was
present through the cytoplasm in the human,
whereas they were restricted to periluminal or
supranuclear region of the epithelial cells in the
chick. The discreapancy of S-100 protein localiza-
tion may be attributed to a different functional
state in the renal epithelium. Molin et al. [27]
suggested an association of S-100 a@ protein with
transport of water, Ca** and electrolyte in cells of
renal tubules.

In human gastrointestinal tract, S-100 a and @
proteins were located in surface mucous cells and
foveolar cells of stomach [9]. S-100 @ protein was
detected in crypt cells of small intestine and
absorptive epithelial cells of human colon [9]. In
the chick, there were a few S-100 protein im-
munoreactive cells in proventriculus and intestine.
Although these positive cells can not be identified
in the present study, they might be endocrine cells

on the distribution.

In mammals, S-100 protein immunoreactive
cells are widely distributed in non-nervous tissues
more than described above: thyroid gland, para-
thyroid gland, adrenal gland, heart, lung, cartilage
and adipost tissue [8, 9, 28-30]. In particular, in
mammalian lymphoid tissues, S-100 protein has
been found in the Langerhans cells of the human
skin [31], histocytes of the human lymph nodes
[32], follicular dendritic cells of the rat, canine,
caprine and human lymphoid organs [33-35], in-
terdigitating reticulum cells of the human lymph
nodes [36], tingible body macrophages of the
canine and caprine lymph nodes [35] and giant
dendritic cells of the guinea pig lymphoid tissue
[37]. S-100 protein might be associated with im-
mune responses in these antigen presenting cells.
However, none of antigen presenting cell types in
the lymphoid organs (thymus, spleen and bursa of
Fabricius) was stained with anti-S-100 serum in the
present study. It is probable that a small number
of S-100 protein immunoreactive cell types in the
chick might be attributed to low levels of S-100
protein below a certain limit of sensitivity of
immunohistochemical method or to species differ-
ence of specificity of anti-bovine S-100 protein
serum against negative cell types. Immunohis-
tochemical study using antiserum against chick
S-100 protein will be needed to clarify this prob-
lem. The function of S-100 protein in various cell
types of the chick remain unknown. Haimoto et al.
[9] showed a differential localization of S-100 a and
S-100 @ proteins among various cell types in non-
nervous human organs and speculated that S-100 a
protein would be related to metabolism, especially
contraction of cardiac and slow-twitch muscles,
and S-100 £ protein in adipocytes might be associ-
ated with lypolysis. Although the heart was not
stained in the present study, the distribution of
S-100 protein in cells of the chick could introduce
new perspective on the biological function of S-100
protein.

ACKNOWLEDGMENTS

The present study was supported in pary by a Grant-in-
Aid for Scientific Research (63480090) from the Ministry
of Education, Science and Culture, Japan.

10

11

12

8;

14

752

REFERENCES

Moore, B. W. (1965) A soluble protein characteris-
tic of the nervous system. Biochem. Biophys. Res.
Commun., 19: 739-744.

Calissano, P., Moore, B. W. and Friesen, A. (1969)
Effect of calcium ion on S-100, a protein of the
nervous system. Biochemistry, 8: 4318-4326.
Boyes, B. E., Kim, S. U., Lee, V. and Sung, S. C.
(1986) Immunohistochemical co-localization of S-
100b and the glial fibrillary acidic protein in rat
brain. Neuroscience, 17: 857-865.

Hachisuka, H., Mori, O., Sakamoto, F., Sasaki, Y.
and Nomura, H. (1984) Immunohistological de-
monstration of S-100 protein in the cutaneous ner-
vous system. Anat. Rec., 210: 639-646.

Shtark, M. B., Gainutdinov, K. L., Khichenko, V.
I. and Starostina, M. V. (1981) S-100, a brain-
specific protein: localization and possible role in
snail nervous system. Cell. Mol. Neurobiol., 1: 289-
299.

Endo, Y. and Endo, T. (1988) Immunohistochemi-
cal demonstration of S-100 protein in the brain
neurosecretory cells of invertebrates (insects and
earthworms). Neurosci. Lett., 90: 11-14.

Donato, R. (1986) S-100 proteins. Cell Calcium, 7:
123-145.

Vanstapel, M. J., Gatter, K. C., Wolf-Peeters, C.,
Mason, D. and Desmet, V. D. (1986) New sites of
human S-100 immunoreactivity detected with
monoclonal antibodies. Am. J. Clin. Pathol. 85:
160-168.

Haimoto, H., Hosoda, S. and Kato, K. (1987)
Differential distribution of immunoreactive S100-a
and S100-8 proteins in normal nonnervous human
tissues. Lab. Invest., 57: 489-498.

Linser, P. and Moscona, A. A. (1981) Developmen-
tal changes in the distribution of S-100 in avian
neural retina. Dev. Neurosci., 4: 433-441.

Goto, S., Matsukado, Y., Uemura, S., Mihara, Y.,
Inoue, N., Ikeda, J. and Miyamoto, E. (1988) A
comparative immunohistochemical study of cal-
cineurin and S-100 protein in mammalian and avian
brains. Exp. Brain Res., 69: 645-650.

Callebaut, M. and Van Nassauw, L. (1986) S-100
protein immunoreactivity in the Golgi zone of pre-
lampbrush avian oocytes. IRCS Med. Sci., 14: 735.
Van Nassauw, L., Harrisson, F., Cras, P. and
Callebaut, M. (1987) Immunohistochemical loca-
lization of S-100 protein, glial fibrillary acidic pro-
tein, and neuron-specific enolase in the pars distalis
of quail, rat, and human hypophyses. Histochemis-
try, 86: 353-358.

Kahn, H. J., Marks, A., Thom, H. and Baumal, R.
(1983) Role of antibody to S100 protein in diagnos-
tic pathology. Am. J. Clin. Pathol., 79: 341-347.

15

16

JL)

18

19

20

21

Vije

23

24

U5)

26

Za i}

Y. Atoyt, K. TAKAYANAGI et al.

Takahashi, K., Isobe, T., Ohtsuki, Y., Akagi, T.,
Sonobe, H. and Okuyama, T. (1984) Immunohis-
tochemical study on the distribution of a and ~
subunits of S-100 protein in human neoplasm and
normal tissues. Virchows Archiv. B., 45: 385-396.
Sternberger, L. A. (1979) Immunocytochemistry.
2nd ed., Wiley, New York, 354 pp.

Nakajima, T., Yamaguchi, H. and Takahashi, K.
(1980) S100 protein in folliculostellate cells of the
rat pituitary anterior lobe. Brain Res., 191: 523-
SSe

Shirasawa, N., Yamaguchi, S. and Yoshimura, F.
(1984) Granulated folliculo-stellate cells and growth
hormone cells immunostained with anti-S100 pro-
tein serum in the pituitary glands of the goat. Cell
Tissue Res., 237: 7-14.

Girod, C., Trouillas, J. and Dubois, M. P. (1985)
Immunocytochemical localization of S-100 protein
in stellate cells (folliculo-stellate cells) of the anter-
ior lobe of the normal human pituitary. Cell Tissue
Res., 241: 505-511.

Girod, C., Trouillas, J., Raccurt, M. and Dubois,
M. (1986) Immunocytochemical localization of S-
100 protein in stellate cells (folliculo-stellate cells) of
the adenohypophysis in the monkeys Macaca irus
and Cercopithecus aethiops. Cell Tissue Res., 246:
237-242.

Girod, C., Durand, N. and Raccurt, M. (1987) .
Immunostaining of a cell type in the islets of
Langerhans of the monkey, Macaca irus by anti-
bodies against S-100 protein. Cell Tissue Res., 247:
11-16.

Uchida, T. and Endo, T. (1989) Identification of
cell types containing S-100b protein-like im-
munoreactivity in the islets of Langerhans of the
guinea pig pancreas with light and electron micros-
copy. Cell Tissue Res., 255: 379-384.

Leduque, P., Scharfmann, R., Basmaciogullari, A.,
Czernichow, P. and Dubois, P. M. (1989) Im-
munocytochemical evidence for the presence of S-
100 protein in insulin-containing cells of cultured
fetal rat islets. Regul. Pep., 24: 143-153.

Michetti, F., Lauriola, L., Rende, M., Stolfi, V. M.,
Battaglia, F. and Cocchia, D. (1985) S-100 protein
in the testis. An immunochemical and immunohis-
tochemical study. Cell Tissue Res., 240: 137-142.
Girod, C., Durand, N. and Raccurt, M. (1986) A
comparative study of immunocytochemical localiza-
tion of S-100 protein in the monkey Macaca irus and
the albino rat testes. Biomed. Res., 7: 333-337.
Kamiya, S., Tsukushi, M., Yamano, S. and Daigo,
M. (1989) S-100 protein-immunoreactive cells in the
bovine ovary. Anat. Rec., 223: 384-386.

Molin, S. O., Rosengren, L., Baudier, J., Hamber-
ger, A. and Haglid, K. (1985) S-100 alpha-like
immunoreactivity in tubules of rat kidney. A clue to

28

29

30

3

32

S-100 Immunoreactive Cells in Chick

the function of a “brain-specific” protein. J. His-
tochem. Cytochem., 33: 367-374.

Zabel, M., Surdyk, J. and Biela-Jacek, I. (1986)
Immunocytochemical study of the distribution of
S-100 protein in the parathyroid gland of rats and
guinea pigs. Histochemistry, 86: 97-99.

Donato, R., Giambanco, I., Aisa, M. C., di Geroni-
mo, G., Ceccarelli, P., Rambotti, M. G. and Spre-
ca, A. (1989) Cardiac S-100,, protein: purification
by a simple procedure and related immunocytoche-
mical and immunochemical studies. Cell Calcium,
10: 81-92.

Cinti, S., Gigolini, M., Morroni, M. and Zingaretti,
M. C. (1989) S-100 protein in white preadipocytes:
an immunoelectronmincroscopic study. Anat. Rec.,
224: 466-472.

Cocchia, D., Michetti, F. and Donato, R. (1981)
Immunochemical and immunocytochemical localiza-
tion of S-100 antigen in normal human skin. Nature,
294: 85-87.

Semba, M. and Kawai, K. (1989) Immunohistoche-
mical reactivity of phagocytic and non-phagocytic
histocytes in lymph nodes with lysozyme, alpha-1-
antichymotrypsin, S-100 protein, alkaline phospha-
tase, and acid phosphatase. Okajima Folia Anat.

39

34

35

36

Si)

755,

Jpn., 66: 61-68.

Cocchia, D., Tiberio, G., Santarelli, R. and Michet-
ti, F. (1983) S-100 protein in “follicular dendri-
tic” cells of rat lymphoid organs. Cell Tissue
Res., 230: 95-103.

Ushiki, T., Iwanaga, T., Masuda, T., Taka-
hashi, Y. and Fujita, T. (1984) Distribution and
ultrastructure of S-100-immunoreactive cells in the
human thymus. Cell Tissue Res., 235: 509-514.
Sugimura, M., Ishimaru, H., Atoji, Y. and Suzuki,
Y. (1987) S-100 protein a subunit immunoreactivity
of follicular dendritic cells in germinal centers of
canine and caprine lymph nodes. Jpn. J. Vet. Sci.,
49: 1183-1185.

Takahashi, K., Yamaguchi, H., Ishizeki, J., Naka-
jima, T. and Nakazato, Y. (1981) Immunohistoche-
mical and immunoelectron microscopic localization
of S-100 protein in the interdigitating reticulum cells
of the human lymph node. Virchows Arch. B. 37:
125-135.

Sugimura, M., Atoji, Y. and Suzuki, Y. (1988)
S-100 protein immunoreactive cells in lymphoid
tissues of duck, guinea pig, dog, pig and cow. Acta
Anat. Nippon., 63: 111.

ear) : = — = tide Oa nas male Su a8) Pe wr ag Set: aah Ste

ae | < amen pert Jeet
! "eens Sitges tans ancl toastnatac, os

ae ae ee a tale
cp Rt ete Tse}

ebladertachuinin ey eiienl NM Ey, Sed | = ig?
ce Ww Aealnr, paaciniy: hen Jencks tein ae

*~

i. (ea ¥ haa ene. ¥ en wi Ae. seeumisue t
‘re: @ et ! on a ——— dpheaiaingy) AL ie

5 abe letras hid Yep taHbeaN uve

| ; ‘ wes none nate nay

Pi ae a
‘ y aes : r :
: : Eas ¢ he a
f : . 4 ee! : ae ay
= cd F x mes 4 Ad
ser as
: x
laa ord
i a :
“ : .
4 ; = is : y
eee oe ~ i bp Se 2s Sets ok
re a ie. d ies
» A ‘
: a y
; ; - st ‘ vee yy ‘
i ~ ; is : ‘
“ 2
— ¥ 2 a ; a
a = ee ‘
= te 1 ;
= “
a i r
0 +, uf b c
oo eM 2
es
: é
= SS 3
Fi |
% o
€
’ : ;
— ‘ 7 i
x f .
‘6? sy" ihe at
5 > = net fi ‘ : ;, , .
- 2 : ' fm
: As
= : 3
_ 5 ds be ‘ s
¥ ns a a 7 :
ye 7 2
E ty Ue & a .
; } " es
& ae 4 bu i
ir - a
1
age is
S : f
y ~ \ z
s
~
:
ies
es 1
\
, ;
7
q 2
i rs x
z \
i
; i
7 it
Mi
4
H : ;
is 5m
fe ee ‘5
2
j
v )
2
p
r ‘ ; ;
- : é
of,
= , j
Fr
—
: ay
4
eee
;
%
i: ¢
\

ZOOLOGICAL SCIENCE 7: 755-761 (1990)

© 1990 Zoological Society of Japan

Two New Species of the Genus Monhystrium Cobb, 1920
(Monhysteridae: Nematoda) from Terrestrial Crabs of
Subfamily Sesarminae (Brachyura) in Japan

KATSUKI YOSHIMURA

Ube Junior College, UBE 755, Japan

ABSTRACT—Two new species of Monhystrium, M. tenuis n. sp. and M. brevis n. sp. (Nematoda:
Monhysteridae), are described from gill chambers of sesarmin crabs (Brachyura: Grapsidae), collected
at Ube, Onoda and Shirahama, Japan. Host crabs of M. tenuis were Parasesarma plicatum, P. pictum,
P. erythrodactylum and Clistocoeloma merguiense, inhabiting the upper littoral zone. M. brevis were
found from Chiromantes haematocheir, C. dehaani, Sesarmops intermedium and Chasmagnathus
convexus, which live mainly in the supralittoral zone. These nematodes supposedly feed on detritus

fouling the gills of these crabs.

INTRODUCTION

Although the nematodes of the family Monhys-
teridae are primarily free living, some monhyster-
ids are known to live epibiotically on terrestrial,
limnetic and marine crustaceans such as Astacus,
Gecarcinus, Cardisoma, Gammarus, Orchestia,
Ligia, etc. [|1-5, 8-11]. Among these nematodes,
three species of the genus Monhystrium Cobb,
1920, i.e. M. transitans Cobb, 1920, M. wilsoni
(Baylis, 1915), and M. inquilinus Riemann, 1969,
have been described from the gill chambers of land
crabs of the family Gecarcinidae (Gecarcinus ruri-
cola, G. lateralis, and Cardisoma guanhumi) from
Carribean coasts [1-3, 10, 11].

In Japan, gecarcinid crabs are restricted to the
south of Tokara strait, though other land crabs of
Sesarminae (family Grapsidae), are widely distri-
buted from the northern part of the Honshu main
island through the Ryukyu Islands [7]. They are
common around river mouths, inhabiting upper
intertidal zone or supralittoral zone according to
their terrestrial adaptation. Among them, Chir-
omantes haematocheir is strongly adapted to ter-
restrial environment and scarcely comes down to
touch sea water.

In order to find out whether Japanese land crabs

Accepted February 13, 1990
Received November 30, 1989

carry such nematodes in their gill chambers, 11
species of Sesarminae and 4 species of Ocypodidae
in the western part of Honshu were examined. As
a result, 2 species of Monhystrium and 1 species of
Gammarinema Kinne and Gerlach, 1953, another
representative of nematodes living on body sur-
faces of crustaceans (Yoshimura, in preparation),
were found from 8 species of Sesarminae. Other 3
sesarmins and 4 ocypodids were not infested by
these nematodes. In this report, descriptions of 2
new species belonging to the genus Monhystrium
are given, aS well as some consideration on the
relationship between these nematodes and their
host crabs.

MATERIALS AND METHODS

The crabs examined are Chiromantes haema-
tocheir, C. dehaani, Parasesarma plicatum, P. pic-
tum, P. erythrodactylum, Perisesarma_bidens,
Sesarmops intermedium, Helice tridens, H. japoni-
ca, Chasmagnathus convexus, Clistocoeloma mer-
guiense, Uca lactea lactea, Scopimera globosa,
Ilyoplax pusilla, and Macrophthalmus japonicus.
The scientific names for these sesarmin crabs are
based on Miyake [7]. Most of them, except
Chiromantes haematocheir and Parasesarma plica-
tum, were collected at several sites along Koto
River in Ube, Yamaguchi Pref., from near the
river mouth up to the limit of the tidal effect

756 K. YOSHIMURA

(about 6km upstream). Chiromantes haema-
tocheir were captured at 50 to 100 m away from the
river. Parasesarma plicatum, P. pictum and P.
erythrodactylum were also obtained at a site 1.2
km from the mouth of Ariho River, in Onoda,
Yamaguchi Pref. In addition, Sesarmops interme-
dium were collected near a small stream, close to
an inlet of Tanabe Bay, in Shirahama on Kii
Peninsula.

The samples of these crabs were obtained from
May to August of 1988. After the crabs were fixed
in 10% formalin, the carapace was removed, and
the gills were picked out with forceps. They were
placed in a petri dish containing a small amount of
water, and were examined under a low magnifica-
tion microscope for nematodes. When nematodes
were found, they were transferred to a dilute
solution of glycerine in 70% ethanol. After etha-
nol and water evaporated, the nematodes were
mounted on a slide in pure glycerine.

DESCRIPTIONS OF SPECIES

Monhystrium tenuis n. sp.
(Fig. 1. a-e)

Materials studied: Holotype, male, S.M.B.L.
Type No. 347, from gill chamber of adult Para-
sesarma erythrodactylum collected at 3.9 km from
the mouth of Koto River in Ube on August 10,
1988. Allotype, female, S.M.B.L. Type No. 348,
from gill chamber of adult Parasesarma erythro-
dactylum collected at 3.9km from the mouth of
Koto River in Ube on August 10, 1988. Paratypes,
4 males and 3 females from gill chambers of adult
Parasesarma erythrodactylum collected at 3.9 km
from the mouth of Koto River in Ube on august
10, 1988.

Description of males’:

31 (holotype)

2. 102 OM. NES

a ne
a=45.0, b=6.6, c=13.7
spic=48 ~m

S'2

Bae 1790 Min095
i oF oF om Lee ee

a=43.8, b=6.6, c=13.4
spic=51 um

b'3

— 191 M 1161
16 24 28 28
a — 4479p — OOnc— sane
spic=48 ~m

J'4

— 202 M 1231
16-26 30 30
a=44.2, b=6.6, c=14.0
spic=54 um

J's

— 181 M 1135
LOW 245 27 26
A=ADdo, Ox, C= 12.)
spic=47 um

1256 um;

1326 um;

1230 «xm;

Body slender, of almost equal diameter through-
out boy length. Cuticle smooth, with minute
sensory setae only in cephalic region. Amphids
circular, about 2 ~m wide, and located 10-12 ~»m
(12 ~m in holotype) behind anterior end, near -
base of buccal cavity. Head blunt and provided
with 10 short cephalic setae, of almost equal
length. No ocellus. Buccal cavity lacks denticles.
Nerve ring at about 60% of esophagus length from
anterior end. Neither ventral gland nor excretory
duct visible. Intestine made up of about twenty
large cells (oligocytous [6]) arranged into two rows
(a ventral and a dorsal). Tail conical, 88-95 ~m
(92 um in holotype) long, or 3.2-3.7 (3.3 in
holotype) times as long as cloacal diameter. Bur-
sa, or caudal ala, present in matured males on each
subventral side around cloaca. Two papillae on
lateral margin of each bursa. A pair of minute

1 The numbers above the line refer to lengths from the
anterior end to the end of the esophagus, the middle
(M) of the body (male) or the vulva (female), and the
anus. Those below the line refer to body widths at the
level of cephalic setae, at the level of the nerve ring, at
the middle of the body (male) or at the vulva (female),
and at the anus. The subsequent measurement refers to
the total body length. a, b, and ¢ represent De Man
ratios. In males, the length of the spicules (spic) is
measured at their chord. In females, the position of the
vulva is expressed as a percentage of the total body
length.

Monhystrium from Sesarmin Crabs 757

woz

widQgc

Fic. 1. Monhystrium tenuis n. sp. a. The total view of the holotype male. b. The total view of the allotype female. c.
The anterior end of the holotype. d. The tail of the holotype. e. The tail of the allotype.

post-cloacal papillae present on subventral sides , (allotype)

near tail end. Only two caudal glands. Single — 195 888 1209
outstretched testis on right of intestine, anteriorly 1424530 22 foie
extends to 173-253 um (213 um in holotype) after a=44.8, b=6.9, c=9.9

base of esophagus. Spicules smooth, slightly Vu=66.0%

cephalate proximally, with thin ventral alae, and

1.7-2.0 (1.7 in holotype) times as long as cloacal 72

diameter. Gubernaculum only weakly developed comme 0) ORIZZ8 1352 um;

foe 26.31 24
a=43.6, b=6.9, c=10.9
Vu=67.3%

and lies parallel to spicule dorsally.
Description of females:

758 K. YOSHIMURA

— 204 972 1320
16.7 928° 32 OP,
a—45 SD 7.1 C— 088
Vu=66.8%

Fa

YO e959 1s
Me) By sy 22
a—A6-2, b—7-45.c— 106
Vu=66.3%

1455 um;

1447 pm;

Body rather longer than in males. Tail 124-136

ym (136 um in allotype) long, or 5.2-6.2 (6.2 in
allotype) times as long as anal diameter, and
smooth, without anal alae nor caudal papillae.
Gonad unpaired, outstretched, and lies to right of
intestine. Vulva elevated. Uterus of allotype
contains a large egg, 72X25 um, which seems to
have undergone the second cleavage.
Etymology: tenuis from their slender body shape.
Remarks: The present new species, Monhystrium
tenuis, is similar to M. inquilinus Riemann, 1969 in
the general body shape and the shape of the
anterior end, but differs from the latter in that the
body is much thinner, the tail is shorter, and the
male lacks precloacal ventral spine.

The presence of a segmented egg in the uterus of
the allotype suggests that the present new species
is viviparous like the hitherto known species of
Monhystrium [3]. This may be also supported by
the occurrence of an embryo at just before hatch-
ing in a female of Monhystrium tenuis from a
Parasesarma sp. collected at Ooshima Island,
Yamaguchi Pref.

The present new species has been found from
Parasesarma plicatum, P. pictum, P. erythrodacty-
lum, and Clistocoeloma merguiense, though much
scarce in Parasesarma pictum. Although I have
examined dozens of Perisesarma bidens, no
monhysterid nematodes have been found from
their gill surface.

Monhystrium brevis n. sp.
(Fig. 2. a-f)

Materials studied: Holotype, male, S.M.B.L.
Type No. 349, from gill chamber of adult Chir-
omantes haematocheir collected at a small hill near

the mouth of Koto River in Ube on May 28, 1988.
Allotype, female, S.M.B.L. Type No. 350, from
gill chamber of adult Chiromantes haematocheir
collected at a small hill near the mouth of Koto
River in Ube on May 28, 1988. Paratypes, 3 males
and 3 females from gill chambers of adult Chir-
omantes haematocheir collected at a small hill near
the mouth of Koto River in Ube on May 28, 1988;
a male and a female from gill chambers of adult
Chiromantes dehaani at 6.3 km from the mouth of
Koto River in Ube on June 17, 1988.

Description of males:

d'1 (holotype)

— 142 M 847

Ta 29) SU a
a—27-4, b=6,6, c—1038
spic=43 um

S2

— 144 M 926

coo an Cee
a=33.8, b=7.0, c=11.6
spic=60 “m

J'3

— 147 M 955
ae 2s) BS) 2)
a—29°7- b=7. cal
spic=48 um

b4

— 130 M 778
LSP 28ie3 Gt28
a—27.9) b—6:7, €—95

1041 pm;

866 um;

spic=46 wm

Js

ie i SS
= 74 98 Ore
a— 32.0 3b — Ope — ileal
spic=57 um

Body short and thick, of almost equal diameter
throughout body length. Cuticle exclusively
smooth, with minute sensillae only in cephalic
region. Amphids circular and about 2 um wide,
situated 10-12 wm (10 4m in holotype) behind
anterior end, near base of buccal cavity. Head
blunt and provided with 10 short cephalic setae of
almost equal length. No ocellus. Buccal cavity

Monhystrium from Sesarmin Crabs 759

Fic. 2. Monhystrium brevis n. sp. a. The total view of the holotype male. b. The total view of the allotype female. c.
The anterior end of the holotype. d. The tail of the holotype. e. The tail of male 2, with fully developed spicules.

f. The tail of the allotype.

lacks denticles in its inner surface. Nerve ring at
59-63% (61% in holotype) of esophagus length
from anterior end. Excretory system indistinct.
Intestine oligocytous, with a ventral and a dorsal
rows of large cells. Tail conical, 81-88 ~m (86 um
in holotype) long, or 3.0-3.8 (3.6 in holotype)
times as long as the cloacal diameter. Bursa
present in matured males on each subventral side
around cloaca. Each bursa supplemented with two
accesory papillae on its lateral margin. A pair of
minute subventral papillae present near tail end.
Only two caudal glands. Single outstretched testis
lies right to intestine and extends anteriorly to 42-
218 um (42 um in holotype) after base of esopha-
gus. Spicules smooth, slightly cephalate proximal-

ly, and 1.7—2.5 (1.8 in holotype) times as long as
cloacal diameter. Middle part of fully developed
spicule concaved ventrally [Fig. 2.e]. Gubernacu-
lum only weakly developed and lies near distal end
of spicule.

Description of females:

¢-; (allotype)

— Bl EP ke

im of op on 7o ea
a—30.9, b=7.5, c=9.9
Vu=63.9%

$2
— 148 650 893

To on ee

760

a=35.8, b=6.8, c=9.2

Vu=64.9%

mae

—Wyib2) 650; G02 |
=pe ss 4
a=33.6, b=6.6, c=9.5
Vu=64.5%

$4

— 134 614 844 |
=e 0s yo ee
a=27.7, b=7.0, c=9.5
Vu=65.1%

“5

== 154y ett "842

Sa oar aD ema ee
a=29.7, b=6.2, c=8.7
Vu=64.2%

General body shape similar to that of males.
Tail 99-109 um (100 um in allotype) long, or 4.1-
5.5 (4.8 in allotype) times as long as anal diameter,
smooth without anal alae nor caudal papillae.
Gonad unpaired, outstretched, and lies to right of
intestine. A large egg present in uterus of the
allotype, measuring 42x19 m and segmented
into the 2-cell stage.

Etymoloty: brevis from the short body length.
Remarks: The present new species, Monhystrium
brevis, resembles M. transitans Cobb, 1920, but it
is quite different from the latter in that it has much
shorter cephalic setae. M. brevis can be easily
distinguished from M. tenuis by its shorter and
stouter body; De Man ratio a of the former is from
27.4 to 35.8 while that of the latter is from 43.6 to
48.2. M. brevis had been collected from Chir-
omantes haematocheir, C. dehaani, and Chasmag-
nathus convexus of Koto River. It was also found
in Sesarmops intermedium from Shirahama, while
S. intermedium of Koto River and Ariho River
seems to be free of them.

DISCUSSION

Two new species of Monhystrium, M. tenuis and
M. brevis, as well as one species of Gammarinema
were recorded from 8 species of sesarmin crabs
(Table 1). Adults of these crabs had the gills dirty

K. YOSHIMURA

TABLE 1. Distribution of Monhystrium tenuis n.
sp., M. brevis n. sp. and Gammarinema sp. in the
gill chambers of sesarmin and ocypodid crabs
from Ube, Onoda, and Shirahama

M...t... M._b. «G.osp.
Chiromantes haematocheir +
Chiromantes dehaani lr =P
Parasesarma plicatum a
Parasesarma pictum a
Parasesarma erythrodactylum = +
Perisesarma bidens
Sesarmops intermedium ts ale
Helice tridens
Helice japonica
Chasmagnathus convexus +
Clistocoeloma merguiense ate

Uca lactea lactea
Scopimera globosa
Ilyoplax pusilla
Macrophthalmus japonicus

with detritus and were also infested by harpacti-
coids and rotifers, but younger crabs had clean gill
surface and were free of such animals. On the
other hand, Perisesarma bidens, Helice tridens, H.
japonica, and 4 species of Ocypodidae, which were
not parasitized by Monhystrium, had clean gill
surface with little detritus. It may safely be said
that two new species of Monhystrium prefer crabs
with dirty gills, probably, feeding on detritus on
the gill surface, which accords with the observation
by Riemann [11] that Monhystrium inquilinus
feeds on such detritus.

Although both Monhystrium tenuis and M. bre-
vis were parasitic on crabs with dirty gills, they
never occurred in the same species. Monhystrium
tenuis were found in such crabs as Parasesarma
plicatum, P. pictum, P. erythrodactylum, and Clis-
tocoeloma merguiense, living in the upper littoral
zone near the river mouth. Monhystrium tenuis
especially prefer Parasesarma erythrodactylum and
Clistocoeloma merguiense which are abundant on
soft substratum. Parasesarma pictum are occa-
sionally found around such habitat, but primarily
they are inhabitants on rocky substratum, and the
infection rate of M. tenuis was very low.

On the contrary, Monhystrium brevis were col-

Monhystrium from Sesarmin Crabs 761

lected from the gill chambers of Chiromantes hae-
matocheir, C. dehaani, Sesarmops intermedium
and Chasmagnathus convexus, living higher in
upper littoral zone or in supralittoral zone. The
infection rate of M. brevis was highest in Chir-
omantes haematocheir, which is highly adapted to
terrestrial environment and scarcely goes down
into sea water. Chasmagnathus convexus, a supra-
littoral inhabitant, was infested by a moderate
number of M. brevis. Chiromantes dehaani is
distributed mainly in the upper littoral zone, and
the density of M. brevis on their gill surface was
very low. In Koto River and Ariho River, Sesar-
mops intermedium are distributed in the upper
littoral zone, and they were not parasitized by M.
brevis. At Shirahama, S. intermedium were col-
lected higher in the supralittoral zone together
with Chiromantes haematocheir, and they were
intensively infested by M. brevis.

In conclusion, Monhystrium tenuis appears to
prefer sesarmin crabs living on soft substratum in
the upper littoral zone near the river mouth, while
M. brevis to such crabs inhabiting the supralittoral
zone without close contact with sea water. The
gills of these crabs are dirty with detritus and both
species of Monhystrium seem to feed on such
detritus.

ACKNOWLEDGMENTS

The author gratefully acknowledge Dr. Eiji Harada,
Seto Marine Biological Laboratory, Kyoto University,
for his valuable suggestion during the completion of the
manuscript. Thanks are also due to Dr. Sievert Loren-
zen, Zoologisches Institut, Universitat Kiel, for his critic-
al reading of the manuscript.

10

ul

REFERENCES

Baylis, H. A. (1915) Two new species of Monhys-
tera (Nematodes) inhabiting the gill-chambers of
land-crabs. Ann. Mag. Nat. Hist., (8) 16: 414-421.
Chitwood, B. G. (1935) Nematodes parasitic in, and
associated with, Crustacea, and description of some
new species and a new variety. Proc. helminth. Soc.
Wash., 2: 93-96.

Cobb, N. A. (1920) One hundred new nemas (type
species of 100 new genera). Contrib. Sci. Nematol.,
9: 217-343.

Gerlach, S. A. (1967) Zwei neue freilebende marine
Nematoden vergesellschaftet mit Crustaceen des
Supralitorals. Verdff. Inst. Meeresforsch. Bre-
merh., 10: 209-215.

Kinne, O. and Gerlach, S. A. (1953) Ein neuer
Nematode als Kommensale auf Brackwassergam-
mariden, Gammarinema gammari n. g. Nn. sp.
(Monhysteridae). Zool. Anz., 151: 192-203.
Martini, E. (1903) Uber Furchung und Gastrulation
bei Cucullanus elegans Zed. Z. Wiss. Zool., 74:
501-556, pls. 26-28.

Miyake, S. (1983) Japanese Crustacean Decapods
and Stomatopods in Color. Vol. II. Brachyura
(Crabs). Hoikusha, Tokyo. (In Japanese)

Osche, G. (1955) Uber die Vergesellschaftung von
Nematoden und Crustaceen, mit einer Beschreibung
von Matthesonema tylosa n. g. n. sp. (Nematoda)
aus dem Kiemenraum einer Assel. Zool. Znz., 155:
253-262.

Riemann, F. (1968) Gammarinema cardisomae nov.
spec. (Nematoda: Monhysteridae) aus dem Kiemen-
raum einer karibischen Landkrabbe. Mitt. Inst. Col-
ombo-Aleman Invest. Cient., 2: 39-43.

Riemann, F. (1969) Nematoden aus dem Kiemen-
raum karibischer Landkrabben: Monhystrium in-
quilinus nov. spec. (Monhysteridae). Veroff. Inst.
Meeresforsch. Bremerh., 11: 239-244.

Riemann, F. (1970) Das Kiemenltckensystem von
Krebsen als Lebensraum der Meiofauna, mit Bes-
chreibung freilebender Nematoden aus karibischen
amphibisch lebenden Decapoden. Veroff. Inst.
Meeresforsch. Bremerh., 12: 413-428.

f%

ME

. aT, sit

=
ake?

yah

wae i]

oi ee veg ye i oH
eee he Kiet (a2 jon Pe
iw
Bint 4,
Se
4
~ i
2 as
“oe cee = A 2 J ‘ 4 :
4 : TCA TUSES ee! 272 P , ‘1
be : we
ve AD ERY S.. bbe , iy
eet ; ~
i iu eh aa Bx :
rs hing G4 yi j Re0 Pe) kee ‘ ‘ . ye byl F er, ee, epee,
i ; Sekt. SR
S . 2 <)3% ‘ =
= 2 ‘ - = a ye * 5 t ea205
ve ‘
ry f ;
Pe ae : cw | ¢ rt ai
&
i: : F
R aries BS as q j ‘ : ye
‘ FA > ©. x ¥
, Cot 5 - as 5 Se i} % t a
be " ~ 4 4 =%
, ioe i? Nae P
! t { ee ra "| i } 7
; A
5 aera) ae] Fs
j
a =
ed a » A
q - S
‘ if D
ne)
~ i
— Sa
oe
j
="
= J
1
J
;

3 ie a as
gaia:
: se ae ARG ay coe

at Sbugatnn ie “hive

- ie Ey 7 6

ae i) a
: ‘iin Bab m0
* noe a = Ns A

nal wy
> ae # Ma
Ma x ee

ugg. me a ‘i 4,
ae silat
iohorcsli
feet no L
Liatiel '

oi ashen sab rat
ri sted PE

*

ZOOLOGICAL SCIENCE 7: 763-766 (1990)

[COMMUNICATION]

© 1990 Zoological Society of Japan

The Induction of D-aspartic Acid in Mouse Lens Protein
by Continuous Gamma-Irradiation

Itsuro TAMANO!, Noriko Fusm', Siro Muraoka!, Kaoru Harapal,

HISAMASA JOSHIMA~

and TAKAAKI ISHIHARA~

Biological Department, College of Arts and Sciences, Chiba University, Chiba 260,
‘Department of Chemistry, University of Tsukuba, Tsukuba 305, and
*National Institute of Radiological Sciences, Chiba 260, Japan

ABSTRACT—The irradiation condition to induce the
increase of D-aspartic acid (D-Asp), which is detected in
aged human and animals, in a short term was studied by
continuous low dose irradiation. The results were
compared with the effect by a single X-ray irradiation.
The continuous gamma-irradiation for 12 months with 83
mGy/day or 29 mGy/day from '*’Cs could induce higher
D/L ratio of aspartic acid (Asp) in the lens protein of
RFM/MsNrs strain of mice. But the irradiation with 10
mGy/day for 14 months did not show the much increase
of D-Asp in the lens protein of mice. In 14.5 months of
post-X-irradiation to whole body with 5 Gy, a little
higher ratio of D/L Asp was observed in the lens protein
of BALB/cA strain of mice. These results indicate that it
is possible to obtain the protein containing D-Asp by
continuous low dose irradiation in a relatively short
period.

INTRODUCTION

Natural proteins in the living organism consist
exclusively of L-amino acids. However, recent
papers reported the existence of D-aspartic acid
(D-Asp) in the proteins from tooth [1, 2], eye lens
[3-5], white matter of brain in aged human [6] and
also in erythrocyte [7]. The presence of D-Asp
was, also, detected in some proteins of lens and
skin of naturally aged mice over 20 months old in
any strain of mice [8] and the proteins containing
D-Asp was biochemically characterized [9]. These
previous results suggested that the existence of
D-Asp in protein could be related with the aging

Accepted October 9, 1989
Received August 4, 1989

process. Radiation is one of the ways for the
acceleration of aging [10]. As a first step to clarify
the relationship between racemization and radia-
tion-induced aging, we have previously carried out
5 Gy whole body external irradiation to mice by
gamma-ray of '*’Cs and the increase of D-Asp was
observed in the lens protein at 18 months after
irradiation [11]. In this paper, the continuous low
dose irradiation for one year and resulting higher
D/L ratio of aspartic acid in mouse lens protein are
reported.

MATERIALS AND METHODS

Animals and irradiation

Animals used in the experiments were male
mice of RFM/MsNrs and BALB/cA strains.
Seventy five RFM/MsNrs mice were irradiated to
whole body continuously every day with 83 mGy,
29 mG or 10 mGy/22 hr/day of gamma rays from
'S7Cs. This irradiation was started from about 3
months old, and continued for about 12 months
(369 days) in both of 83 mGy and 29 mGy and for
about 14 months (417 days) in 10mGy. Twenty
BALB/cA mice, 3 months old, was subjected to
whole body irradiation with 5 Gy of X-ray. The
X-ray irradiation was done by a therapeutic X-ray
machine (Shimazu Seisakusho Co. Ltd., Kyoto),
operated at 200 kVp, 20 mA, with a dose rate of
0.66-0.71 Gy/min in air; filtration was 0.5 mm Al
and 0.5mm Cu; and HVL 1.2mm Cu [8, 11].

764

These irradiated mice were allowed water contain-
ing 3X 10~*N HCl as drinking water and food ad
libitum, and cages were covered with hoods to
avoid any possible infectious contamination.

Preparation of protein from lens

The lens was homogenized with a glass
homogeneizer in 1 ml of cold distilled water and
fractionated into water soluble (WS) and water
insoluble (WI) fractions by centrifugation with
14,000xg for 12min. The WI fraction was
fractionated by adding cold EtOH gradually in an
ice bath and the final concentration of EtOH was
adjusted to 60%. After stirring and standing for 30
min at 4°C, the solution was centrifuged at 20,000

xg for 60 min. Both supernatant (60% EtOH sup
fraction) and precipitate (60% EtOH ppt fraction)

were evaporated to dryness under reduced
pressure.

Determination of the D/L ratio of the amino acid in
lens protein

All glass-wares were baked at 500°C for 3 hr
before use in order to avoid possible contamina-
tion. The sample was hydrolyzed with 6 N HCl at

TABLE 1.
mGy, 29 mGy and 10 mGy/day

T. TAMANOI, N. Fusm et al.

110°C for 22 hr in evacuated sealed tube. The
hydrolysate was dried under reduced pressure.
The residual amino acid was converted to N-
trifluoroacetyl amino acid isopropylesters in the
usual way [11]. The ratio of D- to L-amino acid
was determined by gas-liquid chromatography
(Hitachi 163 Gas Chromatograph) using a chiral
capillary column (Chirasil-val 25 mx<0.25 mm,
Applied Science, USA).

RESULTS AND DISCUSSION

The D/L ratio of aspartic acid from mouse lens
irradiated for 12 months with 83 mGy, 29 mGy per
day, or for 14 months with 10 mGy per day (22 hr)
and non-irradiated one (control samples) are
shown in Table 1. Fractionation of proteins by
ethanol shows that D-Asp is enriched in the
ethanol supernatant (sup) more than the precipi-
tate (ppt). The high D/L ratio of aspartic acid was
detected in 29 mGy and 83 mGy irradiated sam-
ples. As the values in 29mGy were not much
different from those in 83 mGy, it was found that
there was no radiation dose-dependency between
29 mGy and 83 mGy in the induction of D-Asp.

D/L ratio of aspartic acid of lens protein after continuous whole body irradiation with 83

Radi. Dose 83 mGy/day* 29 mGy/day” 10 mGy/day* Control*
Fraction sup ppt sup ppt sup ppt sup ppt
No of Mice 8 10 8 4
0.076 0.057 0.075 0.057 0.059 0.048 0.041 0.051
0.061 0.056 0.079 0.057 0.055 0.048 0.047 0.055
0.079 0.075 0.065 0.053 0.064 0.058 0.051 0.054
0.097 0.051 0.067 0.053 0.059 0.049 0.061 0.053
0.048 0.046 0.062 0.067 0.047 0.048
0.075 0.053 0.089 0.069 0.065 0.057
0.072 0.068 0.102 0.084 0.050 0.047
0.071 0.068 0.139 0.062 0.066 0.055
0.072 0.057
0.077 0.057
Mean 0.072° 0.059 0.083° 0.062 0.058 0.051 0.050 0.053
SD 0.014 0.010 0.023 0.010 0.007 0.005 0.008 0.002

Irradiation term: *” 12 months, ° 14 months, ¢ Non irradiated.

Total dose
Age

: 9 30.6 Gy, © 10.7 Gy, © 4.2 Gy.

: 4° 15 months, © 18 months, ¢ 14 months.

T-test : © significantly differ from control (p<0.05).

D-aspartic Acid in Mouse Lens Protein 765

The D/L ratios of aspartic acid of irradiated sample
with 10 mGy were lower than those with 29 mGy
or 83 mGy in spite of lasting longer irradiation for
14 months instead of 12 months. Thus, the
increase of D/L ratio of aspartic acid can be
obtained by the continuous irradiation with 29
mGy over for one year. The values, 0.04—0.05 of
D/L ratio of aspartic acid, are due to the racemiza-
tion during the hydrolysis of lens protein. As the
difference of D/L ratios of aspartic acid betweeen
10mGy and 29mGy irradiation is statistically
significant in 5%, it suggests the existence of a
threshold between 29 mGy and 10 mGy per day to
increase D-Asp by irradiation. The above results
indicate that it is possible to obtain the protein
containing D-Asp by irradiation in a relatively
short period (12 months).

After 5 Gy acute irradiation, the term for 14.5
months seems to be not long enough for the
induction of D-Asp (Table 2). The previous report
[11] showed the value of 0.09 for the D/L ratio of
aspartic acid in the lens protein at 18 months after
5 Gy acute irradiation. It is necessary to wait much
longer time to observe similar increase of D-Asp
by the use of acute irradiation. The increase of
D/L ratio of aspartic acid does not depend only on
the total dose of irradiation, but the proceeding of
aging process for a longer period of time with (or
after) irradiation. The irradiation of totally 40 Gy
with contituous gamma-irradiation (370 mGy/day
for about 110 days) on BALB/cA mice failed to
induce high value of D-Asp [8]. Our previous

TABLE 2. D/L ratio of aspartic acid of lens protein
at 14.5 months after 5 Gy whole body irradiation

Irradiated* Non Irradiated”
Fraction sup ppt sup ppt
No of Mice 5 +
0.057 0.053 0.056 0.052
0.069 0.052 0.054 0.047
0.064 0.059 ().066 0.052
0.071 0.057 0.057 0.050
0.064 0.056
Mean 0.065 0.055° 0.058 0.050
SD 0.005 0.003 0.005 0.002
Age : * 18 months, ° 15 months.

T-test: © Significantly differ from control (p<0.05).

study indicated that at least more than 20 months
of time is necessary in order to get higher D/L ratio
of aspartic acid in the protein of lens or skin from
naturally aged mice [8, 9].

At the present time it is not clear how D-Asp in
the protein is formed in the short life span of
mouse. This is unlikely to be due to the racemiza-
tion at the body temperature during the animal’s
life time of 2 or 3 years. Because if we could apply
the rate constant for the raceminzaition reaciotn of
aspartic acid of human lens protein by Masters er
al. [4] to the racemization of the mouse lens
protein, the time required to reach the values of
0.07 or 0.10 for the D/L ratio of aspartic acid
would be 9 or 30 years, respectively [8]. It is
possible that D-Asp would come from the food
during the raise for a long term. The food was
supplied to mice with pellets of F-2 and MM-3
which were produced by Funabashi Farm Comp.
(Funabashi, Japan). Then, the D-Asp in mouse
food was analyzed with the same procedure. The
values in D/L ratios of aspartic acid in F-2 and
MM-3 were 0.039 and 0.047 which were caused by
racemization during the hydrolysis of the samples.
Therefore, the increase of D-Asp in the mouse
lens was not due to the supplied food. Thus, the
increase of D-asp would be due to some other
reasons. Recently, Geiger et al. [13] showed that
1% of racemization took place in hexapeptides
containing aspargynyl-glycine (Asn-Gly) residue
after 6 days of incubation under physiological
conditions. This racemization was explained by
the succimide formation of aspargynyl residue in
the peptide. In our previous report [9], it was
discussed that the mouse lens protein containing
D-Asp has 25% of glycine content. Therefore, if
the sequence Asn-Gly is in this protein, D-Asp
formation would proceed rapidly.

The mechanism of formation and the biological
meaning of D-Asp in structural proteins have been
unknown. In order to clarify these problems, it is
necessary to establish a practical way to get
animals containing D-Asp in a short period of
time, because loss of animals tends to result from
the raising of mice for 20 months or more. The
present procedure to obtain animals containing
D-Asp faster by irradiation would be a useful
mean for the study of the formation of D-Asp and

766 T. TAMANOI, N. Fug et al.

aging.

ACKNOWLEDGMENTS

We greatly thank Drs. Koji Kobayashi and Hiroyuki
Nogawa, and Miss Tomomi Kurosawa, Biological De-
partment, College of Arts and Sciences, Chiba Universi-
ty, and Miss Kumiko Fukutsu, Division of Radiation
Hazards, National Institute of Radiological Sciences, for
taking care of animals.

REFERENCES

1 Helfman, P. M. and Bada, J. L. (1975) Proc. Natl.
Acad. Sci. USA, 72: 2891-2894.

2 Helfman, P. M. and Bada, J. L. (1976) Nature, 262:
279-281.

3 Masters, P. M., Bada, J. L. and Zigler, J. S., Jr.
(1977) Nature, 268: 71-73.

4 Masters, P. M., Bada, J. L. and Zigler, J. S., Jr.
(1978) Proc. Natl. Acad. Sci. USA, 75: 1204-1208.

5

113)

Garner, W. H. and Spector, A. (1978) Proc. Natl.
Acad. Sci. USA 75: 3618-3620.

Man, R. H., Sandhouse, M. E., Burg, J. and Fisher,
G. H. (1983) Science, 220: 1407-1408.

McFadden, P. N. and Clarke, S. (1982) Proc. Natl.
Acad. Sci. USA, 79: 2460-2464.

Fujii, N., Tamanoi, I., Muraoka, S., Joshima, H.,
Kashima, M. and Harada, K. (1987) J. Radiat.
Res., 28: 117-125.

Muraoka, S., Fujii, N., Tamanoi, I. and Harada, K.
(1987) Biochem. Biophys. Res. Commun., 146:
1432-1438.

Walburg, H. E. Jr. (1975) Adv. Radiat. Biol., 7:
145-179.

Fujii, N., Tamanoi, I., Joshima, H., Kashima, M.
and Harada, K. (1986) J. Radiat. Res., 27: 183-190.
Knapp, D. P. (1979) Hand-book of Analytical
Derivatization Reactions, John Wiley and Sons,
New York, p. 254.

Geiger, T. and Clarke, S. (1987) J. Biol. Chem.,
262: 785-794.

ZOOLOGICAL SCIENCE 7: 767-770 (1990)

[COMMUNICATION]

© 1990 Zoological Society of Japan

Interaction of C-Reactive Protein with Macrophages in Rat

Wataru Nunomura!””, Hirosot K. WATANABE! and HipEMATsU Hira!

‘Tumour Laboratory and *Nippon Bio-Test Laboratories,
Kokubunji, Tokyo 185, Japan

ABSTRACT — In the primary culture of rat hepatocytes,
C-reactive protein (CRP) was synthesized and released
into the medium. The supernatant of the rat mac-
rophages facilitated the production of CRP, while those
from thymocytes and splenocytes had no effect. The
incubation of macrophages with rat CRP resulted in the
generation of superoxide anion.

These results suggest that there are some interaction
between CRP and macrophages in the rat.

INTRODUCTION

In rat, as in human, C-reactive protein (CRP) is
one of the acute phase proteins [1, 2]. Some acute
phase proteins, such as a -macroglobulin and
fibrinogen, are synthesized in hepatocytes by
exposure to rat macrophage cytokines [3]. On the
other hand, the mouse macrophage was reported
to be activated by human CRP and to generate
superoxide anion which reveals tumoricidal activ-
ity [4, 5]. However, the interaction between CRP
and macrophages in the rat remains unknown.

The normal level of CRP in rat serum is much
higher than that in human serum (about 5,000-
fold), and the rat CRP is somewhat different in
structure from human CRP. For example, rat
CRP has an S-S bridge in its molecule, being
different from human CRP [1]. It is interesting to
investigate whether rat CRP activates the mac-
rophage or not.

MATERIALS AND METHODS
Animals and serum

Adult male and female Wistar rats, weighing

Accepted February 8, 1990
Received September 4, 1989

200-270 g, were obtained from Shizuoka Labora-
tory Animal Center, Japan, and were housed
under constant temperature at 28°C ona 14L: 10D
schedule until examined. Newborn rats (10-day-
old) were produced in this laboratory. Rat serum
was obtained by decapitation and stored at —20°C.

Purification of rat CRP

CRP was purified from rat serum by the lecithin
precipitation method [6] with a slight modification
as described previously [1]. The purity of CRP was
assayed with SDS-PAGE [7] and immunoelec-
trophoresis using antisera raised in rabbits against
the whole rat serum and purified CRP [8]. The
endotoxin activity of the purified CRP prepara-
tions was less than 0.01 ng/ug protein, as deter-
mined with Limulus amebocyte lysate assay kit
(Wako Pure Chemical Industries Ltd., Japan).

Isolation and culture of lymphocytes and mac-
rophages

Macrophages were collected in serum-free
Eagles’ minimum essential medium (MEN, Nissui
Pharmaceutical Co., Ltd., Japan) from peritoneal
exudates of male rats that had been intraperi-
toneally given 2.5 ml of 10% glycogen broth in
0.9% NaCl 5 days before. The cells were incu-
bated for 60 min at 37°C in MEM supplemented
with 5% heat-inactivated fetal calf serum (FCS,
Nippon Bio-Test Laboratories Inc., Japan), and
the monolayer cells were harvested. Thymocytes
and splenocytes were collected from male rats
injected 1 ml of turpentine-oil intramuscularly 2
days before sacrifice. These cells were incubated
under the same condition as for macrophages. The
macrophage, thymocyte and splenocyte (10° cells)

768 W. Nunomura, H. K. WATANABE AND H. HIRAI

were separately cultured in MEM for 24hr at
37°C, and each culture medium was examined for
its cytokine activity.

Isolation and culture of hepatocytes

Hepatocytes were isolated from the livers of
male rats under sterile conditions. The two-step
collagenase perfusion technique after Seglen [9]
was used. After perfusion, the livers were re-
moved and dispersed. The hepatocytes were
suspended in MEM, and nonparenchymal cells
were removed by repeated (3 times) centrifuga-
tions (50g, 5 min). The viability of the hepato-
cytes was 80-90% as revealed by the trypan blue
exclusion test. The isolated hepatocytes were
suspended in MEM containing 5% FCS, 10 mM
N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic
acid (HEPES), penicillin (100 units/ml) and strep-
tomycin (0.1 mg/ml). An aliquot of the cell
suspension (10° cells/0.5 ml- MEM) was plated in a
culture plate (Falcon) with 24 wells of 16mm
diameter. After 6hr incubation at 37°C in 5%
CO>,, unattached cells were removed by gentle
aspiration and 0.5 ml of the fresh supernatant
obtained from lymphocytes or macrophage culture
was added to the hepatocyte culture. The medium
was changed daily.

Quantification of CRP

The levels of CRP in the serum and in the
supernatant of lymphocyte, macrophage or
hepatocyte cultures were determined by an en-
zyme-immunoassay (EIA) according to the
method of Hibi et al. [10].

Measurement of superoxide anion

The macrophages (5X10? cells) suspended in
0.1 ml of MEM containing 5% FCS were transfer-
red into wells of a 96-well microplate (Falcon),
mixed with 0.1 ml of purified CRP of various
concentrations dissolved in 0.01 M phosphate buf-
fer, pH 7.4, containing 0.15 M NaCl (PBS), and
incubated for 2 hr at 37'C. And then, for the assay
of superoxide anion, 16nM ferricytochrome c
(type III, Sigma) in phenolred-free MEM was
added to the macrophage culture and incubated
for 1 hr, with or without 100 nM phorbol myristate
acetate (PMA, Sigma) as a stimulant. The amount

of superoxide anion was determined by measuring
the absorbance of reduced ferricytochrome c at
550 nm with a photospectrometer (228, Hitachi,
Japan) [11]. Results were expressed as nmol!
superoxide anion produced per 10’ macrophages
after subtraction of base-line reading. Calculations
were made from the difference in the absorbance
at 550nm between unstimulated and PMA-
stimulated macrophages. For the positive control,
Lipopolysaccharide (LPS) from Escherichia coli
(strain 055 : BS, final concentration was 100 ng/ml,
Sigma, USA) was used in place of CRP.

In vivo experiment

Newborn rats (10-day-old) were intraperitoneal-
ly injected with 0.1 ml of the supernatant of rat
macrophage described above. As control, rats
were injected with either the same amount of
MEM or none. The serum level of CRP in rats was
examined on the following day of the injection.

RESULTS AND DISCUSSION

The level of CRP in the culture medium of rat
hepatocytes gradually decreased with time as
shown in Table 1. By adding the supernatant of
macrophage culture, the synthesis of CRP was
sustained; however, that of thymocyte or spleno-
cyte culture yielded no such effect. No CRP was
detected in the supernatant of each lymphocyte
and macrophage culture added (data not shown).

The effect of the supernatant of macrophage
culture on the serum level of CRP in 10-day-rats is
shown in Table 2. The serum level of CRP was
significantly increased by injecting of the super-
natant compared with that of the non-treatment,
and of rats injected MEM alone.

Release of superoxide anion from rat mac-
rophages increased as a result of adding CRP to
the culture medium. The maximum release occur-
red by 4 ug/ml of CRP (Fig. 1). In any case no
superoxide anion was detected when bovine super-
oxide dismutase (SOD, final concentration was 600
units/ml, Sigma, USA) was added to the mac-
rophage culture.

The results described above clearly reveal an
interaction between CRP and macrophages. That
is, CRP synthesis in hepatocytes was stimulated by

Interaction of CRP with Mo in Rat 769

TABLE 1.

Supernatant 6

Effect of the supernatant of lymphocyte and macrophage cultures on
CRP synthesis by rat hepatocytes

Time (hr)*
30 54
CRP production (ng/ml/10° cells/24 hr)
None (control) 60:0 s555-85* 30.0+0.6 10263-0257)
Thymocytes 120221329 40.0+3.6 2162203
Splenocytes (HL Dae 9) JgoVae Itosl Voeyae UD)
Macrophages 64.1+15.3 61.8+6.7 58.0+1.0

* After adding supernatant to the wells.

The supernatant of each lymphocyte and

macrophage culture added to the culture of rat hepatocytes synthesizing CRP. The
effect of the addition on the amount of CRP was follwed.
** Mean values from 3 individual experiments are shown (mean+S.D.).

TaBLE2. Effect of supernatant of macrophage
culture on the serum level of CRP in 10-day-old

rat
Treatment* CRP (ug/ml, mean+S.D.)
None 2.58+0.45 (n=6)

Injection of
Supernatant

MEM

4.49+0.72 (n=10)**
3.90+40.23 (n=7)

* Rats were injected i.p. with supernatant of mac-
rophage culture or MEM alone, and examined on
the next day.

** Significantly different from none treatment (p<
0.05).

the supernatant of macrophage culture. In connec-
tion to this evidence, Fuller and Ritchie [12]
reported the supernatant of cultured human
peripheral monocytes and liver macrophages
(Kupffer cell) to increase the synthesis of fibri-
nogen in cultured rat hepatocytes. Geiger et al.
[13] reported that recombinant human IL-6 in-
creased the hepatic mRNA level of a-
macroglobulin, #-fibrinogen, cystein proteinase
inhibitor and a,-acid glycoprotein, and reduced
synthesis of albumin in the rat. Therefore, the
synthesis of CRP in the rat may also be influenced
by soluble factor(s) released from macrophages.
Zahedi and Mortensen [4] reported that human
CRP activated mouse macrophages to enhance
their tumoricidal activity, being independent of
lymphokines and a bacterial lipopolysaccharide.
According to Barna et al. [5], human CRP en-
veloped in liposome elevated the production of
susperoxide anion and generated tumoricidal

100

Oo ©
O oO

BAS
O

Superoxide Anion
(nmol/10’cells/hr)

N
eo)

O
012 4 8

RAT CRP (ug/ml)

Fic. 1. Effect of rat CRP on the release of superoxide
anion from macrophages. Superoxide anion was
determined by the reduction of ferricytochrome c.
Experimental conditions are described in MATE-
RIALS AND METHODS. —@~—; mean+S.D.
from 4 experiments, LPS; lipopolysaccharide of E.
coli.

activity of mouse macrophages in vivo. In the
present study, addition of CRP to the culture of
macrophages resulted in the enhancement of
release of superoxide anion by macrophages.
These findings may suggest that CRP serves as a
physiological macrophage activator, contributing
to the acceleration of a nonspecific host resistance
in inflammatory response. Because the serum

770 W. Nunomura, H. K. WATANABE AND H. Hiral

level of CRP in the rat is very high compared with
that of the normal human [1, 2], there might be
another mechanism for protecting their own tis-
sues from possible excess of superoxide. Superox-
ide dismutase which occurs in all normal tissues
especially rich in liver, might be one of the
candidates.

REFERENCES

1 Nunomura, W., Kawabe, K., Igarashi, M. and
Hirai, H. (1986) Zool. Sci., 3: 1019.

2 DeBeer, ks Gy, Baltz. Me ae Munne EAL
Feinstein, A., Taylor, J., Bruton, C., Clamp, J. R.
and Pepys, M. B. (1982) Immunology, 45: 55-70.

3 Koj A., Gauldie, J., Regoeczi, E., Sauder, D. N.
and Sweeney, G. D. (1984) Biochem. J., 224: 505-
514.

4 Zahedi, K. and Mortensen, R. F. (1986) Cancer
Res., 46: 5077-5083.

5)

10

11

12

13

Barna, B. P., Deodhar, S. D., Gautam, S., Yen-
Liberman, B. and Roberts, D. (1984) Cancer Res.,
44: 305-310.

Hokama, Y., Tam, R., Hirano, W. and Kimura, L.
(1974) Clin. Chim. Acta, 50: 53-62.

Laemmli, U. K. (1970) Nature, 227: 680-685.
Grabar, P. and Williams, C. A. (1953) Biochem.
Biophys. Acta, 10: 193-194.

Seglen, P. O. (1976) Method. Cell. Biol., 36: 309-
312.

Hibi, N., Shima, K., Tashiro, K., Tsuzuki, K.,
Tsukada, Y. adn Hirai, H. (1984) J. Neurol. Sci.,
65: 333-340.

Pick, E. and Mizel, D. (1981) J. Immunol. Method.,
46: 211-226.

Fuller, G. M. and Ritchie, D. G. (1982) Ann. N. Y.
Acad. Sci., 389: 308-322.

Geiger, T., Andus, T., Klapproyh, J., Hirano, T.,
Kishinoto, T. and Heinrich, P. C. (1988) Eur. J.
Immunol., 18: 717-721.

ZOOLOGICAL SCIENCE 7: 771-777 (1990)

[COMMUNICATION]

© 1990 Zoological Society of Japan

A Scanning Electron Microscopic Study of the Blood
Vascular Architecture of the Snake Hypophysis

ICHIRO KOSHIMIZU and YOSHIHIKO OoTA

Biological Institute, Faculty of Science, Shizuoka
University, Shizuoka 422, Japan

ABSTRACT— By means of light- and scanning electron
microscopy, the vascular route in the hypophysial region
was studied. Materials used were four species of adult
snakes (Elaphe quadrivirgata, E. climacophora, E. con-
spicillate and Rhabdophis tigrinus). The infundibular
arteries give off the primary capillary plexus in the
median eminence. Interesting fact is that the primary
capillary plexus in the dorsal median eminence is
continuous with the portal vessels supplying the secon-
dary capillary plexus in the cephalic lobe of the pars
distalis, similarly, the primary capillary plexus of the
veniral median eminence converges to portal vessels
supplying the secondary capillary plexus in the caudal
lobe of the pars distalis. These results suggest that the
pars distalis of snake hypophysis is controlled by a dual
hypothalamic regulatory mechanism.

INTRODUCTION

The regional differentiation of the hypothalamo-
hypophysial complex has been demonstrated in
birds [1-3] and in turtles [4-7]. It is generally
accepted that in birds and reptiles the synthesis
and release of adenohypophysial hormones is
controlled by hypothalamic releasing or inhibiting
factors reaching the pars distalis (PD) via the
hypothalamo-hypophysial portal vessels. In this
respect the vascular arrangement of the hypophy-
sial region is of interest. Previously, the vascular-
ization of the hypophysis of the turtle (Pseudemys
scripta and Geoclemys reevesii) and the snake
(Elaphe quadrivirgata) has been studied by injec-
tion of India-ink and tissue sectioning [8-10]. A
three-dimensional reconstruction of the hypophy-

Accepted January 30, 1990
Received December 27, 1989

sial vascularization in lower vertebrates is, howev-
er, lacking.

In the present study we aimed to analyze the
blood vascular supply of the snake hypophysis by
means of light- and scanning electron microscopy
of corrosion casts.

MATERIALS AND METHODS

Four species of adult male and female snakes,
Elaphe quadrivirgata, E. climacophora, E. conspi-
cillate and Rhabdophis tigrinus were caught in the
fields. The hypothalamic region with intact
hypophysis was fixed in Bouin’s solution in each
animal and prepared for routine histology. Thin
paraffin sections were stained with Heidenhain’s
Azan or Gomori’s paraldehyde fuchsin (AF). For
the study of the vascularization, the animals were
anesthetized with ethyl ether, and dissected to
expose the heart region. After perfusion with
saline, a mixture of India-ink and saline (1: 1) was
injected through the pulsating heart. After de-
capitation, the brains were fixed in Bouin’s solu-
tion, and trimmed to small pieces afterward
remaining the hypothalamus and the hypophysis.
Thick paraffin sections of 100 ~m were prepared
without any staining procedure.

In order to obtain the vascular casts, a commer-
cially available methylmethacrylate medium (Mer-
cox, Japan Vilene Hospital) was injected through
the ventricle of the pulsating heart. Polymeriza-
tion of the resin was achieved within 5 min in situ.
After decapitation, the brains were treated with
20% NaOH or KOH solution, rinsed with water,

Ga I. KOSHIMIZU AND Y. OoTA

and then air-dried. The tissues were sputter-
coated with gold, and examined with a Hitachi
S-450 scanning electron microscope with an accel-
eration voltage of 20 KV.

RESULTS

In the snake, the median eminence (ME) forms
a well developed ampulla on the ventral side of the
infundibulum. The ME is composed of three
layers: an inner ependymal layer, an intermediate
fiber layer, and an outer palisade layer (Fig. 1).
The AF-positive neurosecretory neurons from the
hypothalamic nuclei proceed, almost through the
intermediate fiber layer, to the pars nervosa. The
palisade layer occupies almost ventral half of the
ME. The palisade layer contains primary capillary
plexus deriving from branches of the infundibular
arteries. Some of the capillary vessels penetrate
deeply into the palisade layer of the ME. A
marked accumulation of AF-stainable neurosecre-
tory material is demonstrated along the capillaries
(Fig. 1).

There are no detectable differences in vascular
architecture among four species of snakes investi-
gated. The internal carotid artery branches off
into the infundibular arteries anterolaterally to the

Fic. 1.

ME. In the stalk region of the ME the numerous
branches of the infundibular arteries comprise the
primary plexus (Fig. 2). These capillary vessels are
very elaborate in structure, and they penetrate
deeply into the palisade layer of the ME. The
capillaries form a complicated reticular network of
small vessels. From the primary capillary plexus of
the ME, venules comprising the portal vessels
extend into the PD and supply a secondary
capillary plexus. Snakes have no pars tuberalis,
hence the portal vessels pass from the ME to the
PD through the connective tissue, usually desig-
nated as the pars terminals (Fig. 1). At the pars
terminals region, the primary capillary plexus
unites with several large portal vessels which enter
the PD. In India-ink injected specimen, the portal
vessels are demonstrated distinct dorsal and ven-
tral groups originating from the dorsal and ventral
capillary plexus of the ME, respectively (Fig. 3).
Using the scanning electron microscope on corro-
sion casts, a three dimensional vascular arrange-
ment is clearly confirmed. Within the ME, region-
al specialization of primary plexus angioar-
chitecture is visualized (Fig. 4).

Some portal vessels coming from the capillary
plexus of the dorsal ME run vertically downward,
and break up into a secondary sinusoidal capillary

N Aces <

~ 4 ke

x Sh af
SS phe
2 e

Mid-sagittal section through the infundibulum and hypophysis in Elaphe quadrivirgata. Ca, caudal lobe of

PD; Ce, cephalic lobe of PD; DME, dorsal ME; PI, pars intermedia; PN, pars nervosa; VME, ventral ME. AF

stain. 50.

Hypophysial Portal System in Snake 773

Fic. 2. Vascular cast of pituitary in Elaphe quadrivirgata (ventral view). IA, infundibular artery; PC, primary
capillary plexus; PL, plexus intermedius; S, secondary capillary plexus. X30.

net within the anterior portion of the PD, cephalic
lobe (Fig. 4). On the other hand, other portal
vessels arising from the capillary plexus of the
ventral ME provide an intimate connection to the
secondary capillary net within the posterior por-
tion of the PD, caudal lobe (Fig. 5). The secon-
dary capillaries both from the cephalic lobe and

caudal one converge into several thick vessels (Fig.
5). These vessels are continuous with venous sinus
locating dorsoposterior to the pars nervosa. This
sinus joins between the secondary capillary plexus
of the PD and superficial plexus of the pars
nervosa, and is continuous with hypophysial veins.
The findings described above are schematically

774 I. KOSHIMIZU AND Y. OoTA

Fic. 3.

Mid-sagittal section through the infundibulum and hypophysis of Elaphe climacophora. India-ink injected

specimen. a, primary capillary plexus in the dorsal ME; b, dorsal group of portal vessel; c, secondary capillary
plexus in the cephalic lobe of PD; d, primary capillary plexus in the ventral ME; e, ventral group of portal vessel;
f, secondary capillary plexus in the caudal lobe of PD; PL, plexus intermedius. X50. :

illustrated in Figure 6.

DISCUSSION

The vascularization of bird and turtle hypoph-
ysis has been investigated previously, and the
presence of portal system which supplies the PD
with venous blood from the capillaries of the
primary plexus in the ME has been demonstrated
[11-13]. Concerning the vascularization of the
hypophysial structure of bird, Vitums et al. [1, 2]
have demonstrated the presence of anterior and
posterior primary plexus of which being drained
separately by portal vessels belonging to the
anterior and posterior groups and passing over the
cephalic and caudal lobe of the PD, respectively.
Similar findings have already been demonstrated
in the vascularization of the turtle hypophysial
structure [8, 9].

The present study indicates that the snake ME
has distinct dorsal and ventral regions and is
covered by a well developed capillary network,
corresponding to dorsal and ventral capillary
plexus. Moreover, there are distinguishable dorsal

and ventral groups of portal vessels originating in
the dorsal and ventral capillary plexus of the ME,
respectively. These observations suggest that
there is regional specialization of the vascular
system corresponding to the drainage of the ME
capillary blood containing hypothalamic neuro-
secretory factors.

The portal vessels break up into secondary
capillary plexus in the PD. The PD of the snake
shows distinct histological subdivisions into cepha-
lic and caudal lobes. Using various staining
techniques, including immunohistochemical study,
a characteristic distribution of the PD cells has
been reported [3, 10, 13, 14]. In the previous
study, structural differences between the cephalic
and caudal lobes have been reported in the snake,
Elaphe quadrivirgata, as well as the difference of
distribution pattern of prolactin-immunoreactive
cells in the cephalic lobe and _ thyrotropin-
immunoreactive cells in the caudal lobe [10]. Our
findings on the hypophysial vascular routes show
that some portal vessels coming from the dorsal
ME run directly to the cephalic lobe, while other
portal vessels arising from the ventral ME run

Hypophysial Portal System in Snake VI

Fic. 4. Vascular cast of pituitary in Rhabdophis tigrinus (dorsal view). a, primary capillary plexus in the dorsal ME;
b, dorsal group of portal vessel; c, secondary capillary plexus in the cephalic lobe; IA, infundibular artery; PL,

plexus intermedius. x90.

horizontally towards the caudal lobe of the PD.
Therefore, different hypophysiotrophic neurohor-
mones may be transported from the dorsal and
ventral ME into the cephalic and caudal lobe of the
PD by different vascular route, respectively.
Histologically, the PD of bird and turtle is
divided into cephalic and caudal lobe which
correspond closely to the parts derived from the
oral and aboral lobe of Rathke’s pouch [1-3, 13].
The ME of bird and turtle consists of distinct
anterior and posterior dividions, and separate

vascular system leading to the cephalic and caudal
lobe of the PD is visualized. The vascular

relationships between the ME and the PD is
thought to have evolved as the dual regulatory
stystem. Since similar structural differentiation of
the vascularization is found in the snake, it seems
more probable that the duality of functional link
between the ME and the PD are generalized in the
reptile.

Although anura show well developed vascular
system in the ME, the portal vessels arising from
the ME do not supply distinct areas in the PD as is
demonstrated in birds and reptiles [15].

Fic. 5.

776

I. KOSHIMIZU AND Y. OOTA

Vascular cast of pituitary in Elaphe quadrivirgata (ventral view). IA, infundibular artery; PC, primary

capillary plexus in the ventral ME; PL, plexus intermedius; S, secondary capillary plexus in caudal lobe of PD;

VS, venous sinus. X40.

REFERENCES

Vitums, A., Mikami, S., Oksche, A. and Farner, D.
S. (1964) Vascularization of the hypothalamo-
hypophysial-complex in the white-crowned sparrow,
Zonotrichia leucophrys gambelii. Z. Zellforsch., 64:
541-569.

Vitums, A., Ono, K., Oksche, A., Farner, D. S.
and King, J. R. (1966) The development of the
hypophysial portal system in the white-crowned
sparrow, Zonotrichia leucophrys gambelii. Z. Zell-
forsch., 73: 335-366.

Mikami, S. (1986) Comparative cytology of the
anterior pituitary of higher vertebrates. In “Pars
Distalis of the Pituitary Gland— Structure, Function

and Regulation”. Ed. by F. Yoshimura and A.
Gorbman, Elsevier Sci. Publ. B. V., Amsterdam,
pp. 71-79.

Holmes, R. L. and Ball, J. N. (1974) The Pituitary
Gland. A Comparable Account. Cambridge Univ.
Press, London.

Tsuneki, K. (1976) Neurosecretory axo-axonic
synapses in the median eminence of the turtle,
Geoclemys reevesii. Neuroendocrinology, 20: 59-67.
Oota, Y. (1980) Ultrastructural aspects of the
anterior region of the median eminence of the turtle,
Geoclemys reevesii. Rep. Fac. Sci., Shizuoka Univ.,
14: 75-88.

Oota, Y. (1981) Ultrastructural aspects of the
posterior region of the median eminence of the

Hypophysial Portal System in Snake

777

Fic. 6. Diagram of mid-sagittal section through the infundibulum and hypophysis. a, primary capillary plexus in the
dorsal ME; b, dorsal group of portal vessel; c, secondary capillary plexus in the cephalic lobe of PD; d, primary
capillary plexus in the ventral ME; e, ventral group of portal vessel; f, secondary capillary plexus in the cephalic

10

11

lobe of PD; PL, plexus intermedius; VS, venous sinus.

turtle, Geoclemys reevesii. Rep. Fac. Sci., Shizuoka
Univ., 15: 29-38.

Oota, Y. and Kawada, M. (1986) On the vascular
supply of hypophysis in the turtle, Pseudemys
scripta. Proc. Japan Acad., 62B: 69-71.

Oota, Y. and Koshimizu, I. (1988) Vascular supply
of hypophysis in the turtle, Geoclemys reevesii.
Zool. Sci., 5: 1013-1018.

Oota, Y. and Koshimizu, I. (1987) On the vascular
supply of hypophysis and distribution of prolactin-
and thyrotropin-immunoreactive cells of the ade-
nohypophysis in the snake, Elaphe quadrivirgata.
Proc. Japan Acad., 63B: 297-299.

Green, J. D. (1951) The comparative anatomy of
the hypophysis, with special reference to its blood
supply and innervation. Am. J. Anat., 88: 225-311.

72

13

14

15

Wingstrand, K. G. (1951) The Structure and
Development of the Avian Pituitary. From a Com-
parative and Functional Viewpoint. C. W. K.
Gleerup, Lund.

Saint Giron, H. (1970) The pituitary gland. In
“Biology of the Reptilia. Vol. 3”. Ed. by C. Gans,
Academic Press, London/New York, pp. 135-199.

Sotowska-Brochocka, J. (1976) Identification of
prolactin cells in the pituitary gland of grass snake,
Natrix natrix (L.) using immunofluorescence techni-
que. Bull. Acad. Pol. Sci., Ser. Biol., 24: 107-112.
Lametschwandtner, A., Simonsberger, P. and
Adam, H. (1977) Vascularization of the pars distalis
of the hypophysis in the toad, Bufo bufo (L.)
(Amphibia, Anura). Cell Tiss. Res., 179: 1-10.

=? | |
a | |
4" f R |
+ 3 ,
aa i .
_ i ;
) z 5
" 4 ,
1) ee ,
Le)
\
|
; >
ee
.

ZOOLOGICAL SCIENCE 7: 779-782 (1990)

[COMMUNICATION]

© 1990 Zoological Society of Japan

Immunocytochemical Identification of Four Cell Types
in the Pancreatic Islets of the Japanese
Serow, Capricornis crispus

YASURO ATOJI, YASUAKI TAKADA, YOSHITAKA SUZUKI

and MAKoTo SuGcimMuRA!

Department of Veterinary Anatomy, Faculty of Agriculture, Gifu University,
Gifu 501-11, and ‘Department of Veterinary Anatomy, Faculty
of Veterinary Medicine, Hokkaido University,
Sapporo 060, Japan

ABSTRACT—The pancreatic islet of the Japanese
serow was immunocytochemically examined. The islets
were classified into large and small types, and both types
of islets showed a similar composition of endocrine cells.
The B cells were prominent and located at the periphery
or throughout the islet. The A cells were less numerous
and D and PP cells were sparse. The PP islets showing a
majority of the PP cells were sometimes found. The large
and small islets were detected irrespective of age of
animals. These findings indicate a histological character-
istics of the pancreatic islet of the Japanese serow.

INTRODUCTION

The pancreatic islet contains at least four cell
types of hormone-secreting cells: glucagon (A
cells), insulin (B cells), somatostatin (D cells) and
pancreatic polypeptide (PP cells). The islet shows
a species variety of cell localization and popula-
tioin. In the bovine pancreas, the A and B cells
usually distribute at the center or peripheral region
of the islets [1-3]. Wakuri et al. [4] reported that
the B cells of the Japanese serow, a wild ruminant,
were located at the periphery of the pancreatic
islets. This finding was obtained from aldehyde-
fuchsin staining for the B cells, but was not
confirmed by immunocytochemical technique.
Furthermore, localization of the A, D and PP cells
of the Japanese serow remains unknown. The

Accepted September 27, 1989
Received June 6, 1989

present study was carried out to elucidate the
localization of the endocrine cell types in the
pancreatic islet of the Japanese serow by im-
munocytochemical staining and to compare the
findings with the results obtained from other
mammals including domestic ruminants.

MATERIALS AND METHODS

Many Japanese serows, Capricornis crispus,
were killed in order to protect plants in winter
from 1983 to 1985 in Japan. Among those,
well-preserved 10 animals (7 males and 3 females)
were selected for the present study. They were
estimated to range from fawn to about 10 years old
[5]. Right lobes of the pancreas were removed and
immediately fixed in 10% formalin. Small blocks
were dehydrated with ethanols, embedded in
paraffin and serially cut at 3-S um in thickness.
Immunocytochemical procedure was performed by
the indirect or peroxidase-antiperoxidase (PAP)
method. The primary antisera used were as
follows: polyclonal rabbit anti-glucagon (Polysci-
ence Inc., 1:1,500), monoclonal mouse anti-
porcine insulin (Sekisui Chemical, 1: 1,000),
polyclonal rabbit anti-somatostatin (Polyscience
Inc., 1: 1,000) and polyclonal rabbit anti-porcine
pancreatic polypeptide (Cambridge Medical Di-
agnostics, 1:2,000). Incubation time was over-
night at 4°C for each primary antisera. For the

780 Y. Atoyt, Y. TAKADA et al.

PAP method, the sections were further incubated
with goat anti-rabbit IgG (Tago Inc., 1: 100) as the
second antiserum for 30 min at room temperature.
After incubation with a peroxidase-labeled goat
anti-mouse IgG or PAP complex for 30 min at
room temperature, it was visualized using 3,3’-
diaminobenzidine (25mg/100 ml) and 0.003%
H,O>. Thereafter, sections were lightly counter-
stained with hematoxylin. Immunocytochemical
control was carried out by substituting normal
rabbit (Tago Inc., 1:2,000) or mouse (Cedarlane
Laboratories Ltd., 1: 2,000) serum for the primary
antiserum. The primary antisera preabsorbed by
excess concentration of hormones were also used.

RESULTS

The islets varied in size and an average number

1c

FIG. EL:
somatostatin (c) and PP (d) (arrows). 100.

of islets per area of the pancreatic tissue in 10
animals was about 4.9/mm’. The islets were
roughly classified into large and small types. Large
islets (200-900 um in diameter) were located in
the lobules or in the interlobular connective tissue,
whereas small ones (50-150 4m in diameter) were
round or ovoid in shape and were found solely in
the lobules. No quantitative change in number of
the large islet with aging was found nor sexual
difference of the size of the islet was seen. The
distribution pattern of the four cell types (A, B, D
and PP cells) in the islet was similar in both large
and small islets. The A cells tended to be located
largely in the middle of the islet (Figs. 1a, 2a) and
formed a small population (about 10%). The B
cells were the predominant cell type distributed
throughout the islet (Fig. 1b) or at the periphery
(Fig. 2b), accounting for about 80%. The D cells

1d

Serial sections of a large islet of about a 10-year-old male. Immunostaining for glucagon (a), insulin (b),

Pancreatic Islet of Japanese Serow 781

were scattered throughout the islet (Figs. 1c, 2c)
and made up about 8% of the islet. The PP cells
were very sparse (1-2%) (Figs. 1d, 2d) and islets
without PP cells were frequenctly encountered.
However, in 3 of 10 animals, the PP islets occupied

with numerous PP cells (30-80% of islet cells)
were scattered through the pancreas (Fig. 3a). In
this case, the remaining cells of the PP islet were
usually B cell type (Fig. 3b), and A and D cells
were sparse. Extra-insular cells of four cell types

Fic. 2. Serial sections of a small islet of an adult male. Immunostaining for glucagon (a), insulin (b), somatostatin (c)

and PP (d). X210.

Fic. 3. Serial sections of a PP islet of an adult male. Immunostaining for PP (a) and insulin (b). 340.

782 Y. Atoyi, Y. TAKADA et al.

were sometimes observed in the exocrine acinus
and duct. Neuro-insular complex was foungd in
large islets (about 30%) and contained four cell
types of endocrine cells.

DISCUSSION

The present study demonstrated four types of
endocrine cells in the pancreatic islet of the
Japanese serow. In the four endocrine cells, a
great majority of the B cells in the islet is in
agreement with those of other mammals [6-11].
Previously, Wakuri et al. [4] found a mass of the B
cells at the periphery of the islet in the Japanese
serow. In the present study, however, distribution
pattern of the B cells varied; the B cells showed
peripheral localization or occupied most, if not all,
region of the islet. This discrepancy of localization
probably results from the differences in the fixa-
tion condition, staining method or physiological
state. The PP cells are scattered in the islet and
also exocrine portion of the human, horse, bovine,
echidna, guinea pig, opossum and rat pancreas [6-
10, 12]. In addtion, the PP islet occurred in cattle
and sheep [3, 13]. Nakajima et al. [3] reported a
coexistence of PP and serotonin in the PP islet, and
speculated that serotonin would play roles of
formation, storage and secretion of PP. In the
Japanese serow, the PP islet was observed in 3 of
10 animals, but it is unknown whether the PP islet
contains any monoamines.

Bovine large islets were situated in the inter-
lobular connective tissue and consisted almost of
the B cells [1, 14]. These islets developed their size
maximally at 8-9 month fetus, and then reduced
gradually with development and _ remained
rudimentary until 3 years [1, 14]. Okajima [14]
also reported that the pancreatic islet of the adult

cow is usually the small type and is located in the
lobules. In the Japanese serow, the large islets
located in the lobule or interlobular connective
tissue revealed no apparent aging change in their
number from fawn to 10-year-old and showed a
similar population of four types of endocrine cells
in comparison with the small islets. Thus, the large
islets of the Japanese serow probably persist
through life.

REFERENCES

1 Bonner-Weir, S. and Like, A. A. (1980) Cell Tissue
Res., 206: 157-170.

2 Ehrhart, M., Grube, D., Bader, M. F., Aunis, D.
and Gratzl, M. (1986) J. Histochem. Cytochem., 34:
1673-1682.

3. Nakajima, S., Kitamura, N., Yamada, J., Yamashi-
ta, T. and Watanabe, T. (1988) Acta Anat., 131:
235-240.

4 Wakuri, H., Mutoh, K., Okajima, Y. and Akamat-
su, S. (1980) Kitasato Arch. Exp. Med., 53: 121-
11283.

5 Sugimura, M., Suzuki, Y., Kamiya, S. and Fujita,
T. (1981) Jpn. J. Vet. Sci., 43: 553-555.

6 Grube, D. and Bohn, R. (1983) Arch. Histol. Jap., -
46: 327-353.

7 Baskin, D. G., Gorray, K. C. and Fujimoto, W. Y.
(1984) Anat. Rec., 208: 567-578.

8 Kaung, H. C. (1985) Anat. Rec., 212: 292-300.

9 Girod, C., Durand, N. and Raccurt, M. (1987) Cell
Tissue Res., 247: 11-16.

10 Furuoka, H., Ito, H., Hamada, M., Suwa, T.,
Satoh, H. and Itakura, C. (1989) Jpn. J. Vet. Sci.,
51: 35-43.

11 Miyamoto, H., Mikasa, M. and Ishibashi, T. (1989)
Jpn. J. Zootech. Sci., 60: 255-260.

12 Edwin, N. (1987) J. Zool., 213: 665-671.

13 Reddy, S. and Elliott, R. B. (1984) Aust. J. Exp.
Biol. Med. Sci., 62: 239-244.

14 Okajima, Y. (1981) J. Vet. Med. (Tokyo), 720: 21-
Dale

sa AGH

|
i
|
|

obuiz

D !/ f Published Bimonthly by the Japanese Society of
eve opm en Developmental Biologists
Distributed by Business Center for Academic

Growth & Differentiation Societies Japan, Academic Press, Inc.

Papers in Vol. 32, No. 4. (August 1990)

44. G.N. Cherr, A. I. Yudin, and D. F. Katz: Organization of the hamster cumulus
extracellular matrix: a hyaluronate-glycoprotein gel which modulates sperm access to
the oocyte

45. J. Schilt, M. Delarue and M. Dauca: Immunolocalization of fibronectin and laminin
in the amphibian intestine during spontaneous and triiodothyronine-induced meta-
morphosis

46. H. Mohri, A. Fujiwara, M. Daumae and I. Yasumasu: Respiration and motility in
starfish spermatozoa at various times in breeding season

47. M. Matsuda: Fusion of neural folds in the rhombencephalic region of rat embryos

48. Y.Sakai and S. Yamashina: Spermiation in the mouse: Contribution of the invading
Sertoli cell process to adluminal displacement of the spermatid head

49. A. Mizutani and K. Yanagisawa: Cell-division inhibitor produced by a killer strain
of cellular slime mold, Polysphondylium pallidum

50. Y. Tahara and T. Morinaka: The blood island is a site of formation of the primary
embryo thrombocyte in the chick blastoderm

Abstracts of the papers presented at the 23th Annual Meeting of the Japanese Society of
Developmental Biologists, 1990

Development, Growth and Differentiation (ISSN 0012-1592) is published bimonthly by The Japanese Society of
Developmental Biologists, Department of Developmental Biology, Mitsubishi Kasei Institute of life Science,
Minami-ootani 11, Machida, Tokyo 194, Japan. 1989: Volume 31. Annual subscription for Vol. 32, 1990: U. S.$
148,00, U.S. and Canada: U.S. $ 163,00, all other countries except Japan. All prices include postage, handling
and air speed delivery except Japan. Second class postage paid at Jamaica, N.Y. 11431, U.S.A.

Outside Japan: Send subscription orders and notices of change of address to Academic Press, Inc., Journal
Subscription Fulfillment Department, 1 East First Street, Duluth, MN 55802, U.S. A. Send notices of change of
address at least 6-8 weeks in advance. Please include both old and new addresses. U.S. A. POSTMASTER:
Send changes of address to Development, Growth and Differentiation, Academic Press. Inc., Journal Subscription
Fulfillment Department, 1 East First Street, Duluth, MN 55802, U.S. A.

In Japan: Send nonmember subscription orders and notices of change of address to Business Center for
Academic Societies Japan, 16-3, Hongo 6-chome, Bunkyo-ku, Tokyo 113, Japan. Send inquiries about
membership to Business Center for Academic Societies Japan,4—16, Yayoi 2-chome, Bunkyo-ku, Tokyo 113, Japan.

Air freight and mailing in the U.S. A. by Publications Expediting, Inc., 200 Meacham Avenue, Elmont, NY
11003, U.S.A.

Sophisticated Balance between
Safety and Centrifugation
Capability without Compromise.

Centrifuge in
Integrated with A
Refrigerator

Extra-Quiet
Operation

Ease of Loading/
Unloading
The Rotors

Quick Start/
Quick Stop

High Quality

Triple Safety

Design
Corrosion
Resistance
E HIGH SPEED
wee’ REFRIGERATED
> MICRO CENTRIFUGE
vs move. MR-150
: é aia ‘ili sess aR Sins Aunu ante aaa SS as se : 4 2 |
TOMY CORPORATION SOLE AGENT | |

201 SOLEIL NARIMASU BLDG.. 31-8. NARIMASU 1-CHOME.
ITABASHI-KU. TOKYO 175 JAPAN

TEL:(03)976-341 1 WLEOZ7/ZSiI al WOMNKEO) J
CABLE: TOMYSHO TOKYO FAX:(GIII GII)(03)930-7010

TOMY SEIKO CO.,LTD. MANUFACTURER

2-2-12. ASAHICHO NERIMA-KU. TOKYO 176 JAPAN :
TEL:(03)976-3111 ee .

NARISHIGE

THE ULTIMATE NAME IN MICROMANIPULATION

OUR NEW MODELS WR-88 and MO-102M
MAKE PRECISION MICROMANIPULATION SO EASY!

SOME FEATURES of THE WR-88 WATER ROBOT MICROMANIPULATOR (3-DIMENSIONAL)
* Drift-free, the new WR-88 has a DRIFT movement of less than 2 microns.
x The new WR-88 has a SMOOTH MICRODRIVE MECHANISM.
* An Aqua Purificate remote control ensures totally vibration-free operation.

THARISHIGE) NARISHIGE SCIENTIFIC INSTRUMENT LAB.

(fee Sy, 9-28 KASUYA 4-CHOME SETAGAYA-KU, TOKYO 157, JAPAN
PHONE (INT-L) 81-308-8233, FAX (INT-L) 81-3-308-2005

CABLE : NARISHIGE LABO, TELEX, NARISHIGE J27781

(Contents continued from back cover)

of follicle-stimulating hormone and _ lutei-
nizing hormone with testicular receptors in
the bullfrog, Rana catesbeiana........... 705
Kobayashi, M. and N. E. Stacey: Effect of
ovariectomy and steroid hormone implanta-
tion on serum gonadotropin levels in female
PROMS MIM ee rel yes niate Sas fin staal o.0.se sain aie ales TAS)
Holmes, W. N., Cronshaw, J. and J. L. Redon-
do: Stress-induced adrenal steroidogenesis
in neonate mallard ducklings and domestic
eMC MSM iY ae cnet eats as be ties ole ee (23
Tomooka, Y., Edery, M., Mills, K. T., Bern,
H. A. and J. A. McLachlan: Effects of
androgen on mouse seminal vesicle epithelial
cells in serum-free culture
Koshimizu, I. and Y. Oota: A scanning elec-
tron microscopic study of the blood vascular
architecture of the snake hypophysis (COM-
MAINA 6 (©). | eee ee Hips

Morphology
Ohtsuki, H.: Inner structures of the cerebral
vesicle in the ascidian larva, Styela plicata: a
SIMestudyei ea skh oe ete ees ee ese als eS 739
Atoji, Y., Takayanagi, K., Suzuki, Y. and M.
Sugimura: Immunohistochemical demon-
stration of S-100 protein in the chick non-
NERVOUSAUSSUC ease oe tae sate ne os 747
Atoji, Y., Takada, Y., Suzuki, Y. and M.
Sugimura: Immunocytochemical identifica-
tion of four cell types in the pancreatic islats
of the Japanese serow, Capricornis crispus
(COMMUNICATION): . oc... 2225.5. 8: 779

Taxonomy
Yoshimura, K.: Two new species of the genus
Monhystrium Cobb, 1920 (Monhysteridae:
Nematoda) from terrestrial crabs of subfami-
ly Sesarminae (Brachyura) in Japan ...... 55

ZOOLOGICAL SCIENCE

VOLUME 7 NUMBER 4

AUGUST 1990

CONTENTS

REVIEWS

Koolman, J.: ., Ecdysteroids = <.9....2-.2.5-55 563

Yoshizaki, N.: Functions and properties of
animal lectins: <2. acxnvcadsosen nee 581

ORIGINAL PAPERS

Physiology
Lin, J. T., Toh, Y., Mizutani, M. and H.

Tateda: Putative neurotransmitter in the

ocellar neuropil of American cockroaches
Hagiters siels un hcl ec Wie Sa eee ee Sem gO een 593

Okamoto, K. and T. Tagawa: Aminergic,

cholinergic and peptidergic innervation of
hepatic portal vein in the anuran amphibians

Cell Biology
Waku, Y., Koike, M. and N. Yoshida: Cell
culture of the antennal imaginal disc of the
silkworm, Bombyx mori L. and differentia-
tion of neurons from the culture
Uchiyama, M., Yoshizawa, H., Wakasugi, C.
and C. Oguro: Structure of the internal
gills in tadpoles of the crab-eating frogs,
Rana cancrivora

Biochemistry
Tamanoi, I., Fujii, N., Muraoka, S., Harada,
K., Joshima, H. and T. Ishihara: The in-
duction of D-aspartic acid in mouse lens
protein by continuous gamma-irradiation

(COMMUNICATION) 2. k.2 2 ces ee 763
Immunology
Saad, A. H.: Estradiol-induced lymphopenia

in the lizard, Chalcides ocellatus

Nunomura, W., Watanabe, H. K. and H.
Hirai: Interaction of C-Reactive protein

INDEXED IN:
Current Contents/LS and AB & ES,
Science Citation Index,
ISI Online Database,
CABS Database, INFOBIB

with macrophages in rat (COMMUNICA-
TION)

Developmental Biology
Yang C.-H. and R. Yanagimachi: Changes in
the rigidity of the hamster egg during meiotic
maturation and after fertilization
Kanayama, M. and Y. Kamishima: Role of
symbiotic algae in hatching of gemmules of
the freshwater sponge, Radiospongilla cere-
bellata
Tacke, L. and H. Grunz: Effect of cytochala-
sin B, nocodazol and procaine on binding
and fate of concanavalin A in competent
ectoderm of Xenopus laevis
Tanaka, A. and M.H. Ross: Instability of the
number of segments of unoperated and re-
generated maxillary palpi in the maxillary-
palp-elongate (mpe) German cockroach
mutant

Reproductive Biology
Ueda, J., Hirano, T. and S. Fujimoto:
Changes in protein secretory patterns during
the development of the rat epididymis ...681
Ueda, H., Fukui, Y., Araki, H. and S. Fujimo-
to: Protein secretory patterns during the

development of the rat ovary ............ 691
Endocrinology
Endo, K., Fujimoto, Y., Masaki, T. and K.
Kumagai: Stage-dependent changes in the
activity of the prothoracicotropic hormone
(PTTH) in the brain of the Asian comma
butterfly, Polygonia c-aureum L. ......... 697

Yamanouchi, H. andS. Ishii: Effects of gonado-
tropin-specific antibodies on the interaction

(Contents continued on inside back cover)

Issued on August 15
Printed by Daigaku Letterpress Co., Ltd.,
Hiroshima, Japan

An International Journal

PHYSIOLOGY

CELL and MOLECULAR BIOLOGY
GENETICS

IMMUNOLOGY

BIOCHEMISTRY
DEVELOPMENTAL BIOLOGY
REPRODUCTIVE BIOLOGY
ENDOCRINOLOGY

BEHAVIOR BIOLOGY
ENVIRONMENTAL BIOLOGY and ECOLOGY
SYSTEMATICS and TAXONOMY

published by Zoological Society of Japan

eee | Business Center for Academic Societies Japan
istrnourec’Y vsp, Zeist, The Netherlands

ISSN 0289-0003

ZOOLOGICAL SCIENCE
The Official Journal of the Zoological Society of Japan

Editors-in-Chief: The Zoological Society of Japan:

Seiichiro Kawashima (Tokyo) Toshin-building, Hongo 2-27-2, Bunkyo-ku,

Hideshi Kobayashi (Tokyo) Tokyo 113, Japan. Tel. (03) 814-5675
Managing Editor: Officers:

Chitaru Oguro (Toyama) President: Hiromichi Morita (Fukuoka)
Assistant Editors: Secretary: Hideo Namiki (Tokyo)

Yuichi Sasayama (Toyama) Treasurer: Tadakazu Ohoka (Tokyo)

Hitoshi Michibata (Toyama)

Librarian: Masatsune Takeda (Tokyo)
Miéko Komatsu (Toyama) — a

Editorial Board:

Howard A. Bern (Berkeley) Walter Bock (New York) Aubrey Gorbman (Seattle)
Horst Grunz (Essen) Robert B. Hill (Kingston) Yukio Hiramoto (Chiba)
Susumu Ishii (Tokyo) Yukiaki Kuroda (Mishima) John M. Lawrence (Tampa)
Koscak Maruyama (Chiba) Roger Milkman (Iowa) Kazuo Moriwaki (Mishima)
Richard S. Nishioka (Berkeley) Tokindo S. Okada (Okazaki) Andreas Oksche (Giessen)
Hidemi Sato (Nagano) Hiroshi Watanabe (Tokyo) Mayumi Yamada (Sapporo)

Ryuzo Yanagimachi (Honolulu)

ZOOLOGICAL SCIENCE is devoted to publication of original articles, reviews and communications
in the broad field of Zoology. The journal was founded in 1984 as a result of unification of Zoological —
Magazine (1888-1983) and Annotationes Zoologicae Japonenses (1897-1983), the former official
journals of the Zoological Society of Japan. ZOOLOGICAL SCIENCE appears bimonthly. An annual
volume consists of six numbers of more than 1200 pages including an issue containing abstracts of papers
presented at the annual meeting of the Zoological Society of Japan.

MANUSCRIPTS OFFERED FOR CONSIDERATION AND CORRESPONDENCE CONCERN-
ING EDITORIAL MATTERS should be sent to:
Dr. Chitaru Oguro, Managing Editor, Zoological Science, Department of Biology, Faculty of Science,
Toyama University, Toyama 930, Japan, in accordance with the instructions to authors which appear in
the first issue of each volume. Copies of instructions to authors will be sent upon request.

SUBSCRIPTIONS. ZOOLOGICAL SCIENCE is distributed free of charge to the members, both
domestic and foreign, of the Zoological Society of Japan. To non-member subscribers within Japan, it is
distributed by Business Center for Academic Societies Japan, 6—16-3 Hongo, Bunkyo-ku, Tokyo 113.
Subscriptions outside Japan should be ordered from the sole agent, VSP, Utrechtseweg 62, 3704 HE
Zeist (postal address: P.O. Box 346, 3700 AH Zeist), The Netherlands. Subscription rates will be
provided on request to these agents. New subscriptions and renewals begin with the first issue of the
current volume.

All rights reserved. No part of this publication may be reproduced or stored in a retrieval system in any
form or by any means, without permission in writing from the copyright holder.
© Copyright 1990, The Zoological Society of Japan

Publication of Zoological Science has been supported in part by a Grant-in-Aid for
Publication of Scientific Research Results from the Ministry of Education, Science
and Culture, Japan.

ZOOLOGICAL SCIENCE 7: 783-799 (1990)

© 1990 Zoological Society of Japan

REVIEW

Properties of the Blood-Brain and Blood-Cerebrospinal
Fluid Barrier in the Circumventricular Organs of
the Diencephalic Roof of Teleosts

Yuri Omura!”, Horst-W. KorF’, Mikio Ocurt’

and ANDREAS OKSCHE>

"Laboratory of Fish Biology, Faculty of Agriculture, Nagoya University,
Nagoya 464, Japan, and *Department of Anatomy and Cell Biology,
Justus Liebig University, D-6300 Giessen,

Federal Republic of Germany

INTRODUCTION

In most parts of the brain a blood-brain barrier
is established by tight junctions sealing off the
endothelial cells of the brain capillaries [1-5]. In
addition to these tight junctions, perivascular con-
nective tissue elements, such as collagen fibrils and
macrophages, are also to be considered as structu-
ral components of an effective barrier between the
hemal environment and the cerebrospinal fluid
(CSF), especially in the circumventricular organs
lacking the features of a classical blood-brain bar-
rier [6-8].

Although intravenously administered hydrophil-
ic tracers do not cross the endothelium of regular
brain capillaries [1, 2, 9], they penetrate into the
choroid plexus via the fenestrated capillaries of
this circumventricular organ, and are rapidly in-
corporated into membrane-bounded vesicles lo-
cated in the epithelial cytoplasm [6, 10, 11]. Such
epithelial endocytosis followed by lysosomal se-
questration is considered to be involved not only in
transepithelial transport but also in a mechanism
for intracellular degradation of foreign substances
[4, 10].

By means of intraventricular injection of tracers,
such as India ink, ferritin, horseradish peroxidase

Received April 13, 1990
* To whom all correspondence should be addressed.

(HRP) and latex beads, it is established that a
larger part of motile supraependymal cells, first
described by Kolmer [12], take up such tracer
particles, thus cleaning the ventricular CSF [4, 13-
15]. This leads to the assumption that supraepen-
dymal macrophages may be considered as motile
phagocytes or scavengers and play an active role in
regulating the CSF compartment and associated
CSF-contacting tissues [see 15-17].

The diencephalic roof of teleosts is predomi-
nantly occupied by the photoneuroendocrine
pineal organ [see 17-20], the choroidal dorsal sac
[see 21-24], and the functionally still enigmatic
parapineal organ [see 25-28] (Fig. 1). The former
two circumventricular organs are in open com-
munication with the third ventricle; furthermore,
they are well vascularized by means of numerous
capillaries endowed with fenestrated endothelial
cells [see 17, 23]. These peculiar features of the
diencephalic roof prompted us to investigate the
properties of the barrier between the hemai en-
vironment and the CSF-dominated compartment
in this region in greater detail.

In accordance with our previous and, in addi-
tion, unpublished results, at least three or four
specialized structural components or tissue com-
partments can be proposed to participate in estab-
lishing a blood-CSF barrier in the region of the
diencephalic roof of teleosts: (1) a system of
luminal or circumventricular tight junctions join-

784 Y. Omura, H.-W. Kors et al.

Fic. 1.

Cre

ee = 2S ) Lr

np 4. ae a

Diagrammatic representation of a sagittal section through the telencephalic and diencephalic roof of the

rainbow trout. B bony cranium; C chondrocranium; CH habenular commissure; CP posterior commissure; D
skin; M muscle; NH habenular nucleus; P pineal organ; PP parapineal organ; PY paraphysis; SCO subcommissu-
ral organ; SD dorsal sac; T telencephalon; Th thalamus; TO optic tectum; VT velum transversum; III third

ventricle.

ing adjacent epithelial cells; (2) a system of special-
ized perivascular connective-tissue elements; (3) a
heterolytic system of endocytotic vesicles and lyso-
somes within epithelial cells; and (4) a system of
macrophages inhabiting the luminal surface and
the ventricular compartment.

This review deals mainly with the putative prop-
erties of the blood-CSF barrier in the epithalamic
circumventricular organs of teleosts and analyzes
them in context with the general concept of the
blood-brain barrier.

I. Topographic and Functional Aspects

According to the terminology proposed by
Holmgren [29], the epithalamic area of teleosts is
separated from the telencephalic paraphysis by the
velum transversum; the former is composed of the
pineal and parapineal organs, the dorsal sac and
the subcommissural organ. The present review of
the diencephalic roof of teleosts is focussed on the
pineal and parapineal organs and the dorsal sac. In
contrast, the subcommissural organ, with particu-
lar reference to its location beneath the posterior

commissure, can be regarded as a mesencephalic,
pretectal or tectal structure, at least due to topo-
graphic criteria.

Pineal organ

During the development of teleosts, the poste-
rior epiphyseal evagination, i.e. the pineal organ,
protrudes from the diencephalic roof and extends
its end-vesicle over the telencephalon, while the
anterior evagination, i.e. the parapineal organ,
slightly turns left and remains in situ. In the
well-developed pineal organ, the epithelial paren-
chyma is composed of (i) typical and modified
photoreceptor cells, (ii) uni-, bi- and multipolar
nerve cells, and (iii) CSF-exposed supporting cells
[20, 28, 30-32]. The pineal nerve projects via the
stalk and the posterior commissure to various
brain centers [33-35]. Accumulating evidence
suggests that the pineal nerve is composed not only
of the axons of intrapineal nerve cells but also of
the axonal processes of pineal photoreceptors and/
or CSF-contacting neurons [36].

In the dark-adapted state, the spontaneous dis-
charge from pineal nerve cells is inhibited by light

Blood-Brain Barrier in Fish Epithalamus 785

stimulation; the membrane potentials from pineal
photoreceptor cells are hyperpolarized in response
to flashes of light [37-39]. Measurement of pineal
and plasma melatonin levels by RIA suggested
that the pineal organs of the carp and the pike also
elaborate and release melatonin into the blood
during the darkness [40, 41]. In addition, in-
traperitoneal injection of radioactive indoleamines
and macromolecular proteins revealed that the
pineal organ of the rainbow trout was capable of
rapid uptake of these exogenous substances from
the blood [17, 30, 42].

Although a well-developed vascular system sur-
rounding the epithelial wall of the diencephalic
roof has been described in the rainbow trout [17,
23, 34], until recently there was a lack of exact
information concerning the arterial supply and the
venous drainage in this area. The vascular rela-
tionships between the pineal organ and the dorsal
sac appear to be very complex. In an early study
on these relationships this vascular system was
examined in various teleost species by Friedrich-
Freksa [43]. For example, the pineal organ of the
goby receives bilateral branches from the arteria
pinealis anterior, which form the capillary network
within the end-vesicle, and sends a vena pinealis
via the pineal stalk into the “choroid plexus” (see
next paragraph). Syed Ali et al. [44] recently
studied the pineal vascular system of the rainbow
trout by means of Mercox injections and subse-
quent scanning electron microscopy. According to
their data, there are two different circulatory path-
ways: (1) blood vessels ascending and descending
via the pineal stalk; (2) vascular supply emanating
from both (left and right) lateral aspects and
joining to the former circuit in the pineal end-
vesicle.

In mammals, e.g. in the laboratory rat, the
major blood supply to the pineal organ is provided
by arterial branches from the posterior cerebral
arteries, while the venous drainage is directed into
the great cerebral vein, from which the choroid
plexus of the third ventricle straightly receives the
perfusate from the pineal organ [45, 46]. Con-
sidering such general vascular pattern, it may be
that also in the rainbow trout the pineal secretory
products are transported via a local vascular path-
way into the dorsal sac, and thus gain access to the

ventricular CSF and the brain tissue.

Parapineal organ

In spite of the earlier assumption that the para-
pineal organ is regressed in adult teleosts [see 29],
this tiny and enigmatic organ has been found in
more than 40 species of teleosts [25-28, 29, 47-
52]. The parapineal organ of teleosts has been
considered to be homologous to the parietal eye of
lizards and to the parapineal organ of petromyzon-
ids. The function of the parapineal organ is,
however, still open to discussion, although a lim-
ited number of outer segments of photoreceptor
cells was demonstrated in this vesicular body by
electron microscopy [25, 27] and immunohistoche-
mical use of antibodies against rod-opsin [51, 52].

On the basis of their observations in a large
number of teleost species, Borg et al. [27] con-
cluded that the parapineal organ is variable in size
and location; in some cases it was difficult to
distinguish this organ from the tissue of the adja-
cent habenular nucleus. The parapineal organ of
the rainbow trout, as mentioned elsewhere [28], is
located left to the pineal stalk and dorso-posterior
to the habenular commissure, through which the
parapineal tract penetrates into the left habenular
nucleus. The character of the neural projections of
the parapineal organ is still unproven, although a
close relation to the pineal tract has been indicated
using cobalt-chloride iontophoresis [33].

As mentioned above, opsin-immunoreactive
structures resembling outer segments of photo-
receptors were occasionally found in the para-
pineal organ of several teleost species [51, 52].
According to Ekstrém et al. [36], immunoreactivi-
ties for photoreceptor-specific antigens, such as
opsin and S-antigen, were found not only in a cell
population situated in the parapineal organ but
also within a reasonable number of cells scattered
in the habenular nucleus. This accumulating evi-
dence, thus, prompts us to reflect on the problem
whether the parapineal organ of teleosts might
originate from the habenular nucleus. However,
the functional significance of the parapineal organ
still remains enigmatic. Further studies are re-
quired to elucidate the interrelationships between
the parapineal organ and the habenular nucleus in
teleosts.

786 Y. Omura, H.-W. Kore et al.

Dorsal sac

The diversity in the pattern of ependymal struc-
tures in the teleost epithalamus has been investi-
gated in a large number of species [43, 53, 54].
According to Tsuneki [54], there appear to be two
major types of evolutionary differentiation of the
epithalamic epithelium: (1) the type characterized
by a dorsal sac, and (2) the type leading to the
formation of a choroid plexus. The latter type has
been described only in a limited number of teleost
species [43, 54]. Meanwhile, there is growing
evidence that the dorsal sac of teleosts is homolo-
gous to the choroid plexus of other vertebrates
including mammals [22, 24]. It has been assumed
that, in teleosts, the dorsal sac and the diencephal-
ic choroid plexus may replace each other [43].
Thus, the dorsal sac may have some principal
functions in common with the choroid plexus; it is
generally accepted that the choroid plexus is pri-
marily concerned with the formation of the cere-
brospinal fluid [see 55, 56].

High synthetic activity and protein absorbing
capacity of the dorsal sac have been demonstrated
by use of tracers [22], and by means of enzyme
cytochemistry [24]. Melatonin, the principal secre-
tory product of the pineal organ, has been shown
to occur in the ventricular cerebrospinal fluid,
where it exhibits a marked circadian rhythm [57].
If pineal melatonin, in the rainbow trout, is direct-
ly released into the luminal cerebrospinal fluid via
the open communication with the third ventricle, it
may become highly effective. It is also possible
that the secretory product of the pineal organ
enters the ventricular system via the epithelial wall
of the dorsal sac immediately after its discharge
into the blood of the common vascular system.

II. Experimental Analysis of the Blood-Brain
Barrier in the Diencephalic Roof of Teleosts

Light microscopy

Following intraperitoneal injection of trypan
blue and HRP, the pineal organ, the dorsal sac and
the meninges of the rainbow trout were conspic-
uously stained with the respective tracers (Fig. 2).
The pineal organ, especially the proximal portion

1mm

Fic. 2. Lateral view of the brain of a rainbow trout 8 hr
after intraperitoneal injection of HRP. The tracer is
accumulated predominantly in the pineal organ
(arrow) and moderately in the dorsal sac (arrow-
head) and the meninges. x9.

of its end-vesicle was completely surrounded by
the folded epithelium of the dorsal sac (Fig. 3A).
The well-developed vascular supply, formed by
capillaries of differing diameter, was embedded in
the connective-tissue stroma encompassing the
pineal organ and the dorsal sac (Fig. 3A, B). The
epithelial wall of the sac, thus, seems to share its
vascular supply with the pineal vascular system.
The close proximity of the dorsal sac to the pineal
organ prompts us to consider a functional rela-
tionship between these two diencephalic evagina-
tions of the teleost brain. However, the tentative
functions, especially with regard to secretory inter-
relationships, are largely speculative.

The pineal lumen harbored numerous mac-
rophages, which showed a conspicuous uptake of
intraperitoneally injected HRP (Fig. 3B, C). Trac-
er studies of the dorsal sac also revealed consider-
able numbers of macrophages scattered in close
contact with the apical brush border of the epithe-
lium (Fig. 3D). Comparative studies concerning
the occurrence of ventricular macrophages or sup-
raependymal cells have shown that these cells
appear most abundantly in amphibians, reptiles
and mammals, but less commonly in fishes and
birds [12]. However, our results indicate that a
system of ventricular macrophages, including mac-
rophages in the pineal lumen, and the so-called
supraependymal cells in the dorsal sac and the
epithalamic portion of the third ventricle, are well

Blood-Brain Barrier in Fish Epithalamus 787

Fic. 3. Light micrographs of the diencephalic roof of the rainbow trout. A. Cross-sectioned proximal portion of the
pineal end vesicle (P) fully enveloped by the folded epithelium of the dorsal sac. Numerous free cells (small
arrows) are scattered all over the epithelial brush border. B. Cross section of the pineal stalk (P) embedded in
folded structures of the dorsal sac (SD). Irregularly shaped macrophages (small arrows) in the pineal lumen.
Note numerous capillaries (asterisks) of differing caliber in the connective-tissue stroma between the dorsal sac
and the pineal organ in A and B. C-. Cross section of the pineal end-vesicle 8 hr after injection of HRP. Note
macrophages (arrows) containing HRP-positive granules in the lumen (pl); cl capillary lumen. D. Horizontal
section through the proximal portion of the pineal stalk (P) 24 hr after injection of HRP. Note macrophages
(arrows) containing HRP-granules on the brush border of folded ependymal epithelium underlain by numerous
capillaries (asterisks). III third ventricle. A160, Bx245, Cx 440, Dx 220.

Y. Omura, H.-W. Kore et al.

788

100 um

PP

ee
<<

Ay
et
‘

i

Blood-Brain Barrier in Fish Epithalamus 789

developed in teleosts.

The parapineal organ was found left to the
pineal stalk and dorso-posterior to the habenular
commissure, into which the parapineal tract en-
tered (Fig. 4A). The capillary network around the
parapineal organ appeared to be very scarce. Ex-
cept the basal portion of the capsule and the
surroundings of some capillaries, the parapineal
parenchyma and tract seemed to be almost free of
HRP-accumulations in frozen sections (Fig. 4B,
C). In contrast, the epithelial tissues of the pineal
organ and the dorsal sac from the same specimens
were densely impregnated with HRP-positive
particles, especially 24 hr after intraperitoneal in-
jection of the tracer (Fig. 4D, E). Consequently,
the parapineal organ of the rainbow trout resem-
bles in principle the brain tissues possessing a
blood-brain barrier, which protects them from
exogenous, especially noxious agents.

Electron microscopy

Following intraperitoneal injection, vesicular
endocytosis and lysosomal sequestration of HRP
were finally found in all types of epithelial cells of
the pineal organ and the dorsal sac (Fig. 5A, B, C)
[see also 17]; in contrast, no trace of this material
was observed in the parenchyma of the parapineal
organ (Fig. 6A, B). The accumulation of HRP in
the pineal organ and the dorsal sac appeared to be
primarily related to the existence of fenestrated
capillaries in these regions (Fig. 7A). This holds
also true for the pineal organ of several rodents
[58, 59]. In contrast, no endothelial fenestrae or
pores of the vascular wall were found in the
parapineal organ.

In addition to the macrophages inhabiting the
perivascular spaces, the perivascular fibroblasts

and the capillary endothelial cells exhibited a
distinct lysosomal sequestration of HRP in the
pineal organ and the dorsal sac (Fig. 5C) [see 17].
Moreover, the thin-layered basal lamina and the
interlaced bundles of collagen fibrils surrounding
the pineal organ and the dorsal sac were densely
stained by tannic acid impregnation (Fig. 7B). In
contrast, only an attenuated layer of HRP-reactive
material appeared to occur around the parapineal
organ (Fig.6A, B). This deposit of HRP was
revealed to be mainly related to interlaced bundles
of collagen fibrils.

Intercellular passage of tannic acid was pre-
vented by the tight junctions between the epithe-
lial cells of the pineal organ and the dorsal sac (Fig.
5A, B) [see 17]. Nevertheless, there was substan-
tial evidence speaking in favor of a transepithelial
transport of HRP particles. Macrophages indicat-
ing lysosomal sequestration of HRP were found
not only in the perivascular space and the paren-
chymal epithelium but also in the central lumen of
the pineal organ and along the apical microvilli of
the epithelial cells of the dorsal sac (Fig. 8A). A
number of HRP-labeled vesicles appeared very
close to the apical surface or among the irregularly
shaped microvilli of the epithelial cells.

Despite of a few images indicating an exocytotic
discharge of HRP and ferritin particles into the
ventricular lumen, the macrophages contacting the
epithelial microvilli mainly exhibited signs of
vesicular and lysosomal sequestration of the tracer
particles (Fig. 8A, B). Consequently, these cells
may be involved in an effective barrier mechanism
against exogenous and noxious substances invad-
ing the area of the diencephalic roof and the
epithalamic portion of the third ventricle in the
rainbow trout.

Fic. 4. Light micrographs comparing the parapineal organ with the pineal organ and the dorsal sac in the rainbow
trout. A. Horizontal section immediately dorsal to the opening of the pineal lumen (arrowhead) into the third
ventricle (III). Nerve tract (arrows) leaving the parapineal organ (PP) and joining the habenular commissure
(CH). Asterisks mark capillaries; NH habenular nucleus; P pineal stalk. B and C. Cross-sectioned epithalamic

area 24hr after HRP-injection (frozen sections).

Parenchymal tissue of the parapineal organ free of

HRP-positive material, in accord with the subcommissural ependyma (SCO) and the adjacent brain tissue.
Asterisks indicate capillaries. D and E. Cross-sectioned pineal end-vesicle (P) and dorsal sac (SD) 8 hr (D) or 24
hr (E) after HRP injection (frozen sections). Note macrophages (arrows) displaying strong HRP-reactivity in the

pineal lumen and on the luminal surface of the dorsal-sac epithelium.

Pineal parenchyma contains fine

HRP-positive granules in E. A320, BxX155, Cx315, Dx95, Ex 160.

790 Y. Omura, H.-W. Kore et al.

Fic. 5. Electron micrographs showing the dorsal sac 24 hr after intraperitoneal injection of HRP. A and B. Apical
portion of the dorsal-sac epithelium. Note HRP-positive granules (arrowheads) near the luminal pole of the
epithelial cells (Ep). Arrows point to tight junctions between the epithelial cells; c cilium; mv microvillus. C.
Basal portion of the epithelium. Note secondary lysosomes (arrows) containing tracer particles in macrophage
(Mc) and parenchymal epithelial cells; cl capillary lumen; pvs perivascular space. AX9,900, BX33,400,
Cx 6,900.

Blood-Brain Barrier in Fish Epithalamus 791

Fic. 6. Electron micrographs of the parapineal organ 24 hr after intraperitoneal injection of HRP. A. Peripheral
portion of the parapineal organ. Note the epithelial parenchyma free of tracer particles. Interdigitating processes
of capsular fibroblasts (Fb) extend between parapineal parenchyma and vascular lumen (cl). Arrows point to
secondary lysosomes containing HRP; n nucleus of the parenchymal cell. B. Higher magnificathetion of the
capsular connective tissue. Note bundles of collagen fibrils (small arrows) loaded with HRP close to basal lamina

covering epithelial cells. A 4,200, B x 13,800.

III. Comparative Aspects of the
Blood-Brain Barrier

General aspects

The blood-brain barrier is essential for the
maintenance of the homeostasis of the cerebro-
spinal fluid, and to prevent the entrance of noxious
agents from the blood into the brain [5, 60].

Accordingly, it might be expected that the blood-
brain barrier is a general feature of the central
nervous system in all classes of vertebrates [5].
However, there are some contradictory reports
showing that the blood-brain barrier in some spe-
cies of cyclostomes and elasmobranchs differs in its
structure and function from the respective barrier
in mammals [61-63].

Despite numerous studies using various species

792 Y. Omura, H.-W. Kore et al.

Suen RAS.
: "i

AEROSOL
SES SGN :

Fic. 7. Electron micrographs showing the basal portion of the pineal epithelium. A. 8hr after intraperitoneal
injection of HRP. Large numbers of endocytotic vesicles (small arrows) in the basal portion of supporting cells.
Large arrows mark fenestrated endothelium of the vascular system. B. Basal portion of pineal epithelium after
exposure to tannic acid. Perivascular collagen fibrils (small arrows) and basal lamina (arrowheads) impregnated
with tannic acid; cl capillary lumen; Ed endothelial cell; pvs perivasular space; S supporting cell. AX 17,700,
B X 15,100.

Fic. 8. Electron micrographs showing luminal macrophages (Mc) of the dorsal sac after intraperitoneal injection of
HRP or ferritin. A. 24 hr after intraperitoneal injection of HRP. HRP-particles sequestered in exocytotic and
endocytotic vesicles (arrows) and secondary lysosomes (ly). B. Endocytotic uptake of ferritin particles by a
luminal macrophage. Part of macrophage in contact (small arrows) with apical microvilli of an epithelial cell. C.
Enlarged rectangle in B. Arrows point to endocytotic vesicles containing ferritin particles; mv microvillus; n
nucleus; v vacuole. AX10,200, Bx6,900, Cx 25,800.

793

Fish Epithalamus

1er 1n

Blood-Brain Barr

794 Y. Omura, H.-W. Kore et al.

of vertebrates, there is still a lack of information
concerning the system of the blood-brain barrier in
teleosts. In a previous paper [17] we were able to
demonstrate that in the pineal organ of the rain-
bow trout a barrier for exogenously administered
proteins, e.g. ferritin and HRP, does not exist;
these molecules were rapidly taken up by the
pineal epithelial cells, actively conveyed in a trans-
cellular manner and finally released into the CSF
filling the pineal lumen [17]. In addition, the
existence of fenestrated capillaries in the pineal
organ of the rainbow trout was also proven. In
principle, the lack of a blood-pineal barrier has
been attributed to the leaky vasculature of this
neuroendocrine organ: (1) the presence of fenes-
trated endothelium [see 58, 59, 64, 65] or (2) the
existence of discontinuous junctions between the
endothelial cells [66].

Tight junctions

As mentioned elsewhere [see 1, 2], continuous
tight junctions between adjacent endothelial cells
of cerebral capillaries are the anatomical substrate
of the blood-brain barrier, separating the hemal
environment from the brain parenchyma. This
holds true also for the rainbow trout. In this
teleost species the intercellular passage of tannic
acid was blocked at the luminal junctional complex
of the pineal organ and the dorsal sac [17].

However, there seems to be an intercellular
pathway for passive ion permeation via the tight
junctions between adjacent choroidal epithelial
cells. Electron-microscopic tracer studies revealed
that the tight junctions of choroidal epithelia were
permeated by lanthanum ions, whereas those of
cerebral capillaries significantly blocked the pas-
sive ionic permeation [67, 68]. According to
ultrastructural freeze-fracture studies, the ‘very
tight’ junctions displayed a large number of inter-
connected junctional ridges arranged in series [69].
In contrast, the tight junctions of choroidal epithe-
lia appeared to be a system of interconnected
ridges displaying discontinuities, which was inter-
preted to increase the permeability of the choroi-
dal epithelium [4, 10]. Other studies, however,
have shown that the discontinuities of intercon-
nected junctional ridges are always complemented
by particles or bars [70]. Considering such con-

troversial data, additional experiments by means
of different tracers with smaller molecular weights
and complementary studies by use of the freeze-
fracture technique are required to verify the ‘tight-
ness’ of the tight junctions between the epithelial
cells in the pineal organ and the dorsal sac of
teleosts.

Perivascular connective-tissue elements

According to Brightman [6], a portion of the
HRP particles administered intravenously and pass-
ing the leaky capillaries were found to be affixed
to the collagen fibrils within the perivascular con-
nective-tissue stroma, and appreciable amounts of
these particles moved across the stromal space to
penetrate the epithelial basal lamina of the choroid
plexus. Furthermore, Krisch et al. [8] observed in
the median eminence and the cortex of rats that
bundles of collagen fibrils and numerous phago-
cytes in the leptomeningeal intercellular clefts
were labeled with HRP particles 5 min after intra-
cerebral injection of the tracer. With respect to
these findings, the collagen fibrils may play the role
of a sieve or trap for penetrating exogenous sub-
stances at a rather early stage following systemic
administration, whereas the cellular components
of the connective-tissue stroma, such as mac-
rophages and fibroblasts, may be involved in the
final sequestration and/or heterolysis of the nox-
ious agents. The immuno-phagocytotic involve-
ment of the macrophages inhabiting the perivascu-
lar spaces has been suggested elesewhere [see 7,
17,74).

Endocytotic uptake and lysosomal sequestration

Since the pioneer work by Wislocki and Leduc
[72] the circumventricular organs including the
pineal organ and the choroid plexus have been
considered to represent areas lacking a blood-
brain barrier. However, Dretzki [73] questioned
this assumption, and proposed that in these areas a
partial barrier effect exists between the blood and
the brain parenchyma. In functional terms, this
also holds true for the pineal organ and the dorsal
sac of the rainbow trout. Following intraperitoneal
injection of HRP, a number of dense bodies
containing the tracer appeared in the apical por-
tion of the epithelial cells in these two circumven-

Blood-Brain Barrier in Fish Epithalamus 195

tricular organs of the rainbow trout [17, present
findings]. This accumulation of dense bodies was
correlated with the intracellular demonstration of
acid phosphatase activity, the classical marker
enzyme for lysosomes, in the pineal epithelial cells
[71]. Thus, a lysosomal sequestration and de-
gradation takes place following the endocytotic
uptake of the tracer into the epithelial cells of the
pineal organ and the dorsal sac. This mechanism
exerted by the epithelial cells appears to represent
one of essential components of the effective bar-
rier between the blood- and the CSF-dominated
compartments in the diencephalic roof of teleosts.
The analogy in structure and function between
the dorsal sac of teleosts and the choroid plexus of
mammals was already discussed above. Intra-
venously injected macromolecules, such as HRP,
cytochrome c and microperoxidase, were endocy-
tosed by pinocytotic vesicles at the basolateral
surface of the choroidal epithelium, and thereafter
numerous multivesicular and dense bodies labeled
with these tracers were observed in the apical
portion of the epithelial cells [4, 6, 10, 74]. After
administration of cytochrome c, increase in acid
phosphatase activity was demonstrated at the same
intracellular sites as the tracer activity [10].
Moreover, pinocytotic uptake of HRP and micro-
peroxidase from the CSF and subsequent lyso-
some-bound sequestration were demonstrated in
the choroidal epithelial cells [4, 11]. Such en-
docytotic uptake followed by lysosomal sequestra-
tion in the choroidal epithelium has been consi-
dered to be involved in establishing an effective
blood-CSF or CSF-blood barrier [4, 6, 10].

Transepithelial transport

In contrast to the epithelial endocytotic uptake
of macromolecules from the bloodstream, the exo-
cytotic discharge of the tracers from the apical
surface only occasionally has been observed in the
choroidal epithelium [see 4, 6, 10]. Our results
speak in favor of the existence of a transepithelial
vesicular transport of macromolecular proteins
(ferritin, HRP) in the pineal organ and the dorsal
sac of the rainbow trout; rapid and also compara-
tively slow transepithelial transport from blood to
CSF appears to exist in the two epithalamic cir-
cumventricular organs of the rainbow trout.

The functional significance of this type of trans-
epithelial transport is still open to discussion.
Probably, this transport mechanism may serve to
shift nutrients and biological agents to the CSF,
under participation of the free cells inhabiting the
ventricle, and finally—via an ependymal uptake
mechanism— also to target sites in the brain
tissue. For example, 3H-labeled melatonin, readi-
ly penetrated into the brain after intravenous or
intraventricular injection and selectively concen-
trated within the hypothalamus and the midbrain
[75]. In mammals, using '*°I-melatonin and in
vitro autoradiography the specific melatonin-
binding sites have been identified to occur in the
suprachiasmatic nucleus, median eminence, pars
tuberalis etc. [76, 77]. It has been controversial
whether pineal melatonin is secreted directly into
the third ventricle or actively transported into the
CSF from the blood [47, 59, 64, 78, 79].

In mammals, on the other hand, an uptake from
the CSF followed by intracellular transport into
the blood, i.e. CSF-blood transepithelial transport
via the choroidal epithelium, has been demon-
strated by means of intraventriculo-cisternal injec-
tion of macromolecular tracers [4]. In addition,
the circumventricular organs including the pineal
organ and the choroid plexus have been consid-
ered as a main route for drainage of the CSF
toward the venous system [80, 81]. In order to
maintain a constant CSF milieu, mammals may
employ both blood-CSF and CSF-blood trans-
epithelial transport mechanisms. Whether a CSF-
blood transepithelial transport route exists in the
brain of teleosts has to be further investigated.

Ventricular macrophages

Following systemic administration, the trans-
epithelially conveyed HRP particles finally appear
in secondary lysosomes of the ventricular mac-
rophages. These cells may be involved in an
effective barrier mechanism against exogenous and
noxious substances invading the area of the
diencephalic roof and the epithalamic portion of
the third ventricle in the rainbow trout. In order
to elucidate the phagocytotic role of the supra-
ependymal cells of the choroid plexus, different
tracers have been directly administered into the
ventricular system: India ink in the axolotl and

796 Y. Omura, H.-W. Kore et al.

guinea pig [13]; India ink, thorotrast and ferritinin scavenging debris from ependymal metabolism,
the cat [14]; latex beads in the tegu lizard [15]; _ and (ii) in phagocytosis of noxious foreign parti-
HRP in the rat [4]. Since such particles were cles, e.g. viruses and bacteria, invading the CSF
ingested rapidly by most supraependymal cells, _[15, 16].

they were considered to be engaged (i) in Although systemic administration of exogenous

Ce e &
on
2) 9} \o
a fe) I 10)
@
a)
tjQ
\
o Of\o_o es NAN SLEL TIES DARA RL EARLE SSS SCS coc moe
we A IRN BOE De ay SRT AES AIA NS E ° =o 3
a ROI DU ce oa ree)
a8 e ee

nee ee el ee} °
fi ° 999 0% 5 oO OCR GF euleKaleataleshone ne eee
°

eo 0@8, 6
eooee® aay
ooo

b|

Fic. 9. Diagrammatic representation illustrating the properties of the blood-brain barrier in the diencephalic roof of
the rainbow trout: (for details, see Conclusions). Large arrows indicate endothelial fenestrae. Small arrows point
to pinocytotic vesicles involved in endocytosis and/or exocytosis of the epithelial cells and macrophages; bl basal
lamina; cf collagen fibril; cl capillary lumen; CSF luminal space filled with cerebrospinal fluid; Ed endothelial cell,
Ep epithelial cell; Fb fibroblast; ly lysosome; Mc macrophage; tj tight junction; v vacuole.

Blood-Brain Barrier in Fish Epithalamus

material results in a numerical increase of ventricu-
lar macrophages in the epithalamic portion of the
third ventricle of the rainbow trout, there is still a
lack of knowledge concerning the origin and fate
of the vetricular macrophages. According to Kol-
mer [12], the supraependymal cells are of hema-
togenous origin, migrating from the choroidal
blood vessels, possibly in response to some specific
stimulus, into the CSF. Later this concept was
modified by Carpenter et al. [14]: monocytes from
the bloodstream reach the perivascular stroma via
the thin-walled choroidal capillaries; some of them
remain in situ as tissue macrophages, others reach
the ventricle via the choroidal epithelium (inter-
epithelial route) to become supraependymal cells.
Actually, an invasion of circulating blood mono-
cytes into the neural parenchyma has been demon-
strated by use of the colloidal carbon technique
[82]. _ Considering some recent immunohisto-
chemical findings on brain macrophages [see 83,
84], the ventricular macrophages may originate
from the subarachnoidal vessels which extend into
the stroma of the choroid plexus, but not from the
microglial cells of the brain.

CONCLUSIONS

As mentioned elsewhere [see 17], the per-
meability of the blood-brain barrier of the circum-
ventricular organs has been investigated at the
light- and electron-microscopic levels with the use
of different tracers. The capillaries of most of
these organs, except the subcommissural organ,
are endowed with fenestrated endothelia; such
areas are exposed to the hemal environment.
Therefore, it is generally accepted that these cir-
cumventricular organs are devoid of a blood-brain
barrier. This review, mainly based on our previous
studies and unpublished results on the blood-brain
barrier in the structural differentiations of the
diencephalic roof of the rainbow trout, emphasizes
that in spite of the leaky capillaries an effective
barrier system may exist between the hemal en-
vironment and the ventricular cerebrospinal fluid
compartment in this area. This blood-CSF barrier
may involve the following components (see Fig.
9): (1) a sieve or trap system of perivascular
‘stromal’ connective-tissue elements (collagen

797

fibrils, macrophages, fibroblasts); (2) a system of
circumventricular tight junctions adjoining adja-
cent epithelial cells; (3) a heterolytic system of
epithelial endocytosis followed by lysosomal se-
questration; and (4) a system of macrophages
inhabiting the luminal and ventricular compart-
ments. In contrast to the pineal organ and the
dorsal sac, the parapineal organ of teleosts may be
located ‘inside’ the blood-barrier.

ACKNOWLEDGMENTS

The authors thank Ms. K. Koga and Ms. G. Doll for
their efficient technical assistance. They are also grateful
to Dr. S. Fushiki, Shiga Prefectural Samegai Trout Farm,
for providing the rainbow trout. This work was sup-
ported by grants from the Ministry of Education, Science
and Culture of Japan (M.O., Y.O. 61560210), the Alex-
ander van Humboldt Foundation(Y.O.), andthe Deutsche
Forschungsgemeinschaft (H.W.K., A.O.).

REFERENCES

1 Reese, T. S. and Karnovsky, M. J. (1967) J. Cell
Biol., 34: 207-217.

2 Brightman, M. W. and Reed, T. S. (1969) J. Cell
Biol., 40: 648-677.

3. Leonhardt, H. (1980) In “Handbuch der mikroskop-
ischen Anatomie des Menschen, IV/10: Neuroglia
I”. Ed. by A. Oksche, Springer, Berlin/Heidelberg/
New York, pp. 177-666.

4 van Deurs, B. (1980) Int. Rev. Cytol. 65: 117-191.

5 Goldstein, G. W. and Betz, A. L. (1986) Sci. Amer.
255 (3): 70-79.

6 Brightman, M. W. (1968) In “Progress in Brain
Research, Vol. 29: Brain Barrier Systems”. Ed. by
A. Lajtha and D. H. Ford, Elsevier, Amsterdam,
pp. 19-40.

7 Murabe, Y., Nishida, K. and Sano, Y. (1981) Cell
Tissue Res., 219: 85-92.

8 Krisch, B., Leonhardt, H. and Oksche, A. (1983)
Cell Tissue Res., 228: 597-640.

9 Westergaard, E. and Brightman, M. W. (1973) J.
Comp. Neurol., 152: 17-44.

10 Milhorat, J. H. (1976) Int. Rev. Cytol., 47: 225-
288.

11 van Deurs, B., Mgller, M. and Amtorp, O. (1978)
Cell Tissue Res., 187: 215-234.

12 Kolmer, W. (1921) Anat. Anz., 54: 15-19.

13. Ariéns Kappers, J. (1953) Z. Anat. Entwickl.-
gesch., 117: 1-19.

14 Carpenter, S. J., McCarthy, L. E. and Borison, H.
L. (1970) Z. Zeilforsch., 110: 471-486.

\

15)

16
IY)

18

Ww)

20

21

22

73)

24

25
26
Za])

28

29

30

31

32

33

34
35)

36

37

38

798

Bleier, R., Albrecht, R. and Gruce, J. A. F. (1975)
Science, 189: 299-301. ,

Bleier, R. (1977) J. Comp. Neurol., 174: 359-376.
Omura, Y., Korf, H.-W. and Oksche, A. (1985)
Cell Tissue Res., 239: 599-610.

Oksche, A. and Hartwig, H. G. (1975) In “Brain-
Endocrine Interaction, Vol. II”. Ed. by K. M. Knig-
ge, D. E. Scott, H. Kobayashi and S. Ishii, Karger,
Basel, pp. 41-53.

Oksche, A. and Hartwig, H. G. (1979) In “Progress
in Brain Research, Vol. 52: The Pineal Gland of
Vertebrates including Man”. Ed. by J. Ariéns Kap-
pers and P. Pévet, Elsevier/North-Holland, Amster-
dam/New York/Oxford, pp. 113-130.

Omura, Y. and Ali, M. A. (1981) Cell Tissue Res.,
219: 355-369.

Obermiiller-Wilén, H. (1971) Acta Zool. (Stockh.),
52: 103-115.

Obermiiller-Wilén, H. (1972) Acta Zool. (Stockh.),
53: 219-228.

Jansen, W. F., de Weger, R. A., Woutersen, R. A.,
vna Loveren, H. and van de Kamer, J. C. (1976)
Cell Tissue Res., 167: 467-491..

Jansen, W. F., van Loveren, H., Woutersen, R. A.
and de Weger, R. A. (1976) Histochemistry, 48:
293-306.

Rtideberg, C. (1969) Z. Zellforsch., 93: 282-304.
Korf, H.-W. (1974) Cell Tissue Res., 155: 475-489.
Borg, B., Ekstrém, P. and van Veen, Th. (1983)
Acta Zool. (Stockh.), 64: 211-218.

Omura, Y. (1988) In “Daily Rhythmic Activities in
Aquatic Animals”. Ed. by I. Hanyu and M. Tabata,
Koseisha-Koseikaku, Tokyo, pp. 47-62.
Holmgren, U. (1965) In “Progress in Brain Re-
search, Vol. 10: Structure and Function of the
Epiphysis Cerebri”. Ed. by J. Ariéns Kappers and J.
P. Schadé, Elsevier, Amsterdam/London/New
York, pp. 172-182.

Oguri, M. and Omura, Y. (1973) In “Responses of
Fish to Environmental Changes”. Ed. by W. Cha-
vin, Charles C. Thomas, Springfield, pp. 412-434.
Wake, K. (1973) Z. Zellforsch., 145: 287-298.
Ueck, M., Wake, K. and Kobayashi, H. (1989)
Zool. Sci., 6: 817-831.

Hafeez, M. A. and Zerihun, L. (1974) Cell Tissue
Res., 154: 485-510.

Omura, Y. (1979) Rev. Can. Biol., 38: 105-118.
Ekstrém, P. (1984) J. Comp. Neurol., 226: 321-
335.

Ekstrom, P., Foster, R. G., Korf, H.-W. and Schal-
ken, J. J. (1987) J. Comp. Neurol., 265: 25-33.
Tabata, M., Tamura, T. and Niwa, H. (1975) Vision
Res., 15: 737-740.

Hanyu, I., Niwa, H. and Tamura, T. (1978) Comp.
Biochem. Physiol., 61A: 49-54.

39

40

41

42

43

44

45
46

47
48
49
50
51

52

53)
54

3p)
56

D7

Y. Omura, H.-W. Kore et al.

Tamura, T. and Hanyu, I. (1980) In “Environmen-
tal Physiology of Fishes”. Ed. by M. A. Ali, Ple-
num, New York, pp. 477-496.

Falc6n, J., Guerlotté, J. F., Voisin, P. and Collin,
J.-P. (1987) Neuroendocrinology, 45: 479-486.
Kezuka, H., Furukawa, K., Aida, K. and Hanyu, I.
(1988) Gen. Comp. Endocrinol., 72: 296-302.
Oguri, M., Omura, Y. and Hibiya, T. (1968) Bull.
Jap. Soc. Sci. Fish., 34: 687-690.

Friedrich-Freksa, H. (1932) Z. Wiss. Zool., 141:
52-142.

Syed Ali, S., Korf, H.-W. and Oksche, A. (1987)
Cell Tissue Res., 250: 425-429.

Quay, W. B. (1973) Am. J. Anat. 137: 387-402.
Hodde, K. C. (1979) In “Progress in Brain Re-
search, Vol. 52: The Pineal Gland of Vertebrates
including Man”. Ed. by J. Ariéns Kappers and P.
Pévet, Elsevier/North-Holland, Amsterdam/New
York/Oxford, pp. 39-44.

Hill, C. (1894) J. Morphol., 9: 237-268.
Rtideberg, C. (1971) Z. Zellforsch., 122: 227-243.

van Veen, Th., Ekstrém, P., Borg, B. and Moller,
M. (1980) Cell Tissue Res., 209: 11-28.

van Veen, Th. (1982) Cell Tissue Res., 222: 433-
444. .
Vigh-Teichmann, I., Korf, H.-W., Oksche, A. and
Vigh, B. (1982) Cell Tissue Res., 227: 351-369.
Vigh-Teichmann, I., Korf, H.-W., Nurnberger, F.,
Oksche, A., Vigh, B. and Olsson, R. (1983) Cell
Tissue Res., 230: 289-307.

Holmgren, U. (1959) G6teborgs. Kungl. Vetensk.
Vitterh.-Samh. Handl., B8: 5-66.

Tsuneki, K. (1986) Japan. J. Ichthyol., 33: 27-38.
Dohrmann, G. L. (1970) Brain Res., 18: 197-218.
Milhorat, J. H., Hammock, M. K., Fenstermacher,
J. D., Rall, D. P. and Levin, V. A. (1971) Science,
173: 330-332.

Hedlund, L., Lischko, M. M., Rollag, M. D. and
Niswender, G. D. (1977) Science, 195: 686-687.
Moller, M., van Deurs, B. and Westergaard, E.
(1978) Cell Tissue Res., 195: 1-15.

Welsh, M. G. and Beitz, A. J. (1981) Am. J. Anat.,
162: 343-355.

Lee, J. C. (1971) Progr. Neuropathol., 1: 84-145.

Brightman, M. W., Reese, T. S., Olsson, Y. and
Klatzo, J. (1971) Progr. Neuropathol., 1: 146-161.

Murray, M., Jones, H., Cserr, H. F. and Rall, D. P.
(1975) Brain Res., 99: 17-33.

Bundgaard, M., Cserr, H. and Murrary, M. (1979)
Cell Tissue Res., 198: 65-77.

Reiter, R. J. (1981) Am. J. Anat., 162: 287-313.

Hewing, M. and Bergmann, M. (1985) Cell Tissue
Res., 241: 149-154.

Moller, M. (1986) In “III Reinhardsbrunn Sympo-

67

68

69

70

Th

72

1

74

75

Blood-Brain Barrier in Fish Epithalamus

sium: Circumventricular Organs and Brain Fluid
Environment”. Karl Marx University, Leipzig, p. 53
(Abstract).

Castel, M., Sahar, A. and Frlij, D. (1974) Brain
Res.. 67: 178-184.

Bouldin, T. W. and Krigman, M. R. (1975) Brain
Res., 99: 444-448.

Claude, P. and Goodenough, D. A. (1973) J. Cell
Biol. 58: 390-400.

Noske, W. and Hirsch, M. (1986) Cell Tissue Res.,
245: 405-412.

Omura, Y., Ueno, S. and Ueck, M. (1986) Cell
Tissue Res., 245: 171-176.

Wislocki, G. B. and Leduc, E. H. (1952) J. Comp.
Neurol. 96: 371-413.

Drezki, J. (1971) Z. Anat. Entwick.-gesch., 134:
278-297.

Milhorat, J. H., Davis, D. A. and Lloyd, B. J., Jr.
(1973) Science, 180: 76-78.

Anton-Tay, F. and Wurtman, R. J. (1969) Nature,
221: 474-475.

76
A

78
79

80

81

82
83

84

799

Vanecek, J., Pavlik, A. and Illnerova, H. (1987)
Brain Res., 435: 359-363.

Williams, L. M. (1989) J. Molec. Endocr., 3: 71-75.
Hewing, M. (1982) Cell Tissue Res., 222: 177-185.
Reiter, R. J. (1986) In “III Reinhardsbrunn Sympo-
sium: Circumventricular Organs and Brain Fluid
Environment”. Karl Marx University, Leipzig, p. 59
(Abstract).

Hashimoto, P. H., Gotow, T., Ichimura, T., Naka-
tani, T., Takasu, N., Kodaka, R., Sumitani, S. and
Fukuda, T. (1985) Arch. histol. jap., 48: 173-181.
Hashimoto, P. H. (1986) In “III Reinhardsbrunn
Symposium: Circumventricular Organs and Brain
Fluid Environment”. Karl Marx University, Leipzig,
p. 24 (Abstract).

Ling, E. A. (1978) J. Anat., 125: 101-106.
Murabe, M. and Sano, Y. (1983) Cell Tissue Res.,
229: 85-95.

Matsumoto, Y. and Ikuta, F. (1985) Cell Tissue
Res., 239: 271-278.

ate info

a ae

a0 3

* i Pe oe int

ee rik:

=

=
rides pend Pi

ZOOLOGICAL SCIENCE 7: 801-814 (1990)

© 1990 Zoological Society of Japan

REVIEW

Structure and Action of Molluscan Neuropeptides

Makoto KospayasHt and Yosrro MUNEOKA

Physiological Laboratory, Faculty of Integrated Arts and Sciences
Hiroshima University, Hiroshima 730, Japan

INTRODUCTION

Since Price and Greenberg [1] isolated the
bioactive peptide FMRFamide from the ganglia of
the clam Macrocallista nimbosa and determined its
structure in 1977, a large number of neuropeptides
have been identified in molluscs. The search for
neuropeptides in molluscs is still in progress and
the number of the peptides identified has been
increasing exponentially.

In our laboratory, we have isolated neuro-
peptides and determined their structure of several
kinds of molluscs, including Mytilus edulis (Bival-
via), Fusinus ferrugineus (Prosobranchia) and
Achatina fulica (Pulmonata), and have found
many novel species of neuropeptides. We are also
attempting to isolate neuropeptides from several
non-molluscan animals, from the coelenterates to
the vertebrates.

One of the aims of our peptide search is to
observe the structure and action of neuropeptides
from a comparative point of view. It has been
shown that neuropeptides can be classified into a
number of groups according to their structure and
action. Although the peptides in a particular
group are structurally related to each other, they
are often distributed interphyletically. For exam-
ple, FMRFamide-like peptides are distributed
from the coelenterates to the vertebrates [2].
However, little is known about the evolutionary
relationship among them. In order to throw light
upon this relationship, first of all, it is necessary to

Received May 7, 1990
' To whom all correspondence should be addressed.

search for more FMRFamide-like peptides and to
examine their action, since there seem to be many
more FMRFamide-like peptides yet to be iden-
tified. Similar situations should also be considered
in other peptide groups.

Another aim of our work is to identify the
peptides actually playing roles as neuromediators
involved in the neural mechanisms which we are
trying to elucidate. For example, we are investi-
gating the neural regulatory mechanism of cardiac
activity in the African giant snail Achatina [3, 4],
and recently found a new FMRFamide-like un-
decapeptide which exists in abundance in the atrial
tissue and potentiates the beat of the ventricle of
this animal [5]. This peptide appears to play a
physiological role in the regulation of the heart
beat of Achatina.

Achatina as well as other pulmonate molluscs
possesses a number of large identifiable neurons in
its ganglia. In this respect, Achatina is a very
suitable specimen for neurobiological experi-
ments, and has been used by many investigators,
particularly the Japanese. However, no
neuropeptides were identified from this animal
until Kamatani et al. [6] found an excitatory tet-
rapeptide, achatin-I, in the ganglia of this animal
in 1989. As mentioned later, several neuro-
peptides including the FMRFamide-like undeca-
peptide were recently isolated from Achatina in
our laboratory. These peptides will be useful for
experiments on the neural mechanisms in Acha-
tina.

In this review, we describe the bioactive pep-
tides isolated and purified in our laboratory from
the extracts of muscle tissues and ganglia of three

802 M. KoBAYASHI AND Y. MUNEOKA

species of molluscs, Mytilus edulis, Fusinus ferru-
gineus and Achatina fulica. The structure and
action of these peptides will be compared with
those of other peptides found in various animals by
many other investigators, and it will be attempted
to classify them. We also discuss the possible roles
of the peptides in the three animals.

FMRFamide-RELATED PEPTIDES

The neuropeptide Phe-Met-Arg-Phe-NH>
(FMRFamide) seems to be present in all molluscan
taxa [7, 8]. Further, the peptide may also be
distributed in annelids. The presence of FMRFa-
mide or FMRFamide-related peptide (FaRP) has
been suggested in the leech Hirudo medicinalis {9|
and the earthworm Eisenia foetida [10], and Kra-
jniak and Price [11] recently isolated and sequ-
enced FMRFamide itself in the polychaete Nereis
virens. They proposed that FMRFamide is one of
the more ancient members of the FaRP family.
We isolated FMRFamide from the anterior byssus
retractor muscle (ABRM) of Mytilus and the gang-
lia of Fusinus and Achatina. That is, the peptide
was found to be present in all of the three species
of molluscs which we used for peptide isolation.

The tetrapeptide Phe-Leu-Arg-Phe-NH>
(FLRFamide), as well as FMRFamide, has been
shown to be ubiquitous in molluscs [7, 8, 12],
though the former is always a minor component
[13, 14]. We found FLRFamide in the ganglia of
Fusinus, but failed to find it in the ABRM and the
pedal ganglion of Mytilus. We also failed to find
the peptide in the cerebral and suboesophageal
ganglia and the cardiac tissue of Achatina. In the
latter two molluscs, the amounts of FLRFamide
may be too small to detect by the bioassay methods
we used. In the ganglia of Fusinus, the total yield
of FLRFamide was less than one tenth of FMRFa-
mide, that is, FLRFamide is also a minor compo-
nent in this animal.

In the ABRM, FMRFamide shows a catch-
relaxing or contractile effect depending on its
concentration [15-17]. At 10 °—10-’M, the
peptide relaxes catch tension induced by acetyl-
choline (ACh), whereas at 10°-°M or higher, it
produces a contraction. Further, at 10~?M or
higher, FMRFamide potentiates contractions in

response to electrical pulses of stimulation or ACh
[16]. Of these three kinds of effects of FMRFa-
mide, the potentiating one seems to be of physio-
logical interest. The peptide might be a neuromo-
dulator playing a role in potentiating the contrac-
tion of the ABRM [18].

In the heart of a prosobranch mollusc, Rapana
thomasiana, both FMRFamide and serotonin en-
hance its activity at low concentrations [19, 20].
The excitatory effect of serotonin is blocked by the
ergot alkaloid methysergide, but that of FMRFa-
mide is not affected. Heart excitation by neural
stimulation is also not affected by methysergide
[19]. Using an immunohistochemical method, a
FMRFamide-like substance has been shown to be
present in the atrium of Rapana [21]. These facts
suggest that FMRFamide may be an excitatory
neurotransmitter in the heart of Rapana [21, 22].

In contrast to Rapana, the heart of Achatina, as
well as most other molluscan hearts, seems to be
regulated by serotonin, and this biogenic amine is
considered to be the principal excitatory transmit-
ter [22, 23]. FMRFamide may not be, at least
directly, involved in the regulation of the heart.
Although FMRFamide was found in the ganglia of
Achatina [24], it was not found in the cardiac
tissues. Further, the threshold for a direct excita-
tory effect of FMRFamide on the heart was found
to be as high as 107° M [23].

In the radula retractor of Fusinus, FMRFamide
as well as FLRFamide potentiates twitch contrac-
tions at higher than 10~” M and evokes a contrac-
tion by itself at higher than 10° °M [25, 26]. It is
suspected that these peptides are excitatory
neurotransmitters in this muscle, although strong
evidence is not yet available.

In the radula protractor of Rapana, FMRFa-
mide markedly enhances twitch contractions in
response to short pulses of electrical field stimula-
tion, whereas FLRFamide does not have such an
effect. In the radula retractor of the animal, on the
contrary, FLRFamide enhances twitch contrac-
tions while FMRFamide shows little effect. Since
the effects of FMRFamide and FLRFamide are
not similar, this suggests that specific receptors for
each peptide have evolved in this mollusc. The
modulatory effects of both peptides have been
considered to be brought about by their actions on

Molluscan Neuropeptides

presynaptic sites [27].

The effects of FMRFamide and FLRFamide
have been tested not only on the foregoing muscles
but also on various other molluscan muscles, for
example, the penis retractor muscle and buccal
muscles of Achatina, the proboscis retractor mus-
cle of Fusinus and the pedal retractor muscle of
Mytilus. In all of the muscles tested, except the
radula retractor and protractor muscles of Rapana,
both peptides showed similar excitatory action,
contractile and contraction-potentiating actions.
In addition to these actions, they showed a relaxa-
tion-enhancing action in some muscles, namely,
the ABRM of Mytilus, the penis retractor muscle
of Achatina and the pharyngeal retractor muscle of
a Japanese land snail, Euhadra congenita.

In addition to the foregoing tetrapeptides, two
pentapeptides [28] and four heptapeptides [8, 29,
30] have been found as minor components of
FaRPs in molluscs. The pentapeptides were iso-
lated from Octopus vulgaris and the heptapeptides
from some pulmonates. In our laboratory, we
found two novel molluscan FaRPs, one from Myu-
lus ABRMs and the other from Fusinus ganglia
[18, 26, 31, 32]. As shown in Table 1, both of them
are decapeptides whose structures are closely re-
lated to each other. The Fusinus peptide termi-
nates in FLRFamide, while the Mytilus peptide
terminates in FFRFamide, neither FMRFamide
nor FLRFamide. In this respect, the structure of
the Mytilus decapeptide is unique.

Both decapeptides show FMRFamide-like ac-

TABLE 1.

H-Phe-Met-Arg-Phe-NH,

803

tion in various molluscan muscles. In the ABRM
of Mytilus, the Mytilus peptide potentiates a phasic
contraction in response to electrical stimulation at
higher than 107 '* M and evokes a contraction by
itself at higher than 10~° M. This peptide may be a
neuromodulator having the function of potentiat-
ing a contraction in the ABRM. The Fusinus
peptide is less potent than the Mytilus peptide in
the ABRM [18, 26].

In the radula retractor of Fusinus, the Fusinus
peptide potentiates twitch contractions at higher
than 10~'° M and evokes a contraction by itself at
higher than 10°°M. This peptide may be a
neuromodulator in the radula retractor. The Myti-
lus peptide is less potent than the Fusinus peptide
in the radula retractor [26].

Two hexapeptides having RFamide at their C-
terminal parts have been isolated from the ganglia
of Fusinus [26, 32]. Their primary structures are as
follows:

H-Gly-Ser-Leu-Phe-Arg-Phe-NH>
H-Ser-Ser-Leu-Phe-Arg-Phe-NH>

These peptides do not show FMRFamide-like ac-
tion in molluscan muscles. In the radula retractor
of Fusinus, the peptides do not potentiate but on
the contrary, inhibit twitch contractions. In the
ABRM of Mytilus, they inhibit the relaxation of
catch tension in response to repetitive electrical
pulses of stimulation probably by acting on pre-
synaptic sites [26]. It is very interesting that the
peptides terminating in FLRFamide and FFRFa-

Structure of FMRFamide-related peptides of molluscs

All molluscan taxa

H-Phe-Leu-Arg-Phe-NH,

H-Thr-Phe-Leu-Arg-Phe-NH,

H-Ala-Phe-Leu-Arg-Phe-NH,

Cephalopod Octopus

pGlu-Asp-Pro-Phe-Leu-Arg-Phe-NH,

H-Ser-Asp-Pro-Phe-Leu-Arg-Phe-NH,

Pulmonates

H-Gly-Asp-Pro-Phe-Leu-Arg-Phe-NH>
H-Asn-Asp-Pro-Phe-Leu-Arg-Phe-NH,

H-Ala-Leu-Thr-Asn-Asp-His-Phe-Leu-Arg-Phe-NH,
H-Ala-Leu-Ala-Gly-Asp-His-Phe-Phe-Arg-Phe-NH,

For details, see references [7, 8, 12, 26, 28].

Prosobranch Fusinus
Bivalve Mytilus

804 M. KoBAYASHI AND Y. MUNEOKA

mide exhibit FMRFamide-like action, whereas the
peptides terminating in LFRFamide exhibit the
opposite action to FMRFamide.

Another peptide having RFamide at the C-
terminal part has been isolated from the atria of
Achatina [5]. The primary structure of this peptide
is as follows:

H-Ser-Gly-Gln-Ser-Trp-Arg-Pro-
Gln-Gly-Arg-Phe-NH,

This undecapeptide was termed Achatina cardio-
excitatory peptide-1 (ACEP-1). Although a con-
siderable amount of ACEP-1 appears to be present
in the atrial tissue, it does not have an appreciable
effect on the atrium at 10~*M or lower. In
contrast, the peptide shows a potent excitatory
effect on the ventricle. Further, ACEP-1 potenti-
ates tetanic contractions of the penis retractor
muscle and the buccal muscle of Achatina. In the
former muscle, the relaxation rate of the contrac-
tion is also enhanced. These actions of ACEP-1
are similar to those of FMRFamide. In the central
nervous system of Achatina, however, the action
of ACEP-1 is not similar to that of FMRFamide.
In an identified neuron R-B,, ACEP-1 shows a

potent excitatory action, while FMRFamide shows
a rather inhibitory action (Fig. 1). ACEP-1 might
be an atrial hormone-like substance which brings
about an arousal state in the animal.

A peptide which is predicted by cDNA analysis
to be expressed from neuron L; of Aplysia califor-
nica appears to have nearly the same amino acid
sequence as ACEP-1 [33]. Furthermore, the se-
quence of the C-terminal part of ACEP-1 (-Gln-
Gly-Arg-Phe-NH,) is almost the same as antho-
RFamide (pGlu-Gly-Arg-Phe-NH>), a _ neuro-
peptide isolated from sea anemones [34]. ACEP-
1-related peptides may be distributed not only in
molluscs but also in other phyla.

FaRPs characterized by the C-terminal structure
FXRFamide are found in nematodes, molluscs,
annelids and arthropods [12]. Further, peptides
characterized by the C-terminal structure RFa-
mide are more widely distributed [12]. Genetic
analyses are required to elucidate whether there is
evolutionary relationship among them.

MYOMODULIN-CARP-RELATED PEPTIDES
FMRFamide relaxes catch tension in the ABRM

(Achatina neuron R-B4)

sit N {

\

He
“ih i

\

Hutt
i

I)

eta,
AN TIEER

Hah (i! Rat vm fi eH
“itl ital! HE i! | Ly | Tbe A

Vey FE

wash
ACEP-1
B
I>
=
(==)
TG Ee > 2
‘ y 1 min
40-°M wash
FMRFamide

Fic. 1.
buccal ganglion of Achatina.

Effect of ACEP-1 (A) and FMRFamide (B) on the identified neuron R-B, in the right

Molluscan Neuropeptides 805

of Mytilus at low concentrations and evokes a
contraction at higher concentrations. Muneoka
and Saitoh [17] have examined the structure-
activity relation of FMRFamide on the ABRM.
They have found that the relationship for relaxa-
tion is different from that for contraction, which is
similar to the relationship commonly observed in
other molluscan muscles [20, 35]. From these
facts, it was suspected that Mytilus may possess an
unknown peptidic substance whose function is to
relax catch, and thus we have attempted to hunt
for it. In 1987, a heptapeptide having a potent
catch-relaxing action was found in the pedal gang-
lia of Mytilus in our laboratory [36] and was
termed catch-relaxing peptide (CARP).

The sequence of CARP is homologous with that
of myomodulin which was found by Cropper et al.
[37] to be a neuromodulator controlling the acces-
sory radula closer (ARC) muscle in Aplysia. La-
ter, Vilim et al. [38] found a new myomodulin-
related peptide also in Aplysia and termed it
myomodulin b. In our laboratory, we isolated
another myomodulin-related peptide (Fusinus
myomodulin) in addition to myomodulin itself
from the ganglia of Fusinus [26, 32]. The primary
structures of these myomodulin-CARP-related
peptides (MCRPs) are shown in Table 2.

CARP has been shown to have biological effects
on muscles and neurons in many molluscs [39-41].
In collaboration with others (T. Kiss and L. Hiripi,
Hungary), we recently obtained some results in-
dicating that acetone extracts of the ganglia of a
freshwater bivalve Anodonta cygnea contain a
CARP-like peptidic substance. Further, the pre-
sence of this CARP-like substance has also been
shown immunohistochemically in Polyplacophora,
Pulmonata and Cephalopoda [42, and personal
communications from S. Moffett and from K.
Kuwasawa]. MCRPs seem to be widely distributed
in molluscs.

In Aplysia, the cholinergic motoneuron By,
which innervates the ARC muscle contains
myomodulin and myomodulin b. These peptides
are thought to be co-transmitted with the principal
excitatory transmitter ACh and thought to mod-
ulate contractions of the muscle. The peptides
exogenously applied to the muscle potentiate its
contractions at low concentrations. However,
their actions diverge at higher concentrations.
Myomodulin at higher than 10~’M and myomo-
dulin b at higher than 10~° M depress contractions
[37, 38]. In the ABRM of Mytilus, CARP also
potentiates contractions at low concentrations and
depresses them at higher concentrations. The
threshold concentration of CARP for potentiation
is 3x10 '°—-5x107'°M, being almost identical
with that for relaxation of catch tension. On the
other hand, the threshold concentration for de-
pression varies from 5X10°-?M to 5xX10°°M
depending on the muscle preparations [36, 40]. In
other molluscan muscles, such as the radula mus-
cles of Fusinus and Rapana and the cardiac mus-
cles of Mytilus and Rapana, CARP shows only
inhibitory activity [40]. Myomodulin has also been
found to inhibit contractions of the radula muscles
of Fusinus. The general action of MCRPs on
molluscan muscles seems to be that of inhibition.

The relaxing action of CARP on catch tension in
the ABRM of Myztilus is of physiological interest.
When the ABRM is continuously immersed in a
supramaximal dose of CARP and, during this
immersion, ACh is repeatedly applied for 2 min at
10 min intervals, catch tension recovers gradually
after washing-out ACh. This is probably because
the muscle becomes desensitized to the peptide
during immersion. In such a desensitized state, the
relaxing response to stimulation of low frequency
repetitive electrical pulses is markedly depressed
or abolished, although the relaxing response to
serotonin is little affected [41]. Fujisawa et al. [31]

TABLE 2. Structure of myomodulin-CARP-related peptides of molluscs

H-Pro-Met-Ser-Met-Leu-Arg-Leu-NH,
H-Gly-Ser-Tyr-Arg-Met-Met-Arg-Leu-NH,
H-Pro-Met-Asn-Met-Leu-Arg-Leu-NH),
H-Ala-Met-Pro-Met-Leu-Arg-Leu-NH),

Myomodulin (Aplysia, Fusinus)
Myomodulin b (Aplysia)
Fusinus myomodulin

CARP (Mytilus)

806 M. KOBAYASHI AND Y. MUNEOKA

purified a peptidic catch-relaxing substance from
the ABRMs themselves. The results of sequence
analysis and amino acid analysis have proved that
it is CARP, that is, CARP is present in the
ABRM. Thus, it can be supposed that CARP is a
relaxing transmitter in the ABRM, though the
principal one is considered to be serotonin [18, 26].

It has been shown that CARP has biological
effects not only on molluscan muscles but also on
non-molluscan muscles including a mammalian
muscle, the aorta of guinea pig [26]. In the
echiuroid worm Urechis unicinctus, twitch contrac-
tion of the body-wall muscle is potentiated by
CARP. The threshold concentration is as low as
10-§M. CARP-like peptides might also be pre-
sent in non-molluscan animals [26].

MIP-RELATED PEPTIDES

In 1988, two congeneric hexapeptides that in-
hibited phasic contraction of the Mytilus ABRM in
response to repetitive electrical pulses of stimula-
tion were isolated from the pedal ganglia of the
mussel [43]. These peptides were termed Mytilus
inhibitory peptides (MIPs). Their structures are as
follows:

H-Gly-Ser-Pro-Met-Phe-Val-NH,
H-Gly-Ala-Pro-Met-Phe-Val-NH,

(S?-MIP)
(A?-MIP)

Later, Fujisawa et al. [26, 31] found peptides of
the same structures in the ABRMs themselves. In
addition, they isolated another MIP-related pep-
tide (MIPRP) from the muscles. Although the
structure of this peptide is not completely con-
firmed yet, the probable structure has been prop-
osed as follows [31]:

H-Asp-Ser-Pro-Leu-Phe-Val-NH>

The actions of the three MIPs on contractions of
the ABRM seem to be similar both qualitatively
and quantitatively [18, 26, 43, 44]. The peptides
do not affect catch tension, though they show a
potent inhibitory action on phasic contraction.
The threshold concentration of S?-MIP and A2-
MIP for inhibition was between 107 !°M and 5x
10~'°M. This is probably also the case for the
third MIP.

Fujisawa et al. [26] examined the inhibitory

actions of two MIP-fragment peptides, H-Pro-
Met-Phe-Val-NH> (MIP 3-6) and H-Met-Phe-Val-
NH) (MIP 4-6), on the phasic contraction of the
ABRM. They found that MIP 3-6 was 10-30
times less potent than MIPs and MIP 4-6 was
3,000-10,000 times less potent. From these re-
sults, it is suspected that the essential structure in
MIPs for inhibitory action on the ABRM may be
-Phe-Val-NH), and that the structure -Pro-X-Phe-
Val-NH> may be required for exhibition of potent
inhibitory action. The three MIPs may act on the
common receptor sites. The difference among the
structures of the peptides may be functionally
neutral.

MIPs show inhibitory action not only on Mytilus
muscles but also on various other molluscan mus-
cles and neurons [44-46]. Yongsiri et al. [45]
examined the actions of S*-MIP, A?-MIP and their
fragment peptides, MIP 3-6 and MIP 4-6, on
neurons of Achatina. They observed that among
several identifiable neurons examined, the right
anterior pallial neuron was hyperpolarized by the
peptides. The potency order of the four peptides
was as follows:

S?-MIP > A2-MIP >MIP 3-6 >MIP 4-6

They concluded that the hyperpolarization is due
to an increase in membrane conductance to potas-
sium ions.

Kiss [46] observed that S?-MIP hyperpolarized
the membranes of neurons of Helix pomatia by
inducing voltage-independent small outward cur-
rent due to potassium flux, and that S?-MIP also
modulated the membrane calcium conductance.
He concluded that the peptide exerted its effect on
the Helix neurons by a combination of more than
one type of ionic mechanism.

In collaboration with T. Kiss and L. Hiripi, we
fractionated acetone extracts of the ganglia of
Helix, and obtained more than ten peaks of MIP-
like activity. We also fractionated acetone extracts
of the ganglia of Anodonta and obtained some
peaks of MIP-like activity. MIPRPs may also be
widely distributed in molluscs. We tested MIPs on
the muscles of several non-molluscan animals,
however, thus far no effects have been observed.

Molluscan Neuropeptides 807

SCP-RELATED AND BCP-RELATED
PEPTIDES

The neuropeptides designated small cardioac-
tive peptides A and B (SCP, and SCPx) have been
purified from the nervous system and the peripher-
al tissues of Aplysia [47, 48]. The actions of the
peptides, in particular those of SCPg, have been
extensively investigated in many molluscs. In the
ABRM of Mytilus, SCPg at 10~* or higher po-
tentiates phasic contraction in response to repeti-
tive electrical pulses of stimulation and at 10-° M
or higher produces a contraction [17, 18]. In this
respect, the actions of SCP are similar to those of
FMRFamide, although SCPg does not show any
catch-relaxing action in the ABRM. In the heart
of Achatina, SCP, antagonizes the enhancing ac-
tions of serotonin and neural stimulation on the
beat, in contrast to FMRFamide which further
enhances those actions [23].

In our laboratory, Fujisawa et al. [31] purified
three new species of inhibitory peptidic substances
from acetone extracts of ABRMs. They were
provisionally designated MIP,, MIP; and MIP7.
The HPLC profiles of the peptides indicated that
they are different species from the foregoing MIP},
MIP, and MIP3. The amino acid sequence analy-
ses of the new peptides showed that they have
structures more or less homologous to SCPs as
shown below.

H-Ala-Arg-Pro-Gly-

Tyr-Leu-Ala-Phe-Pro-Arg-Met-NH2 (SCPa)
H-Met-Asn-
Tyr-Leu-Ala-Phe-Pro-Arg-Met-NH, (SCPs)
H-Leu-Ala-Tyr-Pro-Arg-Leu- (MIP)
H-Ala-Ser-His-Ile-Pro-Arg-Phe- (MIP;)
H-Tyr-Ala-Pro-Arg-Phe- (MIP?)

However, the synthesized peptide, H-Leu-Ala-
Tyr-Pro-Arg-Leu-NH) (the amidated peptide hav-
ing a MIP,-sequence), did not show any inhibitory
action but showed a potentiating action on the
phasic contraction of the ABRM. These new
peptides might terminate in -Pro-Arg-Phe(Leu)-
Val-NH>, because the C-terminal Val residue is
often not detected in amino acid sequence analysis
when the peptide sample is small. That is, the new

inhibitory peptides might be members of the fore-
going MIPRPs. If this is the case, MIP, may be
regarded as a peptide related to both SCP and MIP
families. It is now important step up to search for
more peptides related to SCPs and to MIPs and to
investigate their actions as well as their structures.

A number of neuropeptides which have been
designated bag-cell peptides (BCPs) were isolated
from the bag cells of Aplysia [49]. Mostly, they
have the sequence Arg-Leu-Arg-Phe as a fragment
structure [50-52]. For example, the structure of
beta-BCP is H-Arg-Leu-Arg-Phe-His-OH [51]. In
our laboratory, Fujimoto et al. [53] isolated a
peptide related to BCPs from the atria of Achati-
na, the sequence of which has been proposed as
follows:

H-Arg-Leu-Arg-Phe-Ala-(OH)

It has not yet been confirmed whether the C-
terminus of the peptide is free or amidated.
However, the peptide, as well as BCPs, shows
potent cardioexcitatory action in Achatina, and
hence, the C-terminus is surmised to be free.

BUCCALIN-RELATED PEPTIDES

The neuropeptide buccalin was isolated from the
buccal ganglia of Aplysia by Cropper et al. in 1988
[54]. In the buccal neuron B;5 innervating the
ARC muscle, buccalin is co-localized with SCP,
and SCPg. Unlike SCPs which potentiate contrac-
tion of the muscle by acting on postsynaptic sites,
buccalin depresses the contraction probably by
acting on presynaptic sites [54]. In addition to
buccalin, two related peptides, buccalin b (para-
buccalin) and buccalin c, were isolated from the
buccal ganglia of Aplysia [55, 56]. The two pep-
tides were found to exhibit action similar to bucca-
lin on the ARC muscle.

From the ganglia of Fusinus, Kuroki of our
laboratory isolated a peptide having a structure
significantly homologous to Aplysia buccalins. The
structures of the buccalin-related peptides are
shown in Table 3. Although the C-terminus of
Fusinus buccalin has not been confirmed, the
peptide is suspected to be amidated, since both
Fusinus buccalin and Aplysia buccalin similarly
inhibit twitch contractions of the radula retractor

808 M. KoBAYASHI AND Y. MUNEOKA

TABLE 3.

H-Gly-Met-Asp-Ser-Leu-Ala-Phe-Ser-Gly-Gly-Leu-NH,
H-Gly-Leu-Asp-Arg-Tyr-Gly-Phe-Val-Gly-Gly-Leu-NH,
H-Gly-Phe-Asp-His-Tyr-Gly-Phe-Thr-Gly-Gly-Ile-NH),

H-Arg-Met-Asp-Ser-Met-Met-Phe-Gly-Pro-Gln-Leu-(NH;)

muscle of Fusinus. It is not clear whether Fusinus
buccalin acts on presynaptic or postsynaptic sites in
the radula retractor muscle.

RPCH-AKH-RELATED PEPTIDES

In 1974, the crustacean red-pigment-concentrat-
ing hormone (RPCH) was isolated from shrimps
by Fernlund [57]. Two years later, an insect
adipokinetic hormone (AKH), the structure of
which being closely related to RPCH, was isolated
from locusts by Stone et al. [58]. From the time of
these findings to the present date, nine more
species of peptides related to RPCH and AKH
have been found in insects [59, 60]. These arthro-
pod neuropeptides are called RPCH-AKH-family
peptides or RPCH-AKH-related peptides
(RARPs). The structures of the RARPs are
shown in Table 4.

RPCH and AKH show biological activity not
only in arthropod tissues but also in molluscan
tissues. Greenberg et al. [61] have shown that both
the locust AKH and the crustacean RPCH (see
Table 4) are potent excitors of the heart of the
clam Mercenaria mercenaria. Saitoh and Muneoka
[62] have observed that RPCH inhibits or potenti-
ates contractions of Mytilus muscles (the ABRM,
the pedal retractor and the heart). Yanagawa et al.
[27] have also shown that RPCH powerfully po-
tentiates twitch contractions of the radula retractor
of Rapana.

The foregoing facts suggest that RARPs may
also be distributed in molluscs. In fact, Greenberg
et al. [61] have obtained evidence that the ganglion
extracts of Mercenaria contain a peptidic substance
related to RPCH. Therefore, we attempted to
isolate RARP from the ganglia of Fusinus, and
found a peptide closely related to the C-terminal
tetrapeptide fragment of RPCH [63]. The struc-

Structure of buccalin-related peptides of molluscs

Buccalin
(Aplysia)
Buccalin b
(Aplysia)
Buccalin c
(Aplysia)
Fusinus Buccalin

ture of the Fusinus peptide is shown in Table 4.
According to the structure, it was termed
APGWamide.

APGWamide shows biological action on various
molluscan muscles [63]. These actions are qualita-
tively similar to those of RPCH. It is suspected
that APGWamide and RPCH act on the same class
of receptors at least in molluscs, though the actions
of the former are more potent than the latter.

APGWamide potentiates twitch contractions of
the radula retractor of Fusinus in response to short
electrical pulses of stimulation, whereas the pep-
tide inhibits twitch contractions of the radula prot-
ractor. In contrast to the twitches, ACh contrac-
tion of both muscles is not affected by the peptide,
suggesting that the action is on presynaptic sites.
Spontaneous contractions of the crop and the
tetanic contraction of the pharyngeal retractor of
the Japanese land snail Euhadra are inhibited by
APGWamide (Fig. 2). Phasic contractions of the
ABRM nd the pedal retractor of Mytilus in
response to repetitive electrical pulses of stimula-
tion are also inhibited by APGWamide.

In the phasic contraction of the ABRM,
APGWamide inhibits not only tension develop-
ment but also relaxation (Fig. 3A). However, the
peptide does not inhibit but rather slightly potenti-
ates ACh contraction (Fig. 3B), and further, it
does not inhibit relaxation of catch tension in
response to relaxing agents, such as serotonin,
dopamine, octopamine and CARP. Thus, it is
suspected that the peptide exhibits its inhibitory
effects by acting on presynaptic sites. This notion
is supported by the fact that APG Wamide marked-
ly inhibits or eliminates relaxation of catch of the
ABRM in response to stimulation by low frequen-
cy repetitive electrical pulses (Fig. 3C). It has
been well documented that brief repetitive elec-
trical pulses applied to the ABRM relax catch

Molluscan Neuropeptides 809

AAA tha Mu Aa Pal Maa ay

1O-2M =APGWa 1o°8mM
2 min
f
oma
ore
lo” SN ie 46 Sh lO-M

Fic. 2. Effect of APGWamide on spontaneous contractions of the crop (A) and on tetanic
contraction of the pharyngeal retractor muscle (B) of Euhadra. In B, the contraction was
evoked by stimulating the muscle at 10 min intervals with electrical pulses (10 V, 1 msec, 10
Hz, 10 pulses). Each dose of APGWamide was introduced 8 min prior to the onset of

stimulation.
emin 2min
fet | |b ies = i ea
lO” tN APGWa cen Abia
\o-"M APGWa 2 min
lOg

a ore
lO “M eee “ACh Bales: ACh Pulses
ACh

Fic. 3. Effect of APGWamide on phasic contraction in response to stimulation with high
frequency repetitive electrical pulses (15 V, 3 msec, 10Hz, 50 pulses) (A), on tonic
contraction in response to 10~° M ACh (B) and on relaxation of catch tension in response to
stimulation with low frequency repetitive electrical pulses (15 V, 3 msec, 1 Hz, 10 pulses) (C)
in the ABRM of Mytilus. Stimulation with high frequency electrical pulses was applied to the
muscle at 10 min intervals. ACh was applied for 2 min at 20 min intervals. In A and B,
APGWamide was introduced 8 min prior to eliciting the contraction.

tension by stimulating the intramuscular relaxing mide and RPCH on the foregoing molluscan mus-
nerves to release relaxing neurotransmitter [18, cles and compared them with those of APGWa-
64]. mide. The structures of the analogs, FAPGWa-

We tested the effects of two analogs of APGWa- mide and PGWamide, are shown in Table 4. All

810 M. KoBAYASHI AND Y. MUNEOKA

TABLE 4. Structure of RPCH-AKH-related peptides

pGlu-Leu-Asn-Phe-Thr-Pro-Asn-Trp-Gly-Thr-NH,*
pGlu-Leu-Thr-Phe-Thr-Pro-Asn-Trp-Gly-Thr-NH),
pGlu-Val-Asn-Phe-Thr-Pro-Asn-Trp-Gly-Thr-NH,
pGlu-Val-Asn-Phe-Ser-Pro-Gly-Trp-Gly-Thr-NH>

pGlu-Leu-Thr-Phe-Thr-Ser-Ser-Trp-Gly-NH,
pGlu-Leu-Asn-Phe-Ser-Thr-Gly-Trp-NH),
pGlu-Leu-Asn-Phe-Ser-Ala-Gly-Trp-NH)>
pGlu-Leu-Thr-Phe-Thr-Pro-Asn-Trp-NH),
pGlu-Val-Asn-Phe-Ser-Pro-Asn-Trp-NH2
pGlu-Val-Asn-Phe-Ser-Thr-Gly-Trp-NH,

pGlu-Leu-Asn-Phe-Ser-Pro-Gly-Trp-NH,**
H-Ala-Pro-Gly-Trp-NH,

H-Phe-Ala-Pro-Gly-Trp-NH>,
H-Pro-Gly-Trp-NH,

* Locust AKH.

of the peptides showed qualitatively similar actions
in all the muscles examined. Their potency order
was found to be as follows:

APGWamide >FAPGWamide >RPCH
>PGWamide

MYOACTIVE TETRADECAPEPTIDES OF
FUSINUS AND ACHATINA

We have isolated many species of neuropeptides
from the ganglia of Fusinus [24, 26, 32, 63, 65].
One of the peptides was found to be a tetra-
decapeptide that showed excitatory action on the
radula retractor of the animal [65]. This peptide
was provisionally termed Fusinus excitatory pep-
tide 4 (FEP,).

FEP, also showed excitatory action on the penis
retractor muscle and the buccal muscle of Achati-
na, and inhibitory action on the heart of the clam
Meretrix lusoria and the ABRM of Mytilus [65].
From these facts, it can be speculated that FEP,-
related peptides may be widely distributed in mol-
luscs.

Based on this speculation, we attempted to
isolate FEP,-related peptides from the ganglia of
Achatina and found two species of tetra-
decapeptides significantly homologous to FEP4.
They were provisionally termed Achatina excita-

Locusta

Carausius

Romalea
Nauphoeta, Blaberus
Manduca, Heliothis
Schistocerca

Locusta

Periplaneta
Periplaneta
Romalea, Gryllus

Pandalus

Fusinus

Synthetic peptides

** Crustacean RPCH. For details, see references [59, 60].

tory peptide 2 and 3 (AEP, and AEP3). The
structures of these myoactive tetradecapeptides
(MATPs) are as follows:

H-Gly-Phe-Arg-Gln-Asp-Ala-Ala-Ser
-Arg-Val-Ala-His-Gly-Tyr-(NH2)

H-Gly-Phe-Arg-Gly-Asp-Ala-Ala-Ser
-Arg-Val-Ala-His-Gly-Phe-NH2

H-Gly-Phe-Arg-Met-Asn-Ser-Ser-Asn
-Arg-Val-Ala-His-Gly-Phe-NH2

AEP»
AEP3

FEP,

We have not yet been able to confirm whether the
C-terminus of AEP, is amidated. However, since
the three peptides show similar action on several
molluscan muscles, it is probable that the terminus
is amidated.

Fusinus ganglia seem to have at least two more
peptides related to the MATPs. Although the
structure-determination experiments on the two
peptides have not been completed yet, and fur-
ther, the peptides have been found to show not
excitatory but inhibitory activity on the radula
retractor of Fusinus, they are considered to be
MATP-like peptides from the results of amino acid
sequence analyses. Thus, it is probable that there
exists a peptide family of MATPs. Kuroki and
Ikeda of our laboratory have found that FEP, at
low doses potentiates a twitch contraction of the
body-wall muscle of the echiuroid worm Urechis.

Molluscan Neuropeptides 811

MATP-related peptides might also be present in
non-molluscan animals.

ACHATINS

Kamatani et al. [6] isolated a neuropeptide hav-
ing an unusual structure from the ganglia of Acha-
tina and designated it achatin-I. The structure of
achatin-I is H-Gly-p-Phe-Ala-Asp-OH. This is the
first isolation of neuropeptide having D-amino acid
from the nervous system, though two naturally
occurring neuroactive peptides containing D-
amino acids have been reported, 1.e., dermorphin
and Hyp°-dermorphin from skin extracts of the
South American hylide frogs Phyllomedusa
sauvagei and P. rhodei (66, 67]. Kamatani et al.
also isolated another peptide termed achatin-II
from the ganglia. The structure of achatin-II is
H-Gly-Phe-Ala-Asp-OH, a peptide with the same
sequence as achatin-I but consisting only of L-
amino acids.

Achatin-I has been shown to induce a voltage-
dependent inward current due to sodium ions on
the identified neuron of Achatina, periodically
oscillating neuron (PON) [6], which is known to be
the most effective heart excitatory neuron [3]. In
contrast to achatin-I, achatin-II does not show any
effect on the neuron.

In our laboratory, a bioactive peptide was iso-
lated from the atria of Achatina. This peptide was
found to have a potent excitatory action on the
ventricle of the snail, with no significant effect on
the atrium. Amino acid sequence analysis of the
peptide showed that it has the same sequence as
achatins [53]. Therefore, we tested achatin-I and
-II on the ventricle and found that achatin-I en-
hanced its activity at concentrations higher than
10~° M but achatin-II did not show any effect even
at 10-> M. In the atrium, neither achatin-I nor -II
affected its activity. From these results, we con-
cluded that the peptide isolated from the atria is
probably achatin-I, suggesting that achatin-I is
present in the atrium. One of the physiological
roles of achatin-I may be to enhance cardiac
activity by acting on the ventricle as well as on
heart excitatory neurons such as PON.

DISCUSSION AND CONCLUSION

Invertebrate muscles as well as vertebrate
visceral muscles seem to be regulated by multiple
neurotransmitters and neuromodulators. In the
ARC muscle of Aplysia, for example, at least eight
species of neuropeptides, in addition to the classi-
cal neurotransmitters ACh and serotonin, are sug-
gested to be involved in its regulation [37, 38, 47,
48, 54-56, 68-70]. The peptides are SCP,, SCPp,
buccalin, buccalin b, buccalin c, myomodulin,
myomodulin b and FMRFamide. Neurotransmit-
ters and neuromodulators controlling the muscle
movement of an animal are considered to function,
in general, also as neuromediators in the central
nervous system of the animal. Therefore, the
muscles of an animal are excellent bioassay sys-
tems for uncovering new neuropeptides not only in
the peripheral tissues but also in the central ner-
vous system of the animal. Further, if muscles are
used as bioassay systems, the bioactivity of a
number of fractionated samples can be easily and
rapidly examined.

Based on this notion, we have carried out re-
search on neuropeptides of invertebrates by using
their muscles as bioassay systems, and have found
various species of peptides which exhibit biological
action on the muscles. Some of the peptides, such
as CARP, MIPs and ACEP-1 have already been
shown to have effects on central neurons [5, 39, 45,
46]. During the course of the experiments, we
began to consider that the number of species of
neuropeptides regulating an invertebrate muscle
might be far larger than we had previously sus-
pected. Fujisawa et al. [26] have suggested that at
least seventeen species of peptidic substances hav-
ing modulatory effects on the ABRM of Mytilus
are present in the muscle. These substances, in
addition to some classical biogenic amines, may be
involved in the regulation of the muscle [18].
Ikeda et al. [71, 72] separated twenty-five peptidic
substances having biological effects on the body-
wall muscle of Urechis from extracts of the ventral
nerve cords of the animal. These substances might
be involved in the regulation of function of the
body-wall.

The bioactive peptides found in our laboratory
can be classified into several groups. Neuro-

812 M. KoBAYASHI AND Y. MUNEOKA

peptides with the structure of FMRFamide or
FLRFamide at their C-terminal parts have been
found in annelids, molluscs and arthropods [12]. A
peptide terminating in FIRFamide has been iso-
lated from the nematode Ascaris suum by Cowden
et al. [73]. The Mytilus peptide ALAGDHFFRFa-
mide can be regarded as one of the FaRPs from its
actions and the sequence of four amino acids at its
C-terminal part. Thus, FaRPs characterized by the
structure of FXRFamide can be classified into four
subgroups according to the species of amino acid
residue X: FMRFamide, FLRFamide, FIRFamide
and FFRFamide.

The Fusinus inhibitory peptides terminating in
RFamide (Fusinus inhibitory RFa-peptides,
FIRPs) do not show FMRFamide-like actions, at
least on the radula retractor muscle of Fusinus and
the ABRM of Mytilus. However, the structure of
their C-terminal parts, LFRFamide, resemble that
of the above FaRPs. The sequence of four amino
acids at the C-terminal parts of MCRPs also seems
to be related to that of FaRPs.

The foregoing structures of the C-terminal parts
of FaRPs, FIRPs and MCRPs are shown in Table
5. Except the relationship between Phe and Met,
all amino acids in the same position of the sequ-
ence can be changed from one to another by a
point mutation. There may be an evolutionary
relationship among the peptides. New species of
peptides related to the known ones should be
sought after, and further, the precursor genes of
these peptides should be sequenced.

FMRFamide and FLRFamide show almost iden-
tical actions on many molluscan tissues. These two

TaBLE5. Sequence of the four amino acids at the
C-terminal part of FMRFamide-related peptides
(FaRPs), Fusinus inhibitory RFa-peptides
(FIRPs) and myomodulin-CARP-related peptides
(MCRPs)

-Phe-Met-Arg-Phe-NH),
-Phe-Leu-Arg-Phe-NH),

FaRP
-Phe-Ile-Arg-Phe-NH> nage
-Phe-Phe-Arg-Phe-NH>,
-Leu-Phe-Arg-Phe-NH), FIRPs
-Met-Leu-Arg-Leu-NH,

MCRPs

-Met-Met-Arg-Leu-NH,

peptides might act on one-class receptors in the
tissues. In the radula muscles of Rapana, on the
other hand, specific receptors for each peptide
seem to have evolved. In the case of the three
MIPs of the ABRM of Mytilus, they are also
suspected to act on one-class of receptors. In
Helix, however, some classes of receptors for
MIPRPs might exist. In this snail, more than ten
species of MIPRPs are thought to be present. It is
hard to imagine that all of the MIPRPs act on the
same class of receptors. In order to clarify the
physiological role of the MIPRPs in Helix, it is
very important to determine the exact structures of
all of the peptides and to sequence their precursor
genes. It will be necessary to carry out pharmaco-
logical and structural analyses of the receptors for
the MIPRPs.

In summary, it is believed that a great number of
neuropeptides remain unidentified in inverte-
brates. To search for them, invertebrate muscles
are considered to be useful bioassay systems. In
addition to the determination of the structure and
action of neuropeptides, it is essential to investi-
gate their precursor proteins and receptor pro-
teins. These investigations are very important not
only for elucidation of the evolutionary rela-
tionship among the peptides but also for clarifica-
tion of the physiological mechanisms of the system
in which the peptides are involved.

ACKNOWLEDGMENTS

We wish to express our sincere thanks to Professor Y.
Shimonishi and Dr. T. Takao (Osaka University) for
performing the mass spectroscopic measurements on the
peptides identified in our laboratory. We also grateful to
Professor M. J. Greenberg (Florida University) for pro-
viding us with valuable information on FaRPs. The
original work reported here was supported in part by
Grants-in-Aid from the Ministry of Education, Science,
and Culture of Japan.

REFERENCES

1 Price, D. A. and Greenberg, M. J. (1977) Science,
197: 670-671.

2 Greenberg, M. J., Payza, K., Nachman, R. J.,
Holman, G. M. and Price, D. A. (1988) Peptides, 9:
125-135.

3. Furukawa, Y. and Kobayashi, M. (1987) J. exp.

10

ill

12

its)
14

15

16

17

18

19

20

21
22

23

24

755)

26

Molluscan Neuropeptides

Biol., 129: 279-293.

Furukawa, Y. and Kobayashi, M. (1987) J. exp.
Biol., 129: 295-307.

Fujimoto, K., Ohta, N., Yoshida, M., Kubota, I.,
Muneoka, Y. and Kobayashi, M. (1990) Biochem.
Biophys. Res. Commun., 167: 777-783.

Kamatani, Y., Minakata, H., Kenny, P. T. M.,
Iwashita, T., Watanabe, K., Funase, K., Sun, X. P.,
Yongsiri, A., Kim, K. H., Novales-Li, P., Novales,
E. T., Kanapi, C. G., Takeuchi, H. and Nomoto, K.
(1989) Biochem. Biophys. Res. Commun., 160:
1015-1020.

Price, D. A., Davies, N. W., Doble, K. E. and
Greenberg, M. J. (1987) Zool. Sci., 4: 395-410.
Cottrell, G. A. (1989) Comp. Biochem. Physiol.,
93A: 41-45.

Li, C. and Calabrese, R. L. (1987) J. Neurosci., 7:
595-603.

Fujii, K., Ohta, N., Sasaki, T., Sekizawa, Y.,
Yamada, C. and Kobayashi, H. (1989) Zool. Sci., 6:
951-961.

Krajniak, K. G. and Price, D. A. (1990) Peptides,
11: 75-77.

Price, D. A. and Greenberg, M. J. (1989) Biol.
Bull., 177: 198-205.

Price, D. A. (1986) Am. Zool., 26: 1007-1015.
Price, D. A. (1987) In “Neurobiology, Molluscan
Models”. Ed. by H. H. Boer, W. P. M. Geraerts
and J. Joosse, North Holland Publ. Co., Amster-
dam, pp. 208-214.

Painter, S. D. (1982) J. Comp. Physiol., 148: 491-
501.

Muneoka, Y. and Matsuura, M. (1985) Comp.
Biochem. Physiol., 81C: 61-70.
Muneoka, Y. and Saitoh, H. (1986)
Biochem. Physiol., 85C: 207-214.
Muneoka, Y., Fujisawa, Y., Matsuura, M. and
Ikeda, T. (1990) Comp. Biochem. Physiol., in
press.

Kawakami, H. and Kobayashi, M. (1984) Zool.
Sci., 1: 389-397.

Kobayashi, M. and Muneoka, Y. (1986) Comp.
Biochem. Physiol., 84C: 349-352.

Kobayashi, M. (1987) Experientia, 43: 981-986.
Kobayashi, M. and Muneoka, Y. (1989) Biol. Bull.,
177: 206-209.

Hori, K., Furukawa, Y. and Kobayashi, M. (1990)
LOOVRSCUd a) 3 — O84:

Ohta, N., Matsuo, H., Kubota, I., Muneoka, Y.
and Kobayashi, M. (1989) Dobutsu seiri, 6: 213.
Kanda, T., Takabatake, I., Fujisawa, Y., Ikeda, T.,
Muneoka, Y. and Kobayashi, M. (1989) Hiroshima
J. Med. Sci., 38: 109-116.

Fujisawa, Y., Kubota, I., Kanda, T., Kuroki, Y.
and Muneoka, Y. (1990) In “Comparative Pharma-

Comp.

Dil

28

29

30

31

32,

33

34

35)

36

37

38

45

46

813

cology of Neuropeptides”.Ed. by G. B. Stefano and
E. Florey, Manchester Univ. Press, Manchester, in
press.

Yanagawa, M., Fujiwara, M., Takabatake, I.,
Muneoka, Y. and Kobayashi, M. (1988) Comp.
Biochem. Physiol., 90C: 73-77.

Martin, R. and Voigt, K. H. (1987) Experientia, 43:
5S =5435

Price, D. A., Cottrell, G. A., Doble, K. E., Green-
berg, M. J., Jorenby, W., Lehman, H. K. and
Riehm. J. P. (1985) Biol. Bull., 169: 256-266.
Ebberink, R. H. M., Price, D. A., van Loenhout,
H., Doble, K. E., Riehm, J. P., Geraerts, W. P. M.
and Greenberg, M. J. (1987) Peptides, 8: 515-522.
Fujisawa, Y., Kubota, I., Ikeda, T. and Muneoka,
Y. (1990) In “Peptide Chemistry 1989”. Ed. by N.
Yanaihara, Protein Research Foundation, Osaka,
pp. 51-56.

Kanda, T., Kuroki, Y., Kubota, I., Muneoka, Y.
and Kobayashi, M. (1990) In “Peptide Chemistry
1989” Ed. by N. Yanaihara, Protein Research
Foundation, Osaka, pp. 39-44.

Shyamala, M., Fisher, J. M. and Scheller, R. H.
(1986) DNA, 5: 203-208.

Grimmelikhuijzen, C. J. P. and Graff, D. (1986)
Proc. Natl. Acad. Sci. USA, 83: 9817-9821.
Painter, S. D., Morley, J. S. and Price, D. A. (1982)
Life Sci., 31: 2471-2478.

Hirata, T., Kubota, I., Takabatake, I., Kawahara,
A., Shimamoto, N. and Muneoka, Y. (1987) Brain
Res., 422: 374-376.

Cropper, E. C., Tenenbaum, R., Kolks, M. A. G.,
Kupfermann, I. and Weiss, K. R. (1987) Proc. Natl.
Acad. Sci. USA, 84: 5483-5486.

Vilim, F. S., Cropper, E. C., Alevizos, A., Tenen-
baum, R., Kupfermann, I. and Weiss, K. R. (1989)
Soc. Neurosci, Abst., 15: 665.

Kiss, T. (1988) Experientia, 44: 998-1000.

Hirata, T., Kubota, I., Imada, M., Muneoka, Y.
and Kobayashi, M. (1989) Comp. Biochem. Phy-
siol., 92C: 283-288.

Hirata, T., Kubota, I., Imada, M. and Muneoka, Y.
(1989) Comp. Biochem. Physiol., 92C: 289-295.
Matsumura, S. and Kuwasawa, K. (1989) Zool.
Sci., 6: 1095.

Hirata, T., Kubota, I., Iwasawa, N., Takabatake,
I., Ikeda, T. and Muneoka, Y. (1988) Biochem.
Biophys. Res. Commun., 152: 1376-1382.

Hirata, T., Kubota, I., Iwasawa, N., Fujisawa, Y.,
Muneoka, Y. and Kobayashi, M. (1989) Comp.
Biochem. Physiol., 93C: 381-388.

Yongsiri, A., Takeuchi, H., Kubota, I. and
Muneoka, Y. (1989) European J. Pharmacol., 171:-
159-165.

Kiss, T. (1990) Comp. Biochem. Physiol., 95C:

47

48

49

50

51

37)

53

54

5

56

S//

58

Sy)

60

61

814

2OP—212;

Lloyd, P. E., Kupfermann, I. and Weiss, K. R.
(1987) Peptides, 8: 179-184.

Morris, H. R., Panico, M., Karplus, A., Lloyd, P.
E. and Piniker, B. (1982) Nature, 300: 643-645.
Nagle, G. T., Painter, S. D. and Blankenship, J. E.
(1989) Biol. Bull., 177: 210-217.

Rothman, B. S., Mayeri, E., Brown, R. O., Yuan,
P.-M. and Shively, J. E. (1983) Proc. Natl. Acad.
Sci. USA, 80: 5753-5757.

Rothman, B. S., Sigvardt, K. A. and Mayeri, E.
(1985) Soc. Neurosci. Abstr., 11: 482.

Scheller, R. H., Jackson, J. F., McAllister, L. B.,
Rothman, B. S., Mayeri, E. and Axel, R. (1983)
Gell,32:'7=22.

Fujimoto, K., Ohta, N., Muneoka, Y.
Kobayashi, M. (1989) Zool. Sci., 6: 1097.
Cropper, E. C., Miller, M. W., Tenenbaum, R.,
Kolks, M. A. G., Kupfermann, I. and Weiss, K. R.
(1988) Proc. Natl. Acad. Sci. USA, 85: 6177-6181.
Weiss, K. R., Cropper, E. C., Tenenbaum, R.,
Vilim, F. S. and Kupfermann, I. (1988) Soc.
Neurosci. Abstr., 14: 177.

Miller, M. W., Cropper, E. C., Eisinger, K., Vilim,
F. S., Tenenbaum, R., Beushausen, S., Brosius, J.,
Kupfermann, I. and Weiss, K. R. (1989) Soc.
Neurosci. Abstr. 15: 665.

Fernlund, P. (1974) Biochim. Biophys. Acta, 371:
304-311.

Stone, J.S., Mordue, W., Batley, K. E. and Morris,
H. R. (1976) Nature, 263: 207-211.

Penzlin, H. (1989) Naturwissenschaften, 76: 243-
De

Goldsworthy, G. and Mordue, W. (1989) Biol.
Bull., 177: 218-224.

Greenberg, M. J., Rao, K. R., Lehman, H. K.,
Price, D. A. and Doble, K. E. (1985) J. Exp. Zool.
233: 337-346.

and

62

63

64

65

66

67

68

69

70

71

2

M. KoBAYASHI AND Y. MUNEOKA

Saitoh, H. and Muneoka, Y. (1986) Hiroshima J.
Med. Sci., 35: 389-395.

Kuroki, Y., Kanda, T., Kubota, I., Fujisawa, Y.,
Ikeda, T., Miura, A., Minamitake, Y. and
Muneoka, Y. (1990) Biochem. Biophys. Res. Com-
mun., 167: 273-279.

Muneoka, Y. and Twarog, B. M. (1983) In “The
Mollusca”. Vol. 4, Ed. by A. M. S. Saleuddin and
K. M. Wilbur, Academic Press, New York, pp. 35-
76.

Ohta, N., Kanda, T., Kuroki, Y., Kubota, L.,
Muneoka, Y. and Kobayashi, M. (1990) In “Peptide
Chemistry 1989”. Ed. by N. Yanaihara, Protein
Research Foundation, Osaka, pp. 57-62.
Montecucchi, P. C., de Castiglione, R., Piani, S.,
Gozzini, L. and Erspamer, V. (1981) Int. J. Peptide
Protein Res., 17: 275-283.

Montecucchi, P. C., de Castiglione, R. and Erspam-
er, V. (1981) Int. J. Peptide Protein Res., 17: 316-
S218

Weiss, K. R., Cohen, J. L. and Kupfermann, I.
(1978) J. Neurophysiol, 41: 181-203.

Mahon, A. C., Lloyd, P. E., Weiss, K. R., Kupfer-
mann, I. and Scheller, R. H. (1985) Proc. Natl.
Acad. Sci. USA, 82: 3925-3929.

Weiss, K. R., Lloyd, P. E., Cropper, E. C., Frank-
furt, M. and Kupfermann, I. (1986) Soc. Neurosci.
Abst., 12: 947.

Ikeda, T., Kubota, I., Kitajima, Y. and Muneoka,
Y. (1990) In “Peptide Chemistry 1989”. Ed. by N.
Yanaihara, Protein Research Foundation, Osaka,
pp. 45-50.

Ikeda, T., Kubota, I., Kitajima, Y. and Muneoka,
Y. (1990) In “Comparative Pharmacology of
Neuropeptides”. Ed. by G. B. Stefano and E.
Florey, Mancheater Univ. Press, Manchester, in
press.

Cowden, C., Stretton, A. O. W. and Davis, R. E.
(1989) Neuron, 2: 1465-1473.

ZOOLOGICAL SCIENCE 7: 815-820 (1990)

In vitro Thermal Characteristics of Kt and Na™ Induced
Melanophore Responses in a Cold Ocean Teleost,
Pseudopleuronectes americanus

DEREK BURTON and BEVERLEY ANNE EVERARD

Ocean Sciences Centre and Department of Biology, Memorial University
of Newfoundland, St. John’s, Newfoundland, AIC 5S7, Canada

ABSTRACT— Winter flounder, Pseudopleuronectes americanus, off St. John’s, Newfoundland, are
adapted to seasonal cold ocean temperatures between —1° and + 15°C. The miniature neuroeffector
preparations represented by the melanophores and the integumentary sympathetic nerve endings of
scale slips from the flounder dark ocular surface display an in vitro cellular thermal sensitivity range
appropriate to the environmental temperatures experienced seasonally by this flatfish. Thus, both
melanosome aggregation evoked by K* and dispersion elicited by Na* have an optimum temperature
of approximately 20°C. Melanosome aggregation evoked by K*, mediated through noradrenalin
release from the sympathetic nerve endings, is minimally affected by temperatures between 0° and 20°C
with Qio@—10) 2nd Qio(19 20) values of 1.0 to 1.2 over the response range. Melanosome dispersion
evoked by Na™ was refractory at low temperatures with little or no initial response and subsequent
Qio(—10) values up to 2.0. Phentolamine treatment demonstrated that such refractoriness of the
dispersive response is a product of slow noradrenalin clearance rather than an attenuation of the

© 1990 Zoological Society of Japan

intracellular melanosome dispersive mechanism.

INTRODUCTION

A variety of in vitro protocols has been used
increasingly in physiological and pharmacological
experimentation on teleost chromatophores since
Spaeth’s [1] classical studies on Fundulus scale-slip
melanophores. The sympathetic neural elements
and the melanophores of such teleost preparations
constitute a miniature in vitro neuroeffector sys-
tem. These protocols often incorporate isotonic
incubation media based on K* and Na” ions as
stimulants for melanosome aggregation and dis-
persion respectively. Fujii and Oshima [2] have
recently reviewed the accumulating evidence de-
monstrating that K* stimulates the neural compo-
nent of these preparations to release noradrenalin,
which induces melanosome aggregation mediated
through a-adrenoceptors. Recent [3] pharmaco-
logical studies demonstrate that induction of
melanosome movements by K* and Na™ can be
explained by opposing neuronal membrane polar-

Accepted April 2, 1990
Received January 16, 1990

ity changes regulating noradrenalin release.
Although warm temperatures have been shown [1,
4] to aggregate melanosomes dispersed by Na” in
teleost melanophores, there has not been a com-
prehensive comparison of the thermal characteris-
tics of K* and Na™ induced melanophore re-
sponses. The purpose of the current work is to
present such an in vitro comparison for scale-slip
melanophores from the marine flatfish, Pseudo-
pleuronectes americanus, adapted to a seasonal
temperature range from approximately —1 to
+15°C associated with the Labrador current off
Newfoundland.

MATERIALS AND METHODS

Winter flounder, Pseudopleuronectes ameri-
canus, collected by SCUBA divers off the Avalon
Peninsula, Newfoundland, were maintained under
seasonal conditions of temperature and photo-
period in stock tanks supplied with running sea
water. Experiments were started after fish had
been kept for one week in individual “Plexiglas”
black aquaria (400 mm x 225 mm X 203 mm) sup-

816 D. BURTON AND B. A. EVERARD

plied with running sea water under seasonal
temperature and illumination (60 W Im above)
conditions.

Flounder scale slips were gently plucked with
fine forceps from the general background com-
ponent [5, 6] of the integumentary pattern on the
dark ocular side. The scale slips were immersed
immediately in physiological saline solution (PSS)
with the following composition in mM: NaCl,
175.0; KCl, 2.7; MgCl, 6H2O, 0.64; CaCh, 1.53;
NaHCOs;, 5.0; glucose, 5.6. This incubation was
for 30 min with equilibration to the temperature
used subsequently when the scale slip was transfer-
red for initial incubation in melanosome aggregat-
ing fluid (AF) or in dispersing fluid (DF). The
composition of AF in mM was KCl, 180.0;
KHCO3, 5.0; glucose 5.6, and that of DF was
NaCl, 180.0; NaHCO3, 5.0; glucose 5.6. A 10~*
M solution of the a-adrenoceptor antagonist phen-
tolamine mesylate (Ciba-Geigy) in DF was also
prepared. For the initial and subsequent incuba-
tions, each scale slip was maintained for 30 min at
its experimental temperature in the appropriate
medium on a thin glass microslide (0.35—0.4 ml)
mounted on a thermal microscope stage (Bailey
Instruments, Model TS-2) with automatic temper-
ature regulation.

After initial incubation for 30 min in AF or DF
at each temperature, the subsequent experimental
procedures were conducted with either the same
medium or AF-DF reversals with two initial rinses
after each reversal. During incubations, each
medium was replaced, with minimal temperature
variation, after 15 min at temperatures up to 20°C
and at 5 min intervals above 20°C. The pH of all
the incubation media was 7.6—7.8 and the bicar-
bonate adjusted with 5% CO,-95% Oz was a
satisfactory buffer for these preparations [7]. Each
scale slip was used for one experimental procedure
at a single temperature. The degree of melano-
some aggregation or dispersion was estimated at
appropriate times using dermal melanophore
index (DMI) and epidermal melanophore index
(EMI) scales adapted from the Hogben and Slome
[8] scale. In these scales, 1.0 represents complete
melanosome aggregation and 5.0 complete disper-
sion. Protocols for individual scale-slips were
performed randomly for the different tempera-

tures, and mean values for 10 scale slips were used
for constructing DMI and EMI time-response
curves for each temperature. Statistical compari-
sons were made using the Mann-Whitney U test
and the Kruskal Wallis test.

RESULTS

The experimental procedure between 0° and
25°C involved two reversals of incubation media
after an initial incubation in DF. Initial incubation
in DF resulted in complete pigment dispersion
within dermal melanophores consistently at 10°
and 20°C (Fig. 1), but in only 50% and 20% of
scale slips at 0° and 25°C, respectively. Subsequent
dermal melanosome aggregation induced by K* in
AF (Fig. 1) followed an exponential time course,
but with aggregation being attained completely in
only 30 to 40% of these scale slips between 0° and
20°C. However, the overall differences in mean
DMI after 30 min in AF (Fig. 1) were not statisti-
cally significant (Kruskal Wallis test, p>0.05).

Melanophore index

O 10 20 SO
Time (min)

Fic. 1. The effect of temperature on K* induced der-
mal melanosome aggregation after initial incubation
in melanosome dispersing fluid. DMI values=mean
+SEM (n=10 scale slips). For clarity of presenta-
tion some SEM values have been omitted. O-.--O,
0°C; @--@, 10°C; A—4, 20°C; a—a«, 25°C.

Melanophore Thermal Characteristics 817

Between 0° and 20°C, pigment aggregation within
dermal melanophores displayed more or less identi-
cal time course characteristics except that the
initial response for 2 min at 0°C was relatively
slow. Melanosome aggregation between 0° and
20°C was characterized by 50% response time
constants of 4.5 to 4.7 min and by Qjo@_10) and
Qyo10—20) estimates between 1.0 and 1.2 after the
initial 2 min of the response. The response at 25°C
was at first rapid, but it was attenuated (Fig. 1).
After incubation in AF for 30 min at temperatures
from 0° to 25°C the medium was replaced by DF
previously equilibrated to the appropriate temper-
ature (Fig. 2). Subsequent pigment dispersion
induced by Na™* in dermal melanophores was
consistently completed within 30 min at 10° and
20°C, but only 60% of scale slips at 0°C and 20% at
25°C manifested complete dispersion in this time
(Fig. 2). Exponential time-response curves charac-
terised melanosome dispersion induced by Na*

S)

Melanophore index
WG

O 10 20 50
Time (min)

Fic. 2. The effect of temperature on Na‘ induced
dermal melanosome dispersion after replacing the
melanosome aggregating fluid in Fig. 1 with melano-
some dispersing fluid at the same temperatures.
DMI values=mean+SEM for the same groups (n=
10 scale slips) as in Fig. 1. For clarity of presentation
some SEM values have been omitted. Key as in
Fig. 1.

between 9° and 20°C (Fig. 2), with statistically
significant (Kruskal Wallis test, p<0.001) overall
maximal differences after 6min. The Qj (19—20)
for the initial 2 min was 1.5, whereas there was no
observable melanosome dispersion over this time
at O°C. The Qjo_10) estimates during DF incuba-
tion were 2.0 for the response between 2 and 8 min
and 1.4 for the response between 8 and 15 min,
while the Qjoi10-20) was 1.1 for the same time
periods. Time constants for 50% responses were
10.5 min (0°C), 6.4 min (10°C) and 4.5 min (20°C).
Although initial melanosome dispersion at 25°C
was rapid, the response range was attenuated (Fig.
2)

At 30° and 35°C, the experimental procedure
involved only one reversal of AF and DF. At both
temperatures, initial incubation in AF or DF for 30
min resulted in intermediate individual DMI
values. Subsequent reversal of incubation medium
at 30°C did not induce any melanophore responses
in 50% of scale slips in AF and in 60% in DF.
Individual residual responses, up to 6 min in dura-
tion and up to 1 DMI unit in amplitude in AF and
up to 2 DMI units in DF, occurred in the remain-
ing scale slips. At 35°C, there was no response to
either reversal of incubation medium.

Comparable thermal characteristics in AF and
DF were obtained for epidermal melanophores
except that melanosome aggregation in AF was
completed in 70 to 80% of scale slips between 0°
and 20°C. In both incubation media, scale slips
from individual flounder displayed similar time-
response characteristics at each temperature be-
tween 0° and 20°C.

Scale slips from a further five black-adapted
flounder were incubated at 0° or 20°C in DF with
phentolamine after a previous incubation for 30
min in AF at the same temperature. The treat-
ment with phentolamine abolished the thermal
differential in the response to Na* displayed by
controls incubated in DF alone (Fig. 3). Thus,
after 6 min incubation with phentolamine, there
was no significant (Mann-Whiteny U test, p>0.05)
difference between the 0° and 20°C responses.
However, there were significant (Mann-Whitney
U test, p<0.001) differences between the ex-
perimental and control responses at each tempera-
ture as well as between the control responses

818 D. BuRTON AND B. A. EvERARD

Melanophore index

Q 5) 10 1S
Time (min)

Fic. 3. The effect of phentolamine (10~* M) incubation
on Na‘ induced dermal melanosome dispersion at
0° and 20°C after initial incubation in AF. DMI
values=mean+SEM (n=10 scale slips). s—a

0°C; A—4, 20°C; @--@, 0°C with phentolamine;
O-.--O, 20°C with phentolamine.

themselves. The phentolamine treatment also
resulted in complete melanosome dispersion with-
in 15 min at both temperatures (Fig. 3), whereas
DMI values for the controls were 3.5 and 4.7 at 0°
and 20°C, respectively. Likewise, incubation with
phentolamine also eliminated the thermal differ-
ential in the pigment dispersive response of epi-
dermal melanophores to Na‘.

Scale slips initially equilibrated at 10°C in either
AF (DMI and EMI, 1.0) or DF (DMI, 4.7+0.1;
EMI, 5.0) did not display any change in DMI or
EMI as the thermal stage was cooled progressive-
ly, without change in incubation medium, to
—1°C. In contrast, melanophores equilibrated
initially at 20°C in both incubation media (Fig. 4)
displayed changes from their initial high degrees of
melanosome dispersion or aggregation to inter-
mediate levels of melanosome distribution (DMI
3.8+0.2 in DF, 1.9+0.1 in AF; EMI 2.8+0.1 in
DF, 1.7+0.1 in AF) as the thermal stage was
warmed progressively to 36°C without change of
incubation medium. These DMI and EMI changes

Melanophore index

20 24 28 32 56
Temperature (°C)

Fic. 4. Effect of progressive temperature increase on
melanophores initially equilibrated at 20°C to AF
and DF. MI values=mean+SEM (n=10 scale
slips). @—@®, DMI, and @---@, EMI, in DF and
O--O, DMI, and O—O EMI in AF.

in both warm incubation media were statistically
significant (Mann-Whitney U test, p<0.001).
Progressive re-cooling of the thermal stage to 20°C
demonstrated that these effects of temperatures
above 30°C were completely irreversible.

DISCUSSION

These results demonstrate that at the cellular
and tissue levels the thermal sensitivity range in
flounder neuro-melanophore preparations is close-
ly correlated with the environmental seasonal
temperature range to which this species is adapted.
The optimum temperatrue (approximately 20°C)
for these in vitro responses is 5'C above the
seasonal maximum temperature of sea water off
Newfoundland. Incubation of flounder scale slips
with a bicarbonate buffered noradrenalin solution
using a similar in vitro protocol resulted in thermal
characteristics for melanosome aggregation com-
parable to those with K*, including the same
optimum temperature [9], contrasting with faster

Melanophore Thermal Characteristics 819

ageregation and with an optimum temperature of
40°C for tropical Poecilia. Also, low temperatures
do not abolish flounder melanosome translocation
as in reptiles [10] or partly inhibit it as in Poecilia
[9]. The relatively slow translocation rates in
flounder melanophores in vitro are consistent with
the observation that their in vivo responses to
background change are slower than those for many
teleosts [6].

Temperatures between 0° and 20°C have a great-
er influence on melanosome dispersion induced by
Na‘ than on K* induced aggregation. In addition
to the impact of intracellular processes of melano-
some translocation, in vitro time-response charac-
teristics will be influenced by ionic diffusion rates
and by the exocytosis of noradrenalin and its
subsequent clearance, which is mainly by re-
uptake into the nerve terminals [2]. In these
preparations temperatures between 0° and 20°C
have little or no overall effect on the rate of
processes evoked by K™, which are associated with
melanosome aggregation, whereas it is postulated
that the subsequent clearance of noradrenalin in
the Na* medium is inhibited by low temperatures.
The time-response characteristics with phentol-
amine provide compelling evidence demonstrating
that melanosome dispersion induced by Na™ is
attenuated by a relatively slow noradrenalin clear-
ance, particularly at low temperatures. As ionic
stimulation has threshold properties [11], and since
enhanced dispersive responses to phentolamine
are similar at 0° and 20°C some residual activity of
noradrenalin is implicated even at the higher
temperature. This phenomenon requires further
characterization. Results with the a-adrenoceptor
antagonist also indicate that, when released from
the influence of uncleared noradrenalin, the pro-
cess of melanosome dispersion can function well at
the low temperatures experienced by this species.
Our understanding of the mechanism of melano-
some translocation is incomplete [12], although the
relatively narrow thermal range of this activity
parallels that of skeletal muscle of Antarctic fish
[13, 14] conforming with the current [15, 16]
interpretations of melanosome translocation based
on the proteins of the microtubuli and the micro-
trabecular lattice. Transduction of neurotransmit-
ter signal to melanosome movement is also not

well understood, although it has recently been
postulated [17] that inhibition of adenylate-cyclase
mediated by a-adrenoceptor may decrease the
activity of a cyclic AMP kinase involved in the
functioning of melanophore microtubuli. An im-
plication of this mechanism and of recent pharma-
cological observations [3] would be the hypothesis
that K* and Na~* induce melanosome transloca-
tion indirectly by regulating the activity of such a
cyclic AMP kinase through their control of nor-
adrenalin release.

Melanosome aggregation in warm Na~ rich
media in flounder melanophores is consistent with
earlier observations on Salmo and Fundulus (1, 4,
18]. It has been demonstrated that oxygen is not a
factor in this effect of warm temperature [1].
These earlier studies do not refer to any effects of
warm K* rich solutions and the present data invite
modification of Spaeth’s conclusion [1] that
warmth attenuates melanosome dispersion in-
duced by Na~. It can now be stated that sublethal
and lethal temperatures result in intermediate
intracellular distributions of melanosomes
irrespective of their initial distribution.

These results demonstrate that low tempera-
tures, such as are experienced seasonally by winter
flounder, have little or no effect on intracellular
melanosome translocation in this species, but can
influence noradrenalin clearance. The observation
that noradrenalin clearance can be slow in this
species is relevant to the recently reported [19]
changes in malanophore in vitro sensitivity to this
neurotransmitter following experimental adapta-
tion of flounder to a white background. The
particularly slow clearance of noradrenalin at low
temperatures invites questions about the extent of
this phenomenon in vivo and its significance in cold
ocean flatfish autonomic neurophysiology, both in
experimental protocols and under environmental
conditions.

ACKNOWLEDGMENTS

Financial support was provided by an operating grant
from the National Sciences and Engineering Research
Council of Canada, and Ciba-Geigy Canada provided the
phentolamine mesylate gratis. We are also grateful to
summer student Christina Brennan for technical assist-
ance and to Daniel Genge for the graphics work.

820

REFERENCES

Spaeth, R. A. (1913) The effect of temperature on
the melanophores of fishes. J. Exp. Zool., 15: 527-
585.

Fujii, R. and Oshima, N. (1986) Control of chroma-
tophore movements in teleost fishes. Zool. Sci., 3:
13-47.

Burton, D. and Everard, B. A. (1990) A new
analysis of in vitro Na* and K* induction of teleost
melanosome movements. Comp. Biochem. Physiol.
(In press).

Robertson, O. H. (1951) Factors influencing the
state of dispersion of the dermal melanophores in
rainbow trout. Physiol. Zool., 24: 309-323.
Osborn, C. M. (1939) The physiology of color
changes in flatfish. J. Exp. Zool., 81: 479-508.
Burton, D. (1981) Physiological responses of melano-
phores and xanthophores of hypophysectomized and
spinal winter flounder, Pseudopleuronectes america-
nus Walbaum. Proc. Roy. Soc. London., B213: 217-
231. ;

Visconti, M. A. and Castrucci, A. M. L. (1985) Tris
buffer effects on melanophore aggregating re-
sponses. Comp. Biochem. Physiol., 82C: 501-503.
Hogben, L. and Slome, D. (1931) The pigmentary
effector system. VI. The dual character of the
endocrine coordination in amphibian colour change.
Proc. Roy. Soc. London., B108: 10-53.

Burton, D. (1988) A comparison of the effects of
temperature on melanophore in vitro responses to
noradrenalin in two teleost species adapted to cold
ocean and tropical conditions. J. Fish. Biol., 32:
433-442.

10

11

12

13

14

15

16

7!

18

D. BurTON AND B. A. EVERARD

Hadley, M. E. and Goldman, J. M. (1969) Physio-
logical color changes in reptiles. Am. Zool., 9: 489-
504.

Burton, D. (1989) The influence of in vitro sodium
and potassium ion ratio on teleost melanosome
intracellular motility. Experientia, 45: 1105-1108.
Hoar, W. S. (1983) “General and Comparative
Physiology”, 3rd ed. Prentice Hall, New Jersey.
Penney, R. K., Johnston, I. A. and Goldspink, G.
(1979) Temperature adaptation in fish muscle. In
“Animals and Environmental Fitness”, Vol. 2., Ed.
by R. Gilles, Pergamon Press, Oxford, pp. 105-106.
Johnston, I. A. (1985) Temperature adaptation of
enzyme function in fish mucle. In “Physiological
Adaptations of Marine Animals” Ed. by M. S.
Laverack, SEB Symposium 39, Company of Biolog-
ists, Cambridge, pp. 95-122.

Luby, K. J. and Porter, K. R. (1980) The control of
pigment migration in isolated erythrophores of
Holocentrus ascensionis (Osbeck). Cell, 21: 13-23.
Porter, K. R. and Tucker, J. B. (1981) The ground
substance of the living cell. Sci. Am., 244: 57-67.
Andersson, R. G. G., Karlsson, J. O. and Grund-
strom, N. (1984) Adrenergic nerves and the alpha-
adrenoceptor system regulating melanosome
aggregation within fish melanophores. Acta. Phy-
siol. Scand., 121: 173-179.

Smith, D. C. (1928) The effect of temperature on
the melanophores of fishes. J. Exp. Zool., 52: 183-
234.

Burton, D. and Everard, B. A. (1989) An effect of
in vitro chromatic adaptation on melanophore in
vitro sensitivity. Comp. Biochem. Physiol., 93C:
313-316.

ZOOLOGICAL SCIENCE 7: 821-829 (1990)

Effect of Culture Age on Lipofuscin Accumulation and Creatine
Phosphokinase Activity in Spontaneously Beating Rat Heart
Cells, and its Modification by Tocopherol

S. ASHA Devi!, SyUHEI KAN? and SEncHIRO KAWASHIMA~>

Zoological Institute, Faculty of Science, Hiroshima University, Hiroshima 730 and
Zoological Institute, Faculty of Science, University of Tokyo,
Tokyo 113, Japan

ABSTRACT—The relationship between creatine phosphokinase (CPK) activity and lipofuscin (LF)
accumulation in rat heart cells in primary culture was studied. CPK activity showed an increase during
culture from 0.12 +0.03 mU/zg protein (1 day) to 2.20+0.08 (21 days) with a parallel increase in LF (as
autofluorescence intensity, AFI/ug DNA) from 29.0+3.8 (1 day) to 120.0+5.4 (21 days). Heart cells
exhibited a culture age-related decrease in the average beating rate and the beating/non-beating ratio of
myocytes, and the shift in the frequency distribution pattern from that of 12-hr old cultures. Cultures
treated with 5.010 ° IU/ml of alpha-tocopherol acid succinate (vit E) for 6 or 12 days exhibited
increased CPK activity (6 days, 10.4+0.2 mU/yg protein; 12 days, 10.8+0.6). The extent of elevation
of CPK activity was, however, less with a lower concentration of tocopherol (5.0X10~* IU/ml). In
contrast, vit E was effective in reducing LF accumulation. The AFI at 21 days of culture was 46.3+1.3
by 6 days of treatment with 5.0x10~* IU/ml and it was 41.7+0.4 by 12 days of treatment. Lower

© 1990 Zoological Society of Japan

concentration of vit E was also effective.

The present results showed that the alterations in LF

accumulation and CPK activity by vit E treatment were inversely related.

INTRODUCTION

Heart cells in culture exhibit spontaneous con-
tractions, resembling the in vivo myocardium [1].
In older rats, cardiac rate and pacemaker activity
have been reported to decrease [2], similar to the
heart rate in humans [3]. A well known manifesta-
tion of cellular aging in the postmitotic cells is the
progressive accumulation of lipofuscin in the cyto-
plasm [e.g. 4-7], which has also been documented
in the ventricular myocardium of senescent rats,
cats and man [8-10]. Recently, Sohal et al. [7]
have reported the lipofuscin accumulation in cul-
tured rat heart cells, which was accelerated by
oxidative stress.

Earlier reports state that lipofuscin accumula-
tion in the myocardium can be prevented by the
administration of some agents such as centrophe-

Accepted March 3, 1990
Received December 14, 1989

' Present address: Department of Zoology, Bangalore
University, Bangalore 560056, India.

> To whom reprint requests should be addressed.

noxine in guinea pigs [11] and tocopherol in mice
[12].

Heart cells in situ and in culture derive most of
the energy from lipids, especially fatty acids and
hence utilize ATP molecules for their functions
including contraction [13]. Harary et al. [14] have
reported that the major source of energy in beating
cells in culture is derived from lipids and that when
these cells stop beating there is a shift from lipid to
carbohydrate metabolism. There also exists a
quantitative interrelationship between the degree
of depression of contractility and that of ATP and
phosphocreatine levels during inhibition of oxida-
tive phosphorylation [15].

The present work extends these ideas to the
primary culture of rat heart cells with an attempt:
(1) to study changes in the frequency distribution
of spontaneously beating cells as a function of
culture age, (2) to identify changes in lipofuscin
accumulation with culture age, (3) to examine the
relationship between lipofuscin storage and
creatine phosphokinase activity, and (4) to analyze
the effects of alpha-tocopherol on lipofuscin accu-

822 S. ASHA Devi, S. KAN AND S. KAWASHIMA

mulation and creatine phosphokinase activity.

MATERIALS AND METHODS

Culture procedure

Primary culture of heart cells derived from 0- to
1-day-old Wistar/Tw rats was performed according
to the method of Harary and Farley [1] with some
modifications. They used whole hearts, while we
used only the lower third of the ventricle after
removing the major blood vessels and epicardium,
in order to avoid the pacemaker region of the
atria. Tissue was incubated in Hank’s balanced
salt solution (HBSS-) containing 0.1% trypsin
(pancreas protease, Merck, Darmstadt) at 37°C for
30 min. The second incubation lasted for 20 min,
while the rest of the incubations was for 15 min.
Cell suspensions obtained from the first three
incubations were discarded. The subsequent sus-
pensions were dispersed in cold complete growth
medium (CGM) and centrifuged at 600 rpm for 5
min. The resulting pellet was dispersed again in
fresh CGM and plated in 60 mm diameter plastic
culture dishes. Cells were replated after three
hours [16] at a density of 2.0 10* viable cells/ml.
This density was more dilute than those of Harary
et al. [14] (4.0 10° cells in 60 mm dish) and Sohal
et al. [7] (3.0 10° cells in 35 mm dish). Medium
was changed on every alternate days.

Counting of beating

Beats were counted for at least 300 cells in each
age group from 12 hr after replating to 16 days of
culture. Beating to non-beating cell ratio was
determined for 20 myocytes per culture dish in at
least 4 different dishes arising from different sets of

pups.

Vitamin E treatments

Cultures were divided into four groups: Control
1; cells grown in normal CGM for 18 days, Control
2; cells grown in CGM containing 0.005% ethanol
for the last 6 or 12 days of 18-day culture, Exp 1;
cells grown in CGM containing vit E (5<107°
IU/ml of CGM containing 0.005% ethanol). The
treatment lasted from 6 to 18 days of culture (12
days treatment) or 12 to 18 days (6 days treat-

ment), and Exp 2; cells treated with 5<10~4
IU/ml of vit E for the last 6 or 12 days of 18-day
culture. All these groups were cultured until 21
days of culture age after changing to normal CGM
at 18 days.

Creatine phosphokinase (CPK, E.C.2.7.3.2.)

CPK activity was measured at 1, 7, 14 and 21
days of culture and for vitamin E experiments only
at 18 days of culture. The cells were given fresh
CGM 24 hr before the assay of CPK activity. They
were rinsed twice with 3 ml of cold phosphate
buffer (pH 7.2) for 2 min each. Then the cells
were scraped out from the culture dishes by a
rubber policeman and subjected to sonication for 2
min. Samples were centrifuged at 2000 rpm for 10
min. CPK activity was assayed in the supernatant
along with the total protein content.

CPK activity was determined by the coupled
enzyme assay method of Rosalki [17]. Specific
activity was expressed as mU/yg protein.

Lipofuscin (LF)

LF was extracted from the harvested cells in
ethanol-diethyl ether solvent [18] at culture ages of
1, 7, 14 and 21 days and for vitamin E experiments
only at 18 days. The autofluorescence intensity
(AFI) was measured by a spectrofluorophotometer
(RF-540, Shimadzu) at the excitation wavelength
of 360 nm and the emission wavelength of 405-470
nm (maximum emission at 435nm). Full length
emission spectra (350-700 nm) were obtained on a
data recorder, and the area beyond the line drawn
between the two readings at 405 and 470 nm was
measured with the aid of a tablet digitizer. LF
concentration was expressed in terms of AFI/yng
DNA. DNA was measured by thiobarbituric acid
method of Gold and Shochat [19].

Medium and chemicals

CGM comprised of 30% HBSS(—), 30% fetal
bovine serum (FBS, Hazelton Res. Products, St.
Louis) and 40% DME/F-12 mixture with L-
glutamine and 15mM Hepes buffer (pH 7.4).
Trypsin was dissolved in HBSS(—) and prepared
15 minutes before use. The same lot of FBS was
used throughout the present study. All other
chemicals were of the purest and spectral grade.

Aging of Heart Cells in Culture 823

Statistical analyses

Frequency distribution of beating cells was
analyzed by Kolmogorov-Smirnov two-sample
test. The effects of vit E treatment were tested by
single classification ANOVA. All other data were
subjected to Student’s t-test.

RESULTS

Figure 1 shows the recovery of muscle cells
during the steps of incubation with 0.1% trypsin.
At the end of the first 30 min of trypsin incubation,
there was a greater proportion of non-muscle cells
such as blood cells, fibroblasts and smooth muscle
cells than the myocytes (Fig. 1a). Fourth trypsi-
nization yielded a better proportion of myocytes
(Fig. 1b), which were spherical or rod-shaped with
an average contraction rate of 15.1+0.4 beats/
min. Fifth (Fig. 1c) and sixth (Fig. 1d) incubations

resulted in a high proportion of rod-shaped muscle
cells with an average contraction of 23.1+0.2 and
40.5+0.4 beats/min respectively.

For the following experiments, the pooled cells
from fourth to sixth incubations were used. At 24
hr of culture, the myocytes showed flattened pro-
files and were seen attached to the culture dish
(Fig. 2a). The cells were beating asynchronously
with respect to each other (in the average 22.0+
3.1 beats/min) as seen in Figure 3. However, at 2
or 3 days of culture, the protoplasmic connections
were developed among cells and such groups of
cells were beating synchronously (Figs. 2b and 3)
with an average beat of 34.0+2.2 to 36.0+1.3/
min. The increase in the contraction rate was
observed until 5 days of age for the single cells as
well as for the groups of cells, each comprising of
3-4 cells (Fig. 4). At 6 days there were greater
number of groups (Fig. 2c), with the average num-
ber of beats of 32.5+0.9/min for the single cells

Fic: 1.

Recovery of myocytes during several steps of trypsinization at 37°C for 15-30 min each. (a) First enzyme

incubation. Arrow indicates a blood cell. (b) Fourth incubation, arrow indicating beating rod-shaped cells. (c)
Fifth incubation, with more number of beating cells. (d) Sixth incubation resulting in an enriched cell suspension
with a greater proportion of myocytes. Scale bars: 80 um. Phase contrast.

824 S. ASHA Devi, S. KAN AND S. KAWASHIMA

ae es
Pp
7,

ne

Fic. 2. Myocytes in culture. Phase contrast. (a) Two asynchronously beating myocytes at 1 day of culture. Arrows
indicate irregular margins of the cells. (b) Three synchronously beating cells. Arrows indicate the protoplasmic
extensions. (c) Two groups of beating cells, with two cells per group, and a single cell beating asynchronously
with the other cells at 6 days of culture. (d) Cells at 10 days of culture. Beating myocytes (arrows) formed a fiber

(F). Scale bars: 70 ~m.

and 36.0+1.3/min for the groups (Fig. 4). At
more than 8 days of culture the cells had become
more confluent and arranged to form a fibre-like
mass (Fig. 2d), beating at the rate of about 20/min.
Table 1 shows a noticeable feature in the present
cultures that the beating to non-beating ratio de-
clined as culture age advanced.

Distribution of beating frequency of single cells
and groups of cells altogether varied with culture

age. At 12 hr, a large number of cells were beating
in the 2-8 range (Fig. 3). At 1 day of culture, the
peak of distribution shifted toward a greater range.
Five-day cultures had significantly extended their
range of distribution up to 93-99 beats/min (Fig.
3), and as compared to the 12-hr-old cultures the
difference was statistically significant (p<0.001).
The beating cells varied from class 2-8 to 65-71
range at 6 days, 2-8 to 23-29 range at 8 days, and

Aging of Heart Cells in Culture 825

12 hours

FREQUENCY DISTRIBUTION

(@)
olnlNionleo OLN t+ Hl OLD AIO
Re a irae
NnMnomMmonRdeaMNnnonm
Nang NO ORR OD

BEATS /MINUTE

Fic. 3. Changes with age in the frequency distribution
pattern of beating myocytes. Distributions based on
300 cells in each age group. With respect to
12-hour-old cultures, the changes in the distribution
pattern were highly significant (P<0.01, Kolmogo-
rov-Smirnov test).

only 16-22 range was observed at 16 days of
culture (Fig. 3). Vit E treatment had no effects on
the frequency distribution of beating cells (data
not shown).

Lipofuscin

LF concentration increased as a function of

Single cells
b*.*] Groups
p

BSS
(2)

wo
(2)

AVERAGE BEAT MIN”
= ny
(oe) (o)

CULTURE AGE (DAYS)

Fic. 4. Changes in the contraction rate of myocytes as a
function of culture age. Numbers of beats were
calculated in 50 single cells (10 cells/dish) and in 10
groups of cells (two groups/dish). The data of 10
days of culture were omitted from the figure, which
were almost the same as those of 8 days.

TABLE 1. Changes in the ratio of beating to non-
beating myocytes as a function of culture age

Ee haan
12°h (Ppse his 27.2+1.4 2.6
ie! oul aes SES ale 4.3
POG 3:7 aoe 1623 -Eiey Sel
3d Ase - Dee) Sosa dhe) 2.4
4d Tease iy? 28.8+1.2 2.4
8 d Speae ese 45.0+2.8* Ly
16 d Dollaeik7 44:9+1.2 IL

* Values are mean +S.E. of 4 separate cultures.
Difference from the preceding culture age: * P<
0.001 (Student’s t-test).

culture age (Fig. 5). AFI/ug DNA increased from
29.0+3.8 at 1 day to 120.0+5.4 at 21 days (P<
0.001). Cells of Exp 1 groups exhibited a signi-
ficant decrease in their AFI in response to vit E
treatment. The vit E effect was more pronounced
in cultures treated for 12 days than those treated
for 6 days (Table 2). Although heart cells which
received a lower concentration of vit E (Exp 2)
exhibited a decrease in AFI, the change was not so
marked as in the cells of Exp 1, indicating that the
extent of AFI reduction was dependent on the
concentration of vit E (P<0.001, ANOVA).

826 S. ASHA Devi, S. KAN AND S. KAWASHIMA

CPK ACTIVITY(mU/pjg PROTEIN) @---@
LIPOFUSCIN (AFI/ug DNA) @—®

7 7 14 21
CULTURE AGE (DAYS)

Fic. 5. Changes in CPK activity and lipofuscin accu-
mulation as a function of culture age. Each point
represents mean+S.E. (N=4). * P<0.001, differ-
ence from 1-day-old cultures (Student’s t-test).

Creatine phosphokinase

A progressive increase in CPK activity was
observed from 1 day to 14 days of culture (Fig. 5).
In the Exp 1 groups, cells exhibited an increased
CPK activity (Table 2), in response to 5.0 Ome
IU/ml of vit E. A similar but lesser increase was
observed in the Exp 2 group. No significant
changes were detected either in the CPK activity
or in the AFI level between the two control
groups. Our preliminary study has indicated de-
leterious effects of a higher concentration of alpha-
tocopherol acid succinate (5.0 10-7 IU/ml) in
terms of cell growth, and a much dilute concentra-
tion (5.0x10~° IU/ml) failed to elicit any signi-
ficant change neither in the LF concentration nor
in the CPK activity.

DISCUSSION

The results of this study provide evidence for the
age-related elevation in LF content of beating rat
heart cells in primary culture. LF accumulation in
the cytoplasm of postmitotic cells serves as an
important biomarker of aging [e.g. 4-7, 20]. The
increase in LF with culture age coincided with a
decrease in beat rate of single cells and groups of
cells. This age-related decline in the contractility
supports the results of Harary and Farley [1] and
Harary et al. [14] on rat heart cells in culture.
Although they described the histological changes
in the myocytes [1] and the changes in the beating
rate for different periods [14], there has been no
demonstration on the relation between their tryp-
sinization steps and the recovery of a high propor-
tion of muscle cells. Our study afforded changes in
the contraction rates during the different steps of
enzyme treatment. The increase in the average
contraction rate at the end of fifth to sixth trypsi-
nization could be interpreted in terms of a progres-
sive increase in the disorganization of myofibrils of
the myocytes, which gives cells a “pacemaker
state” to initiate the automaticity [21]. A charac-
teristic change as a function of culture age in the
present study was a highly significant difference in
the frequency distribution pattern for beating cells
compared to 12-hr-old cultures, as insured by
Kolmogorov-Smirnov test. Our findings on a wide
range of beating frequency during the initial period
of culture could be due to the presence of cells with
variable membrane sensitivities [22]. The decline
in beating frequency after 8 days of culture might
not be due to the dedifferentiation of the

TABLE 2. Effects of vitamin E on LF accumulation and CPK activity as a function of culture

age of heart cells

ea Oe ET Sy ae com? a E CoM wi E
(days) 5.0 10 5.0 10
LF 6 113.54 3.07 110.512 46.3+1.3 92.34+2.2
12 159.5+10.0 167.4+9.4 41.740.4 73.843.2
CPK 6 2.0+ 0.1 2.0+0.4 10.4+0.2 3.5 40.3
12 2.24 0.1 2.0+0.1 10.8+0.6 3.1+0.1

“ Values are mean +S.E. of 4 different cultures at the culture age of 18 days.

expressed as AFI/ug DNA, and CPK activity, in terms of mU/yg protein.

Lipofuscin (LF) is
Difference among

treatments is statistically significant; P<0.001 (ANOVA).

Aging of Heart Cells in Culture 827

myocytes, since the fibrillar nature of the cells
were Clearly visible at this stage. It is possible that
the decline was not due to the general aging of
myocytes but due to the changes of some factors
which need further investigation. Another culture
age-related manifestation was a decrease in the
beating/non-beating ratio. This was caused by the
relative increase of non-beating single cells as
compared to the beating cells forming groups of
cells. The increase in the ratio from 12 hr to day 2
can be explained as the time the cells need to
recover from trypsin digestion [22].

The heart cells exhibited an increase in CPK
activity until 14 days of culture age. It is known
that the cardiac muscle can utilize ATP molecules
for their contractions [12] and the major source of
energy is derived from lipids, especially fatty acids.
Cells in monolayer can also take amino acids from
the medium for their energy requirement [23], and
phosphocreatine acts as an important phosphagen
at the site of the myofibrils [24]. There is a tight
functional coupling between ATPase and CPK in
the heart plasma membrane. They act as an
enzyme complex using creatine phosphate for im-
mediate rephosphorylation of ADP [25-27].

It is unlikely that the observed increase in CPK
activity with culture age was due either to in-
creased leakage of CPK from damaged cells or to
the proliferation of non-muscle cells like fibro-
blasts, which can result in lowered beat rates [28],
because the DNA level was stable from 1 day to 21
days of culture (42.0—44.5 yg/dish). However, the
non-muscle cells could not be eliminated com-
pletely in spite of the differential plating.

An interesting finding of the present work was a
significant reduction in the LF accumulation in the
cells treated with vit E. Tocopherol comes in
contact with the polyunsaturated fatty acids of
lipids to stabilize the membrane while inhibiting
the free radical reactions [29]. The effects of vit E
treatment have been observed in vivo and in vitro
conditions. Vit E deficient diet for 11 months
starting from 1 month of age in mice showed a
25.3% increase in heart LF content compared to
the control level [30], and vit E-treated mice at 5
months of age exhibited a 53% decrease in LF
content of the heart of the control level [12].

Thaw et al. [31] and Sohal et al. [7] reported that

rate of lipofuscin accumulation was retarded by
antioxidants such as vitamin E. Our results also
demonstrated that the longer the treatment period
and the higher the concentration of vit E, the
greater was the extent of reduction in AFI in
cultured heart cells. Spoerri et al. [11] have
reported a similar trend for the action of cen-
trophenoxine in the myocardium of senile guinea
pigs. Much higher dose of alpha-tocopherol (5 x
10~* IU/ml) interfered with cell growth, and this
could be due to the alterations induced in the
structure and function of serum lipoproteins [32].

It is unlikely that vit E contained in the serum in
the culture medium, if any, was enough to sup-
press LF accumulation, since it is reported that
stored serum lacks the antioxidative activity [33].

A notable outcome in vit E-treated cultures in
the present study was the marked elevation in CPK
activity of treated cells. It is probable that the
lower CPK activity in the controls than vitamin
E-treated cultures at 21 days of culture was due to
vit E deficiency. Studies on human diploid cells
have indicated that vit E deficiency can lead to
high susceptibility of the membranes especially the
mitochondrial membrane to peroxidative damage,
which could possibly spread over the cytoplasm
with simultaneous damage to the macromolecules
such as proteins and interfere with enzyme activi-
ties [29]. Perhaps such a condition was reversed in
our treated cells which exhibited an increased CPK
activity with a concomitant decrease in LF accu-
mulation as compared to the cultures grown in the
normal medium.

ACKNOWLEDGMENTS

Asha Devi was a post-dectoral research fellow of the
Japan Society for the Promotion of Science during April
1988 to March 1989. This work was supported in part by
a Grant-in-Aid from the Ministry of Education, Science
and Culture, Japan (No. 02404007 to S. Kawashima).

REFERENCES

1 Harary, I. and Farley, B. (1963) In vitro studies on
single beating rat heart cells. II. Intercellular com-
munications. Exp. Cell Res., 29: 466-474.

2 Goldberg, P. B., Tumer, N. and Robert, J. (1988)
Effect of increasing age on adrenergic control of
heart rate in the rat. Exp. Geront., 23: 115-125.

10

11

1

13

14

15

16

7

828

Roberts, J. and Goldberg, P. P. (1976) Changes in
basic cardiovascular activities during the lifetime of
the rat. Exp. Age. Res., 2: 487-517.

Ohtani, R. and Kawashima, S. (1983) Reduction of
lipofuscin by centrophenoxine and chlorpromazine
in the neurons of rat cerebral hemisphere in primary
culture. Exp. Geront., 18: 105-112.

Katz, M. L., Robison, W. G. Jr., Herrmann, R. K.,
Groome, A. B. and Bieri, J. H. (1984) Lipofuscin
accumulation resulting from senescence and vitamin
E deficiency: spectral properties and tissue distribu-
tion. Mech. Age. Devel., 25: 149-159.

Davies, I. and Fotheringham, A. (1983) The effect
of lipofuscin on cellular function. Mech. Age. De-
vel., 23: 347-356.

Sohal, R. S., Marzabadi, M. R., Galaris, D. and
Brunk, U. T. (1989) Effect of ambient oxygen
concentration on lipofuscin accumulation in cultured
rat heart myocytes—a novel in vitro model of
lipofuscinogenesis. Free Red. Biol. Med., 6: 23-30.
Malkoff, D. B. and Strehler, B. L. (1963) The
ultrastructure of isolated and in situ human cardiac
age pigment. J. Cell Biol., 16: 611-616.

Jamieson, J. D. and Palade, G. E. (1964) Specific
granules in atrial muscle cells. J. Cell Biol., 23: 151-
WZ,

Goyal, V. K. (1981) Early appearance and rate of
lipofuscin pigment accumulation in human myocar-
dium. Exp. Geront., 16: 219-222.

Spoerri, P. E., Glees, P., and Ghazzani, E. L.
(1974) Accumulation of lipofuscin in the myocar-
dium of senile guinea pigs. Mech. Age. Devel., 3:
311-321.

Kruk, P. and Enesco, H. E. (1981) Alpha-
tocopherol reduces fluorescent age pigment levels in
heart and brain of young mice. Experientia, 37:
1301-1303.

Sobel, H. (1972) Effect of age on cardiac metabo-
lism. In “Recent Advances in Studies on Cardiac
Structure and Metabolism”. Ed. by E. Bajusz and
G. Rona, University Park Press, Baltimore, pp.
101-111.

Harary, I., McCarl, R. L. and Farley, B. (1966)
Studies in vitro on single beating heart cells. IX. The
restoration of beating by serum lipids and fatty
acids. Biochim. Biophys. Acta, 115: 15-22.
Doorey, A. J. and Barry, W. H. (1983) The effects
of inhibition of oxidative phosphorylation and gly-
colysis on contractility and high energy phosphate
content in cultured chick heart cells. Circ. Res. 53:
192-201.

Blondel, B., Roij, I., and Cheneval, J. P. (1971)
Heart cells in culture: A simple method for increas-
ing the proportion of myoblasts. Experientia 27:
356-358.

Rosalki, S. B. (1967) An improved procedure for

18

19

20

Zl

22

23

24

25

26

P|

28

29

30

Sih

Sy

S. ASHA Devi, S. KAN AND S. KAWASHIMA

serum creatine phosphokinase determination. J.
Lab. Clin. Med., 69: 696-705.

Schneider, W. C. (1946) Phosphorus compounds in
animal tissues. III. A comparison of methods for the
estimation of nucleic acids. J. Biol. Chem., 164:
747-751.

Gold, D. V. and Shochat, D. A. (1980) A rapid
colorimetric assay for the estimation of microgram
quantities of DNA. Biochemistry, 105: 121-125.
Brizzee, K. R. and Ordy, J. M. (1981) Cellular
features, regional accumulation and prospects of
modification of age pigments in mammals. In “Age
Pigments”. Ed. by R. S. Sohal, Elsevier/North-
Holland, Amsterdam, pp. 102-151.

Kulikowski, R. R. (1981) Myofibrillogenesis in
vitro: Implications of early cardiac morphogenesis.
In “Perspectives in Cardiovascular Research”. Ed.
by T. Pexieder, Raven Press, N. Y., pp. 367-380.
Mark, G. E. and Strasser, F. F. (1966) Pacemaker
activity and mitosis in cultures of newborn rat heart
ventricle cells. Exp. Cell Res., 44: 217-233.
Canale, E. D., Campbell, G. R., Smolich, J. J. and
Campbell, J. H. (1986) “Cardiac Muscle”. Springer
Verlag, Berlin, pp. 195-218.

Seraydarian, M. W. and Artaza, L. (1976) Regula-
tion of energy metabolism by creatine on cardiac
and skeletal muscle cells in culture. J. Mol. Cell.
Cardiol., 8: 669-678.

Chesky, J. (1977) Comparative aspects of declining
muscle function. In “Aging in Muscle”. Ed. by G.
Kalder and W. Dibattista, Raven Press, New York,
pp. 179-210.

Chesky, J. A., Lafollette, S., Travis, M. and Forta-
do, C. (1983) Effect of physical training on myocar-
dial enzyme activities in aging rats. J. Appl. Phy-
siol.: Respirat. Environ. Exercise Physiol., 55:
1349-1353.

Higuchi, M., Nishii, K. and Takenaka, F. (1979) A
comparison of enzyme activity for energy produc-
tion in the myocardium and conduction system. Jap.
Heart J., 20: 667-673.

Jourdon, P. (1980) Ultrastructure and electrical
activity of newborn rat heart aggregates. Biol. Cell,
37: 149-154.

Packer, L. and Smith, J. R. (1974) Extension of the
lifespan of cultured normal diploid cells by vitamin-
E. Proc. Natl. Acad. Sci., 71: 4753-4767.

Enesko, H. E. and Kruk, P. (1981) Dietary ristric-
tion reduces fluorescent age pigment accumulation
in mice. Exp. Geront., 4: 357-361.

Thaw, H. H., Brunk, U. T. and Collins, V. P.
(1984) Influence of oxygen tension, pro-oxidants
and antioxidants on the formation oflipid peroxida-
tion products (lipofuscin) in individual cultivated
human glial cells. Mech. Age. Devel., 24: 211-223.
Sakagami, H. and Yamada, M. (1977) Failure of

Aging of Heart Cells in Culture 829

vitamin-E to extend the lifespan of human diploid 33 Baker, H., De Angelis, B. and Frank, D. (1988)
cell line in culture. Cell Struct. Function, 2: 219- Vitamins and other metabolites in sera commonly
2271. : used for cell biology. Experientia, 44: 1007-1009.

g ; = 5 “ \ ae ; 3 Mies ub le
mn ; I i = f = ~ a ¢ ~~
- X oS
~ a a T \ r
}
|
7 ;
ig ,
. E
:
; } ; » > ote , § Lae Crue's , eas
7 } * Fy y = i}
ma : . : o), URI Did eng

ie : : ' Z, ick 2 i i, cube feet
; ~ aS PAY

5 -
}
oF 4 , 4
ry N ?
*.
~t

e = k
i]

ZOOLOGICAL SCIENCE 7: 831-839 (1990)

Similarities between the Primary Structures of Exogastrula-
Inducing Peptides and Peptide B Purified from Embryos
of the Sea Urchin, Anthocidaris crassispina

TAKASHI SUYEMITSU, YASUTO TONEGAWA! and KatsuTosHI ISHIHARA

Department of Regulation Biology, Faculty of Science,
Saitama University, Urawa 338, Japan

ABSTRACT— Subsequent to determinations of the primary structures of exogastrula-inducing peptides
A, C and D, the complete amino acid sequence of exogastrula-inducing peptide B, purified from
embryos of the sea urchin, Anthocidaris crassispina, was determined. Peptide B was a mixture of four
peptides, that is, peptides A, A(—), Bn and Bn(—). The C-terminal amino acid sequences of peptides
A(—) and Bn(—) were -Pro-Arg-Thr, while those of peptides A and Bn were -Pro-Arg-Thr-Glu. The
aspartic acid residue at position 36 in peptides A and A(—) was replaced by an asparagine residue in
peptides Bn and Bn(—). Peptides A(—), Bn and Bn(—) were composed of 51, 52 and 51 amino acid
residues and their molecular weights were calculated to be 5625, 5753 and 5624 daltons, respectively.
The exogastrula-inducing peptides can be separated into four groups: the peptide A group with
differences in C-terminal amino acids; the peptide B group with a replacement of the 36th amino acid
residue found in members of the peptide A group; and peptides C and D which have different amino
acid sequences. A summary is presented of the amino acid sequences of all the exogastrula-inducing

© 1990 Zoological Society of Japan

peptides examined.

INTRODUCTION

Exogasrula-inducing peptides are factors that
are present in the embryos of the sea urchin. They
stimulate the extrusion of the archenteron toward
the outside of the embryo and cause exogastrula-
tion when they are added exogenously to embryos
[1-4]. In a previous paper [4], we described the
purification of four exogastrula-inducing peptides
from embryos of the sea urchin, Anthocidaris
crassispina, and named these exogastrula-inducing
peptides A, B, Cand D. The complete amino acid
sequences of exogastrula-inducing peptides A, C
and D have been determined. They were com-
posed of 52, 58 and 53 amino amino acid residues,
respectively, and their molecular weights were
calculated to be 5754, 6464 and 5737 daltons,
respectively [4, 5]. However, the amino acid
sequence of exogastrula-inducing peptide B re-
mained to be determined. It was reported pre-

Accepted March 3, 1990
Received December 11, 1989
" Deceased on October 8, 1989.

viously [4] that amino acid composition of peptide
B was very similar to that of peptide A; the
apparent molecular weight of peptide B was esti-
mated to be 5500 daltons by SDS-PAGE; its
N-terminal and C-terminal sequences were Asp-
Thr- and -Glu-Thr, respectively; and there were
only two lysine residues in the molecule. There-
fore, it seemed probable that digestion by lysyl
endopeptidase would be very useful for the deter-
mination of the amino acid sequence of peptide B.

The present study was undertaken to determine
the amino acid sequence of exogastrula-inducing
peptide B by characterization of the N-terminal
sequence and peptides generated after digestion
with lysyl endopeptidase, in order to compare the
primary structures and biological activities of the
four exogastrula-inducing peptides.

MATERIALS AND METHODS

Purification of exogastrula-inducing peptide B

Exogastrula-inducing peptide B was purified
from embryos of the sea urchin, Anthocidaris

832 T. SUYEMITSU, Y. TONEGAWA AND K. ISHIHARA

crassispina, by the procedure described previously

[4].

Enzymatic digestion and separation of peptides

The reduction and S-pyridylethylation of pep-
tide B were carried out as described by Fullmer [6]
prior to enzymatic digestion and the determination
of amino acid sequences. The S-pyridylethylated
sample (100 nmol) was digested with lysyl en-
dopeptidase from Achromobacter lyticus (EC 3, 4,
21, 50) [7], purchased from Wako Pure Chemical
Ind. (Osaka, Japan), at 30°C for 4 hr with a molar
ratio of enzyme to substrate of 1: 200 in 0.5 ml of
50 mM Tris-HCl, pH 9.0. The resultant peptides
were applied to a column (4.0 mm 1.d. x 250 mm)
packed with TSK gel ODS-120T, purchased from
Tosoh Ltd. (Tokyo, Japan), and separated by
reversed-phase, high-performance liquid chroma-
tography (RP-HPLC) on a JASCO HPLC system
equipped with a TRI ROTAR-VI pump (Japan
Spectroscopic Co., Tokyo, Japan). Peptides were
eluted with a gradient of acetonitrile that con-
tained 0.1% trifluoroacetic acid.

Amino acid analysis

Amino acid analysis of the peptides was carried
out with a reaction-liquid chromatography system,
model 655 from Hitachi, Ltd. (Tokyo, Japan) after
hydrolysis of peptides for 24 hr in 6 N HCl at 110°C
[8]. Cysteine was determined as cysteic acid after
oxidation of the intact peptide by performic acid

[9].

Sequence determination

Direct Edman degradation was performed first
by manual method, described by Chang et al. [10],
for the determination of the N-terminal sequence.
Amino acid sequences were analyzed on a gas-
phase sequencer, Applied Biosystems (Foster
City, CA) 477A Protein Sequencer [11]. To deter-
mine the C-terminal sequence, 20nmol of S-
pyridylethylated peptide was dissolved in 0.2 ml of
0.1M pyridine acetate buffer, pH 5.6, which con-
tained 0.1 mM norleucine as an internal standard,
and denatured by incubation at 60°C for 20 min.
The digestion was started by the addition of 0.2
nmol of carboxypeptidase Y [12] purchased from
Oriental Yeast (Tokyo, Japan) in 5 wl of 0.1M

pyridine acetate buffer, pH5.6, and allowed to
proceed at 25°C for 1hr. Aliquots (25 pl) of the
mixture were removed at appropriate intervals and
5 wl of acetic acid were immediately added to the
samples [13]. The amino acid released were analy-
zed on the Hitachi reaction-liquid chromatography
system.

RESULTS

After the purification of exogastrula-inducing
peptide B by the procedure described previously
[4], the amino acid composition was analyzed. The
composition of peptide B is shown in Table 1,
indicating the presence of two lysine residues per
intact peptide.

The anlysis of the N-terminal sequence of the
S-pyridylethylated peptide B was performed by
automated Edman degradation and the sequence
was determined as far as the 39th residue, as
shown in Table 2, with the exception of the 36th
residue. Two amino acid residues, namely, aspar-
tic acid and asparagine, were detected as the 36th
residue. The N-terminal sequence of peptide B as
far as the 39th residue was identical with that of
peptide A, which was reported previously [4], with
the exception of the 36th residue. In the case of
peptide A, only aspartic acid was detected as the
36th residue. Two lysine residues were detected at
position 18 and 21.

The S-pyridylethylated derivative of peptide B
was digested with lysyl endopeptidase and the
resultant peptides were separated by RP-HPLC.
Four main peaks of protein were obtained, as
shown in Figure 1, and they were designated as
peptides B-1, B-2, B-3 and B-4, respectively. The
amino acid compositions of the material in the four
peaks were analyzed and they are summarized in
Table 3. The N-terminal amino acid sequences of
peptides B-2, B-3 and B-4 were first analyzed by
manual Edman degradation and found to be Asp-
Ser-Val-, Ser-Thr-Phe- and Ser-Thr-Phe-, respec-
tively. These results suggest that peptide B-1
corresponds to residues Cys’?-Glu*’-Lys”’, B-2
corresponds to Asp!-Lys'® and B-3 and B-4 corres-
pond to the C-terminal portions of peptide B.

Peptides B-3 and B-4 were sequenced up to the
31st residue (Table 4), although the amount of the

Exogastrula-Inducing Peptide B

TABLE 1.

833

Amino acid composition of exogastrula-inducing peptides B,

A, A(—), Bn and Bn(—)

Amino acid Analysis*
B
Asp 10.5
Asn
Thr D7
Ser 4.1
Glu 3.8
Gln
Pro 0.8
Gly Wed
Ala 1.0
Cys 625
Val 1.0
Ile 0.9
Tyr 3.8
Phe Dl
Lys Dye
Arg 4.3
Total 53.4

Residues/molecule
Sequence”

A A(—) Bn Bn(—)
5 5 4 4
5 5 6 6
5 5) 5 5
4 4 4 4
2 1 7) 1
2 2, 2 2
1 if 1 1
7 7 7 d
i 1 1 1
6 6 6 6
1 1 1 if
1 1 1 1
4 4 4 4
ju D 2 2
2 2, Z 2
4 4 4 4

52 51 SZ 51

* Values obtained from 24-h hydrolysis of peptide B in 6 N HC! at 110°C.
> Number are taken from sequence data.
“ Cysteine was determined as cysteic acid after perfomic acid oxidation.

31st residue that was released was very small in
both cases. Their amino acid sequences were
identical to each other with the exception of the
15th residue. The 15th residue of B-3 was aspar-
agine, while that of B-4 was aspartic acid. The
amino acid sequence of B-4 was identical to the
sequence from position 22 to 52 in peptide A, as
reported previously [4]. The sequences of B-3 and
B-4 overlap with the N-terminal sequence of the
S-pyridylethylated derivative of peptide B from
position 22 to 39 (Fig. 2). These results suggest
that peptide B is composed of at least two pep-
tides.

The C-terminal residues of B-3 and B-4, that is,
-Thr-Glu are different from the C-terminal re-
sidues of peptide B (-Glu-Thr), which were re-
ported previously [4]. In order to resolve this
discrepancy, the C-terminal sequences of peptides
B, B-3 and B-4 were re-analyzed as described in
Materials and Methods. In the case of the S-

pyridylethylated derivative of peptide B, carboxy-
peptidase Y released glutamic acid and threonine
from the peptide B, in that order, for the first
minute of the reaction. However, the level of
glutamic acid reached only 0.25 mol/mol peptide
after 1 hr, while levels of threonine, arginine and
proline were 1.0, 0.6 and 0.5 mol/mol peptide,
respectively (Fig. 3a). This result suggests that
there are two C-terminal sequences in peptide B,
namely, -Pro-Arg-Thr and -Pro-Arg-Thr-Glu, and
the proportion of the latter is about 25% of the
total. In our earlier analysis of the C-terminal
sequence of peptide B, only glutamic acid and
threonine were detected at 1 hr [4]. This result
explains why we concluded in our earlier experi-
ments that the C-terminal sequence of peptide B
was -Glu-Thr, since we only examined the levels of
glutamic acid and threonine after 1 hr.

In the case of peptide B-3, carboxypeptidase Y
released glutamic acid, threonine, arginine and

834 T. SuYEMITSU, Y. TONEGAWA AND K. ISHIHARA

TABLE 2. N-terminal sequence of exogastrula-
inducing peptide B

Cycle Residue Yield (p mol)?
1 Asp 871
2 Ser 70
3 Val 496
4 Tyr 204
5 Gln 173
6 PeC? Tees
7 Asn 193
8 Arg 343
9 Asp 469

10 Thr 45
iL Asn 154
12 Ser Di
ie) PeC? meqae
14 Asp 599
15 Gly 113
16 Phe 140
7 Gly 119
18 Lys 93
19 PeC? n.q.°
20 Glu 159
ZA Lys 76
22 Ser 11
23 Thr 16
24 Phe 76
25 Gly 64
26 Arg 41
ay) Thr 14
28 Thr 18
29 Gly 76
30 Gln 19
Sil Tyr 21
32 Ile 33}
33 PeC? mG
34 Asn 24
35 PeC? n.q.°
36 Asp 62
Asn 24
3) Asp 100
38 Gly 31
39 Tyr 15

* 1 nmol of peptide was loaded onto the sequencer.
> S-pyridylethylcysteine.
© Not quantitated.

ihe

ae |
|

=

~ 3s
er /20) 2
= FL,
<< 0.
ats

©

1:0

0)
0 10 20
Time (min)

Fic. 1. Fractionation of a lysyl endopeptidase digest of
the S-pyridylethylated derivative of exogastrula-
inducing peptide B. The S-pyridylethylated deriva-
tive (100 nmol) was digested with lysyl endopepti-
dase at 30°C for 4 hr with a molar ratio of enzyme to
substrate of 1: 200 in 0.5 ml of 50 mM Tris-HCl, pH
9.0. The resulting peptides were applied to a
column (4.0mm i.d.x250 mm) packed with TSK
gel ODS-120T. Peptides were eluted with a gradient
of acetonitrile that contained 0.1% trifluoroacetic
acid, at a flow rate of 1 ml/min. The numbers in this
Figure represent peptides B-1, B-2, B-3 and B-4,
respectively, ——, absorbance at 214 nm; ------ , per-
centage of acetonitrile.

proline from the peptide in that order for the first
10 min. After one hour, however, the levels of
glutamic acid, threonine, arginine and proline
reached 0.5, 1.0, 0.9 and 0.6 mol/mol peptide (Fig.
3b). This result suggests that there are two C-
terminal sequences in peptide B-3, that is, -Pro-
Arg-Thr and -Pro-Arg-Thr-Glu, and that the prop-
ortion of the latter is about 50% of the total. By
overlapping the amino acid sequences of the N-
terminal portion of peptide B and peptide B-3, we
determined the complete sequences of the two
peptides, which were named peptide Bn and

Exogastrula-Inducing Peptide B

TABLE 3. Amino acid compositions of peptides produced by digestion of

_ peptide B with lysyl endopeptidase*

Amino aicd B-1 B-2 B-3 B-4
Asp 4.9(5) 4.8(5) 4.9(5)
Thr 0.9(1) 3.4(4) 3.3(4)
Ser 1.7(2) 2.2(2) BE)
Glu 1.0(1) 1.0(1) 1.8 15
Pro 1.2(1) 1.2(1)
Gly 2.0(2) 4.7(5) 4.8(5)
Ala 1.2(1) 1.3(1)
Val 0.8(1)
Ile 1.0(1) 11@)
Tyr 0.5(1) 1.9(3) AG)
Phe 1.0(1) 1.1(1) 1.2(1)
Lys 1.0(1) 0.9(1)
Arg 1.0(1) 3.0(3) 2.8(3)
Total 2.0(3) 14.7(18) el 25.5(30
Sil 31)
Position 19-21 1-18 22-51 22-51
22-52 22-52
Yield (%) 84 52 41 35

835

“ Numbers in parentheses are taken from the sequence.

Bn(—), as shown in Figures2 and 4. The C-
terminal sequence of peptide Bn is -Pro-Arg-Thr-
Glu, while that of peptide Bn(—) is -Pro-Arg-Thr.
The amino acid sequences of Bn and Bn(—) were
consist with that of peptide A as far as the Sist
residue, with the exception of the 36th residue.
The residue at position 36 in peptides Bn and
Bn(—) was asparagine, while that in peptide A
was aspartic acid [4]. The amino acid compositions
of peptides Bn and Bn(—) are shown in Table 1.
The total numbers of amino acid residues were 52
and 51, respectively, and the molecular weights of
the peptides were calculated to be 5753 and 5624
daltons, respectively.

In the case of peptide B-4, carboxypeptidase Y
released threonine, glutamic acid, asparagine and
proline from the peptide in that order for the first
two minutes. The levels of threonine, glutamic
acid, arginine and proline reached 1.0, 0.15, 0.6
and 0.5 mol/mol peptide, respectively, after one
hour (Fig. 3c). This result suggests that there are
two C-terminal sequences in peptide B-4, namely,
-Pro-Arg-Thr and -Pro-Arg-Thr-Glu, and that the
proportion of the latter is about 15% of the total.

By overlapping the amino acid sequences of the
N-terminal portion of peptide B and peptide B-4,
we determined the complete sequences of the two
peptides, as shown in Figures2 and 4. One of
them is peptide A, and the other was named
peptide A(—). Peptide A(—) is missing the C-
terminal residue (Glu) from the C-terminal sequ-
ence of peptide A. The amino acid compositions
are shown in Table 1. The total numbers of amino
acid residues were 52 and 51, respectively, and the
molecular weights of peptides A and A(—) were
calculated to be 5754 and 5625 daltons, respective-
ly. The proportions of peptides A, A(—), Bn and
Bn(—) in peptide B were calculated to be 7%,
39%, 27% and 27%, respectively.

DISCUSSION

The determination of the amino acid sequences
of all the exogastrula-inducing peptides purified
from embryos of the sea urchin, Anthocidaris
crassispina, has been completed, and the sequ-
ences are summarized in Figure 4. The exogastru-
la-inducing peptides can be separated into 4

836 T. Suyemitsu, Y. TONEGAWA AND K. ISHIHARA

TABLE 4. Amino acid sequences of peptides generated by lysyl endopeptidase

B-3 B-4
Cycle a tetramer lai elena enannennenn aaan
Residue Yield(p mol)? Residue Yield(p mol)?

1 Ser 98 Ser 176
2 Thr 156 Thr — 374
3) Phe 430 Phe So)
4 Gly 395 Gly 656
> Arg 51 Arg Wy)
6 Thr 142 Thr 204
of Thr 132 Thr 169
8 Gly Syl5) Gly 401
9 Gln DS Gin 309
10 Tyr 228 Tyr 307
11 Ile 176 Ile 335

12 PeC” Ge PeC? n.q.°
13 Asn 226 Asn Da

14 PeC? n.q.° PeC? n.q.°
15 Asn 141 Asp 202
16 Asp LS) Asp Sil
17. Gly 187 Gly 105
18 Tyr 111 Tyr 185
ID Arg 65 Arg 192
20 Asn 142 Asn 192
74s Asn 168 Asn 189
22 Ala 134 Ala 163
~ Tyr 108 Tyr 52
24 Gly 151 Gly 174
25 Gly 167 Gly 200

26 PeC? n.q.° PeC? n.q.°
27 Ser 30 Ser MIS)
28 Pro TS Pro 88
29 Arg 28 Arg a5
30 Thr 14 Thr 28
ail Glu 4 Glu 3

* 1nmol of peptide was loaded onto the sequencer. ° S-pyridylethylcysteine.
© Not quantitated.

] 10 20
Asp-Ser-Val-Tyr-Gln-Cys -Asn-Arg-Asp-Thr-Asn-Ser-Cys-Asp-Gly-Phe-Gl y-Lys-Cys-Glu-Lys-Ser-Thr-Phe-Gl y-Arg-

N-terminal sequence
B-3, B-4

30 40 50
Thr-Thr-Gly-Gin-Tyr-Ile-Cys-Asn-Cys~peP-Asp-Gly-Tyr-Arg-Asn-Asn-Ala-Tyr-Gly-Gly-Cys-Ser-Pro-Arg-Thr-(Glu) OH

N-terminal sequence

B-3, B-4
Fic. 2. The peptides used to reconstruct the complete sequence. Each peptides is indicated by underlining.

Exogastrula-Inducing Peptide B 837

QO)
U
a
os a
Ck
a @)
O
£
w O05
iS
a
5)
Qo.
cme 30 60
b
ae)
o 1.6 a
©
G
AOS
18)
©
S
O
ie 30 60
Sato! wt “bebe
Oo
2
205
(oz
0 30 60
Incubation time(min)
Fic. 3. Determination of the C-terminal sequences of

peptides B, B-3 and B-4 by digestion with carboxy-
peptidase Y. Twenty nmol of S-pyridylethylated
peptide was dissolved in 0.2 ml of 0.1M pyridine
acetate buffer, pH5.6, which contained 0.1mM
norleucine as an internal standard, and denatured by
incubation at 60°C for 20min. The digestion was
started by the addition of 0.2 nmol of carboxy-
peptidase Y in 5 wl of 0.1 M pyridine acetate buffer,
pH 5.6, and allowed to proceed at 25°C for 1 hr,
Aliquots (25 ul) of the mixture were removed at
appropriate intervals and 5 wl of acetic acid were
immediately added to the samples. The amino acid
released were analyzed on the Hitachi reaction-
liquid chromatography system. a, peptide B; b,
peptide B-3; c, peptide B-4. @, glutamic acid; 0,
threonine; 4, arginine; 4, proline.

groups: the peptide A group, which is composed of
peptides A and A(—); the peptide B group, which
is composed of peptides Bn and Bn(—); and
peptides C and D. Each peptide contains six
cysteine residues per molecule. Remarkably, the
localization of the cysteine residues in each of the
peptides is identical, as shown in Figure 4, pro-
vided that some gaps are inserted into the sequ-
ences of peptide A, A(—), Bn, Bn(—) and C.

It is possible that the difference in C-terminal
amino acids in peptide groups A and B may be due
to the deletion or the addition of glutamic acid. By
contrast, the differences between peptide groups
A and B are due to the amino acid residue at
position 36 and to a significant difference in the
relative amounts of peptides A and B, as previous-
ly reported [4]. It appears, therefore, that these
peptides are encoded by separate genes. These
problem will be resolved after the determination of
the nucleotide sequence of the gene(s) for exogas-
trula-indicing peptides.

The cysteine residues in exogastrula-inducing
peptides may be involved in disulfide bonds, be-
cause peptides that were reduced to break the
disulfide bonds lost their ability to induce exogas-
trulae, as previously reported [4]. The determina-
tion of the position of the disulfide bonds and the
structure of exogastrula-inducing peptides is in
progress.

There are various polypeptides and proteins that
contain six cysteine residues per molecule or do-
main, such as epidermal growth factor (EGF) [14-
16], pancreatic trypsin inhibitor [17-19] and zinc
finger protein [20]. Their common characteristic
feature is that they contain six cysteine residues in
the following patterns: X;CX7CXsCXj9CXCX¢-
CX), (C, cysteine; X,, other amino acids and their
numbers) in the case of EGF; X4CXgCXj>-
CX7CX12CX3CX; in the case of basic trypsin in-
hibitor; XgCX,CX7CX,9CX»CX,7C in the case of
pancreatic secretory trypsin inhibitor; and CX,-
CX¢CX¢CX2CX¢C in the case of zinc finger pro-
tein (C, family). The members of the exogastrula-
inducing groups of peptides, A and B, contain six
cystein residues distrubuted in the pattern XsCX,-
CX;CX,3CXCX,,CX4_5. Note that EGF and ex-
ogastrula-inducing peptide groups A and B contain
the same structure, CXC, and that the others do

838 T. Suyemitsu, Y. TONEGAWA AND K. ISHIHARA
6 13 19 33 35 48

A DsvyQ|c| NRDTNS|c |DGFGk/c|eKsTFeRTTaQYI[C|N[c| pDGyRN-NavGc [C] SPRTE
A(-) psvvq|c| NRDTNS | c DéFak |c EKSTFGRTTGQY1| C|N| C| DDGYRN-NAYGG|C| SPRT
Bn psvvq|c| NROTNS|c |sFek |c EKSTFGRTTEQY YRN-NAYGG| C| SPRTE
#89, : | :

c pTk@G|c| ERATNN| C|NGHED|c|---VQGR-WEQY

D DTVAR|c| ERDTKN| c| DGHGT|c |QLSTFGRRTGQYI| C|F|C|DAGYRKPNSYGG|c| SPSSA

Fic. 4. Amino acid sequence homology between peptides A, A(—), Bn, Bn(—), C and D. Cysteine residues are
emphasized by boxes. Identical amino acid residues, with the exception of cysteine residues, are highlighted by
shading. The sequence are numbered according to positions in peptide. D. —, Gaps inserted into the sequences

of peptides A, A(—), Bn, Bn(—) and C.

not contain this CXC structure. Moreover, all
members of the EGF family contain the CXC
structure [21, 22]. The similarity in the localiza-
tions of cysteine residues in EGF and exogastrula-
inducing peptides suggests that exogasttula-
inducing peptides may be the homolog of EGF in
sea urchins [4, 5, 22]. Exogastrula-inducing pep-
tides are able to induce the formation of the
exogastrulae of sea urchin embryos, when, they
are added outside the embryos. However, their
role in the normal development of sea urchin
embryos remains to be investigated.

ACKNOWLEDGMENTS

Thanks are due to Miss Y. Wakabayashi of Applied
Biosystems Japan Inc. for the amino acid sequence
analysis. This work was supported by a grant to K.
Ishihara (No. 56480013) and a grant to I. Yasumasu (No.
61304009) from the Ministry of Education, Science and
Culture of Japan.

REFERENCES

1 Berg, W. E. and Akin, E. J. (1971) Inhibition of
gastrulation by the blastocoelic fluid from the sea
urchin embryo. Dev. Biol., 26: 353-356.

2 Berg, W. E. (1972) Some characteristic of the
morphologenic inhibitor in blastocoelic fluid from
sea urchin embryo. Exp. Cell Res., 75: 539-541.

3 Ishihara, K., Tonegawa, Y., Suyemitsu, T. and
Kubo, H. (1982) The blastocoelic fluid of sea urchin
embryo induces exogastrulation. J. Exp. Zool., 220:
227-233.

4 Suyemitsu, T., Tonegawa, Y..,. Asami- Yoshizumi,
T., Noguchi, S. and Ishihara, K. (1989) The exogas-

10

11

12

13

trula-inducing peptides in embryos of the sea urchin,
Anthocidaris crassispina—isolation and determina-
tion of the primary structure. Cell Differ. Dev., 26:
53-66.

Suyemitsu, T., Tonegawa, Y. and Ishihara, K.
(1989) Amino acid sequence of exogastrula-
inducing peptide C from the sea urchin, Anthocida-
ris crassispina. Biochim. Biophys. Acta, 999: 24—28.
Fullmer, C. S. (1984) Identification of cysteine-
containing peptides in protein digests by high-
performance liquid chromatography. Anal. Bio-
chem., 142: 336-339.

Masaki, T., Tanabe, M., Nakamura, K. and Soe-
jima, M. (1981) Studies on a new proteolytic
enzyme from Achromobacter lyticus M4971. I.
Purification and some enzymatic properties. Bio-
chim. Biophys. Acta, 660: 44-50.

Benson, J. R. and Hare, P. E. (1975) o-Phthalal-
dehyde: fluorogenic detection of primary amines in
the picomole range. Comparison with fluoresca-
mine and ninhydrin. Proc. Natl. Acad. Sci.
USA, 72: 619-622.

Moore, S. (1963) On the determination of cys-
teine as cysteic acid. J. Biol. Chem., 238: 235-237.
Chang, J. Y., Brauer, D. and Wittmann-Liebold, B.
(1978) Microsequence analysis of peptides and
proteins using 4-N,N-dimethyl-aminoazobenzene-
4’-isothiocyanate/phenylisothiocyanate double cou-
pling method. EFBS Lett., 93: 205-214.

Hewick, R. M., Hunkapiller, M. W., Hood, L. E.
and Dreyer, W. J. (1981) A gas-liquid solid phase
peptide and protein sequenator. J. Biol. Chem.,
256: 7990-7997.

Hayashi, R., Moore, S. and Stein, W. H. (1973)
Carboxypeptidase from yeast. J. Biol. Chem., 248:
2296-2302.

Klemm, P. (1984) Carboxy-terminal sequence de-
termination of proteins and peptides with carboxy-
peptidase Y. In “Methods in Molecular Biology”.

14

15

16

Lg!

18

Exogastrula-Inducing Peptide B

Vol. 1. Ed. by J. M. Walker, Humana Press, New
Jersey, pp. 255-259.

Savage, C. R., Jr., Inagami, T. and Cohen, S.
(1972) The primary structure of epidermal growth
factor. J. Biol. Chem., 247: 7612-7621.

Savage, C. R., Jr., Hash, J. H. and Cohen, S. (1973)
Epidermal growth factor. Localization of disulfide
bonds. J. Biol. Chem., 248: 7669-7672.

Carpenter, G. and Cohen, S. (1979) Epidermal
growth factor. Ann. Rev. Biochem., 48: 193-216.
Kassell, B., Radicevic, M., Ansfield, M. J. and
Laskowski, M.,. Sr. (1965) The basic trypsin inhibi-
tor of bovine pancreas. Biochim. Biophys. Res.
Commun., 18: 255-258.

Greene, L. J. and Bartelt, D. C. (1969) The
structure of the bovine pancreatic secretory trypsin

19

20

21

22

839

inhibitor-Kazal’s inhibitor. J. Biol. Chem., 244:
2646-2657.

Tschesche, J. and Wachter, E. (1970) The structure
of the procine pancreatic secretory trypsin inhibitor
I. Eur. J. Biochem., 16: 187-198.

Evans, R. M. and Hollenberg, S. M. (1988) Zinc
fingers: Gilt by association. Cell, 52: 1-3.

Hursh, D. A., Andrews, M. E. and Raff, R. A.
(1987) A sea urchin gene encodes a polypeptide
homologous to epidermal growth factor. Science,
237: 1487-1490.

Yang, Q., Angerer, L. M. and Angerer, R. C.
(1989) Unusual pattern of accumulation of mRNA
encoding EGF-related protein in sea urchin em-
bryos. Science, 246: 806-808.

at sped ri > f. 3 bs
AVG, zi a jel oh te

wes: ak

“Ky qneae beasts a ae ’ :

lougat acs ny
ee aay

oe

ievatt ryceitsd ty. pe

gat AR

pe ugh
; | th ety

the Se
aay at 40m

ZOOLOGICAL SCIENCE 7: 841-847 (1990)

Comparative Study of the Influence of Head and Tail Grafts on
Axial Polarity in Regeneration of the Freshwater Planarian

SHINGO KurABucH! and YOSHIKAZU KISHIDA~

Dentistry, Nippon Dental University, Tokyo 102, and *Department of Biology,
School of Education, Okayama University, Okayama 700, Japan

ABSTRACT—The influence of head and tail grafts on the prepharyngeal region was studied in
freshwater planarians by comparing the ability to reverse the axial polarity. The original polarity of a 1
mm long fragment was reversed by means of a head graft to the posterior end, which then resulted in the
regeneration of a tail from the anterior cut end. A period of more than 4 days of isolation of the
fragment was required for the manifestation of this ability. If the head graft was removed at the time of
the isolation of the fragment, the anterior and posterior cut ends developed a head, and consequently
the fragments gave rise to a bipolar-head. It the head graft was removed from the posterior end of the
fragment in which a new head had previously regenerated from the anterior cut end, the anterior
regenerating head hindered the regeneration of a second head. In the case of the tail graft to the
anterior end of the host, to regenerate a head from the posterior end the fragment had to be isolated at
least 4 days after transplantation. When the tail graft was removed at the time of the isolation of the
fragments, a bipolar-tail developed. However, if the tail graft was removed from the fragment with a
tail regenerating from the posterior cut end, an additional tail did not regenerate.

These results showed that both tail and head grafts could influence the axial polarity of regenerating
prepharyngeal fragments to nearly the same degree.

© 1990 Zoological Society of Japan

INTRODUCTION

The problem of what mechanism regulates the
axial polarity of planarian body fragments in re-
generation has attracted the attention of many
investigators for a long time [1-6]. Of several
assumptions proposed, the gradient hypothesis put
forth by Child [1, 2] has been the most influential.
This maintains that the head is the region with the
highest metabolic activity in the planarian body
and so acts as a dominant organizer. Subsequent-
ly, the effect of the head on the axial polarity in
regeneration has been examined by a number of
researchers. When a head piece was implanted
into the posterior body region and then amputa-
tion performed nearly anterior to the grafted site,
the anterior cut ends of the host regenerated a tail
instead of a head. This event was explained as
follows: the head graft reorganized the host tissue
anterior to it and caused a reversal of the axial
polarity in the region [7-10]. The function of the

Accepted March 3, 1990
Received December 12, 1989

tail, however, has been largely ignored or has not
been considered in most studies regarding plana-
rian regeneration because it was assumed in gra-
dient theory that the tail has the lowest metabolic
activity and that it is a region which is subordinate
to the anterior body region. Our previous reports,
however, suggested that a tail piece transplanted
into the prepharyngeal region can play an effective
role in the determination of axial polarity in regen-
eration [11, 12]. That is, it is expected that an
identical mechanism may underlie the reorganizing
action of the head and tail on the other tissue.
Under this considerations, the effects of head and
tail grafts on axial polarity in the exact same
prepharyngeal region were comparatively studied.

MATERIALS AND METHODS

Asexual forms of freshwater planarian, Dugesia
japonica, collected from a stream in the vicinity of
Kanazawa City, were employed exclusively as
material for the present experiment. All measured
about 15mm in length. The worms were kept
without food for at least a week prior to surgical

842 S. KURABUCHI AND Y. KISHIDA

A
Fic. 1.

7

Diagrams representing microsurgical procedures in the first (A) and second experimental series (B). A:

Procedure for preparing 1 mm long fragment combined with a head graft (A-3), 1 mm long fragment isolated
from the region anterior to the head graft (A-4) and 1 mm long fragment with regenerating new head, separated
from the head graft (A-7). B: Procedure for preparing 1 mm long fragment combined with a tail graft (B-3), 1
mm long fragment isolated from the region posterior to the tail graft (B-4) and 1 mm long fragment with

regenerating new tail, separated from the tail graft (B-7).

Undulating lines indicate amputation postions.

Rotated black triangular areas are the transplantation sites. P, pharynx.

operation. The general procedures of transplanta-
tion were the same as those described in our
previous reports [9, 12].

Two series of transplantation experiments were
performed as shown in Figure 1. In the first series,
the triangular piece taken from the head region
was implanted into the previously prepared
triangular hole. The hole had a vertex in an
anchored region of the pharynx and base at a level
anterior to it in the prepharyngeal body region,
and consequently orientation of the graft was
adjusted so as to be rotated by 180 degress (Fig. 1,
A-1). The worms which were operated on were
maintained in total darkness overnight at 4°C in a
refrigerator in order to be immobilized. Then the
successful cases wherein the graft tissue united
completely with the host tissue were selected and
their body parts posterior to the grafted site were
removed through two cuts made along oblique
contour lines posterior to the graft to cause the
prepharyngeal piece to combine with the head
graft at the posterior end (Fig. 1, A-2). They were
kept in aged tap water at about 20°C until the next
operation was performed after a different desig-
nated length of time on each group of worms.
Every day for five days, parts of some of the worms
were amputated at a distance of 1 mm in front of
the anterior suture between the graft and the host,
and sequentially 1mm body fragments combined

with the head graft at the posterior end were made
(Fig. 1, A-3). In addition, in some of the pre-
pharyngeal fragments combined with the head
graft at the posterior end, two amputations were
simultaneously performed on the union line be-
tween the host and the head graft, and at a level 1
mm anterior to the grafted region, and thus 1 mm
long fragments situated anterior to the head graft
were produced (Fig. 1, A-4). Furthermore, 1 mm
long fragments which were combined with a head
graft made by according to the procedure men-
tioned above immediately after transplantation of
head pieces (Fig. 1, A-5), were allowed to regener-
ate a new head at the anterior ends, at a tempera-
ture of 20°C over five days (Fig. 1, A-6). Then, the
head graft was removed and consequently 1 mm
long fragments with a regenerating new head at the
anterior end were produced (Fig. 1, A-7).

In the second series, a triangular piece taken
from the tail region was implanted with reversed
orientation into the trianglar hole made in the
prepharyngeal region. Here, the base line was set
at a level 1 mm anterior to the base line of the
triangular hole pierced for the head graft in the
first series and the vertex was directed anteriorly
(Fig. 1, B-1). After complete union between the
graft and host, the part anterior to the graft was
removed to cause the fragments to combine with
the tail graft at the anterior end (Fig. 1, B-2).

Axial Polarity in Planarian Regeneration 843

Every day for five days, several fragments belong-
ing to this group were then amputated at a level 1
mm posterior to the host-graft union line, and the
1mm long fragments with the tail gfaft at the
anterior end resulted (Fig. 1, B-3). The others
were removed from the tail graft to make the 1 mm
long fragments from the body region posterior to
the tail graft (Fig. 1, B-4). Some worms which had
the tail piece transplanted into the prepharyngeal
region were amputated at a position posterior to
the graft at the same time as the removal of the
part anterior to the graft to make the 1 mm long
fragment combine with the tail graft at the anterior
end (Fig. 1, B-5). After the tail regenerated from
the posterior cut ends for 5 days (Fig. 1, B-6), the
grafted tail was removed and subsequently the 1.0
mm long fragments with a regenerating tail at the
posterior end resulted (Fig. 1, B-7).

All specimens in these series were placed in
separate Petri dishes containing aged tap water at
about 20°C for one month, and the characteristics
of the regenerated blastema, both heads and tails,
were examined.

RESULTS

Several fragments in which no regeneration
occurred from the cut surfaces were present in
every group of experiments, but they were omitted
from the data of the present study because they
seemed to be due to wound closure by simple
fusion of the right and left margins of the cut
surface before blastema formation.

TABLE 1.

Series I. Head transplantation

When 1 mm long fragments combined with head
grafts at the posterior end were prepared within
three days after transplantation, all cases regener-
ated a head at their anterior cut end, and they
developed to be bipolar-headed specimens which
had the newly regenerated head on the naterior
side and the grafted one on the posterior side.
When such fragments were isolated on the fourth
day, the tails regenerated at their anterior cut ends
in six (40%) out of fifteen cases; that is to say, in
these cases, the polarity of the anterior blastema
was reversed by the head graft. Being amputated
anteriorly on the fifth day, polarity reversal of the
blastema increased in number; the tail regenerated
from the cut ends in nine cases (69%) out of
thirteen (Table 1).

Next, when the 1 mm long fragments, situated
just anterior to the head graft, were isolated within
two days after transplantation, all of them de-
veloped into normal regenerates as the result of
the regeneration of the head and the tail in anter-
ior and posterior cut ends respectively. But, some
of the fragments made after three days regener-
ated to become a bipolar-head by regeneration of a
new head from both anterior and posterior cut
ends. The incidence of bipolar-head formation
increased with the interval between head trans-
plantation and isolation of fragments increased:
four (36%) out of eleven on the third day, nine
(90%) out of ten on the fourth days, and all
fragments isolated on the fifth day regenerated to

Regeneration from the anterior cut ends of fragments with a

head graft at the posterior end

Days between Number of
transplantation fragments
and isolation examined

0 14

1 11

2 10

3) 10

4 5)

5 13

Total 73

Regenerated blastema

Head Tail
14(100) 0(0)
11(100) 0(0)
10(100) 0(0)
10(100) 0(0)

9(60) 6(40)

4(31) 9(69)
58(79) 15(21)

The numbers in brackets are the percentages.

844 S. KURABUCHI AND Y. KISHIDA

TABLE 2. Regeneration of fragments isolated from the region anterior

to a head graft

Type of regenerates

Days between Number of
transplantation fragments Bipolar Normal
and isolation examined ead worm
0 10 0(0) 10(100)
1 10 0(0) 10(100)
0) 10 0(0) 10(100)
3 11 4(36) 7(64)
4 10 9(90) 1(10)
5 fal 11(100) 0(0)
Total 62 24(39) 38(61)

The numbers in brackets are percentages.

be bipolar-headed as shown in Table 2.

The 1 mm long fragments combined with head
grafts at the posterior ends were isolated im-
mediately after transplantation, and they were
allowed to regenerate a new small head at the
anterior cut ends over five dyas. After the head
grafts were removed from the posterior ends of the
regenerating fragments, the posterior cut ends
frequently regenerated a tail, and so the rate of
formation of the bipolar-head was considerably
low; the bipolar-head appeared in eight cases
(28%) and the normal regenerates in 21 cases
(72%).

Series II. Tail transplantation

In cases which were isolated within three days
after transplantation of the tail piece, all of the
fragments with a tail graft at the anterior ends

TABLE 3.

regenerated a new tail from the posterior cut end.
If the isolation began on the fourth day after
transplantation of the tail piece, some fragments
regenerated a head instead of a tail from the
posterior cut ends; that is, the reversal of polarity
of the blastema occurred in four cases (25%) out of
sixteen. Moreover, if isolated on the fifth day, the
rate of polarity reversal in the posterior blastema
increased to regenerate a head from the cut ends in
eleven cases (69%) out of sixteen.

When the 1 mm long fragments which occupied
the region just posterior to the tail graft were
isolated within three days after transplantation, all
cases developed to normal worms in which a head
regenerated at the anterior end and a tail at the
posterior one. But, if the periods of isolation were
delayed for more than three days, the polarity of
the blastema from the anterior cut ends of frag-

Regeneration from the posterior cut ends of fragments with

a tail graft at the anterior end

Days between Number of
transplantation fragments
and isolation examined

0 10

1 10

Z 10

3 15

4 16

5 16

Total 76

Regenerated blastema

Head Tail
0(0) 10(100)
0(0) 10(100)
0(0) 10(100)
0(0) 15(100)
4(25) 12(75)
11(69) 5(31)
15(20) 61(80)

The numbers in brackets are the percentages.

Axial Polarity in Planarian Regeneration 845

TABLE 4. Regeneration of fragments isolated from the region post-

erior to a tail graft

Days between Number of
transplantation fragments
and isolation examined

0 10

1 10

2 10

3) 10

4 10

5 11

Total 61

Type of regenerates

Bipolar Normal
tail worm
0(0) 10(100)
0(0) 10(100)
0(0) 10(100)
0(0) 10(100)
3(30) 7(70)

10(91) 1(9)

13(22) 48(78)

The numbers in brackets are the percentages.

ments were reversed to develop a tail, and conse-
quently bipolar-tailed specimens were obtained in
several instances: in three cases (30%) out of ten
on the fourth day and ten cases (91%) out of
eleven on the fifth day.

After the 1 mm long fragments with a tail piece
at the anterior ends were allowed to regenerate a
new small tail from the posterior cut ends over five
days, the tail graft was removed from the anterior
ends of the fragments. In this case, the anterior
blastemas reversed their original polarity to be-
come a tail in eight cases (44%) and consequently
the regenerated pieces gave rise to a bipolar-tail
and the remaining ten cases (56%) regenerated a
head according to the original polarity of the
fragment.

DISCUSSION

It is generally accepted that the head graft
always dominates the host body parts at posterior
levels and reorganizes the neighbouring host tissue
to cause a reversal of the anteroposterior axial
polarity [7-10]. The experiments of the head
transplantation in the present study actually de-
monstrate 1) that the influence of the head graft
extends to a greater distance as time goes on; and
2) that it may rapidly proceed to the cut end if
amputation follows at a level adjoining the graft,
though it extends only mildly in worms without
amputation, and 3) that a newly regenerating head
was possessed the ability to prevent the regenera-
tion of an additional head. These results suggest

that either an adult or regenerating head may emit
a substance or substances which inhibit head de-
velopment and these substances may be transmit-
ted to other areas.

On the other hand, there have been few trans-
plantation studies of the tail piece of the planarian
until now, in contrast to studies of the head piece.
The former are limited to the pioneering studies of
Santos [13] and Okada and Sugino [14]. Santos
carried out a transplantation of posterior pieces of
the postpharyngeal region and observed that when
the posterior portion of the host body of the worm
was cut off just posterior to the graft, a head
instead of a tail developed from the posterior cut
surface of the host body. He put fourth the
interpretation that such head development occur-
red as the result of physiological isolation of this
region through severance of the nerve connection
between the head and posterior cut surface. The
present study, however, shows that the polarity in
prepharyngeal fragments from the same level as
used in head transplantation could be reversed by
a tail graft combined to the anterior cut end and
that the presence of the graft in the host for a given
period before isolation of the fragment was critical
for the appearance of such polarity reversal, simi-
lar to the case of head transplantation. Likewise,
as in the case of the head graft, the influence of the
tail graft on polarity was passed on more rapidly in
the amputated worms than in the worms without
amputation, and the regenerating tail possessed
the ability to hinder the regeneration of an addi-
tional tail.

846 S. KURABUCHI AND Y. KISHIDA

Medinhardt [15] postulated in his book on biolo-
gical pattern formation of animals that regenera-
tion of planarians occurs in bipolar fields. These
require two organizing centres, one at each end,
and can be created by two activator maxima, one
controlling the head formation and the other tail
formation. His assumption appears to be sub-
stantiated by the results of the present experiment
in which the head graft and the tail graft have
almost equal potential in the determination of the
polarity of the isolated fragment from the pre-
pharyngeal region at the same level in regenera-
tion. Recently, Chandebois [16-18] reexamined
the reorganization phenomenon of the planarian
tissue from the point of view of the intercalary
regeneration, and found that a head piece united
with a tail piece was restructured by the latter and
concluded that absolute dominance does not reside
in either area. This supports our results that the
tail graft, not only the head graft, can dominate the
prepharyngeal region of the host, even if there is a
head. Chandebois rejects this conclusion but out
study does not.

On the other hand, there are a considerable
number of studies about requlatory substances of
organ formation in planarian regeneration. Wolff
[19] proposed the hypothesis of successive induc-
tion and inhibition of regenerative morphogenesis;
thus, various diffusible inducing and inhibiting
substances play some roles in the morphogenetic
process to regulate organ formation. He and his
colleagues actually found the eye-inducing, the
brain-inhibiting, and the pharynx-inhibiting subst-
ances [20-22]. Ansevin and Wimberly [23] per-
formed an experiment in which the postpharyngeal
pieces were treated with actinomycin, speculating
about several substances which participate in
pharynx formation, for example, pharynx-mRNA,
new polarity-mRNA, a pharynx formation inhibi-
tor and a neutralizer to the above inhibitor. It has
been reported that, in hydras which show an
anteroposterior polarity in regeneration similar to
planarians, some neurosecretory substances which
act as inhibitors and inducers in the formation of
head and foot [24-29] are restricted to nervous
tissue in the head and foot. Also in planarians, the
nervous system is known to play an important role
in the maintenance of the normal proportion of the

isolated fragments during regeneration [30-35],
but these substances have not yet been isolated. It
is quite possible that the axial metabolic gradient
present in the planarian body is closely related to
the nervous system.

REFERENCES

1 Child, C. M. (1911) Studies on the dynamics of
morphogenesis and inheritance in experimental re-
production. I. The axial gradient in Planaria doro-
tocephala as a limiting factor in regulation. J. Exp.
Zool., 10: 265-320.

2 Child, C. M. (1941) Patterns and Problems of
Development. Univ. Chicago Press., Chicago.

3 Morgan, T. H. (1904) The control of heteromorph-
osis in planaria maculata. Wilhelm Roux’ Arch., 17:
693-695.

4 Morgan, T. H. (1904) Regeneration of heteromor-
phic tails in posterior pieces of Planaria simplicissi-
ma. J. Exp. Zool., 1: 385-394.

5 Flickinger, R. A. (1959) A gradient of protein
synthesis in planaria and reversal of axial polarity of
regenerates. Growth, 23: 251-271.

6 Flickinger, R. A. and Coward, S. J. (1962) The
induction of cephalic differentiation in regenerating
Dugesia dorotocephala in the presence of the normal
head and in unwounded tails. Devel. Biol., 5: 179-
204.

7 Sugino, H. (1940) Influence of the head pieces of
planarian implanted in the prepharyngeal region.
Ann. Zool. Japn., 19: 245-253.

8 Ansevin, K. D. (1969) The influence of a head graft
on regenration of the isolated postpharyngeal body
section of Dugesia tigrina. J. Exp. Zool., 171: 235-
248.

9 Kurabuchi, S. and Kishida, Y. (1979) The role of
the nervous system in the planarian regeneration.
III. The influence of the head and the nerve cords on
the blastema regeneration. Ann. Zool. Japn., 52:
179-187.

10 Okada, Y. K. and Kido, T. (1943) Further experi-
ments on transplantation in Panaria. J. Fac. Sci.
Tokyo Imp. Univ., 4: 601-623.

11 Kurabuchi, S. (1979) The influence of a tail graft on
regeneration in the planarian, Dugesia japonica.
Zool. Mag., 88: 8-16.

12 Kurabuchi, S. and Kishida, Y. (1990) Influence of
the tail graft on axial polarity in the planarian
regeneration. J. Exp. Zool., 253: 334-339.

13 Santos, F. V. (1931) Studies on transplantation in
planaria. Physiol. Zool., 4: 111-164.

14 Okada, Y. and Sugino, H. (1937) Transplantation
experiments in Planaria gonocephala (DUGES).
Jap. J. Zool., 7: 373-439.

15

16

i,

18

19

20

“All

22

25

24

25

Axial Polarity in Planarian Regeneration

Meinhardt, H. (1982) Models of Biological Pattern
Formation. Academic Press. London and Tokyo.
Chandebois, R. (1984) Intercalary regeneration and
level interactions in the fresh-water planarian Duge-
sia lugubris 1. The anteroposterior system. Roux’
Arch. Dev. Biol., 193: 149-157.

Chandebois, R. (1985) Intercalary regeneration and
level interactions in the fresh-water planarian Duge-
sia lugubris 11. Evidence for independent antero-
posterior and medio-lateral self-regulating systems.
Roux’ Arch. Dev. Biol., 194: 390-396.
Chandebois, R. (1985) Intercalary regeneration and
level interactions in the fresh-water planarian Duge-
sia lugubris I11. The peripheral localization of the
antero-posterior self-regurating system and the lack
of interactions between the dorsal and vetral faces.
Roux’ Arch. Dev. Biol., 194: 397-403.

Wolff, E. (1953) Les phénomenes d’induction dans
la régénération des Planaires d’eau douce. Rev.
suisse Zool., 60: 540-546.

Lender, T. (1955) Sur quelques propriétés de l’orga-
nisine de la régénéretion des yeux de la planaire
Polycelis nigra. C. R. Acad. Sc. Paris., 240: 1725-
1728.

Lender, T. (1956) L’inhibition de la régénération du
cerveau des planaires Polycelis nigra et Dugesia
lugubris en présence de broyates de tétes ou de
queues. Bull. Soc. Zool. Fr., 81: 192-199.
Ziller-Sengel, C. (1965) Inhibition de la régénéra-
tion du pharynx chez les planaires. In: “Regenera-
tion in Animals”. Ed. by V. Kiortsis and H. A. L.
Trampusch, North-Holl Pub., Amsterdam, pp. 191-
AU

Ansevin, K. D. and Winberly, M. A. (1969) Mod-
ification of regeneration in Dugesia tigrina by Acti-
nomycin D. J. Exp. Zool., 172: 349-362.

Wilby, O. K. and Webster, G. (1970) Studies on the
transmission of hypostome inhibition in hydra. J.
Embryol. exp. Morph., 24: 596-613.

Wilby, O. K. and Webster, G. (1970) Experimental
studies on axial polarity in hydra. J. Embryol. exp.

26

Zi)

28

29

30

31

32

33

34

35

847

Morph., 24: 595-613.

Wolpert, L., Hicklin, J. and Hornbruch, A. (1971)
Positional information and pattern regulation in
regeneration of hydra. Symp. Soc. exp. Biol., 25:
391-415.

Grimmelikhuijzen, C. J. P. (1979) Properties of the
foot activator from hydra. Cell Differ., 8: 267-273.
Schaller, H. C. (1976) Head regeneration in hydra is
initiated release of head activator and inhibitor.
Roux’ Arch., 180: 287-295.

Schaller, H. C. (1978) Molecular Control of Prolif-
eration and Differentiation. Ed. by J. Papaconstan-
tinou and J. Rutter, Academic Press. London, pp.
231-241.

Kishida, Y. and Kurabuchi, S. (1978) The role of
nervous system in the planarian regeneration. I.
Regeneration of body fragments deprived of ventral
nerve cords. Ann. Zool. Japn., 51: 90-99.
Kurabuchi, S. and Kishida, Y. (1978) The role of
nervous system in the planarian regeneration. II.
Regeneration of body fragments with the reversed
nerve cords. Sci. Rep. Kanazawa Univ., 23: 57-63.
Kurabuchi, S. and kishida, Y. (1988) The role of
nervous system in planarian regeneration. In “Re-
generation and Development”. Ed. by S. Inoue,
Proc. 6th M. Singer Symp. Okada Print. Pub.
Comp. Maebashi., pp. 99-110.

Sperry, P. J., Ansevin, K. D. and Tittel, F. K.
(1973) The inductive role of the nerve cord in
regeneration of isolated postpharyngeal body sec-
tions of Dugesia derotocephala. J. Exp. Zool., 186:
159-174.

Sperry, P. J. and Ansevin, K. D. (1975) Deter-
mination in regenerating tissues of Dugesia
dorotocephala: the influence of nerve cord
grafts. J. Embryol. exp. Morph., 33: 85-93.
Sugino, H. (1964) On the role often ventral nerve
cords in the regeneration in the common Japanese
freshwater planarian. Mem. Osaka Univ. Lib. Art.
Educ., Ser. B, Nat. Sci., 13: 109-115.

4

Peete cas

simian tine bie geet

=e ptinabeinieeier

ea. ce

ee ont | (Sie i

at eee z
” alate eae Ute ;
f incsag coh a

hgh ge
Sen We a

cael 4

ety gra Oa
ei ae
Le |
a
ous ey ee
« — or ‘ 5
17 MT 3
- '
” os :
P,
» p
i vt
4 e
4
yy
=
—
}
w i
f

* ee ran Sensi te

3 Ot be pita: ee ‘an
: Tae ee sien eee :
nem te
ener = ania stat

ee
me
a

ae

: ‘a
Ce edie fil a a2
Ele PL Sa sao SEY

at = o
f te eee
hy A © 4 $3 r
eePerrig ‘
i iF al
> eae j
y y ty
7 pa
ott Ee
“|
F ot " rl ¥
‘ 1
‘ 3
7 ry }
f 7
Mend
~ f
5
‘
nd ~
t
>
4 ty =:
= a
j
F
aS =
9 2
Z }
r 2

a

cpanel

ZOOLOGICAL SCIENCE 7: 849-860 (1990)

Changes in Morphology, Buoyant Density and Protein Composition
in Differentiation from the Reproductive Short Form to the
Infectious Long Form of Holospora obtusa, a Macronucleus-

Specific Symbiont of the Ciliate Paramecium caudatum

MASAHIRO FUJISHIMA, KAZUMI NAGAHARA and YOSHIYUKI KOJIMA

Biological Institute, Faculty of Science, Yamaguchi University,
Yamaguchi 753, Japan

ABSTRACT—To know changes of Holospora obtusa of Paramecium caudatum during its differentia-
tion from the reproductive short form to the infectious long form, the bacteria were isolated by Percoll
density gradient centrifugation from homogenates of host cells or isolated macronuclei. H. obtusa
increased its buoyant density in the differentiation from 1.09 g/ml to 1.11 g/ml, 1.13 g/ml and 1.16 g/ml,
sequentially. Elongation completed at the stage of buoyant density 1.09 g/ml. Then, dispersed
nucleoids which were stained with a DNA specific fluorochrome, 4’ ,6-diamidino-2-phenylindole (DAPI)
appeared at the stage of buoyant density 1.11 g/ml. The nucleoids then accumulated in about a
half-length of the bacterial cell and two small dotted regions in the DAPI-positive part began to show
strong fluorescence at the stage of buoyant density 1.13 g/ml. Finally, the space between the two dotted
DAPI-positive regions narrowed and the bacterium differentiated to the infectious long form at the
stage of buoyant density 1.16 g/ml. Only the infectious long form could penetrate the macronuclear
membranes when the bacteria were added to host cells. The reproductive short forms were rapidly lysed
when the short form bearing paramecia were homogenized or the isolated short forms were suspended
in Paramecium extract, suggesting that the bacterial cell wall became resistant to the host lysosomal
enzymes in the differentiation. Two-dimensional SDS polyacrylamide gel analyses of polypeptide
compositions indicated that more than 62% of total polypeptides were replaced to different polypeptide

© 1990 Zoological Society of Japan

species in the differentiation.

INTRODUCTION

The gram negative bacterium Holospora obtusa
is a macronucleus specific symbiont of the ciliate
Paramecium caudatum [1|. The bacterium changes
its morphology in well accordance with the growth
of the host cells. The reproductive short form
(about 1-2 ~m in length) is found predominantly
in the macronucleus of vegetatively growing host
cells and can multiply by binary fission. When the
host cells starve and enter into stationary phase of
growth, the short form stops dividing, elongates its
longitudinal length to 13-15 um, and differentiates
to the infectious long form. Therefore, the bacter-
ium never destroys the host macronuclear en-
velope by their excess multiplications. This dif-

Accepted March 6, 1990
Received February 3, 1990

ferentiation occurs within 2 days after beginning of
starvation. The infectious long form can infect the
macronucleus (=invade into the host macronuc-
leus penetrating the host nuclear envelope), via
the food vacuoles, when homogenates of bacteria-
bearing host cells were added to paramecia, but
reproductive short forms and intermediate forms
in the differentiation cannot infect the nucleus and
some of them were digested in food vacuoles [2-
5]. Therefore, it is suggested that the bacterium
becomes resistant against the host lysosomal en-
zymes during the differentiation. The infectious
long form always penetrates the macronuclear
envelope with one end but never with its other [1,
4]. This special tip can be distinguished from the
other as a dark-looking part by the phase-contrast
microscope and as being homogeneously electron-
translucent by the transmission electron micro-
scope, and is developed during the differentiaiton

850 M. FusJISHIMA, K. NAGAHARA AND Y. KOJIMA

[1, 5, 6]. AH. obtusa infects the macronuclei of
closely related Paramecium species but never in-
fects their micronuclei [4, 7]. Therefore, this
special tip seems to be responsible for nucleus- and
species-specific infectivity of the bacterium. Thus,
easily inducible differentiation in H. obtusa pro-
vides us opportunities to investigate general prob-
lems for endosymbiosis, such as how endosym-
bionts can acquire their abilities for protection
from the host lysosomal enzymes and for recogni-
tion of and invasion to the target organelle of the
host cells. To know changes of H. obtusa in the
differentiation, we isolated bacteria in various
developmental stages of the differentiation by Per-
coll density gradient centrifugations from the host
cells and examined their morphologies, buoyant
densities, protein compositions, and exact timing
for expression of the infectivity in the differentia-
tion.

MATERIALS AND METHODS

Strains and cultures

The cells used in this study was Paramecium
caudatum syngen X (Tsukii, unpubl.), strain
GT703s2. This strain was supplied from Dr.
Hiwatashi’s stock cultures at Tohoku University.
The original Holospora obtusa-bearing strain C101
(syngen unknown) was collected in Munster,
FRG, by Dr. H.-D. Gortz. Strain GT703s2 was
subsequently infected with H. obtusa and used for
obtaining the symbionts in this study. The culture
medium used was 1.25% (w/v) fresh lettuce juice
in modified Dryl’s solution [8] (KH 2PO, was used
instead of NaH»,PO,4-:2H>O) inoculated with a non-
pathogenic strain of Klebsiella pneumoniae one
day before use [9]. In ordinary cultures for H.
obtusa-free cells, several hundred cells were inocu-
lated into 2 ml culture medium and then 4 ml, 10
ml and 10 ml of fresh medium were added on
successive days. But for H. obtusa-bearing cells,
several hundred cells were inoculated into 4 ml
culture medium and 8ml and 14ml of culture
medium were added on successive days to keep
cells at log phase of growth. For mass cultures of
H. obtusa-bearing cells, about 210° cells cul-
tured previously in test tubes were inoculated into

a 5L fiask containing 3.2L of culture medium
inoculated with K. pneumoniae one day before
use. In six to seven days the cultures became the
stationary phase of growth, and majority of H.
obtusa differentiated to the infectious long forms
in the host macronucleus. Cultures were kept at
DSC

Isolation of the reproductive short forms of H.
obtusa and those in the elongation process

The bacteria of the reproductive short forms and
bacteria at various lengths of the elongation pro-
cess were isolated as follows. All processes for the
isolation were performed at 4°C. H. obtusa-
bearing cells from 16 L culture in the late log phase
of growth were strained through eight layers of fine
gauze to remove gross debris. The cells were then
harvested by continuous centrifugation in the 15N
rotor of Tomy RS-20IV centrifuge at 700 x g for 20
min, and washed 3 times by suspending in 100 ml
modified Dryl’s solution and centrifuging in an
oil-test centrifuge (Kokusan, H-210A) at 300xg
for 3 min. The cells were washed finally in 10 ml
TSCM buffer (40mM Tris-HCl, pH7.9, 3mM
CaCl,, 8mM MgCl) [10] containing 0.25 M suc-
rose and re-centrifuged. The pellet was suspended
in 10 ml ice-cold TSCM buffer containing 0.25 M
sucrose, 0.2% (v/v) Nonidet P-40 and 0.1mM
phenylmethylsulfonyl fluoride (PMSF, Sigma) in
20 ml beaker and stirred slowly with a magnetic
stirrer for 5 min. When about 50% of cells were
lysed, the cell suspension was passed through a 20
ml pipette several times to destroy the cell cortex
completely. The lysate was diluted with an equal
volume of TSCM buffer containing 0.25 M suc-
rose, layered over a cushion of 10 ml TSCM buffer
containing 1.6 M sucrose and 0.1mM PMSF in a
50 ml glass centrifuge tube, and centrifuged in the
TS-7 swing rotor (Tomy) at 700g for 10 min.
Macronuclei sedimented to the bottom of the tube
were suspended in 3.5 ml of 0.25 M sucrose and
homogenized with 5-6 strokes in a Teflon
homogenizer. Then, the homogenates were centri-
fuged in 45% (v/v) Percoll (Pharmacia) diluted
with 0.25 M sucrose in the 3N angle rotor (Tomy)
at 40,000<g for 1hr. The reproductive short
forms and bacteria of elongation process formed a
band at the middle of the tube. The band was

Differentiation in Holospora obtusa 851

collected by drown up with a SJ-1211 perista pump
(ATTO), and washed 3 times with 10 ml of Na-
phosphate buffer, pH 7.0, by centrifuging at 4,000
xg for 3min to remove Percoll. The isolated
bacteria were stored suspending in 1 ml of 10 ml
Na-phosphate buffer, pH 7.0, at 4°C.

Isolation of premature long forms of H. obtusa

Premature long forms in the differentiation were
isolated as follows. H. obtusa-bearing cells from 16
L culture were harvested at early stationary phase
of growth and washed with modified Dryl’s solu-
tion as described above. The cells were then
washed finally in 10 ml of 0.25M sucrose and
re-centrifuged. The pellet was then suspended in
an equal volume of 0.25 M sucrose and homoge-
nized in a Teflon homogenizer. Then, the
homogenates were centrifuged in 45% (v/v) Per-
coll as described above. Bacteria formed 4 diffe-
rent bands in the centrifuge tube. Bacteria in each
band were collected, washed and stored at 4°C.

Isolation of infectious long forms of H. obtusa

Infectious long forms of H. obtusa were isolated
as follows. H. obtusa-bearing cells from 16L
culture were harvested at the stationary phase of
growth and washed with modified Dryl’s solution
and 0.25 M sucrose and homogenized as described
above. The homogenates were mixed with 9
volumes of 0.25 M sucrose and centrifuges in the
TS-7 swing rotor at 1,000<g for 10min. The
supernatant containing most of the mitochondria,
cilia and trichocysts was discarded. The pellet
constituted with crude H. obtusa fraction was
made 1 ml by adding 0.25 M sucrose. Then, the
bacterial suspension was centrifuged in 75% (v/v)
Percoll as described above. In some experiment,
the homogenates were directly loaded on 75%
(v/v) Percoll density gradient centrifugation with-
out washing them by 0.25 M sucrose to avoid loss
of holosporas. After the density gradient centri-
fugation, a white band consisted of the infectious
long forms was collected, washed and stored as
described above.

Measurement of buoyant density of H. obtusa

The buoyant density of each band was deter-
mined by a fractionation of the Percoll density

gradient at 0.5ml each from a bottom of the
centrifuge tube and by measuring weight of 200 pl
of each fraction or by adding Density Marker
Beads (Pharmacia) in a centrifuge tube parallel to
the one containing experimental sample.

Light microscopy

Observations of H. obtusa in the host cells or
those isolated from the host cells were made as
follows. A small drop containing the host cells or
the isolated bacteria was placed on a slide glass,
covered with a cover glass and observed with a
Nomarski differential interference-contrast micro-
scope (Olympus BHS-N) at magnifications of x
400 or X1,000. Phase-contrast or fluorescent
microscopies of the isolated symbionts were made
as follows. The symbionts were air-dried on a slide
glass, fixed in an ethanol/acetic acid (3: 1) mixture
for 10 min at room temperature. After drying, 1
pg/ml of DAPI dissolved in deionized water was
dropped onto the preparations and covered with a
cover glass. Observations were carried out with
phase-contrast and fluorescent optics (Olympus,
BH2-RFL) at a magnificatin of < 400.

Infection

Infection of H. obtusa was achieved as follows.
Isolated bacteria and recipient cells, 0.5 ml each,
were mixed in a depression slide in the densities of
3 x 10*-3 x 10° bacteria/ml and 1,500 cells/ml, re-
spectively, at 25°C, and the cells were observed by
a Nomarski differential interference-contrast
microscope.

Two-dimensional SDS-polyacrylamide gel elec-
trophoresis (2D-SDS-PAGE)

Isolated reproductive short forms and infectious
long forms of H. obtusa were lysed by sonication in
the solubilizing solution containing 8.5M urea,
2% (v/v) Nonidet P-40, 2% (w/v) Ampholine and
5% (v/v) f8-mercaptoethanol on ice. Insoluble
materials were removed by centrifuging at 5,000 x
g for 5 min and protein contents of the super-
natants were measured spectrophotometrically
with a Protein Assay (Bio-Rad). 2D-SDS-PAGE
of polypeptides was carried out essentially as de-
scribed by O’Farrell [11]. The first dimension was
cast as 10cm column in 2.5mm diameter glass

852 M. FusISHIMA, K. NAGAHARA AND Y. KOJIMA

tubes using 1.6% (w/v) of pH 5-8 and 0.4% (w/v)
of pH3.5-10 Ampholines (LKB). After pre-
running at 200, 300 and 400 V for 15, 30.and 30
min, respectively, each tube was loaded with 20 ng

e,

e E Re
‘ :
RGB sie

of Holospora protein, and run at 400 and 800 V for
16 and 1 hr, respectively. In case of the infectious
long form, 20 ug protein could be obtained from
about 9.410’ cells. The isoelectric focusing gels

Fic. 1. Photomicrographs of isolated macronuclei of P. caudatum and isolated H. obtusa of buoyant density 1.09
g/ml of the elongation process. A, isolated macronuclei with the reproductive short forms of H. obtusa. B,
isolated macronuclei with H. obtusa in the elongation process. C, a centrifuge tube after 45% (v/v) Percoll
density gradient centrifugation at 19,000 xg for 1 hr. Note, various lengths of bacteria in the elongation process
formed a common band (arrow) of buoyant density 1.09 g/ml. D, a phase-contrast micrograph of H. obtusa
isolated from a band in C. E, DAPI fluorescence of D. Bars, 10 um.

Differentiation in Holospora obtusa 853

were then equilibrated with Laemmli’s lysis buffer
[12] containing bromophenol blue. In the second
dimension, polypeptides were electrophoresed
according to molecular weight on a SDS-
polyacrylamide slab gel consisted of a stacking gel
(120x201 mm) containing 3% (w/v) polyacryl-
amide and a separation gel (1201001 mm)
containing 10% (w/v) polyacrylamide at 15mA
and 30 mA constant current, respectively, until the
tracking dye moved to the end of the gel.
Polypeptides in the gel were stained with silver
using a Silver-stain-kit (Wako).

RESULTS

Reproductive short forms of H. obtusa and those of
elongation process

When the reproductive short forms of H. obtusa
bearing cells were homogenized, the short forms
always disappeared from the homogenates. This
suggests that the short forms are very sensitive
against host lysosomal enzymes. Therefore, at first
we isolated the macronuclei from cell lysate (Fig.
1A and 1B), and then bacteria were isolated from
homogenates of the macronuclei by 45% (v/v)
Percoll density gradient centrifugation (Fig. 1C-
1E). After the centrifugation, the bacteria formed
a white band in the centrifuge tube and buoyant
density of the band was found to be 1.09 g/ml by
using density marker beads (Pharmacia) (see
Materials and Methods). When the macronuclei
were isolated from cells in late log phase of
growth, the nuclei contained exclusively the repro-
ductive short forms (Fig. 1A). However, when the
host cells were starved a little, elongated bacteria
appeared in the isolated macronuclei (Fig. 1B) and
bacterial band became consisted not only of the
reproductive short forms but also of various
lengths of the bacteria of the elongation process
and of fully elongated ones (Fig. 1D and E). This
indicates that the bacteria begin to elongate them-
selves and complete the elongation when their
buoyant density is 1.09 g/ml. Since buoyant densi-
ty of the bacterial band was heavy enough compar-
ing with mitochondria, trichocysts and the food
bacteria, the bacterial band was formed separated
from the bands containing such contaminants.

But, when 75% (v/v) Percoll was used instead of
45%, the bacterial band could not be clearly
separated from the contaminants.

The reproductive short forms and the bacteria of
the elongation process looked entirely dark in
phase-contrast microscopy and entirely DAPI-
positive in fluorescence microscopy. However,
three different morphologies could be recognized
in the fully elongated bacteria under the phase-
contrast microscope; bacteria which looked uni-
formly dark, those which had irregularly dispersed
dark nucleoids, and those which had relatively
regular tandem repeats of dark and blight regions
(Fig. 1D). In all cases, the dark-looking regions in
the bacterial cells coincided with the DAPI-
positive regions (Fig.1E). Since the DAPI
fluorescence becomes bright in the above order,
the differences in morphology of the long forms
may come from the differential degrees of the
nucleoid condensation in the differentiation.

When the isolated bacteria were suspended in a
supernantant of centrifugation at 40,000 xg for 1
hr of the host homogenates, bacteria of short
forms and of various lengths of the elongation
process lysed within 5 min at 4°C, but fully elon-
gated bacteria were not lysed. This shows that
bacteria in the reproductive short form and bacter-
ia in the elongation process were very sensitive to
the host lysosomal enzymes, and that the bacteria
changed their cell wall resistant against the host
lysosomal enzymes during the elongation process.

Premature long forms of H. obtusa

When homogenates of cells in early stationary
phase of growth were centrifuged in 45% (v/v)
Percoll, four distinct bacterial bands appeared.
Bacteria with typical morphologies obtained from
each band were shown in Figure 2. Buoyant
densities of each band were 1.09 g/ml, 1.11 g/ml,
1.13 g/ml and 1.16 g/ml, respectively. All bacteria
in each band were fully elongated ones, and repro-
ductive short forms and those in the elongation
process were not found even in a band of buoyant
density 1.09 g/ml, suggesting that the short bacter-
ia had been lysed in the host homogenates. The
bacteria of buoyant density 1.09 g/ml showed va-
rious degrees of nucleoid condensation as shown in
Figure 1(D and E) and each one of the typical

854 M. FusISHIMA, K. NAGAHARA AND Y. KOJIMA

10 jim

Fic. 2. Photomicrographs of H. obtusa bands after 45% (v/v) Percoll density gradient centrifugation and isolated
bacteria from each bacterial band. Under a thick band of the host debris, four bacterial bands appeared.
Buoyant densities of each band are shown in left of the centrifuge tube. Bacteria are shown with phase-contrast
(left) and fluorescent micrographs (right). Each pair of the bacterial photomicrographs shows the same bacteria.

morphologies was shown in Figure2. DAPI-
positive nucleoid of the bacterium showed a
tendency to accumulate to one end of the bacteria
of buoyant density 1.11 g/ml. Two dotted DAPI-
positive small regions appeared in the nucleoid of
the bacteria of buoyant density 1.13 g/ml and a
space between the two dots becomed narrow in the
bacteria of buoyant density 1.16g/ml. This
heaviest bacteria were morphologically identified
as the infectious long forms (see next paragraph).
These continuous changes in the distribution of the
nucleoid indicate that the changes of the bacterial
buoyant densities in the differentiation proceed in
an order of 1.09 g/ml, 1.11 g/ml, 1.13 g/ml and 1.16
g/ml. It should be noted that, under a phase-
contrast microscope, DAPI positive regions cor-
responded to the dark-looking regions of the bac-
teria of buoyant densities 1.09 g/ml and 1.11 g/ml,
whereas it turned out to correspond to the bright

regioin in the bacteria of buoyant densities 1.13
g/ml and 1.16 g/ml.

Infectious long forms of H. obtusa

When homogenates of cells in stationary phase
of growth were centrifuged in 75% (v/v) Percoll,
one bacterial band consisted of the infectious long
forms was formed (Fig. 3A-D). Buoyant density
of the band was 1.16 g/ml. Comparison of num-
bers of bacteria in the host homogenates and in the
bacterial fraction collected from the Percoll densi-
ty gradient showed that the yields averaged 85%
even if the homogenates were washed with a buffer
prior to Percoll centrifugation (see Materials and
Methods). The bacterial fraction collected from
the Percoll density gradient was essentially free
from contaminations such as mitochondria, tricho-
cysts and the food bacteria (Fig. 3B). The isolated
infectious long forms consisted of three parts

Differentiation in Holospora obtusa 855

= |

Fic. 3. Photomicrographs of a band consisted of the infectious long forms of H. obtusa after 75% (v/v) Percoll
density gradient centrifugation and isolated bacteria. A, a bacterial band in a centrifuge tube. Buoyant density of
the band was 1.16 g/ml. Differential interference-contrast (B), phase-contrast (C) and DAPI fluorescence
photomicrographs of the bacteria of C (D). Bars, 10 um.

under a phase-contrast microscope (Fig. 3C): a
dark-looking part, a refractile part and a small
dark-looking tip, and DNA was localized at the
refractile part (Fig. 3C, D) as observed formerly in
the infectious long forms in the host homogenates

[6]. From this and former results, it is evident that
H. obtusa changes not only its cell length but also
its resistancy against host-lysosomal enzymes,
buoyant density and morphology of the nucleoid in
the differentiation.

856 M. FusisHIMA, K. NAGAHARA AND Y. KOJIMA

- Va |
$i a “ Vd ee
alee:
V4 _—
od ,
¢
V4 W

B

Fic. 4. Two-dimensional SDS-polyacrylamide gel electrophoresis of H. obtusa. Polypeptide compositions of the
reproductive short (A) and the infectious long forms (B). Major polypeptides specific for the reproductive short
form and the infectious long form were indicated by black and white single arrowheads, respectively, and those
common to the two forms were indicated by white-rimmed black arrowheads. At the left of the gels are indicated
the positions of molecular weight markers (BIO-RAD) in kD.

Differentiation in Holospora obtusa 857

Infectivity of the isolated H. obtusa

Isolated bacteria of 4 different buoyant densities
were mixed with P. caudatum cells at densities 1X
10° bacteria/ml and 1500 cells/ml and examined
their infectivity to the macronucleus. It was found
that only the infectious long forms could infect the
macronucleus. The first infected bacterium
appeared in the macronucleus 15 min after the
mixing. The infected bacteria were constricted to
become 7-10 short forms within 2 days after the
infection and began to multiply by binary fissions
in the macronucleus. On the other hand, other
long forms which had been ingested into the food
vacuoles could not leave the vacuoles out to the
host cytoplasm, and were eventually discharged
from the cytoproct. Some of the premature long
forms showed morphological features as if they
had been partially digested in the food vacuoles.
The reproductive short forms and those of various
lengths in the elongation process were digested in
the food vacuoles. From these results, it appears
that at the stage of buoyant density 1.16 g/ml, the
bacterium acquires abilities necessary for leaving
food vacuole, distinguishing some differences of
two kinds of host nuclei and penetrating only the
macronuclear membranes.

We examined an optimum cell density of the
infectious long forms for the infection. Final cell
density of paramecia in the mixture was adjusted
to 1500 cells/ml, and those of H. obtusa were
adjusted to 3 x 10° bacteria/ml, 1 x 10° bacteria/ml,
3x 10° bacteria/ml, 1x 10° bacteria/ml and 3x 10°
bacteria/ml. One hour after the mixing, 50 para-
mecia were observed using a Nomarski differential
interference-contrast microscope at x<400. At
densities of 1x 10° bacteria/ml and 3x 10° bacter-
ia/ml, all cells died within 1 hr. At the densities 1
< 10° bacteria/ml and 3x 10° bacteria/ml, all mac-
ronuclei observed were infected with more than 10
bacteria and the number of infected bacteria per
macronucleus of the latter density of infection was
more than those of the former density. At density
3x 10* bacteria/ml, less than 80% cells were in-
fected. Thus, the optimum density of the infec-
tious long form for infection was found to be 3x
10° bacteria/ml when the density of paramecia was
fixed at 1500 cells/ml. In this condition, the in-

fected bacteria could become the reproductive
short forms and multiplied normally.

Comparison of polypeptides composition between
the reproductive short forms and the infectious long
forms

Polypeptides from isolated reproductive short
forms and infectious long forms were resolved with
2D-SDS-PAGE. From the gels shown in Figure
4(A and B), 296 and 298 polypeptides were recog-
nized in the reproductive and the infectious forms,
respectively, and 185 out of 296 were reproductive-
form specific and 187 out of 298 were infectious-
form specific polypeptides. Number of common
polypeptides recognized in the two forms were
111. Therefore, 483 polypeptide species were
recognized in H. obtusa. These results show that
more than 62% of total polypeptides in each form
were form specific ones and common ones in the
two forms were less than 38%. In Figure 4(A and
B), major polypeptides which were specific for the
reproductive short form (86 polypeptides) and the
infectious long form (119 polypeptides) and those
which were common between the two (55
polypeptides) were marked by different arrow-
heads. Thus, the results indicate that a majority of
the polypeptides of the reproductive short form is
replaced by different polypeptide species during
the differentiation. It should be noted that, during
the differentiation, many acidic polypeptides of
reproductive short form disappeared and _ basic
polypeptides appeared in the infectious long
forms.

DISCUSSION

In the present study, we found that the bacteria
show four different buoyant densities during dif-
ferentiation from the reproductive short form to
the infectious long form. This indicates that the
buoyant density does not increase linearly in time
but increases stepwise. We recently found by
scanning electron microscopy that the bacterium
changed its surface structure in well accordance
with the increases of its buoyant density [13].
Namely, the bacteria of buoyant density 1.09 g/ml
had an entirely rough surface, and those of
buoyant densities 1.11 g/ml and 1.13 g/ml had

858 M. FusyIsHIMA, K. NAGAHARA AND Y. KOJIMA

Phase - contrast

DAPI - fluorescence

H ;
MM qereite
UJ

1.09 —————_ _ 1.11 1.13

Fic. 5. Schemes of morphological changes of H. obtusa
in differentiation from the reproductive short to the
infectious long forms in phase-contrast and DAPI-
fluorescence microscopies. Numbers indicate
buoyant densities of each bacterium (g/ml). Note
that , bacteria change their DAPI-postitive regions
from the dark-looking regions to the blight-looking
regions under a phase-contrast microscope when the
bacterial buoyant density increases from 1.11 g/ml to
1.13 g/ml.

wale-like stripes on their surface, and those of
buoyant density 1.16 g/ml had an entirely smooth
surface except one of the both tips. This special tip
showed a rough surface. Therefore, not only
newly formed nucleoid, the special tip and changes
of protein composition, but also some qualitative
changes in bacterial cell wall may be the cause of
the increase of bacterial buoyant density. Morpho-
logical changes of H. obtusa in the differentiation
are schematically represented in Figure 5. In the
present study, we found that the DAPI-positive
regions of the bacteria changed from the dark-
looking regions to the bright-looking regions
observed under a phase-contrast microscope when
the bacterial buoyant density increased from 1.11
g/ml to 1.13 g/ml. Therefore, at this stage, some
qualitative or structural changes may occur in the
bacterial nucleoids. We could not distinguish the

1.16

special tip in the bacteria of buoyant density 1.13
g/ml. In recent study [13], however, we found that
one of the both tips of the bacterium of buoyant
density 1.13 g/ml was always depressed if the bac-
terium was observed with a scanning electron
microscope without using a critical point dryer.
Therefore, it is suggested that the special tip begins
to differentiate in the bacteria of buoyant density
1.13 g/ml, though the structure is not yet clear in
light microscopy.

We found that majority of bacterial polypep-
tides was replaced by different ones in the dif-
ferentiation. Either one or both of the following
possibilities may be concerned in the phe-
nomenon; (1) the different protein compositions
are derived from different gene expressions in the
bacterial life cycle and/or (2) proteins derived from
host genome may be involved in the bacterial
proteins and these imported protein species and
their amounts may differ at different stages in the
bacterial life cycle. This problem will be solved in
a future experiment that if prokaryote- and en-
karyote-specific inhibitors of RNA and protein
syntheses are added at the beginning of the dif-
ferentiation and their effects on the bacterial pro-
tein compositions are examined by 2D-SDS-
PAGE. Furthermore, detection of protein in-
volved in the special tip of the infectious long form
in 2D-SDS-PAGE in an interesting problem, be-
cause such proteins are expected to have an impor-
tant role for nucleus-specific infection of the bac-
terlum. This protein can be detected by extracting
some of the infectious long form _ specific
polypeptides from gels of 2D-SDS-PAGE to make
their antibodies and by examining their specificity
to the special tip of the infectious long form with
an immuno-electron microscopy.

Comparison of polypeptide pattern between the
reproductive short form and the infectious long
form of H. obtusa has been reported by 1D-SDS-
PAGE [14], but the fact that more than 62% of the
bacterial polypeptides differ in each other was first
discovered in this work. We [15] and Gortz,
Lellig, Miosga and Wiemann (personal com-
munication) recently found that the bacterium
changes its morphology and polypeptide composi-
tion in early infection process, too. An order of
changes of the polypeptide compositions of the

Differentiation in Holospora obtusa 859

bacteria in the infection process is now examined
by us.

In the host cells, binary fission of H. obtusa can
be observed exclusively in the reproductive short
forms, but whether the bacteria of elongation
process and premature long forms are really un-
able to perform the binary fission was unclear. In
the present work, however, we confirmed that only
the reproductive short forms have a furrow for the
binary fission. This suggests that morphological
changes of H. obtusa in the differentiation are an
irreversible phenomenon. This possibility is sup-
ported by the fact that, if elongation of the bacter-
ium began in host macronucleus, the infectious
long forms soon appeared even if excess culture
medium was added to the culture. However, it is
unclear whether the infectious long forms can
become the reproductive short forms in the macro-
nucleus without passing through an ordinary infec-
tion process. G6ortz [16] and Wiemann [17] re-
ported that the reproductive short forms show an
affinity to host macronuclear chromatin and distri-
bute to daughter nuclei by binary fissions of the
host, but the infectious long forms have no such
affinity and are easily excluded in the host cyto-
plasm surrounded by the host nuclear membrane
in the binary fissions. Therefore, it is suggested
that the infectious long forms have no ability to
become the reproductive short forms in the macro-
nucleus.

So far, no killing activity has been known in
Holospora species. In the present study, however,
we found that H. obtusa killed paramecia if the
density of the infectious long forms was 110°
bacteria/ml or more. The facts that the killing
effect is bacterial density dependent and that the
bacteria have an ability to infect the macronucleus
via food vacuoles, suggest that injury of the food
vacuole membranes or the macronuclear mem-
branes by the bacteria may be the cause of the
killing activity. Formerly, we found that the
infectious long form of H. obtusa could not sluice
out of food vacuoles of Blepharisma japonicum,
and that they could not infect macronuclei of P.
bursaria, P. jenningsi, P. trichium, P. woodruffi,
P. calkinsi and P. duboscqui though they could
sluice out of the food vacuoles [4, 7]. Therefore,
using these ciliates, we may clarify whether the

bacterial killing activity is caused by the injury of
the food vacuole membranes or the macronuclear
membranes. What kinds of changes in the starved
host cell trigger the bacterial differentiation re-
mains to be solved.

ACKNOWLEDGMENTS

We sincerely thank Dr. Hajime Ishikawa, Tokyo Uni-
versity, for conducting 2D-SDS-PAGE. This study was
supported, in part, by grants to M. F. from the
Ministry of Education, Science and Culture of
Japan (Nos. 61740422 and 62740428), The Itoh
Science Foundation, The Inamori Foundation,
The Naito Foundation, Nippon Petrochemicals
Co. LTD and The Nippon Life Insurance Founda-
tion.

REFERENCES

1 Gé6rtz, H.-D. (1980) Nucleus-specific symbionts in
Paramecium caudatum. In “Endocytobiology”. Vol.
1, Ed. by W. Schwemmler and H. E. A. Schenk.
Walter de Gruyter & Co., Berlin Ner York, pp.
381-392.

2 Ossipov, D. V., Skoblo, I. I. and Rautian, M. S.
(1975) lota-particles, macronuclear symbiotic bac-
teria of ciliate paramecium caudatum clone M115.
Acta Protozool., 14: 263-280.

3. Fujishima, M. and Gortz, H.-D. (1983) Infection of
macronuclear anlagen of Paramecium caudatum
with the macronucleus-specific symbiont Holospora
obtusa. J. Cell Sci., 64: 137-146.

4 Fujishima, M. and Fujita, M. (1985) Infection and
maintenance of Holospora obtusa, a macronucleua-
specific bacterium of the ciliate Paramecium cauda-
tum. J. Cell Sci., 76: 179-187.

5 Gortz, H.-D. and Wiemann, M. (1989) Route of
infection of the bacteria Holospora elegans and
Holospora obtusa into the nuclei of Paramecium
caudatum. Europ. J. Protistol., 24: 101-109.

6 Fujishima, M. and Hoshide, K. (1988) Light and
electron microscopic observations of Holospora
obtusa: a macronucleus-specific bacterium of the
ciliate Paramecium caudatum. Zool. Sci., 5: 791-
7199:

7 Fujishima, M. (1986) Further study of the infectivity
of Holospora obtusa, a macronucleus specific bacter-
ium of the ciliate Paramecium caudatum. Acta Pro-
tozool., 25: 345-350.

8 Dryl, S. (1959) Antigenic transformation in Para-
mecium aurelia after homologous antiserum treat-
ment during autogamy and conjugation. J. Pro-
tozool., 6 (Suppl.): 25.

9 Hiwatashi, K. (1968) Determination and inheri-

10

itil

12

3)

14

860

tance of mating type in Paramecium caudatum.
Genetics, 58: 387-386. |
Freiburg, M. (1985) Isolaiton and characterization
of macronuclei of Paramecium caudatum infected
with the macronucleus-specific bacterium Holospora
obtusa. J. Cell Sci., 73: 389-398.

Laemmli, U. K. (1970) Cleavage of structural
proteins during the assembly of the bacteriophage
T4. Nature, 227: 680-685.

O'Farrell, P. H. (1975) High resolution two-
dimensional electrophoresis of proteins, J. Biol.
Chem., 250: 4007-4021.

Fujishima, M., Sawabe, H. and Iwatuki, K. (1990)
Scanning electron microscopic observation of dif-
ferentiation from the reproductive short form to the
infectious long form of Holospora obtusa. J. Pro-
tozool., 37: 123-128.

Gortz, H.-D., Freiburg, M. and Wiemann, M.

5

16

1

M. FusISHIMA, K. NAGAHARA AND Y. KosIMA

(1988) Polypeptide differences between infectious
and reproductive forms of Holospora obtusa, an
endonucleobiotic bacterium from the macronucleus
of Paramecium caudatum. Endocyt. C. Res., 5:
233-244.

Fujishima, M. and Kojima, K. (1987) Changes of
protein composition in the life cycle of Holospora
obtusa, a macronuclear specific symbiont of Para-
mecium caudatum. Zool. Sci., 4: 994 (Abstract).
Gortz, H.-D. (1980) Life cycle and infectivity of
Holospora elegans Hafkine, a micronucleus-specific
symbiont of Paramecium caudatum (Ehrenberg).
Protistol., 16: 591-603.

Wiemann, M. (1989) The release of Holospora
obtusa from Paramecium caudatum observed with a
new device for extended in vivo microscopy. J.
Protozool., 36: 176-179.

ZOOLOGICAL SCIENCE 7: 861-869 (1990)

© 1990 Zoological Society of Japan

Analysis of the Contraction-Inducing Factor for Gonadal Smooth
Muscle Contraction in Sea Urchin

S)
NoBuaki TAKAHASHI!*”, Noriyuki SATO””, Norio OHTOMO~, AKIHIRO KoNpbo”,

MASAKI TAKAHASHE and Koxicut KIKUCHI”

!Marine Biomedical Institute, Sapporo Medical College, Higashirishiri, Hokkaido 097-01,
Department of Pathology, Sapporo Medical College, Sapporo 060, *Faculty of
Engineering, Hokkaido University, Sapporo 060, and ° Biotechnology
Research Laboratories, Takara Shuzo Co., Ltd.,

Otsu, Shiga 520-21, Japan

ABSTRACT—The present study deals with the characteristics of contraction factor (CF), so called
Palmer’s substance, on gonadal smooth muscle contraction in the sea urchin. The CF extracted from
various organs of the sea urchin showed seasonal changes in contraction activity. The CF extracted from
the aboral intestine was mainly analyzed. We also could develop a monoclonal antibody, designated as
+ 11-B-2, against the CF. Partial purification of CF was carried out using this monoclonal antibody, and
the CF was considered to be a glycoprotein with a molecular weight of 3,800. An immunohistochemical
analysis of the localization of the CF in the aboral intestine showed that CF is present within the heamal

vessel.

INTRODUCTION

Since oocytes of female sea urchins at their
reproductive period pass through the 2nd meiotic
devision and are remained separate from sur-
rounding somatic cells such as follicle cells, egg-
spawning usually is a final step induced by the
contraction of smooth muscles of the gonad wall.
It is also known that the sea urchin begins gamete-
shedding after some injury especially that in inter-
nal organs. A substance produced at the time of
damage was first investigated half a century ago by
Palmer [1]. However, its chemical nature has been
unknown. The present study deals with character-
istics of the contraction-inducing substance (maybe
Palmer’s substance) for the gonadal smooth mus-
cles, the substance being derived from the diges-
tive tract of the sea urchin. Preliminary descrip-
tions concerning this contraction fractor have
already been described in some papers [2, 3].

Accepted April 5, 1990
Received September 11, 1989
> To whome all correspondence should be addressed.

MATERIALS AND METHODS

Animals

Sea urchins (Strongylocentrotus intermedius)
were collected from the coast of Rishiri Island of
nothern Hokkaido and were used for a bioassay of
the contraction.

BALB/c mice were obtained from CLEA Japan
Inc., Shizuoka, Japan and were used for produc-
tion of monoclonal anti-bodies.

Seawater

Modified Van’t Hoff seawater (ASW) (462 mM
NaCl; 9 mM KCl; 9 mM CaCl; 36 mM MgCl; 17
mM MgSQ,; 20 mM Tris-HCl, pH 8.2) was used as
a basal incubation medium. K*-rich ASW con-
tained KCl at the concentration of 500 mM, NaCl
was removed, concentration of Ca** and Mg’t
was as well as that of ASW. Contraction factor-
ASW (CF-ASW) was prepared by adding 1 ml of
CF fraction of gel filtration to 9 ml of ASW with a
higher concentration (10/9 times) than normal
ASW.

862

Apparatus

The apparatus used to measure the contraction
in the present study is described in the preceding

paper [4].

Recording of bioassay

The recording was carried out at room tempera-
ture which varied from 18 to 25°C.

The gonad contraction was expressed as poten-
tial difference (PD), which was designated in a
previous paper by Okada et al. [5]. In brief,
central parts of gonads at the time of spawning are
known to increase in height. This change in
height, which is evoked by the contraction smooth
muscles of the gonad wall, is measured, and desig-
nated as the potential difference (PD). A hole 3
cm in diameter was made on the oral side of each
sea urchin with solid scissors followed by removal
of the oesophagus by forceps. The animal was
subsequently fixed by pinching the test with large
forceps and immersed in a beaker up to its equator
in filtered natural seawater. The body cavity was
also filled with ASW. A straw previously cut
vertically in half and connected to a strain guage
(SB-1TH, Nihon Koden) was placed on a gonad
and a siphon was set in the central portion of the
body cavity. The gonad was kept in this state for
30 min; during which time the ASW was changed
two or three times. The gonad was first treated
with CF-ASW (crude or fractionated CF). Im-
mediately following a peak of induced contraction
(recorded as PD), the gonad was washed, placed in
ASW solution, and 30 min later a similar treat-
ment was repeated. This bioassay using one sea
urchin was performed about ten times. It was
finally treated with K*-rich ASW. The relative
value of mechanical response (PD) by CF against
that by K*-rich ASW was represented by the
relative response (RR).

Gel filtration

The aboral intestines were isolated with forceps
and immersed in acetone. Its acetone powder was
homogenized with 50mM Tris-HCl, pH 7.5 at.a
concentration of 100 mg per 1 ml of 50 mM Tris-
HCl, pH 7.5 and centrifuged at 40,000 rpm for 60
min. One ml of the supernatant was subjected to

N. TAKAHASHI, N. Sato et al.

gel filtration on Sephadex G-25 column chroma-
tography. Conditions were cloumn size, 1.8 x20
cm, eluant, DW, and flow rate, 50 ml/hr.

Production of monoclonal antibody

BALB/c mice were immunized intraperitoneally
with 0.5 ml of the supernatant of the aboral intes-
tine twice at an interval of 14 days. Five days after
the last immunization, approximately 2x 10°
spleen cells were fused with 410’ NS-1 mouse
myeloma cells by polyethyleneglycol 4000, accord-
ing to the method described by Lemke et al. [6].
After cell fusion, the cells were suspended in 360
ml of RPMI 1640medium with 10% heat-
inactivated fetal calf serum and hypoxanthine (H)-
aminopterin (A)-thymidine (T). The mixture was
seeded into the wells in the presence of feeder cells
obtained from BALB/c mouse spleen. After cul-
tivation for 10 days, the culture medium was
replaced by a medium containing H and T. The
culture media were first screened by the enzyme-
linked immunosolvent assay (ELISA), secondary
by the biotin-avidin horseradish peroxidase
method (ABC method) for antibody activity dis-
playing a positive reaction on the cryostat sections
of the aboral intestine, and finally selected by
inhibition of the CF-induced gonadal contraction.
The positive hybrids were cloned by limiting dilu-
tion. Thus, one hybridoma clone #11-B-2 with
IgG, isotype (Moab #11-B-2) was selected.
Moreover, the actites of the +11-B-2 was obtained
by injecting the 1 x 10’ hybrids to a mouse treating
with Pristant (Aldrich Chemical) and subjected to
further sutdy.

Immunohistochemical staining

Cryostat sections of the aboral intestine of sea
urchin were prepared. The sections were fixed in
acetone for 10 min at 4°C, and were incubated with
or without Moab #11-B-2 for lhr at room
temperature. The sections were washed with PBS
for 20min and reincubated with 200 diluted
serum of biotinylated goat antimouse immunoglo-
bulin (Ig) for 1 hr at room temperature. After
washing with PBS for another 20 min, these sec-
tions were stained by the biotin-avidin horseradish
peroxidase method (Vectastain, Vector Labor-
atorries Inc.).

Gonadal Contraction Factor in Sea Urchin 863

Electrophoresis and Purification

The estimation of molecular weight was per-
formed by electrophoresis, using Swank and
Munkres Gel [7] with the Pharmacia Polypeptide
Molecular Weight (PMW) Electrophoresis Calib-
ration Kit. The supernatant (0.1 ml) of aboral
intestine centrifuged at 40,000 rpm was incubated
with 0.1 ml of Affigel 10 beads and rotated slowly
overnight at 4°C. This was performed in order to
remove non-specific absorbable materials in super-
natant to the Affigel 10 beads. On the other hand,
0.1 ml of Moab #11-B-2 was coupled to 0.1 ml of
another Affigel 10 with goat anti-mouse immunog-
lobulin. It was followed that the 0.1 ml of super-
natant of aboral intestine treated once with the
Affigel 10 beads was added to the Affigel 10 beads
coupled with Moab #11-B-2 and rotated slowly
overnight at 4°C. Then, the Affigel 10 beads were
washed 3 times with a washing buffer consisting of
0.2M Tris-HCl (pH 8.0) and 0.1% NP-40 and
treated with boiling water for 5 min. The super-
natant was separated as purified contraction solu-
tion from the #11-B-2~Affigel 10 beads and was
resuspended in 0.04 ml SDS sample buffer con-
taining 50 mM Tris (pH 7.0) and 2.0% SDS with
5% 2-mercaptoethanol. The sample in SDS buffer
followed by addition 8M urea was applied to
Swank and Munkers gel containing 0.1% SDS
(W/V), 0.075% TEMED (V/V), 0.07% ammo-
nium persulfate (W/V), 0.1M H3PO,, 8M urea,
and Tris at a final pH of 6.8.

RESULTS

Gonadal contraction factor in various organs

It is demonstrated that water extracts of various
organs of the sea urchin such as oesophagus, oral
and aboral intestines, testis, ovary, and radial
nerves induced contractions in the gonad.

The sea urchins as starting materials of CF in
this experiment were collected during the spring.
The stage of the gonad at the collection was an
early growing phase. Various organs were isolated
with forceps and homogenized in distilled water
after measuring the wet weight. Each of the
homogenates was diluted with ASW and subjected

to measurement of PD. Figure 1 displayed the
relative response (RR) in various organs. The
oesophagus among the various organs contained
the greatest amount of gonadal contraction fac-
tor(s) (Fig. 1) with a threshold concentration of
0.01 mg/ml. However, the contraction by the
oesophagus did not attain a maximal tension (RR
=1.0). On the other hand, the contraction by
aboral intestine showed the 1.0 in RR value. The
threshold concentration of the aboral intestine was
0.1 mg fresh wet per 1 ml of ASW and the concen-
trtation at 0.5 RR was 2mg. The present study
uses the aboral intestine for further examinations
of molecular structure of the CF.

Seasonal change of CF activity of aboral intestine
extract

As a result of gel filtration of Sephadex G-25,
two active fractions (CF I and II) in inducing

1.0 @
aboral intestine 2 d
Ee: Ve vee
oO
— Oral
Pa Vis intestine
a o5—
Ss 9
ag
pe
Lh O/ <gonad
fea ie
g é Fo ag Sane TU HT
0.01 0.1 |

RR (PDy/ PDx )
i
2
a
SN
>
E

a ffs Re nerve
om i
dies “shat 01

een rarer ey was Te
( Fresh echt mg/ml )
Concentration of Homogenates

Fic. 1. Relative response of the contraction by water-
extract of various organs to that by K*-rich ASW.
RR; relative response, PD; potential difference.
Points and vertical bars are means+SEM in four
experiments.

864

= (11) OCTOBER
=

ya ea EE) AIS ey Ge) ao
‘a DECEMBER

1.0

jag
a = A750

0.5

APRIL
: :
Soom Cr TO) I ee A anu A mich ale) Dl
e A280
; © Azgo
> _o o_o oe!

Fraction Namnee

Fic. 2. Seasonal variation of contraction activity in wa-
ter extracts of the aboral intestine. The water-
extract was gel-filtrated on Sephadex G-25. Two
peaks (CF I and II) of contraction activity were seen
at autumn and spring. A759, Lowry method; Ango,
absorbance of 280 nm wave-length; A4oo, phenol-
sulfate method. Column conditions: cloumn size,
1.8X20cm; sample size, 1 ml; eluant, DW; flow
rate, 50 ml/hr, and fraction size, 5 ml/tube. Assay
procedure: meaurement of contraction activity was
performed after diluting 1 ml of each fractions to 9
ml of a higher concentration ASW (10/9 times).

gonad contraction were obtained from sea urchin
at the spent stage of the reproductive cycle (Octo-
ber) (Fig. 2). However, with approach to the
breeding season (September) of the following
year, it became possible to obtain only one active
fraction which we designated as the contaction
factor I (CF I). In general, we collected a great
amount of aboral intestine during the early sum-
mer (June and July) and prepared its acetone
powder. Accordingly, our materials appeared to
contain a great deal of CF(I).

N. TAKAHASHI,

N. Sato et al.

Contraction mode and repitition

The manner in which the gonad responded
mechanically to CF(I) was illustrated in Fig. 3a.
Contraction was phasic, with a relaxation period of
about 20min. Therefore, if CF was injected
artifically, ripe sea urchins appear to have the
spawning time of about 20 min.

Another characteristic concerning CF(I) was
capability of being repeated. This property was
convenient for bioassay of the contraction. As
shown in Figure 3b, the gonad was treated with a
crude extract of aboral intestine (AOI) (2.5 mg/
ml) until the time immediately following a peak of
an induced contraction and was then washed with
ASW. Three times after application of AOI, the
contraction attained a maximal PD, that is, the
contraction was saturated. Subsequently, repiti-
tion of contraction by CF was possible ten times.
Duration between any two contractions was 30
min.

Development of Moab +11-B-2 against CF(J)

BALB/c mice were immunized 1.p. with a water
extract (pH 7.5) of the aboral intestine, and their
spleen cells were fused with NS-1 myeloma cells.
The hybrids that could be grown in hypoxanthine-
aminopterin-thymidine selection medium were
first screened for the antibody activity on water
extract of aboral intestine by the ELISA method.
Forty hybrids were selected by its ELISA method
from 120 hybrids. Then, hybrids responding to
frozen tissue sections of the aboral intestine and
gonad by the ABC immunohistochemical method
were screeened. As a result, 10 hybrids were
selected. Finally, hybrids whose antibodies inhibit
functionally the gonadal contraction by CF(I) were
selected. Thus, Moab #11-B-2 with IgG, subtype
was obtained. The ascitic fluid was prepared by
injecting 1x10’ +#11-B-2 hybrids to a mouse tre-
ated with Pristane and purified by Protein A
Separose affinity column which was treated with
binding buffer consisting of 10 mM sodium phos-
phate (pH 8.2) and 0.15 NaCl, was released by pH
3.0 acid solution (100 mM sodium citrate) and was
neutralized by Tris powder. The purified antibody
was concentrated and subjected to further study.
After treating 0.5 ml of the CF(I) with 0.5 ml of

Gonadal Contraction Factor in Sea Urchin 865

(a)

(b)

(1) (3)

AO!
(2.5mg/ ml)

AOI

eon.

20 (min)
£
i;
> ‘| a
eax ny i ( 2min |
l 1 L all
AOI 500 mM K

Fic. 3. Contraction mode and repetition. a, a representative gonad response to CF(I). The contraction was phasic,
with a relaxation period of about 20 min. b, repeated responses by crude extract of the aboral intestine (AOI).
The contraction attained to maximal level at the third time.

ok

{ |
CF( I)
# 11-B-2 IgG,
Fic. 4. Anti-effect of acites derived from #11-B-2 hyb-
ridoma on CF(I). The Moab shows the IgG,
subtype. The ascitc fluid containg the MoAb was
purified by the Protein A Sepharose column. The
purified Moab (0.5 ml) and the CF(I) (0.5 ml) were
mixed during 4 hr and then the activity was mea-
sured.

|
CF(I) CF(I)

the purified antibody (1.16 mg protein amount/ml)
during 4 hours, this Moab clearly inhibited con-
traction induced by CF(I) (Fig. 4).

Localization of CF(1) in aboral intestine

The aboral intestine was a final organ of the

digestive tract and linked to an anus. The aboral
intestine was yellowish brown in its external
appearance and became brown only during spawn-
ing time. The structure of an aboral intestine was
composed mainly of coelomic epithelium (visceral
peritoneum), muscle layer, heamal vessel, connec-
tive tissue, and ciliary column cells and it plays
roles for doing digestion and absorption of foods
(Fig. 5a).

The aboral intestine was frozen by liquid nit-
rogen and sectioned at 4 um by a cryostat. The
section was fixed in acetone and followed by the
ABC method using Moab # 11-B-2 against CF(1).
As shown in Figure 5c, the heamal vessel only was
stained with #11-B-2. Accordingly, CF(I) was
present in the heamal vessel of the aboral intes-
tine. In addition, the portion exhibited a PAS-
positive reaction (Fig. 5b).

Contraction by purified CF(1)

The supernatant (0.1 ml) of aboral intestine
centrifuged at 40,000 rpm was incubated with 0.1

866 N. TAKAHASHI, N. SATO et al.

Fic. 5.

ys
i — « ——y ;

Cc

Immunohistochemical picture of aboral intestine stained by Moab, #11-B-2, against CF. a, haematoxylin-

eosin stain; b, PAS reaction; c, ABC method by #11-B-2. Scale is 50 um.

ml of Affigel 10 beads and rotated slowly overnight
at 4°C. This oparation was performed in order to
remove non-specific absorbable materials in the
supernatant to the Affigel 10. On the other hand,
0.1 ml of Moab #11-B-2 was coupled to 0.1 ml of
another Affigel 10 with goat anti-mouse immunog-
lobulin. It was followed by that 0.4 ml of super-
natant of aboral intestine treated once with the
Affigel was added to the Affigel cupled with Moab
4#11-B-2 and rotated solwly overnight at 4°C.
Then, the Affigel were washed with a washing
buffer (0.2 ml each) 3 times and heated on boiling
water for 5min. Purified contraction factor was
separated from #11-B-2 conjugaed Affigel and

was subjected to the bioassay. The purified mate-
rial markedly induced the gonadal contraction
(Fig. 6).

Molecular weight and brief characteristics of CF(1)

The estimation of molecular weight was per-
formed by electrophoresis, after above-mentioned
purification of CF(I) by affinity chromatography
with the Affigel, using a PMW electrophoresis
calibration kit. A part (0.025 ml) of the purified
CF(1) resuspended in 0.04 ml SDS sample buffer
was applied on the Swank and Munkers Gel. The
molecular markers of this kit consisted of myog-
loblin and its partial digested materials. The

Fic. 6. Gonad response to the purified CF. The CF was
separated from Affigel 10~Moab #11-B-2 complex
by boiling method. CF: crude CF(I), Purified CF:

Gonadal Contraction Factor in Sea Urchin 867

Purified
CK

CF(I) purified by the affinity batch method.

Sas. aS

boiling water

GEA)

CRI)
96°C, 15min

or & Wry —
@
NO
WwW
on

relative electrophoretic mobility of CF was plotted
on a calibration curve and the molecular weight
was estimated. The molecular weight of CF(I) was
estimated as 3,800 (Fig. 7). CF(I) was positive to
coomassie blue and PAS-reaction. In addition,
CF(1) activity did not decrease after a treatment
with boiling water (96'C, 15 min) or autoclave
(120°C, 15 min) (Fig. 8). These indicate that CF(I)
is a heat-stable glycoprotein.

DISCUSSION

Little is known about spawining mechanism(s)
of the sea urchin under natural conditions. As is
well knwon, the ovarian oocytes in the spawning
period pass through the 2nd meiotic division and
are in the state of completion of meiosis, at the
time of the pronuclear stage. When the smooth
muscles in the gonad wall recieve (a) signal for
contraction of unknown internal mechanism, the
sea urchin is able to relase the gametes into sea
water. Thus, induction of the contraction of
gonadal smooth muscles appears to be a key
phenomenon in sea urchin spawning.

On the other hand, artificial treatments that
induce spawning are well known, such as with K
[1, 8, 9] and Ca [1, 10] ions, electric stimulus [11-

Fic. 7. Molecular weight determination using SDS-
PAGE electrophoresis. The purified CF run accord-
ing to adopted from a method by Swank and Munk-
res. 25 ul SDS sample buffer containg the purified
CF was applied to electrophoresis.

autoclave
=
Fah ood
1 We)
| i | N 2min
~ =
sas} oul Lisa
CF(I) CA)
120°C , 15min

Fic. 8. Heat stability of CF. CF activity was not reduced by heat treatment (96°C with biling water for 15 min or
120°C by autoclave for 15 min). One ml of the CF(I) was heated and diluted with ASW and CF activity was

measured.

868 N. TAKAHASHI, N. Sato et al.

14], acetylecholine [15], and damage [1] of shell.
Recently, it has been found that y-aminobutyric
acid (GABA) [4] is also capable of inducing the
artificial spawning. In the experiments using
GABA, we have become aware that a treatment of
gonads with water-extract of the digestive tract of
the sea urchin results in the contraction of gonadal
smooth muscle and the gamete shedding. A con-
traction factor in the extract appeared to be diffe-
rent in molecular weight from acetylecholine and
GABA. Palmer had long ago performed investiga-
tions about nature of the stimulation of injury
which causes sea urchins to spawn. Moreover, in
order to determine from which part of the animal
the possible stimulant is produced, various parts
were separately tested. Gamete shedding occurred
in the following instances: chopped lantern mus-
cle, chapped digestive canal, gonad ground with
sand, ground lanterns, ground empty coronas,
heated sperm and eggs, eggs broken with glass,
and cytolysed eggs and sperm. However, the
chemical nature of the stimulant still remains un-
known.

As described in this paper, our analysis of the
CF(I) seems to indicate that it corresponds to the
substance investigated by Palmer. The present
study showed chemical natures of the substance in
the CF(I). A monoclonal antibody (Moab), desig-
nated as #11-B-2, against CF(I) was produced.
Using this antibody, the CF(I) was estimated to
have a molecular weight of 3,800. It seems a
glycoprotein, since it is positive to coomassie blue
and PAS in SDS-PAGE. The localization of CF(I)
in the aboral intestine was examined by an im-
munohistochemical method and concluded to be in
the heamal vessel. Non-specific staining with the
Moab, #11-B-2, in several organs appears due to
the heamal vessel in various parts.

To clarify the physiological function of the CF in
sea urchin reproduction, consideration will be
given to the following aspects: (1) gamete shedding
and (2) animal motion, because this factor appears
to exert a significant role on reproductive be-
havior.

ACKNOWLEDGMENTS

The authors are grateful to Professor Kyozo Taka-

hashi, Director of Marine Biomedical Institute, Sapporo
Medical College, for his encouragement throughout this

work. We also thank Drs. S. Takahashi, M. Okubo and
A. Yagihashi, Sapporo Medical College, for their skillful
research techniques. This work was partly supported by
a Grant-in-Aid for a Special Research Project in the field
of Biothechnology.

REFERENCES

1 Palmer, L. (1937) The shedding reaction in Arbacia
punctulata. Physiol. Z6ol., 10: 352-367.

2 Takahashi, N., Sato, N., Okubo, M., Yagihashi, A..,
Takahashi, M. and Kikuchi, K. (1987) Analysis of
the gonadal contraction-inducing factor in the aboral
intestine of the sea urchin and development of
monoclonal antibody against this factor. Proc. 2nd
Annual Meeting, Jap. Soc. Basic Reprod. Im-
munol., 85-88.

3. Takahashi, N., Sato, N., Okubo, M., Yagihashi, A.,
Oota, I., Takahashi, M. and Kikuchi, K. (1988)
Further analysis of the gonadal contraction-inducing
factor in the sea urchin intestine using monoclonal
antibodies. Proc. 3rd Annual Meething, Jap. Soc.
Basic Reprod. Immunol., 64-69.

4 Takahashi, N. (1987) Gonad response to y-amino-
butyric acid in the sea urchin. Zool. Sci., 4: 433-439.

5 Okada, Y., Iwata, K. S. and Yanagihara, M. (1984)
Synchronized rhythmic contractions among five
gonadal lobes in the shedding sea urchins: coordina-
tive function of the aboral nerve ring. Biol. Bull.,
166: 228-236.

6 Lemke, H., Hammerling, G. J., Hohmann, C. and
Rajewsky, K. (1978) Hybrid cell lines secreting
monoclonal antibody specific for histocompatibility
antigens of the mouse. Nature (Lond.), 271: 249-
Dale

7 Swank, R. T. and Munkres, K. D. (1971) Molecular
weight analysis of oligopeptides by electrophoresis
in polyacrylamide gel with sodium dodecyl sulfate.
Anal. Biochem., 39: 462-477.

8 Harvey, E. B. (1939) A method of determining the
sex of Arbacia, and a new method of producing
twins, triplets and quadruplets. Biol. Bull., 77: 312.

9 Harvey, E. B. (1940) A note on determining the sex
of Arbacia punctulata. Biol. Bull., 79: 363.

10 Takahashi, N. and Takahashi, M. (1987) Gonad
response to calcium and a comparison of the effects
of calcium, potassium, acetylcholine and y-aminobu-
tyric acid on the sea urchin gonad. Zool. Sci., 4:
441-446.

11 Iwata, K. S. (1950) A method of determining the
sex of sea urchins and of obtaining eggs by electric
stimulation. Annot. Zool. Japon., 23: 39-42.

12 Harvey, E. B. (1952) Electrical method of “sexing”
Arbacia and obtaining small quantities of eggs. Biol.

Gonadal Contraction Factor in Sea Urchin 869

Bull., 102: 284.

13. Harvey, E. B. (1953) A simplified electrical method
of determining the sex of sea urchins and other
marine animals. Biol. Bull., 105: 365.

14 Iwata, K. S. (1962) A simplified electrical method of

15)

determining the sex of sea urchins. Zool. Mag., 71:
301-302.

Iwata, K. S. and Fukase, H. (1964) Artificial
spawning in sea urchins by acetylcholine. Biol. J.
Okayama Univ. 10: 51-56.

Me all utvbler elie) ir ia Sy
A iaials quarcle and the ¢ = phe ea, eae ae
2) tgaction, fact in The 1xiF ek pupae typ aetees "Ease il aa nae
oa iY) ovina weenie t a ah adhere se ie det Ee. ; om
GABA. Paine hy nig homie ao pained Breall

et : » whined ‘ ¥ 2 E 4 ba ieret raw i ea ; 7
ay 3 Cypha, .. hie rests. oe See os shane eas: ea at ge” me sh.

BE

VCA CORSE Vee tii hice $ any Pues

;
2 ai
Fr fas 5 F et % yer
on ‘ anal - ft nds my Coe ee bal ,
eriter . Fi o SSE Vat Ss eA Rel aE o oe Me stele of ti ‘agp ;
PO a os
& : f aS)
=) . ee sch
t ae a) ‘ * ; * a! Wu
the me Mit be. 9) ba ay eee eee eee ‘> Pe aM KOSH Aecudy” ©
‘ kee oe a q a i aio! Se ed ‘la ;
‘ a> renee) ‘i ., ae se $ pa 7 ee Fon 2 P “ Anaya : sa eiats ies are .
were Mes i pe Der oft ssmate | Ie a) Pero ; st cae] Pas
- - 2 eect + > . 4 te rh (oe, See yy Paar
F yr gi - = 2 i ‘ nai api Gita,’ een
een aa oe ia ~= aes LD , 2 at Whee: ,? - ’ c af ee f “
Th, BOE: MAME Seer AS tee Ce ER Terese 6 ae p (heh zen :
ar Saas i ET a sf i Ray sic ieg ‘ Satay tay
4 Cz “af y ' 7 q : ; ep ae | ‘in t c mae a] a :
fe afer La SO" i : ae = it OE. id te eo A i eo = tied ats toe -
hs ' yg es pita Dies J ry EAL PE a CAE = ah 3 ] i 4 ahah sents fois
fi - ~ > oe Ade =, i i Fa ) ‘
. eee Oe! OT rn ek See =) ' oe _ Key sah iain ‘ . ; =e ; z
pe 8 1 iy Sata: SRT Pa oe OU CeRS, ri narhninat §
- me a o c r —
x ; ys - ; - Pi .
i c + i 3 5 bios ys ths, 254
a ee z a - bs Ps a rt
oa 2 = a age i "By ice 09 rie ; :
Ph eae A. lemme
gl =) = Py
i v 7 —s' as : : j i x 7c tr ; :
- =) . .. 4 ae 4 7 a i ‘ 7%. y, 4 J
: View een : ee. 2 BRASTBATE TM (1982)
Det ae ; : 2 ve ee eee he free ee
Tee FS - ; : i : os | “ Ne we et
oa ‘ r Cal “ ’ =,
cal F { ‘
Nemo 04 stk Ae ee . ;
a r » < =i |
a 3 5 A - (au) ’ in .
athe = r es « :
‘ | vas - . ‘S ie rr ivy for ‘ ac: " .
4 3 - ' as
nS Hoga Le =e Aes
ee 7 2 p
~<— > ~
7 i J ‘ ; iu by fy
; ; e 5
Es }.
J = } d ,
Pisa 7 Ee 4 ery, # J
; a
f as i i x : : - = Laity
z . a ;
, z m)
- = " —
: A c is
a ia bo | n Le
rm : = :
a r 2 . r
r re
> y
| -
% 3 ~
tee a 4
\
x
c Z 2
a}
2

ZOOLOGICAL SCIENCE 7: 871-877 (1990)

Stress Effects on Late Pregnancy in the Fluying-Fox,
Pteropus scapulatus

W. RicHARD DuKELow!, C. S. T. Pow, J. H. KENNEDY
and LEN MARTIN

Department of Physiology and Pharmacology, University of
Queensland, St. Lucia, Australia

ABSTRACT—Twenty one flying-foxes (Pteropus scapulatus) were alotted to three experimental
groups. Group I received no handling until the final examination period. Group II was transported,
anaesthetized, and blood sampled approximately 3 weeks before the final examination period. Group
III received the same treatment as Group II but with the incorporation of a laparoscopic examination.
High levels of abortion were noted in all groups (40.0, 20.0 and 45.5% for groups I, II and III
respectively) during the three week experimental period but was not increased by the use of
laparoscopy. Following the second period of examination stress, overall abortion rates for the total
gestation period were 100, 80 and 90% for the three groups respectively, emphasizing the extreme
stress-induced abortion susceptibility of this species. Nevertheless, the usefulness of laparoscopy to
examine selected stages of the gestation was demonstrated. Over 54% of the animals diagnosed
pregnant at the first laparoscopic examination were still pregnant at the subsequent examination 3 weeks
later. Progesterone levels of 15.9 ng/ml or higher were found in all pregnant animals and were never

© 1990 Zoological Society of Japan

above 4.3 ng/ml in nonpregnant animals.

INTRODUCTION

The Order Chiroptera (bats) has over 951 spe-
cies distributed over 187 genera within 19 families
[1]. These can be divided into the two sub-orders
of Microchiroptera and Megachiroptera. The lat-
ter are found in the single family Pteropodidae
consisting of 83 genera and 150 species. In Austra-
lia, they are represented by five genera and nine
species. These flying foxes are characterized as
large (up to one kg), frugivorous bats not posses-
sing echolocation abilities, and as a nonhibernating
species. They typically are found in the tropical
forests where native fruits and flowers are abun-
dant [2].

Pteropus scapulatus (little red flying fox) is the
smallest of the Pteropid flying foxes found in
Australia. This highly nomadic species is wide-

Accepted May 15, 1990
Received January 31, 1990

" Present address and address for reprints: Endocrine
Research Center, Michigan State University, East
Lansing, MI 48824, U.S.A.

spread throughout eastern Australia from arid
inland areas to the coast. P. scapulatus mates in
November and December (the Australian spring)
and usually delivers a single offspring from the
duplex uterus in April or early May (autumn) [3].

Few studies have been conducted on the repro-
ductive physiology of flying foxes until recent years
but successful captive breeding has been achieved
[4, 5]. Some have suggested possible common
neurological pathways between the megachiro-
pterans and primates [6].

The present research was stimulated by analysis
of data collected in this laboratory in 1984 after the
capture of feral P. scapulatus in February with
subsequent movement (in March) of 29 late
pregnant animals. This movement stress resulted
in the subsequent abortion of at least 86% of the
pregnancies [7]. The objectives of the present
studies were to a) determine if laparoscopy could
be used to examine the conceptus in mid-
pregnancy in P. scapulatus and b) determine the
extent to which laparoscopy and associated stres-
sors cause abortion in this species.

872 W.R. DuKELow, C. S. T. Pow et al.

MATERIALS AND METHODS

Husbandry

Adult female P. scapulatus were caught wild at
Monto, Queensland, Australia in March 1986 and
held in open air cages 6X2X2m with males
throughout the breeding season. The animals were
fed daily a chopped fruit (papaya, mango, grapes,
pears, peaches, rock melons and honeydew
melons) diet supplemented with calf and pig milk
replacer 2-3 times each week and water ad
libitum.

Experimental design

Twenty-one adult females were randomly allo-
cated to one of three treatments as follows: Group
I (negative control group) (five animals). These
females were not initially restrained, examined or
transferred. They remained in their colony cage
with the males until March 14, 1989. Group II
(positive control group) (five animals). These
females were restrained and transported on Febru-
ary 20, 1989 from the University of Queensland
Veterinary Research Farm to the departmental
animal house, a distance of 12 km. These animals
were then anaesthetized and a blood sample was
taken as described below. Group III (laparoscopy
group) (eleven animals). These females were
restrained, transported, anaesthetized and blood
sampled like the females in Group II. Additional-
ly, each animal was examined laparoscopically.

Anaesthesia, blood sampling and blood analysis

Each animal in Groups II and III was anaesthe-
tized with a mixture of 10 mg ketamine-HCl (Keta-
mine Injection®, Jurex Pty, Ltd., Riverstone,
N.S.W.) and 2 mg xylazine (Rompun®, Bayer Au-
stralia Ltd., Botany, N.S.W.) i.m. in 0.2 ml
volume. After anaesthesia, 0.5 ml of blood was
drawn from the uropatagial vein [8] with a 25 ga,
19mm needle attachted to a 1 ml tuberculin
syringe. This sample was then placed in a 1 ml
plastic vial containing 5 wl lithium heparin and
gently mixed. The tubes were then placed in a
refrigerator at 4°C until centrifuged the same day
at 1500 g for 10-15 min. Plasma was separated
and stored frozen in 1 ml plastic vials.

Progesterone analysis

The progesterone concentration in plasma was
measured by radioimmunoassay using an anti-
serum to progesterone (prepared by Dr. R. I. Cox,
Division of Animal Production, CSIRO, Prospect,
N.S.W., Australia). *H-progesterone (2.04 TBq/
mmol) was obtained from Amersham Australia.

A plasma pool containing progesterone was
included in each assay to calculate inter- and
intra-assay coefficients of variation. Parallelism
was demonstrated using graded dilutions (1/20 to
1/1) of a plasma sample containing approximately
10 ng of progesterone per milliliter.

For assay, 50 ul plasma samples were diluted to
500 yl with 0.06 M phosphate-gelatin buffer (pH
7.2) and 50 yl aliquots in triplicate were extracted
in 3ml of freshly redistilled hexane (May and
Baker Australia Pty Ltd.) and dried under nitro-
gen (average extraction efficiency 89.6%; n=3).
Standards were prepared and stored in ethanol and
various concentrations (0—2048 pg/tube). These
were subsequently dried under nitrogen in tripli-
cate for each assay. Extracts and standards were
resuspended in 100 wl of buffer and allowed to
stand at room temperature for three hr prior to the
addition of antiserum at 1: 8500 dilution in 100 ul
of *H-progesterone (10,000 cpm) in the same
buffer.

Tubes were incubated overnight at 4°C followed
by separation of bound and free steroid with 300 wl
of 0.625% charcoal-0.0625% dextran in buffer for
10 min at 4°C. Following centrifugation the super-
natant was removed, added to 3 ml scintillation
fluid (toluene, containing 0.3% PPO, 0.03%
POPOP and 5% acetic acid) and radioactivity was
measured in an LKB scintillation counter.

A standard curve was calculated according to a
logit transformation and progesterone concentra-
tions in unknown plasma samples were determined
from this curve and expressed as nanograms of
progesterone per milliliter of plasma. The sensi-
tivity of the assay (caculated as two standard
deviations from the blanks) was typically less than
16 pg/tube. The inter- and intra- assay coefficients
of variation were 8.0% and 7.8% respectively.

University of Queensland, St. Lucia, Australia 873

Laparoscopy

The basic laparoscopic procedure was that de-
scribed for small mammals [9] but modified for use
in the flying fox based on preliminary laparoscopic
examination in this laboratory [5, 7]. After clip-
ping of the abdominal fur, the area was swabbed
with a dilute solution of chlorhexidine gluconate
(Hibitane®, ICI, Villawood, N.S.W.) in 20% etha-
nol. The animal was restrained in dorsal re-
cumbency and the trocar-cannula inserted after
initial incision through the skin, abdominal wall
and peritoneum. The trocar was not forced
through tissue to avoid the possibility of uterine
puncture. The incision was approximately 5-10
mm posterior to the base of the sternum. A 4mm
paediatric laparoscope (Model 7200B, 30°, Karl
Storz and Company, Tuttlingen, West Germany)
was used. Insufflation was with nitrogen via the
cannula gas attachment. A 19 ga 38 mm needle,
attached to a 1 ml tuberculin syringe, was inserted
through the lateral skin and body wall area to
allow manipulation of the internal organs and, if
necessary, to remove urine from the bladder to aid
visibility. Since this needle is 1.08 mm in outer
diameter, it was also used as a measuring device to
approximate uterine size. During laparoscopy
attempts were made to view the ovaries, oviducts
and uterine horns of each animal, to ascertain the
horn of pregnancy and, if no pregnancy was evi-
dent, to look for signs of abortion or resorption. In
a preliminary laparoscopic examination of a
female believed to be nonpregnant, the entire
duplex reproductive tract was easily seen with a
thickening and whitening of the cervical end of the
right uterine horn and a yellow resorption site near
the right tubal uterine junction where implantation
normally occurs [10].

Laparoscopic examinations were completed in
five min or less on each animal, the cannula-
laparoscope removed and the abdominal wall and
skin layers closed with 2—0 synthetic gut sutures.
Of eleven animals receiving laparoscopic examina-
tions, four subsequently (1-3 days) picked sutures
free and were reanaesthetized and resutured with
the addition of stainless steel Autoclips® (Clay-
Adams, Parsippany, NJ, U.S.A.) placed on the
skin. All initial laparoscopic examinations were

completed on February 20-21, 1989.

After examination the animals were returned to
cages in the animal house. All abortions were
recorded.

On March 14, 1989 all animals in the three
groups were brought again for pregnancy assess-
ment by laparoscopic examination and/or palpa-
tion. All animals except those obviously pregnant
by palpation were laparoscopically examined, in-
cluding those known to have aborted. Again, a
blood sample was collected for progesterone
analysis.

Even though the basic research protocol ended
with the March examinations, all pregnant animals
were followed through to abortion or parturition
to further assess the effect of late pregnancy anaes-
thesia, bleeding and laparoscopic stress.

RESULTS

The results of the initial laparoscopic examina-
tions for pregnancy are shown in Table 1.

Based on subsequent examinations (and initial
sample progesterone assays of the females in
Group II) all of the 21 animals were pregnant at
the start of the studies. In Group III, 5 of 11
pregnancies were in the right uterine horn. In
Group I the two females that eventually aborted
were pregnant in the left horn. Thus, where the
side of pregnancy was definitely known, 5 of 13
(38.5%) were in the right horn. This difference
was not statistically significant (P >0.20).

Of the 11 laparoscoped animals, 5 (45.5%)
aborted after a single laparoscopic examination.
These occurred 1, 4, 5, 7 and 16 days after
laparoscopy. Aborted foetuses ranged from 7 to
20 gm. Progesterone values of pregnant animals
ranged from 15.9 to 79.7 ng/ml. There was no
significant relationship between progesterone level
and subsequent abortion. In fact, the female with
lowest progesterone level (number 12) maintained
pregnancy through a second laparoscopic examina-
tion 32 days later. One animal that subsequently
aborted (number 18) was not assayed for pro-
gesterone initially.

The results of the second laparoscopic examina-
tion are shown Table 2. In the negative control
group (1) 3 of 5 animals were still pregnant where-

874 W.R. DuKELow, C. S. T. Pow et al.

TABLE 1. Laparoscopic and endocrinological pregnancy analysis of flying foxes in Group III and outcome
following initial laparoscopy (February) and examination 22 days later

Number)’ | Se? , - pecRwibveeng, accu OSU gen ater intial
aparoscopy

11 376 yes left 14.0 72.8 Aborted, 1 day
12 306 yes right S)5) 1559 No effect
13 384 yes left 11.0 IS No effect
14 292 yes right 9.0 21.4 No effect
15) 426 yes right 17.0 25.4 No effect
16 DP yes left 15.0 32.4 Aborted, 4 days
y/ 345 yes right 15.0 62.7 No effect
18 293 yes left 16.0 — Aborted, 7 days
19 402 yes left IAS) 26.4 Aborted, 16 days
20 406 yes right 10.0 47.5 No effect
21 340 yes left 9.0 66.4 Aborted, 5 days

TABLE 2. Laparoscopic and endocrinological pregnancy analysis of flying foxes in all groups at
final laparoscopic examination

Animal Body Wt. Pregnancy Progesterone Abortion Foetal
Number (2) Octcome (ng/ml) Data Wt(g)
Group I

1 Sie Pregnant 392

2 309 Aborted LES unknown unknown
3 396 Pregnant 101.1

4 316 Aborted 3.4 unknown unknown
5 Si Pregnant 98.2

Group II

6 262 Pregnant 60.2

7 540 Pregnant ce

8 — Aborted March 10 19.0

9 244 Pregnant 67.1

10 283 Pregnant 63.3

Group III

11 329 Aborted 1,2 Keb, 21 12.0

12 300 Pregnant 34.9

13 386 Pregnant onl)

14 306 Pregnant 36.8

15 422 Pregnant 40.0

16 ~ Aborted Feb. 24 10.0

7, 350 Pregnant es)

18 LY: Aborted 4.3 Febs 27, 20.0

19 264 Aborted 30) March 9 8.0

20 35 Pregnant 103.1

Mas SS) Aborted des) Feb. 26 7.0

University of Queensland, St. Lucia, Australia 875

as the other two showed evidence of abortion
(both in the left horn). The pregnant animals were
obviously pregnant (subsequently confirmed by
progesterone levels) and were not subjected to
laparoscopic examinaiton. Of the four remaining
animals in Group II (handled and bled but not
laparoscoped in February) all were obviously pre-
gnant (confirmed by progesterone levels). The
fifth animal had been pregnant and aborted a 19
gm foetus on March 10th. Of the 11 animals in
Group III (see above) six were pregnant at the
time of the final examination. All but one ex-
hibited progesterone levels indicative of pregnancy
and ranging from 34.9 to 103.1 ng/ml. The excep-
tion is animal 17 who had a level of 1.5 ng/ml
progesterone. This animal aborted a 3.3 gm foetus
the next day. This female had a 15 mm diameter
pregnant horn at the initial laparoscopy (Table 1).
Comparison of foetal sizes with females numbered
11, 16 and 18 (who had uterine horn diameters of
14-16 mm and subsequently aborted within 7 days
of laparoscopy) indicates this foetus should be
between 10 and 20 grams. We thus conclude that
the foetus of animal 17 had died but was not
aborted until the day after the second laparoscopic
examination. Abortion rates between Groups I, II
and III were analyzed by factorial Chi-squared
tests for factors and interaction. Laparoscopy
Group III had no significant effect (P<0.10) on
abortion rate. Although animal numbers were low
in Groups I and II, abortion rate was significantly
higher (P< .005) in Group I.

All animals were returned to large outdoor
cages on March 18, 1989 and animals not deliver-
ing by June 15 were considered to have aborted
and the foetus lost. A final analysis of Group I
showed that all animals had been pregnant but
subsequently aborted. None of these animals were
ever laparoscoped while pregnant. In Group II,
two live births occurred on April 23 and May 3
(females 7 and 6 respectively). Neither animal had
ever been laparoscoped. In Group III one live
birth occurred, on May 10, to female 15. This
animal had been laparoscopically examined once
in the initial (February) examination. Interesting-
ly, this female had the largest uterine horn dia-
meter in Group III (Table 1) and thus may have
been the earliest pregnancy in the group. This was

the largest animal both at the initial (February)
and subsequently (March) examinations in Group
III. Only one other pregnant female (number 7 in
Group II) wsa larger at the March examination
and this female also delivered a live young, on
April 23rd.

DISCUSSION

The high fertility rate of P. scapulatus obtained
in captivity is reflected by the 100% overall pre-
gnancy rate at the start of these studies, estimated
conservatively at one-third of the gestation period.
This is in accord with the report of Towers [7] who
reported capture of a large number of P. scapula-
tus in Queensland in February 1984, all of which
were pregnant. Similarly, all 42 mature P.
poliocephalus (the grey headed flying fox) col-
lected between April and September (normal time
for gestation in this species) were pregnant.

In the present studies pregnancy occurred on
both sides of the reproductive tract, apparently
randomly (38.5% on the right). Others [10, 11]
attributed alternation of sides with successive
ovulations to P. giganteus but the evidence for this
was minimal and cannot be confirmed from the
present experiments. Towers [7] reported 9 of 17
(52.9%) P. scapulatus pregnancies were in the
right uterine horn. There is evidence for a local
endometrial reaction to the developing corpus
luteum in a number of Chiropteran species and
Bonilla and Rasweiler [12] proposed a countercur-
rent exchange of ovarian hormones between the
arterial supply of the uterus and the ovarian
venous or lymphatic drainage. Pow and Martin
[13] have demonstrated evidence for such an ex-
change in P. scapulatus.

In the present studies all implantations appeared
to occur cranially, near the tubal-uterine junction,
in accord with the report by Marshall [10].

The laparoscope proved to be a simple and
useful tool to examine the reproductive tract,
although the cumulative effect of capture, trans-
port, anaesthesia, bleeding and _ laparoscopy
obviously resulted in abortion. However the rate
was not significantly higher than in control (Group
I) animals, in fact a significantly greater percentage
of the control animals aborted than those in Group

876 W.R. DuKELow, C. S. T. Pow et al.

II. The present studies do confirm the high suscep-
tibility of these animals to abortion [7]. It is not
evident that laparoscopy caused this stress. This is
not to suggest that abortion cannot occur due to
other natural causes. Since the animals were held
in captivity for almost three years, age may have
been a factor for some females. Nelson [4] re-
ported that small, all-female groups were formed
after conception. Thus the continued presence of
breeding males in the present trials may have
exerted a negative effect. Two of the negative
control animals aborted before any of the group
were transported and moved. The remaining three
animals aborted subsequently without laparo-
scopic examination. In Group II, three of five
animals aborted without laparoscopic examina-
tion. Thus eight of ten animals (80%) aborted, six
of which received handling, anaesthesia and bleed-
ing stress. With animals which were also subjected
to laparoscopy, nine of ten (90%) subsequently
aborted. Of these nine, seven had known dates of
abortion and, of these, only four were within seven
days of the laparoscopy examination. Two of these
animals were laparoscopically examined twice with
abortion after the second examination. One
animal, laparoscopically examined on February
20, and subjected to anaesthesia, palpation and
blood sampling on March 14, subsequently gave
birth on May 10th. Thus there appears to be wide
variations in the susceptibility to these stressors,
but with a substantial level of abortion.

Progesterone levels were an accurate measure of
pregnancy. Towers [7] reported mean nonpreg-
nant progesterone levels in P. scapulatus to be 4.0
+0.5 ng/ml and never over 10 ng/ml. He reported
February pregnancy levels to range from 16 to 135
ng/ml. May pregnancy levels ranged from 102 to
195 ng/ml. In the present study, nonpregnant
animals averaged 2.9ng/ml of progesterone
(ranges 1.5 to 4.3). February pregnant animals
averaged 45.1 ng/ml and ranged from 15.9 to 79.7
ng/ml. March pregnant animals averaged 66.2
ng/ml and ranged from 34.9 to 103.1 ng/ml. No
pregnant animal was found with a progesterone
level of less than 15.9 ng/ml.

It is significant that of eleven animals initially
subjected to complete stress (Group III) six re-
mained pregnant for the subsequent three weeks

indicating, that while the abortion rate is high,
limited observations can be carried out to study
mid-pregnancy development within the uterus.
The effect of handling and laparoscopic stress
during very early pregnancy was not examined in
the present study.

The high incidence of abortion in Groups I and
II confirm the susceptibility of P. scapulatus to
stress-induced abortion and emphasizes the need
for compassionate and careful handling of the
animals in captivity.

ACKNOWLEDGMENTS

We thank Miss Lorraine Little and Drs. Gemma M.
O’Brien and Linda H. Crane for their assistance in the
project. This research was supported, in part, by a travel
grant from the University of Queensland (WRD), NIH
grant HD07534 (WRD) and ARC grant 529-88 (LM).

REFERENCES

1 MacDonald, D. (1984) The Encyclopedia of Mam-
mals. Pub. Facts on File, Oxford.

2 Hall, L. S. (1981) The biogeography of Australian
bats. In “Ecological Biogeography of Australia”.
Ed. by A. Keast, W. Junk Press, The Hague, pp.
1557-1583.

3 Hall, L. S. and Richards, G. C. (1979) Bats of
Eastern Australia, Queensland Museum Booklet
No. 12, Univ. of Queensland Press, Brisbane.

4 Towers, P. A. and Martin, L. (1985) Some aspects
of female reproduction in the grey-headed flying-
fox, Pteropus poliocephalus (Megachiroptera: Pter-
opodidae). Aust. Mammal., 8: 257-263.

5 Martin, L., Towers, P. A., McGuckin, M. A.,
Little, L., Luckhoff, H. and Blackshaw, A. W.
(1986) Reproductive biology of flying-foxes (Chir-
optera: Pteropidae). Aust. Mammal., 10: 115-118.

6 Pettigrew, J. D. (1986) Flying primates? Megabats
have the advanced pathway from eye to midbrain.
Primates, 231: 1304-1309.

7 Towers, P. A. (1987) Reproductive physiology of
female flying foxes (Pteropus spp.) in South East
Queensland. Ph. D. Thesis, 180 pp. University of
Queensland, Brisbane.

8 Grosser, O. (1901) Zur Anatomie und Entwicke-
lungsgeochichte des Gefassytemes der Chiropteren.
Anat. Hefte, Abt. 2, 17: 203-424.

9 Dukelow, W. R. and Ariga, S. (1976) Laparoscopic
techniques for biomedical research. J. Med. Prima-
tol., 5: 82-99.

10 Marshall, A. J. (1953) The unilateral endometrial

iit

12

University of Queensland, St. Lucia, Australia

reaction in the giant fruit bat (Pteropus giganteus,
Brunnich). J. Endocrinol., 9: 42-44.

Wimsatt, W. A. (1979) Reproductive asymmetry
and unilateral pregnancy in Chiroptera, J. Reprod.
Fert., 56: 345-375.

Bonilla, H. D. and Rasweiler, J. J. (1974) Breeding
activity, preimplantation development, and oviduct
histology of the short-tailed fruit bat, Carollia, in
captivity. Anat. Rec., 179: 385-404.

13

14

877

Pow, C. and Martin, L. (1987) Asymmetrical endo-
metrial development in flying foxes (Pteropus spp)
may be explained by local counter-current hormone
exchange. Proc. 19th Ann. Conf. Aust. Soc. for
Reprod. Biol., Sydney., p. 100.

Nelson, J. E. (1965) Behavior of Australian Pter-
opodidae (Megachiroptera). Animal Behavior, 13:
544-557.

ciel Aoki de tects’ fest sorte totais.” Ge |
veda, Acti S het atta) “all-femabe gnlepis ee: Rennes
| EE, fonts Sere OM, Fhius rie sunt eran. ‘i
Aneteding cinles in she: ‘pree ee ee

- Cerne i , : if

OperTR<3 nae} barivie efiecs, 5 TM ae ite Sepertivc =
WARTS atin sais ah Det Leet ge i
ZF oe D,

ra . ' my" oa = i i _ i
were tas insperid Dane ‘ead ‘pha eee.

elena “ihe whee, pomeae 5 Ls 2 AS AS. : his
exit ak ag ie ong. Banik ig ay, } ee ite “ aye: P 7 Pe. Ag honk ie to reiohie

Soong.” ss y : mh eS

ro

WARS 33 ste = CR Cae I AE

~ e a wt) * ie -
‘ LS Oats > ae oe ' 5 ~s io aye oie #
WERT > SL pe wae ae Cha SES
a . - > : ;
Ba rue 1. < ra ay « ~~ Ay ae oe
oe tra _s ’ ee \. - ‘a te OF eee
5 ‘ , = z
: = ” ’
an, 7 “? : > >
5 Se Free 1 x Rw ES RR) Waa eG
pos —
4 hn 7 re 5 = on = : i)
+ : - SE varity
b +t r 2 a ‘. Pe as
*
i i -
be a > ~ oS ue ; mS Py r7
ca . s é *e 4 a vt y
= he a rem ?
3 (iy “ ‘ ' r in. ?
i ale ; :
By 2 eS ee a Tog eek Dass
a - ¥ - ¢¢ - -
= , - « +
Gert ‘ ‘ <5 “ ey t
= Pa ‘7 ae a) ee yy * 4 v4 7
: f i
; -
{ Po i ee fh Pe 3 y fl . mt.)
== * ‘ . = 7 ¢ ;
“wa , = a
; z =a b : ryt,
4 ~~ “ ¢ "
j i ‘ =
7 7 : Fs {
- ~ 12. “te
iS ue
=. j
= ~
— vb “
mS iy a i ?
: a
c
4
' e
o's J Sr
- Pel i
5 ‘
_ yr" ay - = -
uf a > 5
y a 1 —_
= +4 vt
a i os
2 is i
a < 3
ab ra
\
‘ 4 . t
a ate
{
. y
* i
3
4
~
,
‘
¥
‘
=
4
{
iF - i" <4
)
4

ZOOLOGICAL SCIENCE 7: 879-887 (1990)

Strain Difference in Nucleotide Sequences of Rat Glycoprotein
Hormone Subunit cDNAs and Gene Fragment!

YuKIO KATO’, TOSHIHIKO EZASHI, TOSHIAKI Hira and Takako KaATo

Hormone Assay Center, Institute of Endocrinology, Gunma University,
Maebashi, Gunma 371, Japan

ABSTRACT—The cDNAs for common a, LHf and TSH subnits of glycoprotein hormone were
obtained from a cDNA library of Wistar-Imamichi-strain rat anterior pituitary. The a subunit cDNA
contained the entire coding region. The LHf cDNA coded 138 amino acids lacking three residues of the
signal peptide sequence and TSH@ cDNA contained a sequence from residue +20 to the carboxy
terminus residue +118. The three subunit cDNAs differed in their nucleotide sequences from those of
previously reported cDNAs of Sprague-Dawley (SD) rat, but the deduced amino acid sequences of the
three subunits were the same in the two species.

The rat FSH gene was cloned from the genomic library of SD rat using porcine FSH@ cDNA as a
probe. One clone was obtained from 10° plaques and a 1.7-kb EcoRI fragment was hybridized with the
probe. Serial deletion mutants were constructed for the subclones in a plasmid vector and the
nucleotide sequence was determined. The genomic fragment was found to consist of 1663 bp including a
957-bp intron, a 228-bp coding region and a 478-bp 3’-untranslated region. The clone contains part of
the FSHf subunit gene, but encodes most of the mature molecule from amino acid residue 34 to the
carboxy terminus residue 109. The sequence homology between the rat and porcine FSHf genes shows

© 1990 Zoological Society of Japan

high identity (87%) in the coding region but low homology in the intron.

INTRODUCTION

The pituitary glycoprotein hormone family con-
sists of luteininizing hormone (LH), follicle-
stimulating hormone (FSH) and _ thyroid-
stimulating hormone (TSH), which are composed
of a common a subunit and a unique # subunit [1].
The genes for these subunits are located on diffe-
rent chromosomes. Whereas TSH is synthesized in
the thyrotroph, two gonadotropins, FSH and LH,
are synthesized in the same type of pituitary cell,
the gonadotroph. Co-expression of the @ subunits
for FSH and LH was demonstrated previously by

Accepted April 24, 1990

Received March 17, 1990

The nucleotide sequence data reported in this paper
will appear in the DDBJ, EMBL and GeneBank
Nucleotide Sequence Deatbases under the accession
number D00575 for common a, D00576 for LH,
D00577 for FSH and D00578 for TSH8, respectively.
To whom reprint requests should be addressed.
Present Address: Laboratory of Reproductive Bio-
logy, National Institute for Basic Biology, Okazaki,
Aichi 444, Japan

=

WO N

in situ hybridization [2]. FSH and LH cooperative-
ly modulate the gametogenesis in females and
males, whereas TSH stimulates the production of
thyroid hormone. Several studies have demons-
trated that the synthesis of glycoprotein hormones,
as well as their secretion, is modulated in relation
to their biological function in vivo [3, 4]. Howev-
er, it is not yet known how the synthesis and
secretion of these three hormones are modulated
by several factors including hypothalamic hor-
mones, gonadal sex steroids and peptides, and
thyroid hormone. The cDNAs of glycoprotein
hormone subunits are particularly important for
investigation of mRNA levels and the control
mechanisms of gene expression. However, the
availabilities of the cDNAs of the glycoprotein
subunits are still limited. The present series of
studies was aimed at obtaining these cDNAs.
This report describes the determination of the
nucleotide sequences of the three glycoprotein
subunit cDNAs of the Wistar-Imamichi rat and of
the FSH £ gonomic DNA of the SD rat, and their
differences from those of the SD or Holtzman rat

880 Y. Kato, T. EzAsui et al.

reported previously [5-9]. Homology of gene
structures between the rat and porcine FSH £
subunits is also described.

MATERIALS AND METHODS

Isolation of rat glycoprotein subunit cDNAs

Total mRNA from anterior pituitaries of ten
female Wistar-Imamichi-strain rats was prepared
using an RNA extraction kit (Amersham, Bucks,
UK). The cDNA was synthesized according to
Gubler and Hoffman [10] followed by addition of
EcoRI linker and ligation to the EcoRI site of a
vector, Lambda ZAP (Stratagene, La Jolla, CA).
An orignal library of 1.8 x 10° independent clones
was amplified. About 210° clones of the am-
plified library were screened. Probes used were as
follows: porcine cDNAs for common a [11], LH @
[12] and TSH [11] and synthetic oligonucleotides
for common a (100 mer), LHP (148 mer) and
TSH (85 mer) [4, 11]. A multi-priming method
[13] was performed for the probes using a random
labeling kit (Takara Shuzo, Kyoto, Japan). Dupli-
cated filters with absorbed phage DNAs were
prepared on nylon membranes and the alkaline-
denatured DNAs were fixed by UV irradiation as
described in previous paper [11]. The plaque
hybridization was performed at 50°C in 6xSSC (1
x SSC: 0.15 M NaCl and 0.015 M sodium citrate,
pH 7.0) containing 0.02% bovine serum albumin,
0.02% ficoll, 0.02% polyvinylpyrrolidone and 1%
sodium dodecyl sulfate (SDS). The filters were
washed in 1XxSSC containing 1% SDS at 37°C.
The cDNA cloned into the Lambda ZAP vector
was automatically excised into a phagemid cassette
(pBluescript) in the presence of a helper phage
(R408) according to the instruction manual sup-
plied. The nucleotide sequences of the cDNAs
were determined by the chain-termination method
[14] using a-*’P-dCTP for the alkaline-denatured
double-stranded DNA [15] or a fluorescent dye-
labeled primer [16]. The fluorescence-labeled nuc-
leotides were analyzed on an Applied Biosystems
370A automated DNA sequencer (Foster City,
CA, USA).

Isolation of the rat FSHB gene and its nucleotide
sequence

The porcine FSHf cDNA insert [17] was excised
by EcoRI from pUC119 and labeled by a multi-
priming method [13] as described above. A geno-
mic library of a partial HaellI digest of SD rat liver
DNA constructed in Charon 4A was kindly pro-
vided by Dr. J. Bonner [18]. About 1x 10° clones
were screened by plaque hybridization, as de-
scribed above, and one clone was obtained. The
purified phage DNA was digested by EcoRI and
the fragment, which showed positive hybridization
with the porcine FSH@ cDNA, was constructed in
a plasmid vector, pUC119. Serial deletion muta-
tion was performed from both ends of the frag-
ment according to Yanisch-Perron et al. [19] fol-
lowed by sequence determination as described
above.

RESULTS

Isolation of rat glycoprotein hormone cDNAs

The cDNA was synthesized by priming with an
oligo-(dT),2_ 13 for total RNA without enrichment
of the poly(A)-rich RNA using an oligo-dT cellu-
lose column. No bands corresponding to the sizes
of ribosomal RNAs were observed upon agarose
gel analysis of the synthesized cDNAs (data not
shown). This procedure shortened the process and
increased the yield of cDNA clone from a small
amount of mRNA. Two cDNA clones for each of
common a and LH and one for TSHf were
obtained from about 210° clones. These num-
bers do not necessarily mean whole clones present
in the screened ones, since only the clones giving
rather intense hybridization signals were purified
and analyzed.

Nucleotide sequence of the rat common a cDNA

Two of the cDNA clones for common a had the
same length of 643 base pairs (bp) with different
sizes of the poly(A) tail (9 and 14 bases, respec-
tively). The nucleotide and the deduced amino
acid sequences of the common a are shown in
Figure 1. The common a cDNA codes the precur-
sor molecule consisting of a 24-amino acid signal

Cloning of Rat Glycoprotein Hormones 881

] (C)

Ge) 50

AGATCGACAATCAACTGCCCAGAACACATCCTTCCAAAGATCCAGAGTTTGCAGGAGA
100
GCTATGGATTGCTACAGAAGATATGCGGCTGTCATTCTGGTCATGCTGTCCATGGTCCTG
MetAspCysTyrArgArgTlyrAlaAlaValIlleLeuValMetLeuSerMetValLeu
mek = 20 =i @

15.0
CATATTCTTCATTCTCTTCCTGATGGAGACCTTATTATTCAGGGTTGTCCAGAATGTAAA
HisIlleLeuHisSerLeuProAspGlyAspLeullelleGinGlyCysProGluCysLys

Jat +10

200

CTAAAGGAAAACAAATACTTCTCCAAGCTGGGTGCCCCCATCTATCAGTGTATGGGCTGT
LeuLysGluAsnLysTyrPheSerLysLeuGlyAlaProlleTyrGlnCysMetGlyCys

+20 +30

250

TGCTTCTCCAGGGCATACCCGACTCCCGCAAGGTCCAAGAAGACAATGTTGGTTCCAAAG
CysPheSerArgAlalyrProThrProAlaArgSerLysLysThrMetLeuValProLys

+40 +50

350

AATATTACCTCGGAGGCCACGTGCTGTGTGGCCAAATCATTTACTAAGGCCACAGTGATG
AsnIlleThrSerGluAlaThrCysCysValAlaLysSerPheThrLysAlaThrValMet

+60 +70

400

GGAAACGCCAGAGTGGAGAACCACACGGACTGCCACTGTAGCACTTGTTACTACCACAAG
GlyAsnAlaArgValGluAsnHisThrAspCysHisCysSerThrCysTyrTyruHisLlys

+80 +90
450
TCGTAGCTTCCATGTGTGCCAAGGGCTGCGCTGACGACTGCTGACCCGTGCGATGGCACT
Ser*x**
+96
500

GAGTGTCGCCACCTCCTCCTTACCAGACTTCTGACACGCTTCAGTCATACACTGCTGCTT

350

TCCTGTCACATCCCTTATACTTCAGTACCATCGACAGTCTCTTCTCATTAGGGGAAAATG

643

TGTATCTACCATGGTCCCATCAGAAATAAAGCCTTTTCAATCATC(A)n

Fic. 1.

Nucleotide sequence of the rat common acDNA. Deduced amino acids are represented under the nucleotide

sequence. The termination code is indicated by («**). The nucleotides absent in and different from that of SD rat
common a cDNA [5] are indicated by underline and nucleotide above the sequence. Gaps and a polyadenylation
signal (AATAAA) are indicated by (—) and overlining, respectively.

peptide and a 96-amino acid mature molecule.
Upon comparison of the present sequence with the
cDNA sequence of common a from the SD rat
reported previously [5], two nucleotide differences
and an additional 14 bases in the 5’-untranslated
sequence are found, as indicated in Figure 1, but
the amino acid sequences are completely the same.

Nucleotide sequence of the rat LH8B cDNA

One of the two cDNA clones for LHB cDNA
was 495 bp long with 17 bases of poly(A), and was

26 bases longer in the 5’-region than the other (467
bp with 19 bases of poly(A)). The nucleotide and
the deduced amino acid sequence of LH are
shown in Figure 2. The LHS cDNA encoded of
part of the signal peptide and 121 amino acids of
the matuer LHf molecule. The clones lacked 3
amino acids at the N-terminus, but the deduced
amino acid sequence was the same as that of the
SD rat [6]. To construct a full-length cDNA clone,
a segment (indicated in Fig. 2) was chemically
added to the 5’-terminus of rat LH8 cDNA using a

882 Y. Kato, T. EZAsui et al.

1

ATCAAGAATGGAGAGGCTCCAGGGGCTGCTGCTGTGGCTGCTGCTGAGCCCAAGTGTG
MetGluArigLeuGlInGlyLeuLeuLeuTrpLeuLeuLeuSerProSerVa}]

- 20
50 (C)

= 10)
100

GTGTGGGCCTCGAGGGGCCCCCTTCGGCCACTGTGCCGGCCTGTCAACGCAACCCTGGCT
ValTrpAlaSerArgGlyProLeuArgProLeuCysArgProValAsnAlaThrLeuAla

tal

+10
150

GCAGAGAATGAGTTCTGCCCAGTCTGCATCACCTTCACCACCAGCATCTGTGCCGGCTAC
AlaGluAsnGluPheCysProValCyslleThrPheThrThrSerlleCysAlaGlyTyr

af 2 {U)

+30
200

TGTCCTAGCATGGTTCGAGTACTGCCAGCTGCCTTGCCTCCCGTGCCTCAGCCAGTGTGC
CysProSerMetValArgValLeuProAlaAlaLeuProProValProGIinProValCys

+40
250

+50

ACCTACCGTGAGCTGCGCTTCGCCTCTGTCCGCCTCCCTGGCTGCCCACCTGGTGTAGAC
ThrTyrArgGluLeuArgPheAlaSerValArgLeuProGlyCysProProGlyValAsp

+60
(T)

+70

CCCATAGTCTCCTTCCCTGTGGCCCTCAGCTGCCGCTGTGGGCCCTGCCGTCTCAGTAGC
ProlleValSerPheProValAlaLeuSerCysArgCysGlyProCysArgLeuSerSer

+80
350

+90
400

TCTGACTGTGGGGGTCCCAGGACTCAACCAATGACCTGTGACCTTCCCCACCTCCCCGGC
SerAspCysGlyGlyProArgThrGlnProMetThrCysAspLeuProHisLeuProGly

+100
400

eR D
450

CTTCTCCTCTTCTGATGCCCACCCACTAACTCCCCATTCTTCTGGAGCCAGCAGGTGTTC

LeuLeuLeu Phe * *
+120

495

ACCATCCCTCCCAATAAAGGCTTTACAACTGC(A) nh

Fic. 2. Nucleotide sequence of the rat LH@ cDNA. Deduced amino acids are denoted under the nucleotide
sequence. The termination code is indicated by («««). The nucleotides different from that of SD rat LHB cDNA
[6] are indicated by nucleotides above the sequence, respectively. The nucleotide segment which was added
chemically to the clone is boxed. A polyadenylation signal (AATAAA) is indicated by overlining.

synthetic oligonucleotide followed by the polymer-
ase chain reaction.

Nucleotide sequence of the rat TSH cDNA

The TSH cDNA was 369 bp long with 57 bases
of poly(A), and the nucleotide sequence is shown
in Figure 3. The cDNA codes part of the TSHf
molecule from amino acid residues 20 to 118. Two
differences in the nucleotide sequences between
Wistar-Imamichi and SD rats [7] were found in the
3’-untranslated sequence near the poly(A) tail.
One is a nucleotide difference and the other a gap

of three nucleotides. The two TSH8 cDNAs of SD
[7] and Holtzman [8] rats also show a few discre-
panncies.

Nucleotide sequence of the rat FSH gene

The cDNA for FSH was not obtained in the
course of the present cDNA preparation despite
independent screenings. This failure to prepare
cDNA probes for rat glycoprotein subunits,
however, was offset by cloning of the rat FSH
gene fragment. The structure and sequencing
strategy of the SD rat FSH@ gene are shown in

Cloning of Rat Glycoprotein Hormones 883

1 50
GCCTGACCATCAACACCACCATCTGCGCTGGGTATTGTATGACACGGGATATCAATGGC
LeuThrIlleAsnThrThrIleCysAlaGlyTyrCysMetThrArgAsplleAsnGly
+20 +30
100
WAAC TG TPCT OCCCANGTACGCACTCTCTCAGGALGTOIGIACATACAGAGACTICACE
LysLeuPheLeuProLysTyrAlaLeuSerGlnAspValCysThrTyrArgAspPhetThr
+40 7
150
EA GANG AVA GIG NGIGFAVAVA TP ALG © GiGGANG GOGAGACGATLGRTGGl CCl PAT iT GmGG DACCGG
TyrArgThrValGlulleProGlyCysProHisHisValAlaProTyrPheSerTyrPro
+60 +70
200
GTTGCCCTGAGCTGCAAGTGTGGCAAGTGTAACACTGACTACAGCGACTGTACACACGAG
ValAlaLeuSerCysLysCysGlyLysCysAsnThrAsplyrSerAspCysThrHisGlu
+ 80 + 8)
250
GCTGTCAAAACCAACTACTGCACCAAGCCACAGACATTCTATCTGGGGGGATTTTCTGGT
AlaValLysThrAsnTyrCysThrLysProGlnThrPheTyrLeuGlyGlyPheSerGly

+100 eG falles
350
TAACTGTAATGGCAATGCAATCTGGTTAAATGTGTTTACCTGGAATAGAACTAATAAAAT
* KK
(G) (GATS
IGA GANT eS CN

Fic. 3. Nucleotide sequence of the rat TSH& cDNA. Deduced amino acids are denoted under the nucleotide
sequence. The termination code is indicated by (+**). The nucleotides different from the TSH@ cDNA of the SD
rat [7] are indicated above the sequence. Gaps in comparison with the SD rat TSH8 cDNA and a polyadenylation
signal (AATAAA) are indicated by (—) and overlining.

Figure 4A. The clones of the rat FSH gene did
not contain the entire sequence. Three further
independent screenings failed to reveal a clone
containing the entire gene. The nucleotide sequ-
ence of the 1.7-kb EcoRI fragment is shown in
Figure 4B. The coding sequence of the rat FSH8

clone was surveyed by homology with porcine
FSH cDNA. The deduced amino acid sequence
consists of residues +34 to +109, the carboxy
terminus of the mature protein. The position of
intron/exon junction of the rat FSH gene was the
same as those in the human and porcine genes. No

intron-2

SS EEE EEEEEEEEEEEEEEEEEEEEEEEEEEE
—_—_———_ Bgl Il Eco RI
—————— SSS eae
—E
— i a eer
$<—$—____—__ ears SAE WORE Aa 2S
¢——________—_—_ =a
100bp Se
eee ——— ed
Fic. 4A.

884 Y. Kato, T. Ezasui et al.

GAATTCACACTGCTTCTTTGCATTAACACAAGCACAGAGAGAGTTCTCAATAGCCTCCTT
TTACCCATGTTACCTTGTCATTCGTTTAGAATCCCATCACAAACAGAAAAGTAGAACTTC
CTGATGAGATGCACTTTATITAATGATAAAATAGTTTGATCAAACTCTACTTCCCTICAG
TGTATTGCCCATGTCTACGGCTATCAATGAAGTAACATGGCACCGAAATAAAAAGAACAA
CAGCTTCATTTTCTTCCCTCAGCACAAATAGACTGTCCACTGGAAAGTGTTAGAAGGAA&
AATGTGTTGTTTTTTAAAGAAATTCTTATATGCAAAGTATATGTCCTTGGAAGGTAATAG
CACTACAAAGTAAATATCATTTAAAAGGGGAGAACTGTCTGCTAAGCAGTGGAATGATTA
GCATGTTACAAAAGTTCATAAACATATACTATATAATGTATACATTGATTGTATAATGAT
AAGCATTAATATTCATATCATATTATTTTACTTTTTTATAAATGCGAAACAATGTATACA
TTTCCTGAAAGTTAAGCTTTAAAATTTGTAGTTTAATTTGTCATTAATATATCTTTCAAL
TAAGTATGAAAAGGCAACCAAAATTCTAAATCATTATGTCTCTTTTCARCATTCTTCAAT
ATTGATAGTCAAAGTACCTTTGCATTTTGACAAATAGGTAATACTCTGATATGGTCATAG
AAATTATGAAAGAAATTACTTGCATTTTATGTAAGACTTCCAGAACAATTAGCGAGCATT
CCAGAAGAATTGGAAAATAAAGGACACTGCATAGGTGGTCTGTGGGGTTATGIGGTTTTA
ATAGTTAAATGTGACAGCATGCAGAATTAAATTACGGTCTCATTTTAATTAAAATGAACA
AAAACTGCAAATTACATATATTATATATTCTCATCCTCCCTTCCTTGAACTCCTCAGGAT

CTGGTGTATAAGGACCCAGCTAGACCAAACACC
ie

1000
AGAAAGTATGC
Bead tare uae a ied We hee unseen, InLysValCys
+

C
G
1050
GCTGTGCCCGCCACTCAGACTCC
eo Sa ee ace
+

GCAAGTGTGATAGTGACAGCA
lyLysCysAspSerAspSerT
+

omarmMN
O53 @&

(C) 1300
CHGGA GT CCG GCTGTPAAANC OA =f GC OlTPAGGGAC GAP GGG Cl CCA Gi AulminG i: CCrArC ie Giles
(T) 1350
CTGACAGGTAGTGAGGAGGAGCTCAGGACTGGGAGTGCCAGGGCCAGGACTCTATACCAC
1400
DOATECTITDCCCTCCATCOACACTGT TGATT ALCL AAGIOCATT CAC TOLCAACT CAGMGCm

1450 (G)
TGGGGCGGGGGGGTTGGAGGCTTTAATTTTCACCAATCTTAGGAATCTTCTAGAGCAACC
(2) (AT) ea ait (ea)
CTTTCCTTTAGACAAAGGGATGCATGAGTCC--AGGGAGGAAAGGAAAGTGGAAATCTTG

1600
TGAAAGAACTAAACCTAGCATAACCATCTGCTTTTCAGAGCCATCACACTTTTAAAGACT

CCAGC TGATTGCAAGCGAAAGTCTAAAGTCTGCATGGAAGGAATTC
Fic. 4B.

Fic. 4. Partial structure of the SD rat FSH gene. (A). Bold line and squares indicate an intron and an exon,
respectively. The coding region is indicated by a hatched square. Sequence was determined using deletion
mutants. The direction and length sequenced are indicated by arrows under the map. (B). The deduced amino
acid sequences of the SD rat FSH/ subunit are indicated below the genomic sequence. The termination codon is
indicated by («««). Differences from the nucleotide sequence of the Holtzman rat FSH cDNA [9] are indicated
above the nucleotide sequence. A gap is indicated by (—).

Cloning of Rat Glycoprotein Hormones 885

poly(A) addition sequence (AATAAA) was found
in the 3’-untranslated region sequenced. The
genomic fragment of the FSH subunit, excised by
restriction enzyme digestion with Sau3AI+
EcoRI, is suitable for investigaiton of the mRNA
levels of the gonadotropin subunit and its control
mechanisms.

Our present results confirmed that one uniden-
tified nucleotide in Holtzman rat FSH@ cDNA
determined previously by Maurer [9] is cytocine,
and that the amino acid at residue 41 is Ala. Upon
comparison of the nucleotide sequences between
the Holtzman rat cDNA and the SD rat genomic
FSH, differences were found to be present at nine
positions (indicated above the genomic sequence
in Fig. 4B). One of them is located in the coding
region, resulting in a one-base difference at amino
acid residue 39 (Arg for gene and Lys for cDNA,
respectively). The others (8 positions) are gaps
and substitutions in the 3’-untranslated region.

DISCUSSION

This study demonstrated that, among the gly-

coprotein subunits of Wistar and SD rats, common
a, LHf and TSHP have the same amino acid
sequences for the entire region (common a) or
most parts (8 subunits of LH and TSH), and that
their nucleotide sequences show slight differences.
We previously observed similar differences in the
nucleotide sequences among porcine pituitary hor-
mones among different strains [20, 21]. The Wis-
tar-[mamichi rat which shows a precise 4-day
estrous cycle was established from the Wistar
strain by inbreeding in 1961. On the other hand,
the SD rat strain was established by crossbreeding
between femals Wistar rats ans male rats of uncer-
tain strain rat in about 1925. The Holtzman rat
was established by crossbreeding using SD rats,
but the exact strains used are not clear. Therefore
differences may have arisen by allelic polymorph-
ism or differences in individual rats.

While the present study was in progress, Charib
et al. [22] also cloned and characterized genomic
DNA encoding the entire SD rat FSH molecule
using a genomic library, which is the same as that
used in our study. However, their data differed
from ours, and also from those of Holtzman rat

Rat FSHB gene

800 1000 1200 1400 1600

Porcine FSHB gene

Fic. 5. Homology of nucleotide sequences between the porcine and SD rat FSHP
genes. Harr-plot analysis was performed for the porcine (unpublished data) and
SD rat FSHf genes. Each dot represents a position where 15 our of 20 nucleotides

match in both sequences.

886
TaBLE 1. Survey of (U),A occurrence in the 3’
non-coding regions of glycoprotein hormone sub-
unit mRNAs
common a LH TSHP FSH"
rat =f 0 1 il
porcine 13 =3 8 78

Ranks of the listed mRNAs were calculated as
described in [24]. * calculated from the corres-
ponding 3’ non-coding region of rat [22] and porcine
[23] genomic sequences.

FSH@ cDNA. It is unclear why the SD rat genomic
library contains FSH clones differing in their
nucleotides sequence.

When Harr-plot analysis was perfomed between
porcine [23] and Wistar-Imamichi rat FSHP genes,
their coding regions were found to show high
homology (87% identity) but the intron-2 shows a
rather low homology (Fig. 5). On the other hand,
the 3’-untranslated regions show no homology.
Between the human [24] and porcine [22] FSH?
genes, however, high homology was observed in
the intron and 3’-untranslated region. The homol-
ogy in 3-untranslated region suggests that the
rodent FSH gene diverged in a manner rather
different from those of other mammals. Wres-
chner and Rechavi expressed mRNA stability in
terms of the extent of (U),A sequences in the 3”
non-coding region [25]. Table 1 shows the results
of a survey of (U),A occurrence in the 3’ non-
coding region of porcine and rat glycoprotein
mRNAs. According to the evaluation, four rat
glycoprotein subunits and porcine LHP have in-
termediate levels of mRNA stability. In contrast,
the porcine common a, TSH and FSH§ have high
values, suggesting high mRNA lability. In human
[24] and bovine [26] FSH@ mRNAs, two species
differences in the 3° non-coding region were found
and their stabilities were evaluated as intermediate
and highly labile, respectively. It would be in-
teresting to determine whether these differences in
mRNA stability function in the control of gonadot-
ropin biosynthesis.

ACKNOWLEDGMENTS

We are grateful to Mrs. Kyoko Tomizawa for her
technical assistance.

10

11

12

113)

Y. Kato, T. EZASHI et al.

REFERENCES

Pierce, J. G. and Parsons, T. F. (1981) Glycopro-
tein hormones: Strucuture and function. Annu.
Rev. Biochem., 50: 465-495.

Liu, Y. C., Kato, Y., Inoue, K., Tanaka, S. and
Kurosumi, K. (1988) Co-localization of LHf and
FSH@ mRNAs in the porcine anterior pituitary by in
situ hybridization with biotinylated probes.
Biochem. Biophys. Res. Commun., 154: 80-84.
Papavasiliou, S. S., Zmeili, S., Khoury, S., Land-
efeld, T. D., Chin, W. W. & Marshall, J. C. (1986)
Gonadotropin-releasing hormone differentially reg-
ulates expression of the genes for luteinizing hor-
mone a and f subunits in male rats. Proc. Natl.
Acad. Sci., U.S.A., 83: 4026-4029.

Kato, Y., Imai, K., Sakai, T. and Inoue, K. (1989)
Simultaneous effect of gonadotropin-releasing hor-
mone (GnRH) on the expression of two gonadotro-
pin f genes by passive immunization to GnRH. Mol.
Cell. Endocrinol., 62: 135-139.

Godine, J. E., Chin, W. W. and Habener, J. F.
(1982) a subunit of rat pituitary glycoprotein hor-
mones: primary structure of the precursor deter-
mined from the nucleotide sequence of cloned
cDNAs. J. Biol. Chem., 257: 8368-8371.

Chin, W. W., Godine, J. e., Klein, D. R., Chang,
A. S., Tan, L. K. and Habener, J. F. (1983)
Nucleotide sequence of the cDNA encoding the
precursor of the # subunit of rat lutropin. Proc. Natl.
Acad. Sci., U.S.A., 80: 4649-4653.

Chin, W. W., Muccini, J. A., Jr. and Shin, L. (1985)
Evidence for a single rat thyrotropin-@ gene: thyr-
oidectomy increases its mRNA. Biochem. Biophys.
Res. Commun., 128: 1152-1158.

Croyle, M. and Maurer, R. D. (1984) Thyroid
hormones decreases thyrotropin subunit mRNA
levels in rat anterior pituitary. DNA, 3: 231-236.
Maurer, R. A. (1987) Molecular cloning and nuc-
leotide sequence analysis of complementary deox-
yribonucleic acid for the @-subunit of rat follicle
stimulating hormone. Mol. Endocrinol., 1: 717-723.
Gubler, U. and Hoffman, B. J. (1983) A simple and
very efficient method for generating cDNA libraries.
Gene, 25: 263-269.

Hirai, T., Takikawa, H. and Kato, Y. (1989)
Molecular cloning of cDNAs for precursors of por-
cine pituitary glycoprotein hormone common a-
subunit and of thyroid stimulating hormone [-
subunit. Mol. Cell. Endocrinol., 63: 209-217.
Kato, Y. and Hirai, T. (1989) Cloning and DNA
sequence analysis of the cDNA for the precursor of
porcine luteinizing hormone (LH) f subunit. Mol.

Cell. Endocrinol., 62: 47-53.

Feinberg, A. and Vogelstein, B. (1983) A technique
for labelling DNA restriction endonuclease frag-

14

15

16

17),

18

ie

20

Cloning of Rat Glycoprotein Hormones

ments to high specific activity. Anal. Biochem., 137:
6-13. ;

Sanger, F., Nicklen, S. and Coulson, A. R. (1977)
DNA sequencing with chain terminating inhibitors.
Proc. Natl. Acad. Sci., U.S.A., 74: 5463-5467.
Hattori, M. and Sakaki, Y. (1986) Dideoxy se-
quencing method using denatured plasmid tem-
plates. Anal. Biochem., 152: 232-238.

Connell, C., Fung, S., Heiner, C., Bridgham, J.,
Chakerian, V., Herson, E., Jones, B., Menchen, S.,
Mordan, W., Raff, M., Recknor, M., Smith, L.,
Springer, J., Woo, S. and Hunkapiller, M. (1987)
Automated DNA sequence analysis. BioTechni-
ques, 5: 342-348.

Kato, Y. (1988) Cloning and DNA sequence analy-
sis of the cDNA for the precursor of porcine follicle
stimulating hormone (FSH) f subunit. Mol. Cell.
Endocrinol., 55: 107-112.

Sargent, T. D., Wu, J-r., Sala-Trepat, J. M., Wal-
lace, R. B., Reyes, A. A. and Bonner, J. (1979)
The rat serum albumin gene: Analysis of cloned
sequences. Proc. Natl. Acad. Sci., U.S.A., 76:
3256-3260.

Yanisch-Perron. C., Vieira, J. and Messing, J.
(1985) Improved M13 phage cloning vectors and
host strains: Nucleotide sequences of the M13mp18
and pUC19 vectors. Gene, 33: 103-119.

Kato, Y., Hirai, T. and Kato, T. (1990) Molecular
cloning of cDNA for porcine prolactin precursor. J.

21

22

25

24

5)

26

887

Mol. Endocrinol., 4: 135-142.

Kato, Y., Shimokawa, N., Kato, T., Hirai, T.,
Yoshihama, K., Kawai, H., Hattori, M-A., Ezashi,
T., Shimogori, Y. and Wakabayashi, K. (1990)
Porcine growth hormone: molecular cloning of
cDNA and expression in bacterial and mammalian
cells. Biochim. Biophys. Acta., 1048: 290-293.
Charib, S. D., Roy, A., Wierman, M. E. and Chin,
W. W. (1989) Isolation and characterization of the
gene encoding the (f-subunit of rat follicle-
stimulating hormone. DNA, 8: 339-349.

Hirai, T., Takikawa, H. and Kato, Y. (1990) The
gene for the # subunit of porcine FSH: absence of
consensus oestrogen-responsive element and pre-
sence of retroposons. J. Mol. Endocrinol., 5: (In
press).

Jameson, J. L., Becker, C. B., Lindell, C. M. and
Habener J. F. (1988) Human follicle-stimulating
hormone subunit gene encoded multiple messenger
ribonucleic acids. Mol. Endocrinol., 2: 806-815.
Wreschner, D. H. and Rechavi, G. (1988) Differen-
tial mRNA stability to reticulocyte ribonuclease
correlates with 3° non-coding (U),A sequences.
Eur. J. Biochem., 172: 333-340.

Esch, F. S., Mason, A. J., Cooksey, K., Mercado,
M. and Shimasaki, S. (1986) Cloning and DNA
sequence analysis of the cDNA for the precursor of
the @ chain of follicle stimulating hormone. Proc.
Natl. Acad. Sci., U.S.A., 83: 6618-6621.

r 1 Bre. mraseprpe een |
ry RRRRORT BOL cE eae ibrar ib

ees mga oe La eI ete

si Bbiti etal yo 2ei wn it il "
=| aT “Ae ay rai
ee. Ak art A ite ar

aT. eh Y Y oie ies ie pea
‘tp nop Satin. Hel snkI0q R pease ey ote 7a sag

aq bats, tf apnsins avianeaje Ange ao guy see p
an) +e tehtoebkes” =, a9 anions gia BS aca
ees raEy Eat LATS: Sw See frieiel bird! -

5 Biri att <y Aishithes ie.
fo estyenipra cay iet ig seat as
pence 24% eB uett FED: ae x2!
ca ay is tent yoget Ri)

a eel NAS
AR Jae,
‘tt Leet. ShO0 ato

ud
Fit oot Jeary is i

ts wate BRT we v5 ie is ee Peer ae rad
‘Sechelt Ea ce ie TORY Ree roi |
5) wien 3 ay sad - a: PF Sine settee

hie Sante irepevooedty h asi: Ke

1™

?
ae ‘ p 4
Bi cs os gt 2 aro La
~ Abespettttige Geet 7 ae
IWikHE Sonsupoe
tS +
A i
¢ lines 4 7
— d
ie ey Ei ‘ied
¥% a
:
4 ~
.
y
ey L
=
a
a * .
‘
—

ik SRE ES

it” We CH ate

Pe ccks Aki sert, 5.320%
10 a,

bas ro ET) husks =

pone

tei ia © ¢

q ae > gate x

not ig Laat
At eo £%

it ites eS ft

aA

naqeta ie: the

One. Ri
i

F eM

’ Pamtcs! J “ary ee: 2H
>
= Mpeg

‘ a Ny -
ST SOLO f Heep he:

1104 1 ye rh 38 th ‘

ZOOLOGICAL SCIENCE 7: 889-894 (1990)

© 1990 Zoological Society of Japan

Social Condition Influences Sexual Attractiveness
of Dominant Male Mice

SUSUMU HAYASHI

Department of Biology, Faculty of Education, Kagoshima University,
Kogoshima 890, Japan

ABSTRACT—The relationship between social dominance of male laboratory mice and their sexual
attractiveness was investigated by preference testing. Dominant and subordinate male mice used as
stimuli were taken either from the same cage or from different cages. Female mice preferred the odor of
a dominant male to that of a subordinate in both cases. The preference for dominant males disappeared
after the removal of the males’ preputial glands. Three kinds of dominant males were used as stimuli.
Dominant males which had encountered new rivals regularly were more attractive and had heavier
preputial glands than dominant males which had been housed with unchanged intact males. Males
which dominated unchanged intact males were more attractive and had heavier preputial glands than
males which had been housed with castrated males. It is concluded that both attractiveness and weight
of preputial glands of dominant male mice are maintained by rivals.

INTRODUCTION

The odors of male mice can attract female mice
which have cohabitated with a male in adulthood
or have been exposed to male odors for a certain
period before weaning [1]. In some species such as
the hamster [2] and the lemming [3], a female can
distinguish a dominant male from a subordinate on
the basis of olfactory cues. Female mice are
attracted to the odors of dominant male mice [4,
5]. However, the source of the odor which disting-
uishes a dominant male from a subordinate has not
been determined.

Dominant male mice have heavier preputial
glands than subordinates do [6], and the preputial
glands increase when the mouse engages in aggres-
sive behavior [7]. Therefore, the glands are indica-
tive of aggressiveness.

In addition, it is known that the secretions of the
preputial glands attract female mice [8-10]. In this
paper, female preferences for various kinds of
dominant males and the attraction effect of male
preputial glands were investigated.

Accepted March 3, 1990
Received December 14, 1989

MATERIALS AND METHODS

Animal

Mice (Mus domesticus) of the ICR-JCL strain
were used. The animals were born and reared in
our laboratory. Females were separated from
male siblings within three days of birth and a litter
composed of five or six female pups was reared by
a lactating female. When the pups were 11 days of
age, each mother with her litter was put into a
separate cage for 10 days. The cages with female
pups which were to be exposed to male odors were
divided lengthwise by a wire mesh barrier into two
equal compartments. One compartment housed
pups and a lactating female and the other housed
two adult males which were between four to five
months of age. These procedures were employed
to ensure that the female preferences were for
male odors [1, 10]. After the period of cohabita-
tion, the pups and their lactating mother were put
into a normal cage until weaning at 28 days of age.
Thereafter the females were housed in groups of 5
to 6 per cage and they were used as subjects only
once at 100-140 days of age.

890 S. HAYASHI

Preference test

The odor boxes were made of stainless steel
plates, measuring 6X9X5 cm, with an antecham-
ber (6X5X5cm) at one end. The antechamber
side walls and floor were made of stainless steel
plates and the roof was made of a transparent
polycarbonate plate through which the behavior of
a subject entering the antechamber could be
observed. The box was separated from its
antechamber by a stainless steel plate which had 60
holes 2mm in diameter. A mouse in the ante-
chamber could smell odors remaining in the box
through the holes.

Each male used as a stimulus was put into the
box with a small amount of saw dust for 90 min.
Immediately before the test, the male was taken
out of the box, leaving the saw dust in the box.
Two odor boxes which had contained different
males were placed in the right and left rear corners
of the female’s cage.

The females were isolated in a plastic cage
measuring 2417X12cm for four days. On the
second and third days, two clear odor boxes were
placed in the cage for 5 and 10 min, respectively,
to habituate the females to the boxes. Females
were tested on the fourth day. Two clear odor
boxes were placed for 10 min and then they were
replaced by the odor boxes which had each con-
tained a male just before testing. The behavior of
each female was observed for 5 min and the dura-
tion of head protrusion into the antechamber with
nose tip pointed towards the partition between the
box and the antechamber was recorded.

Although each pair of boxes was used twice, all
the data were handled independently.

Experiment 1

The dominant and subordinate males used as
stimuli were four to five months of age and were
prepared following two procedures. The first
(Relative dominance) group of 14 male mice which
had been picked at random (one from each of 14
cages housing six males each) were isolated for six
days. After isolation they were randomly divided
into seven pairs. The two males in each pair were
housed together in a cage for 12 days, but sepa-
rated by a wire mesh barrier to prohibit physical

contact on the second day. Thereafter the barrier
was removed for 20 min daily and the social be-
havior of the two males was observed. At the end
of this period, the male exhibiting the attacks was
designated dominant, and the male which had
been attacked was designated subordinate. There
were no subordinates which were hurt during the
cohabitation period. On the last day of cohabita-
tion, each dominant-subordinate pair was used in
the preparation of the odor boxes. Each pair of
odor boxes was presented to two females. A total
of 14 females were used.

The males of the second group (Independent
dominance) were picked from cages housing three
males each. They were put into new cages every
seven days and social behavior was observed for 30
min in the new cage. Seven dominant males which
had attacked two cagemates and seven subordinate
males which had not shown agonistic behavior and
had been the object of attacks from dominant
males were paired and used as stimuli. Each of the
two stimulus males came from different cages.
Their odors were presented to 14 females (each
pair was tested on two females).

On the day following the test, the preputial
glands of the stimulant males were resected and
weighed. This method is similar to that used in
previous reports [6, 7]. The resected glands were
freed from adhering fatty and connective tissues,
and the fluid contents were gently squeezed out;
then the wet weight of the glands was measured.

About 24 hr after the operation, they were put
into the odor boxes again for 90 min to use as odor
stimuli of another preference test. The same pairs
as the prior test were presented to two females. A
total of 28 females were subjected to the second
preference test. The other procedures were the
same as the prior test.

Experiment 2

Male mice were housed in 41 cages from wean-
ing to 11 weeks of age as an Independent domi-
nance group. They were transferred to a new cage
once a week. At 10 weeks of age, immediately
after they were transferred, aggressive behavior
was observed for 30 min. in order to detect the
dominant male. At 11 weeks of age, each cage
containing three males was divided into three

891 Female Preference for Male Odors

groups. Dominant males which had shown a
variable frequency of aggressiveness were appor-
tioned at random among the three groups. In 17
cages (Stable colony), the three members were not
changed from weaning to testing. In another 12
cages (Unstable colony), members were ex-
changed three times at 11, 13 and 15 weeks of age.
Each male met two strange males and established
new dominant-subordinate relationships at every
exchange. In the other 12 cages (No-rival colony),
dominant males which had been detected at 10
weeks of age were individually housed with two
castrated males from 11 to 17 weeks of age. The
castrated males had been castrated at 5 weeks of
age and were 4 months of age when they were
housed with a dominant male. The two castrated
males were changed at the same time as intact
males in the Unstable colonies. Each dominant
male from 10 Stable colonies was paired with

Dominant

Subordinate

either a dominant male from an Unstable colony
or an intact male from a No-rival colony and five
pairs in each were used as stimuli at 17 weeks of
age. Each pair was presented to two females.

The preputial glands of the other seven domi-
nant males in each group were weighed as in
experiment 1.

RESULTS
Experiment 1

The results were summarized in Figure 1 and
Table 1. Female mice preferred intact dominant
males over intact subordinate males in the Relative
dominance (Wilcoxon matched-pairs signed-ranks
test, P<0.01), and also preferred intact dominant
males over intact subordinate males in the Inde-
pendent dominance (P<0.01) (Fig. 1 and Table
1). After the operation, female mice did not show

Relative
dominance

oe Independent

eee
Zz

dominance

40s

Investigation Time (second)

Fic. 1.

Mean investigation time (sec) and SE in 5-min preference test. White columns show
attractiveness of intact dominant and subordinate males.

Hatched columns show

attractiveness of preputialectomized dominant and subordinate males.

TABLE 1.
Dominane Treatment
Normal (experiment 1)
Relative : ;
Preputialex (experiment 2)
Normal (e iment 1
Independent (expen )

Preputialex (experiment 2)

Female preference for male odors

Dominant Subordinate P
13* 1 0.01

i] d/ NS
12 2 0.01
4 10 0.05

* Number of females which investigated the odor box of a dominant male more frequently than that of a

subordinate one.

892 S. HAYASHI

Unstable

Stahle

No-rival

Investigation Time

(second )

Fic. 2. Female preference for three kinds of dominant males in two way choice situations.
The columns show mean investigation time (sec) and SE.

a reliable preference for preputialectomized domi-
nant or subordinate males in the Relative domi-
nance. However, in the Independent dominance,
preputialectomized subordinate males were more
attractive than preputialectomized dominant males
(P<0.05). The odor of intact dominant males with
preputial glands was investigated more frequently
than that of preputialectomized dominant males
(data from both groups were considered together,
t-test, P<0.01). The attractiveness of intact sub-
ordinate males was similar to that of preputialecto-
mized subordinate males.

The weights of the preputial glands of dominant
and subordinate males in the Independent domi-
nance were 61.6+6.4 mg (mean weight + SE) and
42.6+2.4mg, respectively. Those of dominant
and subordinate males in the Relatice dominance
were 72.4+5.5 mg and 48.4+4.1 mg, respectively.

Experiment 2

As shown in Figure 2, female mice investigated
the odor box which had contained a dominant
male from an Unstable colony more frequently
than the odor box which had contained a dominant
male from a Stable colony (Wilcoxon test, P<
0.01), and investigated the box of a male from a
Stable colony more frequently than that of a male

from a No-rival colony (P<0.01). The weights of
the preputial glands resected from dominant males
of Unstable, Stable and No-rival colonies were
80.4+7.4 mg, 60.6+6.6 mg and 46.0+1.9 mg, re-
spectively. The preputial glands of Unstable col-
ony males were the heaviest (Mann-Whitney U
test, Unstable vs. Stable, P<0.05). Preputial
glands respected from males of the Stable colony
were heavier than those of the No-rival colony
(Stable vs. No-rival, P<0.05).

DISCUSSION

Female mice preferred the odor of a dominant
male to that of a subordinate male. The prefer-
ence disappeared after the removal of preputial
glands. Dominant males which had encountered
new rivals regularly were more attractive than
dominant males which had been housed with un-
changed intact males. Males housed with intact
males were more attractive than males housed with
castrated males.

In experiment 1, female mice discriminated be-
tween dominant and subordinate males which had
been taken from different colonies (Independent
dominance). The social system in these colonies
after transfer to a new cage was a dictatorial

Female Preference for Male Odors 893

system. One animal dominated the other two and
it is in agreement with Butler [4]. These findings
suggest that male odors can be classified into two
behavioral types, dominant and subordinate.

In the case of subordinate mice, the investiga-
tion times before and after preputialectomy were
not significantly different. The lack of an effect of
removing the preputial glands could mean that the
subordinate male emits odors containing little, if
any, preputial odors, that the subordinate emits
little odor or that the odor of the preputial gland of
a subordinate male has little signal value to a
female. A previous study suggests that the weights
of preputial glands from subordinate males are
very light, with little variation [8]. The preputial
glands of subordinates increase if the animal has a
chance to achieve dominance [7, 8] and if the
glands are activated they appear to begin emitting
a preputial odor characteristic of a dominant male.

The attractiveness of a dominant male varied
with social conditions. As shown in experiment 2,
a dominant male mouse which regularly encoun-
tered new rivals was more attractive than a domi-
nant male in a stable colony. Winslow ans Misczek
reported that the frequency of attack declines as a
result of repeated encounters with an identical
animal and recovers when a new rival is introduced
[12]. The habituation of aggressice behavior may
result in the regression of the preputial glands.

Males of the No-rival colony did not develop
larger preputial glands and were less attractive
than males of the Stable colony. Dominant males
from the No-rival colonies were observed acting
aggressively toward castrated males after they
were transferred into new cages but the castrated
males did not show any aggressive reaction. In the
matter of preputial weight, it is known that gland
weight is increased by nontactile stimuli from an
intact male across a wire net barrier [7, 8]. Furth-
ermore, the gland increases more substantially
when periodic encounters with an intact male are
provided. Male odor and/or counterattack from a
cage-mate may be needed for the maintenance of
preputial gland activity.

DeLong reported that in feral condition move-
ments between populations occur mainly in juve-
niles or subadults and those invaders are often
repelled [13]. It seems that among three social

condition in experiment 2 Stable colony resembles
most closely to natural condition.

The precies effects of preputialectomy cannot be
estimated because preputial odor may be charac-
teristic of particular populations, colonies, strains
or species. It is known that females reared by
preputialectomized females prefer preputialecto-
mized females and males over intact females and
males, respectively. On the other hand, females
reared by intact females had reversed preferences
[5]. Therefore, it may be that male preputialec-
tomy reduces the general interest of subjected
females although they recognize the difference
between dominant and subordinate males.

Archer classified aggression into three types:
protective, parental and competitive aggression
[14]. The aggresion dealt with in this paper is
similar to competitive aggression. However, there
is an essential difference. That is, male mice in this
experiment did not seem to fight for female mice
or fight to repel rivals. Aggressive behavior in-
creases male preputial glands. If there are no
rivals, a male mouse cannot maintain increased
preputial glands. Therefore, it seems that male
mice seek rivals in order, as it were, to increase the
weight of their preputial glands, and thereby
attract female mice.

REFERENCES

1 Hayashi, S. and Kimura, T. (1978) Effects of
exposure to males on sexual preferences in female
mice. Anim. Behav., 26: 290-295.

2 White, P. J., Fischer, R. B. and Meunier, G. F.
(1984) The ability of females to predict male status
via urinary odors. Horm. Behav., 18: 491-494.

3. Huck, U. W., Banks, E. M. and Wang, S. C. (1981)
Olfactory discrimination of social status in the
brown lemming. Behav. Neur. Biol., 33: 364-371.

4 Mainardi, M. and Pasquali, A. (1973) Socio-sexual
preferences of the female mouse (Mus musculus) in
relation to the social status of the males. Ateneo
Parmense. Acta Naturalia, 9: 3-12.

5 Sandnabba, N. K. (1985) Differences in the capacity
of male odours to affect investigatory behaviour and
different urinary marking patterns in two strain of
mice, selectively bred for high and low aggressive-
ness. Behav. Process, 11: 257-267.

6 Bronson, F. H. and Marsden, H. M. (1973) The
preputial glands as an indicator of social dominance
in male mice. Behav. Biol., 9: 625-628.

7

10

894

Hayashi, S. (1986) Effects of a cohabitant on
preputial gland weight of male mice. Physiol. Be-
hav., 38: 290-300.

Bronson, F. H. and Caroom, D. (1971) Preputial
gland of the male mouse: Attractant function. J.
Reprod. Fert., 25: 279-282.

Hayashi, S. (1979) A role of female preputial glands
in social behavior of mice. Physiol. Behav., 23: 967-
969.

Hayashi, S. (1985) A preputial odor imprinted on
female mice. J. Ethol., 3: 89-91.

S. HAYASHI

11

12

13

14

Butler, R. G. (1980) Population size, social be-
haviour, and dispersal in house mice: A quantitative
investigaiton. Anim. Behav., 28: 78-85.

Winslow, J. T. and Misczek, K. A. (1984) Habitua-
tion of aggressive behavior in mice: A parametric
study. Aggress. Behav., 10: 103-113.

DeLong, K. T. (1967). Population ecology of feral
house mice. Ecology, 48: 611-634.

Archer, J. (1988) The Behavioural Biology of
Aggression. Cambridge Univ. Press, London, pp.
29-158.

ZOOLOGICAL SCIENCE 7: 895-906 (1990)

© 1990 Zoological Society of Japan

Insemination-Dependent Modification of Circadian Activity
of the Mosquito, Culex pipiens pallens

YOSHIHIKO CHIBA, YUTAKA YAMAMOTO, CHIEKO SHIMIZU, MAKOTO ZAITSU,

MortomicH! Uxki, Makoto YosHn and KENJI TOMIOKA

Environmental Biology Laboratory, Biological Institute,
Yamaguchi University, Yamaguch 753, Japan

ABSTRACT— Male accessory gland (MAG) of the mosquito Culex pipiens pallens produces a chemical
factor, which is conveyed in seminal fluid on copulation and alters the female circadian activity pattern.
This was demonstrated by injecting the gland homogenate (MAGH) into virgin females, as was done in
C. p. quinquefasciatus [1]. Free-running circadian activity of the inseminated female was analyzed for
each individual on the basis of a long recording of more than ten days. It was first demonstrated that the
critical dose of MAGH, to which 50% of mosquitoes respond, was 0.25 to 0.5 gland equivalents and that
MAGH frozen in saline at —20°C for two months was still active.

INTRODUCTION

The activity of mosquitoes is under circadian
control [2-4]. As to the physiological mechanism,
involvement of morning and evening oscillators
has been hypothesized from a kinetic analysis of
individuals by Jones [5] and Clopton [4, 6], though
they give their own definitions of the oscillators.
On the other hand, surgical operations on the
central nervous system yielded a hypothesis that
the midbrain (cerebral lobes) contains the pace-
maker (or a crucial part for the pacemake’s func-
tion) and that an extraocular photoreceptor is
required for entrainment [7, 8]. Another aspect to
the mosquito rhythm is that its waveform in the
female is significantly modified by insemination
(Anopheles gambiae [9, 10]; Culex pipiens pallens
[11]; An. stephensi [12]), and the modification is
caused by the male accessory gland substance (C.
p. quinquefasciatus [{1]).

In the present study using C. p. pallens, we
could roughly reproduce the insemination depen-
dent modification reported for the other mos-
quitoes but found some unique tendencies which
are not fully explained by the earlier hypothesis.
In addition, as a preliminary step to elucidating the

Accepted April 21, 1990
Received March 10, 1990

chemical and physical nature of the rhythm mod-
ifying factor produced by the male accessory
gland, some properties of the crude extract of the
gland, such as dose-dependency and tolerance to
freezing temperature were demonstrated for the
first time. A part of the experimental results,
presented in a symposium [13], has been analyzed
thereafter more profoundly and is included in this

paper.

MATERIALS AND METHODS

Female mosquitoes, Culex pipiens pallens Co-
quillett, were from a laboratory stock which origin-
ated from a field population two years before and
has had field-collected mosquitoes often added
thereafter. The field collections were made in the
Yamaguchi City area. The stock had been kept
continuously in a room lighted from 0400 to 2000
(LD 16:8 h) and kept at 25+0.5°C, the ‘standard’
environmental conditions used also for all experi-
ments except where stated.

Activity recording

Flight or locomotor activity of individual fe-
males was recorded by the computerized system,
including actographs operated photoelectrically, as
described in detail elsewhere [7]. Relative humid-
ity was not controlled, but it was near saturation

896 Y. CuiBpA, Y. YAMAMOTO et al.

because of the presence of water and a sucrose
solution in the activity chamber. |

Experimentation

The first experiment was intended to investigate
the effects of insemination on the entrained circa-
dian activity patterns. The activity was recorded
for females that had spent 7 to 10 days after
emergence with males to be inseminated or with-
out males. In the second experiment, activity
recording started with newly emerged virgin
females. After about 10 days in the activity
chamber, they were marked with different colors
on their thorax for identification and then were
either held for a few days in a cage with several
males for copulaiton or with only females. The
latter served as a control. Thereafter, they were
returned to their activity chambers, to resume
recording. This enabled us to evaluate the effect of
insemination on individual insects. In the third
experiment, to study the effect of insemination on
the free-running circadian activity, mosquitoes
were held in constant darkness (DD) after being
treated in either of the ways adopted in the first
and second experiments. The fourth experiment
was designed to investigate the effect of injection
of the male accessory gland homogenate (MAGH)
intraindividually; the virgin female received injec-
tion around day 10 of recording. MAGHs equiva-
lent to various numbers of male accessory glands
were injected to test the dose-dependency. The
effect of MAGH obtained from frozen accessory
glands was also investigated.

Insemination was verified by dissecting the sper-
matheca of all females used after the recording in
each experiment.

Statistics

Pearson’s correlation coefficient (r) was used as
an index of similarity in the entrained circadian
pattern between mosquitoes; an average activity
per hour, obtained for a definite hour of day from
the multiple-day record, was taken as a variable
and degree of correlation between the variable in
one mosquito and that in another was quantified
by calculating r. The calculation was made for all
combinations of two mosquitoes, and mosquitoes
showing a high value of r with one another were

classified as having a similar temporal pattern (Fig.
1). In a strict sense, r can be an estimate for two
variables with normal distribution. This condition
may not be satisfied here, but the usage of r is
convenient in terms that, as the two entrained
circadian patterns become more similar, the value
of r is expected to be larger.

On the other hand, the average entrained circa-
dian patterns were compared using three kinds of
indices (Figs. 2, 6, and 8): activities in percentage
between 2100 and 2400 (night activity; NA) and
between 0400 and 0700 (morning activity; MA),
and the NA to MA ratio (NA: MA). The average
NA, MA, or NA: MA before treatment, such as
injection of MAGH, was compared with the post-
treatment value using one-tailed t-test. Mos-
quitoes showing NA higher than 20% and MA
lower than 40% were regarded, for convenience,
as having the inseminated pattern, and the number
of them was compared between two different
treatments by chi-square test. These statistics are
summarized partly in Table 1. The other statistical
analyses will be mentioned as occasion demands.

Preparation of male accessory gland homogenate
(MAGH)

Typically, 20 pairs of accesory glands were dis-
sected from 10-day old unmated males into saline
(0.9% NaCl) kept in an ice bath. Centrifugation at
2,000 x g (4°C) for 10 min forced the glands to the
bottom of the centrifuge tube, and supernatant
was discarded. The glands were homogenized with
a glass bar. After the addition of saline, centri-
fugation for 30 min (12,000Xg) at 4°C yielded a
clear pale yellow supernatant. This supernatant,
diluted in some cases with saline, was used as
MAGH for injection, the dose being reported as
male accessory gland equivalents.

To see effects of freezing on MAGH activity,
the accessory gland was kept in saline at —20°C for
60 days before homogenization.

Injection of MAGH or saline

The etherized virgin female was placed ventral
side up on a clay bed and the thorax was fixed by a
band of clay. Each female received 0.5 wl of
MAGH injected through the pleural region of the
fourth to sixth abodominal segments with a capil-

Inseminated Mosquito Circadian Rhythm 897

lary tube drawn to a fine tip. Controls were
sting th 0. ane
injected with 0.5 wl ef saline 40 : 3

RESULTS

w
oO
=
=
ll
—==)—_—_—_

1. Effects of insemination of activity rhythm
Activity under LD 16:8

Fifty seven females from the ‘with female’ cage
all were found virgin in the post-recording dissec-
tion of spermathecae and all 60 mosquitoes from 104 | |
the ‘with male’ cage were inseminated. All mos-
quitoes formed two activity peaks punctually at im
lights-off (evening) and -on (morning) and were 0 | | b
active sometimes in the dark period. The temporal
patterns of the 117 mosquitoes were classified into

three types, A, B, and C using the correlaiton
coefficient (Fig. 1). Type A is characterized by

Activity (%o)
N
Bese

Time of day
Fic. 1. Three types of female mosquito (virgin or in-
seminated) circadian patterns synchronizing with

weak activity in the dark period and the much TEDiIGC 8) Conhdent interwalen G0c2
CO T—E—————————trti“‘O;O;CCCS
Day 1- Eup 1 Control Control 80
—— a 6) Po (N=25) te) (TypeA,N=17)
=, —, => o@
= 1 =i so = (oy 60
= = = eu (exe)
= at 0085
= iL Gell steal J = > % 8
~
housed with 22 aa 2 - $ 40 oe)
e e 2
oe ;
= SS e fo)
20— = Ls 17 —— = 2 o° s % 20 Ss
= ¥ maT —
a= ——" I > S
cB)
me 03
. Bo. 8
ic = gel Experimental Experimental Oo
Beier sers ernnt SSE GO Ss
Sl Mery al ge RN Se eae) ©o (N=33) (TypeA ,N=21) ES
— i = c Q =
—-—T iieiiieac > ea =e C=O 2
a i 711 o O 60 =
7 re LE = “| @ e %
1 = ot = T > O e
= — Tr 0 O Cc
Hh Xe oD
housed with 68 = All ae Oo 40 o
r mir st T z= rol (eo) (cD
a wT wo % S a
= a as =_ °?@ <
——— ee 20 =
AQ) SSeS EES e e? Co &°
a eee
— o e
a a eee oe ry r
———— SS ag) et ( QO,
ZZ 24 125-10 20 40 60 0 20 40 60
Time of day NA:Percent activity between 21°° and 24°°

Fic. 2. Effect of insemination on the circadian pattern. Left: Activity record of a female, virgin throughout the
experiment (the upper panel), and a female inseminated at the halfway point (lower panel). Right: Average night
activity between 2100 and 2400 plotted against average morning activity between 0400 and 0700 for each mosquito
before (open circle) and after (solid circle) the interim treatment (holding experimental females for a few days
with either males or females). Squares: Average for all mosquitoes before (open) or after (solid) the interim
treatment. Triangles: Circadian activity calculated for type A (open) or type C( solid) patterns in Fig. 1. For
further explanation, see text.

898 Y. Curpa, Y. YAMAMOTO et al.

higher morning peak than the eveing one. This
relation between both peaks was reversed in type
C, which is marked also by an intense activity
following the evening peak and declining toward
the end of the dark period. Type B is intermediate
between the two. The frequency distribution over
these three types was quite different between the
virgin and the inseminated females (77=34.996, p
<0.005); 53, 30, and 17% of the virgins showed
types A, B, and C, respectively, compared to 8,
27, and 65% among inseminated females. Thus,
more than half of the virgin females and the
inseminated showed the separate types A and C
patterns, respectively. On the other hand, as many
as 30% showed a common type B.

The intraindividual comparison study revealed a
clear effect of insemination which occurred after
the halfway of recording (Fig. 2). This particular
female housed with females for an interim few
days showed type A pattern throughout the ex-
perimental day (upper left panel). The other
female (lower left panel) changed the activity
pattern drastically from type A to C, after return-
ing from the cage with males; she was verified
inseminated by post-recording dissection. What
was seen in these particular females was demonstr-
able in the others, as shown in the middle and right
panels. The middle panels are based on all mos-
quitoes verified virgin (upper) or inseminated
(lower); insemination caused a decrease of MA
and an increase of NA or NA:MA, while a
decrease of MA occurred also in the controls
(without insemination), to some extent.

These tendencies stand out more when NA and
MA are plotted only for females having type A

TABLE 1.

between pre- and post-treatment were examined by one-tailed t-test.

treatment were compared by chi-square test

pattern before the halfway treatment (right
panels). After the halfway treatment, the number
of females with the inseminated activity pattern,
defined earlier conveniently as showing NA higher
than 20% and MA lower than 40%, was increased
more largely in the experimental group than in
control (Table 1).

Activity under DD

Sufficient data for analysis were obtained from
44 virgin and 27 inseminated females in this experi-
ment. Both groups revealed three kinds of free-
running circadian patterns under DD (Fig. 3).
Pattern I is diphasic, comprising the free-running
evening peak and a second peak. But, some
mosquitoes were less punctual, so the two peaks
were not separated clearly (Pattern II) or showed
only the free-running evening peak (Pattern III).
In addition, there were mosquitoes showing the
other patterns without anything in common also to
each other. These patterns were classified as
‘undefined’ (Table 2). Frequency distributions
over the four types differ between the virgin and
the inseminated, there being more diphasic
females in the former.

To describe the temporal activity pattern in
more detail, we extrapolated the time the activity
starts and ends on the first day (arrow-indicated) of
DD and obtained the free-running period (tau) by
fitting a straight line visually on the daily activity
onsets (line 1, Fig. 3, I, hereafter called onset line)
or offsets (line 2, offset line). In virgins, however,
an additional line 3 (resumption line) was fitted on
the activity resumptions (second peak onsets) for
the first several days of DD. In most inseminated

Statistics for the data depicted in Figs. 2,5, and 7. Differences in NA, MA, and NA:MA

Effects of two kinds of the

Control: Experimental: Injected with: Injected with:
Virgin inseminated saline MAGH MAGH _ frozen MAGH
NA:MA IS (GSB) 4.16 (3.68) 1 O7= 4.40 4.94 2.97
t NA Se nst) Shee (B17) ns 7.18 VSL 8.47
MA Aon) 8.46 (9.61) 3.40 5.66 8.27 7.46
IC: 16.31 (13.78) 12.67 ns

ns: not significant, *: p<0.05, without *: p<0.01
( ): type A

Inseminated Mosquito Circadian Rhythm

899

Inseminated

i

1

:

5

1

,

ahh

|

1
dss a4

rail
il

Per

1]
3
|

Time of day

Fic. 3. Activity records of three individuals with different circadian patterns (1, I, and II) are shown for virgin or
inseminated females. For further explanation, see text.

TABLE 2. Number of females showing three types of free-running circadian pattern under DD (Fig. 3)

N
I II
virgin 44 16 6
inseminated Di 2 11

females, the activity offset was delayed in one
straight line with the passage of days as long as the
recording continued. In four mosquitoes, the
offset changed its moving direction, several days
after DD started, from backward to forward (Fig.
4), requiring two offset lines with different inclina-
tions. The line fitting was done only for the
punctual onset or offset. The results were summa-
rized in Table 3 and Figure 5.

In the virgin female, there was no difference
between taus at activity onset (tau,,) and at offset
(tauogs) (t=1.41, p>0.05), suggesting that both
run in parallel so that daily activity time seldom
changes. In the majority of the diphasic virgin
females (Pattern I), the tau at activity resumption

Types
x7
Ill Undefined
i : 11.43 0.01
9 5 AS) =U),

(tau,;sm) is longer than tao (t=3.38, p<0.01).
The extrapolated times of the activity onset
(ET,,), offset (ET or), and resumption (ET;sm)
were 1920, 0855, and 0058, respectively. Thus, the
daily duration of second peak was, on average,
eight hours in the beginning but was shortened, as
the days pass, to form a triangle active zone.
Consequently, the second peak disappeared in
some mosquitoes, which, thereafter, showed only
the free-running evening peak. Some mosquitoes
still showed the second peak punctually along the
extension of offset line every day with a definite
duration of a few hours to make the diphase with
the free-running evening peak.

In the inseminated female with one straight

900 Y. CnHrsBA, Y. YAMAMOTO et al.

DD >

12 12 12
Time of day

Fic. 4. Activity records, with onset (1) and offset (2) lines, of two inseminated females judged as restoring the
free-running virgin patterns.

TABLE 3. Average values of free-running period (tau) and extrapolated time (ET, clock hour)
for activity onset, resumption, and offset

Virgin Inseminated
Activity
N tau Ea N tau Em
onset 30 23.96 +0.48 1920255 1 238284 1945 + 43
(74 23522 02 2004 + 55)
resumption 7 24.50+0.37 0058+ 63 — a —
offset 10 23ST A taOl53 0855 + 107 12 =24.20+0.41 0233 + 93

4s22417220°23 0148 +32)*
23465 == O12 0529 + 49)**

tau: Free-running period. Et: Extrapolated time.
( ): Females with activity offset changing the free-running direction at the halfway. *: Values before
the direction changed. **: those after.

Numerals following the average are S.D. which, in ET, isin min. For further explanation including
Statistical analysis, see text and Fig. 4 showing the above-tabulated values diagrammatically.

offset line fitted, tauo¢ was longer than tau,, (t= 2.01). Tauog was longer (t=11.24, p<0.01) and
2.86, p<0.01); activity time is lengthened with the | ET ,- occurred much earlier. It is interesting that
passage of days. In comparison with virgin, there — taUggp was not different from the virgin tau, (t=
was no difference in tau, (t=1.40, p>0.05), buta 1.53, p>0.05) and ET o¢ was much closer to ET, sm
subtle difference in ET,, (1950; t=2.02, toos= rather than to ET os in the virgin, though with the

Inseminated Mosquito Circadian Rhythm 901

Time of day
24

Days after start of DD

Inseminated. 1

Inseminated.2

onset
offset

Socccas resumption

Fic. 5. Average lines of activity onset, offset, and re-
sumption for the virgin and inseminated females.
Lower panel: inseminated female showing virgin
pattern restoration.

statistical difference (t=2.35, p<.0.05). The activ-
ity resumption could not be investigated, because
only two were clearly diphasic.

In the four inseminated mosquitoes requiring
the two offset lines (Fig. 4; lower panel, Fig. 5),
taUo¢e Was longer in the earlier days than tau,, (t=
4.85, p<0.01) but, this difference soon dissolved (t
=1.39, p>0.05), so the onset and the offset
gained a parallelism as in the virgin female. The
earlier tauog was not different from that in the
inseminated female with one straight offset (mid-
dle panel, Fig. 5; t=0.15, p>0.05) and also from
virgin tau,,, (upper panel, Fig.5; t=1.55, p>
0.05). Tatiog in the later days was not different
from that of virgins (t=0.22, p>0.05). Tau,, was

a little shorter than that of the straight offset
female (middle panel, Fig. 5; t=2.18, to.95=2.09)
but not different from that of the virgin (t=1.79, p
>0.05). Finally, it may be noteworthy that, if the
later offset line is extended straight back to the first
day of DD (Day 0 in Fig. 5), a triangle inactive
zone is formed with the earlier offset line. Thus, in
brief, four females showed activity pattern similar
to that of the other inseminated females in the
earlier days of DD, but thereafter the virgin pat-
tern was restored in a sense that the onset and the
offset lines became parallel.

2. Injection of MAGH to virgin females

Type A virgin females were selected from the
preliminary recording and were injected, at the
age of 10 days, with MAGH of a given dose
between 1300 and 1500. When the MAGH equiva-
lent injected was, for example, 0.5 accessory
glands, 80% (15/19) of the virgin females changed
their original pattern to type C typical of insemi-
nated female (lower panels, Fig. 6). The three
kinds of indices of activity, NA, MA, and NA
:MA all changed significantly (Table 1) in the
same directions as in inseminated females (Fig. 2).
The effect of injection was not manifested on that
day but appeared at the earliest the next day. On
the other hand, when only saline was injected as a
control (zero dose), most mosquitoes retained the
original type A pattern (upper panels, Fig. 6).
But, as in the virgins from the ‘with female’ cage
(Fig. 2), MA tended to decrease to some extent
and the resultant increase of NA:MA_ was
observed (upper panels, Fig. 6; Table 1).

We tested the other doses to obtain a typical
saturation curve (Fig. 7); number of virgin females
with the inseminated pattern reaches 50% between
doses of 0.25 and 0.5 MAG equivalents. The
inseminated pattern was judged visually from the
computer-plotted activity tracings. Mortality in-
creased from 15 to 40% as the dose increased from
zero- to 4-accessory gland equivalents. The num-
ber of mosquitoes, used to calculate the percen-
tage, ranged from 11 to 24.

MAGH from accessory glands frozen in saline at
—20°C was still as active, enhancing NA: MA and
NA and lowering MA (right panel, Fig. 8, Table
1), as that from normal (unfrozen) accessory

902 Y. CuHIBA, Y. YAMAMOTO et al.

80
Saline
Saline (Type A, N=19)
Day l- ms 2) ee __ 60
wT =
= SS °
—o — °
[= nT ™
> 11 = SS £0) 2
= = 4.9)
eee oe = c
= =~ bi
= — LO Po:
17- TT —_ 2)
(‘Ss
o
a
~ MAGH(0.5gland equivalent) MAGH Po
i= = ma (Type A,N=19) >
an y= ——-—- 60 >
— |_S__ — eiiee aoe _—
= Li |___Fan ey Paes U
Ts SS ©
rr a ee =
| es a TT Cc
> airs =. wove |_aCaaaeaes EAT SF 40 o
a ee ee i 2
ye wo
mm ORB ee ee ao
Se 20 $=
19- a
Sa Se a Sl 0
12 24 12 @) 20 40 60
Time of day NA:Percent activity between 21°° and 2490

Fic. 6. Effects of injection (arrow head) of saline (upper panels) and MAGH (lower panels) on circadian pattern.

For further explanation, see Fig. 2 and text.

100

80

607

407

205

Percent animals showing inseminated pattern

“MAG” equivalents

Fic. 7. Dose dependency of the virgin female activity
modification by the injection of MAGH. Confident
intervals: 95%.

glands was (left panel, derived from the data at
one gland equivalent in Fig. 7). The frequency of
mosquitoes, judged to gain the inseminated pat-
tern in the ‘frozen’ group, was not different from

that in the ‘normal’ group (Table 1).

DISCUSSION

1. Chronobiological models for insemination-
induced modification

Jones and his co-workers investigated the effect
of insemination on circadian activities of the
female mosquitoes, Anopheles gambiae {9, 10] and
C. pipiens quinquefasciatus [5]. They found a
common tendency between the two species that,
under LD 12:12, insemination reduces the even-
ing and morning activity peaks but enhances the
night activity, which declined toward the end of
dark period, and that this modified pattern persists
under DD. These findings were reproduced
roughly in our pallens material which, however,
showed somewhat unique tendencies.

Our main findings are as follows. (1) Circadian
activities in the virgin and inseminated females
were essentially diphasic with an evening peak and

Inseminated Mosquito Circadian Rhythm 903

9 80

SS e3 MAGH (N=19)
‘2 ac (1gland equivalent)
g 60 at 60

S ° @ Oo

c O

v ©

3 40 @ O

S O & e @ @

= @

>

ia © @

3 20 ore

~ O ©

7 @

‘a ®

ie)

< 0 20 40

=

O Homogenate of
frozen MAG (N=12)
O (1gland equivalent)
@
O
Q
g @
0 @
&
@
@
S ee
&
@

60 O 20 40 60

NA: Percent activity between 21°° and 24°°

Fic. 8. Effects of two kinds of MAGH on circadian pattern, one from normal MAG (left) and the other from frozen
MAG. For further explanation, see Figs. 2 and 6, and text.

a second one. (2) In most virgins, the activity
onset line was parallel with the offset line, which
forms a triangle active zone (the second peak) with
the resumption line. The second peak often dis-
appeared after the formation of triangle active
zone. (3) In the inseminated female, tage was
longer than tau,, and was not different from the
virgin taU,,n. However, in some of the insemi-
nated females, tauyg; was soon shortened and the
virgin pattern was restored in the sense that the
activity offset line became parallel with the onset
line.

What is a coherent model for the above-itemized
results? Concerning the virgin circadian rhythm,
Jones [5] postulated the existence of two oscilla-
tors; one controls the eveing peak and the other
controls two slave oscillators underlying, respec-
tively the morning peak and a night one which is
not always manifested. He assumes that insemina-
tion enhances the night peak, but suppresses the
morning one. This view is based mainly on two
observations with C. p.quinquefasciatus virgin or
inseminated; i) the evening and night peaks
showed different free-running periods under DD
and ii) the night and the morning peaks were

equally less dependent on light intensity in contrast
to the evening peak, which recurs with a longer
period at higher light intensity.

From the study with the virgin female of Culiseta
incidens, Clopton [4, 6] also postulated the exist-
ence of evening and morning oscillators, which are
of different nature from Jones’. That is, the
morning oscillator is stable and entrains the even-
ing oscillator, which controls directly the temporal
waveform of overt activity but is so labile as to
often escape control from the morning oscillator
and consequently free-runs with its own period.
Judging from his figures, the waveform is frequent-
ly diphasic as in Jones’s and our material. Thus, a
natural consequence of this postulation would be
that insemination affects the waveform of the
evening oscillator, though Clopton did not allude
to this point.

Both hypotheses follow the Pittendrigh-Daan
model [14] that, in nocturnal rodents, the two
mutually coupled oscillators predominantly con-
trol evening and morning activity (or activity onset
and offset), respectively, though the coupling is
assumed by Clopton to be weak in the mosquito.
Experimental evidence for this model is that taug,

904 Y. CHIBA, Y. YAMAMOTO et al.

and taUos¢ Show reciprocal responses to light in-
tensity, which were observed in quinquefasciatus
and Culiseta.

Our observations do not contradict Jones’
hypothesis, so far as the data shown in the upper
two panels of Figure 5 are concerned. If we regard
the triangle active zone as his morning peak, then
its tau may be measured, in one sense, from the
intermediate between the resumption and the
offset lines, showing a different value from tau of
the evening peak, meeting his hypothesis. Furth-
ermore, his assumption explains to some extent
the day-by-day change in the inseminated wave-
form. That is, the triangle active zone (corres-
ponding to his morning peak) disappeared after
insemination and, instead, the night activity was
intensified and free-running with a different tau
from the evening peak. But, the discontinuation of
morning peak after the triangle active zone forma-
tion in the virgin (upper panel, Fig.5) and the
restoration of the virgin pattern (parallelism be-
tween the activity onset and offset lines) as was
observed in the four inseminated (lower panel,
Fig. 5) remain unexplained.

On the other hand, the linear parallelism be-
tween the activity onset and offset lines in the
virgin female and its restoration in the inseminated
suggests the existence of a mechanism functioning
stably irrespective of virginity (Table 3, and upper
and lower panels, Fig. 5). This reminds us of the
Clopton’s hypothesis that the morning oscillator
stabilizes the circadian system. From this hypoth-
esis, it may be speculated that the entrainment
between the two oscillators continues after insemi-
nation, but that insemination modifies the wave-
form of the evening oscillator. Non-parallelism
between activity onset and offset in 80% of insemi-
nated females (Table 3 and middle panel, Fig. 5)
may not be contradictory to this view, assuming
that the virgin pattern (the parallelism) would have
to be restored, if the recording would continue
longer. However, the day-by-day change in the
waveform (formation of the triangle active or
inactive zone) is not explained by Clopton’s
hypothesis. His material, Culiseta, also often
forms triangle active zone, but, unlike the case of
our material, Culiseta’s triangle was thought to be
a consequence that tau, 1s longer than tauogs.

Thus, the published postulations do not fully
explain our data. But, if these are adequately
combined and are compensated by the other view,
the unexplained parts can largely be reduced. A
useful view in this connection is hinted at by
Aschoff et al. [15] that the waveform of an oscilla-
tor is asymmetric, interacting with a threshold,
which changes as days pass. This may explain the
discontinuation of morning peak in the virgin,
which is unexplained by Jones, and the day-by-day
change of the evening oscillator’s waveform, which
awaits some extension of the Clopton hypothesis.
Furthermore, to explain the restoration of virgin
pattern in the inseminated, the Jones model might
well be so calculated that the night or morning
peak free-runs backward until it attains a definite
phase angle from, for example, the evening peak.
When the virgin pattern is restored may depend on
the relation between tau,, and tauog (the night
activity offset); the shorter the tau, is relatively to
taUor, the sooner the definite phase angle is
attained. This fits our data; tau,, (indicated by the
inclination of onset line) was significantly shorter
in the lower panel of Figure 5 than in the middle,
but there was no difference between the earlier
taUorp (indicated by the inclination of offset line) in
lower panel and tauo¢ in the middle. |

At present, it is hard to conclude as to which of
these plausible explanations is to be preferred.
But in any event, the unique point in our data,
restoration of virgin pattern (the parallelism) in
the inseminated female, could be observed by the
long term recording (20 days at longest) under DD
and by analyzing individual rhythms; Jones’s post-
ulation was based on observations for 4 to 6 days
and the average waveform of multiple individuals.
Restoration was observed in only 20% of the
inseminated females, yet this can not be disre-
garded, because we frequently saw similar phe-
nomena in our previous series of study [11].

Finally, there are two other mechanisms that
deserve future investigation equally to the above-
mentioned. The first is that the morning peak is
advanced by insemination to become the night
peak, as mentioned in Jones’s earlier report on
inseminated Anopheles [9, 10]. The morning peak
moves backward under DD within a limited range
to became parallel eventually with the evening

Inseminated Mosquito Circadian Rhythm 905

peak. The second point is to apply a very simple
view [15] to the whole rhythm; the mosquito
circadian rhythm is controlled by one oscillator
having an asymmetric waveform, different be-
tween the virgin and inseminated females, in-
teracting with a threshold that changes as days
pass.

2. Effects of the male assessory gland homogenate
(MAGH)

The entrained virgin female, injected with
MAGH, showed a circadian pattern (Fig. 6) very
similar to that shown by the inseminated female
(Fig. 2). This demonstrates the reproducibility of
experimental results by Jones and Gubbins [1]
using C. p. quinquefasciatus and means that the
MAGH includes the chemical ‘rhythm modifying
factor(s), RMF’. The preliminary steps of our
study revealed that the activity of this factor de-
pends on dose, the accessory gland equivalent, and
persists during the span of freezing at —20°C.

Immediately after copulation the bursa copulat-
rix is distended with semen. Spermatozoa enter
the spermathecae within a few minutes in Psor-
ophora howardii {16] (cited by [17]) and Aedes
aegypti [18] (cited by [17]); the sperm were
observed to be transferred without MAG secretion
in the latter. Electron micrographs of the male
secretion within the bursa of female Ae. aegypti
reveal male accessory gland cells or their frag-
ments, including numerous mitochondria and va-
rious membranes [19]. Thus, it is highly likely that
the bursa is the first place in the female where
RMF might act to alter the circadian pattern. In
this connection, the following two questions would
be worthy of consideration. Does the RMF itself
move into the circulation of the female through the
wall cell of the bursa or does it stimulate the
release of a second factor by the wall cells, which,
in turn, alters the female’s behavior? The injection
with MAGH into the thorax [5] or the abdomen
(present study) were effective. Furthermore, we
recently found that injection of a fraction of
MAGH [13] and transplantation of MAG (Shimi-
zu and Chiba, unpubl.) into the abdomen were
effective. These facts suggest that the RMF itself
passes through the bursa wall.

There are some chemicals which are transferred

from the male to the female insect on copulation to
modulate her behavior. For example, a prostag-
landin stimulates female crickets to oviposit [20].
A peptide, produced by MAG, makes females
reject copulation in mosquitoes [21] and in Dro-
sophila {22]; the molecular weight was reported to
be 2,000 in the former and 4,000 to 5,000 (esti-
mated from the reported number of component
amino acids) in the latter. The chemical nature of
RMF is a main concern in the next step of our
study.

Another important issue to be addressed is how
the RMF itself, or the second factor, modifies the
female circadian activity. To resolve this issue, the
target organ on which the effective factor acts
should be identified. In our mosquito, we have
already shown that the circadian pacemaker is
probably located in the protocerebral lobe [7]. If
the effective factor acts on the pacemaker, it
should gather to the brain. In this respect, the
report of Young and Downe [21] is informative in
that the peptide controlling female sexual receptiv-
ity concentrated in the head in an experiment using
°H-leucine labeled substances.

The critical dose of MAGH, 1.e., to which 50%
of mosquitoes respond positively, was found to be
0.25 to 0.5 MAG equivalents. It has been known
that a male of Ae. aegypti can successfully insemi-
nate five to eight females sequentially [17], sug-
gesting that only a small fraction of the MAG
content is released during a single insemination. If
this is so in C. p.pallens, then the critical dose we
determined is reasonable.

These studies, including ours may contribute to
the biology of the insect accessory gland that has

attracted so much attention for its role in repro-
duction [23].

ACKNOWLEDGMENTS

We are grateful to Dr. Arden O. Lea, Department of
Entomology, the University of Georgia for reading the
manuscript. We thank Dr. Susumu Y. Takahashi, Facul-
ty of Liberal Arts and Science, Yamaguchi University for
his valuable advice from a biochemical viewpoint.
Thanks are also due to Dr. Katsuhiko Endo of our
laboratory for his technical help. This study was sup-
ported by Grants-in-Aid for Scientific Research from the
Japan Securities Scholarship Foundation and the Minis-
try of Education, Science and Culture of Japan

906

(61840024).

10

11

iV)

REFERENCES

Jones, M. D. R. and Gubbins, S. J. (1979) Mod-
ification of female circadian flight-activity by a male
accessory gland pheromone in the mosquito, Culex
pipies quinquefasciatus. Physiol. Entomol., 4: 345-
Soul.

Chiba, Y. (1964) The diurnal activity of the mos-
quito, Culex pipiens pallens in relation to lighting
condition. I. Activity under the continuous lighting
condition. Sci. Rep. Tohoku Univ. Ser. IV. (Biol).,
30: 67-75.

Jones, M. D. R., Hill, M. and Hope, A. M. (1967)
The circadian flight activity of the mosquito Anoph-
les gambiae: phase setting by the light regime. J.
Exp. Biol., 47: 503-511.

Clopton, J. R. (1984) Mosquito circadian and
circa-bi-dian flight rhythms: a two-oscillator model.
J. Comp. Physiol. A, 155: 1-12.

Jones, M. D. R. (1982) Coupled oscillators controll-
ing circadian flight activity in the mosquito, Culex
pipiens quinquefasciatus. Physiol. Entomol., 7: 281-
289.

Clopton, J. R. (1984) Mosquito circadian flight
rhythms: differential effects of constant light. Am. J.
Physiol. , 247: 960-967.

Kasai, M. and Chiba, Y. (1987) Effects of optic lobe
ablation on circadian activity in the mosquito, Culex
pipiens pallens. Physiol. Entomol., 12: 59-65.
Chiba, Y. and Tomioka, K. (1987) Insect circadian
activity with special reference to the localization of
the pacemaker. Zool. Sci., 4: 945-954.

Jones, M. D. R. and Gubbins, S. J. (1977) Mod-
ification of circadian flight activity in the mosquito
Anopheles gambiae after insemination. Nature, 268:
731-732.

Jones, M. D. R. and Gubbins, S. J. (1978) Changes
in the circadian flight activity of the mosquito
Anopheles gambiae in relation to insemination,
feeding and oviposition. Physiol. Entomol., 3: 213-
juph).

Chiba, Y., Shimizu, C. and Kasai, M. (1985) Mod-
ification of female moquito circadian activity by a
series of reproductive behaviors. In “Neurosecretion
and the Biology of Neuropeptides”. Ed. by H.
Kobayshi, H. A. Bern and A. Urano, Japan Sci.
Soc. Press, Tokyo/Springer-Verlag, Berlin. pp. 497-
503.

Rowland, M. (1989) Changes in the circadian flight

13

14

15

16

77,

18

19

20

7

D2

23

Y. Cuipa, Y. YAMAMOTO et al.

activity of the mosquito Anopheles stephensi associ-
ated with insemination, blood-feeding, oviposition
and nocturnal light intensity. Physiol. Entomol., 14:
77-84.

Chiba, Y., Yoshii, M., Yamamoto, Y., Tomioka,
K. and Takahashi, S. Y. (1989) Physical and che-
mical properies of a male secretory factor modifying
female circadian pattern in the mosquito Culex
pipiens pallens. In “Host Regulated Developmental
Mechanism in Vector Arthropods.” Ed. by D.
Borovsky and A. Spielman, Florida Medical En-
tomology Laboratory, Vero Beach, Florida, pp.
199-205.

Pittendrigh, C. S. and Daan, S. (1976) A functional
analysis of circadian pacemakers in nocturnal ro-
dents. V. Pacemaker structure: a clock for all sea-
sons. J. Comp. Physiol., 106: 333-355.

Aschoff, J., Gereche, A., Kureck, H., Rieger, P.
P., Saint Paul, U. V. and Wever, R. (1971) Inter-
dependent parameters of circadian activity rhythms
in birds and man. In “Biochronometry”. Ed. by M.
Menaker, Natl. Acad. Sci., Washington DC, pp. 3-
Die

Lum, P. T. M. (1961) The reproductive system of
some Florida mosquitoes. II. The male accessory
glands and their role. Ann. Ent. Soc. Amer.,. 54:
430-433. (cited by Clements, 1963)

Clements, A. N. (1963) International Series of
Monographs on Pure and Applied Biology. vol. 17.
The Physiology of Mosquitoes. Pergamon Press.
Oxford, London/New York/Paris.

Schwartz, P. H. (1961) Behavior of spermatozoa of
Aedes aegypti (L.). M. Sc. Thesis, University of
Maryland, 45 pp. (cited by Clements, 1963)

Jones, J. C. and Sheffield. H. G. (1970) The fine
structure of the seminal bursa of Aedes aegypti.
Mosq. News, 30: 270-271.

Loher, W., Ganjian, I., Kubo, I., Stanley-
Samuelson, D. and Tobe, S. S. (1981) Prostaglan-
dins: Their role in egg-laying of the cricket Tele-
ogryllus commondus. Proc. Natl. Acad. Sci. USA.,
78: 7835-7838.

Young, A. D. M. and Downe, A. E. R. (1987) Male
accessory gland substances and the control of sexual
receptivity in female Culex tarsalis. Physiol. En-
tomol., 12: 233-239.

Chen, P. S. (1988) A male accessory gland peptide
that regulates reproductive behavior of female D.
melanogaster. Cell, 54: 291-298.

Leopold, R. A. (1976) The role of male accessory
glands in insect reproduction. Ann. Rev. Entomol.,
21: 199-221.

ZOOLOGICAL SCIENCE 7: 907-912 (1990)

© 1990 Zoological Society of Japan

Role of the Medullary Raphe Nucleus in Regulating Sexual
Behaviors in Female Rats

KOREHITO YAMANOUCHI

Neuroendocrinology, Department of Basic Human Sciences, School of Human
Sciences, Waseda University, Tokorozawa, Saitama 359, Japan

ABSTRACT— Role of the medullary raphe nucleus in regulating female sexual behaviors was examined
in female rats. Radiofrequency lesions either of the raphe obscurus nucleus (ROBL) or the raphe
magnus nucleus (RMGL) were made in ovariectomized rats. Subsequently, female sexual behavior
tests were started 2 days after the subcutaneous implantation of a Silastic tubing containing E, and
carried out daily for 5 days. The ROBL females showed high scores of lordosis quotient in the first test,
in comparison to those of RMGL, Sham-operated and non-brain surgery control females. In addition,
incidence of ear-wiggling behavior in the ROBL group was higher than that in the control group. These
results suggest that the raphe obscurus nucleus may play an inhibitory role in regulating not only lordosis

but also soliciting behaviors in female rats.

INTRODUCTION

Inhibitory and facilitatory signals for the induc-
tion of female sexual behavior are known to de-
scend from the forebrain, such as the septum and
the ventromedial hypothalamus, to the lower
brainstem [1, 2]. Further, neuroendocrinological
and electrophysiological studies suggest that the
midbrain central gray is a critically important
center of female sexual bahaviors and a possible
acting site of the forebrain influence [3-5]. The
pontine medial periventricular gray and adjacent
areas have also been reported to be a minimally
required neural component for the display of lor-
dosis behavior, based from results showing that its
destruction caused a severe loss of lordosis [6, 7].

In the medulla oblongata, the lateral vestibular
nucleus and the nuclei of reticular formation are
thought to be closely associated with descending
facilitatory mechanisms for lordosis regulation,
since destruction of these nuclei resulted in impair-
ment of lordosis and, conversely, their electrical
stimulation facilitated lordosis behavior [8-11].

Recently, we have reported that lesions in the
dorsomedial pontine tegmentum including the dor-

Accepted April 26, 1990
Received April 18, 1990

sal raphe nucleus facilitated lordotic response [12].
The dorsal raphe nucleus contains amount of sero-
tonergic neurons which send axons to the forebrain
and cortex [13]. It is generally thought that
serotonin plays an inhibitory role in the regulation
of female sexual behaviors, because an injection of
a serotonin receptor agonist decreased sexual re-
ceptivity, whereas a synthesis inhibitor of seroto-
nin potentiated lordosis in female rats [14-16]. In
this study, in order to elucidate the role of the
medullary raphe nucleus which send serotonergic
fibers to the spinal cord, in female sexual behaviors
regulation, either the raphe obscurus nucleus or
the raphe magnus nucleus was destroyed and sex-
ual behaviors were observed in female rats.

MATERIALS AND METHODS

Wistar female rats (230-255 g), housed under a
controlled photoperiod (14: 10 hr, light : dark) and
temperature (23-25°C), were castrated and sub-
jected to the brain surgery. Two different raphe
lesions were made stereotaxically in the medulla
oblongata by a radiofrequency lesion generator
(Radionics Inc. Burlington, MA) under ether
anesthesia. An incisor bar having been set at 5 mm
below the interaural line, a electrode (0.7 mm) was
lowered 9.5 mm from the bregma level at a point

908 K. YAMANOUCHI

12.5 mm posterior to the bregma on the midline
(raphe obscurus nucleus lesion, ROBL, 14 anim-
als) or lowered 10.5mm from the bregma at a
point 10.5mm posterior to the bregma on the
midline (raphe magnus nucleus lesion, RMGL, 8
animals). To produce these radiofrequency le-
sions, the current was applied and the temperature
at the electrode tip was kept at 50-52°C for 1 min
in both the ROBL and the RMGL groups. For
comparison purposes, castrated controls (CON-
TROL, 11 animals) and castrated sham-operated
controls (SHAM, 7 animals) were prepared. In
the SHAM group, the electrode was lowered to
the level of the raphe obscurus nucleus in the same
manner as for the ROBL group, but no corrent
was applied.

Four to 5 weeks after the operation, all females
were implanted subcutaneously with a Silastic tub-
ing (3 cm in length, 1.57 3.18, i.d.xo.d.; Dow-
Corning No. 602-285) containing estradiol-17
(E,; Sigma). The behavioral testing was initiated 2
days after E,-implantation, and carried out daily
for 5 days. In the behavioral tests, each ex-
perimental female was placed in an observation
cage with two vigorous males and the lordosis
quotient (LQ, number of lordosis/10 mounts, x
100) was recorded. The presence or absence of
soliciting behaviors (ear-wiggling and hopping)
also was recorded for each animal.

At the end of the test, all animals were sacrificed
Each brain then was removed and fixed in 10%
formaline solution. Subsequently, frozen sections
stained with cresylechtviolet were made in order to
determine the precise localization of the lesion.
LQs were analyzed using the F-test and then by the
t-test or the Cochlan Cox method. The data of
incidence and the median value of soliciting be-
haviors were evaluated statistically by means of an
¥° analysis and the Mann-Whitney U-test.

RESULTS

The mean LQs in each group are shown in
Figure 1. In the first test (2 days after E>-
implantation), only 5 of 11 CONTROL females
showed lordosis behavior and the mean LQ of this
group was 14.5+5.8. In the SHAM and RMGL
females, incidences of lordosis also were low (1 of

100

C)
|
»
(S
v
»
S 50
CG © CONTROL (11)
co) 4 SHAM(7)
2 @ ROBL (14)
me) @ RMGL(8)
‘=
(eo)
zi

0

2 3 4 5 6
days after E, administration
Fic. 1. The mean LQ and S.E. of each group in S tests.

Behavioral tests were started 2 days after sub-
cutaneous implantation of a E,-containing Silastic
tubing and carried out daily for 5 days. The mean
value of ROBL group is significantly higher than
those of the other groups (P<0.05) in the first test
(day 2 after E,). Abbreviations: ROBL, Raphe
obscurus nucleus lesion; RMGL, Raphe magnus
nucleus lesion.

7 and 2 out 8 animals, respectively), the mean LOs
being 11.4+5.8 and 5.0+3.8, respectively. In
contrast, 12 out of 14 ROBL females displayed
lordotic responses and the mean LO (45.7+8.4)
was significantly higher than those of the other
groups (p<0.05). In the second test, mean LQs of
the SHAM and RMGL groups were comparable to
that of the CONTROL group (see Fig. 2). In the
ROBL group, the high LQ value (80.8+5.1) was
obtained, but was not statistically different from
those of other groups. In the subsequent three
tests, the mean score of LQ in each group in-
creased gradually with no statistically differences
among the groups.

Incidences of ear-wiggling and hopping in the
first test and the median for the onset of soliciting
behavior (day of E,-implantation=day 0) are
summarized in Table 1. Only one female in the

Medullary Raphe and Lordosis 909

Fic. 2. A photomicrograph of representative frontal section of the raphe obscurus nucleus lesion (ROBL)

(Cresylechtviolet stain).

TABLE 1.
onsets of soliciting behaviors

Incidence (posi./total)

Incidence of lordosis, ear-wiggling (EW) and hopping (HOP) in the first test and

Day of onset (median)

Group

Lordosis EW HOP EW HOP
CONTROL 5/11 1/11 0/11 4
SHAM 1/7 0/7 0/7 4 5
ROBL 12/14 8/14* 2/14 a 5
RMGL 2/8 1/8 0/8 3 355)

* P<0.05 vs CONTROL and SHAM (77 test).

** P<0.05 vs CONTROL (U-test).

CONTROL group showed ear-wiggling in the first
test. Incidence of ear-wiggling of the SHAM and
RMGL groups were comparable to that of CON-
TROL. In the ROBL group, 8 out of 14 animals
displayed ear-wiggling, being significantly higher
than those in the CONTROL and the SHAM
groups (P<0.05). The median for the onset of
ear-wiggling in the CONTROL and SHAM groups
was day 4 (the third test). The onset of ear-
wiggling of ROBL group was advanced, the me-
dian being day 2 (P<0.05, U cal=25.5 vs CON-
TROL). In all groups, onset of hopping behavior
was delayed, in comparison with ear-wiggling (see
Table 1). The medians for the onset of hopping
were almost same in all groups. In the behavioral

testings, locomotor activities in the lesioned
females did not seem to significantly differ from
those observed in the castrated controls, although
systematic analyses of such activities, except for
female sexual receptivity, were not pursued in the
present experiment. There were no statistical
differences in the mean body weights among the
experimental groups.

The locations of the lesoins were determined
histologically on reference to the Paxinos and
Watson [17]. The ROBL was located midline of
the medulla oblongata at the level of the inferior
olive nuclei (Fig. 2). The ROBL damaged the
raphe obscurus nucleus and adjacent area. In all
ROBL females except 2 animals, the rostral tips of

910 K. YAMANOUCHI

the raphe obscurus nucleus at the level of the facial
nucleus were intact. In 3 ROBL females, the
caudal ends of the raphe magnus nucleus seemed
to be lesioned. The caudal part of the raphe
obscurus nucleus, being close to the pyramidal
decussation, also remained intact in all ROBL
females. The ROBL extended dorsally to the
medial longitudinal fascicles. The ventromedial
area of the medulla oblongata including the medial
part of the inferior olive nuclei and the dorsal
tissue of the raphe pallidus nucleus were destroyed
in most ROBL females. In 2 ROBL rats, the
lesions extended ventrally to the dorsomedial part
of the pyramids.

In the RMGL group, the lesions were in the
ventromedial area of the medulla oblongata, from
the level of the superior olive nucleus to the level
of the facial nucleus. The RMGL damaged the
raphe magnus nucleus and a part of the raphe
pallidus nucleus. In most RMGL females, the
lesions penetrated to the dorsomedial part of the
pyramids. In both the ROBL and RMGL rats, the
medial lemniscus appeared to be partially dam-
aged.

DISCUSSION

In the present experiment, the ROBL potenti-
ated lordosis behavior and ear-wiggling but the
RMGL did not show any significant effects. The
raphe obscurus nucleus in the medulla may partici-
pate in an inhibitory system, not only of lordosis
but also of soliciting behavior in female rats, the
raphe magnus nucleus may not. This may indicate
that there is a clear functional difference in mod-
ifying female sexual behaviors between the nucleus
raphe obscurus and magnus. Both of the two
raphe nuclei contain serotonergic cells which send
their axons to the spinal cord [13, 18]. Serotonin
has been suggested to play an inhibitory role in
regulating lordosis behavior from the results of
experiments with using drugs that are known to
modify activity of the serotonergic neurons [14-
16]. The serotonergic fibers originated from the
raphe obscurus nucleus terminate to the ventral
and intermediolateral horn and these from the
raphe magnus nucleus to the dorsal horn [13].
Thus, there is a topographical difference in bulbos-

pinal serotonergic fibers between these two me-
durally raphe nuclei. The facilitatory effect of the
ROBL could be due to the interruption of the
descending serotonergic fibers to the ventral and
intermediate horn. However, the expected reduc-
tion in serotonin contents in the spinal cord was
not seen in estrogen-treated female rats during
high levels of a lordosis responding [19]. A dys-
function of the spinal serotonergic neurons by an
intrathecal injection of neurotoxin for the sero-
tonergic neurons did not enhance lordosis [20]. It
may still be premature to conclude that facilitation
of female sexual behavior by ROBL is due to
destroy the descending serotonergic system. On
the other hand, there is a considerable body of
evidence indicating that the ascending serotonergic
systems are involved in the lordosis inhibiting
mechanisms [21-23]. In our previous results, it
have also been showed that lesions in the dorsome-
dial midbrain tegmentum including the dorsal
raphe nucleus which send axons to the forebrain
potentiated female sexual behaviors [12].
Another possible explanation for facilitatory
effects of ROBL on female sexual behaviors is
concerned with non-serotonergic ascending mech-
anisms of the raphe obscurus nucleus, based from
the facts that 65% of the neurons in the raphe
obscurus nucleus and 85% of these in the raphe
magnus were non-serotonergic [24]. Further, ana-
tomical evidence suggests that many axons of the
raphs obscurus nucleus terminate in the mesen-
cephalon [25]. In the midbrain, amount of neural
substrates are involved in the regulation of lordosis
and soliciting behaviors [26]. Especially, the mid-
brain central gray has been demonstrated to play
an important role in regulation of both lordosis
and soliciting behaviors [4, 27-29]. In the lower
part than the midbrain such as the pons, the system
regulating soliciting behavior seems to be dissoci-
ated with the system regulating lordosis, because
destruction of the dorsomedial pontine tegmentum
[6, 7] or the interruption of the ventral norad-
renergic fiber bundle [30] resulted in loss of lordo-
sis but not of soliciting behaviors. Informatively, a
similar dissociation phenomenon has not been
found in the midbrain level. From the present
results, simultaneous enhancement of soliciting
and lordosis behaviors following the ROBL indi-

Medullary Raphe and Lordosis

cates that inhibitory influences of these behaviors
coexist in the raphe obscurus nucleus. In this
conjecture, it can be speculated that the raphe
obscurus nucleus operates through these ascending
efferent fibers and affects the neural substrates
which regulate both lordosis and soliciting be-
haviors such as the midbrain central gray. Further
experiments are necessary to detect the kind of
neurons which govern an inhibitory influence on
sexual behaviors in the raphe obscurus nucleus.

ACKNOWLEDGMENTS

The author wishes to express his hearty thanks to
Professor Yasumasa Arai, Juntendo University School of
Medicine, for his valuable suggestions. This study was
supported by a Grant-in-Aid from the Ministry of Educa-
tion, Science and Culture of Japan (63540600) and a
Research Grant from Waseda University (63A—120).

REFERENCES

1 Pfaff, D. W. and Schwartz-Giblin, S. (1988) Cellu-
lar Mechanisms of female reproductive behaviors.
In “The Physiology of Reproduction, Vol. 2.” Ed by
E. Knobil and J. Neill, Raven Press, New York, pp.
1487-1568.

2 Yamanouchi, K., Matsumoto, A. and Arai, Y.
(1985) Neural and hormonal control of lordosis
behavior in the rat. Zool. Sci., 2: 617-627.

3. Edwards, D. A. and Pfeifle, J. K. (1981) Hypotha-
lamic and midbrain control of sexual receptivity in
the female rat. Physiol. Behav., 26: 1061-1067.

4 Sakuma, Y. and Pfaff, D. W. (1980) Mesencephalic
mechanisms for integration of female reproductive
behavior in the rat. Am. J. Physiol. , 237: R285-290.

5 Sakuma, Y. and Pfaff, D. W. (1980) Facilitation of
female reproductive behavior from mesencephalic
central gray in the rat. Am. J. Physiol., 237: R278-
R284.

6 Yamanouchi, K. and Arai, Y. (1982) Dissociation
between the display of lordosis and soliciting be-
haviors in female rats with lesions of the dorsome-
dial pontine tegmentum. Physiol. Behav., 28: 155-
IS),

7 Yamanouchi, K. and Arai, Y. (1983) Forebrain and
lower brainstem participation in facilitatory and
inhibitory regulation of the display of lordosis in
female rats. Physiol. Behav., 30: 155-159.

8 Modianos, D. and Pfaff, D. W. (1976) Brain stem
and cerebellar lesions in female rats. II. lordosis
reflex, Brain Res., 106: 47-56.

9 Modianos, D. and Pfaff, D. W. (1977) Facilitation
of the lordosis reflex in female rats by electrical

10

11

12

13

14

13)

16

7)

18

19

20

74)

jap

23

911

stimulation of the lateral vestibular nucleus. Brain
Res., 134: 333-345.

Modianos, D. and Pfaff, D. W. (1979) Medullary
reticular formation lesions and lordosis reflex in
female rats. Brain Res., 171: 334-338.

Zemlan, F. P., Kow, L-M. and Pfaff, D. W. (1983)
Effect of interruption of bulbospinal pathways on
lordosis, posture, and locomotion. Exp. Neurol., 81:
177-194.

Yamanouchi, K. and Arai, Y. (1985) The role of
mesencephalic tegmentum in regulating female rat
sexual behaviors. Physiol. Behav., 35: 255-259.
Tork, I. (1985) Raphe nuclei and serotonin contain-
ing systems. In “The Rat Nervous System, Vol. 2,
Hindbrain and Spinal Cord”. Ed by G. Paxinos,
Academic Press, New York, pp. 43-78.

Arai, Y. and Yamanouchi, K. (1979) Inhibition of
lordosis by ergocornine in estrogen-progesterone
primed ovariectomized rats. In “Integrative Control
Functions of the Brain, Vol. 2.” Ed. by M. Ito, N.
Tukahara, K. Kubota and K. Yagi, Kodansha,
Tokyo/Elsevier, Amsterdam, pp. 353-355.
Meyerson, B. J. and Lewander, T. (1970) Serotonin
synthesis inhibition and estrous behavior in female
rats. Life Sci., 9: part I: 661-671.

Zemlan, F. P., Ward, I. L., Crowley, W. R. and
Margules, D. L. (1973) Activation of lordotic re-
sponding in female rats by suppression of sero-
tonergic activity. Science, 179: 1010-1011.
Paxinos, G. and Watson, C. (1986) The Rat Brain
in Stereotaxic Coordinates, 2nd ed., Academic
Press, New York.

Anderson, E. G. (1983) The serotonin system of the
spinal cord. In “Handbook of the Spinal Cord, Vol.
1, Pharmacology”. Ed. by R. A. Davidoff, Marcel
Dekker, New York, 1983, pp. 241-274.

Kow, L.-M., Zemlan, F. P. and Pfaff, D. W. (1979)
Attempts to reinstate lordosis reflex in estrogen-
primed spinal female rats with monoamine agonists,
Horm, Behav., 13: 232-240.

Hansen, S. and Ross, S. B. (1983) Role of descend-
ing monoaminergic neurons in the control of sexual
behavior: Effects of intrathecal infusions of 6-
hydroxydopamine and _ 5,7-dihydroxytroptamine.
Brain Res., 268: 285-290.

Foreman, M. M. and Moss, R. L. (1978) Role of
hypothalamic serotonergic receptors in the control
of lordosis behavior in the female rat. Horm. Be-
hav., 10: 97-106.

Luine, V. N., Frankfurt, M., Rainbow, T. C.,
Biegon, A. and Azmitia, E. (1983) Intrahypothala-
mic 5,7-dihydroxytryptamine facilitates feminine
sexual behavior and decrease [*H]-imipramine bind-
ing and 5-HT uptake. Brain Res., 264: 344-348.
Luine, V. N., Renner, K. J., Frankfurt, M. and
Azmitia, E. (1984) Facilitated sexual behavior re-

24

75)

26

27

912

versed and serotonin restored by raphe nuclei trans-
planted into denervated hypothalamus. Science,
226: 1436-1439.

Wiklund, L., Leger, L. and Persson, M: (1981)
monoamine cell distribution in the cat brain stem. A
fluorescence histochemical study with qualification
of indolaminergic and locus coeruleus cell groups. J.
Comp. Neurol., 203: 613-647.

Taber, E., Brodal, A and Walberg, F. (1960) The
raphe nuclei of the brain stem in the cat, I. normal
topography and cytoarchitecture and general discus-
sion. J. Comp. Neurol., 114: 161-188.

Pfaff, D. W. (1980) Estrogens and Brain Function,
Neural Analysis of Hormone-Controlled Mamma-
lian Reproductive Behavior, Springer-Verlag, New
York.

Floody, O. R. and O’Donohue, T. L. (1978) Le-

28

29

30

K. YAMANOUCHI

sions of the mesencephalic central gay depress ultra-
sound production and lordosis by female hamsters.
Physiol. Behav., 24: 499-508.

Muntz, J. A., Rose, J. D. and Shults, R. C. (1980)
Disruption of lordosis by dorsal midbrain lesions in
the golden hamster. Brain Res. Bull., 5: 359-364.
Yamanouchi, K., Nakano, Y. and Arai, Y. (1984)
Mesencephalic central gray as supraspinal neural
substrates for lordosis reflex: Deprivation of sero-
tonergic influence by p-chlorophenylalanine. Zool.
Sci., 1: 126-131.

Hansen, S., Stanfield, E. J. and Everitt, B. J. (1981)
The effects of lesions of lateral tegmental norad-
renergic neurons on components of sexual be-
haviour and pseudopregnancy in female rats.
Neuroscience, 6: 1105-1117.

ZOOLOGICAL SCIENCE 7: 913-921 (1990)

© 1990 Zoological Society of Japan

Apocrine Gland of the Infraorbital Gland of the Japanese
| Serow, Capricornis crispus

YASURO ATOJI and YOSHITAKA SUZUKI

Department of Veterinary Anatomy, Faculty of Agriculture,
Gifu University, Gifu 501-11, Japan

ABSTRACT—The apocrine gland of the infraorbital gland of the Japanese serow, Capricornis crispus,
was investigated by electron microscopy and lectin histochemistry. The infraorbital gland was occupied
largely by the apocrine gland, which consisted of columnar secretory cells and myoepithelial cells. The
secretory cells contained round or ovoid nucleus, well-developed Golgi apparatus, mitochondria and

filaments.

Rough-surfaced endoplasmic reticulum (RER) varied in shape: elongated, round or

fragmentary. There were three types of granules in the cytoplasm: large lucent, large dense and small
dense. Large apical caps were smooth-surfaced and rose from a luminal surface. Apocrine tubules
showed four staining patterns by concanavalin A. Peanut agglutinin revealed very active apocrine
secretion of secretory cells. Myoepithelial cells contained myofilaments, glycogens and mitochondria.
These data are available for a comparison with those of wild Artiodactyla.

INTRODUCTION

Many even-toe ungulates have various special-
ized skin organs: the interramal gland, infraorbital
(preorbital or antorbital) gland, caudal gland,
metatarsal gland, tarsal gland, interdigital (pedal)
gland or inguinal gland are examples [1]. Some are
thought to serve as a scent organ for communica-
tion, especially, in reproductive behaviour via an
olfactory tract among animals [2]. The specialized
skin gland generally consists of the apocrine gland
and/or sebaceous gland, and its volume in relation
to the whole glandular tissues greatly varies among
the species [3, 4]. Morphological evidence of skin
glands of wild ungulates to date is almost limited to
light microscopic observation owing to the difficul-
ty of obtaining sufficient materials. Richter [5]
stressed that the ultrastructure of specialized skin
glands is very available for classification of organs
and analysis of the secretion mechanism of secre-
tory cells.

The Japanese serow, Capricornis crispus, is an
endemic ungulate and has three kinds of special-
ized skin glands: the infraorbital gland, interdigital

Accepted March 22, 1990
Received December 19, 1989

gland and preputial gland [6-9]. The animal
frequently rubs secretions of the infraorbital gland
on twigs or tree trunks as scent marking related to
territory [10]. The infraorbital gland is a very
well-developed cutaneous organ, and secretions
from a secretory pore are a sticky fluid [11]. Each
infraorbital gland is composed of a thin inner
sebaceous gland and a thick outer apocrine gland.
The sebaceous gland shows sex dimorphism and its
lipogenesis is revealed by electron microscopy
[12]. On the other hand, the apocrine gland
exudes glycoconjugate secretions by two secretory
modes of apocrine secretion and exocytosis [13]. It
is also reported that secretory cells of the apocrine
gland show a close functional relationship between
secretion and the Golgi apparatus by lectin his-
tochemistry [13]. However, the fine structure of
the apocrine gland, which would help to under-
stand cellular function, has not fully elucidated.
The present study deals with the histology of the
apocrine gland of the infraorbital gland of the
Japanese serow by electron microscopy and lectin
histochemistry.

MATERIALS AND METHODS

A number of Japanese serows were sacrificed in

914 Y. ATOJI AND Y. SUZUKI

each winter from 1979 to 1985 as part of studies on
the conservation and control of the Japanese serow
by the Agency of Culture Affairs, Japan. Thirty
adult Japanese serows (15 males and 15 females)
were selected for this study. All the specimens
were fresh or well preserved until dissection. After
gross dissection, the infraorbital glands were
photographed, removed, and fixed in 10% forma-
lin for light microscopic observation. Paraffin
sections of 5 um thick were stained with hematoxy-
lin and eosin, azan and lectins.

Lectin histochemistry was conducted as follows.
Arachis hypogaea agglutinin (PNA) and Canavalia
ensiformis agglutinin (Con A) were obtained from
Vector Laboratories (USA). Paraffin sections
were deparaffinized and endogenous peroxidase
was treated by 0.3% HO; prior to lectin staining.
Sections were pre-incubated with 1% normal goat
serum and then incubated with biotinylated lectins
for 2 hr (Con A) or overnight (PNA) at 4°C. The
optimal concentration of lectins was 10 g/ml (Con
A) or 25 ug/ml (PNA). Sections were rinsed in
0.01 M phosphate-buffered saline (PBS) and incu-
bated with avidin-biotin-peroxidase complex
(ABC) (Vector Laboratories, USA) for 30 min at
room temperature. After rinsing in PBS, the
reaction product was visualized by incubation with
PBS containing 3, 3’-diaminobenzidine (25 mg/100
ml) and 0.003% H,O, for 5-10 min, then washed
in tap water and mounted. Controls were as
follows: 1) oxidation with 1% periodic acid for 10
min before lectin staining, 2) substitution of un-
labelled lectins for biotinylated lectins and 3) in-
cubation with each lectin containing 0.1M D-
galactose (PNA) or 0.1 M a-methyl-D-mannoside
(Con A).

For electron microscopy, the infraorbital gland
of both sexes was fixed in 2.5% glutaraldehyde in
0.1M phosphate buffer at pH 7.2 (4°C). After
rinsing in phosphate buffer, small blocks were
postfixed in 1% osmium tetroxide in 0.1 M phos-
phate buffer at pH7.2 (4°C), dehydrated with
graded ehanol series and embedded in epoxy resin.
Ultrathin sections were cut by a Porter-Blum II
ultramicrotome, stained with uranyl acetate and
lead citrate and examined with a Hitachi HU-12
electron microscope.

RESULTS

The infraorbital glands were located in large
depressions of the lacrimal bone, maxilla and
zygomatic bone, and a secretory pore (2-3 mm in
diameter) opened on the surface of the hairy skin
in each side. In longitudinal sections, the gland has
a dense hair pouch which extended to the pore at
the dorsal surface (Fig. 1). The pouch was usually
filled with clear or white secretions. The glandular
tissues were made up of a thin and yellowish white
internal layer and a thick and brown external
layer. The internal sebaceous layer showed
apparently sexual dimorphism: the female’s
sebaceous gland consisting of the ordinary and
modified types was evidently wider than the
male’s. The external layer was composed of the
coiled apocrine gland and connective tissue. Both
sebaceous and apocrine glands opened into hair
follicles of the pouch. Apocrine tubules varied in
size (70-230 um in diameter) and consisted of a
single secretory cell layer and a spindle-shaped
myoepithelial cell layer (Fig. 2). Secretory cells of
the apocrine gland exhibited active apocrine secre-
tion in hematoxylin and eosin preparations. Lectin
staining using PNA further provided more appa-
rent feature of apocrine secretion (Fig. 3). In Con
A sections, apocrine tubules showed four staining
patterns. First, all secretory cells were strongly
stained (Fig. 4a). Nuclear envelopes were also
reactive. Second, some cells lost staining intensity
of the whole cytoplasm (Fig. 4b). Third, almost no
secretory cells and nuclear envelopes were stained
(Fig. 4c). Fourth, the basal half of the cytoplasm
and nuclear envelopes became positive (Fig. 4d).
Myopithelial cells did not stain with Con A.

On the electron microscopic observation, tall
secretory cells had oval nuclei which were usually
located in the basal half of cells, sometimes in the
apical third (Fig. 5). Rough-surfaced endoplasmic
reticulum (RER) was distributed throughout the
cytoplasm and varied markedly in shape: elon-
gated, circular or fragmental profiles (Figs. 6, 8, 9,
11). The Golgi apparatus was situated in the
supranuclear region of the cytoplasm and display-
ed variable forms with flattened saccules, vesicles
or mixed type (Figs. 6, 9). Many mitochondria and
filament bundles were distributed in the cytoplasm

SINS)

Infraorbital Gland of the Japanese Serow

he infraorbital gland. Secretions are

int

f apocrine glands

jor component o

hows a ma

ion s
<3

1 sect

Longitudina

Fic. 1.

h. Azan. :

ine tu

seen in a pouc

x 370

le columnar epithelium. HE.

ined with a sing

]

1S
ine secretion

bule

Apocr

Fig. 2:

x 180.

bserved by PNA staining.

1S O

Active apocr

Fic: 3.

916 Y. Ato! AND Y. SUZUKI

Fic. 4. Four different staining patterns of tubules with Con A. Secretions in a lumen (c, d) are virtually
unstained. 280. . |

Fic. 5. Apocrine tubule consists of tall secretory cells and flat myoepithelial cells. Secretory cells contain
mitochondria, stacks of elongated RER, granules and the Golgi apparatus. 3,200.

Fic. 6. Luminal surface of a secretory cell is covered with microvilli and there are large dense granules (D) and large
lucent granules (L) in the cytoplasm. Arrows indicate tight junctions. 6,900.

Fic. 7. Apical caps of secretory cells are large and their surface is smooth. Microvilli are found at lateral walls below
apical caps. Arrow: tight junction. 8,700.

Fic. 8. Myoepithelial cells are irregular-shaped and contain abundant myofilaments and glycogen particles. 6,500.

Infraorbital Gland of the Japanese Serow

918 Y. ATOJI AND Y. SUZUKI

Fic.9. Apical cytoplasm contains well-developed Golgi lamellae, round RER and large lucent granules. x 12,500.

Fic. 10. A row of small dense granules (thin arrows) is seen along the lateral wall of a secretory cell. D: large dense
granule. L: large lucent granule. Thick arrows:filament bundle. 12,500.

Fic. 11. Note two different shapes of RER at the base of neighbouring secretory cells. 11,400.

(Figs. 5, 10). Three kinds of granules were de-
tected in the cytoplasm. The first was large lucent
granules found in the Golgi region or in the apical
portion (Figs. 6, 9). They measured 500-1,400 nm

in diameter and frequently contained dense debris.
The second was large electron dense granules
measuring 800-—1,600 nm in diameter (Figs. 6, 10).
The third was small granules containing dense

Infraorbital Gland of the Japanese Serow 919

materials (150-200 nm in diameter) (Fig. 10). This
type was subject to crowding and was distributed
near the basal or lateral cell membrane. Neigh-
bouring columnar secretory cells were bound by
the junctional complex consistent with terminal
bars at a luminal surface (Fig. 6) and by desmo-
somes and interdigital foldings of cell membranes
at the lateral walls (Figs. 9, 11). Luminal surface
was covered densely by microvilli (Fig. 6) or
formed an apical protrusion for apocrine secretion
(Fig. 7). Apical caps, which contained numerous
free ribosomes, round RER or debris in the floccu-
lent matrix, were as large as the cell’s own dia-
meter and their luminal surface was smooth. In
this case, microvilli were observed on lateral walls
between apical caps and _ thight junction.
Myoepithelial cells lay between secretory cells and
basal lamina, and contained irregular-shaped nuc-
lei located in a central or upper region (Fig. 8).
Numerous myofilaments, mitochondria and gly-
cogen particles were found in the cytoplasm.

DISCUSSION

The present study showed that the apocrine
gland occupied a huge area in the infraorbital
gland of the Japanese serow. Dik-diks (Madoqua
phillipsi and M. guentheri), duikers (Cephalophus
monticola, C. ogilbyi, C. silvicultor, C. natalensis
and Sylvicapra grimmia) or antelope (Neotragus
pygmaeus) have a well-developed infraorbital
gland in which the sebaceous gland, located in the
central region, occupies a large area [5, 14-16]. In
deer species, the infraorbital gland is usually small,
rudimentary, or absent [17-19]. The Japanese
serow always secretes a clear and thick fluid from
the infraorbital gland. Yokohata et al. [20] found
that the apocrine region contained hydrocarbons,
glycerids, phospholipids and sterol esters. Atoji et
al. [13] reported two secretory mechanisms of
exocytosis and apocrine secretion of secretory cells
and glycoconjugate secretions containing galac-
tose, mannose, N-acetyl-galactosamine and N-
acetyl-glucosamain in apocrine tubules. Active
apocrine secretion is also confirmed in the present
study. These evidences probably indicate that
most secretions of the infraorbital gland probably
originates from the apocrine portion.

RER of apocrine secretory cells is a round or
short cistern in the human axillary [21] and cerumi-
nous [22-24] glands or has a long, flattened shape
in the anal scent gland of the woodchuck, Marmota
monax [25]. The present study clearly demons-
trated various RER shapes in secretory cells of the
infraorbital gland in the Japanese serow; fragmen-
tal, circular or elongated forms. In general, newly
synthesized glycoproteins in RER are transported
to the Golgi apparatus and then elaborated into
secretory gaunules, plasma membrane or serum
[26, 27]. The previous study showed a direct
involvement of the well-developed Golgi appar-
atus to the secretory modes of apocrine secreion
and exocytosis in the apocrine gland of the infraor-
bital gland [13]. Therefore, the variety of RER in
the present study would reflect indirectly the secre-
tory activity of the apocrine gland. This idea is
further supported by a variable staining of apoc-
rine tubules with Con A which has been shown a
marker lectin for RER in many cell types [28-30].
Con A negative stage in secretory cells (Fig. 4c)
might be consistent with those which contain abun-
dant fragmentary RER or free ribosomes in the
cytoplasm.

Apocrine secretory cells of the infraorbital gland
contained large lucent granules, large dense gra-
nules, and small dense granules in the cytoplasm.
Large lucent granules of the apocrine gland are
released into a lumen through microvilli-rich
luminal surface by exocytosis [13]. This granule is
similar in appearance to the lemur antebrachial
organ [31]. Secretory granules for exocytosis are
large dense in the woodchuck anal gland [25] and
small dense in the human apocrine gland [21, 22].
In the present study, release of large dense gra-
nules was not observed. Small dense granules of
the infraobital gland were located near the lateral
or basal membrane. Smith and Hearn [25]
observed membrane-bound small dense granules
at the basal cytoplasm of the woodchuck. In the
canine apocrine gland, small dense granules were
found in cistern of RER [32]. It is, therefore,
speculated that two dense granules of the infraor-
bital gland are derived from RER.

The luminal surface of secretory cells is usually
covered with numerous microvilli when apocrine
secretion does not occur [21, 25]. Surface of apical

920 Y. ATON AND Y. SUZUKI

caps has been reported to be smooth in the apoc-
rine gland of the human [23], woodchuck [25] and
Japanese serow [13]. In the present electron
microscopic observation, apical caps containing a
few cell organelles are large and project into the
lumen as if thrusting microvilli aside on the lateral
walls. In human apocrine gland, they are smaller
in size and contain small vesicles [21-23]. The
apical caps are formed by the bradual swelling of
glycoconjugate-rich apical cytoplasm from the
luminal surface with microvilli [13].

Myoepithelial cells of the apocrine gland contact
firmly secretory cells by desmosomes of foldings of
the plasma membrane [32]. Atoji et al. [13]
reported a relationship of soybean agglutinin
(SBA) binding between myoepithelial cells and
secretory cells. SBA always binds myoepithelial
cells, except that only the Golgi apparatus of
secretory cells reacts strongly with SBA. This
might indicate a prolonged involvement of
myoepithelial cells in secretory cells in addition to
contraction by a adrenalin [33].

Secretions from the infraorbital gland of the
Japanese serow is a mixture derived from the
apocrine gland and the sebaceous gland. The
histological composition of the sebaceous gland
shows sex dimorphism; the male’s gland only con-
sists of the ordinary type and the female’s one has
the ordinary and modified types [6]. The modified
sebaceous gland is much larger in size than the
ordinary type and morphological changes of
lipogenesis gradually progresses in large acini [12].
That is, the female’s secretions seem to contain
extra lipid components compared with the male.
In fact, evidence obtained from lipid analysis of
the infraorbital gland supports this suggestion [20].
Therefore, the difference of contents in secretions
would be useful for discriminating the sex, when
the animal sinffs secretions rubbed on objects.

Another sexual difference in the adult Japanese
serow is the weight of the infraorbital gland; the
male gland is significantly heavier than the female
[6]. Few data on sexual difference of the infraor-
bital gland in other Artiodactyla are available,
because the gland is very difficult to obtain. Mos-
sing and Kallquist [19] reported that the male’s
infraorbital gland of the reindeer Rangifer taran-
dus is larger than the female and is most developed

during the rutting season. Further studies on wild
ruminants will be expected.

REFERENCES

1 Gosling, L. M. (1985) The even-toed ungulated:
order Artiodactyla. In “Social Odours in Mam-
mals”. Ed by R. E. Brown and D. W. MacDonald,
Clarendon Press, Oxford, pp. 550-618.

2 Izard, M. K. (1983) Pheromones and reproduction
in domestic animals. In “Pheromones and Repro-
duction in Mammals”. Ed. by J. G. Vandenbergh,
Academic Press, new York, pp. 253-285.

3 Sokolov, V. E. (1982) Comparative morphology of
skin of different orders. In “Mammal Skin”. Ed. by
University of California Press, Berkeley, pp. 43-
S12.

4 Flood, P. (1985) Sources of significant smells: the
skin and other organs. In “Social Odours in Mam-
mals”. Ed. by R. E. Brown and D. W. MacDonald,
Clarendon Press, Oxford, pp. 19-36.

5 Richter, J. (1973) Zur Kenntnis der Antorbital-
driisen der Cephalophinae (Bovidae, Mammalia).
Z. Saugetierkunde, 38: 303-313.

6 Kodera, S., Suzuki, Y. and Sugimura, M. (1982)
Postnatal development and histology of the infraor-
bital glands in the Japanese serow, Capricornis
crispus. Jpn. J. Vet. Sci., 44: 839-843.

7 Uno, K., Sugimura, M., Susuki, Y. and Atoji, Y.
(1984) Morphological study on vagina, vestibule
and external genitalia of Japanese serows. Res. Bull.
Fac. Agr. Gifu Univ., 49: 183-195.

8 Atoji, Y., Suzuki, Y. and Sugimura, M. (1989) The
preputial gland of the Japanese serow Capricornis
crispus: ultrastructure and lectin histochemistry.
Acta Anat., 134: 245-252.

9 Atoji, Y., Suzuki, Y. and Sugimura, M.. (1988)
Lectin histochemistry of the interdigital gland in the
Japanese serow (Capricornis crispus) in winter. J.
Anat., 161: 159-170.

10 Masui, M. (1987) Social behaviour of Japanese
serow, Capricornis crispus crispus. In “The Biology
and Management of Capricornis and related Moun-
tain Antelopes”. Ed. by H. Soma, Crom Helm,
London, pp. 134-144.

11 Kanomata, K. and Izawa, M. (1982) Some observa-
tions on estrus in female Japanese serows, Capricor-
nis crispus. J. Jpn. Assoc. Zool. Grad. Aquar., 24:
61-63.

12 Atoji, Y., Sugimura, M. and Suzuki, Y. (1989)
Sebaceous glands of the infraorbital gland of the
Japanese serow Capricornis crispus. J. Submicrosc.
Cytol. Pathol., 21: 375-383.

13 Atoji, Y., Hori, Y., Sugimura, M. and Suzuki, Y.
(1987) Lectin histochemical study on the infraorbit-

14

15

16

LW)

18

19

20

21

22

23

24

Infraorbital Gland of the Japanese Serow

al gland of the Japanese serow (Capricornis crispus).
Acta Morphol. Neerl.-Scand., 25: 201-213.
Richter, J. (1971) Untersuchungen an Antorbital-
driisen von Madoqua (Bovidae, Mammalia). Z.
Saugetierkunde, 36: 334-342.

Kuhn, H. J. (1976) Antorbitaldriise und
Tranennasengang von WNeotragus pygmaeus. Z.
Saugetierkunde, 41: 369-380.

Mainoya, J. R. (1978) Histological aspects of the
preorbital and interdigital glands of the red duiker
(Cephalophus natalensis). E. Afr. Wildl. J., 16: 265-
Dia

Quay, W. B. (1955) Histology and cytochemistry of
skin gland areas in the caribou, Rangifer. J. Mamm.,
36: 187-201.

Quay, W. B. and Muller-Schwarze, D. (1970) Func-
tional histology of integumentary glandular regions
in black-tailed deer (Odocoileus hemionus colim-
bianus). J. Mamm., 51: 675-694.

Mossing, T. and Kallquist, L. (1981) Variation in
cutaneous glandular structures in reindeer (Rangifer
tarandus). J. Mamm., 62: 606-612.

Yokohata, Y., Sugimura, M., Suzuki, Y., Naka-
mura, T. and Atoji, Y. (1985) Histology and lipid
analysis of sebaceous portion of infraorbital gland in
female Japanese serows. Res. Bull. Fac. Agr. Gifu
Univ., 50: 185-191.

Kurosumi, K. and Kurosumi, U. (1982) Further
studies on the ultrastructure of the human axillary
apocrine sweat glands. Okajima Folia Anat. Jpn.,
58: 305-324.

Kurosumi, K. and Kawabata, I. (1976) Transmis-
sion and scanning electron microscopy of the human
ceruminous apocrine gland. I. Secretory glandular
cells. Arch. Histol. Jpn., 39: 207-229.

Testa-Riva, F. and Puxeddu, P. (1980) Secretory
mechanisms of human ceruminous glands: a trans-
mission and scanning electron microscopical study.
Anat. Rec., 196: 363-372.

Shugyo, Y., Sudo, N., Kanai, K., Yamashita, T.,

25

26

27

28

29

30

3)

32

38)

921

Kumazawa, T. and Kanamura, S. (1988) Morpholo-
gical differences between secretory cells of wet and
dry types of human ceruminous glands. Am. J.
Anat., 181: 377-384.

Smith, J. D. and Hearn, G. W. (1979) Ultrastruc-
ture of the apocrine-sebaceous anal scant gland of
the woodchuck, Marmota monax: evidence for
apocrine and merocrine secretion by a single cell
type. Anat. Rec., 193: 269-292.

Fleischer, B. A. (1983) Mechanism of glycosylation
in the Golgi apparatus. J. Histochem. Cytochem.,
31: 1033-1040.

Griffiths, G. and Simons, K. (1986) The trans Golgi
network: sorting at the exit site of the Golgi com-
plex. Science, 234: 438-443.

Pinto da Silva, P., Torrisi, M. R. and Kachar, B.
(1981) Freeze-fracture cytochemistry: localization
of wheat-germ agglutinin and concanavalin A bind-
ing sites on freeze-fractured pancreatic cells. J. Cell
Biol., 91: 361-372.

Atoji, Y., Hori, Y., Suzuki, Y. and Sugimura, M.
(1988) Concanavalin A binding sites of rough endo-
plasmic reticulum containing intracisternal microtu-
bules in canine neurons. J. Comp. Pathol., 99: 55-
63.

Roth, J. (1983) Applicaiton of lectin-gold com-
plexes for electron microscopic localization of glyco-
conjugates on thin sections. J. Histochem.
Cytochem., 31: 987-999.

Kneeland, J. F. (1966) Fine structure of the sweat
glands of the antebrachial organ of Lemur catta. Z.
Zellforsch., 73: 521-533.

Kurosumi, K., Shibasaki, S. and Ito, T. (1984)
Cytology of the secretion in mammalian sweat
glands. Int. Rev. Cytol., 87: 253-329.

Sato, K. (1980) Pharmacological responsiveness of
the myoepithelium of the isolated human axillary
apocrine sweat gland. Brit. J. Dermatol., 103: 235-
243.

| Simles: scary). re a AR Uber dele

- verlgank aller sb aleve: oe pied ‘wae ae aS
TU aie baat autre ees it LB Ue -

Tide 2gnEoneT bas ioe uma poder
vy ed Eigen fotesrbaadt 10" satire

cil 1 ne Nw

RF i

.. =

Sahn toe ¥ ath, i i 4

fo oe bit ry hen cube.
= epbnadgnor yeahs Pe Meh; Sieh eight ieee tii

a chess spite greta uke a ; )

ange er Smeg. | Al esl: :

pi a Rea (fRet ae GT
tejtverstsionc et sleet Bi
: Bans sae QIS

paki, Peis

ole nati ea

re

> ay ‘
aa ;
vos i

=
= oa
tad {
4
si Uo
7 a
¢
} Re :
= \
J } i
Fs
u s
ye)
t
é € 4
’
l
é iad Ss
} 7 » a
12
y
e vi } Ks
Ne
i
om “a

if ae its

a AMS sah

ope mala

Cin fein Hoh
a hee oe

Pal

‘ .

a
/ ; a
DCR

ety ‘2

peers ek a i se :
Vedat
sienaieen Es
ce iff at
sihecut a ‘
TAPES Eh
ae
br ks
ay PH) i‘ Tite a, -saptiseR
t
} 1 >: é
:
~% 4
a y
B.
we e)
we : .
¥
: ae

ZOOLOGICAL SCIENCE 7: 923-931 (1990)

© 1990 Zoological Society of Japan

Structural and Carbohydrate Histochemical Aspects of the
Snout Skin of the Opossum, Didelphis virginiana Kerr

WILFRIED MEYER and AZUMA TSUKISE

Institut fiir Zoologie, Tieraérztliche Hochschule Hannover, Biinteweg 17, D-3000
Hannover 71, F.R. Germany, and Department of Veterinary Anatomy,
College of Agriculture and Veterinary Medicine,

Nihon University, Fujisawa 252, Japan

ABSTRACT—This study describes the general structure of the snout skin in the common American
opossum (Didelphis virginiana Kerr). Additionally, carbohydrate histochemical methods, including
lectin histochemistry, have been used to determine functional characteristics of the snout, in particular
as related to the epidermis. As an unique feature of the opossum snout, strongly reacting glycoconju-
gates (neutral glycoproteins) were present in the intercellular substances among the flat cells of the
stratum corneum. These glycoproteins exhibited the following spectrum of terminal saccharide
residues: $-N-acetyl-D-glucosamine, N-acetyl-a-D-galactosamine, a-D-galactose, a-L-fucose, ~-D-
galactose-D-N-acetyl-galactosamine. The specific functions of such substances in the corneal layer of

the opossum snout skin are discussed.

INTRODUCTION

The mammalian snout has undergone typical
morphological and functional adaptations which
have led to manifold functional properties. In this
connection, the snout is a delicate “sense organ”
because its skin is studded with numerous sensoric
structures, such as free nerve endings or sensory
end organs, including sinus hair follicles [1-4].
Additionally, the snout may be involved in olfacto-
ry or even digestive processes as connected with
specialized nasolabial glands [5-8].

However, knowledge on the structure and func-
tion of the integument is still relatively scarce
where marsupials, and, in particular, the Didelphi-
dae are concerned [9, 10]. Nevertheless, some
information is available on specific body regions of
the latter group, such as the pouch [9], the snout
shape [11] and innervation [12], the foot pads [13,
14], and the scrotum [15]. The present study
supplies a first general description of snout skin
architecture in the opossum, but focusses on car-
bohydrate histochemical features as, probably, re-

Accepted March 26, 1990
Received December 6, 1989

lated to specific snout functions in this marsupial
species.

MATERIALS AND METHODS

Six adult common American opossums (Didel-
phis virginiana Kerr) (40%, 2°) were ex-
amined in the present study. After the animals
were sacrificed, snout skin samples were dissected
out and fixed in 10% formalin in 95% ethanol,
10% formalin containing 2% calcium acetate [16],
and Bouin’s fluid for 48 hr 4°C or room tempera-
ture. The tissue specimens were then carefully
dehydrated in a graded series of ethanol, embed-
ded in paraffin wax and cut at a thickness of 8 um.
For staining purposes sections were deparaffinized
in xylene, hydrated through descending concentra-
tions of ethanol and stained with various pro-
cedures for the histochemical study of glycoconju-
gates including combined control procedures.

The histological staining of choice was haema-
toxylin (after Delafield) and eosin (H.E.), addi-
tionally some sections were stained according to
Humason and Lushbaugh [17]. For the demon-
stration of glycoconjugates the following histo-
chemical staining were emplyed: Periodic acid

924 W. MEYER AND A. TSUKISE

Schiff (PAS) ([{18]; Schiff’s reagent after Barger
and De Lamater [19]); alcian blue (AB pH 1.0)
[20]; AB pH2.5 [21]; dialysed iron-ferrocyanide
(DI-FCY) [22]; AB pH 2.5 -PAS [23], and diffe-
rent lectins labelled with horseradish peroxidase
(PO) (purchased from E.Y. Labs./Medac) in con-
centrations of 10-20 «g/ml in 0.1 M PBS (pH 7.2)
for 30 min at 20°C and 2 hr at 4°C.

The lectins used were the following-concanava-
lin A (Con A), peanut agglutinin (PNA), Dolichos
biflorus agglutinin (DBA), Ricinus communis
ageglutinin-I (RCA-I), tinin-I (UEA-I), soy bean
agglutinin (SBA), wheat germ agglutinin (WGA),
and Limulus polyphemus agglutinin (LPA) (for
residue demonstration see [24-26]). The activity
of the peroxidase employed for labelling was re-
vealed by a diaminobenzidine-hydrogen peroxide
system (DAB, purchased from Sigma) [27].

For lectin controls, the following saccharides
were added at a final concentration of 0.01 M to
the respective lectin solutions: a-methyl-D-
mannoside for Con A, lactose for PNA, N-acetyl-
D-galactosamine for SBA and DBA, £-D-
galactose for RCA-I, a-D-galactose for MPA, L-
fucose for UEA-I, n-acetyl-D-glucosamine for
WGA, and N-acetyl-neuraminic acid for LPA;
additionally unconjugated lectins were substituted
for lectin-PO conjugates; and sections were ex-
posed to the PO and DAB system without lectin
preincubation. To detect endogenous peroxidase
activity in tissues, certain control sections were
reacted with DAB only.

Chemical modification by sulfation [28] was per-
formed on some sections, prior to staining with AB
pH 1.0. Enzyme digestion with e-amylase (from
Bacillus subtilis, Sigma; 1 mg/ml in 0.1 M_ phos-
phate buffer (pH 7.0) at 37°C for 3 hr) was con-

ducted on some sections, prior to PAS staining.
For this enzyme digestion experiment, two types of
controls were done: some sections were incubated
in buffer solutions without the enzyme under iden-
tical conditions of temperature and duration, and
other sections were kept intact without any incuba-
tion procedures [5]. |

RESULTS

The snout of the opossum is typically long and
pointed. Frontally one right and one left upper lip
can be identified, as separated by a nasal field, i.e.
the rhinarium and the upper lips have fused cen-
trally to form a planum nasolabiale. This structure
is divided medially by a more or less deep cleft
leaving the nostrils laterally, each with a clear
dorsal sulcus alaris (Fig. 1). The nose and the
planum nasolabiale of the opossum are unpig-
mented, glabrous and relatively smooth regions,
although they show a fine surface pattern of dis-
crete penta- to octogonal elevations with a dia-
meter of approximately 200-800 4m (rate pegs;
Fig. 2).

Light microscopically, the skin of the nose and
the planum nasolabiale is rather thick (about 300
yam in the pegs), with a distinct undersculpture of
interdigitating epidermal pegs and ridges and der-
mal papillae (Figs. 3, 4). The thickest epidermis
shows a stratum basale of cuboidal cells, followed
by 18-20 layers of roundish and polyhedral spi-
nous cells, and 5-6 layers of elongated and spin-
dle-shaped cells of the stratum granulosum. The
stratum corneum consists of 10-15 layers of
strongly flattened cells and has an overall thickness
of only 30-40 um (Fig. 4). Among the cells of the
stratum basale of the thicker epidermal ridges

Fic. 1.

Shape of the opossum snout (A-frontal, B-ventral, C-lateral view (after Barge [11] and original).

Histochemistry of Opossum Snout Skin 925

Fic. 2. Surface view of the opossum snout (x 120).

TABLE 1. Carbohydrate histochemical reactions in the skin layers of the opossum snout

Reactions Epidermis Dermis Hypodermis
Stratum Stratum Stratum Stratum Connective Blood Connective
corneum granulosum spinosum basale tissue vessels tissue

PAS 4* 2-3? 0-1 0 3-4 3 2

AB (pH 1.0) 0 il 0 0 2-3 1-2 2

AB (pH 2.5) Oh 1-2? 0 0 D. 1-2 2

DI-FCY 1-2 1-2 2 2 1-2 D 1-2

AB (pH 2.5)-PAS 4_S* 2-2? 0-1? 0 3-4 3 2-3

Sul-AB (pH 1.0) 1 1-2 0 0 3 2 2-3

Amy-PAS 4* 2-3° 0-1? 0 3-4 ) 2

PO-Con A-DAB 2-3 3-4 2-3 2-3 2-3 2-3 2-3

PO-RCA-I-DAB 0-1 2-3 2 1-2 2 2 D

PO-PNA-DAB 1-3? 2-4 2-3 D 2. 1-2 2

PO-SBA-DAB 2-4? 4* 2-4° 2-3? 3 1-2 2-3

PO-MPA-DAB 3-4? 4° 2-4° 2 2-3 2-3 2-3

PO-DBA-DAB 0-1 2 1-2 1 2-3 1-2 2-3

PO-UEA-I-DAB 1-2? 4* 3-4? 0 2-3 1-2 2-3

PO-WGA-DAB 1-3? 3-4° 2-3 2 2-3 1-2 2-3

PO-LPA-DAB Me 2-3 2 1-2 2 Dy 2

Reaction intensities: 0=no reaction, 1=very weak, 2=weak, 3=moderate, 4=strong, 5=very strong a:
intercellular substance.

926 W. MEYER AND A. TSUKISE

Fic. 3.

General structure of the opossum snout (semi-schematic sagittal section near mid-line; E-epidermis,

D-dermis, CT-connective tissue sheath of nasal conchae, NC-cartilage of nasal conchae, P-palatinal
bone, JO-Jacobson’s organ, M-lip musclature; blood vessels near nasal concha and in dermis black).

often Merkel cells can be found, sometimes form-
ing a single-layered row along the deepest part of
the ridge (Fig. 5). In general, the cells of the
granular layer show only very few cytoplasmic
keratohyalin granules throughout the whole snout
skin. This is in contrast to this epidermal cells type
in the hairy skin of the opossum.

The dermis of the opossum snout lacks any
glands. It exhibits a loosely structured stratum
papillare with a network of thinner but irregularly
crossing collagen fibre bundles and intermingled

elastic fibres, and numerous blood capillaries
which reach the top of the dermal papillae (Figs. 3,
6). Obviously these capillaries are accompanied by
fine nerve fibres. The broad stratum reticulare
with numerous densely interwoven collagen fibre
bundles is characterized by thicker blood vessels
which ascend from a deep dermal vascular net-
work, marking the transitional zone of this skin
layer with the rigid connective tissue sheath of the
nasal conchae or their cartilage, respectively (Fig.
7). The dermis of the opossum also contains

O27,

Histochemistry of Opossum Snout Skin

?
>)

—~
we
5 a
Pry WY
Gy 4s
ae
aA
}
A
canes
ca
ie
ww
na =
3 &
25
a &
6
2°
OE
@) 5}
= a
Oo
sO
7)
ae
=I
ae
DN
+ ©
al}
n &
oe
a. ob
25
nS
t=} 35)
ay
3 (6)
~ YD
S| on
5 {I
ans
n 2
cs
3 6
oO §
5
5
—
a
oN)
Ao
E's
oO 8
~m™ O
are
Ms
—
3
ty
~
“ft
~Q
n

Fics. 4, 5.

x 890).

, the dermis incorporates several bun-

dles of muscle fibres. At the most medial part of
the transitional region of the nasal septum with the

nasolabiale

h are concentrated along

1C

several mast cells, wh

larger blood vessels. In the basal part of the upper
lip and the intersected nasal field of the planum

928 W. MEYER AND A. TSUKISE

ie

26 BA

?

se tYemey
Ae ; ma

ae 4 ae
nop, aH 2
oe. i
os
ee .

ge Ae

palatinum, these muscle fibres closely neighbour to
large blood vessels surrounding Jacobson’s organ
(Fig. 3).

The carbohydrate histochemical results are
summarized in Table 1 and documented in Figures
8-12. It is evident that the reactions for PAS, AB
pH1.0 and 2.5, DI-FCY and AB pH2.5-PAS
were generally weak to moderate in the cytoplasm
of the epidermal cells and in the connective tissue
of dermis and hypodermis. In contrast to these
observations, especially prominent, i.e. strong to
very strong reaction staining was restricted to the
intercellular substances between the flat cells of
the stratum corneum (PAS, AB pH 2.5-PAS; Fig.
8). Digestion with a-amylase did not diminish the
PAS staining intensity of all the epidermal strata.
Sulfation increased the intensity of the AB pH 1.0
reaction in the stratum cornerm and stratum
granulosum, and in the connective tissue of dermis
and hypodermis. AB pH 2.5-PAS, AB pH 1.0 and
DI-FCY staining made it possible to differentiate

WET HN S PO
7 “5 i DAV toutes ~/

att

aw A = . * % 3 a :
+x 4 vi wR eS ie ;
nt

Fics. 6,7. Structure of stratum papillare (6) and dense network of connective tissue around nasal conchae (7) in the
opossum (silver impregnation, phase contrast, E-epidermis; 6 and 7- x 350).

several mast cells in these layers.

All epidermal layers reacted positively with PO-
labelled lectins (Figs. 9-12). Strong colouring was
clearly confined to intercellular substances, par-
ticularly between the cells of the stratum corneum,
the stratum granulosum and the upper layers of the
stratum spinosum (PNA, SBA, MPA, UEA-I,
WGA). A cyoplasmic tinging for PO-lectins was
achieved in the cells of the stratum granulosum
and the upper and medium layers of the stratum
spinosum (Con A, RCA-I, PNA, DBA, UEA-I,
MPA, WGA, LPA). However, this intracellular
epidermal staining pattern was somewhat weaker
than in the intercellular epidermal material. In the
dermis and hypodermis, connective tissue fibres
and fibre bundles, respectively, exhibited weakly
to moderately positive reactions. Only the endo-
thelial cells of dermal blood vessels sometimes
stained more intensely (Con A, MPA).

Histochemistry of Opossum Snout Skin 929

12-PO-UEA-I; 8-640, 9-12- x 580).

DISCUSSION

The overall structure of the snout, including the
planum nasolabiale of the opossum (Didelphis
virginiana Kerr) to some extent compares to that

> *
>.

of carnivorous mammals. This comprises the lack
of tubular glands as well as the relatively thin
stratum corneum, or the sensoric equipment [1, 4,
12, 29]. The fine sculpturing of the snout surface,
however, closely resembles to that found in the

930

bovine muzzle [30].

Carbohydrate histochemistry revealed very high
amounts of neutral glycoproteins as present in the
intercellular substances among the corneal cells.
Although such substances can, for example, also
be found among the cells of the stratum corneum
of the porcine planum rostrale or the bovine
muzzle [5, 8], the opossum here shows a rather
unique feature, because in the other species, this
material could have been derived from the secre-
tions of eccrine and apocrine glands. In the
opossum such glycoproteins may be produced by
the epidermal corneal cell itselves, exhibiting
several saccharide residues, as indicated by the
PO-lectin staining procedures: $-N-acetyl-D-
glucosamine, N-acetyl-a-D-galactosamine, a-D-
galactose, a-L-fucose, $-D-galactose-D-N-acetyl-
galactosamine. Such a pattern of terminal sugars
differs somewhat from that found in the stratum
corneum of the porcine snout and the bovine
muzzle [5, 8], or in the corneal layer of the hairy
skin of other mammalian groups including man
[31-34].

The role of intercellular glycoconjugates of
outer epidermal layers is, probably, not only re-
lated to functions as cementing and/or barrier
substances as proposed for different mammalian
species [5, 8, 35, 36], but may also include a certain
water storage capacity to keep the snout moist.
Sufficient moisture on the snout may be an impor-
tant prerequisite for successful food finding by
taste reception via a microlayer of mucus to which
odorant molecules are continuously supplied [37].
In this connection, however, the secretional fluid
of the nasal glands should, additionally, be taken
into account [8]. Moreover, such intercellular and
moistened glycoproteins among the rather few
layers of the stratum corneum of the opossum
snout skin, could act as “shock absorbing” sub-
stances, and thereby improve the mechanorecep-
tive sensibility of the snout. This view is empha-
sized by the fact that the opossum as a predomi-
nantly nocturnal animal would need an exquisitely
sensitive tactile apparatus for orientation [12].

ACKNOWLEDGMENTS

The skillful technical assistance of Mrs. Birgit Hass-

W. MEYER AND A. TSUKISE

further is gratefully acknowledged.

Nn

10

11

12

1S

14

IS

REFERENCES

Walter, P. (1961) Die sensible Innervation des
Lippenund Nasenbereiches von Rind, Schaf, Ziege,
Schwein, Hund und Katze. Z. Zellforsch. mikrosk.
Anat., 53: 394-410.

Nafstad, P. H. J. (1972) On the dermal innervation.
Z. Anat. Entw. Gesch., 135: 337-349.

Halata, Z. (1975) The mechanoreceptors of the
mammalian skin. Ultrastructure and morphological
classification. Adv. Anat. Embryol. Cell Biol., 50:
1-77.

Schwarz, R. and Meyer, W. (1990) Haut (In-
tegumentum, Cutis Derma) und Hautanhangs-
organe. In “Die Anatomie von Hund und Katze”.
Ed. by J. Frewein and B. Vollmerhaus, P. Parey
Verlag, Hamburg/Berlin, (In press).

Tsukise, A., Meyer, W. and Schwarz, R. (1983)
Histochemistry of complex carbohydrates in the skin
of the pig snout, with special reference to eccrine
glands. Acta anat., 115: 141-150.

Tsukise, A., Fujimori, O. and Yamada, K. (1988)
Histochemistry of glycoconjugates in the goat
nasolabial skin with special reference to eccrine
glands. Acta. anat. 132: 150-158.

Tsukise, A., Meyer, W., Fujimori, O. and Yamada,
K. (1989) The cytochemistry of glycoconjugates in
the planum nasolabial gland of the goat as studied by
electron microscopic methods. Histochem. J. (In
press).
Meyer, W. and Tsukise, A. (1989) Histochemistry
of glycoconjugates in the skin of the bovine muzzle,
with special reference to glandular structures. Acta
anat., 134, (In press).

Barbour, R. A. (1977) Anatomy of marsupials. In
“The Biology of Marsupials”. Ed. by B. Stonehouse
and D. Gilmore, Macmillan Press, London/Basing-
stoke, pp. 237-272.

Sokolov, V. E. (1982) Mammal skin. Univ. Cali-
fornia Press, Berkeley/Los Angeles/London.
Barge, J. A. J. (1936) Zur Morphologie der Lippe.
II. Uber Lippenbildung bei den Marsupialiern. Z.
Anat. Entw. Gesch., 105: 264-287.

Munger, B. L. (1965) The intraepidermal innerva-
tion of the snout skin of the opossum. J. Cell Biol.,
26: 79-97.

Forthey, J. A. (1973) Cytology of eccrine sweat
glands in the opossum. Am. J. Anat., 136: 205-220.
Winkelmann, R. K. (1964) Nerve endings of the
Nroth American opossum (Didelphis virginiana): a
comparison with nerve endings of primates. Am. J.
Phys. Anthropol., 22: 253-258.

Tsukise, A., Meyer, W. and Ikeda, T. (1985)
Carbohydrate histochemical investigation in the

16

7/

18

19

20

2M

Vip

23

24

U5)

26

Histochemistry of Opossum Snout Skin

scrotal skin of the common American opossum
(Didelphis marsupialis L.). Cell. Mol. Biol., 31:
357-364. ;

Leppi, T. J. (1968) Morphochemical analysis of
mucous cells in the skin and slime glands of hag-
fishes. Histochemie, 15: 68-78.

Humason, G. L. and Lushbaugh, C. C. (1960)
Selective demonstration of elastin, reticulum, and
collagen by silver, orcein, and aniline blue. Stain
Technol., 35: 209-214.

Spicer, S. S.. Horn, R. G. and Leppi, T. J. (1967)
Histochemistry of connective tissue mucopolysac-
charides. In “The Connective Tissue”. Ed. by B. M.
Wagner and D. E. Smith, Williams and Wilkins,
Baltimore, pp. 251-303.

Culling, C. F. A. (1974) Handbook of Histopatho-
logical and Hisochemical Techniques. 3rd ed., But-
terworths, London/Boston.

Lev, R. and Spicer, S. S. (1964) Specific staining of
sulphate groups with Alcian blue at low pH. J.
Histochem. Cytochem., 12: 309.

Pearse, A. G. E. (1968) Histochemistry, Theoreti-
cal and Applied. 3rd ed., Vol. 1, Churchill, London.
Yamada, K. (1973) The effect of digestion with
Streptomyces hyaluronidase upon certain histoche-
mical reactions of hyaluronic acid-containing tissues.
J. Histochem. Cytochem., 21: 794-803.

Mowry, R. W. (1963) The special value of methods
that colour with acidic and vicinal hydroxyl groups in
the histochemical study of mucins. With revised
directions for the colloidal iron stain, the use of
alcian blue 8 GX and their combination with
periodic acid-Schiff reaction. An. N.Y. Acad. Scie
106: 402-423.

Pearse, A. G. E. (1985) Histochemistry, Theoreti-
cal and Applied. 4th ed., Vol. 2: Analytical Tech-
nology, pp. 754-785. Livingstone, Edinburgh.
Alroy, J., Ucci, A. A. and Pereira, M. E. A. (1988)
Lectin histochemistry: and update. In “Advances in
Immunohistochemistry (Neoplasms diagnosis)”. Ed.
by R. A. De Lellis, Raven Press, New York, pp. 93-
Se

Spicer, S. S. and Schulte, B. A. (1988) Detection
and differentiation of glycoconjugates in various cell
types by lectin histochemistry. Bas. appl. His-

Zh)

28

29

30

31

32

33

34

35

36

31)

931

tochem., 32: 307-320.

Yamada, K. and Shimizu, S. (1977) The histo-
chemistry of galactose residues of complex carbohy-
drates as studied by peroxidase-labeled Ricinus com-
munis agglutinin. Histochemistry, 53: 143-156.
Yamada, K. and Hoshino, M. (1972) Morpho-
chemical analysis of epithelial mucosubstances in the
gall bladder of the fowl (Gallus domesticus). His-
tochemie, 29: 120-128.

Muller, G. H., Kirk, R. W. and Scott, D. W. (1983)
Small Animal Dermatology. W.P. Saunders Co.,
Philadelphia/Tokyo.

Mackie, A. M. and Nisbet. A. M. (1959) The
histology of the bovine muzzle. J. agric. Sci., 52:
376-379.

Ookusa, Y., Takata, K., Nagashima, M. and Hira-
no, H. (19839 Distribution of glycoconjugates in
normal human skin using biotinyl lectins and avidin-
horseradish peroxidase. Histochemistry, 79: 1-7.
Nemanic, M. K., Whitehead, J. S. and Elias, P. M.
(1983) Alterations in membrane sugers during
epidermal differentiation: visualization with lectins
and role of glycosidases. J. Histochem. Cytochem.,
31: 887-897.

Schaumburg-Lever, G., Alroy, J., Ucci, A. and
Lever, W. F. (1984) Distribution of carbohydrate
residues in normal skin. Arch. Dermatol. Res., 276:
216-223.

Meyer, W. (1986) Die Haut des Schweines (The
porcine integument). Schliitersche Verlagsanstale,
Hannover.

Brody, I. (1968) The epidermis. In “Handbuch der
Hautund Geschlechtskrankheiten”. Ed. by J. Jadas-
sohn, Ergw. I/1, Springer Verlag, Berlin/Heidel-
berg/New York, pp. 1-142.

Hashimoto, K., King, L. E., Yamanishi, Y.,
Beachey, E. H. and Maeyens, E. (1974) Identifica-
tion of the substance binding pemphigus antibody
and concanavalin A in the skin. J. invest. Der-
matol., 62: 423-435.

Beets, M. G. J. (1971) Olfactory response and
molecular structure. In “Chemical Senses 1: Olfac-
tion (Handbook of Sensory Physiology. Vol. IV)”.
Ed. by L. M. Beidler, Springer Verlag, Berlin/
Heidelberg/New York, pp. 257-321.

ZOOLOGICAL SCIENCE 7: 933-938 (1990)

© 1990 Zoological Society of Japan

Trematodes of the Genus Orientocreadium (Digenea:
Orientocreadiidae) from Freshwater
Fishes of Japan

TAKESHI SHIMAZU

Nagano Prefectural College, 49-7 Miwa 8-chome,
Nagano 380, Japan

ABSTRACT—From Japanese freshwater fishes, Orientocreadium chaenogobii sp. n. and O. pseudobag-
ri Yamaguti, 1934 (Orientocreadiidae), are described and illustrated. Data on their hosts, geographical
distribution and life cycles are provided. The new species from the rectum of Chaenogobius laevis (type
host) and Ch. urotaenia (the freshwater type) (Gobiidae) of Hokkaido most closely resembles O.
pseudobagri and O. siluri (Bykhovskii et Dubinina, 1954) Yamaguti, 1958, but differs from them in the
oral sucker being smaller than the ventral, the vitellaria commencing at the ventral sucker level and

other morphological features.

INTRODUCTION

This paper, the sixth in a series on the digenetic
trematodes of the Japanese freshwater fishes, deals
with two species, including a new one, of the genus
Orientocreadium Tubangui, 1931 [1] (Oriento-
creadiidae).

MATERIALS AND METHODS

The materials and methods have been described
in the first paper [2].

RESULTS AND DISCUSSION

Family Orientocreadiidae Yamaguti, 1958

Orientocreadiinae Yamaguti, 1958 [3], p.171 (type
genus, Orientocreadium Tubangui, 1931 [1]).

Orientocreadiidae: Skrjabin and Koval’, 1960 [4], p. 17;
1963 [5], p. 84.

Diagnosis. Body elongate to fusiform, spinose
or reportedly smooth, may or may not have uni-
cellular glands [6] in forebody. Oral sucker simple,
subterminal. Prepharynx distinct. Pharynx com-
paratively large. Esophagus usually short, rarely
long, bifurcating in front of ventral sucker; intes-

Accepted March 28, 1990
Received January 31, 1990

tinal ceca terminating at near posterior extremity
of body. Ventral sucker in anterior half of body.
Testes two, entire or lobate, median, tandem or
oblique, in posterior half of body. Cirrus pouch
clavate, enclosing internal seminal vesicle, prosta-
tic complex and cirrus, usually extending post-
eriorly to ventral sucker; cirrus spined internally or
not; external seminal vesicle present. Genital
atrium shallow. Genital pore median, immediately
in front of ventral sucker. Ovary entire or tri-
lobed, almost median, between ventral sucker and
anterior testis. No seminal receptacle. Laurer’s
canal present. Ootype complex usually postova-
rian. Uterus serving as seminal receptacle in
proximal coils, extending backward as far as, or to
near, posterior extremity of body, overreaching
ceca laterally; metraterm well developed, spined
internally or not. Eggs operculate, small, numer-
ous, segmented, fully embryonated or not. Vitel-
laria in lateral fields of hindbody, continuous or
not, usually reaching to posterior extremity of
body, may or may not unite to form a lattice work
in younger worms. Excretory vesicle tubular or
Y-shaped, reaching to posterior testis. Intestinal
parasites of freshwater fishes and reptiles.
Miracidia nonoculate, bearing a pair of flame
cells; epidermal cell formula 3, 3. Armatae cercar-
iae produced in daughter sporocysts in Lymnaea
snails; flame cell formula 2 [((3+3+3)+(3+3+3)]

934 T. SHIMAZU

=36. Metacercariae encysting in aquatic animals
such as mollusks, a parasitic olygochaete and fishes
and even in their daughter sporocysts.

Discussion. This diagnosis is based on Skrjabin
and Koval’ [5], Yamaguti [7], Tang and Lin [8],
Besprozvannykh [9] and the present study.

All the available descriptions of the existing
genera and species, particularly those from India,
in the family were based on relatively poor mate-
rial. Dayal [10, 11] described the presence of a
small seminal receptacle in his new species, Neoga-
naga barabankiae and N. secunda. However, it is
suggested that what he called the seminal recepta-
cle is not a true one but a chamber made of the
distal portion of the oviduct and the proximal
portion of Laurer’s canal (see this paper). Tang
and Lin [8, fig. 2] described Laurer’s canal and its
expanded proximal portion as filled with sperm.
Fischthal and Kuntz [6], considering the shape (I-
or Y-shaped) of the excretory vesicle to have a
generic significance, recognized two genera as
valid (but either in the family Plagiorchiidae Lthe,
1901, or partly in it and partly in the family
Brachycoeliidae Looss, 1899): Orientocreadium
Tubangui, 1931 [1] (syn., Paratormopsolus
Bykhovskii et Dubinina, 1954 [12]), with an I-
shaped excretory vesicle; and Ganada Chatterji,
1933 [13] (syn., Neoganada Dayal, 1938 [10], Niza-
mia Dayal, 1938 [14], Ganadotrema Dayal, 1949
[11], Macrotrema Gupta, 1951 [15]), with a Y-
shaped excretory vesicle. Tang and Lin [8] and
Besprozvannykh [9] described and figured a dis-
tinctly Y-shaped excretory vesicle in the cercarial
stages of O. batrachoides Tubangui, 1931 [1], and
O. pseudobagri Yamaguti, 1934 [16], respectively.
In the adult stages of both species, on the contrary,
the excretory vesicle is I-shaped [1, 6, 16, 17]. This
discrepancy remains to be explained. In the pres-
ent study I failed to definitely determine the shape
of the excretory vesicle in O. pseudobagri. All the
described genera and species need further studies
of their morphology and life cycles. Yamaguti [7]
synonymized all the six other genera with Orien-
tocreadium, but I follow Fischthal and Kuntz [6]
for the time being.

The family occurs in Japan, China, the Philip-
pines, India, Russia, Poland, Czechoslovakia, the
Sudan, Egypt and the Rhodesias [7, 18-20].

Genus Orientocreadium Tubangui, 1931

Orientocreadium Tubangui, 1931 [1], pp. 417-418 (type
species, O. batrachoides Tubangui, 1931).

Paratormopsolus Bykhovskii et Dubinina, 1954 [12], p.
789 (type species, P. siluri Bykhovskii et Dubinina,
1954).

Diagnosis. Orientocreadiidae. Esophagus short.
Testes entire. Ovary entire. Excretory vesicle I-
shaped.

Discussion. The following are Japanese species.

Orientocreadium chaenogobii sp. n.
(Figs. 1-3)

Material examined. 1) Lot 1. Twenty-two gravid
whole-mounts (NSMT-PI 3636-3637) from the rec-
tum of Chaenogobius laevis (Gobiidae) from Lake
Toro near Kushiro, Hokkaido, on October 10,
1981, and July 22, 1984. The specimens (3637)
were collected by Nagasawa.

2) Lot 2. One immature and 3 gravid whole-
mounts (NSMT-PI 3639) from the rectum of Ch.
laevis from the Chitose River at Ebetsu, Hok-
kaido, on August 2, 1984.

3) Lot 3. Fifty-eight gravid whole-mounts
(NSMT-PI 3064 and 3638) from the rectum of Ch.
urotaenia (=Ch. annularis) (the freshwater type)
from Lake Toro on June 28, 1984.

Description. Ten gravid whole-mounts of lot 1
measured (Figs. 1-3). Body fusiform, 0.99-2.13
by 0.46-0.63; forebody 0.43-0.71, 30-44% of total
body length. Tegumental spines covering all over
body though becoming sparse posteriorly. Uni-
cellular glands present in forebody anterior to
cecal arch; their ducts uncertain. Oral sucker
0.12-0.20 by 0.13-0.20. Prepharynx thick, 0.05-
0.09 long. Pharynx 0.06-0.11 by 0.09-0.12, with 4
anterior muscular protuberances. Esophagus
short, 0.03-0.08 long, bifurcating about midway
between two suckers; intestinal ceca thick, slightly
sinuous, terminating near posterior end of body.
Ventral sucker 0.14—-0.24 by 0.15-0.26; sucker
width ratio 1: 1.20-1.50.

Testes oblique, entire, usually contiguous, rarely
separated by uterus, 0.08—-0.27 by 0.12-0.24, pre-
equatorial in hindbody. Cirrus pouch club-shaped,
curved, 0.20-0.35 by 0.06-0.09, extending poste-
riorly between ventral sucker and ovary; internal

Trematodes of Orientocreadium 935

Fics. 1-3.

Orientocreadium chaenogobii sp. n. 1: Adult worm, holotype (lot 1), ventral view. 2: Terminal genitalia,

paratype (lot 1), ventral view. 3: Ovarian complex, paratype (lot 1), dorsal view.

Fics. 4-6. O. pseudobagri. 4: Adult worm, holotype (lot 1), ventral view.

view. 6: Ovarian complex in lot 3, dorsal view.

5: Terminal genitalia in lot 2, ventral

(Scale bars: 0.5 mm in Figs. 1 and 4; 0.2 mm in Figs. 2-3 and 5-6.)

seminal vesicle oval, small; pars prostatica oblong,
with a well-developed sphincter at posterior end;
prostatic cells numerous, densely surrounding pars
prostatica and internal seminal vesicle; cirrus long,
slender, spined internally, slightly protrusible; ex-
ternal seminal vesicle saccular, voluminous, be-
hind ventral sucker. Genital atrium wide, shallow.
Genital pore median, in front of ventral sucker.
Ovary globular, to right of median line, closely
facing ventral sucker across cirrus pouch, 0.08-
0.19 by 0.14-0.20. Laurer’s canal running poste-
riorly, opening dorsally about anterior border of
anterior testis; its proximal portion and distal
portion of oviduct making an expanded chamber,
sometimes storing a small number of sperm in it.
Ootype complex postovarian. Uterus occupying
all available space of hindbody; metraterm well
developed, shorter than cirrus pouch, spined inter-
nally. Eggs numerous, operculate, not fully

embryonated though segmented, 32-38 by 18-22
ym. Vitellaria variable in shape and size, distrib-
uted continuously on ventral and lateral sides of
ceca from ventral sucker level to short distance in
front of cecal termination (or about halfway be-
tween posterior testis and posterior end of body),
separated there. Excretory vesicle reaching to
posterior testis, probably I-shaped; excretory pore
terminal.

Hosts: Chaenogobius laevis (type host) and Ch.
urotaenia (the freshwater type) (Gobiidae).

Site of infection: Rectum.

Localities: Lake Toro (type locality) and the
Chitose River, Hokkaido.

Specimens: Holotype, NSMT-PI 3636; 83 para-
types, NSMT-PI 3064 and 3636-3639.

Discussion. This new species, Orientocreadium
chaenogobii sp. n., most closely resembles O.
pseudobagri Yamaguti, 1934 (see below), and O.

936 T. SHIMAZU

siluri (Bykhovskii et Dubinina, 1954 [12]) Yama-
guti, 1958 [3], but differs from them in the oral
sucker being smaller than the ventral (the sucker
width ratio being 1: 1.20—1.50), the diagonal testes
being located usually contiguous and _pre-
equatorial in the hindbody, the submedian ovary
more closely facing the ventral sucker, Laurer’s
canal running posteriorly, and the vitellaria com-
mencing at the ventral sucker level.

There is a gravid whole-mount (NSMT-PI 3640)
of the new species which Nagasawa obtained from
“the intestine of Moroco percnurus sachalinensis”
(Cyprinidae) from Lake Toro on August 8, 1981.
This specimen was not used for the above descrip-
tion because the records of its host and site of
infection on its label may possibly be erroneous.

The species parasitizes Chaenogobius laevis and
Ch. urotaenia (the freshwater type) and possibly
Moroco percnurus sachalinensis in Hokkaido,
Japan [this paper]. It life cycle in unknown.

Orientocreadium pseudobagri Yamaguti, 1934
(Figs. 4-6)
Orientocreadium pseudobagri Yamaguti, 1934 [16], pp.
334-336, fig. 39.

Macroderoides asiaticus Belous in Skrjabin and Antipin,
1958 [21], pp. 519-525, fig. 149.

Material examined. 1) Lot 1. Four gravid
whole-mounts (holotype and paratypes) (MPM
Coll. No. 22290) of Orientocreadium pseudobagri
of Yamaguti [16] from the intestine of [the gigi,
Pelteobagrus nudiceps| (Bagridae) from Lake
Biwa, Shiga Prefecture, on July 15, 1927. The two
paratypes lack the anterior or the posterior part of
the body. Yamaguti recorded the host as “Pseudo-
bagrus aurantiacus.” The Japanese name for the
fish with this scientific name is the “gibachi.” The
label says that the type series was found in the
“gigi”, or P. nudiceps, from Lake Biwa, and
moreover the “gibachi” does not inhabit the lake.

2) Lot 2. Five gravid whole-mounts (NSMT-PI
3633-3634) from the intestine of P. nudiceps from
Lake Biwa at Onoe on May 4, 1979, and June 3,
1980. The specimens (3633) were collected by
Nagasawa.

3) Lot. 3. One gravid whole-mount (NSMT-PI
3635) from the intestine of Silurus lithophilus

(Siluridae) from Lake Biwa at Onoe on June 3,
1980.

Description. 1) For the original description and
figure for O. pseudobagri, see Yamaguti [16].

From lot 1, the holotype and paratype measured
(Fig. 4). Body elongate, 1.18-1.50 by 0.22-0.34,
broadest at about ovarian level; forebody 0.47-
0.55, 37-40% of total body length. Tegumental
spines present all over body though becoming
sparse posteriorly. Unicellular glands present in
region anterior to cecal arch in forebody; their
ducts not worked out. Oral sucker 0.09—0.14 by
0.11-0.16. Prepharynx thick, 0.08-0.09 long.
Pharynx 0.07-0.09 in diameter, with 4 anterior
muscular protuberances. Esophagus short, 0.03-
0.04 long, bifurcating at posterior third of fore-
body; intestinal ceca somewhat sinuous, reaching
to near posterior end of body. Ventral sucker
usually smaller than oral, 0.08—0.14 by 0.12-0.13;
sucker width ratio 1: 0.85-1.07.

Testes subglobular, tandem, usually separated
by uterus, in middle third of hindbody, 0.07-0.09
by 0.10-0.12. Cirrus pouch clavate, curved, 0.16-
0.26 by 0.05—-0.06, extending posterior to ventral
sucker; internal seminal vesicle oval, small; pars
prostatica oblong, with a well-developed sphincter
at its posterior end; prostatic cells numerous,
densely surrounding pars prostatica and internal
seminal vesicle; cirrus long, slender, spined inter-
nally, slightly protrusible; external seminal vesicle
saccular, voluminous, straight or curved, usually
reaching to ovary. Genital atrium fairly wide,
shallow. Genital pore median, in front of ventral
sucker. Ovary spherical, pretesticular, almost me-
dian, standing fairly apart from ventral sucker,
0.08-0.12 in diameter. Laurer’s canal present.
Ootype complex between ovary and anterior tes-
tis. Uterus extending medianly to posterior end of
body, coiled preovarian; metraterm well de-
veloped, shorter than cirrus pouch, spined inter-
nally, surrounded by gland cells. Eggs numerous,
operculate, 30-32 by 16-20 um (collapsed). Vitel-
line follicles variable in shape and size, extending
continuously on ventral and lateral sides of ceca
from ovarian level to some distance in front of
cecal termination (or about halfway between pos-
terior testis and posterior end of body), separated
there. Excretory vesicle reaching to posterior

Trematodes of Orientocreadium 937

testis; its shape not definitely determined (prob-
ably I-shaped); excretory pore terminal.

2) From lots 2-3, 5 gravid whole-mounts mea-
sured (Figs. 5-6). Body 1.10-2.84 by 0.23-0.55;
forebody 0.43-0.79, 28-39% of total body length.
Oral sucker 0.09-0.17 by 0.10-0.19. Prepharynx
0.05-0.09 long. Pharynx 0.08-0.12 by 0.05-0.12.
Esophagus 0.04—-0.10 long. Ventral sucker usually
smaller than oral, 0.09-0.15 by 0.11-0.18; sucker
width ratio 0.82—1.00. Testes 0.06-0.28 by 0.07-
0.30. Cirrus pouch 0.16—-0.40 by 0.04-0.12. Ovary
0.08-0.16 by 0.09-0.20. Laurer’s canal running
anteriorly, opening dorsally at ovarian level; its
proximal portion and distal portion of oviduct
making an expanded chamber, sometimes contain-
ing a small number of sperm. Eggs not fully
embryonated though segmented, 30-44 by 18-22
pm. Vitellaria sometimes interrupted at testicular
levels and confluent posteriorly. Other features
similar to those of lot 1.

Discussion. The present study proposes a slight
emendation of the original description for this
species by Yamaguti [16], as follows: tegumental
spines covering all over body though becoming
Sparse posteriorly; unicellular glands present in
forebody; pharynx with 4 anterior muscular pro-
tuberances; internal seminal vesicle oval, small;
cirrus and metraterm having spines internally; and
eggs operculate. He described the excretory vesi-
cle as a wide tube extending to the posterior testis,
which I failed to confirm in lot 1.

Bykhovskaya and Kulakova [22] listed Mac-
roderoides asiaticus [21] from Pseudobagrus fulvi-
draco of Primorye, USSR, as a synonym of O.
pseudobagri. | agree to this treatment because this
fluke [21] has a large external seminal vesicle and is
closely similar to O. pseudobagri in general mor-
phology, final host and locality (see below). O.
siluri was synonymized with O. pseudobagri by
Fischthal and Kuntz [6], whom Ejsymont [19]
followed. It seems to me, however, that O. siluri
[12] is different from O. pseudobagri in the oral
and the ventral sucker being subequal in size, the
unarmed cirrus and metraterm and the vitellaria
commencing at the posterior border of the ventral
sucker. The trematode that Ejsymont [19] re-
ported as O. pseudobagri from Poland more close-
ly resembles O. siluri than O. pseudobagri.

This species has been recorded from Pelteobag-
rus nudiceps and Silurus lithophilus of Shiga Pre-
fecture, Japan [16, this paper]; Parasilurus | = Silu-
rus| asotus, Percottus glehni and Pseudobagrus
fulvidraco of Primorye, USSR [9, 21, 23-25]; and
P. fulvidraco of Fujian, China [26]. Besprozvan-
nykh [9] and Ermolenko and Besprozvannykh [25]
made experimental and field studies of its life cycle
in Primorye, USSR. Xiphidiocercariae [or arm-
atae cercariae] are formed in daughter sporocysts
in the first intermediate host, Lymnaea peregra
amurensis. Metacercariae encyst in several species
of fishes and mollusks. The final host is Percottus
glehni. Other final hosts there are Parasilurus |=
Silurus| asotus and Pseudobagrus fulvidraco [21,
23, 24].

REFERENCES

1 Tubangui, M. A. (1931) Trematode parasites of
Philippine vertebrates, III: Flukes from fish and
reptiles. Philippine J. Sci., 44: 417-422, pl. 1.

2 Shimazu, T. (1988) Trematodes of the genus Allo-
creadium (Allocreadiidae) from freshwater fishes of
Japan. Bull. Natl. Sci. Mus., Tokyo, Ser. A, 14: 1-
ZN

3 Yamaguti, S. (1958) Systema Helminthum. Vol. 1
(2 parts), Interscience Publishers, New York, xii+
1575 pp.

4 Skrjabin, K. I. and Koval’, V. P. (1960) Superfami-
ly Lepocreadioidea Cable, 1956. In “Trematodes of
Animals and Man”. Ed. by K. I. Skrjabin, Izdatel’-
stvo AN SSSR, Moskva, 18: 17. (In Russian)

5 Skrjabin, K. I. and Koval’, V. P. (1963) Family
Orientocreadiidae Skrjabin et Koval’, 1960. In
“Trematodes of Animals and Man”. Ed. by K. I.
Skrjabin, Izdatel’stvo AN SSSR, Moskva, 21: 83-
115. (In Russian)

6 Fischthal, J. H. and Kuntz, R. E. (1963) Trematode
parasites of fishes from Egypt. Part VII. Oriento-
creadium batrachoides Tubangui, 1931 (Plagior-
chioidea) from Clarias lazera, with a review of the
genus and related forms. J. Parasitol., 49: 451-464.

7 Yamaguti, S. (1971) Synopsis of Digenetic Trema-
todes of Vertebrates. 2 vols., Keigaku Publishing
Co., Tokyo, 1074 pp., 349 pls.

8 Tang, C.-c. and Lin, S.-m. (1973) On the life history
of Orientocreadium batrachoides Tubangui, with a
consideration on the phylogeny of the superfamily
Plagiorchioidea. Acta Zool. Sinica, 19: 11-22, pls.
1-3. (In Chinese with English summary)

9 Besprozvannykh, V. V. (1984) The life cycles of
Orientocreadium pseudobagry Yamaguti, 1934, and

10

11

12

13)

14

15)

16

17

938

Allocreadium baueri Spassky et Roitman, 1960
(Trematoda) from fishes in Lake Khanka. In “Para-
zity Zhivotnykh i Rastenii”. AN SSSR, Dal’nevos-
tochnyi Nauchnyi Tsentr, Biologo-Pochvennyi Insti-
tut, Vladivostok, pp. 71-77. (In Russian)

Dayal, J. (1938) Studies on the trematode parasites
of fishes. A new trematode Neoganada barabankiae
(nov. gen., nov. spec.) from Clarias batrachus. Proc.
Indian Acad. Sci., Sec. B, 7: 132-137.

Dayal, J. (1949) Trematode parasites of Indian
fishes, Part II. Indian J. Helminthol., 1: 93-116.
Bykhovskii, B. E. and Dubinina, M. N. (1954)
Materials on the systematics of digenetic trematodes
of the family Acanthocolpidae Liihe, 1909. Zool.
Zh., 33: 788-793.

Chatterji, R. C. (1933) On the trematode parasites
of a Rangoon siluroid fish Clarias batrachus (Lin-
naeus 1785). Bull. Acad. Sci. Alahabad, 3: 33-40.
Dayal, J. (1938) Studies on the trematode parasites
of fishes. A new trematode Nizamia hyderabadi, n.
gen., n. sp., from the intestine of a fresh-water fish,
Ophiocephalus punctatus. Proc. Natl. Acad. Sci.,
India, 8: 53-58.

Gupta, S. P. (1951) Studies on the trematode
parasites of food-fishes of U.P. A new trematode,
Macrotrema macroni n. gen., n. sp., from the intes-
tine of a fresh-water fish, Macrones cavasius (Ham.)
of the sub-family Leptophallinae Dayal 1938. Indian
J. Helminthol., 3: 101-108.

Yamaguti, S. (1934) Studies on the helminth fauna
of Japan. Part 2. Trematodes of fishes, I. Jpn. J.
Zool., 5: 249-541.

Beverley-Burton, M. (1962) Some trematodes from
Clarias spp. in the Rhodesias, including A/Jlo-
creadium mazoensis n. sp. and Eumasenia bang-
weulensis n. sp., and the comments on the species of
the genus Orientocreadium Tubangui, 1931. Proc.
Helminthol. Soc. Wash., 29: 103-115.

T. SHIMAZU

18

I)

20

ZN

22

23

24

25

26

Zithan, R. (1968) O naleze a variabilite Oriento-
creadium siluri (Bychowsky et Dubinina, 1954) a
Crepidostomum auriculatum (Wedl, 1857) v Cesko-
slovensku. Biologia, Bratislava, 23: 857-862.
Ejsymont, L. (1970) Parasites of the sheatfish,
Silurus glanis L., from the river Biebrza and its
tributaries. Acta Parasitol. Pol., 17: 203-216.
Institute of Hydrobiology, Hubei Province, China
(1973) An Illustrated Guide to Fish Diseases and
Pathogenic Fauna and Flora in Hubei Province.
Kexue Chubanshe, Beijin, 456 pp. (In Chinese)
Skrjabin, K. I. and Antipin, D. N. (1958) Genus
Macroderoides Pearse, 1924. In “Trematodes of
Animals and Man”. Ed. by K. I. Skrjabin, Izdatel’-
stvo AN SSSR, Moskva, 14: 517-537. (In Russian)
Bykhovskaya, I. E. and Kulakova, A. P. (1987)
Parasitic Metazoa (Part 2). In “Key to the Parasites
of Freshwater Fish of the USSR”. Ed. by O. N.
Bauer, Izdatel’stvo Nauka, Leningrad, Vol. 3, pp.
141-143. (In Russian)

Zmeev, G. Ya., (1936) Trematodes and cestodes of
fishes of the Amur River. Parazitol. Sb., No. 6: 405-
436. (In Russian) Cited by Ejsymont [19] and Yama-
guti [7].

Strelkov, Yu. A. (1971) Trematodes from the fishes
of the river Amur’s basin. Parazitol. Sb., No.-25:
120-139. (In Russian)

Ermolenko, A. V. and Besprozvannykh, V. V.
(1987) On pathogenic effect of some metacercariae
of trematodes on freshwater fishes from the South of
Far-East of the USSR. Parazitologiya, 21: 159-162.
(In Russian with English summary)

Wang, P.-q., Sun, Y.-l., Zhao, Y.-r., Zhang, W.-h
and Wang, Y.-l. (1985) Notes on some digenetic
trematodes of vertebrates from Wuyishan, Fujian.
Wuyi Sci. J., 5: 129-139. (In Chinese with English
summary)

ZOOLOGICAL SCIENCE 7: 939-946 (1990)

© 1990 Zoological Society of Japan

Dracoderidae, a New Family of the Cyclorhagid Kinorhyncha
from the Inland Sea of Japan’

RoBERT P. HIGGINS and YOSHIHISA SHIRAYAMA

Department of Invertebrate Zoology, National Museum of Natural History,
Smithsonian Institution, Washington, D.C. 20560, U.S.A., and
Ocean Research Institute, University of Tokyo,

Nakano-ku, Tokyo 164, Japan

ABSTRACT—A new family, Dracoderidae, is erected to include Dracoderes abei, n. sp., n. gen.,
described from subtidal muddy sediment of the Inland Sea of Japan at Mukaishima Island. This species
is distinguished from all other known kinorhynch species by the presence of alternating laterally
displaced dorsal spines on segments 4-10 and a middorsal spine on segment 11, and on segment 12 of
males. In addition, this species has an unique series of five 3-element oral styles alternating with four
single-element oral styles. No midterminal spine or lateral terminal accessory spines are present in
either sex. The male has three penile spines as in Echinoderes; one is exceptionally long and easily
mistaken for a lateral terminal accessory spine, the other two penile spines are short, one with cuticular

banding.

INTRODUCTION

The first report of a member of the Phylum
Kinorhyncha from Japan was Echinoderes masudai
Abe, 1930 from Gogoshima Island near Hiroshi-
ma. In the 60 years since Abe’s paper [1] was
published, Tokioka [2] reported two Echinoderes
species and Sudzuki [3-5] found three kinorhynch
species representing two genera from various loca-
lities of the Japanese coastal waters.

The description of Echinoderes masudai by Abe
[1] was too poor that we considered its precise
redescription is necessary. In April 1986, we thus
carried out bottom dredgings in the area close to
the Mukaishima Marine Biological Station of the
Hiroshima University to collect other specimens of
E. masudai. The area was selected because it
appeared to be a relatively undisturbed habitat
reasonably close to the type locality of the species.

This paper is to describe a new kinorhynch
species found from the dredge samples. The
species was easily classified as a member of the

Accepted April 4, 1990
Received February 3, 1990
' The Kinorhyncha of Japanese Coastal Waters. 1.

order Cyclorhagida. But on the basis of its charac-
teristics in dorsal spines, oral styles and penile
spines, it was considered necessary to erect a new
genus and family to address the new species.

MATERIALS AND METHODS

The specimens described in this paper were
collected by the authors from subtidal (3 m) sandy
mud found in the Mukaishima yacht harbor, south-
west of the Mukaishima Marine Biological Station,
17 April 1986. Collections were made with a
Higgins Meiobenthic Dredge [6].

Kinorhynchs were removed from the sediment
by the bubbling technique [6]. Laboratory proce-
dures followed a standard protocol [7]. Specimens
were transferred by an Irwin Loop from a 10%
formalin fixative to a 2% solution of glycerin in
deionized water, the water was allowed to evapo-
rate, leaving the specimens in glycerin from which
they were transferred to a small drop of Hoyer’s-
125 mounting medium on a double-coverglass Hig-
gins-Shirayama slide mount [6]. Specimens were
observed by differential interference and phase
contrast optics, measurements are given in micro-
meters (“m), standard kinorhynch abbreviations

940 R. P. HIGGINS AND Y. SHIRAYAMA

are used.

Specimens mentioned in this paper are depo-
sited in the National Museum of Natural History,
Smithsonian Institution, indicated by USNM cata-
log numbers and by the senior author’s reference
numbers (RH).

DESCRIPTION
Order Cyclorhagida Zelinka, 1896
Suborder Cyclorhagae Zelinka, 1896
Dracoderidae, n. fam.

Diagnosis. —Segment 2 consisting of 14 placids,
midventral placid widest, others of different
widths. Segment 3 consisting of complete ring of
cuticle. Segments 4-13 consisting of two sternal
plates articulating midventrally with each other
and laterally with single rounded tergal plate.
Midterminal and lateral terminal accessory spines
absent in adult (juveniles unknown). Dorsal
spines on segments 4-10 alternatingly offset lateral
to midline, middorsal spine present on segment 11
of both male and female, and on segment 12 of
male only. Cuspidate lateral spines absent. Mid-
ventral and four other oral styles well cuticula-
rized, with three elements (two-jointed); remain-
ing (alternate) four oral styles with thin cuticle,
one element (unsegmented).

Type genus. — Dracoderes n. gen.

Included genera. —One, Dracoderes, n. gen.

Dracoderes, n. gen.

Diagnosis. —\dentical with family diagnosis.

Type species. — Dracoderes abei, n. sp.
Etymology. —The genus name is derived from the
Greek (drakon) dragon plus (deres) neck. The
gender of the genus is musculine.

Dracoderes abéi, n. sp.
(Figs. 1-5)

Diagnosis. —Identical with genus diagnosis.
Etymology.—This species is named in honor of
Yoshio Abe, the first Japanese scientist to study a
kinorhynch.

Type Material.—Holotype, adult female, RH
2086.1, USNM 235446, col. R. Higgins and Y.

Shirayama, 17 April 1986, sandy mud, 3 m depth,
Mukaishima yacht harbor (southwest of Mukaishi-
ma Marine Biological Station), 34° 21.5’N., 133°
13.0°E. Allotype, adult male, RH 2079.1, USNM
235447, 10m depth, other data same as for
holotype.

Description: Holotype, adult female (Figs. 1, 5a-
c), trunk length 276 ~m; MSW-6 (maximum ster-
nal width at segment 6) 76 um, 28% of trunk
length; SW (standard width at segment 12) 66 um,
24% of trunk length.

Segment 2 (neck) apparently consisting of 14
placids of varying width. Length of placids not
measurable because of withdrawn head. Midven-
tral placid (placid 1) ca. 16 ~m wide, adjacent
placids (placids 2) ca. 8 zm wide, placids 3 ca. 8
ym wide, placids 4 and 5 (both lateral and dis-
torted by compression) not measurable, placids 6
ca. 5ym, placids 7 ca. 8m, middorsal placid
(placid 8) ca. 5 um.

Segment 3 (first trunk segment) a ring of cuticle
31 um long (measured laterally in optical section),
with prominent pachycyclus. Subdorsal, round to
oval cuticular scars occur slightly posterior to
midlength of segment with presumptive sensory
spots slightly more lateral and nearer anterior
margin of ventral surface. Perforation sites (of
cuticular hairs) few and scattered. Cuticular pores
(also in male, Fig.5e, PO), possibly complex
sensory spots, present on either side of ventral
midiine. Posterior margin of cuticle consisting of
pectinate fringe as in all succeeding segments.

Segment 4 and all succeeding segments consist-
ing of single, oval-arched dorsal (tergal) plate with
each lateroventral margin articulating with lateral
margin of one of two single lateroventral (sternal)
plates to form distinct tergal-sternal articulation
zone; two lateroventral plates articulating midven-
trally. Pachycyclus of tergal and sternal plates well
developed. Length of segment 4, 32 wm. Single
dorsal spine, 40 um long, located lateral to dorsal
midline; area of attachment of spine with distinc-
tive anterolateral extensions as in all succeeding
dorsal spines. Moderately spaced perforation sites
appearing to be in three poorly organized horizon-
tal rows on tergal plate, and in two poorly orga-
nized horizontal rows on each sternal plate. Mod-
erately robust, lateral spine, 16 ~m long, present

Dracoderidae, a New Kinorhyncha Family 94]

nT iit ri ri

: i
100 um be ii vinyl ma

va mu =
eects iy

) hin) i ng =

Fic. 1. Dracoderes abei n. gen., n. sp., holotypic female, USNM 235446, neck and trunk segments; a, ventral view;
b, dorsal view.

about midway near lateral margin of tergal plate. | and near posterior margin of each sternal plate,
Sensory spot present slightly mesial to the midline | second sensory spot or possibly a cuticular pore

942 R. P. HIGGINS AND Y. SHIRAYAMA

Ni - = mnt
TN L;
: aaa ar R 100 um Pmnntiony whyy nyguntnfvny 10 ‘
[ Ones ae r S (o Seite? ee pe sf =. aati
iT) Hatt Aine ee FPRETEATYAv rvril) vwanvnnvA TNT
(Pa enn wy mv t eh oor sic ealets ama
aur ana = Pemeno el Tan yyy viyyneuyeattTin
Plame 200 a4 rr monte Fiat “ i Rees.
OF “2,2 20 fon hein ta sth —— |
Anna veN TR AMRanl i nnn TTT ATUNTTE PNTdn Nn RL
NG x55 9 co: oN Dues PR
pcr ee =
| IPB RTERSe oT (MuopToeTOTEO Aeal \ WAC THTvvearinnn|| rT TNH NNOABT TTI
BON SEN tes p OP O35 F wan Pts dna 54.10 ens
; TITETU TURTLE Wanna OL Te TAR

EUTERT Orde kins FetHttH!

°o

< atively ee 8, })
nh ipY ER? a

a b

Fic. 2. Dracoderes abein. gen., n. sp., allotypic male, USNM 235447, neck and trunk segments; a, ventral view; b,
dorsal view.

located about midlength near the lateral margins.

sial margin of each sternal plate; pore canal ex-
Presumptive cuticular pore located near anterome-

tending posteriorly from pore. Sternal muscle

Dracoderidae, a New Kinorhyncha Family 943

scars, poorly defined in anterior segments, present
posterolaterally to presumptive cuticular pores.

Segment 5, 32 um long, with prominent dorsal
spine, 44 um long, situated laterally to middorsal
line, alternate to position of dorsal spine on seg-
ment 4. No lateral spines present. Perforation
sites appear similar to pattern on segment 4.
Cuticular scar, possibly a pore, situated subdorsal-
ly, usually hidden by pectinate fringe of preceding
segment; other cuticular structures such as sensory
spots similar to those of preceding segment. Pre-
sumed muscle scars present near midlateral mar-
gins of tergal plate.

Segment 6, 30 um long. Maximum sternal width
(a measurement of the anterior margin of the
widest pair of sternal plates), 76 wm, 28% of trunk
length, occurrings at segment 6. Prominent dorsal
spine, 48 um long, situated laterally to middorsal
line, alternate to position of dorsal spine on pre-
ceding segment. No lateral spines present. Other

(Vyywiyerverwvyr cry 1 VV
Ge

50 um )
ah ee ON
ie

Fic. 3. Dracoderes abei n. gen., n. sp., allotypic male,

USNM 235447, right half, segment 12 and 13, ven-
tral view.

SO um

Fic.4. Dracoderes abei n. gen., n. sp., allotypic male,
USNM 235447, mouth cone, ventral view.

cuticular structures similar to those described for
segment 5.

Segment 7, 34 um long, with prominent dorsal
spine 52 um long, situated laterally to middorsal
line, alternate to position of dorsal spine on pre-
ceding segment. Lateral spine (Fig. 5a, L7), 15
ym long, narrowly acicular, probably an adhesive
tube, present adjacent to lateral margins of tergal
plate. Other cuticular structures similar to those
described for segment 6.

Segment 8, 34 um long, with prominent dorsal
spine, alternately situated as in previous segments.
Perforation sites appear organized in only two
rows. Robust lateral spine (Fig. 5a, L8), 32 ~m
long, present adjacent to lateral margins of tergal
plate. Unlike other kinorhynchs, cuticle of tergal
plate appears diverted around base of spine, later-
al margin of adjacent sternal plate near base of
lateral spine with ca. 5-um incision (Fig. 5a, SI)
extending posteromesially; lateral spine thus
appearing to be positioned between the tergal and
sternal plates rather than on lateral margins of

944 R. P. HIGGINS AND Y. SHIRAYAMA

Fic.5. Dracoderes abei n. gen., n. sp.; a, holotypic female, USNM 235446, segments 7-9, ventral view; b, same,
segments 8-11, dorsal view (broken line indicates middorsal plane); c, same, segments 11-13, ventral view; d,
allotypic male, USNM 235447, mouth cone, optical section; e, same, segments 1—4 (head, neck, and first two
trunk segments), ventral view; f, same, segments 11-13, ventral view. Phase contrast photomicrographs all to
scale shown ina. D9, D10, D11, dorsal spines; LTS, lateral terminal spine; L4, L7, L8, lateral spines; MC, mouth
cone; OO, oocyte; OS, oral styles; PO, pore; PS, pharyngeal styles; P1, P3, penile spines; SI, segment incision
(on segment 8); SP, spermatozoa.

Dracoderidae, a New Kinorhyncha Family 945

tergal plate (Fig. 1a). Other cuticular structures
similar to those described for segment 7.

Segment 9, 34 um long, has prominent dorsal
spine (Fig. 5b, D9), 52 um long, alternately situ-
ated as in previous segments, and lateral spine, 30
yam long, similar to that of preceding segment;
segment otherwise similar to segment 8.

Segment 10, 32 um long, with prominent dorsal
spine (Fig. 5b, D10), 44 um long alternately situ-
ated as in previous segments, lateral spine, 30 ~m
long, similar to that of preceding segment; seg-
ment otherwise similar to segment 9 but with fewer
perforation sites.

Segment 11, 33 um long, with middorsal spine
(Fig. 5b, D11), 54 um long and lateral spine, 28
ym long, similar to that of the preceding segment
but without cuticular incision at the lateral margins
of sternal plates. Pachycyclus appearing narrower
than in preceding segments and with fewer per-
foration sites.

Segment 12, 33 um long, lacking both dorsal and
lateral spines. Standard width (a measurement
across the anterior margin of segment 12), 66 um,
24% of trunk length. Segment otherwise similar to
preceding segment but with fewer perforation
sites, no apparent muscle scars, and less developed
pachycyclar structure.

Segment 13, 30 um long, without dorsal spine.
Lateral terminal spines (Fig. 5c, LTS), 250 um
long, nearly equal to trunk length (LTS/TL 91%),
extending posterolaterally from anterolateral mar-
gins of segment. Perforation sites not apparent;
large oval muscle scar located near anteromesial
margin, usually covered by pectinate fringe of
preceding segment; other cuticular scars not appa-
rent. Posterior margin of sternal plates rounded,
extremely thin cuticle preventing accurate deter-
mination of limits of plates. Tergal plate bifurcat-
ing into distinctively shaped tergal extensions,
rounded laterally, pointed distally, apparent sen-
sory spot near the lateral margin adjacent to apex.
Mesial border of tergal extensions proceeding
directly anterior from apex before curving mesially
to junction at ventral midline.

Allotype, adult male (Figs. 2-4, 5d-f), trunk
length 248 um; MSW-8 76 um, 31% of the trunk
length; SW 544m, 22% of the trunk length.
Lateral terminal spines 290 um long, 117% of the

trunk length.

The lateral and dorsal spines of the male are
similar to those of the female. The major differ-
ences between the female and male include an
additional middorsal spine, 28 um long, on seg-
ment 12; 3 penile spines at anterolateral margin of
each sternal plate of segment 13; relatively longer
(LTS/TL 117%) lateral terminal spines; and slight-
ly different dorsoventral trunk outline which
reaches a maximum sternal width more posteriorly
(MSW-8, MSW/TL 31%) than in the female and
tapers to a relatively narrower standard width
(SW/TL 22%). The largest and most prominent
penile spine (Fig. 5f, P1) is 60 ~m long and super-
ficially resembles the lateral terminal accessory
spine present in females of the family Echinoderi-
dae. A second penile spine with an extremely thin
cuticle is ca. 26 um long. The third penile spine
(Fig. 5f, P3), ca. 20 um long, has an equally thin
cuticle with slightly thicker bands (Fig. 3).

The head (segment 1) of the allotypic male (Fig.
5d, e) was everted and the mouth cone (Figs. 4,
5d) extended. The scalid numbers and arrange-
ment are not easily determined, but the clearly
extended mouth cone exhibited some characteris-
tics unique to the phylum. Nine oral styles (Fig.
5d, OS), located on all but the middorsal (sector 1)
B-radii [8], surround the mouth cone. The mid-
ventral oral style (B-6) and all other even-sector
(B-2, B-4, B-8, and B-10) oral styles consist of
three elements: a pointed distal element ca. 5 ~m
long, a median element ca. 11 um, and a basal
element ca. 18 wm long. A cuticular thickening
appears on the external surface of each basal
element; short, single, small lateral spines are
present on either side of its attachment to the
mouth cone. The four odd-sector oral styles (B-3,
B-5, B-7, and B-9) have a extremely thin cuticle
and consist of a single, unsegmented element with
two small lateral spines on either side of the base.

DISCUSSION

Although superficially Dracoderes abei appears
to be a member of the genus Echinoderes, close
examination reveals several characters which re-
quires the erection of a new genus which could be
assigned to either the family Centroderidae or the

946 R. P. HicGins AND Y. SHIRAYAMA

family Echinoderidae. At the same time a suite of
unique cahracters requires that a new family be
established for Dracoderes.

On the basis of the complete, undivided ring of
cuticle which comprises the segment 3 (the first
trunk segment) and the division of the remaining
segments into a tergal and two sternal plates,
Dracoderes could be assigned to the Centroderi-
dae. Character states which support this argument
include: 1) the presence of only 14 placids, of
varying widths, as found in Campyloderes (16
placids in all other cyclorhagids); 2) two distinctive
pores, possibly complexed sensory spots, near
posteroventral margin of segment 3 as found in
Condyloderes ; and 3) the presence of dorsal spines
on segments 4-11 (and 12 in male) in the adult.

Character states which support the assignment
of Dracoderes to the family Echinoderidae in-
clude: 1) the absence of a midterminal spine in the
adult; and 2) the absence of a lateral terminal
accessory spine, as in males of all species of
Echinoderes and both sexes in E. coulli.

Three pairs of penile spines, found in the male
of Dracoderes, are common in the monotypic
family Echinoderidae. Penile spines are unknown
in the three genera comprising the Centroderidae.
In one genus (Centroderes) and the monotypic
family Zelinkaderidae, males possess a flexible
subdorsal (and sometimes middorsal as well) spine
with banding similar to that noted in one of two
smaller pairs of penile spines in Dracoderes. Only
two penile spines have been noted in the homa-
lorhagids.

Although no other kinorhynch taxon is known
to possess five 3-element and four single-element
oral styles, 3-element oral styles are present in the
cyclorhagid families Zelinkaderidae, Semnoderi-
dae, and Catertidae as well as in the homalorhagid
family Neocentrophyidae. Single-element oral
styles are present in the cyclorhagid genus Condyl-
oderes and the homalorhagid family Pyc-
nophyidae. Other cyclorhagid genera, including
Echinoderes, have 2-element oral styles with the
exception of the unique fused oral style complex in

Campyloderes.
The lateral, alternate displacement of dorsal

spines as well as the shape of the opening of the
tergal cuticle through which each dorsal spine
emerges are additional characters of Dracoderes
unique within the phylum. Furthermore, no other
taxon has the incised lateral sternal margins found
in segments 8-10 of Dracoderes. Thus, a separate
family has been erected to include the monotypic
genus Dracoderes.

ACKNOWLEDGMENTS

We sincerely thank Dr. Yoshiaki Hirano and other
members of the staff of the Mukaishima Marine Biologic-
al Station for their help in arranging a productive visit to
this station. Our colleagues, Dr. Janet Reid and Mr.
Birger Neuhaus generously reviewed this manuscript and
made valuable suggestions for which we are most grate-
ful. This research was supported by the University of
Tokyo Scientific Research Promotion Fund which is
gratefully acknowledged.

REFERENCES

1 Abe, Y. (1930) Das Vorkommen von Echinoderes in
den japanischen Gewasern. J. Sci. Hiroshima Univ.,
ser. B, 1: 39-43.

2 Tokioka, T. (1949) Notes on Echinoderes found in
Japan. Publ. Seto Mar. Biol. Lab., 1: 67-69.

3. Sudzuki, M. (1976) Microscopical marine animals
scarcely known from Japan. I. Micro- & meio-faunae
around Kasado Island in the Seto Inland Sea of
Japan. Proc. Jap. Soc. Syst. Zool., 12: 5-12.

4 Sudzuki, M. (1976) Recent portraits of wild biota in
Japan. II. The inland sea of Japan around Kasado
Island, Yamaguchi Prefecture. Obun Rons6o, 7: 11-
32 (In Japanese).

5 Sudzuki, M. (1979) Some aspects of the haline
interstitial biota from Ryukyu Shoto, subtropical
chain islands, Southwest Japan. Sesoko Mar. Sci.
Lab. Tech. Rep., 6: 37-50.

6 Higgins, R. P. and Thiel, H., editors. (1988) Intro-
duction to the Study of Meiofauna. Smithsonian
Institution Press, Washington. 487 pp.

7 Higgins, R. P. (1983) The Atlantic barrier reef
ecosystem at Carrie Bow Cay, Belize, II, Kinorhyn-
cha. Smithson. Contr. Mar. Sci., 18: 1-131.

8 Higgins, R. P. (1990) Zelinkaderidae, a new family
of cyclorhagid Kinorhyncha. Smithson. Contr. Zool.,
500: 1-26.

ZOOLOGICAL SCIENCE 7: 947-954 (1990)

© 1990 Zoological Society of Japan

Systematic Position of Stephanella hina (Bryozoa: Phylactolaemata),
with Special Reference to the Budding Site and
the Attachment of Sessoblasts

HIDEO MUKAI

Department of Biology, Faculty of Education, Gunma University,
Maebashi, Gunma 371, Japan

ABSTRACT—The systematic position of Stephanella hina Oka was revised on the basis of a
morphological observation, and a proposal was made to place the genus Stephanella in its own family,

the Stephanellidae.

Stephanella differs from the other genera of the Phylactolaemata in that buds

appear across a wide region of the cystid, and the sessoblast attaches to the substratum on the

deutoplasmic side.

INTRODUCTION

Stephanella hina, belonging to a monotypic
genus of the Phylactolaemata, was first described
by Oka [1] from Tokyo. Later, Toriumi [2]
redescribed the species based on many colonies
from the Tohoku district of Japan. Recently, this
species was also found to occur in northeastern
North America [3].

Stephanella hina is characterized by branching
colonies ranging from flat to massive in form, a
stolon-like elongation of cystids, circular floato-
blasts and slightly oval sessoblasts [2]. As to the
systematic position of this species, there are sever-
al different views [1, 3-5]. In this paper, I describe
the morphological features of S. hina, with special
reference to the budding site and the attachment of
sessoblasts, and discuss the systematic position of
this species.

MATERIALS AND METHODS

In winter and early spring, from 1978 to 1983, a
number of Stephanella hina colonies were obtained
from a pond at Kasagake, near Kiryi, Gunma
Prefecture, Japan. Colonies grew on the under-
sides of logs, twigs, leaves and other substrata

Accepted April 7, 1990
Received January 10, 1990

submerged in the water. The specimens collected
were fixed either in 10% formalin for preservation
or in Bouin’s solution for histological examination.
Several colonies, collected by Dr. M. Toriumi
from some localities near Sendai, Miyagi Prefec-
ture, Japan, from 1949 to 1951, were also ex-
amined.

Most observations were carried out with pre-
served material. For histological purposes, par-
affin sections, 7 um thick, were stained with De-
lafield’s hematoxylin and eosin.

MORPHOLOGICAL OBSERVATION

Colony and polypide

Colonies are flat or more or less massive in gross
morphology, show a branching configuration, and
are rather small, a few to several centimeters in
major axis. Each unit of the colony is a zooid,
which consists of a polypide and a cystid. The
cystid, representing the body wall, is elongate and
tube-like. The polypide is located in the distal end
of the cystid, and the rest of the cystid is stolon-like
in appearance. The cystids of neighboring zooids
form a continuous branch with no septa between
the zooids. The cystid is composed of an inner
living part called the endocyst and an outer non-
living part called the ectocyst. The ectocyst is
gelatinous, colorless and hyaline, and covers the

948 H. Muxkal

endocyst in a thick layer. The ectocyst is too soft
to hold a definite configuration by itself, and in
fact, the ectocysts of all zooids in a colony are
agglutinated to form a common gelatinous mass in
which the colony proper is embedded.

In relatively young colonies, the branches open
widely and the polypides are rather sparse. Such
colonies are repent. Although the polypidal parts
arise from the substratum, the stolon-like parts of
the endocysts attach entirely to the substratum
(Fig. 1). On the other hand, old colonies, especial-
ly those growing on a limited substratum, are
compact and crowded with polypides. In such
colonies, the polypides arise from the substratum
to a height of 1-1.5 cm (Fig. 2A). The stolon-like
parts of the endocysts are U-shaped, with a poly-
pide at the top of the ascending limb. The de-
scending limb (or the proximal part) of the U-
shaped endocyst is attenuate; the bending bottom
is close to the substratum, but rarely attaches to it.

Fic. 1. Part of a repent colony (preserved specimen) of
Stephanella hina. Polypides, buds (only those clear-
ly recognized as such under a stereomicroscope),
mature or nearly mature statoblasts are depicted.
Ectocyst is not shown. Small buds are marked by
arrowheads. Asterisks indicate parts probably
broken in the field. fb, floatoblast; rec, rectum; sb,
sessoblast; st, stomach; ts, tentacle sheath. Scale
bar: 1 mm.

The ascending limb contains many floatoblasts.
The polypides are always upright when ex-
panded (Fig. 2B). The lophophore is small and

Fic. 2. Compact colonies of Stephanella hina. A.
Shows upright zooids arranged radially around a
twig (preserved specimen). B. Surface view of a

living specimen. Scale bars: 1 mm.

Systematic Position of Stephanella hina 949

Bid D

bee

Fic. 3. Distal parts of creeping branches in repent colonies (preserved specimens) of Stephanella hina. Zooids and
buds (only those clearly recognized as such under a stereomicroscope) are depicted, and small buds are marked by
arrowheads. Statoblasts and ectocyst are not shown. A micrograph of the boxed area in A is shown in Fig. 4A.

Scale bar: 1 mm.

horseshoe-shaped with short lophophoral arms. In
a compact colony, the tentacle number ranges
from 34 to 45 with an average of 40.8 (n=20
polypides, preserved). In healthy polypides, the
epistome is reddish-brown and the stomach is pale
brown. When polypides are contracted into the
cystids, their digestive tracts are usually straight,
but are sometimes folded in repent colonies.

Budding site

Distal parts of branches in repent colonies are
shown in Figure 3. Buds appear in the terminal
part of the branch and in the lateral parts of the
endocyst. Hereafter, a bud formed in the terminal
part of a branch and that formed in the lateral wall
of an endocyst will be referred to as a terminal bud
and a lateral bud, respectively. A branch grows
forward by successively producing terminal buds,
one by one. When a lateral bud develops into a
zooid, branching occurs.

The positional relationship between a polypide
and its lateral buds is not constant. In the distal
part of a branch, one or two lateral buds are
usually located close to each polypide, at the front,
lateral or rear of the polypide. Moreover, lateral

buds are sometimes found between polypides, or
in the stolon-like part of the endocyst. Such lateral
buds are most frequently encountered in some
endocysts several polypides distant from the grow-
ing tip of a branch (see boxed area in Figs. 3A,
4A). The lateral buds in a cystid are not uniform in
developmental stage. Lateral buds in the proxim-
al, stolon-like part of the endocyst are generally
younger than the few lateral buds near the poly-
pide. The formation of buds in a region of the
cystid far from the polypide is unusual in the
Phylactolaemata. The histological section of such
a lateral bud is shown in Figure 4B.

In old compact colonies, as described above,
most endocysts are not directly adherent to the
substratum, though embedded in the ectocyst jel-
ly. In such endocysts, no lateral bud is found. In
the distal parts of branches free of the substratum,
polypides grown from terminal buds are serially
connected by U-shaped endocysts (Fig. 4C).

Floatoblast

The zooid forms a number of floatoblasts around
the funiculus. Mature floatoblasts are set free from
the funiculus and accumulate in the coelom. It is

950 H. MuKAalI

Fic. 4. Budding (preserved specimens) of Stephanella hina. A. Part of a repent colony (boxed area in Fig. 3A)
showing the presence of lateral buds here and there along the endocyst (ectocyst removed). B. Longitudinal
section of the stolon-like part of the cystid, showing a lateral bud attached to the endocyst (en). The ectocyst (ec)
with debris stuck to it is seen. C. Terminal portion of a branch free of the substratum in a compact colony. Scale

bars: A, 1 mm; B, 0.1 mm; C, 1 mm.

=

Fic. 5. Floatoblasts of Stephanella hina. A. Mature floatoblasts. The fenestra is larger on the deutoplasmic side than
on the cystigenic side. B. Surface view of the deutoplasmic valve with a small projection in the center. Scale bars:

A, 0.5 mm; B, 0.1 mm.

Systematic Position of Stephanella hina 951

not rare for 20-30 floatoblasts to be found in the
coelom of a zooid. The floatoblasts are circular or
nearly circular with a diameter of 350-390 »m, and
are buoyant with the marginal float filled with gas
(Fig. 5A).

To distinguish between both sides of a stato-
blast, the terms “cystigenic” and “deutoplasmic”
are used in the sense of Hyman [6]. The deuto-
plasmic side refers to the surface with which a
statoblast attaches to the funiculus during its de-
velopment, whereas the cystigenic side refers to
the opposite surface. As is well known, in a
developing statoblast, the capsule is initially sec-
reted on the cystigenic side and eventually closes in

the center of the deutoplasmic side [7].

A tiny projection is present in the center of the
deutoplasmic valve (Fig. 5B). The central projec-
tion, so designated in this paper, corresponds to
the place where the capsule closes and where the
statoblast is connected to the funiculus during its
development. This is easily ascertained by ex-
amining nearly mature floatoblasts still attached to
the funiculus. The float covers the capsule more
broadly on the cystigenic side than on the deuto-
plasmic side. The fenestra, the central area not
covered by the float [8], is slightly reticulated and
almost circular on both sides; 110-130 um in di-
ameter on the cystigenic side and 165-190 um on

Fic. 6. Sessoblasts of Stephanella hina. A. Frontal view of four mature sessoblasts. B. Surface view of frontal valve.
C. Surface view of basal valve with a small projection in the center. D. Histological section of a developing
sessoblast (capsule-secreting stage). The marginal lamella (arrow) leans upward, indicating that the upper side is
the frontal side. The capsule is thinnest just below the asterisk, indicating that the lower side (hence the basal
side) is the deutoplasmic side. E. Histological section of a mature sessoblast, showing a thick cementing layer

(ce). Scale bars: A, 0.5 mm; B-E, 0.1 mm.

952 H. MuKAI

the deutoplasmic side.

Sessoblast

Sessoblasts are formed only in the repent parts
of the endocysts, and are fixed to the substratum.
The sessoblasts are usually oval (but occasionally
nearly circular), measuring 415-460 ~m and 330-
375 ym in major and minot axis, respectively (Fig.
6A). A narrow peripheral lamella or annulus leans
toward the frontal side. The frontal valve is
reticulated (Fig. 6B). Some irregular tubercles as
described by Toriumi [2] are present on the surface
of the frontal valve. The basal valve is simply
reticulated, and has a small projection in the
center (Fig. 6C). This central projection is similar
in appearance to that of the floatoblast, suggesting
that the basal side of the sessoblast is the deuto-
plasmic side. This view was ascertained by histolog-
ical study. Since the capsule closes in the center of
the basal surface, this is the deutoplasmic side
(Fig. 6D). Thus, the sessoblast of stephanella hina
attaches to a substratum on the deutoplasmic side.
Under and around a mature sessoblast, the ecto-
cyst is dark brown, opaque and hard. This pig-
mented, hardened ectocyst serves as a cementing
layer, fixing the whole basal surface of the sesso-
blast to the substratum (Fig. 6E). A special struc-
ture for attachment, or a so-called attaching appa-
ratus is not formed from the basal valve of the
sessoblast.

DISCUSSION

Stephanella hina is unique in that it possesses
budding ability in a wide region of the cystid. Buds
appear in the terminal part of the branch (terminal
buds) and in the lateral walls of the cystid (lateral
buds). Oka [1] and Toriumi [2] have already noted
the presence of buds in the stolon-like part of the
cystid. A branch of S. hina grows forward by
successively producing terminal buds one by one,
which is essentially the same as in other species
with branching type colonies [7, 9, 10]. As for the
lateral bud, no definite budding site is found in
relation to the polypide. Lateral buds formed in
the stolon-like part of the cystid are, however,
usually smaller and younger than the few buds
formed close to the polypide, suggesting a belated

appearance of the former compared with the
latter.

In most species of the Phylactolaemata, the
budding site is confined to the distal part of the
cystid on the ventral side of the maternal polypide,
where buds arise successively in an ordered man-
ner and migrate forward or laterally as they grow
[7, 9, 10]. As a result, in Plumatella species having
branching type colonies, lateral branches issued
from a maternal zooid are arranged proximo-
distally in the order of appearance [10]. In only
one species, Fredericella sultana, has it been re-
ported that the cystid can produce buds in other
parts than the normal budding site. In this species,
a secondary polypide appears in the basal part of
the cystid of the ancestrula germinated from a
statoblast [9] and the cystids of ordinary zooids can
produce additional polypides when artificially cut
5 a, WAT

The terms “cystigenic” and “deutoplasmic” are
used after Hyman [6] to distinguish between both
sides of a statoblast (floatoblast or sessoblast). As
pointed out in this paper, the deutoplasmic side is
characterized by the presence of a central projec-
tion. The central projection in floatoblast or
sessoblast of Stephanella hina has been demon-
strated by some workers [3, 8, 13], but none of
them have made references to what the projection
means biologically. Since a floatoblast usually
floats with the cystigenic side up, this side is
customarily called “dorsal” and the opposite side
“ventral”. However, the dorsal and ventral sides
are sometimes confused in the literature.
Although both Wood [8] and Smith [3] report that
the central projection of the floatoblast of S. hina
is in the dorsal valve, this is incorrect.

The sessoblast of Stephanella hina is unique in
that it attaches to a substratum on the deutoplas-
mic side and that it lacks a sclerotized attaching
apparatus. It is attached directly to the cementing
layer, a specialized ectocyst, by the whole deuto-
plasmic surface, and no sclerotized structure is
extended from the basal deutoplasmic valve. The
detailed process in the formation of sessoblats was
only recently studied in Plumatella casmiana [14]
and P. emarginata [15]. In these species, the
sessoblasts attach to a substratum on the cystigenic
side. Along the submarginal zone of the cystigenic

Systematic Position of Stephanella hina 953

valve, a chitinous layer or the so-called attaching
apparatus grows out and becomes connected to the
underlying cementing layer. Essentially the same
mode of attachment for sessoblasts has been ascer-
tained in some other species of Plumatella (P.
repens, P. vorstmani and P. fruticosa) and Gelat-
inella toanensis (unpublished observations).

As to the systematic position of Stephanella
hina, there are several different views. This spe-
cies was placed in the family Plumatellidae by both
Oka [1] and Toriumi [4]. Lacourt [5] erected the
family Stephanellidae to comprise a single species,
S. hina. He assumed a phylogenetic relation
between S. hina and Cristatella mucedo, based on
the growth pattern of colonies (according to him,
leading to spindle-shaped colonies in both species)
and the shape of floatoblasts (circular in both
species). Later, Smith [3] challenged the proposal
of Lacourt [5] and re-placed S. hina in the Plum-
atellidae based chiefly on statoblast morphology.
In my opinion, S. hina should be placed in its own
family, the Stephanellidae, on the basis of peculiar
features different from those adopted by Lacourt
[5]. The family Stephanellidae is characterized by
widely distributed budding ability in the cystid and
by the attachment of sessoblasts on the deutoplas-
mic side.

Stephanella seems to be related on one hand to
Plumatella. These two genera produce branching
colonies and have both floatoblasts and sesso-
blasts. On the other hand, Stephanella may be
related to Fredericella because, as mentioned be-
fore, at least F. sultana retains blastogenic ability
in parts of the cystid other than the normal bud-
ding site. Contrary to the statement of Lacourt [5],
I do not find any similarity between S. hina and
Cristatella mucedo in the form of colonies. The
circular shape of the floatoblasts in these two
species seems to be superificial without phylogenet-
ic significance (see also [16]). A circular outline is
also known in the floatoblast of Hyalinella orbi-
sperma [17] and in the deutoplasmic valve of the
floatoblast of Asajirella gelatinosa [16, 18}.

SYSTEMATIC ACCOUNT
Family Stephanellidae Lacourt, 1968

Revised diagnosis: Colony branching; ectocyst

soft and gelatinous; endocyst elongate and stolo-
nial in proximal part; budding site in terminal of
branch and in lateral walls of cystid; both floato-
blast and sessoblast are formed; sessoblast attaches
to a substratum on the deutoplasmic side.

Genus Stephanella Oka, 1908
Stephanella hina Oka, 1908

Stephanella hina Oka, 1908 [1]; Toriumi, 1955
[2]; Toriumi, 1956 [4]; Lacourt, 1968 [5]; Smith,
1988 [3].

At the present time, this is the only species
belonging to the family Stephanellidae.

ACKNOWLEDGMENTS

I am grateful to Dr. Makoto Toriumi for kindly
supplying invaluable specimens. Thanks are also due to
Dr. Tsukané Yamasaki of Tokyo Metropolitan Universi-
ty for his critical reading of the manuscript.

REFERENCES

1 Oka, A. (1908) Ueber eine neue Gattung von
Siisswasserbryozoen (Stephanella n.g.). Annot.
Zool. Japon., 6: 277-285.

2 Toriumi, M. (1955) Taxonomical study on fresh-
water Bryozoa, XI. Stephanella hina Oka. Sci. Rep.
Tohoku Imp. Univ., Ser. 4, 21: 131-136.

3 Smith, D. G. (1988) Stephanella hina (Ectoprocta:
Phylactolaemata) in North America, with notes on
its morphology and systematics. J. N. Am. Benthol.
Soc., 7: 253-259.

4 Toriumi, M. (1956) Taxonomical study on fresh-
water Bryozoa, XVII. General consideration: In-
terspecific relation of described species and phy-
logenic consideration. Sci. Rep. Tohoku Imp.
Univ., Ser. 4, 22: 57-88.

5 Lacourt, A. W. (1968) A monograph of the fresh-
water Bryozoa-Phylactolaemata. Zool. Verh.
Uitgeg. Rijksmus. Natl. Hist. Leiden, 93: 1-159.

6 Hyman, L.H. (1959) The lophophorate coelom-
ates-phylum Ectoprocta. In “The Invertebrates”.
McGraw-Hill Book Co., New York, Vol.5, pp.
275-515.

7 SBrenk Pe 71953) Etude sur les Phylactolémates.
Ann. Soc. R. Zool. Belg., 84: 301-440.

8 Wood, T. S. (1979) Significance of morphological
features in bryozoan statoblasts. In “Advances in
Bryozoology”. Ed. by G. P. Larwood and M. B.
Abbott, Academic Press, London, pp. 59-74.

10

11

12

13

14

954

Brain, P. (1936) Contribution a l’étude de la répro-
duction asexuée des Phylactolémates. Mém. Mus.
R. Hist. Nat. Belg., Ser. 2, 3: 569-625.

Mukai, H., Fukushima, M. and Jinbo, Y. (1987)
Characterization of the form and growth pattern of
colonies in several freshwater bryozoans. J. Mor-
phol., 192: 161-179.

Braem, F. (1908) Die geschlechtliche Entwicklung
von Fredericella sultana nebst Beobachtungen tiber
die weitere Lebensgeschichte der Kolonien. Zoolo-
gica (Stuttgart), 20: 1-38.

Otto, F. (1921) Studien tiber das Regulationsver-
mO6gen einiger Stsswasserbryozoen. Arch. Entw.-
Mech. Org., 47: 399-442.

Wiebach, F. (1975) Specific structures of sessoblasts
(Bryozoa: Phylactolaemata). Docum. Lab. Géol.
Fac. Sci. Lyon, 3: 149-154.

Mukai, H. (1982) Development of freshwater bryo-

H. MuKai

15

16

iL7/

18

zoans (Phylactolaemata). In “Developmental Biolo-
gy of Freshwater Invertebrates”. Ed. by F. W.
Harrison and R. R. Cowden, Alan R. Liss Inc., New
York, pp. 535-576.

Mukai, H. and Kobayashi, K. (1988) External
observations on the formation of statoblasts in Plum-
atella emarginata (Bryozoa, Phylactolaemata). J.
Morphol., 196: 205-216.

Mukai, H. and Oda, S. (1980) Comparative studies
on the statoblasts of higher phylactolaemate bryo-
zoans. J. Morphol., 165: 131-155.

Bushnell, J. H. (1965) On the taxonomy and
distribution of freshwater Ectoprocta in Michigan.
Pt. 3. Trans. Am. Microsc. Soc., 84: 529-548.
Oda, S. and Mukai, H. (1989) Systematic position
and biology of Pectinatella gelatinosa Oka (Bryozoa:
Phylactolaemata) with the description of a new
genus. Zool. Sci., 6: 401-408.

ZOOLOGICAL SCIENCE 7: 955-959 (1990)

© 1990 Zoological Society of Japan

A New Species of the Genus Paramoera (Crustacea: Amphipoda)
from the Intertidal Zone of Hokkaido, Northern Japan

AKIRA HIRAYAMA

Laboratory of Biology, Department of Liberal Arts, Asia University,
Musashino-shi, Tokyo 180, Japan

ABSTRACT—A new species of Paramoera from the intertidal zone at Shiriuchi, Hokkaido, northern
Japan, is described. The ventral projection of cephalic lobes in the new species is singular to the genus
Paramoera, and the intimation that the new species is one of the genus Relictomoera, being separated
from Paramoera on the basis of this character, is abolishes through re-examination of the known species
of Relictomoera because they have usual anterocephalic lobes anteromedially protruding as in members
of Paramoera and related genera: the genus Relictomoera, therefore, is unvalid, and Relictomoera relicta
and R. tsushimana must be replaced to the former genus, Paramoera. Besides the cephalic lobes, the
characteristic that the pleonal epimeron 3 of the new species prominently expands backward has not
previously been known in any species of the genus Paramoera and the related genera.

INTRODUCTION

Pontogeneid gammarids swarming in the inter-
tidal zone of Shiriuchi, Hokkaido, were collected
by Y. Hanamura, who sent me eight specimens for
identification. These specimens, being all female,
clearly show characteristics of the genus Para-
moera except for one character of head, namely
the possesion of ventral projection in the anter-
ocephalic lobes. This appearance intimates that

this gammarid closely relates to the known species
of Relictomoera, which has been separated from
the genus Paramoera by J. L. Barnard and G. S.
Karaman [1] on the basis of this character and
accomodates two species, R. relicta (Uéno, 1971)
[2] and R. tsushimana (Uéno, 1971) [3]. Thus, I
re-examined the cephalic lobes of these two spe-
cies in Uéno’s type-specimens and noticed his
mis-observation. The result is discussed in re-
marks of the present species.

Fic. 1.

Accepted February 19, 1990
Received October 13, 1989

Paramoera hanamurai sp. nov. Holotype (female, 4.5 mm).

956 A. HIRAYAMA

N1

D,E,M1

B1,J,J1,L1
H,1,K

A,B,C,L

M,N,O
0.1 mm F,G,J2,K1,N1,01

Fic. 2. Paramoera hanamurai sp. nov. Holotype (female, 4.5mm). A: Head. B and B1: Antenna 1 and accessory
flagellum. C: Antenna 2, D: Upper lip. E: Lower lip. F: Maxilla 1. G: Maxilla 2. H: Right mandible. I: Left
mandible. J, J1 and J2: Maxilliped, inner plate and outer plate. K and K1: Gnathopod 1 and palm. L and LI:
Gnathopod 2 and propod. M and M1: Pereopod 3 and dactyl. N and N1: Pereopod 4 and dactyl. O and O1:
Pereopod 5 and dactyl.

A New Paramoerid from Hokkaido 957

Op2emim =RIFK ,

0.5 mm H :
0.5 mm A,B,C,D,E,F,G ~

OFlemm G1 --0:0 mm G1

Fic. 3.

Paramoera hanamurai sp. nov. Holotype (female, 4.5 mm) and paratype no. 1 (female, 3.4 mm: C and C1).

A: Pereopod 6. B: Pereopod 7. C and Cl: Pereopod 7 and dectyl in the paratype no. 1. D: Pleonite 1. E:
Pleonite 2. F: Pleonite 3. G and G1: Pleopod 2 and coupling spines. H: Urosome. I: Uropod 1. J: Uropod 2. K:

Uropod 3. L: Telson.

All the specimens are deposited in the collection
of Asia University.

Paramoera hanamurai sp. nov.
(Figs. 1-3)

Type series

Holotype: Female, 4.5mm, collected from the
intertidal zone in Shiriuchi of Kami-iso District,
Hokkaido, northern Japan; 22 May 1984; coll.
Yukio Hanamura. Paratypes: seven female speci-
mens (nos. 1-7), collected together with the
holotype. Holotype and a part of paratype no. 1
(3.4mm) are mounted on slide glasses in gum-
chloral medium. Collection number: Asia 1.

Description of holotype (female)

Body. 4.5mm long. Head: rostrum very small;
two upper and lower plates of anterocephalic lobes

slightly overlapped each other, the lower one
extending forward ventrally. Gills present on
pereonites 2-6. Pleonal epimeron 3 prominently
expanding posteriorly, armed with two upper and
ventral notches posteriorly, with 6 additional
spines on anterior half of ventral margin.
Antennae. Antenna 1 shorter than antenna 2;
peduncular segments 1-2 armed ventrodistally
with 2 and 1 spines respectively; accessory flagel-
lum uni-segmented, vestigial, with 2 long setae
apically; majority of flagellar segments with one
aesthetasc ventrodistally. Antenna 2: gland cone
of peduncular segment 2 extending to proximal
0.75 point of peduncular segment 3, tapering;
peduncular segment 3 with ventrodistal spine.
Mouthparts. Lower lip: inner plates almost
coalescent with outer plates; outer plates with
about 7 short setae on inner angle of shoulder.
Maxilla 1: inner plate with 7 plumose facial setae;
outer plate with 10 paired, comb-like spines, of

958 A. HIRAYAMA

which 3 are bified; palp biarticulate, with 4 conical
spines apically. Inner plate of maxilla 2 with facial
row of 5 long setae. Both mandibles similar except
for lacinia mobilis and number of accessory spines;
incisor ornamented with 5 teeth; lacinia mobilis in
left mandible ornamented with 5 teeth, in right
mandible bifid, following 6 serrate accessory
spines, plus one bifid spine in left mandible; ter-
minal segment of palp subequal to penultimate
one, brush-like, finely setose on inner side. Maxil-
liped: inner plates with 3 conical spines and 5
plumose setae; outer plates reaching to proximal
3/8 point on second segment of palp, with 8
tooth-like spines gradually getting slender, 3 pro-
ximal spine bifid; length ratio of palpar segments
Dad EO

Gnathopod 1. Rather feeble, subchelate, almost
uniform in width. Coxa 1 rectangular, smaller than
coxa 2. Length ratio of segments from basis to
propod 14:5:6:7:12. Palm undefined, with 3
rows of 3 spines on inner side. Dactyl reaching to
middle row of spines on palm when closed.

Gnathopod 2. Subchelate, sublinear, distinctly
longer than gnathopod 1. Coxa 2 subrectangular,
serrate on anterior half of ventral margin. Length
ratio of segments from basis to propod 11:2:3:6
:7. Palm slightly concave, undefined. Dactyl
fitting into concavity of palm when closed.

Pereopods 3-4. Similar. Coxa 4 slightly concave
along upper half of posterior margin. Posterior
length ratio of segments from basis of propod
11:3:8:6:6 in pereopod 3, 45:15:32:22:27 in
pereopod 2. Propod with 1 spine and 2 unequal
setae posterodistally. Dactyl small, with claw-like
apical part.

Pereopods 5-7. Anterior length ratio of seg-
ments from basis to propod approximately 6:1:3:
3:3 in pereopod 5, 12:2:8:9:9 in pereopod 6;
propod in pereopods 5-6 provided with 3 unequal
locking spines; dactyl in pereopods 5—6 small, with
claw-like apical part. Basis of pereopod 5 winged
posteriorly, subrectangular, with 4 small or minute
teeth aroung medial part of posterior margin; in
pereopod 6 most expanding backward at proximal
0.33 point of posterior marginal length, thin post-
erior plate decreasing in width on distal 0.66 of
length, distinctly serrate posteriorly. Pereopod 7,
missing propod and dactyl, similar to pereopod 6

except for more slender and less serrate.

Pleopod 1. Rami equal in length, 1.4 times as
long as peduncle; proximal segment of inner ramus
with 3 bifid setae; terminal swimming setae 0.64
time as long as rami.

Uropods. Uropod 1, excluding terminal spines,
not extending beyond peduncle of uropod 3; rami
equal in length, about 0.5 time as long as peduncle,
with apical set of 2 spines, one spine stout and
elongate. Uropod 2: rami with a terminal set of 3
spines, central spine stout and elongate; inner
ramus slightly longer than outer ramus. Uropod 3
extending far beyond uropods 1-2; rami lan-
ceolate, aremd with spines and plumose setae
marginally; inner ramus longer than outer ramus.

Telson. Not extending beyond peduncle of
uropod 3, cleft to about 0.8 time its length; lobes
apically bifid, with pair of setae on apical notch,
lacking spines.

Paratype

Paratypes are the same as the holotype in exter-
nal appearances except for the propod and dactyl
of pereopod 7 missed in the holotype. The propod
and dactyl of pereopod 7 in the paratype no. 1
specimen (3.4mm in body length ) is similar to
those of pereopod 4.

Etymology. The specific name hanamurai is in
honour of Mr. Y. Hanamura who collected the
specimens of this new species.

Remarks

The present species has two outstanding charac-
teristics in the molphology of the head; the two
upper and lower plates of anterocephalic lobes are
slightly overlapped each other, and the lower one
protrudes forward. This latter appearance is re-
miniscent of the cephalic lobes of the known
species of the genus Relictomoera, which has been
separated from the genus Paramoera by J. L.
Barnard and G. S. Karaman [1] on the basis of this
appearance and accomodates only two species,
namely R. relicta (Uéno, 1971) [2] and R. tsushi-
mana (Uéno, 1971) [3]. Thus, I re-examined the
cephalic lobes of these two species in Uéno’s
type-specimens: ten of twenty-seven specimens in
R. relicta because all the specimens are in a glass
tube, and the holotype in R. tsushimana. This

A New Paramoerid from Hokkaido 959

observation made it clear that the cephalic lobes of
these two species show anteromedially protrudent
contour usually observed in members of Para-
moera and related genera [1-22] and that they
must be replaced to their former genus, Paramoera
[2-3]. On the other hand, the generic characteris-
tics in the present species well agree with difinition
of the genus Paramoera |4, 7-8, 11]. Therefore,
the present species can be treated as a new species
of the Paramoera on the basis of these characteris-
tic cephalic lobes here. Furthermore, the feature
that the pleonal epimeron 3 of the new species
prominenlty expands backward has not previously
been known in any species of the genus Paramoera
and the related genera [2—4, 6-22].

ACKNOWLEDGMENTS

I express sincere thanks to Mr. Yukio Hanamura of
Enkai Chosa Kaihatsu Co. Ltd., Sapporo, for giving me
the opportunity to examine the present material. I thank
Dr. Masatsune Takeda of National Science Museum,
Tokyo, for giving me the convenience to examine Uéno’s
specimens. Thanks are also due to Prof. Yoshihide
Suzuki of Asia University, Tokyo, for providing me
working space and facilities.

REFERENCES

1 Barnard, J. L. and Karaman, G. S. (1982) Clas-
sificatory revisions in gammaridean Amphipoda
(Crustacea), part 2. Proc. Biol. Soc. Wash., 95:
167-187.

2 Uéno, M. (1971) The fauna of the insular lava caves
in west Japan. VII. Subterranean Amphipoda. Bull.
Nat. Sci. Mus. Tokyo, 14: 161-170.

3 Uéno, M. (1971) Subterranean Amphipoda from
the Islands of Tsushima. Bull. Nat. Sci. Mus. Tokyo,
14: 195-199.

4 Barnard, J. L. (1952) Some Amphipoda from
central California. Wasman J. Biol., 10: 9-36.

5 Barnard, J. L. (1969) The families and genera of
marine gammaridean Amphipoda. Bull. U. S. natn.
Mus., 271: 1-545.

6 Barnard, J. L. (1969) Gammaridean Amphipoda of
the rocky intertidal of California: Monterey Bay to
La Jolla. Bull. U. S. natn. Mus., 258: 1-230.

7 Barnard, J. L. (1972) The marine fauna of New
Zealand. Mem. N. Z. Oceanogr. Inst., 62: 1-216.

8

10

11

12

13

14

15

16

17

18

19

20

21

juju

Barnard J. L. and Drummond, M. M. (1987) Rec-
tification of Amphoediceros willisi Fearn-Wannan
(1968), genus and species remorved to Paramoera
Miers. Proc. R. Sco. Vict., 99: 13-18.

Barnard, K. H. (1916) Contributions to the crusta-
cean fauna of South Africa: The Amphipoda. Ann.
S. Afr. Mus., 15: 105-302.
Barnard, K. H. (1932)
Report, 5: 1-326.
Bellan-Santini, D. et Ledoyer, M. (1974) Gammar-
idens (Crustacea-Amphipoda) des Iles Kerguelen et
Crozet. Tethys, 5: 635-708.

Bousfied, E. L. (1958) Fresh-water amphipod
crustaceans of glaciated North America. Canadian
Field Naturalist, 72: 55-113.

Chevreux, E. (1906) Crustacés amphipodes: Ex-
pédition Antarctique Frangaise (1903-1905) com-
mandée par le Dr. Jean Charcot. Sci. Nat.: Doc.
Sci.: 1-100.

Griffiths, C. L. (1974) The Amphipoda of southern
Africa part 2. The gammaridea and Caprellidea of
south west Africa, south of 20°S. Ann. S. Afr. Mus.,
62: 169-208.

Ledoyer, M. (1982) Faune de Madagascar. 59(1).
Crustacés amphipodes gammariens. Familles des
Acanthonotozomatidae a Gammaridae. Editions du
CUNT Re S598:

Nicholls, G. E. (1938) Amphipoda Gammaridea.
Australas. Antarct. Exped. 1911-14, Sci. Repts.
ser. C-Zool. and Bot., 2: 1-145.

Schellenberg, A. (1931) Gammariden und Caprel-
liden des Magellangebietes, Siidgeorgiens und der
Westantarktis. Zool. Res. Swed. Antarc. Exped.,
1901-1903, 2: 1-290.

Stebbing, T. R. R. (1906) Amphipoda I: Gammar-
idea. Das Tierreich, 21: 1-806.

Stephensen, K. (1927) Papers from Dr. Th. Morten-
sen’s Pacific Expedition 1914-16. XL. Crustacea
from the Auckland and Campbell Islands. Vidensk.
Medd. fra Dansk naturh. Foren, 83: 289-390.
Stephensen, K. (1944) Some Japanese amphipods.
Vidensk. Medd. fra Dansk naturh. Foren, 108: 25-
88.

Thurston, M. H. (1974) Crustacea Amphipoda from
Graham Land and the Scotia Arc, collected by
operation Tabarin and the Folkland Islands depen-
dencies survey, 1944-59. Br. Antarct. surv. Sci.
Rept., 84: 1-89.

Uéno, M. (1933) Three noticeable freshwater Crus-
tacea of Hokkaido. Annnot. Zool. Japan., 14: 115-
12%

Amphipoda. Discovery

SS —

at lois ex Cs aC aR ere
2 _ haga awa ie Te es eK roa. ale Se a
<r” pliant sabia int dna Seema C7 eae:

an em int ‘ae. hie’ it tf Sdn, A NO: 5 < een

tam cavhoaidiane: RAR ORL, ibe BegeO.-B os

=e Bist (CAR cat a) ais. ener e 3 ni a cae eA a
; Jak Nath te roe ine 165 rea neal cee eT: “ant a
: : ae : Cn Tae ae” a

; oy Sinnerie cehaoale aie de + aig © * if ps i
acascinibrget hick Aste AMES 2s AE Ie er Mae rath a
E: cn Es aD ae ae ts ; pee, 6 aaa unpue Maas Pan ‘1a: ent bist vt on

) '
a ee Ee fel 2 tet MI 09 be
A pe a abut ont “apes ad. pe fi voting: nyowang OFF Se
ee ASF, pats ar (3 ee Be, ait
“ee vali Mid ihe ABs satis "nats ee SS pinta ie she tenotieny
es ty ie oF ot oe #4 . eo 2 : ‘oe
. at CU cee eer a 4 . ‘arta i reers oe OF consiereenag Si
Sah e sat | edit ; - “es oo + IRL oo P sj
hasten ee he ik ed a ada ince eke oie. TT nd tidenss ava tiz Dana
; 3 a F nb miat = SLY. a ay hog
‘ gavags 5 ele A eRe AL OAT tats on. ie
sents HieRT ES pete ct Abaca 4, | 2
¥ Aareny FL Bee ett t. or iP Ea ty a4 are kT es ee Su cle Ray ae) a: ot . F uA ih eB vi o Z Sis ere
a ‘ im "i f 5 i
. ya rere Bs i: Resin lal . 5h
a bis ses PARRA .s id A St a ae, J Sn ; eat Wo ue toe , itt 45 odie
“eR Bar Popheaiieey IEE? 1 fp brat ees + {hose et peronod aaa
(oa) Os cece ree PRIGNIIAeE aie RM) I ‘ oe epee ere
> : 7 ‘ee ‘
> PRIS = Ri ile ue Pm oe iu \ ; Ne}
i Pans Fos SE y Pa} PEM Here ety]
) ut | a ri fn i 5 ix dj 3 1] in =) ft as Nee a i
: 2 15S ee pa
€ SI 2} ad ‘ * ry) a . ies .
ee : é : i
> ‘ : .
ib i ;
y teres 16 ; } ae
i gertcest sata quest
5 4 i : > a04- ey
sh ee i ed ‘
;
f. \
= re & :

ZOOLOGICAL SCIENCE 7: 961-965 (1990)

[COMMUNICATION]

© 1990 Zoological Society of J apan

Diurnal Structural Changes in Rhabdomeric Microvilli of the
Compound Eye in the Blue Crab, Callinectes sapidus

YOSHIHIRO Tou!

Department of Biology, Yale University,
New Haven, CT 06511, USA

ABSTRACT—Structural changes of rhabdoms were
examined in the compound eyes of the blue crab
Callinectes sapidus cultured under a regular light cycle
(L:D=13:11). Retinulae in most eyes contained rhab-
doms with straight microvilli of uniform diameter.
Retinulae with flattened thick microvilli were also found
throughout the day, though their occurring frequencies
are low. Late in the light period retinulae occurred with
thabdomeric regions entirely occupied by cytoplasm
filled with tubules. In such retinulae the receptor cells
contained extensive synthetic organelles, and the volume
filled with rhabdomeric membranes increased by a factor
of approximately two around dusk. These data are
discussed in relation to cellular mechanism of the
photosensitive membrane turnover.

INTRODUCTION

It is well known that in certain crustacean eyes
the rhabdoms are larger during night than during
day [1-7]. The increase in rhabdom size at dusk
includes assembly of new rhabdomeric mem-
branes. In the crab Leptograpsus variegatus rhab-
doms are totally lysed before dusk and new
thabdoms were rapidly assembled after dusk [8].
In the carb Hemigrapsus penicillatus total lysis of
thabdoms were rarely found around dusk, but
rhabdomeric membranes appeared to be internal-
ized by fusion of their constituent microvilli
throughout the day but especially around dusk [6].
Neither total lysis nor fusion of microvilli have
been reported in other crustaceans. For further

Accepted March 3, 1990
Received September 20, 1989

" Present address: Department of Biology, Faculty of
Science, Kyushu University, Fukuoka 812, Japan

understanding of the cellular mechanism under-
lying the turnover of photosensitive membranes,
morphological changes of rhabdomeres should be
examined for more crustacean species.

In the present study diurnal structural changes
of rhabdomeres have been examined in the eye of
the blue crab Callinectes sapidus, and the data are
discussed in relation to membrane turnover, and
are compared with those reported for other crusta-
ceans.

MATERIALS AND METHODS

Blue crabs, Callinectes sapidus obtained from a
commercial supplier were maintained in a labora-
tory aquarium under a regular light cycle (L: D=
11:13) with the light phase extending from 07 : 00
to 18:00 local time. The intensity of the light
during this period was 150 lux. After at least one
week of acclimatization to the laboratory aqua-
rium at 20°C eyes were sampled at 17 different
times throughout the day (Table 1), and fixed for
electron microscopy (EM). Methods used for EM
preparation have been described in detail in a
previous paper [9].

RESULTS AND DISCUSSION

The retinal structure of the Callinectes eye has
already been described [9]. The ommatidium
contains eight retinular cells: one cell of these
contributes only to the extreme distal part of the
rhabdom, and the remaining seven regular retinu-
lar cells contribute to almost the entire length of

ene eee
areas
era

pererner Sa

Tan
ee

eee

Sppcorescrnlebaset
Sy

8 € Se Mos Sua
eR = BE scp ONION
SE Sc Se ER NIE : g
een ee oe es et ean % oes eon

RON ag peo Set RSE
a agers aceon So
Po OO Ee, at

eee
>!

er
pty]

os
oes:

Wee:

bi

Rie seawe
ae oe

a eneaes

Diurnal Rhabdom Changes 963

TaBLeE 1. Sampling times of eyes, and numbers of
eyes with well organized rhabdomeric microvilli
(R), those with thick or flattened microvilli (F),
and those with no rhabdomeric microvilli (N).
Three eyes at 18:10 (asterisks) consist of R and F
rhabdomeres as shown in Fig. 6, but they are
included in F

Times of Numbers

Sampling of Eyes
R F N

07 : 00 (Light on)

07 : 30 3 1

08 : 00 3

10:00 3

13:00 3

US ye 1s) 2

17:00 2 1 2

1730 2

18: 00 (Light off)

18:10 1 3

19: 00 1

20330 1

2S 1

23230 1 1

00 : 00 4

01:00 1

03 : 00 1

05 : 00 4 2

06 : 30 2)

the rhabdom. In this paper only that part of the
trhabdom formed by the seven regular retinular
cells is described. The rhabdom appears as a
square column (380-500 «m in length) surrounded
by vacuoles but connected to the cell bodies of the
surrounding seven retinular cells. Cross sectioned
areas of rhabdom measured at the mid-level are
20-34 um? at noon and 42-83 um” at midnight [9].
The retinular cells located on the apposed sides
and those on the other apposed sides of the square

column alternately project stacks of microvilli.
Thus, the rhabdom appears in vertical sections as
alternate piles of perpendicularly aligned microvil-
lar masses (Fig. 1).

Forty seven compound eyes were sampled at 17
different times of the day (Table 1). In 31 eyes
trhabdoms consisted of closely packed straight
microvilli of uniform size (53-72 nm in diameter)
as Shown in Figure 1. Such eyes occurred through-
out the day.

In 11 other eyes rhabdoms contained mostly
thick or flattened microvilli (Fig. 2). Flattened
microvilli occurred throughout the day, but were
more frequent around dawn and dusk. Similar
microvilli have been reported for the Hemigrapsus
eye, and transition between microvillar mem-
branes and cytomembranes by fusion of microvilli
and re-division of thicker ones have been discussed
[6].

In 2 eyes at dusk (L 10:00) the rhabdomeric
region consisted of intricated bundles of clear
tubules (Figs. 3-5). The tubules were aligned
parallel to the usual microvillar direction. They
were, however, likely to be extracellular space
between microvilli or intracellular tubules, but not
microvilli. Cytoplasms around tubules, which
appeared more electro-dense, were interconnected
and continued to the parent cytoplasm, suggesting
disappearance of usual microvilli in the rhab-
domeric region (Fig. 4). The parent cytoplasm of
these retinular cells also contained bundles of
tubules together with extensive Golgi complexes
and rough and smooth surfaced endoplasmic re-
ticulum (ER) (Fig. 5). Extensive presence of such
synthetic organelles as ER and Golgi complexes
suggests that these cells may be in an early stage of
thabdomeric membrane assembly.

In 3 eyes at the beginning of the dark period (D
00:10) a more advanced stage of membrane
assembly was found. In such eyes some rhabdo-

Fic. 1.

A longitudinal section of a normal rhabdom (RH) sampled at midnight (D 06:00).

Perirhabdomeric

vacuoles are indicated by asterisks as well as in subsequent figures. 7,600.

Fic. 2.
Fic. 3.
with tubules.
organelles. x<8,000.

A longitudinal section of a rhabdom consisting of flattened microvilli sampled at dawn (D 12:30). 28,000.
A cross section of a dusk (L 10:00) retinula whose rhabdomeric region (RR) is replaced by cytoplasm filled
The parent cytoplasm also contains bundles of tubules (arrows) and extensive cytoplasmic

Fic. 4. A higher magnification of a rhabdomere region shown in Fig. 3. Note interconnections of cytoplasm (thin
arrows) around tubules and their continuation (thick arrows) to the parent cytoplasm. Xx 16,000.

with tubules (arrows); R, rough surfaced ER; S, smooth surfaced ER. x 14,000.

Fic. 6. A cross section of a rhabdom with well organized rhabdomeres (N) and those with thicker microvilli (A)
sampled just after darkening (D 00:10). The thicker microvilli contain vacuoles, and their parent cytoplasm
(AC) contains more vacuoles and vesicles (arrows) than those (NC) with the normal rhabdomere. 11,000.

meres consisted of uniform microvilli and some
consisted of thick or flattened microvilli. The
cytoplasm of the thick microvilli often appeared
clear suggesting poor development of axial fila-
ments. This may be related to some function of
axial filaments in morphological changes of rhab-
domeric microvilli reported in some invertebrate
photoreceptors [10]. Rhabdomeres of different
profiles occurred in mosaic in some rhabdoms (Fig.
6). Since more vacuoles occurred in retinular cell
cytoplasm with thicker microvilli than in those with
well organized ones (Fig. 6), these vacuoles may
move towards the rhabdomeric region to be
assembled into future rhabdomeric membranes.

Retinulae without rhabdomeric microvilli (Fig.
3) may be comparable to totally broken down
trhabdomeres of the Leptograpsus eye [8]. This
occurrence in only two of the 12 Callinectes eyes
sampled before dusk (L 08:15 to L 10:00, Table
1) is suggestive of a rapid renewal also in this crab.

It is not known, however, whether rhabdomeric
microvilli were enzymatically lysed or some parts
were internalized by fusion as mentioned above.
Organelles known to be involved in degradation
such as multivesicular bodies and residual bodies
do not show marked increase before dusk as
compared with those around dawn: the number of
multivesicular bodies counted in cross section of

Diurnal Rhabdom Changes 965

the retinula was 32.3+6.6 (n=17, n: number of
retinulae examined) at dawn (L 01:00) and 6.8+
3.5 (n=14) at noon (L 06:00), whereas it was 9.6
+2.3 (n=10) in the retinula without rhabdomeres
(L 10:00) and 10.9+4.4 (n=10) in the normal
retinula (L 10:30) before dusk.

One possibility is that some old rhabdomeric
membranes may be internalized as vesicles or
tubules with partial degradadtion. The occurrence
of retinulae with well organized microvilli in more
than half of the pre-dusk eyes (7/12) suggests that
both degradation and renewal of rhabdomeric
microvilli are rapid, and the two processes occur
simultaneously before dusk. The tubules in the
rhabdomeric region in Figure 3 may be just de-
rived from the internalized microvillar mem-
branes: they were the first drastic sign of rhabdom
changes at dusk and retinulae such as shown in
Figure 6 followed Figure 3.

Some of the internalized old microvillar mem-
branes may be re-used for the renewal of night
type rhabdomeres. If both freshly synthesized
cytomembranes and some internalized old rhab-
domeric membranes are assembled into large night
thabdoms and they are evenly distributed in new
microvilli, a turnover rate of the rhabdomeric
membrane is estimated as follows. If one first
assumes an approximately 2-fold increase in mem-
brane volume of the rhabdom around dusk, and
second that all internalized membranes would be
used for synthesis, then cytomembranes as much
as daytime microvillar membranes would newly be
assembled for the night rhabdom. Since the night
thabdom loses a half of its membranes at the
coming of dawn, a fresh membrane assembled at a
given dusk would leave the rhabdom turnovered
by a half decay time of 24 hr in the Callinectes eye.

The total lysis of the rhabdom was _ hardly
observed in the present study, but this does not

mean that such a change would not occur in the
Callinectes eye. If such a drastic change is a brief
event, and controlled by culture conditions, the
chance for sampling such lysed rhabdoms is low.
Repeated observation of crab eyes by sampling at
shorter intervals like 15min around dusk and
comparison of structural changes of rhabdoms
under different culture conditions are need for a
more complete description of membrane turnover
in crab photoreceptors.

ACKNOWLEDGMENTS

This work was supported by financial aids from NIH
Grant EY-02929 to Prof. T. H. Waterman, Department
of Biology, Yale University. The author is deeply
indebted to Prof. Waterman for giving him an opportun-
ity to carry out this work in his laboratory, and for his
critical discussions and reading of the manuscript.

REFERENCES

1 Nassel, D. R. and Waterman, T. H. (1979) J.
Comp. Physiol., 131: 203-216.

2 Hafner, G. S., Hammond-Soltis, G. and Tokarski,
T. (1980) Cell Tiss. Res., 206: 319-332.

3 Stowe, S. (1981) J. Comp. Physiol., 142: 19-25.

4 Waterman, T. H. and Piekos, W. B. (1981) J. Exp.
Zool., 217: 1-14.

5 Arikawa, K., Kawamata, K., Suzuki, T. and
Eguchi, E. (1987) J. Comp. Physiol., 161: 161-174.

6 Toh, Y. (1987) J. Electron Microsc., 36: 213-232.

7 Eguchi, E., Arikawa, K., Ishibashi, S., Suzuki, T.
and Meyer-Roshow, V. B. (1989) Zool. Sci. 6: 241-
250.

8 Stowe, S. (1980) Cell Tissue Res., 211: 419-440.

9 Toh, Y. and Waterman, T. H. (1982) J. Ultrastruct.
Res., 78: 40-59.

10 Blest, A. D. (1988) In “Advances in Insect Physiol-
ogy”. Vol. 20, Ed. by P. D. Evans and V. B.
Wigglesworth, Academic Press, London/New York,
pp. 1-53.

ges
: Fits as

e: ASPs, mene

hd ef

5

oan

xs lade
PEST ri
» “*

,

:

=

von
r

i Te

‘e

ZOOLOGICAL SCIENCE 7: 967-971 (1990)

[COMMUNICATION]

© 1990 Zoological Society of Japan

Serum and Urine Osmolyte Concentrations during Acclimation
to Various Dilutions of Seawater in the Crab-Eating Frog,
Rana cancrivora

Minoru UcuIvyAMA, TsUYOSHI OGASAWARA!*, TETSUYA HIRANO!,

SAKAE KIKUYAMA’, YUICHI SASAYAMA*

and CHITARU OcuRo*

Department of Oral Physiology, School of Dentistry at Niigata, The Nippon
Dental University, Niigata 951, ‘Ocean Research Institute, University of
Tokyo, Tokyo 164, *Department of Biology, School of Education,
Waseda University, Tokyo 169, and *Department of Biology,
Faculty of Science, Toyama University,

Toyama 930, Japan

ABSTRACT—The crab-eating frog, Rana cancrivora,
inhabits brackish ponds where salinity and osmolality
fluctuate markedly. Serum osmolyte concentrations were
significantly higher than those of urine in R. cancrivora
caught in the field. In acclimation experiments to various
dilutions of seawater (SW), the frogs were soaked in each
medium (tap-water, 50% SW and 75% SW) for 48 hr.
All frogs survived during acclimation in each medium.
Serum osmolality in each group was higher than that of
the acclimated medium. Copious diluted urine was
observed in urinary bladders of frogs acclimated to
tap-water. On the other hand, a lower volume and
higher concentration of bladder urine osmolytes were
observed in 75% SW frogs. Serum Na, K and Cl
concentrations were always higher than those of urine in
all groups. However, urine Ca and Mg concentrations
were higher than those of serum in 75% SW acclimated
frogs, although serum Ca and Mg concentrations were
higher than those of urine in tap-water acclimated frogs.

INTRODUCTION

The crab-eating frog, Rana cancrivora, is the
only amphibian which lives naturally in seawater.

Accepted April 21, 1990
Received March 15, 1990

* Present address: Department of Biology, School of
Sciences, Kanagawa University, Kanagawa 259-12,
Japan

It inhabits mangrove swamps of south-east Asia,
where environmental salinity fluctuates markedly
because of alternate flooding and evaporation.
Although a number of amphibians are capable of
tolerating a moderate saline environment, they are
generally unable to survive for more than a few hr
in external media of more than 350 mOsm/kg [1].
In contrast, R. cancrivora is able to tolerate
concentrations of 80% seawater (about 800
mOsm/kg) [2]. In previous investigations it has
been reported for this species that plasma osmo-
lyte (e.g. Na, K, Cl, urea and free amino acids)
concentrations increase in high-salinity media [2-
4]. Internal osmolality was thus always kept higher
than that of the external medium [2-4]. It has also
been reported that urine osmolality of R. cancri-
vora adapted to seawater is always lower than that
of plasma and external media under experimental
conditions [2, 3]. However, plasma and urine Ca
and Mg concentrations have not been measured.
Furthermore, there are no available data on the
osmolyte concentrations of plasma and urine in
this species in the field.

The present project was undertaken to clarify
the hormonal control mechanisms of mineral
metabolism in R. cancrivora adapted to a brackish

_——

968

water environment. The present study is the first
step towards obtaining an understanding of osmo-
lyte metabolism in R. cancrivora. Blood and urine
osmolyte concentrations (Na, K, Ca, Mg, Cl, N
components) were examined in intact R. cancri-
vora, which had been freshly caught in the field.
Changes in serum and urine composition were also
studied on acclimation to various dilutions of
seawater.

MATERIALS AND METHODS

The crab-eating frog, Rana cancrivora, was
collected around prawn culture ponds at Ang Sila
near Bangkok, Thailand. Animals of both sexes,
weighing 8-70 g, were used in the present study.
These ponds are canalized from the Gulf of Siam.
Water was collected from the ponds and the canal
for salinity, mineral and osmotic concentration
measurements.

Experiment on acclimation

Six forgs were weighed, and urine and serum
samples were collected within 4 hr of capture. The
rest of the frogs were divided into 3 groups (each
consisting of 14 frogs) and acclimated to tap-water
and various dilutions of aged seawater for 48 hr at
room temperature (23-26C). During the ex-
perimental period the animals were not fed. Aged
seawater drawn from the Gulf of Siam was diluted
with aged tap-water. This experiment was con-
ducted at the Marine Science Center of Chulalong-

TABLE 1.
Salinity Osmolality
(%o) (mOsm/kg)
Pond (4/21)? 35 1030
Canal (4/21) 36 1035
Pond (4/27) 19 Sti
Canal (4/27) fet 306
Coastal water 33 945
at Ang Sila
Tap water ND 46
50% SW owe. 475
75% SW DD? 709

“ date of sampling (21 and 27 April 1987).

> data are calculated from the value of chloride.

M. UcuiyaMA, T. OGASAWARA et al.

korn University, Bangkok, from 21 to 25 April
1987.

Blood and urine collection

The frog was weighed and the lower abdomen
was pressed gently to obtain a urine sample. After
opening the body wall, the tip of the heart was cut,
and the blood was collected in non-heparinized
tubes and centrifuged immediately.

Analytical methods

Environmental water, serum and urine samples
were immediately frozen and shipped to Japan,
and then stocked in a freezer until analysis.
Salinity was measured with a salinometer (Salinity
refractometer S/Mill, Atago). Ca, Mg, K and Na
concentrations were determined with a Nichiden-
Varian atomic absorption spectrophotometer
(Model 1200) after appropriate dilution. Chloride
was determined with a chloride meter (Jookoo,
Model C-4OPA). Osmolality was measured with a
semimicro osmometer (Yanako). Urea and
ammonia were determined by colorimetry using
Urea N B-Test Wako and Ammonia Test Wako,
respectively (Wako Chemicals). The data were
evaluated using Student’s t-test.

RESULTS AND DISCUSSION

Salinity, osmolality and mineral concentrations
of the habitat are shown in Table 1, confirming
that the salinity fluctuates markedly. The salinity

Composition and osmolality of habitat and diluted seawater used in the experiments

Na Ca Cl
(mM) (mM) (mM)
463.9 9.60 541.0
488.0 10.00 563.0
247.8 5.80 294.0
148.0 3.60 165.0
448.0 9.40 511.0

0.4 0.65 ND
228.0 6.00 261.0
368.5 8.40 394.0

Serum and Urine Osmolytes of a Frog in SW

969

TaBLE 2. Serum osmolyte concentrations in R. cancrivora in the field and during acclimation to

diluted seawater

75% SW

Group Field Tap water 50% SW
Osmolality 541.3 339.3 SD Dey
(mOsm/kg) + 39.0° (4)? + 10.0° (6) SE, 477 (9)
Na (mM) 220.10 S27 189.15

SE. (OG (9) + 1.97% (13) se SFE ie)
K (mM) 7.62 Wed) 6.96

+ 0.50 (6) sp O53) (Us) + 0.34 (14)
Ca (mM) 3.43 2.34 2.70

+ 0.24! (6) ae) Waikey (CLS) + 0.18" (13)
Mg (mM) nat 0.59 il 7)

se ES) + 0.05% (13) + 0.05® (14)
Cl (mM) 8.3 94.7 160.3

+ 4.68 (4) + 6.1% (6) + 2.28 (6)
Urea (mM) 163.0 24.8 Siiei

3) 5) + 4.9% (6) +: < 3.1°2),(7)

Values are mean+SE.

* Numbers in parentheses are numbers of animals used.

b-<-d Significantly different from field group; ° P<0.05, © P<0.01, * P<0.001.

734.8

+ 7.8°8 (6)
227.96

+ 3.248 (13)
13.02

1258 (8)
2.55

+ 0.10° (13)
1.60

= O13" (12)

203.2
+ 4.6% (6)

°£2 Significantly different from tap-water group; © P<0.05, ‘ P<0.01, ® P<0.001.

TABLE 3.
diluted seawater

Group Field Tap water 50% SW
Osmolality 390108 88.9 20723
(mOsm/kg) + 22.2” (6)? te 5.4° 592) a od ibs (GE)
Na (mM) 5.60 5.83 TES

+ 1.00* (6) + 0.99* (12) + 0.88" (14)
K (mM) 13.24 0.98 2.01

+ *2.25® (5) J OO (GD) + 0.49% (14)
Ca (mM) 233), 2.34

+ 0.208 (5) + 0.04% (12) + 0.26" (14)
Mg (mM) 1.42 0.03 0.84 |

+ 0.238 (5) ae OOS Gay Se Ose Gi)
Cl (mM) 44 6 0.9 4.7

= 2 (6) ee O24) + 0.9°8* (7)
Urea (mM) 136.5 DES MeSer

a8 LS) ae Sus) (0) 41 (7)
Ammonia (mM) 134.18 DL 53.4°8

ate le) se IL (6) ae ADH)
Urine volume? at eas a5 ar

Values are mean+SE.

* Numbers in parentheses are numbers of animals used.
> Extent of urine content in urinary bladder (+, less than 0.4 ml; ++, 0.4-1.0 ml; + + +, more than

1.0 ml).

“© Significantly different from field group; ° P<0.05, ¢ P<0.01, © P<0.001.

Urine osmolyte concentrations in R. cancrivora in the field and during acclimation to

75% SW

354.5
MP

H+

12.72
3.36" (10)

9.36
2.498 (10)

I+

H+

Se
+ 0.959" (10)

3.69

+ 0.65™ (10)
el

ge DiC)

96.0
11.3°8 (5)

100.1
+ 48.4 (5)

+

+

da Significantly different from tap-water group; ‘ P<0.05, ® P<0.01, mePES 0.001.
‘LK Significantly different from serum values of corresponding group; ' P<0.05, ! P<0.01, * P<

0.001.

970 M. UcuiyAma, T. OGASAWARA et al.

of pond and canal water on 21 April was higher
than coastal water from the Gulf of Siam. This
suggests that salinity of the ponds and canal was
concentrated by evaporation. After heavy rain on
27 April, salinity decreased to about half of the
concentration on 21 April. Many R. cancrivora
were captured around the brackish ponds, showing
that they were adapted well to this osmotically
changeable environment. Serum and urine con-
centrations are shown in Tables 2 and 3. In intact
frogs caught in the field, serum osmolyte concen-
trations were higher than those of urine, but
hypotonic to the ambient water (Tables 1-3). On
the other hand, the study on salinity acclimation
showed that serum osmolality of R. cancrivora is
always higher than that of the external medium.
This is consistent with previous studies [2-5]. In
the previous and present experiments on salinity
acclimation, frogs were maintained in diluted
seawater. In those conditions, frogs may maintain
their body fluid hypertonic relative to the medium.
In the present acclimation study, serum Na and Cl
concentrations were very high, though lower than
those of 50% and 75% seawater. This is consistent
with the previous studies, which showed that a
large part of serum osmolality depends on serum
Na, Cl and urea concentrations [1-3]. Thus, it was
confirmed that serum urea concentration contrib-
utes a large part of the serum osmolality in a saline
environment. Other unknown osmolytes also
increased in a saline environment, because serum
osmolality was higher than the total indicated by
osmolyte concentrations in Table 2. On the other
hand, most of the frogs captured in the field stayed
out of the salt water. Therefore, osmolality of
their body fluid may be lower than that of the
environment, although osmolyte concentrations of
their body fluid were very high compared with
values for common amphibian species in fresh
water [1].

In a tolerance study, Gordon et al. [2] reported
that R. cancrivora preferred environmental media
of less than 50% seawater. They also observed
that the frogs died when they were transferred
directly from fresh water to 70% seawater. In the
present study, all frogs seemed to be in good
condition for 48 hr when they were acclimated
directly from about 20% seawater to high salinity,

50% or 75% seawater. As shown in Table 2,
serum Na, K, Cl, Mg and urea concentrations
changed in parallel to mineral concentrations in
the environment. In contrast, serum Ca concen-
tration was stable, although there were large
differences in environmental concentration. The
present study indicates that serum Ca level was
controlled accurately, although environmental Ca
levels fluctuated.

Although accurate measurements of urine flow
rate were not made, as the salinity of the environ-
ment increased, the urine became more concen-
trated as a result of decreased urine production.
Thus copious diluted urine was present in the
urinary bladders of frogs acclimated to tap-water,
whereas there was a lower volume and higher
osmolyte concentrations in urine of 75% seawater
frogs. In all the urine osmolytes, there were
significant differences between the tap-water
group and the 75% seawater group. Considerable
quantities of metabolites other than urea were
present in the urine in all groups as the total
contribution of Na, K, Ca, Ma, Cl and urea to
urinary osmolality was only about 40%. These
results are fairly consistent with the results of a
previous study [2].

In comparison with serum and urine osmolyte
concentrations, serum Na, K and Cl concentra-
tions were always higher than those of urine in all
groups (Tables 2, 3). On the other hand, urine Ca
and Mg concentrations were higher than those of
serum in 75% seawater acclimated frogs, although
serum Ca and Mg concentrations were higher than
those of urine in tap-water acclimated frogs. This
may suggest that net kidney tubular absorption of
monovalent ions occurred in frogs of all acclimated
media. In contrast, net tubular absorption and
secretion of bivalent ions may occur in tap-water
and 75% seawater acclimated frogs, respectively.
These results are consistent with those for clear-
ance studies of marine teleosts and euryhaline
teleosts adapted to seawater, which indicates that
overall net tubular absorption of Na, K and Cl and
net tubular secretion of Ca and Mg occur [6]. It is
known that corticosteroids, such as cortisol, are
important for seawater adaptation in euryhaline
teleosts [see 7]. Thus some control mechanism by
hormones may be involved in the seawater adapta-

Serum and Urine Osmolytes of a Frog in SW 971

tion of R. cancrivora. More studies are needed to
clarify this hormonal control. Recently, Sasayama
et al. [8] studied the histology of the parathyroid
gland and the ultimobranchial gland. Further
studies are also needed to understand the mechan-
ism of control of serum and urine Ca and Mg
concentrations; calcemic hormones such as para-
thyroid hormone and calcitonin may be involved in
the process.

ACKNOWLEDGMENTS

This work was supported by a Grant-in-Aid for
Overseas Scientific Research (62041035) from the Minis-
try of Education, Science and Culture of Japan. The
authors thank Prof. T. Piyukarncharna and the staff of
the Marine Science Department and the Marine Science
Center, Chulalongkorn University, Bangkok, for their
hospitality and help.

REFERENCES

Balinsky, J. B. (1981) J. Exp. Zool., 215: 335-350.
Gordon, M. S., Schmidt-Nielsen, K. and Kelly, H.
M. (1961) J. Exp. Biol., 38: 659-678.
Schmidt-Nielsen, K. and Lee, P. (1962) J. Exp. Biol.
39: 167-177.

Gordon, M. S. and Tucker, V. A. (1968) J. Exp.
Biol., 49: 185-193.

Colley, L., Rowe, R. W., Huggins, A. K., Elliott, A.
B. and Dicker, S. E. (1972) Comp. Biochem.
Physiol., 41B: 307-322.

Hickman, C. P., Jr. and Trump, B. F. (1965) In “Fish
Physiology, vol. 1”. Ed. by W. S. Hoar and D. J.
Randall, Academic Press, New York, pp. 91-239.
Maetz, J. (1969) Gen. Comp. Endocrinol., Suppl. 2:
299-316.

Sasayama, S., Oguro, C., Ogasawara, T., Hirano, T..,
Harumi, T. and Wissesang, S. (1990) Gen. Comp.
Endocrinol., in press.

’

———

2 :
She aad “ess

oe sth sri

inge .
is
*
1
=
?
i]
— i.
&

ZOOLOGICAL SCIENCE 7: 973-977 (1990)

[COMMUNICATION]

© 1990 Zoological Society of Japan

The Effect of Glucose and Phosphate on Mouse Two-Cell
Embryos to Develop in Vitro

Ket Aoxt!, MASAHISA NAKAMURA~*, HIDEO NAMIKI.,

SHOICHI OKINAGA’ and Ktyosut ARAP

‘Department of Biology, School of Education, Waseda University,
Shinjuku-ku, Tokyo 160, and *Department of Obstetrics and
Gynecology, School of Medicine, Teikyo University,
Itabashi-ku, Tokyo, 173, Japan

ABSTRACT— The removal of either glucose or phos-
phate from the complete HTF medium significantly
improved the rate of development of non-inbred (ICR)
mouse 2-cell embryos to 3- and 4-cell stages. When
neither glucose nor phosphate was present, further
improvement in the development of embryos beyond the
2-cell stage was obtained. These observations suggest
that the presence of glucose and phosphate in the
medium during the 2-cell stage is detrimental to embryo-
nic development.

INTRODUCTION

The earliest phase of development of the mouse
embryo involves many distinct morphological and
physiological changes. Successful culture of 2-cell
mouse embryos in vitro through the blastocyst
stage would provide valuable information regard-
ing cell development and differentiation. Howev-
er, with the exception of some inbred and F1
strains [1, 2], embryos from non-inbred mice do
not develop into blastocysts when cultured from
the 1- or 2-cell stage, but cease development at the
2-cell stage in the chemically defined medium; a
phenomenon referred to as the ‘2-cell block’. For
2-cell embryos to develop through the blastocyst
stage in vitro, several prerequisites have been
reported independently; a lactate/pyruvate ratio

Accepted April 4, 1990
Received March 5, 1990
* To whom reprints request should be addressed.

approaching 120 [3], the addition of ethylene-
diamine-tetraacetic acid (EDTA) to the culture
medium [4, 5] and low oxygen tension [6].

In vitro blocks to embryonic development are
not unique to the mouse. Cattle embryos cease
development between the 8- to 16-cell stages [7]
and hamster embryos between the 2- and 4-cell
stages [8]. Schini and Bavister [9] reported that the
removal of glucose and phosphate from the culture
medium eliminates the hamster 2-cell block, allow-
ing 25-75% of embryos to develop to the 8-cell or
morula stage in vitro. According to Chatot et al.
[10], in a CZB medium with no glucose but
containing a lactate/pyruvate ratio of 116, 1mM
glutamine and 0.1mM EDTA, 83% of mouse
2-cell embryos developed beyond the 2-cell stage,
while only 34% did in a CZB medium with
glucose. The object of this study was to examine
the effect of glucose and phosphate on in vitro
development of mouse 2-cell embryos.

MATERIALS AND METHODS

Culture medium

The chemically-defined medium used in all
experiments was the HTF medium [6] (Table 1).
Thirty five-hundredth mM phosphate was used in
this study in place of 0.37 mM phosphate because
the concentration of phosphate was changed easily
when 0.35 mM phosphate was used. Media were

974 K. Aoki, M. NAKAKURA et al.

TABLE 1. Media formation

Concentration (mM)
Component

HTF* CZBP
NaCl 101.60 81.62
KCl 4.96 4.83
KH,PO, 0.37 1.18
MgSO,:7H,0 0.20 1.18
NaHCO; 25.00 De) M2
CaCl,:2H,O 2.04 1.70
Glucose 2.78 5
Na-lactate 21.40 SH)
Na-pyruvate 0.33 OF
Glutamine — 1.00
BSA (mg/ml) 4.00 5.00
Penicillin G (U/ml) 100.00 100.00
Streptomycin (mg/ml) 0.05 0.70
Lactate/pyruvate 64.85 115297

* Quinn et al. [6]. ° Chatot et al. [10]

filter-sterilized through 0.2 ~m Steradisc 25 (Kura-
bo; Osaka, Japan) and gassed with 5% CO, in air.
Only fresh media were used in the experiments.

Collection and culture of embryos

In all experiments, random-bred ICR female
mice (10-12 weeks old) were used. ICR mice were
superovulated with intraperitoneal injections of 5
iu. PMSG (Teikoku Zoki; Tokyo, Japan) and 5
i.u. hCG (Sankyo Zoki; Tokyo) at intervals 48 hr
apart. At 16-17hr post-hCG the females were
killed by cervical dislocation, the oviducts dis-
sected out and the cumulus masses released into
the HTF medium containing 4mg/ml bovine
serum albumin (BSA). For fertilization in vitro
male mice were killed 14 hr after the females were
injected with hCG, the vasa deferentia and epi-
didymides dissected out and one of each placed in
a 0.4 ml drop of pregassed HTF medium with 4
mg/ml BSA [6]. The sperm were squeezed out
gently and allowed to capacitate for 2 hr at 37°C.
AT 18-19 hr post-hCG 10 yl of the sperm suspen-
sion was added to each 0.4ml drop of HTF
medium in Falcon 35-mm dishes (Falcon; Lincoln
Park, NJ, USA) to give a final sperm concentra-
tion of 150sperm/ul. Eggs and sperm were

incubated together for 5 hr at 37°C under liquid
paraffin oil (Squibb; Princeton, NJ). After the
incubation the eggs were washed twice in the fresh
medium pergassed with 5% CO, in air at 37 C, and
transferred into 0.4 ml of the conditioned medium
covered with liquid paraffin oil and incubated at
37°C under 5% CO, in air. Cultures were
terminated at 48 hr and the embryos were scored
for developmental stage. The number of nuclei
per embryo was confirmed by observation under
an Olympus microscope equipped with Nomarski
differential interface contrast (model BH-2; Olym-
pus Optical Co., LTD, Tokyo). Data were
analyzed by Y” tests.

RESULTS

The effects of the presence of glucose, phos-
phate and EDTA on the development of mouse
2-cell embryos were examined. At the beginning
of culture, all the embryos were at the 1-cell stage.
Approximately 86 to 96% of 1-cell embryos
developed into the 2-cell stage. There was no
significant difference in the rates of the first
cell-division occurred between the complete HTF
medium and the conditioned medium. At the end
of culture, therefore, 1-cell embryos were not
scored for the analysis of phosphate and glucose
effects on embryonic development. After culture
of 48 hr, development of 2-cell embryos through
the 3- and 4-cell stages in the complete HTF
medium (containing 2.78mM glucose and 0.35
mM phosphate, but without EDTA) was signi-
ficantly lower (29.3%; P<0.005) than in the
medium lacking either glucose (39.8%) or
phsphate (44.7%) (Table 2). These results were
almost identical to those obtained when 0.07 mM
phosphate was used in place of 0 mM phosphate.
There was no significant difference in the frequen-
cy of elimination of the 2-cell block obtained by
removing either glucose or phosphate from the
complete HTF medium (39.8% vs 44.7%).
However, the elimination of the 2-cell block was
significantly increased when both glucose and
phosphate were removed, as compared to the
samples in which either of them was present
(68.4% vs 39.8% or 44.7%, respectively; P<
0.005) (Table 2).

Two-cell Block in Mice 975
TABLE 2. Effects of glucose and phosphate on 2-cell embryo development
EE SE EE EE EE eeeEESEIO OO eee eee

2.78 mM Glucose 0 mM Glucose

Phosphate (mM) Phosphate (mM)
0 (%) 0.07 (%) 0.35 (%) 0 (%) 0.07 (%) 0.35 (%)
2 Cell 104 (55.3) US (S551) 162 (70.7) 36 (31.6) 2 (BU) 112 (60.2)
3 Cell 25) (SES) 7 (21) 24 (10.5) pes (NG)e3)) 9 (10.3) 25) (Be)
4 Cell 59 (31.4) 45 (32.1) 43 (18.8) 56 (49.1) 51 (58.6) 49 (26.3)
3 Cell+4 Cell 84 (44.7)? 62 (44.3)? 67 (29.3) 78 (68.4)* 60 (69.0)* 74 (39.8)
Total 188 (100) 140 (100) 229 (100) 114 (100) 87 (100) 186 (100)

KH.PO, was used for varying concentration of phosphate. “ Compared with 0.35 mM phosphate; P<0.005.

TABLE 3. Effects of EDTA, glucose and phosphate on 2-cell embryo development

2.78 mM Glucose 0 mM Glucose

Phosphate (mM) Phosphate (mM)
0 (%) 0.07 (%) 0.35 (%) 0 (%) 0.07 (%) O35) (Fa)
ZiCell 18 (25.4) B2e(27E)) 83)(3925) 9 (10.3) S727) 14 (13.7)
3 Cell S 7 @ae3) sy (C257) 24 (11.4) 5 ( 5.7) T(G939) 11 (10.8)
4 Cell 45 (63.4) 71 (60.2) 103 (49.0) T3E(83:9) (W735) 1 G3)
3 Cell+4 Cell 53 (74.6)? 86"(72-9): 127 (60.5) 78 (89.7)* 62 (87.3) 88 (86.3)
Total 71 (100) 118 (100) 210 (100) 87 (100) 71 (100) 102 (100)

EDTA (0.1mM) was added to all the incubation media.

phosphate.

KH,PO, was used for varying concentration of

>< Compared with 0.35 mM phosphate; b, P<0.05 and c, P<0.025.

Similar results were obtained when 0.1mM
EDTA was added to the HTF medium. One-tenth
mM EDTA was chosen for this study, because 20-
100 ~M EDTA supported the development tn vitro
of more than 80% of ICR one-cell embryos
through the 4-8 cell stages [5]. In the presence of
0.1mM EDTA, 60.5% of 2-cell embryos de-
veloped beyond the 2-cell stage in the complete
HTF medium, while only 29.3% did in the absence
of EDTA (P<0.005) (Tables 2 and 3). When
either glucose or phosphate was removed, more
than 74.6% of the 2-cell embryos developed to 3-
and 4-cell stages. That is, when glucose was
omitted from the complete HTF medium, but 0.1
mM EDTA was present, 86.3% of 2-cell embryos
developed beyond the 2-cell stage (86.3% vs
60.5%; P<0.005) (Table 3). With the omission of
phosphate, 74.6% showed like development
(74.6% vs 60.5%; P<0.05) (Table 3). No signi-

ficant difference was observed between the effects
of removing, alternately, glucose and phosphate
(P>0.05). When both glucose and phosphate
were removed from the HTF medium, 89.7% of
the 2-cell embryos developed beyond the 2-cell
stage (Table 3). This rate (89.7%) was not signi-
ficantly different from that obtained when either
glucose (86.3%) or phosphate (74.6%) alone was
removed (Table 3). When 0.1mM EDTA was
added to the HTF medium containing 0.07 mM
phosphate, there was no difference in the rates of
development of 2-cell embryos beyond the 2-cell
stage between this medium and the medium
containing no phosphate (0.07 mM vs 0 mM phos-
phate) (Table 3).

DISCUSSION
The present study shows that the 2-cell block of

976 K. Aoki, M. NAKAKURA et al.

the in vitro development of ICR mouse 2-cell
embryos is partially overcome by the removal of
glucose and/or phosphate from the culture
medium. Chatot et al. [10] have reported that a
significant improvement in embryo development
beyond the 2-cell stage is achieved by the removal
of glucose from the CZB medium containing 0.1
mM EDTA and 1 mM glutamine (see Table 1).
According to their research, 83% of 2-cell embryos
developed beyond the 2-cell stage with the remov-
al of glucose, while only 34% did in medium
containing glucose. In the present study, 86.3% of
2-cell embryos developed to the 3- and 4-cell stages
in the HTF medium which contained 0.1mM
EDTA but no glucose (Table 3). When glucose
was added to the HTF medium containing 0.1 mM
EDTA, 60.5% of the 2-cell embryos developed
beyond the 2-cell stage (Table 3), while only 34%
showed such development in the CZB medium
containing 1 mM glutamine [10]. This difference
might be due to the effect of glutamine on embryo
develepment. Chatot et al. [10] incubated mouse
embryos under 5% CO, 5% Oy, and 90% Nob,
which is reportedly beneficial for culture of some
blocking strains of embryos [4], whereas embryos
in our study were incubated under 5% CQ) in air.
Chatot et al. [10] also cultured embryos in a
medium approaching a lactate/pyruvate ratio to
120, while we used a ratio of 65. Therefore, the
difference between our and their results (60.5% vs
34%) could instead be due to the fact that the
actions of these substances are inhibited by glu-
cose.

The removal of glucose improved the rate of
development of 2-cell embryos to 3- and 4-cell
stages by approximately 10% (Table 2). With the
removal of glucose and, in addition, a phosphate
level lowered from 0.35 mM to 0 mM, a significant
improvement in this development was observed,
when no EDTA was present. This is the first
report, to the best of our knowledge, that shows
that the removal of glucose and phosphate from
the culture medium is beneficial to the develop-
ment in vitro of mouse 2-cell embryos beyond the
2-cell stage. With regard to hamster embryos,
Schini and Bavister [9] reported that the removal
of glucose and phosphate from the culture medium
eliminates the 2-cell block. Our results are consis-

tent with theirs despite the difference in species
used in the experiments. With the addition of 0.1
mM EDTA to the glucose and phosphate-free
HTF medium, further improvement in the de-
velopment of 2-cell embryos beyond the 2-cell
stage was obtained (from 68% to 90%), which
indicates that the removal of glucose and phos-
phate does not completely eliminate the 2-cell
block. It may be speculated that the chelation of
some metal other than Ca** and Mg’ is critical to
the beneficial action of EDTA, because the con-
centration of both ions in the culture medium is
much higher than that found in the medium
containing EDTA [4]. The removal of both
glucose and phosphate would not affect the chela-
tion of some metals.

In embryos from inbred strains of mice, the
glycolytic pathway is blocked through the 8-cell
stage and the development of embryos in vitro is
supported by pyruvate and lactate until that stage
[3, 11]. Therefore, the mechanism by which
glucose and phosphate block the development of
mouse 2-cell enbryos to the 4-cell stage remains
unclear. Nevertheless, it is noteworthy that both
glucose and phosphate cause a 2-cell block in
mouse embryos. Since our method for culture of
mouse embryos is quite conventional, futher in-
formation regarding the 2-cell block may easily be
obtained by following our simple methodology
rather than by adopting some more complicated
one. Further investigations are certainly needed to
explain the mechanism underlying the 2-cell block
produced by these compounds.

REFERENCES

1 Whitten, W. K. and Biggers, J. D. (1968) J.
Reprod. Fert., 17: 399-401.

2 Goddard, M. J. and Pratt, H. P. M. (1983) J.
Embryol. Exp. morph., 73: 111-133.

3 Cross, P. C. and Brinster, R. lis (1973) Expl. Cell
Res., 77: 57-62.

4 Abramczuk, J., Solter, D. and Koprowski, H.
(1977) Dev. Biol., 61: 378-383.

5 Hoshi, M. and Toyoda, Y. (1985) Jpn. J. Zootech.
Sci., 56: 931-937.

6 Quinn, P. and Harlow, G. M. (1978) J. Exp. Zool.,
206: 73-80.

7 Thibault, C. (1966) Ann. Biol. Anim. Biochim.

Two-cell Block in Mice 977

Biophys., 6: 159-164. Lewis, J. L. and Torres, I. (1989) J. Reprod. Fert.,
8 Bavister, B. D., Leibfried, M. L. and Lieberman, 86: 679-688.

G. (1983) Biol. Reprod., 28: 235-247. 11 Biggers, J. D. (1987) In “The Mammalian Preim-
9 Schni, S. A. and Bavister, B. D. (1988) Biol. plantation Embryo: Requlation of Growth and

Reprod 39: 1183-1192. Differentation in Vitro”. Ed. by B. D. Bavister.

imnGhatat. ©. Ly Ziomek, © A> Bavister, B.D-, Plenum Publishing, New York. pp. 1-22.

980 N. OHKUBO

Fic. 1.
from that of the same organ in A.

The surface granulated around exobothridial seta
and the terrace.

Lateral side of prodorsum expanding to form a
curved edge resembling well known lines S and L
in the family Galuminidae. These terms are also
applied here. Line S representing the upper corner
of the edge (see a-b in Fig. 2A), and its posterior
half is developed as a keel. Line L representing
the lower coner of the edge, whole indistinct.

Lateral podosoma Acetabular tectum I widely
developed with a round protrusion anteriorly
where line S connects, extending backward as a

Ay

Ramusella floralis sp. n. A. Dorsal aspect; B. Ventral aspect; C. Sensillar head from a view point different

narrow ridge beyond pedotectum I. Flat swelling
situated between acetabular tectum I and posterior
end of lateral rostral margin. Ventral part of
pedotectum I becoming a keel, terminating more
proximally than lateral prodorsal margin, while
dorsal part extends backward as a narrow granu-
lated ridge.

Acetabulum II wide, opening posteriorly. Ace-
tabular tectum II forming a semicircular ridge.
Pedotectum II absent.

Acetabulum III very narrow, opening anterior-
ly. Acetabular tectum III absent. A long plate

A New Ramusella from Japan 981

ad,

oe E

Fic. 2. Ramusella floralis sp. n.

D

A. Lateral aspect (a-b showing schematic section); B. Rostral tip in antero-

dorsolateral view; C. mouth parts; D. Genital region (the left genital plate removed); E. Anal region.

projecting laterally, surrounding acetabula II and
III. The plate somewhat triangulate in ventral
view, having epimeral seta 3c on its upper side.
The region of podosomal constriction between
acetabula II and III well granulated.

Discidium projected laterally, having dorsal and
ventral slopes: the former steep and the latter
gentle. In lateral view, an edge observed as the
boundary between the two slopes (Fig. 2A). A
straight acute ridge running from the base of
discidium to just behind bothridium, being some-
times observed in dorsal view (Fig. 1A: the ridge
on the left side and only discidium on the right
side). Acetabulum IV as narrow as acetabulum
III, opening laterally at the top of a truncated

cone.

Epimeral region Epimeres I, Il and HI+IV
convex. Epimere II proximally with a backward
swelling. Posterior border of epimere III+IV
producing a keel. Grooves separating epimeres;
sternal and sejugal ones very deep. The bottom of
sternal groove narrow and flat, having two oval
hollows at crossings. A short longitudinal ridge in
front of sternal groove. Setal formula: 3—1-3-3.
Setae la, Ic, 2a and 3a the shortest, equal in
length. Seta 7b medium sized. Setae 3b, 4a, 4b
and 4c very long and slender. Seta 3c the longest,
thick and strongly barbed.

Ano-genital region Genital aperture nearly
rectanguler, though only slightly wider posteriorly,

982 N. OHKUBO

FiG. 3.
(antiaxial).

1.5 times longer than wide. Anterior edge of
genital plate protruding beyond the aperture (Fig.
2D). Genital plate convex at antero-median
region where 3 short setae are located. Two more
short setae situated near posterior corner, 1.5
times longer than the anterior three and about as
long as epimeral seta Ja.

Anal aperture as long as wide. Anal, aggenital
and adanal setae about as long as the epimeral seta
lb. Setae ad; situated a little behind the level of
the anterior margin of anal aperture.

Notogaster Ten pairs of setae arranged as in Fig.
4, where relative distances between adjacent setae
or between seta and notogastral border based on

Ramusella floralis sp. n. A. Leg I (proximal); B. Leg II (antiaxial); C. Leg III (proximal); D. Leg IV

the measurements of one dried specimen are
shown. All setae except cy long and selender.

Three pairs of lyrifissures im, ih and ips are
present; lyrifissures za and ip could not be de-
tected:

Legs Trochanteral setation 1-1—-2-1. Femoral
setation 5—5—3-2; dorsal setae thick. Genual
setation 2(1)-1(1)-1(1)-1; solenidion oIII short
with a blunt tip; seta dIV very thin. Tibial setation
4(2)—4(1)-3(1)-3(1); setae /” III and IV shorter
and thicker than setae /’ III and IV, respectively;
solenidion ¢IV very long. Tarsal setation 16(2)—
12(1)-11-10; a leaf-shaped seta on leg IV.

Type-series Holotype (NSMT-Ac 10300 on 5

A New Ramusella from Japan 983

Fic. 4. Ramusella floralis sp. n. Schema of setal posi-
tion on notogaster, indicated by figures which are
measured on one specimen. One numerical value of
figures equals 2.5 um.

slides) and 12 paratopotypes (on slides and in
spirit); Kagamigahara, Gifu, July 20, 1989, col-
lected by T. Shimohata. All specimens will be
deposited in the collection of the National Science
Museum (Nat. Hist.), Tokyo.

Remarks The new species has two features
exceptional for the genus Ramusella: absence of
interlamellar setae and bilateral pectination of
sensilli. By these features, it can easily be
distinguished from all its known congeners. Apart
from the pectination, the following species are
similar to the new species in the shape of sensilli:

R. assimillis (Miheléié, 1956) [2] and R. tuberculata
Mahunka et Topercer, 1983 [3]. These species are
characterized as follows: 1) Notogastral and ven-
tral setae cilated, 2) setae c> absent, 3) mutual
distance of setae r> wider and 4) setae ad; situated
more anteriorly.

In Japan, one representative of this genus has
been known: R. sengbuschi Hammer, 1968 [4]
(= Oppia tokyoensis Aoki, 1974 [5]). This species
is easily distinguished from the new species by the
following additional characters: 1) Body smaller
and slender, 2) longitudinal line outside lamellar
seta distinct from dorsal view, 3) seta c2 absent, 4)
epimeral setae short, 5) mutual distance of aggen-
ital setae shorter and 7) adanal setae ad; preanal in
position.

ACKNOWLEDGMENTS

The author is indebted to Mr. T. Shimohata who
offered the specimens of the new species.

REFERENCES

1 Subias, L. S. (1980) Oppiidae del complejo “cla-
vipectinata-insculpta” (Acarida, Oribatida). Eos, 54:
281-313.

2 Miuihelcic, F. (1956) Oribatiden Stideuropas V. Zool.
Anz., 162: 154-174.

3 Mahunka, S. and Topercer, E. (1983) Some new
oribatids from Czechoslovakia (Acari). Folia ent.
hung., 44: 229-237.

4 Hammer, M. (1968) Investigations on the oribatid
fauna of New Zealand, with a comparison between
the oribatid fauna of New Zealand and that of the
Andes Mountains, South America. Part III. Biol.
Skr. Dan. Vid. Selsk., 16: 1-96, 33 pls.

5 Aoki, J. (1974) A new species of oribatid mite found
in the middle of Tokyo. Bull. Natn. Sci. Mus. Tokyo,
17: 283-285.

= oe if rive vee jou

~~:

ni Ipaiee 7 ng batenhe:: VR bm Yyotle a

Coo
‘Uers it :

Lasse “in

ie iii 25st
fer raed

ee
ie’ wiley “lear

ZOOLOGICAL SCIENCE 7: 985-987 (1990)

[COMMUNICATION]

© 1990 Zoological Society of Japan

Lasioglossum blakistoni sp. nov., the Northernmost
Representative of the Palaeotropic Subgenus
Ctenonomia (Insecta, Hymenoptera,
Halictidae)

SHOICHI F. SAKAGAMI and MeEryo MUNAKATA!

Zoological Section, Institute of Low Temperature Science,
Hokkaido University, Sapporo 060, and
‘1-40-12, Kaji, Hakodate 041, Japan

ABSTRACT— Lasioglossum (Ctenonomia) _ blakistoni
sp. nov. differs from most Ctenonomia species by smooth
and sparsely punctured mesoscutum. Attaining north-
ward Shimokita Peninsula, the northernmost area of
Honshu, this species is regarded as the northernmost
species of the large palaeotropic subgenus Crenonmia.

Ctenonomia Cameron is a large palaeotropic
subgenus of the enormous halictine genus Lasio-
glossum Curtis [1, 2] and represented in Japan only
by L. kumejimense (Matsumura et Uchida) con-
fined to S. E. Islands (Amami-Oshima to Iriomote
Is.). This paper briefly describes another consub-
gener from Honshu, leaving details elsewhere.

L. blakistoni sp. nov.

Fasily separated from most consubgeners by
smooth and very sparsely punctured mesoscutum.
L. callorhinum (Cockerell, 2, Thailand) [3] is
closest but PP (=punctures) finer (¢ 12-16 um),
WIS (=width of interspaces=IS)/¢ PP 0.5-3.0, IS
finely shagreened, propodeal dorsum more irregu-
larly ridged, and tergum 1 on disc with finer and
sparser PP (¢+4 um, WIS/¢ PP 4.0-6.0 or more)
(Ebmer, in litt.). L. compressum (Bliithgen, ?.,
Darjiling, Nepal) [4] has similar mesoscutal sculp-

Accepted May 9, 1990
Received March 27, 1989

ture but smaller (wing plus tegula 4.5 mm), head
shorter (width/length 35:31) and frons above
broadly smooth as on mesoscutum. L. xystonotum
(Vachal, 2, Sikkim to Sumatra) has mesoscutum
sparsely punctured but tessellate and differs from
L. blakistoni in many other features [2].

Female. Body 7.5mm, wing plus tegula 6.7
mm. Black; distitarsi, tergal margins, tegula,
mandible partly with brownish tint; mesoscutum
and terga with bluish enamel luster; veins brown,
stigma and subcosta darker. Hairs whitish, often
partly yellow-tinted, scutellum with sparse admix-
ture of dark hairs; tergum 1 without basal tomental
patches, T>_4 with basal fasciae (often hidden on
T,). Head width/length 52:51. Occiput not
carinate; face reticuloareolate, each areola+25
ym wide; supraclypeus with ¢ PP+25-35 um, IS
imperceptively etched and fairly shining, WIS+
1.0 marginally, 2.0-3.0 medially; clypeus above
similarly sculptured, below sparser and coarser;
clypeal tooth mild. Mesoscutum anteriorly not
projecting and not truncate, disc with ¢ PP 25-35
ym, IS very wide (WIS/ ¢ PP 2.0-5.0), smooth and
shining; mesoscutellum similar, mediolongitudi-
nally depressed. Propodeal dorsum coarsely longi-
tudinally ridged, IS smooth and shining; premargi-
nally crescently delimited; posterolaterally strong-
ly and rectangularly carinate, posterolateral corner
distinctly slanting, glabrous and tessellate; post-
erolateral carina continuous to posterior margin

986 S. F. SAKAGAMI AND M. MUNAKATA

but widely interrupted medially. As in other
consubgeners, inner hind tibial spur with ror De
oblique teeth and basal ones longer than width of
spur. Tergum 1 with basal slope smooth and
shining; disc smooth with fine PP (¢ 10m),
WIS>¢ PP; IS smooth, only marginally super-
ficially lineolate; T, with PP denser and coarser (¢
15 um), IS coriaceous.

Male. Similar to female. Clypeus marginally
narrowly yellow. Head width/length=50: 48.
Antenna attaining propodeum; flagellomeres long,
F, +2.0 times longer than wide. Mesoscutal and
mesoscutellar PP denser. Posterolateral corner of
propodeal dorsum more slanting, smooth and
shining; posterolateral carina weaker. Sternal
margins transverse and simple. Sternal hairs
simple and on S, not particularly dense.

Specimens examined (all from Honshu, Japan).
Holotype ? Aomori Pref., Shimokita Peninsula:
Zaimoku 20ix 1967, Y. Nakamura. Paratypes
Aomori Pref., Shimokita Pen.: Zaimoku (97 ¢ 30
v—27 1x, 130 16 vii-271ix 1967, Y. Nakamura),
Ikukoma (92 30v-4x, 59 14ix-4x 1967, S.
Nishimura), Sai (152, 6vi-17ix, 4¢ 9-171x
1968, S. Okada), Akagawa (2492 24 v-Six, 419
18 ix—5 x 1968, Y. Harada), Yagen (27% 8 vi-4 x,
9% 22 ix—4 x, 1969, M. Miyoshi), Okuyagen (2? 8
vi 1969, K. Kasajima); Tsugaru Distr. (all M.
Yamada): Nagasaka (22 19 ix 1987, Zatodishi (1 $
19 v 1948), Mt. Kudoji (2? 7 vi 1981), Dake (1
5 x 1980), Ohkawa (1 $ 22 ix 1985), Hisayoshi (1 ?
23 ix, 1 10 x 1980), Itadome (2 3 26 v 1984, 25 vi
1981); Tochigi Pref., Nikko (2¢ 101ix 1984, O.
Pellmyr); Gifu Pref., Tokuyama (1 10 x 1976, K.
Yamauchi), Takawashi (1 2 7x, 1¢ 20 ix 1974, K.
Okumura), Hida-Hagiwara (1¢ 17x 1974, K.
Okumura), Akigami (22 81x 1978, Y. Morimo-
to); Fukui Pref. (both Y. Haneda) Suwara, Ohno
(12 31 v 1981), N6go-Hakusan (16 14 ix 1982);
Kyoto Pref., Ashi (12 8 vi 1986, T, Kakutani),
Kibune (1% 8 x 1986, M. Kato).

Holotype and 5? 5 from Shimokita in En-
tomological Institute, Hokkaido University (Sap-
poro), 12 18 (% Zaimoku, 20ix 1967, 9
Tokuyama) in Ebmer collection (Linz, Austria),
72 6% from Shimokita, 3? from Tsugaru and all
specimens from prefectures other than Aomori in
Sakagami collection (Sapporo), all other Shimoki-

ta specimens in Munakata collection (Hakodate)
and all other Tsugaru specimens in Yamada
collection (Kuroishi).

This species is dedicated to T. Blakiston, a
British trader, who stayed in Hakodate (1863-
1883) and first pointed out the zoogeographical
importance of Tsugaru Strait between Honshu and
Hokkaido based on the distribution of higher
vertebrates [5]. Unlike mammals and birds the
distribution of many insect groups are less affected
by the Blakiston’s line. Among 48 non-parasitic
halictine bees known from Aomori Pref. [6, 8],
only three species are unknown from S. Hokkaido:
Lasioglossum (Evylaeus) japonicum (Dalla Tor-
re), L. (E.) hirashimae Ebmer et Sakagami and L.
(C.) blakistoni. Another species which has been
regarded as showing the same chorologic pattern,
Halictus (Seladonia) aerarius Smith, was recently
recorded from Tobetsu n. Hakodate (22, vi, viii
1987).

Being a palaeotronic group, Ctenonomia is only
sparsely distributed in the Palaearctics, mainly in
the southern marginal areas. The known north-
ernmost limit is S. Kansu from which L. (C.)
sinicum was described [7] based on a single male
(as Halictus sinicus). The exact locality is not given
but probably south of 40°N. Thus, L. blakistoni is
considered the northernmost representative of
Ctenonomia. The northern coast of Shimokita
Peninsula (+41.5°N), where L. blakistoni was
abundantly collected, latitudinally does not much
differ from southernmost Hokkaido. Neverthe-
less, no specimen of L. blakistoni was included
among over 29,000 halictine specimens collected
there during 1960-1989 by one of us (M.M.).
Thus, the northern limit of L. blakistoni is prob-
ably determined not climatically, and this species is
considered one of the rare halictine bees of which
chorologic pattern is limited by the Blakiston’s line
as a geographical barrier. The abundance of L.
blakistoni, the member of a tropical group, in
Shimokita Peninsula under a cool temperate cli-
mate is noteworthy. The area also represents the
northern limit of the Japanese honeybee Apis
cerana japonica and is famous as the world
northern limit of monkeys and apes by the
presence of the natural population of Macaca
fuscata.

Lasioglossum blackistoni sp. nov. 987

496-510.
ee DOMENTS BE @ockercliedi al) SAN (1937) eAmier, Mus. Nowite 950:
We thank P. A. W. Ebmer for his kind information on 1-12.
Lasioglossum callorhinum, and all collectors whose 4 Blithgen, P. (1926) Zool. Jahrb., Syst., 51: 375-698.
names are given in the text. 5 Blakiston, T. (1883) Trans. Asiatic Soc. Jpn., 11:
126-140.
6 Yamada, M. and Sakagami, S. F. (1988) J. Aomori-
Ken Biol. Soc., 25: 10-21.
IIE aN Gls 7 Bluthgen, P. (1934) Arkiv for Zoolgi., 27A (13) 1-
1 Michener, C. D. (1965) Bull. Amer. Mus. Nat. Hist., 23.
130: 1-362. 8 Yamada, M., Munakata, M. and Sakagami, S. F.

2 Sakagami, S. F. (1989) J. Kansas entom. Soc., 62: (1990) J. Aomori-Ken Biol. Soc., 27: 35-40.

' ia te ne

. ronpielaesh he ier aa

Ye sti Sslenen re

i % :
| | 7
es fs me ( D iecetioniae:
.
j
5
- ry - —5
| gor pia b= Bae Tos i
¥
at 4 heh , oti a Peet Lib al —
Ps 7 “a ‘
a a8 : rial
; ee ey gas || eh F Can bE Ny a S rtit ome hep has? TiaTpeTh) sf
th, Lane fs ; , 2 ¢ be es = Ti Tis 28 SDR eS a8 i
; My Swi? ree? se : z . " : “ : | 2
~\ Yo Wwess 4 an ial ae oe ae = : ; j , :
ul 7" S. . 4 2 < ~ j e é + 2° et re 1}
7 cay i ‘ t
> Ls %, heer " Ar Pe ey %, i ht hes «3 nen : Arh, " =
rj , : : ]
a ne ‘ “ j bh ;
: : hi ’ & ‘ fs ~, 7 a0 re U s
nae 2
e ~~ » 4 2 ;
c 4 j i ; A q Fee 3 i
7 : Lat
i}
~ i = ‘i
ae oe ‘i , :% pice
: x , a ‘ : :
: :
re
ve . 4 Hy Re ' = 5 be oe “EN é
2
= ‘ y a i : ; i
¥ ' : x ‘ f
_ . ; "
Wy : ; :
ke +: 3 Fate % ‘ rs ’ jet
= > = ;
= 2 Y ; |
= = :
~= i i
- i |
= Tt ty oi ; i . |
c j |
a q
Lay 3 ? eis | .
af - 4 S ;
3 % a t \
¢
= 5 |
= _ a ;
¥ - ,
) 2
_ ;
; |
eT i
A ¥
eae : -
} eF |
~ : ; ;
7
5 ° 4
i
re , J
ms
| = baa
_ os |
= ve
—
’ 7 ‘
7 x
4
t
1 ;
- \
y
-
(

D eve / O m e Nn f Published Bimonthly by the Japanese Society of
Pp Developmental Biologists
Distributed by Business Center for Academic

Growth & Differentiation Societies Japan, Academic Press, Inc.

Papers in Vol. 32, No. 5. (October 1990)

51. REVIEW: N. Osumi-Yamashita and K. Eto: Mammalian cranial neural crest cells
and facial development

52. R.D. Burke: Ontogeny of an extracellular matrix component of sea urchins and its
role in morphogenesis

53. T. Shimizu, H. Kinoh, M. Yamaguchi and N. Suzuki: Purification and characteriza-
tion of the egg jelly macromolecules, sialoglycoprotein and fucose sulfate glycoconju-
gate, of the sea urchin, Hemicentrotus pulcherrimus

54. A. Fujiwara, K. Kataoka, K. Mikami-Takei, E. Tazawa and I. Yasumasu: Vegeta-
lization induced by procaine and tetracaine in sea urchin embryos

55. M. Sakai: Heat-induced reversal of dorsal-ventral polarity in Xenopus eggs

56. H. W. M. van Straaten, K. C. Hooper and M. Bernfield: Hyaluronan disappears
intercellularly and appears at the basement membrane region during formation of
embryonic epithelia

57. B. Sadaghiani and J. R. Vielkind: Explanted fish neural tubes give rise to
differentiating neural crest cells

58. H. Shirai, N. Hosoya, T. Sawada, Y. Nagahama and H. Mohri: Dynamics of
mitotic apparatus formation and tubulin content during oocyte maturation in starfish

59. S. Fujiwara and N. Satoh: Patterns of segregation of mitochondria into muscle
lineage cells during embryogenesis of the ascidian Halocynthia roretzi

60. S. Asada, C. Tateishi and T. Nakazawa: Mitochondrial protein synthesis during
embryonic development of cricket

61. I. Mabuchi and H. Takano-Ohmuro: Effects of inhibitors of myosin light chain
kinase and other protein kinases on the first cell division of sea urchin eggs

62. E. J. Carroll, Jr., R. L. Heath and K. Izuishi: Extreme toxicity of ethanol during
activation of sea urchin eggs

Development, Growth and Differentiation (ISSN 0012-1592) is published bimonthly by The Japanese Society of
Developmental Biologists, Department of Developmental Biology, Mitsubishi Kasei Institute of life Science,
Minami-ootani 11, Machida, Tokyo 194, Japan. 1989: Volume 31. Annual subscription for Vol. 32, 1990: U. S.$
148,00, U.S. and Canada: U. S. $ 163,00, all other countries except Japan. All prices include postage, handling
and air speed delivery except Japan. Second class postage paid at Jamaica, N.Y. 11431, U.S.A.

Outside Japan: Send subscription orders and notices of change of address to Academic Press, Inc., Journal
Subscription Fulfillment Department, 1 East First Street, Duluth, MN 55802, U.S.A. Send notices of change of
address at least 6-8 weeks in advance. Please include both old and new addresses. U.S.A. POSTMASTER:
Send changes of address to Development, Growth and Differentiation, Academic Press. Inc., Journal Subscription
Fulfillment Department, 1 East First Street, Duluth, MN 55802, U.S. A.

In Japan: Send nonmember subscription orders and notices of change of address to Business Center for
Academic Societies Japan, 16-3, Hongo 6-chome, Bunkyo-ku, Tokyo 113, Japan. Send inquiries about
membership to Business Center for Academic Societies Japan,4—16, Yayoi 2-chome, Bunkyo-ku, Tokyo 113, Japan.

Air freight and mailing in the U.S. A. by Publications Expediting, Inc., 200 Meacham Avenue, Elmont, NY
11003, U.S.A.

NARISHIGE

THE ULTIMATE NAME IN MICROMANIPULATION

OUR NEW MODELS WR-88 and MO-102M
MAKE PRECISION MICROMANIPULATION SO EASY!

SOME FEATURES of THE WR-88 WATER ROBOT MICROMANIPULATOR (3-DIMENSIONAL)
* Drift-free, the new WR-88 has a DRIFT movement of less than 2 microns.
* The new WR-88 has a SMOOTH MICRODRIVE MECHANISM.
* An Aqua Purificate remote control ensures totally vibration-free operation.

THARISHIGE) NARISHIGE SCIENTIFIC INSTRUMENT LAB.

\ debe | 9-28 KASUYA 4-CHOME SETAGAYA-KU, TOKYO 157, JAPAN
PHONE (INT-L) 81-308-8233, FAX (INT-L) 81-3-308-2005

CABLE : NARISHIGE LABO, TELEX, NARISHIGE J27781

(Contents continued from back cover)

Hayashi, S.: Social condition influences sexu-
al attractiveness of dominant male mice

Chiba-Y.. \Y. Yamamoto, ©: Shimizu; M.
Zaitsu, M. Uki, M. Yoshii and K. Tomioka:
Insemination-dependent modification of
circadian activity of the mosquito, Culex
PEMCMSEDONICHIS on So) oe tes wos ernie bw oly sions 895

Yamanouchi, K.: Role of the medulary raphe
nucleus in regulating sexual behaviors in
RSIRAIS TS A ea eae ease 907

Morphology —

Atoji, Y. and Y. Suzuki: Apocrine gland of
the infraorbital gland of the Japanese serow,
GADCORMUS CHISPUS oo ao ca vse tees sou tees 913

Meyer, W. and A. Tsukise: Structural and
carbohydrate histochemical aspects of the
snout skin of the opossum, Didelphis virgi-
FANART tere. ws so Saas peas Was lew ane be 923

Taxonomy
Shimazu, T.: Trematodes of the genus Orien-
tocreadium (Digenea: Orientocreadiidae)
from freshwater fishes of Japan .......... 933
Higgins, R. P. and Y. Shirayama: Dracoderi-
dae, a new family of the cyclorhagid
Kinorhyncha from the Inland Sea of Japan

Mukai, H.: Systematic position of Stephanella
hina (Bryozoa: Phylactolaemata), with spe-
cial reference to the budding site and the
attachment of sessoblasts

Hirayama, A.: A new spepcies of the genus
Paramoera (Crustacea: Amphipoda) from
the intertidal zone of Hokkaido, northern
DAD ANN Parone ee ests notes oie, ee eorepe oneal or adroit 955

Ohkubo, N.: A new species of Ramusella
(Acari: Oribatei) from Japan

Sakagami, F. S. and M. Munakata: Lasiog-
lossum blackstoni sp. nov., the northernmost
representative of the palaeotropical sub-
genus Cfenonomia (Insecta, Hymenoptera,
FPANCH GAG) 2, atic acmactenre iket nue a 40% 985

ZOOLOGICAL SCIENCE

VOLUME 7 NUMBER 5

OCTOBER 1990

CONTENTS

REVIEWS

Omura, Y., K. Horst-W., M. Oguri and A.
Oksche: Properties of the blood-brain and
blood-cerebrospinal fluid (CSF) barrier in

the circumventricular organs of _ the

diencephalic roof of teleosts” "ae.-45.44547 783
Kobayashi, M. and Y. Muneoka: Structure
and action of molluscan neuropeptides ...801

ORIGINAL PAPERS

Physiology

Burton, D. and B. A. Everard: In vitro charac-
teristics of K* and Na* induced melano-
phore responses in a cold ocean teleost,
Pseudo-pleuronectes americanus
Toh, Y.: Diurnal structural changes in rhab-
domeric microvilli of the compound eye in
the blue crab, Callinectes sapidus (COM-
MUNICATION)
Uchiyama, M., T. Ogasawara, T. Hirano, S.
Kikuyama, Y. Sasayama and C. Oguro:
Serum and urine osmolyte concentrations
during acculimation to various dilutions of
seawater in the crab-eating frog, Rana
cancrivora (COMMUNICATION)

Cell Biology

Devi, S. A., S. Kan and S. Kawashima:
Effect of culture age on lipofuscin accumula-
tion and creatine phosphokinase activity in
spontaneously beating rat heart cells, and its
modifications by tocopherol

Biochemistry
Suemitsu, T., Y. Tonegawa and K. Ishihara:
Similarities between the primary structures
of exogastrula-inducing peptides and peptide
B purified from embryos of the sea urchin,

Anthocidaris crassispinag ~~ +. eee 831
Developmental Biology
Kurabuchi, S. and Y. Kishida: Comparative

study of the influence of head and tail grafts
on axial polarity in regeneration of the fresh-
water planarian
Fujishima, M., K. Nagahara and Y. Kojima:
Changes in morphology buoyant density and
protein composition in differentiation from
the reproductive short form to the infectious
long form of Holospora obtusa, a macro-
nucleus-specific symbiont of the ciliate Para-
mecium caudatum
Aoki, K., M. Nakamura, H. Namiki, S. Oki-
naga and K. Arai: The effect of glucose
and phosphate on mouse two-cell embryos
to develop in vitro (COMMUNICATION)

Reproductive Biology
Takahashi, N., N. Sato, N. Ohtomo, A.
Kondo, M. Takahashi and K. Kikuchi:
Analysis of the contraction-inducing factor
for gonadal smooth muscle contraction in sea
urchin
Dukelow, W.R., C. S. T. Pow, J. H. Kennedy
and L. Martin: Stress effects on late preg-
nancy in the flying-fox, Pteropus scapulats

Endocrinology
Kato, Y., T. Ezaki, T. Hirai and T. Kato:
Strain difference in nucleotide sequences of
rat glycoprotein hormone subunit cDNAs
and gene fragment

Behavior Biology

(Contents continued on inside back cover)

INDEXED IN:
Current Contents/LS and AB & ES,
Science Citation Index,
ISI Online Database,
CABS Database, INFOBIB

Issued on October 15
Printed by Daigaku Letterpress Co., Ltd.,
Hiroshima, Japan

: December 1990 —
”

Tal

An International Journal

Proceedings of the

Sixty-First Annual Meeting of the
Zoological Society of Japan
October 3-5, 1990

Niigata

Vol. 7 No. 6 December 1990
published by Zoological Society of Japan

Business Center for Academic Societies Japan

Betabued by VSP, Zeist, The Netherlands

ISSN 0289-0003

gripe FM lee 3

nn bert? Pe
ie *” ZOOLOGICAL SCIENCE

\ The Official Journal of y Zoological Society of Japan

me SAAR ED oo?
Ea Asa Ten ae The Zoological Society of Japan:
BENCHED LSE (Tokyo) Toshin-building, Hongo 2—27-2, Bunkyo-ku,
Piideshi Kopayastt 7 (rekyo) Tokyo 113, Japan. Tel. (03) 814-5675
Managing Editor: @ihcere

Chitaru Oguro (Toyama) President: Hiromichi Morita (Fukuoka)
Assistant Editors: Secretary: Hideo Namiki (Tokyo)
Yuichi Sasayama (Toyama) Treasurer: Tadakazu Ohoka (Tokyo)

Hitoshi Michibata (Toyama) Librarian: Masatsune Takeda (Tokyo)
Miéko Komatsu (Toyama) oo ae

Editorial Board:

Howard A. Bern (Berkeley) Walter Bock (New York) Aubrey Gorbman (Seattle)
Horst Grunz (Essen) Robert B. Hill (Kingston) Yukio Hiramoto (Chiba)
Susumu Ishii (Tokyo) Yukiaki Kuroda (Mishima) John M. Lawrence (Tampa)
Koscak Maruyama (Chiba) Roger Milkman (Iowa) Kazuo Moriwaki (Mishima)
Richard S. Nishioka (Berkeley) Tokindo S. Okada (Okazaki) Andreas Oksche (Giessen)
Hidemi Sato (Nagano) . Hiroshi Watanabe (Tokyo) Mayumi Yamada (Sapporo)

Ryuzo Yanagimachi (Honolulu)

ZOOLOGICAL SCIENCE is devoted to publication of original articles, reviews and communications
in the broad field of Zoology. The journal was founded in 1984 as a result of unification of Zoological °
Magazine (1888-1983) and Annotationes Zoologicae Japonenses (1897-1983), the former official
journals of the Zoological Society of Japan. ZOOLOGICAL SCIENCE appears bimonthly. An annual
volume consists of six numbers of more than 1200 pages including an issue containing abstracts of papers
presented at the annual meeting of the Zoological Society of Japan.

MANUSCRIPTS OFFERED FOR CONSIDERATION AND CORRESPONDENCE CONCERN-
ING EDITORIAL MATTERS should be sent to:
Dr. Chitaru Oguro, Managing Editor, Zoological Science, Department of Biology, Faculty of Science,
Toyama University, Toyama 930, Japan, in accordance with the instructions to authors which appear in
the first issue of each volume. Copies of instructions to authors will be sent upon request.

SUBSCRIPTIONS. ZOOLOGICAL SCIENCE is distributed free of charge to the members, both
domestic and foreign, of the Zoological Society of Japan. To non-member subscribers within Japan, it is
distributed by Business Center for Academic Societies Japan, 6—16—3 Hongo, Bunkyo-ku, Tokyo 113.
Subscriptions outside Japan should be ordered from the sole agent, VSP, Utrechtseweg 62, 3704 HE
Zeist (postal address: P.O. Box 346, 3700 AH Zeist), The Netherlands. Subscription rates will be
provided on request to these agents. New subscriptions and renewals begin with the first issue of the
current volume.

All rights reserved. No part of this publication may be reproduced or stored in a retrieval system in any
form or by any means, without permission in writing from the copyright holder.
© Copyright 1990, The Zoological Society of Japan

Publication of Zoological Science has been supported in part by a Grant-in-Aid for
Publication of Scientific Research Results from the Ministry of Education, Science
and Culture, Japan.

ZOOLOGICAL SCIENCE 7: 989-998 (1990)

EMITHSON aw
JUN 27 199]
Ww

LIBRARIES

REV

Morphological Aspects of Ascidian Fertilization

MAKOTO FUKUMOTO

Biological Institute, College of General Education,
Nagoya City University, Nagoya 467, Japan

ABSTRACT—The spermatozoa of ascidians have an acrosome(s), albeit a small one. In Ciona
intestinalis, an acrosome reaction occurs via vesiculation in fundamentally the same way as observed in
mammalian spermatozoa. After the acrosome reaction, apical processes protrude from the apex of
the sperm head in the perivitelline space. Gamete fusion takes place between some of these apical
processes and egg plasmalemma, resulting in the incorporation of the sperm into the egg from the
anterior tip of its head in the same way that it does in other marine invertebrates. Fertilization in C.

© 1990 Zoological Society of Japan

intestinalis has characteristics of both mammals and marine invertebrates.
fertilization which has not been previously described.

ascidian fertilization.

INTRODUCTION

From the phylogenetic view point, the progeni-
tors of ascidians might have been in an important
position in an evolutionary pathway from inverte-
brates to vertebrates. For this reason, we can
expect the ascidians to provide us with many
challenging and attractive problems in the various
fields in biology. Studies on fertilization in asci-
dians have shown that there is a series of events
which are probably necessary for the successful
achievement of gamete fusion between the sperm
and egg. Many of the biochemical events associ-
ated with ascidian fertilization are described in two
recent reviews [1, 2]. This review will focus
morphological aspects of fertilization, mainly in
Ciona intestinalis.

The Acrosome of Ascidian Spermatozoa

The differentiated spermatozoon of Ciona intes-
tinalis is schematically illustrated in Figure 1.

There are a number of papers which have ques-
tioned the presence of an acrosome in differenti-

Received May 9, 1990

This is a new type of
Some speculations are also presented on

ated ascidian spermatozoa: in Corella parallelog-
ramma [5], in Ascidia malaca |6], in Ciona intesti-
nalis [7], in Halocynthia roretzi [8], in both Molgu-
la impura and Styela plicata [9], in Microcosmos
sabatieri [10] and Ascidia malaca, Ascidiella asper-
sa and Phallusia mammillata [11].

Cloney and Abbott [12] first clearly showed
membrane bound vesicle(s) which they referred to
as the putative acrosome in the anterior tip of the
sperm head in Ascidia callosa. They argued that
the claim that the spermatozoa of ascidians lack an
acrosome should be reconsidered. Before the
discovery of a putative acrosome in A. callosa,
there had been several papers which reported the
presence of an acrosome in ascidian spermatozoa:
in Ascidia nigra [13], in Diplosoma listerianum and
Lissoclinum pseudoleptoclinum [14], in Polysy-
ncraton lacazei and Trididemnum cereum [15] and
in C. intestinalis {16, 17]. However, these papers
were not convincing because they did not demons-
trate the existence of a membrane bound acrosom-
al vesicle.

The view that ascidian spermatozoa have no
acrosome can be attributed to the difficulty of
getting good fixation and to the greatly reduced
size of the ascidian acrosome [12]. Cotteli et al.

990 M. FuKUMOTO

Schematic Illustration of Ciona intestinalis Spermatozoon

Fic. 1.

The spermatozoon of C. intestinalis has architectural features that are characteristic of ascidian spermatozoa

which have been previously described [12, 24]. It has an elongated head (approximately 4 ~m long) with a
wedge-shaped tip and a mitochondrion which is closely applied laterally to the nucleus. An acrosome (A) is
present at the anterior region of the head, which appears as a flattened vesicle (about 150 nm x 160 nm x 60 nm).
At the anterior-most tip of the head, apical substance (AS) which is an accumulation of granular material,
approximately 5—7 nm in diameter, is recognized. Fuzzy materials (surface ornamentation) decorate the external
surface of the plasmalemma enclosing the head. Nuclear pores (NP) are present. They appear sometimes in the
anterior region of the head. Tubular structures (TS), approximately 200 nm in diameter, are present in the
mitochondrion just inside the inner membrane running antero-posteriorly. Such mitochondrial inclusions have
been reported in some species of ascidians [3, 4, 23, 69, 71]. N, nucleus; T, tail.

[18] reported that an acrosome existed in C. intesti-
nalis sperm. They found vesicle(s) at the apex of
spermatids in spermiogenesis. Unfortunately, they
did not clearly show an acrosomal vesicle in ma-
ture spermatozoa. Subsequently a membrane
bound vesicle(s) has been found in the anterior
region of the spermatozoa of a number of species
of ascidians (Table 1).

Recently, Lambert and Koch [2] have argued
that the vesicles in ascidian spermatozoa should be
called “apical vesicles” until the criteria for posi-
tive identification as acrosomes can be more close-
ly met.

However, if a vesicle(s) is present in the proper
location for an acrosome, this would be one kind
of evidence for the existence of an acrosome,
because acrosomes are anteriorly located vesicles
in the sperm of other animals. Furthermore, the
vesicle in ascidian spermatozoa is not an oddity
which is only found in one or two species, but a
feature of all species examined in the Orders of
Pleurogonan and Enterogonan ascidians (Table 1).
For these reasons, the vesicles in ascidian sperma-
tozoa have been referred to as “acrosome” [20, 24,
DSi

Acrosome Differentiation

During acrosome differentiation in other animal
species, proacrosomal vesicles derived from the
Golgi complex(es) coalesce to form one acrosomal
vesicle [26-30]. During spermiogenesis in Halocy-
nthia roretzi, Kubo et al. [8] have reported that the
acrosomal vesicle is formed from coalescing Golgi
derived vesicles. However, it is difficult to con-
clude from their photographs that an acrosome is
actually formed by the fusion of the Golgi vesicles.
They were not able to demonstrate the existence of
an acrosome in mature sperm at the completion of
spermiogenesis.

During spermiogenesis in Styela plicata, Pyura
haustor [23], Molgula manhattensis [31] and in
Diplosoma macdonaldi (Fukumoto, submitted),
small vesicles appear in the blister at the anterior
region of spermatids and fuse to form an acrosome
during further differentiation. In M. manhattensis,
vesicles (approximately 50 nm in diameter) which
are at least three or four in number appear in the
blister at the apex of the spermatid in early stages
of spermiogenesis. They fuse with each other to
form an acrosomal vesicle during middle stages of

Fertilization in Ascidians 991
TABLE 1. Ascidian species with acrosome(s) examined
Order Suborder Family Genus and species References
Enterogona aplousobrachia Clavelinidae Clavelina huntsmani [«]
Clavelina picta [+]
Distaplia occidentalis [*]
Polyclinidae Aplidium californicum [*]
Didemnidae Diplosoma macdonaldi [+]
Phlebobranchiata Cionidae Ciona intestinalis [19, 20]
Ciona savigney [*]
Perophoridae Perophora formosana [*]
Perophora annectens [«]
Ecteinascidia turbinata [=]
Corellidae Corella pacifica [«]
Ascidiidae Ascidia nigra [*]
Ascidia mentula [*]
Ascidia callosa [12]
Phallusia mammillata Zils 24
Pleurogona Stolidobranchiata styelidae Styela plicata [23]
Styela clava [*]
Cnemidocarpa finmarkiensis [23]
Botryllus schlosseri [23]
Metandrocarpa taylori [«]
Pyuridae Pyura haustor [23]
Pyura vittata [«]
Boltenia villosa [24]
Herdmania momus [24]
Halocynthia roretzi [=]
Molgulidae Molgula manhattensis [24, 31]

* Fukumoto unpublished.

spermiogenesis. The acrosome is_horseshoe-
shaped just after fusion and rounds up during
further differentiation. The fully differentiated
acrosome is approximately 80x 80x40 nm and is
positioned at the apex of the sperm head. The
differentiation of the acrosome in M. manhattensis
is schematically illustrated in Figure2. At in-
termediate stages of spermiogenesis in M. manhat-
tensis, the vesicles in the blister attach to the inner
surface of the blister at essentially the same posi-
tional level [31]. This suggests that the inner
surface of the blister functions as a scaffold for the
vesicles. This has led me to postulate that by the
middle spermiogenesis, a specialized area which
has the property of binding specifically with the

vesicles and which functions as a scaffolding for
them differentiates in the plasmalemma enclosing
the blister. This area encircles the blister at a
definite positional level as a band as shown in
Figure 3. During intermediate stages of sper-
miogenesis, vesicles make contact with the inner
surface of this band. This band confines the field of
vesicle movement, making it easier for them to
recognize each other, and/or this band generates a
contractile force which causes the vesicle to move
along the inner surface of the band. After contacts
are made between these vesicles, they fuse to each
other on the inner surface of this band resulting in
the formation of a horseshoe-shaped acrosome
which transforms into a sphere during further

992 M. FuKUMOTO

Fic. 2. Schematic illustration showing the differentia-
tion of the acrosome in M. manhattensis. In early
spermatide (A, B), the plasmalemma at the apex
expands to form a blister which is covered with fuzzy
extracellular material. Vesicles (50-60 nm in dia-
meter) are present in the blister. They are mod-
erately electron-dense and at least three or four in
number (a, b). Midway through spermiogenesis (C,
D), these vesicles attach to the inner surface of the
plasmalemma enclosing the blister (c). These vesi-
cles, thereafter, fuse with each other along the inner
surface of the plasmalemma and form a horseshoe-
shaped acrosomal vesicle (d1) which transforms into
a sphere (d2). In mature spermatozoa (E), the
acrosome is a slightly depressed sphere positioned at
the apex (e). The illustrations marked by small
letters represent the transverse sections at the level
indicated in those marked by capital letters, respec-
tively (Fukumoto [31]).

| \

Fic. 3. Schematic illustration showing a hypothetical
scaffold which facilitates vesicle contact and fusion.
See details in text (Fukumoto [31]).

development [31].

With respect to the origin of the vesicles in the
blister, no direct evidence has been found to show
that they are derived from the Golgi apparatus. In
view of the fact that early spermatids have a fairly
well developed Golgi apparatus and Golgi derived
vesicles, it seems safe to assume that they are
derived from Golgi vesicles [23, 31].

The Acrosome Reaction

In some marine invertebrates, the acrosome
reaction occurs via exocytosis of acrosomal subst-
ance followed by the formation of one of more
acrosomal processes which are responsible for the
fusion of the sperm with the egg plasmalemma.
Exocytosis occurs through an opening formed by
the fusion between the acrosomal outer membrane
and the plasmalemma enclosing the acrosome [32-
34].

In Ciona intestinalis, | have reported acrosome
fragmentation, which has been referred to as an
acrosome reaction [19]. It became clear, however,
that I mistook the apical substance, which some-
times appears as small vesicles, for acrosome frag-
mentation, because I had overlooked the presence
of the apical substance at that time [20].

It has been suggested that the acrosome reaction
in C. intestinalis is similar to that of mammalian
spermatozoa, mainly because it seems to involve

Fertilization in Ascidians 993

Fic. 4. Possible explanation of the morphological
changes during the acrosome reaction in C. intestina-
lis spermatozoa. 1. An intact acrsome is present at
the apex of the head. The acrosomal outer mem-
brane is tightly apposed to the overlying plas-
malemma (arrow). Apical substance (AS) is located
at the anterior-most tip of the head. N, nucleus. 2.
The acrosome reaction occurs through the fusion
between the acrosomal outer membrane and the
overlying plasmalemma (arrow). 3. Membrane fu-
sion proceeds along the peripheral margin of the
acrosome, which causes vesiculation. 4. Vesicula-
tion is completed. Thin connections are found
between the vesicle and the inner surface of the
acrosomal inner membrane. 5. The vesicle has come
off. The acrsomal inner membrane is exposed and is
contiguous with the plasmalemma enclosing the
apex (Fukumoto [25]).

just the exocytosis of the acrosomal substance [35].
Unfortunately, membrane fusion between the
acrosome and the overlying plasmalemma has not
been clearly shown.

Recently, it has been found that caffeine or
theophylline induces morphological changes in the
acrosome of C. intestinalis spermatozoa, which I
have referred to as an acrosome reaction [25]. An
acrosome reaction occurs by fusion between the
acrosomal outer membrane and the plasmalemma
enclosing the acrosome. The fusion seems to
proceed along the peripheral margin of the acro-
some, resulting in vesiculation [25]. Figure 4
provides an interpretation of the acrosome reac-
tion in C. intestinalis.

Vesiculation typically occurs during the acro-
some reaction in mammalian spermatozoa [36-
38]. In this respect, the acrosome reaction in C.
intestinalis spermatozoa is similar to the acrosome
reaction of mamalian spermatozoa [25].

The claims 1) that the acrosome reaction, which
consists of the break down of the plasmalemma at
the tip of the head, the opening of the acrosomal
vesicle, the blebbing of the acrosomal inner mem-
brane and the formation of tubules which make
contact with the fibrillar network of the chorion
occurs at the surface of the chorion after species
specific binding of the spermatozoa [1, 16, 17, 39-
41], and 2) that the acrosome reaction is triggered
in vitro by a fucosyl-containing glycoprotein com-
ponent extracted from the chorion [42, 43] have
been made for C. intestinalis. However, these
claims are not convincing because they are not
backed up by convincing pictures documenting the
acrosome reaction.

Fertilization in Ascidians

The binding of the spermatozoa to the chorion
involves the tip of the sperm head [16, 17, 44] or
the fuzzy material on the plasmalemma of the
sperm apex [20, 45] and the outer fibrous struc-
tures of the chorion. There is evidence that this
binding is mediated by an enzyme-substrate reac-
tion between a-L-fucosidase on spermatozoa and
fucosylated glycoproteins located on the fibrous
structures of the chorion [46].

With the electron microscpe, it is difficult to

994 M. FukKUMOoTO

detect gamete fusion and morphological changes in
spermatozoa during the process of fertilization,
especially gamete fusion, because of the difficulty
of getting polyspermy, even when inseminating
with a high concentration of sperm. Recently, it
has been found that polyspermy can be induced in
Ciona intestinalis, when caffeine- or theophylline-
treated eggs are inseminated by fairly high concen-
trations of spermatozoa [20]. This has provided us
with useful material for observing how gamete
fusion occurs and any structural changes taking
place in C. intestinalis spermatozoa prior to gamete
fusion in the perivitelline space.

An acrosome reaction (vesiculation) occurs on
the surface of the chorion or after the penetration
of the sperm apex through the chorion in the
perivitelline space [45]. In the pervitelline space,
several processes protrude from the apex of the
sperm head in both cases of normal and polysper-
mic fertilization [20, 45]. These processes typically
are about 100 nm in length and 40 nm in diameter
[20]. In other marine invertebrates, the plas-
malemma enclosing the acrosomal process(es) are
derived from the acrosomal inner membrane [32].
However, it is not clear that the plasmalemma
enclosing the process in C. intestinalis is derived
from the acrosomal membrane, because some of
the processes protrude from a region which is a fair
distance from the former location of the acrosome.
For this reason, these processes have been desig-
nated as “apical process” [20, 45]. Gamete fusion
between the sperm and the egg plasmalemma
occurs by means of the apical process(es), resulting
in the incorporation of the spermatozoon into the
egg cytoplasm beginning at the anterior tip of the
sperm head [20, 45]. This suggests that the apical
process is functionally homologous to the acrosom-
al process of other marine invertebrates. In
marine invertebrates, the formation of acrosomal
process(es) is due to the rapid polymerization of
actin molecules which are stored as a subacrosomal
or periacrosomal substances [32-34, 47, 48] or by
rapid projection of previously polymerized actin
filaments [49, 50]. With respect to apical process
formation, it is worth mentioning that actin is
probably present in the apex of the sperm head in
Boltenia villosa and Cnemidocarpa finmarkiensis
[51], although negative results have been obtained

in Ascidia ceratodes [51] and in Phallusia mammil-
lata (22).

In C. intestinalis spermatozoa, an accumulation
of electron-dense material [41] or granular mate-
rial [20], about 5-7 nm in diameter, is observed at
the anterior-most tip of the head, which is referred
to as apical substance [20]. Such an accumulation
of electron-dense substance was first reported at
the apex of the differentiated spermatozoa in
Ascidia callosa, where it was assumed to corres-
pond to the periacrosomal substance found in
some animal species [12]. Further morphological
and immunohistochemical studies are needed to
clarify the precise role of the apical substance.

The apical processes have never been observed
at the apex of spermatozoa in which acrosome
reaction has been induced by caffeine [25]. It
seems safe to assume that the factor(s) that elicit(s)
the structural changes at the apex of spermatozoa
exist(s) in the perivitelline space of C. intestinalis
[20].

In P. mammillata, several vesicles (up to eight)
have been found at the apex of the sperm head 5
22]. It has been suggested that some of these
vesicles fuse with the overlying plasma membrane —
and release their contents prior to passage of the
sperm through the chorion and their remains could
be recognized in the spermatozoa in the peri-
vitelline space. Honegger [22] speculated that
these vesicles contained chorion lysin(s) and/or
additional enzymes that participated in the process
of sperm-egg fusion. It has been suggested that
sperm-egg fusion in P. mammillata occurs between
the plasmalemma of the post-acrosomal region of
the sperm head and the egg membrane, as
observed in mammalian fertilization [21, 22]. In C.
intestinalis, in contrast to the findings in P. mam-
millata, the acrosome reaction occurs via vesicula-
tion, in fundamentally the same manner as
observed in mammalian spermatozoa [25] and
gamete fusion takes place between the apical pro-
cesses of the sperm head and egg plasmalemma,
resulting in the incorporation of the spermatozoa
into the egg from the anterior tip of the sperm
head, in the same way that it does in other marine
invertebrates [20, 45]. The suggestion was made
that fertilization in C. intestinalis has characteris-
tics of both mammals and marine invertebrates

Fertilization in Ascidians 995

Fic. 5. A half-SEM (Bar indicates 100 «m) and half-drawing representation explaining the morphological aspects of
fertilization in C. intestinalis. The egg is enclosed by a chorion (Ch) to which spermatozoa approach through the
cleft between follicle cells [52] and bind by their anterior tip as a prerequisite step for fertilization including self
and non-self recognition [16, 17, 40, 46, 53, 54]. A single layer of highly vacuolated follicle cells (FC) which are
thought to be involved in sperm attraction [55] adheres to the outside of the chorion. Test cells (TC) which are
still enigmatic in their role at fertilization are located in the perivitelline space close to the inner surface of the
chorion.

(1) The spermatozoon binds to the surface of the chorion by the anterior tip of its head; (2) The acrsome
reaction occurs on the surface of the chorion or after the penetration of the sperm apex through the chorion. A
mitochondrion is cast away prior to penetration of the chorion (see [56, 57] for mitochondrial translocation or
shedding); (3) In the perivitelline space, apical processes protrude from the apex of the sperm head; (4) Gamete
fusion takes place between some of the apical processes and egg plasmalemma, resulting in the incorporation of
the sperm from the anterior tip of its head. A, acrsome; ApP, apical process; AS, apical substance; M,
mitochondrion; PvS, perivitelline space.

[25]. Figure 5 summarizes and provides a morpho-
logical interpretation of the events during fertiliza-
tion in C. intestinalis. The ascidian egg is enclosed by a relatively thick
and tough noncellular chorion (vitelline coat) com-
posed of a thick network of interwoven microfibrils
[17]. This presents a barrier to penetration by the

Some Speculations on Ascidian Fertilization

996

M. FUKUMOTO

TABLE 2. Acrosome and head volumes®* and their ratio in some echinoderm and ascidian spermatozoa

(Fukumoto [24])

Volume (2°) of

Species Acrosome (A) Head (H) A/JH (%) References

Echinodermata

Thyone briareus 0.10 1.0 10.0 [34]
Marthasterias gracilis 0.11 1.0 11.0 [48]
Pisaster orchraceus 0.16 1.0 16.0 [48]
Asterias forbesii 0.08 1.0 8.0 [66]
Strongylocentrotus purpuratus 0.03 1.4 Deak [67]
Ascidia

Pyura haustor 0.0004 0.84 0.05 [23]
Styela plicata 0.001 0.64 0.16 [23]
Styela clava 0.001 0.64 0.16 [24]
Cnemidocarpa finmarkiensis 0.001 0.50 0.20 [24]
Botryllus_ schlosseri 0.0005 1.10 0.05 [24]
Boltenia villosa 0.0008 0.64 0.12 [24]
Herdmania momus 0.001 0.65 0.15 [24]
Molgula manhattensis 0.0005 0.36 0.14 [24]

* All figures are estimated based on the illustrations in the references cited.

spermatozoa. They might utilize a chorion lysin(s)
in order to make a hole in the chorion. Proteases
have been detected in Ciona intestinalis spermato-
zoa [7]. Further evidence that spermatozoa con-
tain proteases which are neccessary for fertilization
in Halocynthia roretzi has been obtained [58-64].
At present, however, it is not clear where these
proteases are localized in spermatozoa. In some
marine invertebrates and mammals, the lysin(s)
are thought to be stored in the acrosome [38, 65].
The acrosome in ascidian spermatozoa has a great-
ly reduced size. Table 2 shows a comparison of
acrosome size in ascidians with that of some echi-
noderms. In spite of the fact that the ascidian egg
has a relatively thick chorion, the volume of the
acrosome in ascidians in extremely small compared
to that in echinoderms. The volume ratio (%) of
the acrosome to the head is roughly estimated to
be in the range of 0.1-0.2% for ascidians, while it is
in the rage of 2.5-16% for echinoderms. Furth-
ermore, the acrosome in ascidians is not as com-
pact as it is in other organisms [24]. These facts
suggest that the acrosome in ascidian spermatozoa
is not a major place for storing lysin(s). During the
process of fertilization in C. intestinalis, the acro-
some reaction occurs on the surface of the chorion

or after penetration of the sperm apex through the
chorion [20, 45]. This suggests that the acrosome
reaction is not required for passage through the
chorion [45]. I have suggested that the poorly
developed ascidian acrosome might react at an
appropriate step in fertilization and participate
mainly in the fusion of the gamete _plas-
mamembranes, and that the chorion lysin(s) are
intercalated into the plasmalemma enclosing the
anterior region of the sperm head [23]. In this
context, we should pay attention to the fact that
the plasmalemma enclosing the sperm head is
externally decorated by a fuzzy extracellular mate-
rial or surface ornamentation [7, 19, 20, 23, 24, 31,
68, 69]. It is the fuzzy extracellular material at the
apex of spermatozoa, which first makes contact
with the chorion at fertilization [19, 20, 45]. In C.
intestinalis spermatozoa, the region responsible for
the binding to the chorion is a Con A binding site
which is exclusively restricted to the plasmalemma
at the tip of the sperm head [16, 17, 70]. This
suggests that the fuzzy material at the tip of the
sperm head is chemically different from that
ornamenting other part of the head. In Perophora
annectens spermatozoa, the fuzzy extracellular
material is restricted exclusively to the plas-

Fertilization in Ascidians 997

malemma enclosing the anterior quarter of the
apical structure. This coincides with the region
where the fragmented proacrosomal vesicles prob-
ably were incorporated during the process of sper-
miogenesis [69]. It has been proposed that the
fuzzy extracellular material (surface ornamenta-
tion) at the tip of the sperm head in ascidians is the
site where the lysin(s) are found and that it plays
an important role in sperm-chorion interactions at
fertilization [20, 23, 24, 45, 71].

Concluding Remarks

The ascidian spermatozoa have an acrosome(s),
albeit a small one. In Ciona intestinalis spermato-
zoa, the acrosome reaction occurs via vesiculation,
in fundamentally the same way as observed in
mammalian spermatozoa. It occurs on the surface
of the chorion or after passage through the chorion
in the perivitelline space, which suggests that the
acrosome reaction is not necessary in order for
sperm to bind to or traverse the chorion. It has
been proposed that the fuzzy extracellular material
(surface ornamentation) at the tip of the sperm
head in ascidians is the site where the lysin(s) are
found and that it plays an important role in sperm-
chorion interactions at fertilization. In the peri-
vitelline space, apical processes protrude from the
apex of the sperm head. Gamete fusion occurs
between some of these processes and egg plas-
malemma, resulting in the incorporation of the
sperm into the egg from the anterior tip of its head,
in the same way that it does in other marine
invertebrates. In this respect, the apical process
may be analogous to the acrosomal process(es) of
other marine invertebrates. Fertilization in C.
intestinalis has characteristics of both mammals
and marine invertebrates, which is a new type of
fertilization.

Studies of the events associated with fertilization
have been done on ascidians with external fertiliza-
tion. There are many other groups of ascidians in
which fertilization is internal. In order to develop
a satisfactory understanding of the general
mechanisms of ascidian fertilization, the study of
the morphological aspects of internal fertilization
is indispensable and urgent.

ACKNOWLEDGMENTS

The author is most grateful to Professor Gary Freeman
and Mrs. Judy Lundelius, of the University of Texas at
Austin, for their valuable suggestions and for reading the
manuscript. He is indebted to Miss Megumi Yoshida, a
medical student at Nagoya City University, for her help
in preparing the manuscript. This work was supported in
part by a Grant-in-Aid from the Ministry of Education,
Science and Culture, Japan.

REFERENCES

1 De Santis, R. and Pinto, M. R. (1988) Zool. Sci., 5:
919-924.

2 Lambert, C. C. and Koch, R. A. (1988) Develop.
Growth and Differ., 30: 325-336.

3 Fukumoto, M. (1979) J. Ultrastruct. Res., 68: 1-5.

4 Burighel, P., Martinucci, G. B. and Magri, F. (1985)

Cell Tissue Res., 241: 513-521.

Franzen, A. (1976) Zoon, 4: 115-120.

Villa, L. (1977) Acta Embryol. Exp., 2: 179-193.

Woollacott, R. M. (1977) J. Morph., 152: 77-88.

Kubo, M., Ishikawa, M. and Numakunai, T. (1978)

Acta Embryol. Exp., 3: 283-295.

9 Villa, L. (1981) Acta Embryol. Morphol. Exp., 2:
69-85.

10 Villa, L. and Tripepi, S. (1982) Acta Embryol.
Morphol. Exp., 3: 201-215.

11 Villa, L. and Tripepi, S. (1983) Acta Embryol.
Morphol. Exp., 4: 157-168.

12 Cloney, R. A. and Abbott, L. C. (1980) Cell Tissue
Res., 206: 261-270.

13 Schabtach, E. and Ursprung, H. (1965) J. Exp.
Zool., 159: 357-366.

14 Tuzet, O., Bogoraze, D. and Lafargue, F. (1972)
Ann. Sci. Nat. Zool., 14: 177-190.

15 Tuzet, O., Bogoraze, D. and Lafargue, F. (1974)
Bull. Biol., 108: 151-167.

16 Rosati, F. and De Santis, R. (1978) Exp. Cell Res.,
112: 111-119.

17 De Santis, R., Jamunno, G. and Rosati, F. (1980)
Dev. Biol., 74: 490-499.

18 Cotelli, F., De Santis, R., Rosati, F. and Monroy,
A. (1980) Develop. Growth and Differ., 22: 561-
569.

19 Fukumoto, M. (1984) J. Ultrastruct. Res., 87: 252-
262.

CON NN

20 Fukumoto, M. (1988) J. Cell Sci., 89: 189-196.

21 Honegger, T. G. (1986) In “Progress in Develop-
mental Bioilogy”. part B, Alan R. Liss, New York,
pp. 85-89.

22 Honegger, T. G. (1986) Dev. Biol., 118: 118-128.

23 Fukumoto, M. (1983) Develop. Growth and Dif-

24

25

26

Zi)

28

29

30

31

32

35)

34

35

36

37

38

39

40

41

42

43

44

45

998

fer., 25: 503-515.

Fukumoto, M. (1986) Int. J. Invert. Reprod. Dev.,
10: 335-346.

Fukumoto, M. (1990) Develop. Growth and Dif-
Haltes AV-2 SES).

Burgos, M. H. and Fawcett, D. W. (1955) J.
Biophys. Biochem. Cytol., 1: 287-300.

Fawcett, D. W. and Hollenberg, R. D. (1963) Z.
Zellforsch., 60: 276-292.

Dan, J. C. and Shirakami, A. (1971) Develop.
Growth and Differ., 13: 37-52.

Takaichi, S. and Dan, J. C. (1977) Develop.
Growth and Differ., 19: 1-14.

Eckelbarger, K. J. (1984) J. Ultrastruct. Res., 89:
146-164.

Fukumoto, M. (1985) J. Ultrastruct. Res., 92: 158-
166.

Colwin, L. H. and Colwin, A. L. (1961) J. Biophy.
Biochem. Cytol., 10: 231-254.

Dan, J. C. and Hagiwara, Y. (1967) J. Ultrastruct.
Res., 18: 562-579.

Tilney, L. G., Kiehart, D. P., Sardet, C. and Tilney,
M. (1978) J. Cell Biol., 77: 536-550.

Usui, N., Takahashi, I. and Hoshi, M. (1987) In
“New Horizons in Sperm Cell Research” Ed. by
Mohri, H., Japan Sci. Soc. Press, Tokyo/Gordon
and Breach Sci. Publ., New York, pp. 265-274.
Barros, C., Bedford, J. M., Franklin, L. E. and
Austin, C. R. (1967) J. Cell Biol., 34: cl—c5.
Yanagimachi, R. and Noda, Y. D. (1970) J. Ultras-
truct. Res., 31: 465-485.

Stambough, R. (1978) Gamete Res., 1: 65-85.
Dale, B., Denis-Donini, S., De Santis, R., Monroy,
A., Rosati, F. and Taglietti, V. (1978) Biol. Cellu-
laire, 32: 129-133.

Monroy, A. and Rosati, F. (1983) Gamete Res., 7:
85-102.

Rosati, F., Pinto, M. R. and Casazza, G. (1985)
Gamete Res., 11: 379-389.

De Santis, R., Pinto, M. R., Cotelli, F., Rosati, F.,
Monroy, A. and D’alessio, G. (1983) Exp. Cell
Res., 148: 508-513.

De Santis, R. and Pinto, M. R. (1987) Develop.
Growth and Differ., 29: 617-625.

Rosati, F. and De Santis, R. (1980) Nature, 283:
762-764.

Fukumoto, M. (1990) Invert. Reprod. Develop.,

M. FUKUMOTO

46

47

48
49

50

51

a2
33)

54

55

56

57

58

59

60

61

62

63

64

65

66
67

68
69
70

71

17: 147-154.
Hoshi, M., De Santis, R., Pinto, M. R., Cotelli, F.
and Rosati, F. (1985) Zool. Sci., 2: 65-69.

Summers, R. G. and Hylander, B. (1974) Cell
Tissue Res., 150: 343-368.

Tilney, L. G. (1978) J. Cell Biol., 77: 551-564.
DeRosier, D. J., Tilney, L. G., Bonder, E. M. and
Frankel, P. (1982) J. Cell Biol., 93: 324-337.
Tilney, L. G., Fukui, Y. and DeRosier, D. J. (1987)
J. Cell Biol., 104: 981-993.

Lambert, C. C. and Lambert, G. (1984) Dev. Biol.,
106: 307-314.

Bates, W. R. (1979) Can. J. Zool., 58: 1942-1946.

Kawamura, K., Fujita, H. and Nakauchi, M. (1987)
Develop. Growth and Differ., 29: 627-642.

Kawamura, K., Fujita, H. and Nakauchi, M. (1989)
Develop. Growth and Differ., 31: 493-501.

Miller, R. L. (1975) Nature, Lond., 254: 244-245.
Lambert, C. C. (1982) Amer. Zool., 22: 841-849.

Lambert, C. C. and Lambert, G. (1983) Gamete
Res., 8: 295-307.

Hoshi, M., Numakunai, T. and Sawada, H. (1981)
Dev. Biol., 86: 117-121.

Sawada, H., Yokosawa, H., Hoshi, M. and Ishii, S.
(1982) Gamete Res., 5: 291-301.
Sawada, H., Yokosawa, H. and Ishii, S. (1984) J.
Biol. Chem., 259: 2900-2904.

Sawada, H., Yokosawa, H., Someno, T., Saito, T.
and Ishii, S. (1984) Dev. Biol., 105: 246-249.
Sawada, H., Kawahigashi, M., Yokosawa, H. and
Ishii, S. (1985) J. Biol. Chem., 260: 15694-15698.
Sawada, H., Yokosawa, H., Numakunai, T. and
Ishii, S. (1986) Experientia, 42: 74-75.

Yokosawa, H., Numakunai, T., Murao, S. and Ishii,
S. (1987) Experientia, 43: 925-927.
Haino-Fukushima, K. and Usui, N. (1986) Dev.
Biol., 115: 27-34.

Dan, J. C. (1960) Exp. Cell Res., 19: 13-28.
Longo, F. J. and Anderson, E. (1969) J. Ultras-
truct. Res., 27: 486-509.

Georges, D. (1969) J. Microscopie, 8: 391-400.
Fukumoto, M. (1984) J. Cell Sci., 66: 175-187.
Casazza, G., De Santis, R. and Pinto, M. R. (1988)
Develop. Growth and Differ., 30: 147-158.

Fukumoto, M. (1981) J. Ultrastruct. Res., 77: 37-
58,

ZOOLOGICAL SCIENCE 7: 999-1011 (1990)

© 1990 Zoological Society of Japan

REVIEW

Neuromuscular Transmission in Molluscan Hearts

RoBErT B. HILL and KIvOAKI KUWASAWA

Department of Zoology, University of Rhode Island, Kingston,
Rhode Island 02881, USA, and Department of Bioilogy,
Tokyo Metropolitan University, Fukazawa,
Setagaya-ku, Tokyo 158, Japan

Postjunctional Responses

At the time when we began our collaboration,
unitary postjunctional potentials had been re-
corded from myocardial cells of two opisthobranch
species [1-3]. This was close to the beginning of
the remarkable outpouring of knowledge concern-
ing CNS control of aplysiid circulation which re-
sulted from mega-cell neuronal studies conducted
in the laboratory led by E. R. Kandel [4]. We
wished to lay the foundation for a complementary
study of neural control of molluscan hearts at the
mini-cell cardiac smooth muscle level, with an
emphasis on interpretation of postjunctional re-
sponses. Our collaborative work began with the
use of an isolated preparation consisting of the
visceral ganglion, cardiac nerve, and an innervated
cardiac trabeculum from the prosobranch Busycon
canaliculatum [5]. Spontaneous activity in the
cardiac nerve elicited unitary inhibitory junctional
potentials (IJPs), from a resting potential near
reversal potential (—60 mV). Simple spontaneous
IJPs could fuse into compound IJP’s, but when the
cardiac nerve was stimulated large driven com-
pound IJPs appeared, at about three times the
amplitude of unitary IJPs. In spontaneous activity,
unitary IJPs could be attributed to nerve action
potentials (NAPs) in single axons while large com-
pound IJPs could be attributed to multiaxonal
repetitive inhibition of the postjunctional muscle
cell. This offered a mechanism for neural inhibi-

Received July 2, 1990

tion of the myocardium. Stimulation of the cardiac
nerve at a frequency just high enough for summa-
tion of compound IJPs held the myocardial cell
membrane at a slightly hyperpolarized level.

Although unitary IJPs, corresponding to single
NAPs, could be observed at resting potential,
similar unitary excitatory junctional potentials
(EJPs), also corresponding to single NAPs, could
only be observed after resting transmembrane
potential diffrence had been reduced. This hap-
pened following a shock to the cardiac nerve which
elicited a large long-lasting compound EJP. As the
compound EJP decayed, unitary EJPs appeared at
a transmembrane potential of about 40 mV, and
amounted to about one sixth of the amplitude of
the preceding compound EJP. Unitary IJPs and
EJPs were not confined to any “pacemaker” re-
gions, but appeared to indicate diffuse control by
innervation throughout the myocardium. Some
observations at this time indicated central/
peripheral interactions. For instance spontaneous
NAPs were abolished if the cardiac nerve were
severed distally to a structure which we call the
pseudo-ganglion, the “cardiac ganglion” of Carl-
son [6].

Studies of postjunctional! potentials in myocar-
dial cells, which are not spontaneously beating,
enable individual facilitating and summating EJPs
and IJPs to be distinguished, in the absence of
action potentials [7]. This approach showed that
strong excitation of the B. canaliculatum myocar-
dium can be induced by interaction of direct
excitation with rebound excitation, at the end of a

1000

period in which neural stimulation had evoked
myocardial IJPs.

A general picture had then emerged for the
hearts of opisthobranch gastropods, in which there
can be excitation by summation of EJPs and
inhibition by summation of IJPs but driving by
individual EJPs can only occur within a limited
range of frequencies [8]. This general picture also
proved to apply to the prosobranchs, which may be
considered a better parallel to vertebrates, since at
least 1500 axons may innervate the heart [8].
Ventricular smooth muscle of B. canaliculatum can
be tetanized by direct stimulation, but not by
stimulation of the cardiac nerve; the cardiac
smooth muscle failing to follow high frequencies of

nerve stimulation. Compound IJPs and EJPs
elicited by stimulating the cardiac nerve were slow,
lasting approximately one second, and were also
small, compared to postjunctional responses of
other molluscan smooth muscles. This may indi-
cate that electrical coupling between myocardial
smooth muscle cells is closer than coupling in, for
instance, buccal or ABRM smooth muscle cells.

Conventional postjunctional responses have
been described in cardiac muscle and arterial mus-
cle of Aplysia californica [4]. The innervation of
the heart can elicit fast EJPs in auricular muscle,
which can elicit a beat of the heart. Similarly, the
CNS elicits fast EJPs in circular muscle fibers of
the abdominal aorta, and three types of fast EJPs
in longitudinal muscle fibers of the anterior aorta.
The cardiac motor neurons of A. californica can be
classified as heart excitors and heart inhibitors [9].
One heart excitor (LDyp) elicits fast cardiac EJPs
in a phasic pattern, in which brief CNS activity
excites only one extra beat. Another heart excitor
(RByg) fires repetitively to elicit a slow, graded
cardiac excitation. By comparison, multiaxonal
excitation of the B. canaliculatum heart appears to
span the functions of these identified central motor
neurons of A. californica. The heart inhibitors of
A. californica (LDyy and LDy) briefly inhibit
the heart.

The central heart excitor neuron (LDypf) evokes
unitary discrete fast facilitating EJPs, which can be
recorded from cardiac muscle cells on the luminal
surface of the A. californica heart, both in the
auricle and in regions of the ventricle innervated

Hitt, R. B. AND K. KUWASAWA

by the auricular nerve [10]. These EJPs can induce
premature beats, as can EJPs induced in the heart
of Dolabella auricularia [2]. The central heart
inhibitor neurons (LDy;) in A. californica do not
appear to evoke widespread IJPs [10], in any
mirror image relationship to the EJPs evoked by
LDyr. However, IJPs were recorded from ven-
tricular and auriculo-ventricular (AV) valve mus-
cles of Aplysia kurodai and D. auricularia {1, 11].
These single identifiable heart inhibitor neurons of
aplysiids are thus functionally similar to the mul-
tiaxonal inhibitory innervation in B. canalicu-
latum.

Afferent Activity

Spontaneous activity in the cardiac nerve of B.
canaliculatum was not abolished by severing the
nerve between the heart and the visceral ganglion
[8]. This activity proved to correspond to afferent
nerve impulses of several kinds.

a. There is a burst of afferent nerve activity
during each contraction of a spontaneously beating
ventricle. b. “On” spikes appeared when a quies-
cent ventricular trabeculum was stretched. Beats
occurring during a stretch were of increased fre-
quency and duration. c. Spontaneous cyclic affe-
rent spikes which appear in the cardiac nerve of a
quiescent ventricle increase in duration and fre-
quency of bursts during stretching. Correlation of
spontaneous activity and responses to stretching
suggested that each contraction normally sets up
afferent stretch receptor activity in the cardiac
nerve. :

At the time when we first described afferent
activity in the cardiac nerve of B. canaliculatum [8,
12] very little was known about afferent signals
from gastropod hearts, and nothing else had been
published describing afferent impulses from
myocardia in which unitary postjunctional poten-
tials had been detected. However, afferent activity
and feedback modulation of the heart of Helix
pomatia were soon reported [13, 14]. In 1973
central neural interactions involved in cardiac con-
trol had been worked out carefully in the opisthob-
ranch A. californica [15], but observations of unit-
ary postjunctional potentials in cardiac muscle and
recording of afferent stretch-responses in the car-

Transmission in Molluscan Hearts 1001

diac nerve had perhaps proceeded further in the
prosobranch B. canaliculatum [5,7, 8]. We there-
fore began to try to carry knowledge of the cardiac
reflex system back toward the central nervous
system (in collaborative work with H. Neal, [16]).
We found that afferent activity, originating in an
intraventricular branch of the cardiac nerve, could
be traced through the visceral ganglion and on into
the parietal-visceral connectives. Afferent activity
in the parietal-visceral connectives was correlated
with efferent activity in intra-ventricular branches
of the cardiac nerve. Stimulation of an in-
traventricular branch induced afferent activity in
other branches, which was abolished when the
cardiac nerve was served distal to the pseudogang-
lion, but enhanced when the cardiac nerve was
severed between the pseudoganglion and the
visceral ganglion. This seems to indicate that the
pseudoganglion receives an inhibitory influence
from the CNS.

In B. canaliculatum, the afferent limbs of a
cardiac reflex arc have been established. In re-
sponse to a beat or stretch, afferent signals pass as
far as the parietal-visceral connectives and efferent
impulses appear in the intra-ventricular branches
of the cardiac nerve. Such efferent activity, which
can inhibit or excite cardiac muscle by inducing
facilitating and summating IJPs or EJPs, appear to
be generated in the neuropile-like pseudoganglion
and to be under inhibitory control by the visceral
ganglion. Since the predominating effect of the
cardiac nerve is to elicit myocardial IJPs [7, 8], we
suggested [16] that each contraction of the heart
may drive its own termination, in a reflex arc
through the pseudoganglion, modulated through
the visceral ganglion by higher centers.

Afferent input to the CNS from the heart has
also been recorded in H. pomatia [13, 14, 17],
Achatina fulica [18], and Aplysia depilans [19].
Touch, pressure, osmotic stimuli or chemical sti-
muli applied to the heart induced increased affe-
rent activity, or a change in background afferent
activity in the intestinal nerve of H. pomatia [13,
17]. A number of cells scattered throughout the
CNS of H. pomatia may participate in cardiac
reflexes [20]. Stimulation of the heart or pericar-
dium sets up a firing pattern in identifiable neuron
V12, which corresponds to efferent activity in the

cardiac nerve and inhibition of the heart [14].

A local reflex in the intestinal nerve of Helix
aspersa can inhibit the heart, after removal of the
CNS [21]. Apparently the reflex is mediated by
cells in the intestinal nerve trunk.

Identifiable heart excitor (RByg) and heart in-
hibitor (LDyy;) motor neurons of A. depilans are
identical to those of A. californica [19]. Afferent
sensory input from the heart of A. depilans
affected a number of identifiable neurons. After a
delay, cardiac afferent activity also affected RBygE
and LDy;. Thus it is quite possible that a cardiac
reflex arc may be found in A. depilans and A.
californica.

Ionic Basis of Transmitter Action

Irisawa in 1978 [22] included myocardial cells of
molluscs in a review of cardiac pacemaker cells,
considering that any myocardial cell could act in
the diffuse wandering pacemaker site. He re-
viewed the early literature concerning dependence
of resting potential, and of slow diastolic depolar-
ization, in molluscan myocardial cells, on [K*]o,
[Na*],, [Ca**],, and [Cl~],. Before that, Gers-
chenfeld [23] had reviewed the literature con-
cerned with ionic permeability changes activated
by receptors for ACh, dopamine (DA), and 5HT
on molluscan neurons. (Both Irisawa and Gers-
chenfeld commented on the lack of quantitative
work). The rather extensive early literature on
ACh responses in molluscan literature, was useful-
ly systematized by Gerschenfeld in 1973 [23].
Common responses were: D-responses blocked
by d-tubocurarine (dTC), hexamethonium and
strychnine; rapid H-responses which reversed at
—60 mV and were blocked by dTC and strych-
nine; and slow H-responses which reversed at — 80
mV and were blocked by tetraethylammonium and
methylxylocholine. The ionic mechanisms of these
neuronal responses appeared to involve changes in
permeability mainly to Na* (D-responses), K*
(slow H-responses), and Cl (fast H-responses).
However, there is a rather complex pattern of
physiological variation among molluscan neuronal
membranes and it is not easy to fit the gastropod
neuronal ACh receptors into the vertebrate clas-
sification of nicotinic and muscarinic receptors

1002 HILL, R. B. AND K. Kuwasawa

[23].

The concept of D-responses and H-responses to
acetylcholine may be extended to molluscan car-
diac muscle. In earlier literature [24] bivalve
myocardial cells have been reported to respond to
ACh with either a Na-dependent D-response or a
Cl -dependent H-response. Ventricular myocar-
dial tissue of D. auricularia has a D-response to
ACh, although ventricular myocardial tissue of A.
californica can show an H-response [25]. D-
responses to ACh resemble EJPs, while H-
responses to ACh resemble IJPs, in the myocar-
dium of D. auricularia [26]. Both IJPs and H-
responses are blocked by dTC [26, 27].

Bivalve hearts may be excited or inhibited by
ACh [28]. Elliott [29] compared ACh responses,
in cardiac muscle cells of Mercenaria mercenaria,
Mytilus edulis and Crassostrea virginica, to re-
sponses described in ganglion cells of A. californi-
ca. The H-response of Mercenaria ventricle
appears to be identical to the “slow, potassium-
mediated cholinergic response” of Aplysia neurons
[30[. Mytilus ventricle shows a D-response to ACh

5-HT 10° M

[29, 31]. Elliott [29] used iontophoretic application
of ACh, while recording intracellularly, and found
the D-response primarily sodium-dependent, thus
resembling the sodium-dependent ACh response
of Aplysia neurons. Crassostrea ventricle shows a
biphasic rapid D-response/slow H-response to
ACh [29]. The D-phase is due to an increase in
chloride permeability, resembling the chloride-
mediated ACh responses of Aplysia neurons,
while the H-phase is due to an increase in potas-
sium permeability. Elliott [29] found that the rapit
sodium-mediated D-response, the rapid chloride-
mediated D-response, and the slow potassium-
mediated H-response of molluscan cardiac muscle
fibers correspond to the 3 types of ACh responses
of Aplysia ganglion cells.

Studies of Chemical Transmitters

We have recently directed our collaborative
work to studies of the mode of action of chemical
neurotransmitters. Some of this work will be
briefly summarized in this section. |

ARR AIUUAAAAAA
AAMAAANNAARAAOALAAAARAAAAAAAIA

ACh 10M

'

Fic. 1.

Octopus hummelincki: Continuous records of transmembrane potential of a ventricular muscle cell.

Diastolic depolarization and cardiac rate were increased by SHT(10~° M). Hyperpolarization by ACh(10-* M)
caused cardiac inhibition by abolishing diastolic depolarization.

Transmission in Molluscan Hearts

CEPHALOPODA: Specimens of Octopus hum-
melincki were obtained from a commercial sup-
plier. Electrophysiological recordings were made
from auricles and systemic ventricles in prepara-
tions from animals anesthetized by injection of
isotonic MgCl, solution into the vena cava.

Both 5HT and ACh affect membrane potential
of ventricular myocardial cells of O. hummelincki
(Fig. 1). SHT acts like an excitatory neurotrans-
mitter, increasing the rate of rise of prepotentials
and thus heart rate. ACh resembles an inhibitory
nerotransmitter; hyperpolarizing, slowing the rate
of diastolic depolarization, and thus inhibiting the
heart. Continuous stimulation led to maintained
hyperpolarization or maintained depolarization of
myocardial membrane potential (Table 1). During
maintained depolarization, the rate of diastolic

TABLE 1. Membrane potentials of myocardial cells
of Octopus hemmelincki

Quiescent resting

Active cycle
potentials (mV)

Maximum (mV)

50.0+8.2 (4)
49.9+3.9 (52)

45.8+6.7 (17)*
48.0+5.7 (37)

Auricle
Ventricle

* mean+SD.The number of preparations is indi-
cated in parenthesis.

depolarization increased and the amplitude of the
action potentials increased (Fig.2A). During
maintained hyperpolarization there was a “vagal
escape”; that is, continued slow diastolic depolar-
ization (Fig. 2B). When the cut peripheral end of
the visceral nerve was stimulated with a glass
suction electrode, both IJPs and EJPs could be
recorded from either ventricular or auricular
myocardial cells (Fig. 3). EJPs (Fig. 3A1) or IJPs
(Fig. 3B1) are relatively slow. Repetitive stimula-
tion to the visceral nerve at 2Hz elicits facilitating
ventricular entricular EJPs (Fig. 3A2). At higher
frequencies, the EJPs facilitate and summate (Fig.
3A3). Similarly, repetitive stimulation at 5Hz
elicits IJPs, which summate to maintain membrane
hyperpolarization (Fig. 3B2). These results with
excitatory and inhibitory junctional potentials indi-
cate that neural excitation and inhibition are medi-
ated by junctional activity involving multiaxonal
innervation from the visceral nerve. Compound

1003
A

ee ee =
eee US ee EP os é “TAR
Pe a Ses Rey
er ee :
~ wg
ee)
LY

'
a:
{

so AS on Gee
We ee
VAVAYRYAYA\

5mvV

10sec

Fic. 2. Octopus hummelincki: Ventricular myocardial
action potentials.
A. Excitatory effects of stimulation of the visceral
nerve between arrows.
B. Inhibitory effects of stimulation of the visceral
nerve, between arrows.

EJPs corresponded one-to-one to stimuli of
maximal intensity. Four peaks could be detected
in one such long-lasting compound EJP. These
peaks corresponded one-to-one to stimulus pulses,
indicating that the compound EJP was composed
of, at least, four unitary junctional potentials,
produced by polyneuronal innervation.

Cardiac bioassay studies using the isolated syste-
mic heart of Sepia officinalis show myocardial
excitation by noradrenaline (NA) and related com-
pounds, as well as by S5HT and analogs [32].
Similar studies show cardioinhibition by ACh and
related compounds [33]. In a recent review of
extensive bioassay studies, it was concluded that
ACh is the natural cardioinhibitory transmitter for
the systemic heart of S. officinalis, while NA is the
probable cardioexcitatory transmitter [34].

ACh also slows the systemic heart and other
rhythmic components of the circulatory system of
Octopus dofleini and induces peripheral vasodila-
tion [35]. NA, however, also shows the systemic
heart and decreases peripheral resistance. SHT
excited systemic and branchial hearts of O.
dofleini.

1004 Hitt, R. B. AND K. KUWASAWA

[amv

auricle

250 msec

ventricle

500 msec

ventricle

500 msec

Fic.
. Excitatory junctional potentials.
A unitary EJP from the auricle.

B

IJP

|2my

250 msec

auricle

IJPs

wna nw sn

ventricle [rsteeeeecoer ae eee er]

1 sec

Octopus hummelincki: Microelectrode recordings from single myocardial cells.

Summating and facilitating repetitive EJPs from the ventricle (stimulation at 5 Hz).

Inhibitory junctional potentials.
A unitary IJP from the auricle.

3
A
ile
2. Repetitive facilitating EJPs from the ventricle (stimulation at 2 Hz).
3)
B
1
Z,

Summating and facilitating repetitive IJPs from the auricle (stimulation at 7 Hz).

BIVALVIA: We have conducted studies of the
mechanisms of action of chemical neurotransmit-
ters in the hearts of M. mercenaria and Mercenaria
stimpsoni (unpubl.).

Stimulation of the cardiac nerve of M. mercenar-
ia probably excites both excitatory and inhibitory
axons, since a long-lasting hyperpolarization fol-
lows the cessation of excitatory stimulation. EJPs
in the auriculo-ventricular valve of M. mercenaria
may be serotonergic, since they are reduced by
treatment with UML. IJPs in the ventricle of M.
mercenaria are probably cholinergic, since they are
blocked by Mytolon and by dTC. The atrio-
ventricular valve is excited by 5HT and hyperpola-
rized by ACh, but ACh has a strongly positively

modulatory effect on excitation by 5HT (Fig. 4).
These results, obtained by using isolated cardiac
neuromuscular preparations from bivalves may be
compared to earlier results obtained in studies of
peripheral neuromuscular control of opisthob-
ranch myocardia. Physiological diversity in the
molluscs, not surprisingly, extends to a wide spec-
trum of cardiac responses to acetylcholine, which
may be an inhibitory transmitter for overall func-
tion of the whole heart in some species, but
excitatory in others [28].
GASTROPODA: In a number of opisthob-
ranchs, ACh may also have differing effects in
differing regions of the heart, even though the
overall cardiac effect is “inhibitory”. We say

Transmission in Molluscan Hearts 1005
-8
5-HT 10 M
i OTA CACACACACACATACACAC ACA
—48mV

ACh 10°M

—-44mV
hops oo ae

ACh 10°M + 5-HT 10M

-52mV

ty

60 sec | 10 mV

Fic. 4. Mercenaria mercenaria: The AV valve responds to SHT and to ACh. Beating is induced by SHT(10-° M),
while ACh(10~° M) strongly hyperpolarizes, but ACh and 5HT interact in an interesting combined action. ACh
acts faster, and begins to hyperpolarize, but in the presence of ACh the onset of 5HT-action is faster, SHT
depolarizes more markedly, and 5HT has a more intensely positive inotropic and chronotropic action.

Ventricle
—52
{
Ai ye
Auricle
—46
| 20 mV
WNT T —56
—53 |
&t) ————_———__—_—_—-.-—- aes
d-1¢ —————_ 1 sec

Fic. 5. Simultaneous intracellular recordings were made from the ventricle (upper trace) and AV valve (lower trace)
of Dolabella auricularia while ACh (10° M) and dTC (10> M) were applied in the bathing medium. Numerals
show membrane potential at the points indicated (arrows). dTC depolarized the AV valve, which had been
hyperpolarized with ACh. dTC thus blocked the H-response to ACh in the AV valve, but not the D-response to
ACh in the ventricle.

1006

“inhibitory” in quotation marks, because relaxa-
tion of the myocardium to accept a greater volume
of venous inflow can be part of an excitatory
pattern at the level of the whole organism [3, 4, 9,
10, 15]. An instance of differing effects in different
regions is seen in the heart of D. auricularia, where
muscle fibers of the ventricle are depolarized by
ACh but muscle fibers of the AV valve are hyper-
polarized (cf. Fig. 5). In Dolabella, cardiac muscle
cells showing neurally induced IJPS are mainly
confined to the AV valve [1, 11]. In the AV valve,
effects of applied ACh are very similar to the
effects of neurally evoked hyperpolarization. De-
pression of the appearance of action potentials
progresses during a train of IJPs, even at a fre-
quency at which IJPs do not summate to maintain
hyperpolarization. The equilibrium (reversal)
potential for IJPs in AV valve muscle fibers is very
close to maximum hyperpolarization by ACh.
Ventricular muscle fibers are mainly depolarized
by ACh. A previous review [11] has summarized
the relation of responses to ACh to such factors as
pH, En, pacemaker localization, and phase of the
action potential. Literature surveyed [11] suggests
that in certain molluscan hearts the depolarizing
effect of ACh may be linked to an increase in
Na-conductance, while a hyperpolarizing effect
may be linked to an increase in Cl -conductance.
The membrane potential of myocardial fibers of D.
auricularia {11, 36] can be controlled by varying
[K*],. In zero-[K*],, the valve shows an aug-
mentation of the hyperpolarizing phase of the
normally biphasic ACh-response, while the ventri-
cle shows an augmentation of the depolarizing
phase. ACh may increase permeability to both
sodium and potassium, both in the valve and in the
ventricle [11]. The observed dependence of the
ACh-potential on [K~], might result from a redis-
tribution of Cl”. This redistribution would follow
changes in [K*],, as predicted from a Donnan

HILL, R. B. AND K. KUWASAWA

equilibrium.

Pharmacological agents have been used to
analyze the cellular mechanisms of excitatory and
inhibitory neural control of opisthobranch hearts
[26, 27]. Liebeswar et al. [37] have identified two
cholinergic cardioinhibitor motoneurons and one
serotonergic cardioexcitor motoneuron in A. cali-
fornica. Multiple responses to ACh in molluscan
neurons and cardiac muscle cells have been re-
viewed [27]. There seems to be general agreement
that ACh evokes: (1) a depolarizing response
(D-response) blocked by hexamethonium which
represents increased Gna; (2) a depolarizing or
hyperpolarizing response (H- or D-response)
blocked by dTC, which represents increased Gq;
and (3) a slowly hyperpolarizing response (slow
H-response) blocked by methylxylocholine, which
represents increased G,, and which is not blocked
by dTC. Cholinergic inhibition of the heart of A.
californica is not blocked by dTC [37]. The next
step seemed to be to determine whether or not
dTC would block IJPs. In fact, Kuwasawa and
Yazawa [27] tested dTC, atropine, hexametho-
nium and tetraethylammonium (TEA), but only
dTC blocked IJPs in the AV valve. dTC blocks —
IJPs, which have been inverted into depolarizing
junctonal potentials in Cl" -free medium, but does
not block simultaneous EJPs in the ventricle,
evoked by the same neural stimulation.

Pharmacological analysis of postjunctional re-
sponses and responses evoked by bath and ion-
tophoretic application of ACh was continued in a
later study of cardiac muscle fibers of D. auricular-
ia and Pleurobranchaea novaezealandiae [26]. In
both opisthobranchs, both D-responses and H-
responses to ACh correspond to decrease in mem-
brane resistance. These ACh-responses could be
differentially antagonized by mytolon, dTC, hex-
amethonium, and methylxylocholine, as follows.

Blocked by by by by

dTC Mytolon Methylxylocholine Hexamethonium
AV-valve H-response H-response — =
ventricle = D-response = D-response

Transmission in Molluscan Hearts

1007

Figure 5 is an example of simultaneous recording from the ventricle and AV-valve, showing effects of
ACh on both the sites and effects of dTC on the effects.
These effects may be related to ionic currents, as follows.

Blocked in Na-free solution
AV-valv
Ventricle D-response

Inverted in Cl -free solution

H-response

Pharmacological agents differentially blocked IJPs, as follows.

dTC Mytolon
D. auricularia

AV-valve blocks IJPs

P. novaezealandiae

AV-valve blocks IJPs

Methyxylocholine Hexamethonium

blocks IJPs — —

blocks IJPs — —

IJPs in the AV valve of both D. auricularia and P. novaezealandiae were inverted in Cl -free solution.

Significance of the Study of Molluscan
Cardiac Neurotransmitters

Walker [38] has reviewed the advantages of the
molluscs for studies of transmitters and modula-

tors. A wide range of pharmacological and ionic
analyses have been applied to a number of neuro-
muscular systems, which have frequently also been
used for studies of the structure-activity rela-
tionships of transmitter analogs and antagonists,
which may characterize receptors. Molluscan pre-
parations also provide valuable membrane models
[39]. Irisawa [24] demonstrated that bivalve
myocardium consists of classic unitary muscle,
with action potentials for which both Na* and
Cat* participate in spike potentials (without an
overshoot). These (and other) molluscan myocar-
dia provide convenient material for studying mem-
brane polarization dependence of the action
potential, Na* and Ca‘ * participation in a multi-
ionic spike, Ba** replacement of Ca**, nexus
transmission from cell to cell, Cat* control of
Na* inactivation, and other fundamental elec-
trophysiological phenomena [24].

The role of classic neurotransmitters in the
control of molluscan hearts has been reviewed a
number of times. If we refer to tables published in
1966 [40], we see that neurally elicited cardiore-
gulatory events had been reported in 40 species

among the Amphineura (Polyplacophora and
Aplacophora), Gastropoda, Bivalvia and Cephalo-
poda. Acetylcholine had been identified in ganglia
and hearts of 17 species, among the Gastropoda,
Bivalvia and Cephalopoda, while 5HT had been
identified in nervous tissue and hearts of 13 spe-
cies, among the Amphineura, Gastropoda, Bival-
via, and Cephalopoda [40]. At that time, there
was an extensive earlier literature reporting bioas-
says which supported a regulatory role for classic
neurohumors [for earlier reviews, see 41, 43].
Physiological and pharmacological evidence sup-
ported the suggestion that, for the heart of M.
mercenaria, ACh was the inhibitory cardioregula-
tory neurotransmitter while SHT was the excita-
tory cardioregulatory neurotransmitter [40]. At
that time (1966), there were indications that an
unidentified cardioexciter “substance X” was also
present as a neurosecretory product in gastropod
ganglia. More about that later, but (in 1969)
Greenberg [44] soon proposed that there were
isoreceptors on molluscan cardiac muscle cells for
classic neurotransmitters (ACh, SHT, and
catecholamines) as well as polypeptide cardiactive
substances. His review of the then current litera-
ture [44] still repays thoughtful rereading.

Since the early muscle bath bioassay studies of
putative neurotransmitters, many investigators
have used molluscan cardiac tissue for studies of

1008

many facets of the classic neurotransmitters, ami-
no acid neurotransmitters, and neuropeptides [38],
but only a few laboratories have directed attention
to the physiological roles of these chemical subst-
ances in control of functioning molluscan hearts [4,
26, 37, 45-49]. Jones’s [50] review in “The Mollus-
ca” may be consulted for progress by 1983 in
assessing the effects of classic neurotransmitter
substances on resting membrane potentials and on
action potentials of molluscan hearts. From the
literature reviewed [50] it appears that action
potentials are never over-shooting. Nevertheless,
modulation of the form of the action potential by
applied neurohumors regulates force [50] and car-
diac output [45, 47, 50-52]. However, there is an
extraordinary range of physiological variability in
responses of molluscan hearts to neurohumors
such as ACh and 5HT [S50].

Ultrastructural studies of gastropod myocardia
[53] have revealed dense innervation of the atrium
with nerve fibers distended with dense granules,
which may well be both motor and neurosecretory.
Similarly, methylene blue and EM preparations of
the H. pomatia heart showed a rich network of
swollen fibers in the atrium, which might well be
neurosecretory [54, 55]. Subsequently, a number
of cases have been identified in which neurosecre-
tory products are indeed discharged at the heart
[53, 56-58]. Molluscan hearts may also be of
particular interest in that they offer an opportunity
to study stepwise excitation by biologically active
amines, nucleotides, and peptides [59]. That is,
regulation involves the CNS, motor neurons, mod-
ulatory neurones, and biochemical processes [55]
at transmembrane and intracellular sites [59, 60].
Certainly, FMRFamide and related cardioactive
peptides have profound effects on molluscan
hearts [61-63] at the low concentrations1 in which
they circulate in the blood [64].

Cardioactive neuropeptides of gastropods were
reviewed by Lloyd in 1982 [65]. At that time,
cardioactive neuropeptides from gastropod neural
tissue could be grouped according to molecular
weight. In order from lowest to highest molecular
weight, these were (1) small cardioactive dipep-
tides (SCP)s, (2) medium cardioactive peptides
(MCPs), and (3) large cardioactive peptides
(LCPs). Of these, only the LCPs (released at a

HILL, R. B. AND K. KUWASAWA

neurohemal site in the atrium) were believed to be
circulating neurohormones involved in cardiac reg-
ulation. Lloyd [65] pointed out that the megacell
gastropods were particularly suited for dif-
ferentiating the actual physiological roles of these
neuropeptides, from their diverse pharmacological
actions. The heart of H. aspersa was excited by 5
or more gastropod neuropeptides [65]. A family of
5 peptides certainly exists in the CNS of Lymnaea
stagnalis, since all these have been sequenced, and
more may have been found by immunoreactivity,
but it remains to be determined how many of these
peptides participate in cardiac control [66]. Hep-
tapeptides appear to act on the H. aspersa heart at
different receptor sites than FMRFamide recep-
tors [67].

A brief history of molluscan cardiactive
neuropeptides may be in order here, although
several reviews have already covered the volumi-
nous literature extensively [38, 50, 57, 61, 62, 65,
68, 69].

An unidentified cardioexciter substance, found
in gastropod ganglia, was dubbed “substance X” in
1966 [40]. However, substance X soon proved to
comprise several substances identified by chroma-
tography as “peaks A, B, and C” [70-72]. Peak C
was then identified as the tetrapeptide, H-Phe-
Met-Arg-Phe amide(FMRFamide) by Greenberg
and Pride [68, 73, 74]. Jones [50] has reviewed the
numerous instances in which FMRFamide has
proved cardioactive in bioassays on molluscan
hearts. In a series of paper, from 1973 to 1980
[reviewed in 50] Greenberg, Price, and Painter
have suggested that FMRFamide may be a
neurosecretory substance, synthesized in ganglia
and transported in the visceral nerve to a release
site in the atrio-ventricular region of moluscan
hearts.

The history of the identification and bioassay of
other molluscan cardioactive peptides was re-
viewed in 1983 by Jones [50]. Many of these
substances are cardioactive [50]. Once released
into the hemolymph, the cardioactive
neuropeptides may act as long-term “regulators”
of cardiac and visceral muscle [68]. An instance of
this may be the “humorigenic” action of the blood
of Aplysia dactylomela [47].

At this time, the many valiant attempts to

Transmission in Molluscan Hearts

review the invertebrate neuropeptides have been
bedevilled by a plethora of publications describing
bioassays of a relatively few completely characte-
rized substances from invertebrate neural tissues
and immunoassays of a large number of substances
from vertebrate peptide families. These two
groups of substances were reviewed and dubbed
“native” and “naturalized” by Greenberg and
Price, in 1983 [69]. In this review, we have
concentrated our attention on “native” molluscan
neuropeptides which have been shown to have a
cardioregulatory role since 1983. We will, howev-
er, try to direct attention to the value of the study
of molluscan cardioregulatory neuropeptides, in
what Lukowiak and Murphy [75] have called the
“explosion” of research on neuroactive peptidergic
systems.

Price [76] has shown that cardioexcitatory
FRMFamide is probably present throughout the
phylum Mollusca. However, another tetrapeptide
(FLRFamide) may contribute to cardioexcitation
in several species, as may several heptapeptides in
the pulmonates. By 1988, a FMRFamide-related
family of 8 to 10 molluscan peptides had been
recognized [77]. These all have a more or less
similar distinctive C-terminal sequence, and show
a large range of potency in bioassays on the
ventricle of M. mercenaria.

In a limited number of preparations, physioilo-
gical roles of peptides related to FMRFamide are
beginning to be worked out [reviewed in 78].
FMRFamide has a cardioexcitatory effect on the
heart of Rapana thomasiana which resembled the
effect of serotonin, but is mediated by an action on
different receptors. In contract, FMRFamide and
SCPg appear to modulate the cardioexcitatory
neurotransmitter of A. fulica, FMRFamide en-
hancing cardioexcitation while SCP, antagonizes
excitation. The ionic mechanisms of FMRFamide-
related peptides have been studied in molluscan
central ganglion cells, but not in postsynaptic
regions of molluscan cardiac muscle [78]. The
catch-relaxing peptide (CARP) of the pedal gang-
lion of M. edulis has been determined to be H-Ala-
Met-Pro-Met-Leu-Arg-Leu-NH >. It inhibits the
hearts of M. edulis, Meretrix lusoria and Tapes
japonica [79]. CARP resembles another hep-
tapeptide, myomodulin of A. californica [80].

1009

Molluscan hearts are thus of interest for the
comparative study of evolution of control systems
analogous to, but not homologous with, control
systems in vertebrate hearts [81].

ACKNOWLEDGMENTS

This work was supported in part by a Grant-in-Aid for
International Scientific Research Program from the
Ministry of Education, Science and Culture, Japan (No.
02044122) and a fund for Special Research Project at
Tokyo Metropolitan University. Contribution from Shi-
moda Marine Research Center, No. 506.

REFERENCES

1 Kuwasawa, K. (1967) Sci. Rep. Tokyo Kyoiku
Daigaku Sec. B, 13: 111-128.

2 Kuwasawa, K. and Matsui, K. (1970) Experientia,
26: 1100-1101.

3 Mayeri, E., Kupfermann, I., Koester, J. and Kan-
del, E. R. (1971) Am. Zool., 11: 667.

4 Koester, J. and Koch, U. T. (1987) Experientia, 43:
972-980.

5 Kuwasawa, K. and Hill, R. B. (1973) Life Sci., Part
I, 12: 365-372.

6 Carlson, A. J. (1905) Biol. Bull., 8: 123-159.

7 Kuwasawa, K. and Hill, R. B. (1972) Experientia,
28: 800-801.

8 Kuwasawa, K. and Hill, R. B. (1973) In “Neuro-
biology of Invertebrates, Tihany 1971”., Ed. by J.
Salanki, Pergamon Press and Akadémiai Kaido,
Budapest, pp. 143-165.

9 Kandel, E. R. (1976) “Cellular Basis of Behavior.
An Introduction to Behavioral Neurobiology”., W.
H. Freeman, San Francisco, pp. 1-727.

10 Mayeri, E., Koester, J.. Kupfermann, I., Liebes-
war, G. and Kandel, E. R. (1974) J. Neurophysiol.,
37: 458-475.

11 Kuwasawa, K. (1979) Am. Zool., 19: 129-143.

12 Kuwasawa, K. and Hill, R. B. (1970) Biol. Bull.,
139: 427.

13. S.-R6zsa, K. (1972) Ann. Biol. Tihany, 39: 39-48.

14 S.-R6zsa, K. and Salanki, J. (1973) J. Comp.
Physiol., 84: 267-279.

15 Koester, J., Mayeri, E., Liebeswar, G. and Kandel,
E. R. (1973) Fed. Pro¢., 32: 2179-2187.

16 Kuwasawa, K., Neal, H. and Hill, R. B. (1975) J.
Comp. Physiol., 96: 73-83.

17 S.-Rézsa, K. and Salanki, J. (1973) Ann. Biol.
Tihany, 40: 95-108.

18 S.-R6zsa, K. (1979) Comp. Biochem. Physiol., 63A:
435-445.

19 S.-Rozsa, K., Salanki, J., Vér6, M., Kovacevic, N.

20
21

22
23

24
25
26
27
28
29
30
31
32

33
34

3)

36
Si)

38

39

40

41

42

43

to.

45

46

1010

and Konjevic, D. (1980) Acta. Physiol. Acad. Sci.
Hung., 55: 31-40. |
S.-Rozsa, K. (1979) Am. Zool., 19: 117-128.
Bagust, J., Fitrzsimons, J. T. R. and Kerkurt, G. A.
(1979) Comp. Biochem. Physiol., 62A: 397-407.
Irisawa, H. (1978) Physiol. Rev., 58: 461-498.
Gerschenfeld, H. M. (1973) Physiol. Rev., 53: 1-
119.

Irisawa, H., Irisawa, A. and Shigeto, N. (1969)
Experientia Suppl., 15: 176-191.

Hill, R. B. and Yantorno, R. E. (1979) Am. Zool.,
19: 145-162.

Kuwasawa, K., Yazawa, T. and Kurokawa, M.
(1987) Experientia, 43: 986-990.

Kuwasawa, K. and Yazawa, T. (1980) Adv. Physiol.
Sei, 22: 113=132%

Greenberg, M. J. (1965) Comp. Biochem. Physiol.,
14: 513-539.

Elliott, E. J. (1980) J. Physiol., 300: 283-302.
Elliott, E. J. (1979) J. Comp. Physiol., 129: 61-66.
Shigeto, N. (1970) Am. J. Physiol., 218: 1773-1779.
Kling, G. and Schipp, R. (1987) Comp. Biochem.
Physiol., 87C: 251-285.

Kling, G. (1987) Zool. Jb. Physiol., 91: 13-25.
Kling, G. and Jakobs, P. M. (1987) Experientia, 43:
511-525.

Johansen, K. and Huston, M. J. (1962) Comp.
Biochem. Physiol., 5: 177-186.

Nomura, H. (1965) Japan. J. Physiol., 15: 253-269.
Liebeswar, G., Goldman, J. E., Koester, J. and
Mayeri, E. (1975) J. Neurophysiol., 38: 767-779.
Walker, R. J. (1986) In “The Mollusca, Vol. 9,
Neurobiology and Behavior, Part 2”. Ed. by A. O.
D. Willows, Academic Press, Orlando and London,
pp. 279-485.

Meech, R. W. (1986) In “The Mollusca, Vol. 9,
Neurobiology and Behavior, Part 2”. Ed. by A. O.
D. Willows, Academic Press, Orlando and London,
pp. 189-277.

Hill, R. B. and Welsh, J. H. (1966) In “Physiology
of Mollusca, Vol. II”. Ed. by K. M. Wilbur and C.
M. Yonge, Academic Press, New York and London,
pp. 126-174.

Krijgsman, B. J. and Divaris, G. A. (1955) Biol.
Revs. Cambridge Phil. Soc., 30: 1-39.

Ripplinger, J. (1957) Ann. Sci. Univ. Besancon,
Zool. et Physiol., 8: 3-179.

Jullien, A., Cardot, J., Ripplinger, J. and Joly, M.
(1959) Ann. Sci. Univ. Besancon, Zool. et Physiol.,
12: 67-82.

Greenberg, M. J. (1969) Experientia Suppl., 15:
250-265.

Smith, P. J. S. and Hill, R. B. (1986) J. exp. Biol.,
123: 243-253.

Smith, P. J. S. and Hill, R. B. (1987) J. exp. Biol.,

61

62

63

64

65
66

67

68

69

70

fll

72

8

74

as

HILL, R. B. AND K. KUWASAWA

127: 105-120.

Smith, P. J. S. (1987) Experientia, 43: 956-965.
Kobayashi, M. (1987) Experientia, 43: 981-986.
Wilkens, J. L. (1987) Experientia, 43: 990-994.
Jones, H. D. (1983) In “The Mollusca, Vol. 5,
Physiology, Part2”. Ed. by K. M. Wilbur,
Academic Press, Orlando, pp. 189-238.

Hill, R. B. (1974) J. exp. Biol., 61: 529-539.

Hill, R. B. (1974) J. exp. Biol., 61: 629-637.
Nisbet, R. H. and Plummer, J. M. (1969) Experien-
tia Suppl., 15: 47-68.

Cottrell, G. A. and Osborne, N. (1969) Comp.
Biochem. Physiol., 28: 1455-1459.

Cottrell, G. A. and Osborne, N. N. (1969) Ex-
perientia Suppl., 15: 220-231.

Gainer, H. (1972) Brain Res., 39: 403-418.
Lloyd, P. E. (1978) J. Comp. Physiol., 128: 269-
276.

Lloyd, P. E. (1978) J. Comp. Physiol., 128: 277-
283.

S.-Rozsa, K. (1969) Experientia Suppl., 15: 69-77.
Koester, J., Dieringer, N. and Mandelbaum, D. E.
(1979) Am. Zool., 19: 103-116.

Greenberg, M. J., Painter, S. D., Doble, K. E.,
Nagle, G. T., Price, D. A. and Lehman, H. R.
(1983) Fed. Proc., 42: 80-86.

Price, D. A., Davies, N. W. Doble, K. E. and
Greenberg, M. J. (1987) Zool. Sci., 4: 395-410.
Price, D. A., Cobb, G. G., Doble, K. E., Kline, J.
K. and Greenberg, M. J. (1987) Peptides, 8: 533-
S33.

Nagle, G. T. (1982) Life Sci., 30: 803-807.

Lloyd, P. E. (1982) Fed. Proc., 41: 2948-2952.
Ebberiuk, R. H. M., Price, D. A., van Loenhout,
H., Doble, K. E., Riehm, J. P., Geraerts, W. P. M.
and Greenberg, M. J. (1987) Peptides, 8: 515-552.
Payza, K. (1987) Peptides, 8: 1065-1074.
Greenberg, M. J. and Price, D. A. (1979) Am.
Zool., 19: 163-174.

Greenberg, M. J. and Price, D. A. (1983) Ann.
Rev. Physiol., 45: 271-288.

Frontali, N., Williams, L. and Welsh, J. H. (1967)
Comp. Biochem. Physiol., 22: 833-841.

Agarwal, R. A. and Greenberg, M. J. (1969)
Comp. Biochem. Physiol., 31: 835-850.

Agarwal, R. A., Ligon, P. J. B. and Greenberg, M.
J. (1972) Comp. Gen. Pharmacol., 3: 249-260.
Price, D. A. and Greenberg, M. J. (1977) Science,
191: 670-671.

Price, D. A. and Greenberg, M. J. (1977) Prep.
Biochem., 7: 261-282.

Lukowiak, K. and Murphy, A. D. (1987) In
“Neuromethods, Vol. 6, Peptides”. Ed. by A. A.
Boulton, G. B. Baker and Q. J.Pitman, Human
Press, Clifton, N. J., pp. 439-476.

76
5)

78

79

Transmission in Molluscan Hearts 1011

Price, D. A. (1986) Am. Zool., 26: 1007-1015.
Greenberg, M. J., Payza, K., Nachman, R. J.,
Holman, G. M., and Price, D. A. (1988) Peptides,
oS —135.

Kobayashi, M. and Muneoka, Y. (1989) Biol. Bull.,
177: 206-209.

Hirata, T., Kubota, I., Takabatake, I., Kawahara,

80

81

A., Shimamoto, N. and Muneoka, Y. (1987) Brain
Res., 422: 374-376.

Kanda, T., Kuroki, Y., Kubota, I., Muneoka, Y.
and Kobayashi, M. (1990) In “Peptide Chemistry,
1989”. Ed. by N. Yanaihara, Protein Research
Foundation, Osaka, pp. 39-44.

Greenberg, M. J. (1979) Am. Zool., 19: 175-190.

1013

Proceedings of the
Sixty-First Annual Meeting of the

Zoological Society of Japan

October 3-5, 1990
Niigata

Suffixal letters of abstract number refer to the abbreviated
subfields of zoology

PH : physiology

GE : genetics

BI : biochemistry

DB: developmental biology

EN : endocrinology

CM: cell biology and morphology

BE : behavior biology and ecology

TS : taxonomy and systematics

; -.\ <dpeealyaiddas si «<a yis1 mado taeiade to eiient
? ~ * ; i ae na - i o 5s : * z a) UF H —_
See Sees ae _ = ¥ROIROS 10 ZO
: ca ; : ; 4 , ; ben 4 | = ar ‘
; of ae .2 ome Maoloieyle <2 Te
; Bs ’ } ay ; 2S ie ae Any ©
ie , - -yuenietjon > [a >
2 ; f : = ; ;
‘ | Bio rin lstnomaowysb ad
1 r F ; : 4 ee I200n 5 1 Ney
z nor bas vector! fies > Mio
yaeiaoy bas yecioid yerveise eh Cc Sr
7 . fy ? a > " oa |
‘ ; Seve Ii SAIOTORS: Gs
al | is | : :
: ) ‘
| : Ss
' X
x ~
Fa 1

ZOOLOGICAL SCIENCE 7: 1015-1016 (1990)

[THE ZOOLOGICAL SOCIETY PRIZE]

© 1990 Zoological Society of Japan

Embryonic Induction and Cell Differentiation
During the Amphibian Development

MAKOTO ASASHIMA

Department of Biology, Yokohama City University,
Kanazawa-ku, Yokohama 236

Since H. Spemann and H. Mangold (1924)
discovered the “organizer”, the morphogenetic
center in the newt embryos, many attempts have
been made to isolate the inducing substance which
causes cell differentiation and morphogenesis in
the presumptive ectoderm (undifferentiated cell
mass) of the embryos. I have approached this
embryonic induction problem in a wide variety of
ways, from experimental morphology to molecular
biology.

First, I proved that the capacity to induce a
secondary set of axial structures is not a property
of the egg cortex. Rather, as embryogenesis
proceeds through cleavage division to blastulation,
first, endodermal cells, then later the dorsal side of
the embryo progressively acquire primary
embryonic induction capacity.

Using the molecular approach, I focused on
discovering the nature of the mesodermal inducing
substance. Initially I isolated mesodermal induc-
ing or vegetalizing inducing substances from chick-
en embryo’s extract, carp swim bladder and the
embryo itself. Using these materials, some prop-
erties and features of these factors were eluci-
dated; protein in nature, molecular weight 25,000-
30,000, homodimer (monomer 12,000-13,000).
But with those factors, the amino acids sequence
of the protein was not determined. From those
basic experiments I developed the idea that it
would be better to isolate the mesodermal induc-
ing factors from various types of mammalian
cultured cell lines. Presently there exists a vast
number of cell lines which are of different origins,
secrete different proteins. Mesoderm induction
was assayed by examining the differentiation of
mesoderm-type features in cultures of presumptive

ectoderm of the newt and Xenopus embryos. The
assay system also examined erythroid differentia-
tion activity (EDF activity) in order to test the
relationship between mesoderm induction and
activin A (=EDF). Of 22 mammalian cell lines
examined, six strains were positive for both
mesoderm-inducing activity and EDF activity.
The conditioned medium of one of them, K562
myelomonocyte, has the highest inducing activity.
Then I purified the mesodermal inducing subst-
ance using by biochemical methods such as ammo-
nium sulfate precipitation, DEAE-Toyopearl col-
umn chromatography and reverse phase HPLC.
Finally, I succeeded in isolating the pure substance
as a single band by SDS-PAGE. I determined its
molecular weight to be 25,000 daltons. This
mesodermal inducing substance has the same
amino acid sequence of activin A (or Erythroid
Differentiation Factor=EDF) and consists of a
homodimer of 12,500 daltons.

Recently, the mesoderm-inducing effects of the
transforming growth factor (TGF-f) family of
proteins have been widely examined.

Independent of those efforts I determined that
activin A possesses mesodermal inducing activity.
I used the purified activin A made by recombinant
cells. First I reported that activin A is the most
effective of the range of inducing substances,
having its activity at the concentration of 1 ng/ml.
At the concentration of 10 ng/ml, it has 100%
mesodermal inducing activity. In the explants I
can see notochord, muscle and mesenchymal
tissues in high frequencies.

Though presently several mesodermal inducing
factors such as bFGF, TGF-f,, TGF-f;, ECDGF,
and so on are known, I have shown that activin can

1016

induce the mesodermal tissues at the highest
frequency by the lowest concentrations of the
factor.

Next, Xenopus liver genomic DNA library was
screened using a rat activin A cDNA as a probe to
identify the Xenopus homologue of the mamma-
lian activin gene. Some independent genomic
DNA clones were isolated. The reduced amino
acid sequence of these genes all showed almost

complete conservation of the distribution of cys-
teine residues. By Northern analysis we could
detect some of their mRNAs in the embryo. One
of them is expressed from stage 9 to the tailbud
stage. Recently, I have also tried to isolate the
activin protein itself from the early embryos. This
may demonstrate that the activin and its related
proteins are the natural inducer of embryonic
development.

ZOOLOGICAL SCIENCE 7: 1017 (1990)

[THE ZOOLOGICAL SOCIETY PRIZE]

© 1990 Zoological Society of Japan

Mechanism of Cytokinesis in Animal Cells

IssEI MABUCHI

Department of Biology, College of Arts and Sciences, University of Tokyo,
Komaba, Meguro-ku, Tokyo 153

Since cell division is an indispensable process for
living organisms, it has been a great topic in
biology. The animal cell divides into two by
constriction of the cleavage furrow. It has been
apparent that the cleavage furrow actively exert a
force for the constriction. The structure called the
contractile ring or the contractile arc has been
found in equilateral cleavage or in unilateral
cleavage, respectively, and shown to be mainly
composed of actin filaments. We have been
studying at a molecular level the mechanisms by
which the force for the constriction is produced
and the contractile apparatuses are formed.

The interaction between actin and myosin fila-
ments has been known long to produce the
contractile force in skeletal muscle. In non-muscle
cells, however, even the presence of these proteins
had not been established in the early 1970s except
Physarum plasmodium and blood platelet. We
studied ATPases in sea urchin eggs to clarify motor
proteins for cleavage, and found myosin. We
produced antibodies against starfish egg myosin
which abolished the interaction between actin and
myosin. When the antibodies were microinjected
into live starfish blastomeres, cleavages of these
cells were inhibited. This was the first evidence
that myosin is involved in cytokinesis. On the
other hand, this experiment exploited a novel
approach to function of proteins in the cell. The
involvement of myosin in cytokinesis has recently
been re-confirmed in yeast and Dictyostelium
discoideum, by gene targeting.

Next we became interested in the dynamic
property of the contractile ring. This structure had
been reported to appear at the onset of furrowing,

reduce its volume during contraction and finally
disappear after cleavage. This dynamics could not
be explained solely from the property of actin
itself. We searched for factors which regulate
supramolecular structure of actin in the echi-
noderm eggs, and found nine proteins which
modulate assembly properties of actin in vitro.
Among these proteins, alpha-actinin was shown to
be concentrated in sperm-reception cones in imm-
ature oocytes, fertilization cone and cortical layer
in fertilized eggs, and the cleavage furrow.

Even after these studies, both the dynamics and
the mode of contraction of the contractile ring or
arc were still mysterious. Perhaps the most
important approach to these problems is isolation
of the contractile apparatus. This had not been
successful for many years. Recently, we have been
able to isolate the cleavage furrow from newt eggs
by a surgical method. The isolated furrow was
contractile in vitro in the presence of ATP. The
contraction was inhibited by various actin- and -SH
poisons which re-confirmed the role of the actin-
myosin system in the furrowing. The ultrastruc-
ture of the contractile arc was also investigated in
detail at the first time. Evidence that the arc
filaments are formed by bundling of the unorga-
nized cortical actin filaments was obtained. Furth-
ermore, interestingly, the isolated furrow fraction
contained some unique proteins. We have also
succeeded in isolating the cleavage furrow from
sand dollar (Clypeaster japonicus) eggs and found
that the fraction contained unique proteins. Cur-
rently we are trying to identify these proteins. This
approach should lead us to further understanding
of the mechanism of cleavage.

¢ ah
ne
Rev od
s
are!
—
: !
:
2
Wy
‘ ae;
ot
G
/
aes! =

Si ade ‘ob “th: i UE varie! ratuodlompque

~ sioddehoel ar emaliong =

vine bee ciate soins xi : cane

jon bingo % fy
“hte i bat

qr RE ten yiaing Sonigtaxs 3A
oHT ak besiryaz o¥E> igen

Hota -AaIOT - ate Bee}
ae: wm noe bis) spnaaey
ot Pmwoite ae ‘ibie-Rt qin: 2 a peistote ae afi grooms,
af ti ond neitysose rs 18g! i baleviasongy aq?
he) oa! isoinioa® bit onde mivinisitivish: TOS Ftp
| weeny agevnob. eft-bas agn> bg citi} ni
bite ee yey ott dist. apis we: alt’ * mir tek

ne 3 ‘myabon:

Bios ‘ont sieeve ett Ye nner wrt Te ehant’ SH

teors sa eyes His v2 woe OH Sheen sisw 316

ie
Sot oF do sends fabeqad
Bed IAF: ‘batt gmt
nDed sv8n ey y Fistya9 ~ oi

ead “ ati i AUTH *

miei PRCT Sa ;
ae BIL TRS Site HIGH: S43 16

ewe S¥ tas PSE tee aay
et he avi wit uiteir ved ths site

nm bodse as 4¢ Poet | nee a, wilt

‘ ae n* Fes i ype ih :
247 Ct HSisice! ost } be « J
I aT A ' : ie a
‘ab 414 4 ra > : i “af7) re ori xq * aaa 5 3a eset)

5 & , ’
PTC pel “ith ee EG fey UVR ERY Te Ae Sie
‘ = ’
+, emt s :
af =f :
rae MRF 4 D1 Se rR ak areas at ;
£ ee 5 rs
PY re a 4
a) 3 lee sa -. eee t F iia
e “4
4 : i . u ea 7
&
‘ - = ;
rt 9 fer . 7 at: Fe ie
~! 2? “i
s -. ~
x ot! t 4rha ie5 } t
a ~
37 38) Ci : f L $ Mi ; }
é -
5 sc 4 ) y ; ? f
% *
é ae
f t,
J / if i es
j
; ‘al .
Pat lll \ A \ Vans 8 tee ie 24
ys 4 P
my ‘ ;
fs Pi i E ves |
yet
rt 7 yee i ry
“ i
=hig as
et - | Ss ees
§ han F

Al ah tT pertoly tofis tBoqtysall < She af ‘shen ng

ebmsgen Sieluibom-

SRD ONTS Tg. oh

Jal jalednig, bold tal a
iste evel 2 iff
4 d shan rH le pen
3 Wi. 7) a0) J ie ba Hin ‘7 Cae.

og teat ae

> @e wt pink. sabivib
Head cad 1 Worat
a ‘rises viovins: worth
‘pg bSliso ourtouate oT
Apes tat AB. aliioein oD

suit’ ‘svat i
‘od pinebisih sat ‘Set ries 1
‘te nuboig 23 tee a
"hanrrel vik. ttihcki ay

Bt ane fis ben fiiwe OW
| Gi asic WOLD
Hoven hy

oA
= he SUD

BrUrSik Vy a o.1 ie } ISH siaiat tS | oa
sax MITCT - (hse orf Gt is

“4

CIV Bee Teh 4AiRee
“ © ol * mise - Jaa & Wad on

pests, CAV ioevolar sat

soup ak mad anibee fitne ott 4
io 2agevedlo. siden a
rata ¥ 4s | . trl}
Si ct eatsiarg 6 aod
sng Ge
! atten plrTs nde od
ayer to send Salas 120GR OF Ber

PH 1
THE LIGHT EFFECT ON PIGMENT MOVEMENT OF
MEDAKA XANTHOPHORES BY REAGENTS
I. KAWAI. Aichi Pref. Coll. of Nursing,
Nagoya.

The light effect on pigment movement by
some kinds of reagents was examined using

denervated xanthophores in isolated scales
of medaka, Oryztas lLattipes. Pigment
movement was observed under two lights,
one of 100 lux, 500 nm, which did not
cause a light response, the other of 9000
lux Halogen, which caused a light
response. In summer, light accelerated the
response of K* aggregation but blocked
noradrenaline aggregation and theophylline
dispersion after a few minutes. But in
winter, the light reduced K* aggregation.
These effects in summer were differnt from
those in winter. The pigment movement
of xanthophores in usual microscopic
observation was mixed by the effect of
reagents and the microscopic light. It
was concluded that the observation in dark
condition, which did not cause a light
response, was required when we wished to
examine only the effect of the reaget.

PH 2

THE BACKGROUND ADAPTATION AND VISUAL FIELD
OF THE FLATFISH, Paralichthys olivaceus.
M. Fujimoto, T. Arimoto, F. Morishita and
K. Yamada. ZOOM “ENSit sj, BAGS OI SEs,
Hiroshima Univ., Hiroshima.

The adaptive response of the flatfish,
Paralichthys olivaceus, to several
different background patterns, which had
been printed on a long strip of plastic
film beneath the fish, was observed and
photographed. The image was processed by a
computer and the paleness indices were
calculated for comparison. The intact fish
showed different body shades and patterns
depending on the background pattern. The
skin showed maximal pallor on a white
background. On a small checkered pattern,
the skin color changed to light-brown with
many small dispersed white dots. On a
large checkered pattern, the skin was
bleached and several conspicuous’ black
patches appeared. On a black background,
the tint of the skin became darker. The
adaptive changes in the fish with a
covered left eye, observed by following
the same procedure aS with the intact
fish, were not very different from those
of the intact fish. While the fish with a
covered right eye was prevented from
showing these changes. The results suggest
that the left eye of the fish does not see
the immediate background. This difference
in the visual field is also supported by
an anatomical study.

Physiology 1019

PH 3

EFFECTS OF INOSITOL POLYPHOSPHATES AND
DIACYLGLYCEROL ON MELANOPHORES IN PRIMARY
CULTURE OF THE TILAPIA, OREOCHROMIS
NILOTICUS.

H.Wakatabi, N.Oshima and R.Fujii. Dept. of
BLOMOLEeCMIL, SCioy MACS OF SEto5s Uomo
Univ., Funabashi.

It has recently been known that pigment
motility in chromatophores is dependent on
the change, not only in the cAMP but also
fim Ehe Cacv tevela in Ene Gy tosiol | in
order to confirm this concept, two cate-
gories of substances, products of hydro-
lysis of PIP5, i.e., DAG (OAG) and IP3 and
the derivatives of the latter, many gt
them being known to be involved in Ca i
turnover, were applied on permeabilized
melanophores in primary culture of the
tilapia. IP3 induced melanosome aggrega-
tion through its Ca cs mobilizing action.
Netthen SiR now lh, had» such an yact ony.
The results may be dug to the uncapiaba lie y
of IP, to mobilize Ca ty IP,-induced Ca
entry into the cyipep Laveut is dependent on
the preceding Ca + mobilization. OAG
inhibited the pigment-aggregating effects
of NE and MCH to a certain degree.
Presumably, OAG-stimulation of PKC may
result in the activation of Ca *+_ATPase,
and of beta-adrenocepfor, i.e., the
decrease in cytosolic Ca v ievei ama Ene
increase in the cAMP level, respectively.
The present circumstances indicate that
inositol polyphosphates and DAG may
function as physiolosical mediators
triggering the pigment movements.

PH 4

REEXAMINATION OF THE EFFECT OF PROLACTIN ON
THE TELEOST FISH CHROMATOPHORES.

N. Oshimal, K. Kittal and H. A. Bern? 'Dept.
of Biomolecular Sci., Fac. of Sci., Toho
Univ., Funabashi, 2Dept. of Integrative
Biol., Univ. of California, Berkeley, USA.

Effects of the pituitary hormone pro-
lactin on the teleost fish chromatophores
were studied in vitro. As materials, scales
isolated from the dorsal part of the medaka
(Oryzias latipes), tilapia (Oreochromis
niloticus) and paradise goby (Rhinogobius
giurinus) were used. In addition to a pair
of tilapia (Oreochromis mossambicus) pro-
lactins, the larger PRL (PRL188) and smaller
PRL (PRL177) isolated by reversed-phase HPLC
at Univ. of California, growth hormone of
tilapia (tGH), ovain PRL (oPRL), Urotensin
I, II (UI, UII) and TL 2A (tPRL-associated
fraction) were also put to use. After the
application of each hormone to the isolated
scales, photomicrographs were taken contin-
uously and the responses of chromatophores
were analyzed.

Tilapia PRL177 and oPRL were found to
have an effect of dispersing pigment within
xanthophores, whereas tPRL1g8g, tGH, TL 2A,
UI and UII were of no effect. However, con-
trary to the suggestion proposed by Sage in
1970, prolactin appears to have little ef-
fect on the yellowing of fish actually,
Since the minimum effective concentration
of tPRL177 and oOPRL is higher than 100 nM
and the xanthosome-dispersing action of the
peptides is weaker than that of &-MSH.

1020 Physiology

PH 5

BETA-ADRENOCEPTORS IN THE MELANOPHORE OF
THE GOBY, TRIDENTIGER TRIGONOCEPHALUS.
H. Katayama. Mukaishima Marine Biological
Station, Fac. Sci., Hiroshima Univ., Mitugi-
gun, Hiroshima-ken.

Effects of several B-adrenergic agonists
in the absence and in the presence of s-
adrenergic antagonists on the pigmentary
state of denervated melanophores in isolat-
ed caudal fins of the fish were studied to
determine subtype of $-adrenoceptors of the
melanophores. Pigment of melanophores was
made to aggregate within the cells by 100
UM verapamil in the presence of 10 UM phen-
tolamine during the experiments. Isoproter-
enol (ISO), epinephrine (E), norepinephrine
(NE) and sulbutamol (SAL) all caused re-
dispersion of pigment in the presence of
verapamil and phentolamine. The order of
the pigment-dispersing potency was ISO>E>NE
>SAL. Propranolol effectively inhibited
the pigment-dispersing effects of all these
B-agonists. Metoprolol, a selective antag-
onist of 8,-receptors, considerably inhib-
ited the effect of NE and partially inhib-
ited the effect of E. By contrast, metopro-
lol did not inhibit the effect of SAL, the
selective 8,-agonist. These results suggest
that both §8,- and 8.,-adrenoceptors exist in
the melanophores. The relative numbers of
8,- and 8,-receptors, as a percentage of
the total population of S8-receptors,were
estimated to be 5% and 95%, respectively,
from analyses of the effects of E in the
presence of butoxamine or metoprolol.

PH 6

CHANGES IN THE LIGHT REFLECTING PROPERTIES
OF THE IRIDOPHORE WHICH DISPLAYS MOTILTY
OF THE ORGANELLE-AGGREGATION AND
DISPERSION TYPE

R.Fujii, H.Hayashi, J.Toyohara and H.
Nishi. Dept. of Biomolecul. Sci., Fac. of
Sci., Toho Univ., Funabashi.

When light-reflecting platelets were
aggregated into the perikarya of the
motile iridophores of some gobiid fish,
including the dark sleeper, Odontobutis
obscura obscura, the cells assumed bluish
tone. Those cells looked more whitish
with some of yellowish shade, when the
platelets were dispersed throughout the
cytoplasm. Spectral reflectance measure-
ments confirmed this phenomenon. The
bluish tone exhibited by the platelet
aggregate is considered to be due to the
presence of piles of platelets, which were
formed during the process of the organelle
aggregation. The multilayered thin-film
interference phenomenon occurring in the
piles of platelets leads to the generation
of color with high chroma. The condition
should be in favor of generating spectral
reflectance peak around blue region.
Optical analyses were made to explain the
sequence. When adapted to a dark
background, the fish became dark with a
sign of blue, being different from other
gobiid fishes, in which brownish hue
always dominates. The bluish tone may
have some ethological significance, such
as fomethe: antia-slandseintemsipeic i tac
identification in their social behavior.

PH7

ORGANIZATION OF REFLECTING PLATELETS IN
VERTEBRATE IRIDOPHORES

Y. Kamishima. Dept. of Biol., Facult. of
Sci., Okayama Univ., Okayama

The prominent monochromatic coloration
characteristic to iridophores rises from
uniform sets of reflecting platelets which
are arranged in a fixed way. However, it
is yet to be understood how the organelles
are fashioned in the uniform size and
aligned in a fixed position.

In developing iridoblasts, a double
walled envelope (the limiting membrane) of
the platelet was frequently observed. The
double wall was formed by folding of a
flattened cistern which was derived from
the ER. This means that the space where
the reflecting matter such as guanin
accumulates is not a lumen of the cistern
but a cytoplasmic portion encircled by the
folding of a flattened cistern. In this
way, thickness of the platelet means the
distance between the inner membranes of
the folded cistern. In the earlier devel-
opmental stage, intersecting bars were
observed between the inner membranes of
platelet envelope. The deposition of
guanin grew on these bars, so that the
length of the bars might determine the
thickness of the matured platelet.

There are a bundle of fibrous elements
at the margin of the platelet rows. These
fibers resembled to the intermediate fila-
ment and were hardly found in matured
cells. They appeared attached to the
developing platelet.

PH 8

LONGITUDINAL BAND AND THE SHOALING AND
SCHOOLING BEHAVIORS IN ROSE BITTERLING.
H. Nishi and R. Fujii. Dept.. ote Balomoltec:
Sei., Fac. of “Sci-..,. oho Unaiwie, Hunalbalsias:
Many bitterling specwes) dalcsip tay
brightly colored vertical bandon ahe
posterior half of the trunk. Among then,
the rose bitterling, Rhodeus yocessatus
ocellatus is known to possess a beautiful
blue-green vertical band. We found that
it was mainly composed of iridophores and
melanophores. The iridophores were found
physiologically inactive, and consequently
rapid color change of the band could not
be detected. The band, on the other hand,
looked differently in color when viewed
from different angles. Observations on
their schooling behavior indicated that
the fish made a lattice-like formation
horizontally rather than three dimen-
sionally, keeping their mutual distance
and the angle of formation within definite
ranges. The fish whose longitudinal band
was masked with Indian ink: never made a
shoal. It was interesting that an individ-
ual belonging to other bitterling species

than R. o. o., which had no colored band,
could get into the shoal of the latter
species. It may be concluded that, in R.
Oo. O., the perception by visual means of

the colored vertical band of other indi-
viduals is very important for making a
school. Since the color of the band does
not change actively, however, the fish may
not employ its hue for their subtle commu—
nication among those of the same species.

PH 9

INVOLVEMENT OF INHIBITORY GTP-BINDING PRO-
TEIN (Gi) IN PIGMENT-AGGREGATION RESPONSE
OF CULTURED FISH MELANOPHORE.

A.Shimada, F.Morishita and K.Yamada. Zool.
WMSiG5, ISS OIE Weao, WUleoslobine ~ Wins, -
Hiroshima.

Using cultured melanophores of the gold-
fish, Carassius auratus, the subtype of d-
adrenoceptors mediating pigment-aggrega-
tion response and involvement of Gi in the
Signal-transduction system were investi-
gated. All the adrenergic d-agonists exam-
ined induced a concentration-dependent
aggregation of pigment. The order of pig-
ment aggregating potency was norepineph-
rine, naphazoline, clonidine, methoxamine
and phenylephrine. Both yohimbine and
prazosin inhibited the cell response to
norepinephrine with the inhibitory potency
of yohimbine being about 100-fold greater
than that of prazosin. IAP, a pertussis
toxin shown to block the Gi-mediated
Signal-transduction system in mammalian
cells, inhibited the cell response to nor-
epinephrine in a concentration- and time-
dependent manner. The cell response to
100 nM norepinephrine attenuated to about
25% of the maximal after 6-h treatment
with 1 ug/ml IAP and totally disappeared
after 24-h treatment. The results indi-
cate that the subtype of M@-adrenoceptors
in cultured goldfish melanophores is Q,,
stimulation of which induces aggregation
of pigment through mediation of Gi.

PH 10

ULTRASTRUCTURE IN RELATION TO THE MOTILITY
OF IRIDOPHORES OF THE PARADISE GOBY ,
RHINOGOBIUS GIURINUS.

T. Iga, H. Ishida and A. Matsuno. Dept.
of Biol., Fac. of Sci., Shimane Univ.,
Matsue

Iridophores in the dermis of the
Paradise goby, Rhinogobius giurinus were
motile. Their motility involve the
translocation of platelets within the
cells. By the movement of platelets, the
iridophores change their light reflecting
areas and also spectral reflectance from
the cells. Thus, the iridophores take
part in color changes of the goby. The
ultrastructure of the motile iridophores
and the changes in the cell shape related
to the motility were studied with electron
microscopy. Cross sectional profile of
iridophores showed that in an iridophore
in the expanded state, the platelets were
scattered randomly throughout the cell,
so that the cell appeared to be rather
flat. In the punctate state, the plate-
lets were gathered, in groups or in stacks
with regular arrangements, in the centro-
sphere, which appeared to be ovoid in
shape. The processes from which the
platelets were lost remained there without
retracting. Abundant microfilaments were
observed in the cortical cytoplasm.
Microtubules were also seen around the
central regions of the cells. But, they
appear to be rather few in number, while
developing well in melanophores.

Physiology 1021

PH 11

CYTOSKELETON OF MELANOPHORES OF ORYZIAS
LATIPES: A COMPARATIVE STUDY OF MOBILE AND
IMMOBILE MELANOPHORES.

M. Obika. Dept. BalOlls 5 Ikeal@ “Wholakws ,
Yokohama.

In melanophores of color mutants of the
medaka dm and dm-2, melanosomes always
remain dispersed, incapable of responding
to stimuli that produce pigment aggrega-
tion in wild type melanophores. In melano-
phores of dm X dm-2 hybrid (Fi); mobility
of pigment granules is restoréd. Since the
morphology of the cells in situ is
different between mobile and immobile
types, an attempt was made to see if the
impaired mobility results from structural
modification of cytoskeletal framework.
Electron microscopy revealed that the
Mutants, especially dm-2, have bundles
of intermediate filaments in melanophores.
Immunofluorescence staining indicated that
these filaments are vimentin. Mobile me-
lanophores of the hybrid were only weak-
ly immunoreactive to vimentin antibody.
When cultured on a cover slip, melanopho-
res of the mutants tend to lose their
dendritic morphology, becoming a sheet of
rather simple outline similar to mobile
cells in culture. Though the presence of
melanosomes interferes with the
visualization of fluorescence images, it
appears that the localization and density
of microtubules and actin are similar in
all types of melanophores in culture.

PH 12

TRANSPORT OF SUCROSE SOLUTION FROM THE
CROP TO THE MIDGUT IN THE BLOWFLY,
PHORMIA REGINA M.

A.Shiraishi, T.Koga! and T.Yamamoto2.
IDepart. of biol., Fac. of Sci., Kyushu
Univ, Fukuoka. 2Chikushi Women's Junior
College, Dazaifu.

Tonic discharge and burst discharge
of impulses were recorded from the intact
recurrent nerve during the crop emptying
in the blowfly. Burst discharge of im-
pulses consisted of efferent impulses from
the brain and three different afferent
impulses from the periphery. They were
classified into two types. Type 1 burst
were classified into four subtypes and
type 2 bursts were also classified into
two subtypes. Type 1 burst operated in
reverse mode to pass a certain fluid vol-
ume back from the crop to the oesophagus.
Type 2 burst operated to pass a certain
fluid volume through the proventricular
valve into the midgut.

Tonic discharge of impulses mainly
consisted of efferent impulses from the
brain and of one type of afferent impulses
from the stretch receptor 3 located at the
crop duct. Firing rate of tonic efferent
impulses periodically decreased during
which fluid in the crop passed back
through the crop duct valve into oesopha-
gus. The periodic cycle was 90 sec, 327
sec and 1800 sec in the flies starved or
raised on 0.1 M and 2.0 M sucrose respec-
tively.

1022 Physiology

PHS

DISCRIMINATION OF COLOURED PAPERS IN THE
FLIES TRAINED TO MONOCHROMATIC LIGHTS.
T.Fukushi Dept. Gf Biol. Mayagiicoll:
of Educ., Sendai.

Individual blowflies, Lucilia cuprina,
were trained to a monochromatic light by
presenting a sucrose droplet on a patch

of white paper illuminated by the
monochromatic lieime (420 ico O20) sone
every 20 nm) and then tested for

Valsitinge icOLour marks! anmamtesits cist ior
many marks consisting of four colours
(blue; igneen, yellows andi ysced)) an) “the
arena under illumination with a halogen
lamp (150 W).

The flies visited frequently to blue
marks after training to monochromatic
lights of 440 to 480 nm, to yellow after
training to 520 to 580 nm and to green
after training to 500 nm. After training
to 620 nm and in the darkness the flies
visited frequently to red marks,
suggesting that the wavelength of 620 is
the limit of spectral sensitivity in the
fly.

Colour vision in the fly is discussed
on the basis of the spectral sensitivity
curves of three types of photoreceptor
(R1-6, R7 and R8) obtained from
intracellular recordings in other flies.

PH 14

RHYTHM OF FEEDING BEHAVIOR INDUCED BY BLOOD
INFUSION INTO THE SUCKER CAVITY IN ADULT
LAMPREY (Ichthyomyzon unicuspis).

R, KawaSaki. Col, Biomed, Technol, Niigata
Univ, Niigata, 951 Japan,

Lampreys take food and oxygen simulta-
neously (in the larva ) or separately (in
the parasitic adult) by the use of rhythmic
activity of prebranchial and branchial sys-
tem. Adult and juvenile lampreys were anes-
thetized and spinalized by pithing behind
the medulla. The head was held and sucker
was placed over a hole in a disc set in
lake water at 10 C. EMGs (head and branchial
muscles) and sucker vacuum were monitored.
Blood of fish or human (10 ul in each push)
was infused into the oral cavity through re-
peating-dispenser apparatus. Infusion of the
blood initiated startle ,pumping ,and suc-
ceedingly ,feeding behavior. when the feed-
ing decreased in rate ,additional infusion
of the blood increased in frequecy and in
amplitude of suctioning. Profiles of the
feeding induced by blood may be somewhat di-
ferent from that induced by skin extract;
"laceration work' and 'blood suckling' on
the prey fish skin may be different in the
feeding profiles. Independecy of the two
rhythms ( feeding and respiration ) was con-
firmed by additinal recording of respiratory
mechanogram; feeding events were seen ap-
pearing in all of the phases of respiratory
cycle. ln conclusion, another rhythm _ gen-
eration system for feeding behavior is de-
veloped during metamorphosis in lampreys.

In addition, observation of head muscle con-
traction and sucker vacuum changes confirmed
that suction generator muscles are mainly
apicalis-retractors (m.cardioapical and m.
styloapical). The piston cartilage is actu-
ally a immobile bar, supporting the movement
of the apicalis through a flexible joint.

PH 15

VISUAL FACTORS ATTRACTING MEDAKA
K.Taneda, T. Nishimoto and T. Matsuoka.
Dept. (of Biol. Hace om (Seis, | KochalmUnraer
Kochi

The distance between a pair of medaka
(66,52 or 22)in a circular pail was mea-
sured every 10 sec. Time-change in the
distances of each other showed that they
ranged within 10cm with slight fluctuations
in every case. The time-change in the dis-
tances between blinded specimens, however,
fluctuated markedly within a wide range.
These results suggest that visual sense
plays an important role in the schooling
behavior of the medaka. The behavior of a
test fish (male medaka), which was swimming
freely around a glass beaker containing a
dummy or a live fish, was videotaped. The
period of time when the test fish was
stayed near the side of the beaker (within
5cm) was expressed as the attraction of an
object. The degree of attraction of any
live fish was considerably high, while the
degree of a dead fish or a dummy was very
low. However, it was markedly increased
when the dead fish or dummy was artifi-
cially moved. An avoiding reflex by the
test fish in response to a sudden decrease
in illumination was investigated when the
fish was in various states — solitary,
grouped or stationary near an object. The
fish tended to turn in the opposite direc-
tion from the neighboring fish. On the con-
trary, it tended to turn toward a motion-
less object. A significance of the above
differences in response was discussed.

PH 16

THE PART OF THE BRAIN WHICH IS NECESSARY
FOR THE NORMAL SWIMMING BEHAVIOUR OF THE
HAGFISH.

S. Ooka! and H. Kabasawa% ‘Atomigakuen Jr.
Coll., Tokyo, *Aburatsubo Marine Park Aq.,
Miura.
Various parts of the brain of the hagfish,
Eptatretus burgeri, were cut or removed,
and the swimming behaviour of them was
examined by the observation with eyes and
by recording with an infra-red light photo-
cell system. After the examination of
behaviour, the brains were fixed, sectioned
and stained with Nissl method to confirm
the position of operations. Animals showed
the normal swimming behaviour after linear
surgical cuts across the full depth of the
brain at the border between the mesencepha-
lon and the diencephalon. The cut between
the mesencephalon and the rhombencephalon
caused abnormal swimming behaviour. When
the dorsal part of the mesencephalon (optic
tectum) was removed, the animals showed the
normal behaviour, but when the dorsal and
middle parts of mesencephalon (optic tectum
and the dorsal part of mesencephalic
tegmentum) were removed, the animals showed
an abnormal behaviour. The animals whose
brain were surgically removed except the
rhombencephalon and the mesencephalic
tegmentum showed the normal behaviour.
These results indicate that the nucleus of
the medial longitudinal fassiculus in the
dorsal part of mesencephalic tegmentum
plays an important role in the normal
swimming behaviour.

Physiology 1023

PH 17

IS THE TERMINAL NERVE SYSTEM INVOLVED IN
THE CONTROL OF REPRODUCTIVE BEHAVIOR OF THE
DWARF GOURAMI?
N. Yamamoto, Y. Oka and S. Kawashima. Zool.
Inst... Fac. of Sci., Univ. of Tokyo, Tokyo.
It has been speculated that the terminal
nerve (TN) system is involved in the
control of reproductive behavior of some
vertebrates because of its immunoreactivity
to gonadotropin-releasing hormone (GnRH).
However, experimental evidence is’ scanty.
To clarify whether the TN system is
concerned with reproductive behavior or
not, we used a tropical fish, the dwarf
gourami, and examined the effect of
bilateral electrolytic lesion of TN cells
on the male reproductive behavior
patterns. One and two days before, and one
and two weeks after the operation, the male
was paired with a female, and reproductive
behavior was observed for an hour. Two
weeks after bilateral TN lesion, most of
the GnRH-immunoreactive (ir) fibers
disappeared, leaving GnRH-ir somata in the
preoptic area and their axons toward
pituitary intact. Although male nest-
building and clasping-quotient (number of
male clasping/number of female thrusting,
%) tended to decrease one week after the TN
lesion, there was a variety among
individuals; some male performed active
reproductive behavior, whereas others none
at all. A more detailed study is necessary
tO ifeeaclr BB ~ iin COMmEIUSGI OM Om wine
functional role of the TN system in fish.

PH 18

THE ORGANIZATION OF RECEPTIVE FIELDS OF
SPIKING LOCAL INTERNEURONES IN THE LOCUST
WITH EXTEROCEPTIVE INPUTS FROM BOTH THE
MIDDLE AND HIND LEGS.

T.Nagayama, Zool. Inst., Fac. of Sci.,
Hokkaido Univ. Sapporo.

Two groups of spiking local inter-
neurones have so far been identified both
physiologically and morphologically in the
locust Schistocerca gregaria, one with
somata at the ventral midline and one with
somata in an antero-medial region of the
metathoracic ganglion. Both interneurones
receive exteroceptive inputs from the
hindleg and make direct output connection
onto the leg motor neurones.

This study demonstrates that only tne
interneurones of an antero-medial group
receive additional mechanical inputs from
the ipsilateral middle leg. These meso-
thoracic fields are either wholly
excitatory or wholly inhibitory and may
correspond in position to the field on the
ipsilateral hindleg. Furthermore, some
interneurones are also excited or
inhibited by inputs from the contralateral
middle and hind legs.

The antero-medial interneurones
integrate the local inputs from a hindleg
in parallel with midline interneurones and
play a role in the expression of local
reflexes. Their wider receptive fields,
however, suggests that they also influence
the coordination between the legs in
posture and locomotion.

PH 19

PHYSIOLOGICAL PROPERTIES OF DUM NEURONS OF THE
COCKROACH THORACIC GANGLION.

H. Washio, Lab. Neurophysiology, Mitsubishi
Kasei Inst. Life Sci. Machida, Tokyo.

Intracellular recordings were made from
single somata of the dorsal unpaired median
(DUM) neurons of the thoracic ganglion of the
cockroach, Periplaneta americana. In contrast
to the cell bodies of insect motoneurons,
cockroach DUM neurons have been found to give
large and overshooting action potentials
followed by a large afterhyperpolarization
(AHP). The AHP was reduced, abolished and
reversed by progressive hyperpolarization. The
AHP was completely abolished by removal of
extracellular Ca2+ or addition of Co2+. The
involvement of both K+ and Ca2+ in the
generation of the AHP suggests strongly that
the AHP is mediated by a Ca2+-activated Kt
conductance.

Very small spontaneous depolarizing and
hyperpolarizing junction potentials were
recorded from the soma membrane. On
comparing these potentials with the
spontaneous potentials of the leg muscle
fibers, there are no synapses on the DUM
somata. Octopamine (OA) had an excitatory
effect on the DUM neurons. Application of OA
(10-5M) increased the frequency of spontaneous
action potentials which were accompanied by
about 50% increase in the apparent membrane
resistance. Thus, neuromodulators such as OA
might play important roles in the synaptic
mechanism by altering the synaptic strength
and cellular properties.

PH 20

MORPHOLOGICAL STUDIES ON OUTPUT
CONNECTIONS OF ASCENDING INTERNEURONES ONTO
UROPOD MOTOR NEURONES IN CRAYFISH
M.Sato! and M.Hisada2. lBiology, Rakuno

Gakuen Univ., Ebetsu, Hokkaido. 2Zool.
Inst., Fac. of Sci., Hokkaido Univ.
Sapporo.

Intersegmental ascending interneurones
Ooriginatng from the terminal abdominal
ganglion of the crayfish receive sensory
inputs from the uropods. In addition to the
function of these interneurones as the
mechanosensory interneurones, we found that
many of them also function as pre-motor
interneurones to the uropod motor neurones
through the local circuits.

Using a double marking method (whole-
mount silver intensification and HRP
marking), we investigated connections of
ascending interneurones onto mortor
neurones morphologically by electron-
microscopy. Direct connections of an
ascending interneurone onto a motor neurone
were found in a limitted ventral area and
in a very small number. This indicates that
though the ascending interneurones' have
direct connections to the motor neurones
through chemical synapses, they are likely
to be of small importance in the local
circuits subserving the motor output
generation.

1024 Physiology

PH 21

INTRACELLULAR RECORDING AND STAINING OF
TERMINAL NERVE CELLS IN THE BRAIN OF THE
DWARF GOURAMI IN VITRO.
Y.Oka and T.Matsushima. Zool. Inst., Fac.
of Sci., Univ. of Tokyo, Tokyo

We have previously shown in the dwarf
gourami that terminal nerve (TN) cells are
a major component of the GnRH system and
that the TN-GnRH system is structurally
independent from the preoptic/hypophysial-
GnRH system which facilitates gonadotropin
release from the pituitary. Thus, we have
hypothesized that GnRH may function as an
important neuromodulator. A whole-brain in
vitro preparation has been developed to
facilitate studies on the ionic and
biochemical mechanisms of the TN-GnRH
neuromodulatory system. In the present
study, we examined the spontaneous
electrical activity of TN cells using both
extra- and intracellular recording
techniques and determined the morphology
of individual cells by intracellular
injections of either Lucifer Yellow or
biocytin. TN cells showed an endogenous
regular beating discharge pattern (1-20
Hz). Anatomical observation of
intracellularly-labeled cells revealed
multiple axonal branches that projected
to those areas where we had previously
demonstrated dense GnRH-immunoreactive
fibers. The present results suggest that
individual TN cells may modulate neural
activity in a wide variety of structures
via these multiple axon branches.

PH 22

THE FREQUENCY CHARACTERISTICS OF THE
MEMBRANE POTENTIAL FLUCTUATION AND THE

MOTOR OUTPUT IN THE LOCAL NONSPIKING

INTERNEURON OF CRAYFISH
-Takahashi,;, M.Hisada and M.Takahata“
Zool. Inst.,,Fac. of Sci., Univ. Hokkaido,
Sapporo and “Inst. Appl. Electr., Univ.
Hokkaido., Sapporo.

The roll of the local nonspiking

interneurons (LNSNs) in the terminal
ganglion of crayfish were studied in the
aspect of the frequency characteristics.
The LNSNs receive various postsynaptic
inputs, graft them dynamically into its
membrane potential, and control motoneurons
(MNs). The spontaneous fluctuation of
membrane potential was analyzed by fast

Fourier transform. The power spectrum

showed low pass characteristic. The cutoff
frequency was 43.9 + 22.6 Hz (n=9).

To estimate signal transmission from
LNSNs to MNs, white-noise modulated current
was intracellularly injected into the LNSNs
and the output spike train of the MNs was
monitored at the axon. The cross-
correlation between the above input and
output was calculated. The transmission
function which also showed low pass
characteristics had lower cutoff frequency
(Goe8 = 2BoV ji (M=S2))) Gham tinat @f tine
spontaneous fluctuation.

By these low pass characteristics the
discrete post synaptic potentials in the
DECMOWOreLINSINS) swe men almaitied= sini ats
frequency so that they can control the body
movement continuously and in a graded way.

PH 23

PHYSIOLOGICAL PROPERTIES OF NONSPIKING
INTERNEURON DENDRITES IN CRAYFISH.

M. Takahata and S. Shimizu. Sect. of Sensory Information
Processing. Res. Inst. of Applied Electricity. Hokkaido Univ..
Sapporo.

Passive electrical properties of the nonspiking interneuron
dendrites have been studied in comparison with those of mo-
toneuron dendrites using intracellular recording and staining
techniques in crayfish. The nerve cell was penetrated by a
glass microelectrode loaded with the fluorescent dye. Lucifer
yellow (3% in 1M LiCl). The site of electrode penetration
into the cell was visually confirmed under the dissecting mi-
croscope during intracellular recording by illuminating the
Lucifer-filled cell in situ with the blue light for excitation.

The mean input resistance of nonspiking intemeuron den-
drites (27MOhm) was significantly higher than that of the
fast motoneuron dendrites (7MOhm: P<0.001). The mean
time constant of nonspiking interneurons (46msec) was
longer than that of fast motoneurons (8msec: P<0.001). The
slow motoneurons showed the intermediate value of mean
input resistance (12MOhm), although they had the mean
time constant of 47 msec which was as long as that of non-
spiking interneurons. All interneurons and motoneurons
showed no significant rectification in the currently examined
range (<!.0nA) of hyperpolarizing current. The mean resting
potential of nonspiking interneurons was significantly lower
than that of motoneurons confirming our previous observa-
tion. The high input resistance and the long response time
constant of the dendrites of nonspiking interneurons might be
suggestive of their function as the integrative element oper-
ating independently of the motoneuron dendrites which are
dedicated to synaptic integration for individual motoneurons.

PH 24

MODULATION OF THE EXCITABILITY OF WIND
SENSITIVE INTERNEURONS IN FLYING OR WALKING
CRICKETS
Y. Baba, K. Hirota and T. Yamaguchi. Dept,
of Biol., Fac. of Sci., Okayama Univ.
Okayama

Intracellular recordings from the
terminal abdominal ganglion were made in
tethered-flying or suspended-walking
crickets to examine the influence of
behaviors on the excitability of identified
wind sensitive interneurons: five giant
interneurons (MGI, LGI, 9-2, 9-3, 10-2),
two local spiking interneurons (LSI-3, LSI-
9) and four local non-spiking interneurons
(LNI-1, LNI-2, LNI-5, LNI-6). The membrane
potential of MGI, LGI and LSI-9 started to
increase at the initiation of flight or
walking, and their responses to wind were
suppressed during flight as well as during
walking. The spontaneous firing of 9-3 and
10-2 increased as soon as the animal
Started flying or walking, but their
responses to wind decreased during these
behaviors. Both the membrane potential and
spontaneous firing of 9-2 and LSI-3 were
hardly affected by either of flight or
walking. The fluctuation of membrane
potential of local non-spiking interneurons
occurred spontaneously even in a quiescent
state. However, both the frequency and
amplitude of the fluctuation of membrane
POLeENELall VO ea Nie— lye Ni Sy eeercln Clam Neat
decreased not only during flight, but also
during walking. Those of LNI-2 increased
during the whole period of flight or
walking.

Physiology 1025

PH 25

AXOTOMY-INDUCED LONG-LASTING FIRING IN AN
ANAL MOTONEURON OF THE CRAYFISH.

A. Muramoto. Fukushima Biomed. Inst. of
Environm. & Neopl. Dis., Futabagun.

I presented evidence that axotomy can
induce a long-lasting discharge activity
(afterdischarge, AD) of an anal motoneuron
(AML) , which is capable of driving anal
contractions of the crayfish, Procambarus
Clarkii. The mechanism of production of
this AD was examined. Activity of AML was
recorded from the posterior intestinal
nerve (PIN) extracellularlly by a suction
electrode in the isolated nerve cord prepa-
ration, which composed of the 5 thoracic
ganglions (T1-T5) and the 6 abdominal ones
(A1-A6) with the PIN. Transection of the
connectives between Tl to Al had no influ-
ence on a AML activity, but that between Al
to A6é could always evoke a AD which lasts
over a long period of time (7 sec-67 min).
Moreover, in the nerve segment of the PIN
separated from A6, cutting of its stump
could produce a AD similar to that before
separation. In view of the previous study
that AML is a t-shaped unipolar and its
soma exists within A6, the ascending axon
of AML runs through the nerve cord to Al
and the AD may occur due to axotomy of AML
itself and not dependent on synaptic effect
from other neurons, since when the cord was
bathed in high-Cca&t, high-Mg** Ringer's solu-
tion or in Ca‘-free, high-Mg*t one, the
response of AD similar to that in normal
Ringer's solution could be observed even
after 4-h adaptation of these solutions.

PEN 26

LEVEL OF TYROSINE CONTENT OF THE SUBOESO-
PHAGEAL GANGLION IN THE PHYTOPHAGOUS
LEPIDOPTEROUS LARVAE

T.SHIMIZU-, M.KIKUCHI“, T.KUSANO-,

F.MIYAGAWA- and y . TSUKADA
Navetionall Iystitute of Health, Shinagawa,
Tokyo 141, “SRL, Hachioji, Tokyo 192

It is well known that suboesophageal
ganglion (SG) is involved in the control
of various activities such as walking,
respiration, stridulation, head movements,
grooming and mandibular movements.
Previously, we presented that SG obtained
from wandering stage larvae, Mamestra
brassicae, contains dopa, octopamine,
tyrosine and tyramine using HPLC with
native florescence detector analysis (Zool.
S@i5 > 69 U2, AOS) e6 Ane aS Aes shinee
tant, therefore, that investigation of the
biological function of the SG in the same
phytophagous larvae, Aedia leucomelas
Linne and Agrius convolvuli (Linnaeus).

We found in both insects high level of
tyrosine content of the SG in the larvae
from the post-wandering stage to just
before pupation as compared with that in
the last instar larvae until pre-wandering
stage (during feeding stage). Relationship
between mandibular movements (feeding) and
tyrosine content of the SG is also
discussed.

PH 27

SEROTONIN-IMMUNOREACTIVE INTERNEURON IN
THE PINEAL ORGAN OF RIVER LAMPREY,
LAMPETRA JAPONICA.

S.Tamotsu, M.Samejima, C.Nagashima* and
Woomierenicas Io DejNeo IDeyealolls - *Central
Lab. of Histology, Hamamatsu Univ. Sch.
Med., Hamamatsu.

The pineal organ of the river lamprey
is a photosensitive system consisted of
photoreceptors, supporting cells and nerve
cells. Histological characteristics of
the pineal nerve cells have been so far
insufficiently studied. Serotonin-immuno-
reactive (SI) neurons were recently
observed in the pineal organ of lamprey.
In the retina SI cells were found in the
interneurons (amacrine cells). In order
to study whether these SI neurons in the
pineal organ are interneurons or ganglion
cells, the double labeling method of
retrogradely transported tracer and
immunocytochemical techniques was applied.
The former labeled the ganglion cells and
the latter identified the SI interneurons
and ganglion cells. It was foun that the
SI nerve cells in the end-vesicle of the
pineal organ were not labeled retrograde-
ly. Some of them made contact (synapse?)
with adjacent serotonin-immunoreactive
photoreceptor cells. These findings
suggest that they are interneurons. SI
cells around the atrium are yet to be
identified as interneurons.

PH 28

FUNCTIONAL REGENERATION OF NEURAL RETINA OF
NEWT I: PHYSIOLOGICAL AND MORPHOLOGICAL
STUDIES.

T.SAITO AND Y.KANEKO. LMSies > Or JNO, Sei
Univ. of Tsukuba, Tsukuba, IBARAKI.

Normal and regenerated retinas at periods
ranging from 19 to 35 days after removal of
the original retina were enzimatically
dissociated into single cells. Solitary
photoreceptor, bipolar and ganglion cells
in the normal retina could be identified
under the microscope. The solitary cells
dissociated from the regenerated retina at
periods between 24 and 35 days after the
operation sometimes included neurons that
morphologically resembled photoreceptor,
bipolar and ganglion cells in the control
retina, though the identification of these
cell types became more difficult at earlier
regenerating times. At regenerating periods
GANG ENG! scone Oe eeromen22) daysi atte, athe
operation, the retina was usually observed
as a Single cell layer of mostly
depigmented cells. Solitary cells
dissociated from this early regenerating
retina were characterized by round cell

bodies with no _ processes. These cells
produced action potentials with a prolonged
long duration as the membrane was
Gepolarized. In 80% L-15 culture medium at
26 °C, some of these cells remained viable
for up to 30 days and generated extensive
fields of neurites. They elicited

overshooting action potentials of short
duration similar to those of ganglion cells
in the control retina.

1026 Physiology

PH 29

FUNCTIONAL REGENERATION OF NEURAL RETINA OF
NEWT II: VOLTAGE-GATED CURRENTS OF THE
SOLITARY GANGLION CELLS.

Y.KANEKO AND T.SAITO. Wied > Oye Jeol, Syekte
Univ. of Tsukuba, Tsukuba, IBARAGI.

Ionic basis of the membrane currents
underlying action potentials of solitary
ganglion cells dissociated from regenerated
retinas at periods between 19 and 35 days
after the removal of the retina were
examined with voltage-clamp techniques, anda
compared with those of the control retina.
(1) Solitary ganglion cells of the control
retina consisted of four voltage-gated
currents: a transient inward Nat-current
(INa),a sustaind inward Ca+t-current (Ica),
a delayed outward Kt-current (IK), anda
4-AP-sensitive outward current. The mean
value for the maximum Ina and that of I,
for depolarization from -80 mV to +20mvV
were 788 and 394 pA (n=43). The magnitude
of Ica varied widely from cell to cell(mean
value=153 pA; n=6). The transient outward
current was not studied, because of its
large variability among cells. (2) At early
regenerating periods (19-22 days after the

surgery), INa, Ica and Ix were already
present in the cells that could be identi-
fied as neurons. The transient outward

current tended to develop somewhat later.
The mean values for the maximum Ina, Ik and
Ica were 152 (n=19), 85 (n=19) and 81 pA
(n=3), respectively. Within the following
two weeks of regeneration, these four
currents all increased and reached to the
control values.

PH 30

REVERSAL POTENTIALS OF LIGHT RESPONSES IN
CHROMATICITY-TYPE HORIZONTAL CELLS OF THE
CARP RETINA.

K.-I. Takahashi and M. Murakami. Dept.
Physiol., Keio Univ. Sch. Med., Tokyo.

To reveal ionic mechanisms underlying
light responses in the chromaticity-type
(C-type) horizontal cells, the reversal
potentials were measured using the Ca
action potential method (Murakami &
Takahashi; J. Physiol., 1987). In a
modified solution containing high Ca, Ba,
and some K channel blockers, the C-type
cells produced long-lasting Ca action
potentials where spectral responses
appeared in opposite polarities to those
at the resting state like a mirror image.

A glutamate (Glu) response induced at
the Ca action potential was hyperpolariz-
ing, whereas depolarizing when Glu was
applied at the resting state. The peaks
of these responses of both polarities
coincided well at -5 mV, indicating that
this level is the reversal potential of
the Glu response. Also at this level, no
light responses were detected, indicating
that the reversal potentials of the light
and Glu responses were the same. The
above observations and the reversal
potential level strongly suggest that the
C-type cells receive Glu-mediated inputs
from cones, and that the synapses on
horizontal cells have non-selective
permeability to both Na and K.

PH 31

MULTISENSORY CONTROL OF COMPENSATORY
EYESTALK MOVEMENTS IN THE CRAYFISH
(PROCAMBARUS CLARKIT)

Y. Okada, M. Ebihara, H. Furudate and T.
Yamagucha: Dept. of Biol... Hace OfeScie
Okayama Univ., Okayama

When the intact animal is tilted in roll,
the eyestalks deviate in a way as to
compensate for the imposed tilt even in
darkness. When the bilaterally statolith-
ectomized animal is suspended in air and in
a quiescent state, the orientation of the
eyestalks in space depends on the position
of a single light: the direction and
magnitude of the deviation of eyestalks are
determined by the position of light in the
vertical plane. When the bilaterally
statolithectomized animal is tilted in roll
under a vertical light or the walking legs
of the bilaterally statolithectomized
animal are moved up and down by rolling the
substrate, the compensatory eyestalk
movements occur for the absolute or
relative imposed tilt. In hilatearallv
Statolithectomized animal, «

light and a contact sensation to wa.--
legs inhibit the compensatory eyestalk
movements induced by substrate and visual
Stimuli, respectively. These results
indicate that when the simultaneous change
in gravitational, visual, and substrate
inputs induces synergically the
compensatory eyestalk movements, but when
any one of these inputs keeps stationary,
it acts as an inhibitory input to ene
compensatory eyestalk movements.

PH 32

WHITE NOISE ANALYSIS OF CERCAL AFFER-
ENTS IN THE COCKROACH

Y. Kondoh, J. Okuma, and Y. Hasegawa

Wako Research Center, Honda R&D Co. Ltd., Wako, Saitama

Responses of cercal afferents in the cockroach, Perzplaneta amer-
icana, were evoked by a Gaussian white noise modulation of
wind. First- and second-order kernels were computed by cross-
correlating the spike discharge against the white noise input to
analyze the response dynamics of cercal sensory afferents. We
found that the summation of the first and second- order models
produced by convolving the first- and second-order kernels with
the stimulus inputs could predict each impulse in spike trains.

First-order kernels from twelve afferents were biphasic, and
closely matched the time differential of a pulse. Second-order
kernels had two diagonal peaks and two off-diagonal valleys. This
four-eyes configuration was basically invariant irrespective of the
stimulus angle whereas the kernels varied in their amplitude. The
differential waveform of the first-order kernel was reversed when
stimulated on the opposite side. In either side, the valley of
the first-order kernel was opposed by the diagonal peak of the
second-order kernel, implying that the cercal afferents had a half-
wave rectifier property. Both kernels had a similar characteristics
in frequency domain: the time between the peak and a follow-
ing valley of the first-order kernel was constant, and had a mean
time of 4.640.1ms (n=12), whereas the time between two di-
agonal peaks of the second-order kernels was 4.7+0.1ms. Thus,
the cercal sensory cells act as a bandpass filter with a peak fre-
quency of about 105Hz, and are velocity sensitive because of the
biphasic first-order kernel. The peak time, however, varies greatly
from 1.9 to 7.2ms, which would reflect the spatial distribution of
the corresponding hairs on cercus and serve to reduce the time
resolution.

Physiology 1027

REUS3

A FREEZE-FRACTURE STUDY OF THE SYNAPTIC SITE
IN THE OUTER PLEXIFORM LAYER OF THE RETINA.
S.Hidaka, Dept. of Physiol., Tokyo Women's Med. Coll.,
Tokyo.

The outer plexiform layer (OPL) of the vertebrate retina
includes the synaptic interactions between photoreceptors and
bipolar and horizontal cells. In catfish (/ctalurus punctatus) the
synaptic connectivity patterns in the OPL have been resolved in
thin-sectioned specimens (Hidaka et al.,'86). I have investigated
the internal structure of the membrane at specialized neural
contacts in the OPL with the freeze-fracturing technique, which
splits cell membranes and thus exposes their internal structure.

In rod and cone terminals, the plasma membrane along the
slope of the presynaptic ridge of the photoreceptor ending
contains synaptic vesicle sites. The rod- and cone-horizontal cell
dendrites forming the lateral compoments of the triad have a
linear array of intramembrane ectoplasmic fracture face (EF)
particles in the site facing the arcifom density (AD) and rows of
plasmic face (PF) particles in the proximal region to AD. The
synaptic ridge is the site of the signal transmission from the
photoreceptor to the horizontal cell. At the "spinule" contact
region between distal invaginating horizontal cell dendrites and
photoreceptor terminals, the horizontal membrane has densely
aggregated PF particles and the adjoining membrane of the
photoreceptor terminals contain an aggregate of PF and EF
partilces. The spinule contact might be a feed-back site from the
horizontal cell to the photoreceptor.

The plasma membrane of the central dendrite of the off-center
bipolar cell facing the presynaptic ridge contains an aggregate of
PF particles. At the contact region between the basal surface of
photoreceptor endings and the lateral dendrites of both types of
bipolar cells, the receptor membrane has clusters of PF particles
and the adjoining membrane of the bipolar dendrites contains an
aggregate of EF particles. The central dendrite and the lateral
dendrites of the bipolar cells are thought to be the synaptic site
concerning the response-polarity.

PH 34

MORPHOGENESIS OF THE PHOTORECEPTOR OUTER
SEGMENT DURING DEVELOPMENT OF MOUSE REITNA.
S. Obata and J. Usukura, Dept. of Anatomy,
Sch. of Medicine, Nagoya Univ., Nagoya.

The disk membrane formation of photo-
receptor cells in developing mice retinas
were studied by rapid freezing, freeze-
substitution, freeze-etching, immunocyto-
chemistry and biochemical techmiques. In
freeze-substituted preparations, the apical
swelling of connecting cilium in early
developing stages (postnatal 5-8 days)
contained several small vesicles. In the
following stages, similar vesicles were
observed at the basal region of developed
outer segments. These vesicles are consid-
ered to be precursors of disk membranes
Since they contained opsin. Rapid freezing
was able to arrest pinocytotic vesicle
formations and their fusion in cytoplasm.
We are proposing that disks are formed by
fusion of small vesicles and following
compression. Actin filaments seemed to be
involved in such formation precess. Indeed,
Many actin filaments were located exclu-
Sively at the basal region of outer seg-
ments or distal ends of connecting cilia
throughout the development. Mostly, barbed
ends of actin filaments were on the
membranes. However, direction of actin
filaments in early stages was much more
variable than in adult.

PH 35

LIGHT AND ELECTRON MICROSCOPIC INVESTI-
GATION OF CELLULAR ARCHITECTURE IN THE
RETINA OF THE FIREFLY SQUID, WATASENIA
SCINTILLANS III

'M.Michinomae, =H.Masuda, =M.Seidou and
NR WaLEOs 5 Dees Or JIBIOl. 4 IEG. Oit Sets,
Konan Univ., Kobe Hyogo. 2)Dept. of Biol.,
Fac. of .Sei.; Nara Women's Univ. Nara.
S\IMoG. Olt JRLOl,., Wees Ow Sis, Oseliel
Univ., Tyonaka Osaka.

ee

We investigated the retina of the firefly
squid, Watasenia scintillans. The outer
segment(OS) of the small area of the ven-
tral retina is thick (600 um), and com-
posed of three types of OS of photo-
receptor cells. While the other dorsal
Pact Vor Tetina as) than. (300 %2 mm); and) is
only one type of OS (Seidou et al. 1990).

In the present study, we found the
twisted arrangement of the OS around 10
wm from the distal end in the ventral
retina. The twisting cause the rearrange-
ment of rhabdomes, resulting in increase
of the smaller sized rhabdomes.

Seidou et al.(1990)J Comp Physiol A 166
769-773

PH 36

EXPRESSION OF A REPORTER GENE IN THE
CNS OF DROSOPHILA LETHAL MUTANTS
INDUCED BY P-ELEMENT INSERTIONAL
MUTAGENESIS.

A. Komatsu!, E. Machiyamal!, Y. Sano2, R. Ueda?
and D. Yamamoto?.!Dept. Physiol., Tokyo
Women's Med. Coll., 2Mitsubishi-Kasei Inst. Life
Sci.

Mutation was induced by transposition of
Drosophila P-element, P-lwB, which contains a
reporter gene, lacZ of Escherichia coli, and a weak
constitutive promoter. The expression of the lacZ
gene was detected with a histochemical technique
which provides blue precipitate within cells
containing the gene product, B-galactosidase. At the
wandering stage of 3rd instar larvae, all of 129
lethal strains screened showed reporter gene
expression in the CNS; 72 out of them expressed
regional specificity of the expression within the
CNS; and 14 out of them revealed neuron-type-
specific expression. While most of the strains were
embryonic lethal, six strains showed pupal lethality.
In one of them, homozygotes had stunted imaginal
discs expressing no fB-galactosidase activity and a
CNS expressing the activity within neurons located
at the frontal end of brain. This result suggests a
possibility that the underdevelopment of discs in this
mutant may be caused by a secondary effect of
defective neuronal function.

1028

PH 37

Morphological study of the photoreceptive sites on
the genitalia of a butterfly, Papilio xuthus.
Yumiko Miyako, Kentaro Arikawa, and Eisuke Eguchi.
Department of Biology, Yokohama City University,
22-2 Seto, Kanazawa-ku, Yokohama 236, Japan.

Two pairs of photoreceptive sites (P1,P2) have been found

electrophysiologically on the genitalia of Papilio xuthus®). It
has been suggested that the photoreceptive organs lie beneath
transparent cuticle on the scaphium (male P1) and the papila
analis (female P1). The photoreceptive cells have not yet been
identified, however.

In the present study, we first examined series of semithin
sections (0.4um thick, epoxy-embedded, toluidine blue-
stained) of Pls of both sexes and found a single ovoid struc-
ture (ca.30um x 40um) that contained a nerve cell body in
each of the Pls. The cell body sends an axon (5~7um in di-
ameter) to a nerve branch from which the photoresponse is
recorded by a suction electrode.

We also examined ultrathin serial sections of the ovoid
structure by electron microscopy. The nerve cell body in the
structure bears a thick axon proximally, whereas at the distal
end it bears several sensory processes each of which has nu-
merous microvilli at the tip. The distal portion of the ovoid
structure, therefore, is occupied with the microvilli. Since the
microvilli are not very well organized, the cross sectional
profile of the microvillar region is extremely labyrinthine. The
ultrastructure of the cell is very similar to an extraocular pho-
toreceptive cell reported in an earthworm, Lumbricus ter-
restris),

(1) Arikawa et al. Nature, 288,700-702, 1980.
(2) Rohlich et al. Z Zellforsch, 104,435-357, 1970.

PH 38
Arrangement of spectral receptors in the omma-
tidia of Papilio xuthus (Lepidoptera) determined by
measurement of the polarization sensitivity

Kouichi Bandai and Kentaro Arikawa.

Department of Biology, Yokohama City University,
22-2 Seto, Kanazawa-ku, Yokohama 236, Japan.

In the retina of a butterfly, Papilio xuthus, five types of

spectral receptors were identified by intracellular recording.
These types have their sensitivity maxima in the UV (365nm),
violet (400nm), blue (460nm), green (520nm), and red
(600nm) range of the spectrum“). In order to determine the ar-
rangement of these receptors in an ommatidium, especially in
the distal retinular layer, responses to different wavelengths
and e-vectors were recorded intracellularly from each of the
receptors.

In the distal part of the ommatidium, four cells contribute
microvilli to the rhabdom. In the normal position of the but-
terfly, cells 1 and 2 form a dorsal-ventral pair, and cells 3 and
4 form a posterior-anterior pair.

From the phase of the polarization sensitivity curves
(P.S.curves) and its relation to the orientation of the mi-
crovilli, we conclude that cells 1 and 2 are UV, violet and/or
blue receptors and cells 3 and 4 are green receptors. This re-
sult supports the idea that the arrangement of the spectral re-
ceptors is different between the ommatidia.

The phases of two P.S.curves which were recorded from a
single green receptor at wavelengths of the primary peak
(520nm) and of the secondary peak (365nm) are the same.
The phase of P.S. curves from UV receptors differ from those
of green receptors by 90°. This suggests that the secondary
peak was caused by green receptor itself, but not by UV re-

ceptors.
(1) Arikawa et al. Naturwiss,74:297-298, 1987.

Physiology

PH 39

Valva-opening response induced by light stimulation

of the genital photoreceptors of a butterfly, Papilio.
Kentaro Arikawa, Takuya Ai, and Eisuke Eguchi.

Department of Biology, Yokohama City University,
22-2 Seto, Kanazawa-ku, Yokohama 236, Japan.

A swallowtail butterfly Papilio has two pairs of extraocular

photoreceptive sites on the genitalia’. Each site contains only

one photoreceptive neuron (genital photoreceptor, GP) whose

soma bears many microvilli?. The GPs send axons to the last
abdominal ganglion and are functionally connected with some
abdominal motoneurons”). The biological significance of the

GPs, however, has not yet been determined.

In response to light stimulation of the genitalia, a male but-
terfly opens its valva, a pair of lobes for clasping female's geni-
talia during copulation. Based on the following results, we
show that the valva-opening response (VOR) is somehow con-
trolled by the GP activity.

1. VOR is elicited by light stimulation of the genitalia, not of
the eyes. Even a decapitated animal can perform the VOR.

2. The action spectrum of VOR peaks at violet which is similar
to the spectral sensitivity peak of the GPs.

3. GPs have inhibitory and excitatory effects on the activity of
motoneurons that innervate valva-opener and -closer mus-
cles. Action spectrum of these effects also peaks at violet.

A male repeatedly opens and closes its valva at the beginning
of copulation to ‘determine’ the correct position and then clasps
the female with the valva to stabilize the posture. Involvement of
the GPs in the VOR indicates that the butterfly's mating be-
havior is, at least in part, controlled by the "hind sight”.

1) Arikawa, Eguchi, Yoshida & Aoki. Nature,288:700,1980.
2) Miyako, Arikawa & Eguchi. Zool Sci, this issue, 1990.
3) Arikawa & Aoki, J Insect Physiol ,30:853,1984.

PH 40

THE FORMATION OF MYELOID BODIES IN
RETINULAR CELLS OF PUPAL COMPOUND EYES OF
SILKWORM MOTHS (BOMBYX MORI) EXPOSED TO A
CONSTANT BRIGHT LIGHT.

Shihoko Maeda and Eisuke Eguchi.

Dept of Biol, Yokohama City Univ,

22-2 Seto, Kanazawa-ku, Yokohama 236.

Numerous well developed myeloid bodies
whose structure comprises massed parallel
and flattened membranes are formed in the
retinular cells of the developing compound
eyes of pupae removed from the cocoons of
Silkworm (Bombyx mori) irradiated with
constant and abnormally bright light
(10,000 lux). Under normal conditions,
such well developed myeloid bodies have
not been observed in arthropod compound
eyes. They have only been found in the
retinular cells of lepidopterans (Manduca
sexta) reared with carotenoid deficient
diet (White and Bennett, Cell Tissue
Res, 257:519-528,1989) . In this experiment,
however, silkworms were fed with normal
diets containing carotenoids. Therefore
the formation of myeloid bodies in our ex-
periment is caused by abnormally strong
illumination. The fine structure of
myeloid bodies and their relationship to
endoplasmic reticulum are discussed.

PH 41

THE EFFECT OF RED LIGHT TO AN ANOMALOUS
BEHAVIOUR (PYOKORI) IN D. MELAMOGASTER
‘K.Matsubara, 7M.Michinomae, Y.Kawakami,
Y. Tanabe, °?E.Nakasima 1)Toyonaka 4th
Junior High School, Osaka. 2)Dept. of
BiGlo, RACs Om SEs, I<sOmein Why, 5 hole
EWOBO,. S)DAOGS Of Ikatte Selo, Winiyys oie
Osaka Pre., Sakai Osaka.

An anomalous behaviour mutant of
"pyokori" in D. melamogaster does jump

with the stimulation of flicker light.
After emergence, the pyokori behaviour(P)
is strongly affected by environmental
light. When the flies that lost of P by
darkness processing are exposed to white
light for short period, the P is recov-
ered. However, the P is not recovered,
even if the flies that lost of P by red
light irradiating are exposed to white
light. The visual pigments of both are
analyzed with HPLC. In the former, the
11-cis 3-hydroxyretinal (3-OH) is disap-
peared with the darkness processing. How-
ever, the all-trans 3-OH is kept yet.
When white light is irradiated to this,
the 11-cis is formed simultaneously, and
then P is recovered. In the latter,
1i1-cis and all-trans 3-OH are well re-
served during the red light irradiating.
However, the P is disappeared completely.
These results suggest that the irradiat-
ing of red light cause to denaturation of
the visual substances, resulting in de-
crease of pyokori behaviour.

PH 42

EFFECT OF CAROTENOID DEFICIENCY ON CIRCADIAN
RHYTHM IN HATCHING OF THE SILKWORM, BOMBYX
MORI.

k. Sakamoto and I. Shimizu. Res. Sect. of
Environ. Biol., Labo. for Plant Ecol., Fac. of
Science, Kyoto Univ., Kyoto.

The silkworm larvae were reared on a
carotenoid(CARO)-deprived artifical diet, and
hatching rhythm of the offspring larvae was
investigated in order to examine an involve-
ment of CARO and/or vitamin A in the circadian
rhythm of insect. The hatching patterns of the
larvae(MO) from the CARO-deficient moths, from
lutein-injected moths(MO+lutein) whose visual
photosensitivity was confirmed to recover, and
from moths which had been reared on mulberry
leaves(ML) were examined, respectively. The
Magnitude of phase-delay or phase-advance in
the circadian rhythm of CARO-deprived larvae
was not siginificantly different from CARO-
rich population(ML). However, the first peak
prouced by a Single light-pulse exposed in DD
appeared significantly faster than ML-group
and supplement of lutein delayed the peak
nearly to that of ML. When the eggs were kept
in LD 4:20 regime, CARO-deprived larvae
emerged at around light-on with two peaks: one
was found in scotophase and one just after
light-on. These two gates free-run in DD-
condition. CARO-rich larvae(ML and MO+lutein)
showed only one peak just after light-on.

Physiology 1029

PH 43

THE EFFECT OF LIGHT ON THE OPSIN SYNTHESIS
IN THE EYE OF DROSOPHILA.

H.Nagatani, M.Ozaki, F.Tokunaga, K.Ozaki.
Depot “Biola hacia OL Sica), Osaka Univ. ,
Toyonaka.

In order to study the effect of light
on the opsin synthesis of Drosophila, we
partially purified rhabdomal membranes of
photoreceptor cells. Opsin-deficient flies
(ninaE mutants or flies reared on vitamin
A-free (VA-) medium) lacked the 35kDa
peptide which was found in the purified
membrane of normal flies. This indicated
that the peptide was identical with opsin.

Neither chromophore nor opsin is syn-
thesized in VA- flies. By feeding these
flies with all-trans retinal in the dark,
all-trans 3-hydroxyretinal (AT-RAL) is
accumulated in their heads. Because blue
light irradiation starts such flies syn-
thesizing rhodopsin, it is suggested that
the light causes opsin synthesis as well
as 11-RAL formation. In the present study
we showed that the incorpolation of -
Met into 35kDa peptide was trigered by
blue light irradiation. The results di-
rectly indicated that opsin synthesis was
initiated with blue light.

Opsin synthesis was also accelerated
with blue light in VA+ flies. However, the
newly synthesized opsin appeared at 37kDa
in the irradiated flies. When green or red
light was given just after blue light,
the opsin band was observed at 35kDa.
Synthesis of 37kDa opsin after blue irra-
diation is probably due to the large
amount of metarhodopsin in the retina.

~

PH 44

SOME PROPERTIES OF SUBSTANCES CONTRIBUTING

TO THE EXPERIENCE-DEPENDENT DEVELOPMENT OF

FLY VISUAL FUNCTION

K.Mimura, Nagasaki Univ., Fac. Liberal Arts
Nagasaki 852.

It has been demonstrated using HPLC with
electrochemical detection that two optic
lobe-dominant substances (biogenic amines
and/or related substances) increased tran-
siently during the early period of post-
emergence in visually experienced flies,
although these have not been identified
yet. As the transient increase was not seen
in flies reared in continuous darkness, it
was suggested that the two substances were
concerned with experience-dependent and
plastic development of visual function. The
transient increase of these substances was
not seen when anisomycin, an inhibitor of
protein synthesis, was injected into the
fly immediately after emergence. This re-
sult proves that these substances particl1-
pate in a mechanism of the plastic develop-
ment of vision. Upon light deprivation of
the unilateral compound eyes immediately
after emergence, the transient increase of
the two substances was not observed in the
optic lobe of the deprived side, but only
in the intact side. This shows that the
plastic development mechanism is localized
only in the optic lobe. These results sup-
port the facts revealed previously by
behavioral experiments.

1030

PH 45

42KD PROTEIN IN RHABDOMERIC PHOTORECEPTOR
OF CEPHALOPOD

H.Uchiyama, K.Narita, M.Seido and Y.Kito.
Does Ope lastoilo, iMretes Oi Sens, Obese iis
Toyonaka, Osaka.

Previously we studied the nucleotides
contents in the rhabdomeric photoreceptor
and concluded the cyclic GMP or cyclic
AMP did not participate in the primary
phototransduction process. In this study,
we searched G protein in the rhabdomeric
photoreceptor of cephalopod. Any protein
like transducin in vertebrate photo-
receptor could not be detected by the
SDS-PAGE analysis of the membrane
preparations. However, we noticed that in
the supernatant fraction of 15,000g
centrifugation, the 42KD band was reduced
in the illuminated membranes. Two
Dimensional Electrophoresis showed that
the 42KD band contained at least two
proteins. Some part of the 42KD band
reacted with the
Quantitative analysis by DNase I could

actin-antibody.

not demonstrate a significant difference
in light and dark preparations.

PH 46

PHOTOSENSITIVITY OF CONE VISUAL CELLS
TevoshizaWa. — DEpiis ols Biophys.) dCs
of Sci., Kyoto Univ., Kyoto.

Several distinctive features of cones
compared to rods were explained in terms
Of the properties of chicken iodopsin.
lodopsin was found to be similar in photo-
sensitivity to, but lower in quantum yield
than bovine rhodopsin. Picosecond laser
photolysis of jodopsin at room temperature
confirmed the formation of bathoiodopsin
but failed to detect-any eaylierm inter-
mediate corresponding to photorhodopsin.
These results are consistent with the
lOWweR quantum. ya eld. of iodopsain:
Nanosecond laser photolysis demonstrated
rapid formation and decay of Meta I]
jodopsin. Since Meta II rhodopsin has
been shown to have an enzymatic activity,
this result may partly account for the
rapid generation and decay of the cone
RECEDECE POLENE dl wine iacks, likes Meta
II rhodopsin, Meta II iodopsin was shown
to catalyse the exchange reaction of GDP
to GIP on bovine rod transducin, so that
the rapid decay may also account for the
low sensitivity of cones. lIodopsin was
also phosphorylated in the presence of ATP
under irradiation by bovine rhodopsin
kinase. The low Km of the photo-phospho-
rylation obtained suggests the rapid
recovery of the cone receptor potential.
In conclusion, the low sensitivity and
rapid response of cones to light could be
explained by lower amplification in the
enzyme cascade process.

Physiology

PH 47

CLONING OF CHICKEN GENOMIC DNA MOLECULES
ENCODING V}SUAL PIGMENTS.

F.Tokunaga 4 T. Iwasa“, M.Miyagishi®, and
S.Kayada’. Dept. of, Brolt , = hacemotiuw scm
Osaka Univ., Toyonaka, “Dept. of life Sci.,
Himeji Inst. of Tech., Himeji, Byol. Seas.
Lab, Kao Co. ,kiD =, Lochara vandias) opiates
Phys., Fac. of Sei., Tohoku Univ., Sendai.

From comparison of sequence homology
among rhodopsin and color visual pigments
of human, it was predicted that visual
pigments have evoluved along three lines
(rhodopsin, blue sensitive pigment and
gree-red sensitive pigment). Genonic
library of chicken was screened by DNA
fragments encoding chiken rhodopsin, human
blue sensitive pigment and red sensitive
pigments and eleven groups of colonies were
classified by the grade of southern blot
hybridyzation. Nucleotide sequence of one
of them was sequenced (19D). With the
fragment of 19D as a probe, cDNA library
was screened and clone 201 was selected as
a positive clone. The nucleotide sequence
of the clone 201 was sequenced. The
sequence was very much similar to that of
human and fish red sensitive pigment.
Therefore the clone 201 was estimarted to
be chicken red sensitive pigment.
Comparison of amino acid sequences among
red and green sensitive pigments of human,
fish and chicken showed that amino acid
residues close to N-terminal substituted
very much among them, but resirdues inside
the membrane and at c-terminal region are
highly conserved.

PH 48

PHOTORECEPTION BY THE PINEAL COMPLEX:
IMMUNOHISTOCHEMISTRY AND HPLC.

H. Masuda, M. Otand) 2S Ontshal You Bulkaday,
T. Yoshizawa and M. Michinomae. Dept. of
Biol, Each of Scala, )Nazalwomentsp a Uneaven
Naa Dept. “of Biophystip iace ml OiwmS Cole,
Kyoto, “Unive, Kyoto. Depa OE Ola tac.
of Sci., Konan Univ., Kobe.

We investigated the photopigments of
the pineal complex in birds, reptiles and
amphibians using immunohistochemistry with
anti-serum against bovine rhodopsin (anti-
Rh-Ab) and monoclone antibodies against
chicken iodopsin (anti-Io-mAb) and the
HPLC analysis of retinoids.

The photoreceptor cells in the pineal
organ of the Japanese quail (Coturnix
coturnix japonica) exhibited immuno-
reactivity to the anti-Rh-Ab, but not to
the anti-Io-mAb. The HPLC analysis
revealed the existence of retinal and
retinol. All photoreceptor cells in the
parietal eye and pineal organ of the
Japanese lizard (Takydromus
tachydromoides ) were immuno-negative to
the anti-Rh-Ab. On the other hand,
immunoreactivity to the anti- Io-mAb was
observed in some photoreceptor cells of
the pineal organ. In the bullfrog (Rana

catesbeiana), the frontal organ and
pineal organ were immuno-positive to both
of the anti-Rh-Ab and anti-Io-mAb. On

HPLC analysis, retinol and 3-dehydro
retinol were found in the pineal organ and
carotenoids were found in the frontal
organ.

Physiology 1031

PH 49

IMMUNOHISTOCHEMISTRY ON THE LOCALIZATION
OF VISUAL PIGMENTS IN THE RETINA OF BIRDS
AND REPTILES.
A, TKaweitta, Wo Onslow, Wo imwUxewley = etal We
MoshizavalDeptyTor Bvoll, Hac. | Of "SCI,
Nara Women's Univ., Nara, Dept. of Bio-
phys., Fac. of Sci., Kyoto Univ., Kyoto.
Localization of cone visual pigments
were compared immunohistochemically by
using monoclonal antibody against chicken
iodopsin (anti-Io mAb) and anti-serum
against bovine rhodopsin (anti-Rh Ab)in
the retina of chickens and turtles. In
the chicken retina, the outer segments of
cone cells with red (single cone) and
green (double cone) oil droplets (Od) were
labeled by anti-Io mAb, while those of
cone cells with yellow and clear Od
(single cone) were by the anti-Rh Ab.
Neither the anti-Io mAb or the anti-Rh Ab
labeled the cone cells with blue Od.
These results indicate that iodopsin
localizes in cones with red and green Od,
and that the visual pigment in cones with
blue Od is different from those in cones
with yellow and clear Od. In the turtle,
the outer segments of cone cells with red
(single cone) and yellow (double cone) Od
were stained by both anti-Io mAb and anti-
Rh Ab. The outer segments of cones with
pale-green Od (single cone) were labeled
only by the anti-Rh Ab, and those with
Orange and clear Od were not stained by
either of the antibodies. Thus, at least
three visual pigments exist in cones of
the turtle.

PH 50
Synthesis of 3-hydroxyretinal in the cytosol of the
butterfly compound eye
Youji Shimazaki and Eisuke Eguchi.

Department of Biology, Yokohama City University,
22-2 Seto, Kanazawa-ku, Yokohama 236, Japan.

The visual pigment chromophore of a butterfly, Papilio
xuthus, is 3-hydroxyretinal (RAL3). Most of the RAL3 had
been thought to exist within the rhabdom made up by the
photoreceptive membrane. However, the eye of Papilio con-
tains more RAL3 than the amount calculated from rhabdom
volume and particle density on the rhabdomeric membrane:
i.e. RAL3 also exists outside the rhabdom (Eguchi et al, J
Comp Physiol, 165:589,1989).

In the present study, we firstly measured the RAL3
amounts in the precipitation (ppt) and in the supernatant (sup)
of a Papilio eye homogenate (25,000g,30min) by HPLC. In
the ppt, which contains most of the rhabdomeric membrane,
70 to 75% of the total amount of RAL3 was detected, whereas
the sup contained the rest (25 ~ 30%). The sup also contained
a considerable amount of 3-hydroxyretinol (ROL3).

The relative amount of all-trans RAL3 in the sup and the
ppt is much higher in the light adapted eye. We irradiated the
sup of the light adapted eye with different monochromatic
lights or kept under constant darkness for 2 hrs, and found
that blue light (470nm) resulted in an increase of 11-cis and
all-trans RAL3s by a factor of ca.3 and 2 respectively.

Finally, we added 11-cis retinol (not found in the butterfly
eye) to the sup. The non-native chemical was also converted,
at least in part, into 1 1-cis and all-trans retinal.

These results indicate that the sup contains some blue light-
dependent system that synthesize RAL3. ROL3 in the sup is
one of the candidate substances for the RAL3 precursor.

PH 51

MOLECULAR STRUCTURE OF SQUID RETINAL-
BINDING PROTEIN (III).
K.Ozaki, A.Terakita, M.Ozaki, R.Hara and
T.Hara. DEE O12 BWilOls,, wae. O12 SEs 5
OsakasUnivew seLovondkaL eee
In order to study the relationship
between molecular structure and function
of squid retinal-binding protein (RALBP),
we tried the expression of cDNA for RALBP
in E. coli. Little amount of RALBP poly-
peptides were yielded when EF. coli cells
were transformed with the full length cDNA
in pTTQ18 vector and incubeted at 37°C. We
then produced the deletion mutant which
lacks the polynucleotide at 5' end of cDNA
covering the whole 5' nontranslated region
and the following 50 nucleotids of trans-
lated sequence. Although this mutant gave
an better expression, the product comp-
letely lost the retinoid-binding activity
probably due to the thermal denaturation
of protein. To avoid the denaturation, we
carried out the expression of RALBP at
lower temperature WSHEMIGn ain) els evalet aco
translation system as follows. The full
length cDNA was subcloned in pBS vector to
obtain mRNA of RALBP. The mRNA was then
translated at 20°C in a rabbit reticulo-
cyte translation system. The assay with
the 3H-labeled retinol showed that the
synthesized RALBP had binding activity to
externally added retinol. When mRNA from
the 5'-deleted cDNA was used for transla-
tion, the product was: not capable of bind-
ing retinol. This result suggests that N-
terminal polypeptide is indispensable for
functional expression of RALBP.

PH 52

AXIAL DIFFUSION OF RETINOL IN SINGLE RODS
FOLLOWING ,BLEACHES OF RHODOPSIN.

Ko Azuma , M., Azuma , A.E.Walter and
T.P.Williams’. Dep. of Biol., Osaka
Med.Coll., Takatsuki, Osaka 569, Dep. of
Health Sci. ,Osaka, Kyoiku Univ.,Hirano-ku,
Osaka 547, and Inst. of Mol. Biophys.,
Florida State Univ.,Tallahassee, Florida
32306-3015, USA

Microspectrophotometer(MSP)measurements
of isotated single rods of frog Rana
pipiens indicated that the } max of
rhodopsin in rod outer segments (ROSs)
was 504.2 + 2.2 (SD) nm and the value of
linear dichroism (Ai/Ay) at the ),max of
rhodopsin was 4.4 + .85 (SD). MSP measure-
ments also suggested that all-trans-
retinol formed from bleached rhodopsin at
an irradiated region gradually diffused
into the non-irradiated region along the
ROS long axis. High-pressure liquid chro-
matography (HPLC) analyses of retinoids
in isolated retinas before and afte irra-
diation indicated that stored retinol was
less in dark-adapted retinas and the final
photoproduct of rhodopsin, all-trans-
retinol, remained stable in the retina
for at least 4 hr at 25°C.

These results are consistent with the
accepted model of the visual cycle which
requires diffusion of retinol across
multiple membranes.

1032 Physiology

PH 53

STRUCTURE IN THE PROTEIN MOIETY OF
RETINOCHROME AND ITS DERIVATIVES
A.Kishigami A N, Sekiya . K.Yoshihara
R.Hara , T.Hara , and F.Tokunaga
Dept. Of Biol. ac. Ofsscin,, Osaka) Unive,
Toyonaka, Suntory Inst. for Bioorganic
Res., Mishimagun, Osaka, Japan.
EAST aac RSENS BS Se ae TP AV ATE LE AE SA a Ne
A photopigment, retinochrome, is located on
the myeloid bodies in teh visual cells of
cephalopods. Spectrophotometry for retino-
chrome and its derivatives in digitonin
solution (pH 6.5) was carried out to deter-
mine the conformation of the protein. Cir-
cular dichroism (CD) measurements in the UV
region (190-250 nm) indicated that retino-
chrome has a high helical content which is
not altered accompanying the conversion of
retinochrome to metaretinochrome. The CD
spectrum of aporetinochrome was identical
with that of regenerated retinochrome,
suggesting that the insertion of the chro-
mophore to aporetinochrome does not change
the secondary structure. The absorption
spectrum in the near UV region (250-350 nm)
was analyzed with fourth derivative to
seperate the absorption bands due to aro-
Matic amino acid residues of retinochrome.
Fourth derivative spectrum indicated that
tryptophan residues are located in a hydro-
phobic core in retinochrome, and that this
environment does not change after photocon-
version and regeneration. These data sug-
gest that retinochrome is a transmembrane
protein in the myeloid body.

’

PH 54

A NEW VISUAL PIGMENT BASED ON 4-HYDROXY
RETINAL GENERATED FROM BOVINE OPSIN.

Y.Kitol M.SeidoulK.Narital
M. Sawada“ K. Yoshihara?

Inept. OM IOl,, HAGs Cie SEs, OsSelei
2

M.Michinomae~

and J.Partridge?

Univ. Toyonaka, “Dept. of Biol., Konan

Univ. Kobe, “Inst. of SUNBOR, Mishimagun,

4 Dept. Gis yAoYoils, wredie@ll, Wino 7 lswealisieoibe

Both (S)-11-cis-4-hydroxyretinal from
eyes of a squid, Watasenia scintillans
and the chemically synthesized enantiomer
generated a 486 nm A4 visual pigment (VP)
from bovine opsin. Shape of abs.

spectrum, Molar Ext. coefficient and
photosensitivity of the A4 VP were very
Siiubilewe teoy Aelovomyss ~~ @ae.. elas} Vil | WAP o On

iyradiation at 7 €C, Meta I (470 nm) and
meta II (376 nm) intermediate appeared,
being stable as those of Al VP. The CD at
a band was almost identical but about
twice at £@ band. The 4-hydroxyretinal

from A4 VP generated with enantiomer,

showed the negative CD. This indicates
that bovine opsin can discriminate
echamalkiitty “of Sthe “retinal: Above facts

suggest not only the squid but’ some
vertebrates may utilize this chromophore
for VP adaptation to shorter wavelengths.

PH 55

POLARIZATION SENSITIVITY OF RETINULA CELLS
OF OCTOPUS AND CRAYFISH.

ae Gleadall', Ab Hariyama and Y Tsukahara?.
INIBB, Okazaki; 2Res. Center for Applied

Informatn. Sci., Tohoku University, Sendai.
The octopus and the crayfish both have a
remarkably high polarization sensitivity
(PS), based on just two receptor channels,
which are orthogonal. Possible functions
for such high PS values (probably »>10 in
the octopus, and in the crayfish
Procambarus may be as high as 40) include
enhancement of visual discriminations
based on contrast detection. Intra-
cellular recordings from the eyes of
intact 'summer' crayfishes reveal a peak
sensitivity in the retinula cell at close
to 600nm, regardless of the e-vector
quality of the light stimuli. However, the
consistent presence of a shoulder at ca.
480nm implies that the crayfish provides a
good model system for studies of (i) the
absorption characteristics of the
rhodopsin molecule, and/or (ii) the
derivation of contrast functions, which in
the two-channel system can be expressed in
terms directly related to the PS value. It
is argued that the unusually high PS of
the crayfish indicates that contrast
discrimination (analogous to colour
vision) may be important. Further, the
possible involvement of retinula cell R8
in deriving contrast is implicated, and
other possible functions may include a
role in adaptation. The latter function in
the squid may be one role of the photo-
sensitive vesicles.

PH 56

THE PRIMARY STRUCTURE OF CHICKEN RED
SENSITIVE CONE VISUAL PIGMENT, IODOPSIN.
T.Okano, O.Kuwata, Y.Imamoto, K.Kokame,
D.Kojima, H.Matsumoto, A.Morodome,
Y.Fukada, Y.Shichida, T,Yoshizawa. Dept.
of Biophys.,; Fac:.0f,SGio..KyGeo Unive.
Kyoto.

A purified iodopsin was digested Dy
CNBr or several proteolytic enzymes
(chymotrypsin, lysyl endopeptidase, and
Staphylococcus aureus protease) into
fragments, the amino acid sequences of
which were determined. A partial sequence
of the C-terminal region (FRNCILQLFGKKV
DDGSEV) was utilized for synthesizing an
Oligonucleotide probe (56mer), Dy which
the iodopsin cDNA (1339 bases) was
determined. The amino acid sequence of
iodopsin deduced from the cDNA (362
residues) included all the sequences
determined by the peptide analysis, and
Showed 80%, 42% or 43% homology to that of
human red-sensitive cone pigment, cattle
or chicken rhodopsin, respectively. The
hydropathy profile of iodopsin implied
that it has seven transmembrane x -
helical segments like rhodopsin. — The
cleavages in the middle of the segment VII
by chymotrypsin suggest that unlike
rhodopsin, iodopsin may have a hydrophilic
pocket near the seventh segment, that 1s
consistent with the accessibility of the
chromophore binding site of iodopsin to
various kinds of chemicals.

Physiology 1033

PH 57

Primary and higher-order chemosensitive
neurons in the cricket (Grylius
bimaculatus)

M. Yokosuka and T. Yamaguchi*

Tochigi Res. Lab., Kao Corp., Tochigi and
*Dept. of Biol., Fac. of Sci., Okayama
Univ. , Okayama

Five types of sensory hairs were identified
on the basis of their morphological
characterstics in the anatennae of
crickets. Backfilling of the antennal nerve
with NiCl2 revealed that the majority of
primary sensory neurons have terminal
arborizations within the ipsilateral
antennal lobe of the deutocerebrum, and a
small number of primary sensory neurons run
round the outside of antennal lobe,
projecting in the ipsilateral half of the
suboesophageal ganglion. The sexual
dimorphism was found in the glomerular
organization of the antennal lobes: there
are 16 spheroidal glomeruli in each lobe of
both sexes, but one macroglomerulus only in
the male. Many primary sensory neurons
responded in dose-dependent manner to
chemicals such as NaCl, LiCl, and Glucose.
The higher-order interneurons recorded from
the nerve connectives between the
Suboesophageal and prothoracic ganglia
responded also to the chemicals in the same
way as the primary sensory neurons. The
wind stimulation to the cerci had no effect
on the responses of primary sensory neurons
TOnmmeMNew Chemnucalis )eiwhiwle= sate) had can
inhibitory effect on the responses of
higer-order interneurons

PH 58

ENHANCEMENT OF XENOPUS GUSTATORY RESPONSES
TO THE MIXTURE OF ACID AND SALT.
S.Yamashita!, Y.Sunayamal, Y.Yamamotol and
H.Kageura?2. DEVOW5 Cit HLOIES,, COILS Orit Wpllo),
Arts, Kagoshima Univ., Kagoshima, 2Dept. of
Biol., Fac. of Sci. , Fukuoka Univ., Fukuoka,

The stimulating effects of the mixture of
acid and salt on chemoreceptors were exam-
ined by recording the neural activity of
the glossopharyngeal nerve of the African
clawed toad (Xenopus laevis).

Responses to HCl were greatly enhanced
over the concentration range of 0.0005 M to
0.01 M by the presence of NaCl of concen-
trations between 0.003 M and 0.3 M.

Responses to the mixture increased with
decreasing pH of HCl or with increasing
concentration of NaCl, and the greatest
relative enhancing effect was detected at
about pH 3.0.

Enhancing effects by the mixture were
Similar at the same pH, regardless of the
sorts of acids, if it contained the same
salt of the same concentration.

HCl solutions mixed with 0.3 M chloride
salts resulted in similar responses, but
those mixed with 0.3 M sodium salts caused
large differences in responses, suggesting
the contribution of the anion of the salt
to the enhancing effects by the mixture.

The order of the enhancing effects due to
chaotropic anions which form water-anion
bond with water molecules was F<Cl<Br<I <
ClO4<SCN. These ions may neutralize the
positive receptor membrane so that Ht ions
will be easily bound to the receptor sites.

PH 59

STRETCH RECEPTORS OF A MARINE ISOPOD,
BATHYNOMUS DOEDERLEINI, WITH SPECIAL
REFFERENCE TO THE MORPHOPHYSIOLOGICAL
COMPARISON WITH STRETCH RECEPTORS OF A
TERRESSTRIAN ISOPOD, ARMADILLIDUM VULGARE

A. Niidal, T. Yamaguchil, T. Yazawa2, and
KR, Tamassia Angee, Blois, ees Sesh,
Okayama Univ., Okayama and Dept. Biol.,
Fac. Sci., Tokyo Metropolitan Univ., Tokyo

Segmentally arranged stretch receptors
were found in the thorax and abdomen of a
marine isopod, Bathynomus doederleini.
Their spatial arrangement was similar to
that of stretch receptors of a terrestrial
isopod, Armadillidum vulgare (Niida et al,
1990a,b): all the stretch receptors have
paired structure in which the receptor
muscles of each pair run closely together.
NO difference between sarcomere length of
one member and that of another member in
each pair of thoracic receptor muscles of
Bathynomus and both thoracic and abdominal
receptor muscles of Armadillidum was
measured. However, the sarcomere length of
one member was much shorter than that of
another member in each pair of abdominal
stretch receptors of Bathynomus. These
anatomical features were good agreement
with the physiological features that the
Slowly adapting spike discharges occurred
in response to stretch stimuli in each
pair of thorasic stretch receptors of
Bathynomus and all stretch receptors of
Armadillium, while both the slowly and
fast adapting spike discharges were evoked
by stretch stimuli in each pair of
abdominal stretch receptors of Bathynomus.

PH 60

FEEDING MOTOR OUTPUTS AND TASTE RESPONSES
OF THE CEREBRAL NEURONS IN THE SLUG,
Limax flavus.

A. Iwama, T. Sekiguchi and A. Mizukami.
Tsukuba Research Center, SANYO Electric
Co.,Ltd., Ibaraki.

As a step in approaching the neural
basis of feeding behavior in the slug, feed-
ing motor outputs induced by taste stimuli
to the lip were recorded from the external
lip nerves (ELNs) and the buccal nerve roots
(BRs). Cerebral neurons relating to feeding
motor outputs were examined intracellularly.

In the intact central nervous system
(CNS), patterned rhythmic activities were
observed by the taste stimuli in the ELNs
and BRs. When we cut the cerebro-buccal
connective (CBC) on one side, taste stimuli
to the side could not induce the rhythmic
activities, whereas the activities were
observed by the stimuli to the another side.
The elicited rhythmic activities were
recorded in both side of ELN. Rhythmic
activities similar to those induced by taste
stimuli were also observed in the isolated
buccal ganglia (BG). In addition, we found
cerebral neurons which sent their axon into
ELN, showed rhythmic activities by taste
stimulus to ipsi- or contralateral lip, and
received only ipsilateral sensory input.

These results suggest that the sensory
input which induces feeding rhythm is
transmitted through ipsilateral CBC to the
BG, the feeding motor rhythm is produced by
the network in the BG, and the rhythm is
sent back to ELNs bilaterally.

1034 Physiology

PH 61

RESPONSES OF BRAIN AND THORACIC GANGLION
NERVE TO BOMBYKOL IN THE MALE SILKWORM MOTH
TT Shabuyasands Nesug.c. mnSt. OF B1Ole isCis.5,
Univ. of Tsukuba, Tsukuba.

We already reported on responses of the
protocerebrum neurons to Bombykol in the
male silkworm moth brain. Responses of
the neurons by intracellular recording app-
eared remarkable tonic type of impulse.

The dose-response curve pattern strongly
resembled to that of wing vibration to Bom-
bykol. To know responses of output nerves
to Bombykol, we observed mass responses of
the protocerebrum in the brain and the tho-
racic ganglion nerves. Wave discharges of
the brain to Bombykol were recorded well,
and they mainly increased discharges in du-
ration according to concentration increase.
The thoracic ganglion consisted of the pro-
,ymeso- and meta-ganglion has many nerve
branches to the fore-, mid- and hind-legs,
and also to the fore- and hind-wings. In
the output nerves to legs, the fore- and
mid-leg nerves responded with phasic type
of impulse, however hind-leg nerves respo-
nded with tonic type of impulse. These
impulse discharges increased in frequency
with concentration increase of Bombykol.
Long duration of discharges was regorded
from nerves to the hind-wing at 10 . The
discharges may be initiated wing vibration
in their mating dance. Sometimes, from
nerves to the hind-leg and hind-wing, rhy-
thmic discharges were observed. This type
of discharge may be useful for orientation
toward a female moth.

PH 62

VOLTAGE-DEPENDENT BLOCK OF SINGLE cAMP-
ACTIVATED ION CHANNEL CURRENTS BY DIVALENT
CATIONS IN OLFACTORY RECEPTOR CELLS.

No Suvsplicat, wool, Weste,, PAG Ce Silo;
Hokkaido Univ., Sapporo.

Cyclic AMP-activated ion channels of
frog olfactory receptor soma were studied
by measuring single channel currents from
inside-out patches. Channels were activa-
ted by exposing the internal surface of
the patch to 10 uM cAMP in a microflow
Superfusion chamber. In symmetrical
solutions containing 115 mM Na and no
divalent cations, unitary current-voltage
relation exhibited a weak outward
rectification and gave a unit conductance
of 23 pS at + 60 mV. Addition of divalent
cations of physiological concentrations
(1 mM Ca and 1.6 mM Mg) on the external
side of the membrane blocked the Na
current in the negative voltage range,
giving a strong outward rectification.
Addition of divalent cations(0.1 uM Ca and
1.25 mM Mg)on the internal side of the
membrane also blocked the current in the
positive voltage range and gave a strong
inward rectification. Mg ions(with 0.1 uM
Ca) added on the internal side of the
membrane blocked the current ina
saturable manner with the Ki /2 Ge 30) > 50
uM. Results suggest that the cAMP-
activated ion channels in olfactory
Beicepitomucelks iconitaginegwal haigh walt fainabty,
binding site for divalent cations which is
located within the channel pore.

PH 63

NEURAL CIRCUIT OF THE CENTRAL PATTERN
GENERATOR OF THE STOMATOGASTRIC GANGLION IN
SQUILLA.

K.Tazaki. Biol.Lab., Nara Univ.Educ., Nara.

Movements of the posterior cardiac plate
(pcp) and pyloric regions of the stomach in
Squilla are controlled by a central pattern
generator of the stomatogastric ganglion in
which most of the neurons are identifiable.
The neurons in the pcp-pyloric system, pcp
constrictor (PCP). pyloric. cons bua chorea)
and pyloric dilator (PD) motorneurons, and
pyloric interneuron (AB) can _ produce. se-
quentially patterned motor outputs. PCP
neuron activity is followed by activity of
PY neurons, and then by activity of PD and
AB neurons. These motorneurons and inter-
neurons are connected by either electricai
or inhibitory chemical synapses, and thus
make a characteristic neuronal network to
be expressed in a wiring diagram. All
neurons involved in the pcp-pyloric system
exhibit endogenous properties to produce a
Slow, regenerative potential which leads ta
burst discharges. The cellular and network
properties of the pcp-pyloric system in
stomatopods are remarkably similar to those
of the homologous pyloric system in
decapods although the former species belong
to much more phylogenetically primitive
crustacean group than the latter. The
stomatogastric nervous system of Crustacea
provides a model system for systematic
comparative analysis of the neural circuit.

PH 64

AFFERENT NEURONS AND THEIR ACTIVITIES IN THE
SWELLING LOCATED IN THE BASEMENT OF FORE-
WING ON THE CRICKET, GRYLLUS BIMACULATUS.
T.Tsurugai and N.Ai. Dept. of Biol., Tokyo
Gakugei Univ., Koganei, Tokyo.

The swelling, new findings, located in the
basement of fore-wing in cricket is continu-
ously touched to the tergum of prothoracic
segment during flight movement of the fore-
wing. Afferent neurons extend from meso-
thoracic ganglion to the swelling along N1Cla
neuron branching,that is Nl lateral root of
the mesothoracic ganglion.

Innervation of the afferent neuron of the
swelling was investigated by back-fill method
and traditional procedure of transverse sec-
tion (15ym) of paraffin-embeded specimen of
the mesothoracic ganglion.

Neural contacts of these afferent pathes to
the other neurons, that is the dorsal, ventral
and median intersegmental tract, were ipsi-
laterally distinguished. Neuron branch of the
both side one was clearly contacted with the
dorso-midline of ganglion at the distal end.

Mechanical stimulation given to this swell-
ing surface regularly induced on afferent path
with phasic and tonic impulse train as unitary
discharges.

However, when the free flight was evoked by
air-puff stimulation to head, an induced spike
train with short duration was rised on the
instant of just initiation phase on flight
movement.

Excitation of swellig in fore-wing may be
arised “on signal” for flight movement.

But, as already reported, the innervation
and neural activity from hind-wings swelling
have a little difference from the fore-wing
ones (1982).

PH 65

NEURONAL LABELING OF THE GLOSSOPHARYNGEAL
NERVE WITH FLUORESCENT DYES (DiI AND Di0O)
DNSREX ED) TISSUES.

T. Nagai. DEDE6 5. PIMSTOlses Tevkivo,~ Wimarwe
SiGhtas Media ne On VOnmlnsat.

We have shown with Co-lys_ labeling
(Nagai and Matsushima, '89) that
Organization of the glossopharyngeal (IX)
nerve ganglion in axolotls differs from
that in other vertebrates: ganglion cells
of the IX nerve are not clearly separated
from those of the vagus (X) nerve. To
label ganglion cells. of two nerves
differentially, fluorescent carbocynanine
dye, Dil or DiO (2.5% in 50% EtOH and 50%
DMSO), was applied to the peripheral cut
end of either of the IX or the X nerve in
formaldehyde-fixed tissues. They were
maintained in fixative at 37° C for up to
two months, embedded in gelatin, sectioned
on a vibratome, and examined with
epifluorescence. DiI produced a = bright
red-orange fluorescence when viewed with
rhodamine filters. Peripheral axons of the
cells were clearly labeled and pathways of
the central axons to the brainstem were
also discerned. In the cell body, the cell
envelope was more intensely fluorescent
than the cell nucleus, suggesting lateral
diffusion of the dye in the plasma
membrane. Through fluorescein filters, Dil
fluoresces yellow, while DiO fluoresces
green. However, visualization of the IX
and the X nerve cells respectively with
two dyes was not successful, because
diffusion of DiO was much slower than Dil.

PH 66

Sensory organs in the cerebral vesicle of
the ascidian larva, Botryllus schrosseri.
He OMEsukicosinst. Boles back “Educe. O71 ta
Univers Olitiay.

Sensory organs in the cerebral vesicle
of the larva of Botryllus schrosseri were
investigated by SEM. The cerebral vesicle
of this species was attached to the anter-
lor right surface of the sensory center of
the visceral ganglion situated at the
center of the larval trunk. A sensory
organ, photolith, which has been reported
to be sensitive to both light and gravity
was seen on the postero-ventral wall of
the cerebral vesicle. The photolith con-
sisted of a pigment cup (statocyte) and a
light-sensitive organ which invaded into
the pigment cup. The pigment cup was at-
tached to the ventral wall of the cerebral
vesicle with a narrow junction. Two pro-
tuberances which sent many fibrilar pro-
jections from thir apical ends to the
posterior surface of the pigment cup were
newly found. The distal end of the light-
sensitive organ was divived into 4~5
branches in the pigment cup. The top of
the branches contained several piled men-
branes. On the surface of the light-sen-
sitive organ, except at the top of its
branches, many ridges, consisting of small
particles were found. Several protuber-
ances, resembling the hydrostatic pressure
receptor in other species existed in the
larval cerebral vesicle of 8B. schrosseri.

Physiology 1035

PH 67
CHEMOSENSILLUM OF THE BLOWFLY, PHORMIA

REGINA; ISOLATION AND PROTEIN COMPOSITION

M. Ozaki, K. Ozaki and F. Tokunaga. Dept.
Of PBOle webaG Or SCal eaUMan. Of eOSaikal,
Toyonaka.

On the labellum and the tarsus of the
fly, many chemosensilla are found. Four
taste cells having different functions
extend their sensory processes into each
sensillum, making a sensory unit.

We succeeded to isolate the sensilla
from the labella by modifying the method
of Kaissling, which has been applied to
the anntenal sensillum isolation in the
silk moth. The sensilla include no cell
bodies of the taste cells and no other
tissues and cells, but preserve the
sensory processes of the taste cells.

Being compared with the extract of
whole labella, the extract of isolated
sensilla included limited kinds of
protein. Among the proteins, a major
soluble protein of 25 kDa in the molecular
weight was first isolated and determined
its N-terminal amino acid sequence. It was
an acidic protein and 7 of 35 amino acid
residues were glutamic acid. This protein
was detected not only in the labellum but
also in the tarsus, but not detected in
the haustellum, the proximal part next to
the labellum. Labellum and tarsus are
chemosensillum rich but haustellum has no
chemosensilla. Thus, the soluble protein
was likely to be concerned with the
mechanism of taste reception of the fly.

PH 68

A NOVEL OXYTOCIN-LIKE PEPTIDE ISOLATED FROM
THE NEURAL COMPLEXES OF TUNICATE, Styela
plicata.

M.lwakiri', A.Sugiyama'!, T. Ikeda”, Y.Muneoka?
and [.kubota*. ‘Dept. of Biol., Fukuoka
Univ. of Educ., Munakata, *Fac. Integrated
Arts and Sci., Hiroshima Univ., Hiroshima
and *Suntory Bio-Pharma Tech Center, Gunma.

By using strips of the body-wall muscle
of Styela plicata as bioassay systems,
several bioactive substances were suggest to
be present in methanol extracts of the
neural complexes of the animal. One of the
substances was shown to be a peptide having
an excitatory action on the Styela muscle.
The peptidic substance was purified by a
HPLC system, and its structure was
determined by amino acid sequences analysis,
amino acid analyses and FAB-MS analysis.

The determined structure is as follows:

| uae ssa, Uae alles Ean Naciaa! Gaed
H-Cys-Tyr-Ile-Ser-Asp-Cys-Pro-Asn-Ser-
Arg-Phe-Trp-Ser-Thr-NH2

That is, the substance was found to be a
oxytocin-like peptide. The peptide was
termed SOP (Styela oxytocin-like peptide).

The examinations of actions of SOP on
other animals are now in progress.

1036 Physiology

PH 69

REFLEX ACTIVITY OF CARDIOARTERIAL-VALVE
NERVES IN RESPONSE TO ANIMAL MOVEMENTS AND
EXTERNAL STIMULI IN BATHYNOMUS DOEDERLEINI.
J. Okada and K. Kuwasawa Dept. of Biology,
Tokyo Metropolitan University, Tokyo.

Five pairs of lateral arteries arise
Ee ielN@) In@QeSlieie'a Cardio-arterial valves
composed of a pair of flaps are located at
the junction between the heart and
arteries. The valves are innervated by
lateral cardiac (valve-dilator) nerves
(LCNs) containing one or two axons.

Tactile stimuli applied to the body
surface decreased impulse rate of LCNs.
The Sth HENS (GueNS») to the valve of the
swimmeret artery was activated when
Sswimmeret-beat or when they were manually
protracted. LCN5 firing rate was diminished
by stimuli applied to abdominal ganglionic
nerves containing axons from tactile
receptors of the body surface, while it was
activated by stimuli to swimmeret nerves
containing axons of mechanoproprioceptor-
like processes. The stretch receptor-like
processes were observed at the base of
Sswimmerets by means of forward-filling of
swimmeret nerves with CGOrr= incl INalte 6
Excitatory effects of swimmeret protraction
on LCN5 disappeared after axons of the
processes were severed from the periphery.

These results suggest that the LCN

receive inhibitory input from tactile
receptors, and that the’ LCN receive
excitatory input from mechanoproprioceptors
which are activated by movements of

swimmerets and walking legs.

PH 70

EXCITATORY AND INHIBITORY PATHWAYS TO
CARDIO-ACCELERATORY NEURONS IN THE ISOPOD
CRUSTACEAN BATHYNOMUS DOEDERLEINI.
K.Tanaka and K.KuwaSsawa Dept. of Biology,
Tokyo Metropolitan University, Tokyo.

We have identified two pairs of the lst
and 2nd cardio-acceleratory (CAl and CA2)
neurons in, respectively, the 2nd and 3rd
thoracic ganglia (TG2 and TG3).

Excitatory and inhibitory postsynaptic
potentials (EPSPs and IPSPs) were recorded
from both CAl and CA2 neuron somata in
response to stimulation of abdominal
ganglionic nerves and external stimuli to
the body. In single-ganglion (TG2 or TG3)
preparations, stimuli applied to the lst
and 2nd roots of TG2 or TG3 evoked EPSPs
in CA neurons. EPSPS were recorded from
these neurons in response to extention of
thoracic limbs and protraction of swimmerets.

In ganglionic preparations which associated
with the lst thoracic ganglion (TG1), stimuli
applied to the lst to 3rd roots of TG2 or TG3
evoked IPSPs in CA neurons. Tactile stimuli
applied to spines of thoracic limbs, hairs
of swimmerets and the surface of the
tergites evoked IPSPs in the neurons. But
in preparations lacking in TGl, all those
stimuli did not evoke IPSPs in the neurons.

From these results, it is suggested that
EPSP input to CA neurons was transmitted in
TG2 or TG3 itself from mechanoproprio-
ceptors located in the thoracic limbs and
swimmerets, and that IPSP input to CA
neurons was sent from tactile receptors
through TGl.

PH 71

EFFECTS OF PUTATIVE NEUROTRANSMITTERS ON
THE HEART OF THE ISOPOD CRUSTACEAN,
BATHYNOMUS DOEDERLEINI.

T. Yazawa, K. Tanaka and K. Kuwasawa.
Dept., of Biol. , Tokyo Metropol itane Unease

Any neurotransmitters of extrinsic
inhibitory and acceleratory nerves and
intrinsic cardiac ganglionic neurons of
the heart have not yet been identified in
any isopod species. Various substances
were administered to isolated hearts. ACh,
= |FlI and octopamine showed prominent
positive chronotropic and inotropic
effects at 10 nM, although none of them
showed direct depolarizing effects on the
myocardium. Glutamate (1000 nM) and
quisquarate (100 nM) showed positive
inotropic and tonotropic effects, and
depolarizing effects on the myocardium.
Effects of inhibitory and acceleratory
nerves on the heart were blocked by,
respectively, picrotoxin and atropine.
Myocardial force of spontaneous heart beat

was weakened by 2-amino-5-
phosphonovalerate. GABA immunoreactive
axons were found in extrinsic cardio-
regulatory processes. Glyoxylic acid-

induced fluorescent processes were not
found inside the heart. It is likely that
the extrinsic inhibitory cardioregulatory
nerve is GABAergic. We may imply that ACh
and glutamate-like substances are neuro-
transmitters OIE 5 respectively, the
extrinsic accelerator nerves and the
intrinsic cardiac gnglionic neurons.

dal WZ

SEROTONERGIC INNERVATION OF THE HEART AND
ARTERIES IN SOME MOLLUSCAN SPECIES.

K. Kuwasawa, S. Matsumura and M. Kurokawa.
Dept. of Biology, Tokyo Metropolitan

University, Tokyo.

We studied serotonergic mechanisms of the
cardiovascular system in a variety of
phylogenetically selected molluscan species,
Liolophura japonica (Polyplacophora), Aplysia
kurodai, Pleurobranchaea novaezealandiae
(Opisthobranchia), Mercenaria stimpsoni
(Bivalvia) and Octopus vulgaris (Cephalopoda).
In the species neural processes showing
immunoreactivity to rabbit anti-serotonin
antibody were found in whole mount
preparations from the heart and arteries.

We observed effects of serotonin-
perfusion on heart beat by means of
mechanical recording and on arteries by
means of blood-flow recording and/or
mechanical recording. Serotonin (at
concentrations more than 10-19 to 1078 M)
produced excitatory effects on the heart of
the species. Serotonin (more than 10°°M)
produced longitudinal relaxation which
decreased blood-flow rate in the anterior
artery of Liolophura, Aplysia and
Pleurobranchaea, and contraction which
increased the rate in the dorsal aorta of
Octopus. The arteries of Aplysia and
Octopus showed rhythmic contraction by
perfusion with serotonin (more than 10°°M
iO LO-7 MM).

These results may indicate that serotonin
is not only a cardio-regulatory transmitter
but also a vaso-regulatory transmitter in
the species used in this study.

Physiology 1037

BEETS

INHIBITORY INNERVATION OF THE CARDIO-
ARTERIAL VALVE OF HOMARUS AMERICANUS.

T. Kuramoto. Shimoda Marine Research Center
Univ. of Tsukuba, Shimoda, Shizuoka 415.

AS pain Of filvapis in thie candio—stenna 1
arterial valve of the American lobster is
innervated by the segmental nerves of both
thoracic and abdominal ganglia. Electrical
Seamieataon was! app lased | tom ealch fot. thie
peripheral nerves while membrane potential
of the flap muscle was measured with glass
microelectrodes (30 Mohm). The stimulation
elicited hyperpolarizing responses (3-5 mV
im anplitnde ang 1,522 iin Gitieaiei@m)) aim
the muscle with a delay (25 ms). Only one
kind and unitary potential was evoked in
each flap by the single nerves. Therefore,
the valve function may be regulated in the
manner of polyneural and multiterminal
innervation by similar, inhibitory motoneu-
rons in the thoracic and abdominal ganglia.

By repetitive stimulation (1-50 Hz), the
inhibitory junction potential (IJP) exhi-
bited a remarkable facilitation (2-10 hold
increase in amplitude). The facilitation
continued for minute-long periods. This
long-lasting facilitation was distinctive
feature in the nerve-muscle transmission.

Summation of the IJP caused hyperpolari-
zing the muscle membrane from -60 to -90
mY. Simultaneous recordings of the poten-
tial and tension of the flap muscle showed
that the tension follwed each change in the
potential. The IJPs suppressed spontaneous
and evoked contractions of the flaps.

PH 74

PHYSIOLOGICAL CHANGE OF THE HEART BEAT
IN JUVENILE STAGE OF THE ISOPOD CRUSTACEAN
LIGIA EXOTICA.

ins MamaAgwu Slots IbnSto ‘Ou IiOll, Sens, Winiwyo oi
Tsukuba, Tsukuba.

The heart beat of the Ligia juveniles
was examined. The juvenile (manca stage)
just released from the brood chamber of a
parent was about 3 mm in body length and
its heart rate was around 300 per minute.
Each action potential of the heart muscle
followed by a heart beat consists of a slow
potential (50 to 60 mV) and overshooting
(10 to 20 mV) spike potentials superimposed
One aie Maximum hypererpolarization of the
ACELOM DOVEMENAI WAS Ox Aoome 7/0) iO IO) w/e
By application of TTX (10 nM to 1la&M), the
spike potential disappeared and the heart
continued to beat associated with remaining
oscillatory slow potentials. Morphological
data revealed that the cardiac ganglion is
present in the heart and stimulation of the

ganglionic trunk induced excitatory
junctional potentials in the heart muscle,
whereas no spontaneous activity was
recorded from it. A brief pulse applied
intracellularly to the heart muscle changed
the beat interval and reset the beat
rhythm. The heart rate increased by

depolarizing the heart muscle and decreased
by hyperpolalizing. The results indicate
that the heart beat of Ligia juveniles is
myogenic.

After a few weeks, the heart beat of the
juvenile was neurogenic. Thus the heart
beart of Ligia changed physiologically from
myogenic to neurogenic in juvenile stage.

PH 75

REGULATION OF BUCCAL MUSCLE MOVEMENT BY
CENTRAL NEURONS IN ACHATINA FULICA.

M. Yoshida and M. Kobayashi.

Physiol. Lab., Fac. of Integrated Arts and
Sci., Hiroshima University, Hiroshima.

We have identified five pairs of buccal

ganglia neurons controlling buccal muscles
in Achatina. Of these neurons, B4 showed
the most potent effect on the contraction
of the radula retractor. Pharmacological
experiments suggested the excitatory
transmitter of B4 to be an acetylcholine
(ACh). Each neuron of a pair of ventral
cerebral distinct neurons (v-CDNs) in the
cerebral ganglia sent it's processes to
buccal ganglion and buccal nerves and,
when fired, it could enhance both B4- and
ACh-evoked contraction of the retractor.
These effects of v-CDN could be mimicked
by application of serotonin to the muscle.
We also found many neural fibers and vari-
cosities in the retractor as well as cell
bodies of v-CDN showing serotonin-like
immunoreactivities. Thus the enhancing
effects of v-CDN on muscle contraction may
be mediated by serotonin-like substance
which is released from terminals of v-CDN
and acts on the muscle directly. Firing
of v-CDN produced gradually rising and
slowly decaying EPSPs in identified moto-
neurons in the buccal ganglia. v-CDN in
Achatina may be homologous to serotonergic
cerebral giant cells, which have been
known to modulate feeding movement in
several gastropods, since v-CDN shares
several common features with them.

PH 76

MODULATORY EFFECTS OF THE NEUROPEPTIDE APGW-
AMIDE AND ITS ANALOGUES ON MOLLUSCAN MUSCLES.
Y.Muneokal,y.Kurokil, T.Ikedal, Y.Fujisawal,
I,Kubota2, H.Minakata3 and K.Nomoto3. JlFac.
of Integrated Arts and Sci., Hiroshima Univ.,
Hiroshima, 2Suntory Bio-Pharma Tech Center,
Oura-gun, Gunma and 3Suntory Institute for
Bioorganic Research, Mishima-gun, Osaka.

The structure of APGWamide first isolated
from the ganglia of a prosobranch mollusc
Fusinus is closely related to that of the
C-terminal tetrapeptide fragment of the
crustacean hormone RPCH. In some molluscan
muscles, the actions of APGWamide have been
shown to be qualitatively similar to those
of RPCH. In the present experiments, we
examined the effects of APGWamide, RPCH and
related compounds (FAPGWamide, PGWamide, GW-
amide and Wamide) on several molluscan
muscles, such as the ABRM of Mytilus. The
compounds ,except Wamide,showed a relaxation-
inhibiting, a contraction-inhibiting ora
contraction-potentiating effect on them.
In some of the muscles, the actions of the
compounds were suggested to be presynaptic.
In all the muscles examined, the potency
order of the active compounds was GWamide>
APGWamide >FAPGWamide > RPCH >PGWamide. That
is, GWamide was found to be more potent than
APGWamide. In the APGWamide-containing
neurons, the peptide might be processed by a
dipeptidylaminopeptidase,and the product GW-
amide might be released as a neuromodulator.
We also tested fifteen GWamide analogues,
such as Gly-Trp-OH,Gly-tryptamine and D-Ala-
Trp-NH2, on the molluscan muscles, but could
not detect any bioactivity of them.

1038 Physiology

PH 77

A NOVEL BIOACTIVE PEPTIDE ISOLATED FROM THE
PROSOBRANCH MOLLUSC FUSINUS FERRUGINEUS
K.Kurokil, I.Kubota’, T.Ikedal,
-Fujisawa’, H.Minakata and Y.Muneokal.
Fac. Integrated Arts and Sci., Hiroshima
Univ., Hiroshima. “guntory Bio-Pharma Tech
Center, Gunma and Suntory Inst., Bioorgan-
ic Reserch, Osaka.

A novel peptide that potentiates twitch
contraction of the radula retractor muscle
of Fusinus ferrugineus was isolated from
the ganglia of the animal. The primary
structure of the peptide, which was provi-
sionally termed FEP-(Fusinus excitatory peptide 5),
was determined to be as follows:

H-Leu-Ser-Ser-Phe-Val-Arg-Ile-NH

The structure of tke ESR ADOIEIGS frag-
ment. of FEP-, FVRlIlamide, is closely related
to those o the peptides QFYRIamide and
EFLRIamide which have been shown to be
encoded in a gene of Lymnaea with FMRFamide
and FLRFamide. The fragment structure also
seems to be significantly homologous to those of
FMRFamide-related peptides. Therefore, we synthe-
sized FEPs, FVRlamide and FVRFamide and examined
their actions on some molluscan muscles (the
radula retractor muscle of F. ferrugineus, the
ABRM of Mytilus edulis and the crop of Euhadra)
to compare with those of FMRFamide. The obtained
results showed that the actions of the peptides
having Rlamide at their C-terminal parts were
rather opposit to those of the peptides having
RFamide. Although there might be evolutional
relation between FEPs and FMRFamide, their physi-
Ological roles seem to be different.

PH 78

FIVE MEMBERS OF MYTILUS INHIBITORY
PEPTIDES (MIPS) ISOLATED FROM THE ABRM OF
MYTILUS EDULIS
Y.Fujisawa I.Kubota“, T. Ikeda! and Y.
MuneoKa~. lpac. of Integrated Arts and
eciences, Hiroshima Univ., Hiroshima and
Suntory Bio-Pharma Tech Center, Gunma

Two homologous neuropeptides termed
Mytilus inhibitory peptides (MIPs), which
show potent inhibitory effects on the
contractions of the anterior byssus re-
tractor muscle (ABRM), have been isolated
from the pedal ganglia of Mytilus edulis.
In the present study, we isolated these
peptides (MIP, 2) and also three more
analogs (MIP 3 ; 5) from the acetone ex-
tract of the ABRMs by HPLC purification
and bioassays. They were;

MIP: GSPMFVamide MIP»: GAPMFVamide

MIP3: DSPLFVamide MIP: YAPRFVamide

MIP-: ASHIPRFVamide
All ape five peptides showed similar
inhibitory effects on the contractions
caused by electrical stimilation, acetyl-
choline and FMRFamide application ,and had
no effect on the catch tension of the
ABRM. From the experiments with C-termi-
nal fragments, it was shown that the
PXFVamide was important for the inhibitory
effect, although the essential structure
was FVamide. We suppose that all the five
MIPs are involved in the physiological
regulation of the ABRM. The MIP system
may offer a good example for the evolution
of the gene of family peptides with rela-
tion to their receptors.

PH 79

STRUCTURAL DIVERSITY OF MIP (MYTILUS INHIBITORY
PEPTIDE) ANALOGUES ISOLATED FROM SOME PULMO-
NATES.

T.Ikeda!, T.Kiss*, L.Hiripi2, I.Kubota®, Y.Kamatani‘,
H.Minakata* K.Nomoto~*, and Y.Muneoka~. Fac.,
Integrated Arts and Sci., Hiroshima Univ., Hiroshi-
ma, “Balaton Limnological Research Inst., Hungar-
ian Academy of Sci., Tihany, Hungary, 3suntory
Bio-Pharma Tech Center, Gunma and “Suntory Inst.,
Bioorganic Research., Osaka.

It has been reported that MIPs, such as GSPM-
FVa and GAPMFVa, show an inhibitory action not
only on Mytilus muscles but on some pulmonate
muscles and neurons. In the present experiments,
therefore, we atempted to isolate MIP analogues
from acetone extracts of the ganglia of the
pulmonates, Helix pomatia and Achatina fulica. We
obtained twelve species of MIP analogues from
Helix and seven species from Achatina. The
probable structures of them are as follows.

Helix pomatia:
GAPAFVa AAPRFVa GAPMFVa GAPLFVa
GSPYFVa GAPYFVa RAPYFVa SVPIFVa
GVPYFVa GPPMFla AAPFFVa RAPFFVa
Achatina furica:
AAPKFVa GAPKFVa GAPVFVa GAPYFVa
AAPYFVa GPPMFVa GAPFFVa

In addition to these MIP-family peptides, some
other analogues were suggested to be present in
Helix, and several other analogues in Achatina.
That is, each of the animals seems to have more
than ten species of MIP analogues. It is very
interesting that such a wide structural diversity
of family peptides having the same number of
amino acid residues is found. It is required to
clarify the physiological and evolutional meanings
of the structural variety of the family peptides.

PH 80

THREE NOVEL PEPTIDES ISOLATED FROM THE ATRIA OF
THE AFRICAN GIANT SNAIL.

K. Fujimoto!, M. Kobayashi |, I. Kubota?, Y. Kamatani®,
H. Minakata® and K. Nomoto’. 'Physiol. Lalo, , lr:
Integrated Arts & Sci., Hiroshima Univ., Hiroshi-
ma, “Suntory Bio Pharma Tech Center, Gunma and
Suntory Inst. Bioorganic Res., Osaka.

Heart beat of the African giant snail, Achatina
fulica, is known to be regulated by multiple
transmitters or modulators including neuropep-
tides. In the present study, three novel peptides
were isolated from the atria of this animal, which
showed the cardio-excitatory action. One of them
termed ACES-1 was considered to be the same tetra-
peptide as achatin-I, purified from the ganglia of
this snail (Kamatani et al. 1989). Experimental
results suggest that achatin-I enhances the heart
beat in two ways; centrally by stimulating the
heart excitatory neurons, and peripherally by
increasing the beat amplitude directly. The
second one termed ACES-2 was a pentapeptide close-
ly related to bag cell peptides of Aplysia. The
third one was a novel undecapeptide, the structure
of which was determined to be H-Ser-Gly-Gln-Ser-
Trp-Arg-Pro-GIn-Gly-Arg-Phe-NH,. This peptide was
termed Achatina cardio-excitatory peptide-1 (ACEP-
I). ACEP-1 enhanced the contraction of penis
retractor and buccal muscles and also excited the
identidfied neurons controlling buccal muscle
movement of the snail. From these diverse activi-
ties, ACEP-1 is assumed to be an atrial hormone-
like substance.

Physiology 1039

PH 81

EFFECTS OF ADP AND VANADATE ON THE RATE OF
RELAXATION IN CHEMICALLY SKINNED MOLLUSCAN
SMOOTH (CATCH) MUSCLE.

N. Ishiit, A.W.M. Simpson” and C.C.
Ashley”~. *Dept. of Sports Sci., Fac. of
Arts and Sci.,Univ. of Tokyo, Tokyo,
2Univ. Lab. of Physiol., Oxford, U.K.

To elucidate the mechanisms underlying
the ‘catch" contraction, we studied the
effects of various substances that affect
the actomyosin ATPase on the mechanical
properties of saponin-treated fibers from
a molluscan catch muscle, the anterior
byssus retractor muscle of Mytilus edulis.
Isometric contraction and relaxation were
induced by changing rapidly [Ca**] with
either the ‘calcium jump" technique with
controlled Ca**-buffer concentration or
the laser-flash phtolysis of a caged
chelator Diazo-2. Upon a rapid drop in
[Ca7*], maximally activated fibers showed
a slow relaxation during which no tension
redevelopment was observed after a quick
release, suggesting that the fibers were
in the catch state. An additional, rapid
relaxation was seen at the onset of the
laser-flash photolysis. ADP (10mM) slowed,
whereas P; (10mM) accelerated these
processes. cAMP (10 *M) slowed the ear-
liest process, although it accelerated the
following slower processes. Vanadate (1mM)
interrupted the catch and brought about a
rapid relaxation. The results suggest that
the catch is caused by the actomyosin
crossbridges, particularly those in the
state of actomyosin-ADP intermediate.

PH 82

MAXIMUM VELOCITY OF SHORTENING IN INTACT,
SINGLE CRAYFISH SLOW MUSCLE FIBERS.
T.Tameyasu, Dept.Physiol.St.Marianna Univ.
Sch.Med., Kawasaki.

The time course of shortening at zero
external load was studied by the slack test
method during tetanic contractions in more
than 10 isolated, single fibers from the
crayfish slow muscle (SL 7-9 um). The fiber
predominantly exhibited a biphasic shorten-
ing expressed by a combination of two
straight lines. The velocity in the initial
phase of the biphasic shortening (V1) was
14.3 + 3.3 (SD) um/half sarcomere (h.s.)/s
at 20°C, while that in the second phase
(V2) was 7.3 + 1.5. Lowering temperature
decreased both V1 and V2 with Qj9=1.4 for
Vl and 2.0 for V2. Lowering the external Ca
decreased both Vl and V2 with a substantial
decrease in the tetanic tension. Pre-
stretching the fiber up to 1.5 times the
resting length increased Vl, but not V2.
The distance shortened during the initial
high-velocity phase was 0.14 um/h.s. in the
range of 0.08-0.21 um/h.s. The results sug-
gest that the initial high-velocity phase
was brought about by an acceleration of
sliding between thick and thin filaments by
a compressive force which might develop
during a tetanic contraction. The slow
velocity phase was thought to reflect the
cross-bridge turnover rate with little exo-
geneous stress on the myofilaments.

PH 83

SHORTENING OF RABBIT PSOAS MUSCLE FIBERS
INDUCED BY PHOTOLYSIS OF CAGED-ATP.

Q.Abe*, T. Yamada“, K.Takahashi- and H.Sugi2
Zool. Inst., Fac. of Sci., Univ. of Tokyo,

Tokyo and “Dept. of Physiol., Sch. of Med.,

Teikyo Univ., Tokyo

To study the relation between the distance
of myofilament sliding and the amount of ATP
utilized, single glycerinated fibers from
rabbit psoas muscle were first put into rigor
state, and then activated to shorten under a
small load (less than 1% of Poy) by photolysis
of caged-ATP, the fiber shortening being
recorded with a high-speed video system (200
frames/s). To prevent diffusion of ATP out of
the fiber, the fiber was immersed in silicone
oil, so that the released ATP was only uti-
lized for myofilament sliding.

In response to the caged-ATP photolysis,
the fiber shortening started after a delay of
20-100 ms, which may correspond with ATP-
induced breaking of rigor linkages, and even-
tually stopped when the released ATP was used
up for myofilament sliding. The fiber short-
ening took place uniformly along the entire
fiber length. The distance of fiber shorten-
ing decreased with decreasing amount of re-
leased ATP. The minimum shortening of the
fiber (about 2%) was observed with 50-70 uM
ATP. Assuming the myosin head concentration
within the fiber to be 150 uM, the distance of
actin-myosin sliding induced by hydrolysis of
one ATP molecule is estimated to be 20-30 nm,
if some cooperativity between two myosin heads
is taken into consideration.

PH 84

MECHANISM OF FORCE ENHANCEMENT AFTER
STRETCH IN VERTEBRATE SKELETAL, MUSCLE.
Teizo Fsuchiya » Paul Edman*, Haruo
Sugi-, ~Dep. ERystol. | Sch. Med., Teikyo
Univ. Tokyo, “Dep. Pharm. Univ. Lund,
Sweden.

It has been known that stretch of
active muscle at a moderate velocity
causes force to increase above the
isometric level and this phenomenon is
called force enhancement after stretch.
To know the mechanism of this phenomenon,
a living single skeletal muscle, tibialis
anterior, of the frog was stretched
during tetanus in the low temperature (2-
3° C) and effects of stretch velocity,
amplitude of stretch, sarcomere length,
hypertonicity and temperature were
investigated. The residual force above
isometric force 4 or 5s after stretch was
constant irrespective of stretch velocity
and change of velocity during stretch if
the stretch amplitude was constant. The
residual force was higher at longer
sarcomere length in the range between
2.0-3.0um. The ratio of residual force
to isometric force was very much constant
at low and high temperature.
Hypertonicity by 98mM sucrose had no
effects on this ratio of the residual
force. The possibility that the passive
elastic element parallel to cross-bridges
was involved in this phenomenon was
discussed.

1040

PH 85

EFFECT OF SCN AND K IONS ON THE LENGTH-TENSION
RELATION AND THE RESTING AND ACTION POTENTIALS
IN RAT PAPILLARY MUSCLE.

D.Kondo, T.Kobayashi, and H.Sugi Dept. of Physiol.,
Sch. of Med., Teikyo Univ., Tokyo.

It has been shown that, in rat papillary muscle, the
contractile force is potentiated by more than 25% by
25%-SCN and 5.7mM-K solutions, and the potentiation
is associated with membrane depolarization in both
cases (Iwamoto et al. 1989). We examined the effect of
SCN and K ions on the contractile force and on the
resting and action potentials over the length range
from 0.7 to 1.0Lmax where the Starling's law of the
heart is obvious.

The resting and action potentials did not change
significantly when the muscle length was varied
from 0.7 to 1.0Lmax in the standard solution. In 25%-
SCN and 5.7mM-K solutions, the resting potential de-
creased by 5-20mV with the same amount of decrease
in the action potential amplitude. These changes
‘remained, however, unchanged by the change in
muscle length from 0.7 to 1.0Lmax. On the other hand,
the contractile force in response to a single electrical
stimulus decreased from the maximum value to nearly
zero with decreasing muscle length from 1.0 to
0.7Lmax in both the standard and the SCN and K
solutions, so that the degree of potentiation of
contractile force remained the same irrespective of
muscle length. These results indicate that (1) the
Starling's law of the heart may not be mediated by the
length-dependent depolarization, and (2) the SCN-
and K-induced force potentiation is not associated
with the shift of the active force-length relation.

PH 86

ACTIN-BOUND NUCLEOTIDE AND SLIDING MOVEMENT.
N.Oishi!, A.Muraoka2, and H.Sugi2 !Radioisotope
Research Center and 2Dept. of Physiol., Sch. of Med.,
Teikyo Univ., Tokyo.

It has been shown that the substitution of the
bound nucleotide may cause some conformational
change in actin. In an attempt to study the possible
role of actin-bound nucleotide in the _ sliding
movement between actin and myosin, we prepared
several kinds of nucleotide-substituted F-actin from
rabbit muscle. For effective displacement of bound
nucleotides, we prepared G-actin containing ADP first
by the quick dispersion of actin subunit from F-actin
under sonic vibration, and then polymerization was
induced by salt and phalloidin in the presence of the
ATP analogs : adenylyl imidodiphosphate(AMPPNP),
8-bromoadenosine triphosphate(BrATP), or inosine
triphosphate(ITP). HPLC analysis of the nucleotide
content in the F-actin thus prepared showed that
over 93% of the nucleotides were replaced for
AMPPNP, over 99% for BrADP or IDP.

In order to measure the sliding mevement of the
nucleotide-substituted F-actins on heavy mero-
myosin(HMM) or on myosin subfragment-1(S1),
thodamine-phalloidin was introduced as _ fluorescent
probe in place of phalloidin, and assayed by an in
vitro system on a _video-enhanced fluorescent
microscopy. No appreciable difference of the sliding
velocities were observed among the _ nucleotide-
substituted F-actins and the normal F-actin. They
slided continuously on HMM or S1 at the velocitiy of
about 10 pm/sec at 30°C. The results suggest that the
actin-bound nucleotide may not play a significant
role in the sliding movement between actin and
myosin.

Physiology

PH 87

FORCE-VELOCITY RELATION OF THE ATP-DEPENDENT
SLIDING BETWEEN ACTIN AND MYOSIN AS
DETERMINED BY AN JN VITRO FORCE MOVEMENT
ASSAY SYSTEM.

H.Sugi, S.Chaen and K.Oiwa Dept. of Physiol., Sch. of
Med., Teikyo Univ., Tokyo.

To eliminate the gap between muscle physiology and
muscle biochemistry, we constructed .a new in vitro
force-movement assay system, in which myosin-coated
polystyrene beads (diameter, 2.8m) were made to
slide on the actin filament arrays (actin cables) in giant
algal cells and subjected to constant centrifugal forces
by means of a centrifuge microscope. The bead move-
ment was recorded with a video system (30frames/s)
combined with a stroboscopic light source.

The bead continued moving on actin cables with a
constant velocity under a centrifugal force directed op-
posite to the bead movement. The steady-state force-
velocity (P-V) relation thus obtained was hyperbolic in
shape for small loads, but deviated from hyperbola for
large loads. Similar results were obtained for the beads
with Po values ranging from 2 to 39 pN, indicating that
the P-V relation reflects the properties of individual
myosin heads rather than the change in number of
myosin heads involved. Unexpectedly if the bead was
subjected to centrifugal forces in the same direction as
the bead movement, the velocity of bead movement
decreased with increasing centrifugal forces, though the
bead eventually detached from actin cables. This work
was performed in collaboration with Drs Kamitsubo
(Hitotsubashi Univ.) and Shimmen (Himeji Univ. of
Tech.).

PH 88

MECHANISM OF PROPAGATION OF ACTION POTENTIAL
AND THE ELECTRICAL CONSTANTS IN. THE
SWIMBLADDER MUSCLE OF A TELEOST SEBASTICUS
MARMORATUS.

T.Kobayashi, D.Kondo, A.Goto and H.Sugi Dept. of
Physiol., Sch. of Med., Teikyo Univ., Tokyo.

The swimbladder muscle of a teleost fish Sebasticus
marmoratus exhibits extremely rapid twitches to pro-
duce sounds for communication. To study the mode of
nervous control in this muscle, we examined its elec-
trical properties as well as its mode of motor nerve
innervation with histochemical techniques.

The action potential showed little of no overshoot
with a tendency to attenuate with distance when ini-
tiated by intracellular stimulation at one point on the
muscle fiber. The above decremental propagation of
action potential was consistent with the very large
membrane capacitance (23-32uF/cm2) and the very
low membrane _ resistance (180-250Q+cm2). On the
other hand, if the motor nerve innervating the
muscle was stimulated, the action potential was set up
at the area of nerve insertion and propagated along
the fiber length with a velocity of 7m/s. In accor-
dance with the decrementless propagation in
response to motor nerve stimulation, histochemical
studies showed the presence of many motor nerve
terminals distributed along the entire fiber length.
This, together with the very high velocity of action
potentials propagation, indicates that the action
potential propagates along the motor nerve.
Histochemical studies also showed that this muscle
consists of only one type of muscle fibers.

Physiology 1041

PH 89

Ca-SPIKE AND DEPOLARIZING AFTERPOTENTIAL OF
PYLORIC MUSCLES .OF THE STOMACH IN SQUILLA.

C.Chiba and K.Tazaki. Biol.Lab., Nara Univ.
Educ., Nara.

The constrictor muscles of the pyloric
stomach of Squilla display an endogenous
oscillator property. They generate spikes
by synaptic inputs of the motorneurons lo-
cated in the stomatogastric ganglion. The
spike of these muscles is followed by a
Sustained depolarizing afterpotential which
leads to repetitive spike discharges. They
can generate oscillatory potentials with
repetitive spikes after ablation of the
stomatogastric ganglion. The spike remains
unaffected by bath application of tetrodo-
toxin. It can be aqenerated in salines in
which Ba or Sr is substituted for Ca, but
blocked by Mn or Coa, suaqgesting that the
Spike is Ca-depencent. The sustained depo-
larizing afterpotential is observable in
the Co-saline, but not in the Na-free sa-
line. The experimental entrainment of
rhythmic activity of the muscles shows that
the duration of the sustained denolarizing
afterpotential depends on the frequency of
stimulation: it decreases with increasing
frequency. We suagest that the sustained
depolarizing afterpotential may be related
to voltage-dependent and time-dependent Na
current. The pyloric constrictor neuromus-
cular system appears to consist of a motor-
neuron oscillator and a muscle oscillator.

PH 90

INTRACELLULAR CALCIUM TRANSLOCATION DURING THE
CONTRACTION-RELAXATION CYCLE IN THE SWIMBLADDER
MUSCLE OF THE SCORPIONFISH.

S. Suzukil, N. Hino?, Y. Oshimi! and H. Sugi!. Dept. of
Physiol., Sch. of Med., Teikyo Univ., Tokyo and 2Dept. of
Physiol., Sch. of Med., Juntendo Univ., Tokyo.

To give information about the intracellular Ca trans-
location during the contraction-relaxation cycle in the
skeletal muscle, the change of Ca distribution along the
sarcomere at various physiological states in the
swimbladder muscle fibers of the scorpionfish was
examined by electron probe X-ray microanalysis of

cryosections. In the resting fibers, the Ca concen-
tration was highest at the boundary of the A- and I-
bands where the terminal cisternae were located. In

the contracting fibers, on the other hand, the Ca
concentration decreased significantly at the boundary
of A- and I-bands, and increased significantly at the A-
band region. This result indicates that Ca stored in the
terminal cisternae in the resting fibers is released into
the myoplasm to cause contraction. In the fibers frozen
di OMGesece safter » the sonsct) of srelaxation,, the €a
concentration increased at the A-band region, though
the isometric tension already decreased to a half of the
maximum tension. In the completely relaxed fibers, the
Ca concentration decreased gradually at the A-band
region, and increased again at the boundary of the A-
and I-bands. Since the A-band region contains also the
fenestrated collars and longitudinal tubules of the SR in
the cryosections of about 200 nm thick, the change of Ca
distribution during and after the relaxation indsicates
that Ca released from the terminal cisternae can be
taken up mainly at the fenestrated collar and
longitudinal tubules, and then slowly return to the
terminal cisternae.

PH 91

STIFFNESS OF THE ARM OF THE STARFISH
LINCKIA LAEVIGATA.

T. Motokawa.

Dept. Ot. BIOL COMO SG. .

Univ. of the Ryukyus, Okinawa.

Arm stiffness of the starfish Linckia
laevigata was measured by bending test,
regarding the arm as a cantilever. The
arm was well simulated with a hollow
cylinder whose cross sectional size was
constant in the proximal region, tapering
uniformly towards the tip in the distal
region. When the proximal end (arm height
hz) is fixed and weight P is applied to
the distal end (height h,), the stiffness
(E) of the arm is given by the following
equation we derived from beam theory:

PL?
E= (14H a 3)
3wW1

where L is arm length (length of non-
tapered part being 1,), W is deflexion at
distal end, I is the second moment of area
which we calculated as 0.05h2*, A is
Cyr eand Herts) (h=—ha) hye) obidtehnes's: of
Linekia arm was 8.04 MPa (geometric mean,
log SE=0.035, N=70). It increased to 20.5
MPa when the arm was mechanically
stimulated.

PET 92

EFFECTS OF CALCIUM ON THE MECHANICAL
PROPERTIES OF THE CAPSULE WALL OF ISOLATED
NEMATOCYSTS.

M. Hidaka. Dept. of Biology, Univ. of the
Ryukyus, Okinawa.

lie is not clear whether Ca-free
solutions induce nematocyst discharge by
increasing the osmolarity of intracapsular
fiuid or “by softening the capsule wail,
especially ithe: SUsitopper is. reealont. The
effects of Cae on the mechanical
properties of the capsule wall of isolated
nematocysts were examined. Nematocysts
isolated from acontia of the sea anemone
Calliactis polypus were immersed in

solutions COmudiiminge 0, 205 30, and
40(W/V)% polyethylene glycol (MW 1450) and
various concentrations of Ca The

shorter and longer circumferences of the
capsule were plotted against the osmotic
pressure of the bathing medium to calculate
the Young's modulus of the capsule wall.
The moduli of the capsule wall in the
longitudinal and circumferential directions
WweRee Loh Bimcl lod Gk, wesipxeCwielays The

modulus Ol thie weicaps ule swarl! was not
Significantly affected by Gay These
results suggest that Ca affects the
osmolarity of the intracapsular ie JE pistG!
rather than the mechanical properties of
the capsule wall. X-ray microanalysis

showed that calcium was evenly distributed
in the capsule of undischarged nematocysts
and not localized at the capsule wall or
the 'stopper!' region.

1042

PH 93

PROPERTIES OF A HIGH-MOLECULAR WEIGHT
CALCIUM BINDING PROTEIN ISOLATED FROM
PLASMA MEMBRANE-ENRICHED FRACTION OF A
MOLLUSCAN SMOOTH MUSCLE

T.Yamanobe*, T.Mimura’ and H.Sugi’.

*Central Lab. Analyt. Biochem. and “Dept.
Physiol., Sch. Med., Teikyo Univ., Tokyo.

In the anterior byssal retractor muscle
(ABRM) of Mytilus edulis, Ca ions are
accumulated at the inner surface of the
plasma membrane, and are released into the
myoplasm to cause contraction (Atsumi and
Sugi, 1976). To study the mechanism of Ca
accumulation and release at the plasma
membrane, we have isolated a 450K Ca-
binding protein characteristic of the
plasma membrane-enriched fraction of the
ABRM (Yamanobe, Mimura and Sugi,1985). Ca
binding assay shows that,in the presence of
8mM Mg, the 450K protein binds 10 mole Ca/
mole with a binding constant of 6.0x10°M .
Env thespresence vot Al PGhe .ADE ormscGlPe athe
Ca binding capacity of the 450K protein was
decreaced in the pCa range of 5~7. In
contrast, cAMP increased the binding
capacity at the pCa range of 6~7. With the
phenol-sulfuric acid and the peroxidase-
coupled lectins methods, the 450K protein
was shown to contain sugars( about 9% ) and
the N-linked complex carbohydrate chains.
These results are consistent with the view
that the 450K protein may be involved in
the regulation of mechanical activity of
the ABRM in physiological pCa range.

PH 94

PRIMARY STRUCTURE OF CaVPT, THE TARGET OF
CA2+-VECTOR PROTEIN OF AMPHIOXUS.

T. Takagi & J. A. Cox* Biological Institute,
Faculty of Science, Tohoku University, Sendai,

*Department of Biochemistry, University of

Geneva, Switzerland

CaVPT, a target protein of Ca2*-vector
protein (CaVP) from amphioxus muscle, was
purified from its complex with CaVP after
dissociation by 6 M urea. The amino acid
sequence of CaVPT has been determined. It is
composed of 243 residues. From the sequence
three domains can be infirred: A collagen-like
N-terminal segment that resembles the N-
terminal segment of skeletal muscle myosin light
chain kinase (MLCK); next to it is located a-
helical segment which likely binds the CaVP.
This segment is followed by two immuno-
globulin (Ig)-folds, typically belong to the C2
subclass and _ particularly resembles those
present in the neural cell surface adhesion

molecules. Recently the presence of Ig folds of
this type has been reported in some intracellular

muscular proteins, namely in smooth muscle
MLCK, C protein and titin, as well as in the
nematode twitchin. We suggest that CaVPT acts
on the structure of the thick filament in muscle
or regulates via other Ig fold-containing protein.

Physiology

PH 95

VISUAL DISCRIMINABILITIES IN THE JUVENILE GOLDFISH.
K. Ohnishi. Dept. of Physiol., Nara Med. Univ.,
Kashihara

To examine the development of visual discrimin-
abilities in the goldfish (Carassius auratus) ,
brightness, pattern and color discriminability were
compared between the juvenile (4-5 cm) and the adult
fish (9-11 cm) by using an instrumental training
technique (Y maze). In brightness discrimination
training, the juvenile group (n=12) and the adult
group (n=6) showed very similar trained performances
(not significantly different). In pattern discrimi-
nation training, the adult group (n=9) acquired the
learned performances but the juvenile group
(4 learners, 3 non-learners) acquired the learned
performances very slowly (significantly different).
The juvenile group trained with green vs. blue (0
learner, 11 non-learners) and one trained with red
vs. blue (5 learners, 10 non-learners) did not show
the learned performances, but the performances of
the juvenile learners were not significantly differ-
ent from those of the adult group (n=6) trained with
red vs. blue. The juvenile group trained with red
vs. green (9 learners, 3 non-learners) showed the
learned performances, which were very similar to
those of the adult group (n=7). There were no dif-
ferences between the performances of the juvenile
(n=9) and the adult group (n=10) in an active shock-
avoidance training task. It is considered that the
visual discriminabilities of the juvenile fish are
immature specifically in pattern and color (green
vs. blue, red vs. blue) discrimination but not in
brightness discrimination and general learning
ability in the juvenile fish may be not inferior to
that of the adult fish.

PH 96

DII LABELLING OF NASOPALATINE NERVE WHICH
CONTROLS MOUSE VOMERONASAL PUMPING ACTION
T. Hatanaka, Dept.coif) Biol sy] pacer

duc., Univ. of Chiba, Chiba.

An ingress of odorants into the
vomeronasal organ and egress from it are
driven by a peristasis like movement of the
cavernous vascular tissue which runs along
the organ. To know the motor control
mechanism of this vasomotor movement, as a
first step, an innervation pattern of naso-
palatine nerve was studied by fluorescent
Dil labelling.

A crystal of Dil embedded in a fixed
tissue made a dye marking on the nerve,
which could be observed by light micro-
scopy, but branching details had to be
observed by fluorescence microscopy. When
embedding was done in the part of the
vomeronasal organ, several vomeronasal
nerve branches which arose the portion of
embedding to the olfactory bulb, and the
nasopalatine nerve which run distally and
proximally from that portion along the
vessel at dorsolateral side of the organ
were labelled. These anterograde and
retrograde labellings of the nasopalatine
nerve demonstrated several branching to the
vessel wall. An embedding near the naso-
palatine nerve out of the organ labelled
the nerve anterogradely into the organ.
thie > efficiency Jol »DiieWabeliiineeewals
reconfirmed, and this nasopalatine nerve
innervation was assumed to controle the
vasomotor movement.

PH 97

MEMBRANE POTENTIAL RESPONSE TO A LOCAL-
IZED THERMAL STIMULUS IN RELATION TO
CULTURE TEMPERATURE IN PARAMECIUM

T. Tominaga and Y. Naitoh, Inst. Biol.
Sci., Univ. Tsukuba, Tsukuba.

Paramecia accumulate in a region having
temperature identical with the culture
temperature (Tc). The accumulation must
be accomplished, if paramecia exhibit
avoiding response when they swim away
from the culture temperature region. To
understand membrane potential-mediated
mechanism for the thermoaccumulation, we
examined a potential response to a sud-
den raise in temperature. A paramecium
cultured at 25°C showed a depolarizing
response to the thermal stimulus at AZINE
(experimental temperature; Te), while it
showed a small ambiguous response at
15°C. To clarify the mechanism for the
modification of the response by Te, we
examined a potential response to a ther-
mal stimulus applied to a localized area
of a paramecium. When Te was Tc, a
stimulus to any one end of the cell
produced a depolarization. When Te was
lower than Tc, the stimulus to the an-
terior end produced a depolarization,
while that to the posterior end a hyper-
polarization. The Te-dependent pos-
terior hyperpolarization was affected by
Tc, while not the anterior depolariza-
tion. Thus it is concluded that the an-
terior depolarization is modulated by
the posterior hyperpolarization so that
Pparamecia accumulate at Tc.

PH 98

PROTEOLITIC TREATMENT OF CILIA ON THE
CIIATED SHEETS OF TRITON-GLYCEROL-
EXTRACTED PARAMECIUM CAUDATUM.
K. Okamoto and M. Noguchi. Dept. of
BHO mac. Ot SiCAn, eLOyama Unive, LOvamal.
We examined the direction that the
cilia pointed when observed from above for
understanding the control mechanism of
the beating direction of cilia by Ca2+ and
cyclic nucleotides. The ciliated sheets
of cell model which stuck to the glass
surface were perfused successively with
reactivation solution. We have reported
that cAMP competed with Ca2+ ion in
determining the orientation of cilia, and
CGMP tended to change the ciliary
orientation toward 3 o'clock. When a
ciliated sheet were perfused with the
solution containing elastase or
V8 protease in the presence either Ca2t,
CAMP or cGMP, ciliary orientation was not
affected by these proteases. When
perfused with reactivation solution
containing trypsin, cilia gradually
changed their orientation, and finally
pointed toward the direction about 12
o'clock irrespective of the concentration
of Ca2+, cAMP or cGMP. Althogh ciliary
sensitivity to cyclic nucleotides depended
on their generating part of the cell
surface, these differential sensitivity
disappeard by trypsin. This suggests
that the common component for the
cyclic nucleotides sensitive mechanisms
lose its function during trypsin
digestion.

Physiology 1043

PH 99

ROLES OF CALCIUM ION ON THE CILIARY
RESPONSE {9 FOL Be CED IN CILIATES
Y. Mogami and H. Machemer Dept., Biol.
Ochanomizu Univ. Tokyo and “Dept., Biol.,
Ruhr-Univ., Bochum, FRG.

Voltage-dependent motor responses of
transverse cirri of Stylonychia were
investigated under voltage-clamp. The
cirri which were immotile at the resting
potential performed posterior inclination
responding to hyperpolarization and
anterior beating to depolarization. With
increase in the amplitude of depolariza-
fiion., anterior beaving occurred) awicth
longer latency and cirri became inactive
near Ca equilibrium potential (ca. 120 mV
depending on the Pere and ae

concentrations of and
respectively). Devolarization ae ae
equilibrium poten, ial which would cause

ENG @iweirili Ow Ca bs the artificially
reversed electro-chemical gradient,
induced the posterior inclination almost
identical to that induced by
hyperpolarizing voltage steps. Latency of
the response decreased with increasing the
positivity of the potential, indicating a
mirror image relation of the latency
change at the physiological potential
shift. The extremely large depolarization
could also induce the hyperpolarization
specific motor response in marginal cirri.

These evidences suggest that the decrease
in intraciliary Ca concentration
triggers the hyperpolarization-induced
motor responses in ciliates.

PH 100

HIP IPA Cw Ss OTF CaZzt ON Wisi “DIC RUFC wWw Ic OW" Oy
ROTATIONAL MOVEMENT OF DEMEMBRANATED SEA
URCHIN SPERMATOZOA.

S. Ishijima and Y. Hamaguchi. Biological
laboratory, Tokyo Institute of Technology,
O-okayama, Meguro-ku, Tokyo.

Sea urchin spermatozoa generally rotate
about their longitudinal axis as they swim
freely in a medium. Our measurements of
the direction of the rotation in sea
urchin spermatozoa showed that most of sea
urchin spermatozoa rotated in clockwise
when viewed from its anterior end and the
remainder underwent in anticlockwise.
Moreover, some spermatozoa occasionally
changed their rotational direction. These
results suggest that the mechanism
regulating the direction of rotational
movement of spermatozoa is not rigid. To
find the factor controlling the direction
of the rotational movement, we examined
rme GEereces Of CAm Om cme CliseeSici@m! Cre
rotational movement of demembranated sea
urchin spermatozoa.

When concentration of Catt in the
Geactivation solution decreased,
proportion of spermatozoa rotating in
clockwise decreased while proportion of
spermatozoa in anticlockwise increased.
Since the rotational movement of
JoO@QicmMasconuOe 18 CUS ieO© wits widisSor
dimensional displacement of beating
flagella, these results suggest that the
handedness of bending waves of sea urchin
sperm flagella changes by Ca

1044

PH 101

FLAGELLAR WAVEFORM OF SEA-URCHIN
SPERM FOLLOWING AN ABRUPT CHANGE IN
BEAT FREQUENCY.

Dan Eshel!, Chikako Shingyojiz, Kenjiro
Yoshimura2, Ian R. Gibbons! and Keiichi
Takahashi2. 1Pacific Biomedical Research Center,
Univ. of Hawaii, Honolulu and 2Zool. Inst., Fac. of
Sci., Univ. of Tokyo, Tokyo.

The flagellum of a sea urchin sperm whose head
is held with a suction pipette and vibrated laterally
beats in synchrony with the imposed vibration over a
frequency range of 30 to 80 Hz (Nature, 325: 351-352,
1987). An abrupt shift in frequency of the imposed
vibration is immediately followed by a corresponding
change in the beat frequency. We analyzed the
transient flagellar waveform of live Hemicentrotus
pulcherrimus sperm following abrupt changes in
vibration frequency (from 38 to 65 Hz and from 65 to
38 Hz) by recording the flagellar movement at 200
frames/sec with a high-speed video system. We
found that the rate at which new bends are initiated
at the base of the flagellum changes immediately on
changing the vibration frequency so that the beat
frequency matches the vibration frequency.
However, the bend angle and the bend length
change only gradually as the new bend waves are
propagated. It took more than 4 beat cycles for the
bend angle to attain the new steady state. These
results suggest that the mechanism regulating the
flagellar beat frequency is independent of that
controlling the bend angle and the wavelength.

PH 102

EFFECT OF EXTERNAL FORCE ON THE DEVELOP-
MENT OF BENDS IN SEA URCHIN SPERM
FLAGELLA.

Dan Eshel!, Kenjiro Yoshimura2, Chikako Shingyoji?,
Ian R. Gibbons! and Keiichi Takahashi2. 1Pacific
Biomed. Res. Cent., Univ. of Hawaii, Honolulu, Zool.
Inst., Fac. of Sci., Univ. of Tokyo, Tokyo.

The effect of external force on the development of
flagellar bends was studied by using a piezoelectrically
driven micropipette that held the head of a sea urchin
sperm by suction. When the pipette was vibrated
laterally with a constant sinusoidal waveform the
flagellum beat in synchrony with, and in the same
plane as that of the movement of the pipette. A lateral
shift of the pipette at a constant velocity of up to 1.2
um/msec which interrupted the vibration for 1.5 beat
cycles delayed the initiation of a bend that was due to
occur 0.5 beat cycle after the onset of the shift. The
delay was longer when the shift was in the direction of
the reverse bend than when it was in the direction of
the principal bend. When the shift was in the direction
of the reverse bend the maximal angle of the reverse
bend initiated at the onset of the shift was smaller than
that of the bend initiated during constant vibration.
This effect was not observed in the principal bend
initiated at the onset of a shift in the direction of the
principal bend. The time courses of the bend angles
indicate that the development of the reverse bend angle,
but not that of the principal bend angle, is impeded
after 0.5 beat cycle from the beginning of the shift,
suggesting that the bend angle does not fully develop
without the initiation of the next bend.

Physiology

PH 103

A NEW FLAGELLAR MOTILITY MUTANT, OC-10, IN
CHLAMYDOMONAS.

S. Nakamural, E. Kawanishit, Sc Nakamura?,
S. Watanabe2 and M. K. Kojimal. 1Dept. of
Biol., Fac.. of Scils) 2Dept of Bioline mace
of Edu., Toyama Univ., Toyama, 3Gen. Edu.
Div., Ryukyu Univ., Okinawa.

A new paralyzed mutant, OC-10, was iso-
lated in Chlamydomonas reinhardtii. Wild-
type cells usually swim forward with a mean
velocity of abou 120 um/sec. On the other
hand, OC-10 cells can not swim and show
very little flagellar movement. However,
when OC-10 flagella were demembranated and
reactivated with ATP or ADP, they recovered
vigorous motility. The beating patterns
of the reactivated axonemes were somewhat
different from those of the wild-type.

The rate of flagellar regeneration of OC-10
was 2-3 times slower than that of the wild-
type. Although electron microscope exami-
nation showed no abnormality in OC-10 fla-
gella, SDS PAGE revealed a change in the
motility of 92 kDa protein in OC-10 axoneme.
Temporary dikaryons with quadriflagella
were constructed between OC-10 and the wild-
type cells, however OC-10 flagella could
not regain motility. Genetic analysis
showed that OC-10 phenotype resulted from

a single gene mutation. From these results,
we suppose that OC-10 has defects in the
bridges which link the outer doublet micro-
tubules to flagellar membrane, and the
bridges are more rigid than those of the
wild-type. Therefore, OC-10 flagella can
not move.

PH 104

MOTILITY OF THE BARNACLE SPERM

Ye Nakata!, ales Nakazawa! and M. Okuno?.

Dept. of Biol., Fac. of Sci., Univ. of Toho

Dept. of Biol., Coll. of Art. and Sci., Univ. of
Tokyo

Sperm of the barnacle, B. Tetraclita, appears the
string shape. Nucleus runs parallel to the axoneme
for 80 percent of the length. Electron microgra-
phy showed that the cross section of the nucleus
was a half moon. The central pair of the axoneme
localized almost parallel to the nuclear bowstring.
In artificial sea water (0.5M NaCl, 10mM KCl, 10mM
CaClo, 55mM MgsSO,, 10 mM HEPES-NaHCO, (pH 8.0)
), sperm swam straight forwardly with sinusoidal
bending waves. The swimming velocity, amplitude
of bend, wave length and beet frequency were
177um/s, 3.4um, 24.6um and 3.0Hz, respectively.

When dry sperm was suspended in 50 percent or
higher ASW, more than 60 percent sperm repre-
sented the normal motility. However, only 10
percent sperm represented normal motility in 40
percent or lower ASW. The pH optimum in motili-
ty appeared at pH 8 (90 percent motility). When
the concentration of Ca“* in ASW was below 50uM,
only 20 percent sperm represented normal motility.
The percent motility increased with Ca concen-
tration and attained 80-90 percent at 3mM CaClo.
Verapamil inhibited sperm motility even in the
presence of high concentration of CaClo. The
motility was completely lost at 800uM verapamil.

These results suggested that calcium was re-
quired on the initiation of the sperm motility. We
are planning to investigate the influx of calcium
upon the initiation of sperm motility and the effect
of other divalent cations.

Physiology 1045

PH 105

INVOLVEMENT OF ZINC IN THE REGULATION OF
SPERM MOTILITY IN STARF{SH AND SEA URCHIN.
.-Morisawa , N.Funayama and H.Michibata .
Misaki Marine Bigl. St.,Fac. Sci., Univ.
of Tokyo, Miura, Biol. Inst., Fac. Sci.,
Toyama Univ., Toyama.
Several previous studies have provided evi-
dences for possible involvement of heavy

metals in sperm motility of starfish (Kino-
Shita,. 1956; Mohri et al., 1990) and sea
urchin ‘(Utida & Nanao, 1956; Clapper et al.

,1985). The present study presents evi-
dences which confirm the above studies by
the accurate measurement of sperm motility
using the automated semen analyzer, Cell-
Soft (CRYO Resources, Ltd.). Sperm motility
of the starfishes, Asterina pectinifera and
Asterias amurensis, was activated by metal
chelators, histidine, EDTA and EGTA. In
eontrast, chelators including zinc specific
one EDDA prevented sperm motility of the
sea urchin, Hemicentrotus pulcherrimus. The
chelator dependent inhibition was prevented
by zinc. On the other hand, measurement of
metal ion contents uSing neutron activation
analysis (Ins. for Atomic Energy of Rikkyo
Univ.), the most sensitive technique for
detection of zinc, showed that zinc content
in the spermatozoa of starfish decreased
after movement of spermatozoa, whereas zinc
content in the spermatozoa of sea urchin
showed no change. These findings suggest
that the removal or addition of zinc that
is inhibitory in starfish or acceleratory
in sea urchin to motility, regulates sperm
motility in echinoderm.

PH 106

INTRACELLULAR ca2t INCREASES ACCOMPANIED
WITH INITIATION OF PUFFER SPERM MOTILITY

- Oda, M. Morisawa , and Y. Kudo’.

Misaki Marine Biological Station, Fac. of
Sci., Univ. of Tokyo, Miura, “Mitsubishi-
Kasei Inst. Life Sci., Tokyo
In the most of marine teleosts, sperm mo-
tility is initiated by an increase in ex-
ternal osmolality (Morisawa & Suzuki, 1980),
We previously reported that coexistence of
ionophore, A23187 and Ca“* can induce the
initiation of puffer sperm motility without
an increase in external osmola}ity. This
suggests that intracellular Ca uihas, aukey:
OILS se@ie ielmE) aliqaliedeyealoyal Cpe eeliol eye motil-
ityeimethe: present study, using Gacy £lu0=
rescence probe, quin-2 and the fluorescence
spectrophotgmeter, an increase in intra-
cellular Ca“* was practically detected when
sperm motility was initiated in the hyper-
tonic solution. Intracellular Ca“* concen-
tration was 21.4 nM in the isotonic solu-
tion and it increased to 99.0 nM in the
hypertonic solution ( REET nM ). When
changes in intracellular Ca YO A diuneAS2
loaded sperm cell was analysed by the fluo-
rescence microscope/video-camera system,
Ca“* increased transiently and concomitant-
ly with motility initiation in hypertonic
osmolality, and then, it fell to low level
within 10 seconds. These results suggest
that increase in external osmolality causes
a transient increase in intracellular Ca“*
and Ca“* may induce the initiation of fla-
gellar beating in marine teleost sperm.

PH 107

A NEW GLIDING ACTIVE DYNEIN ATPASE FROM
OUTER ARM DEPLETED AXONEMS
S. Wada, M. Okuno and H. Mohri

Colle sob Arts iG oCin a Unive Of Lokyo.-hokyo.

We prepared the gliding active ATPase,
which was referred to as IA, from the outer
arm depleted axonemes of sea urchin sperm
flagella by selective extraction and by col-
umn chromatography. This ATPase was supposed
to be derived from the inner arms since the
anti-21S antibodies didn’t cross-react with IA
IA had one major high molecular weight peptide
which was the same mobility as A B-chain of
21S outer arm dynein on SDS-PAGE. We compared
the properties of the gliding activities
between AB dynein (one of the outer arm com-
ponent) and IA.

At the low Mg-ATP concentration (0. 01mM-
0.1mM) AB failed to attach the microtubules
but IA induced gliding of microtubules. On the
other hand, IA couldn’t glide microtubles at
the high Mg-ATP concentration (0. 5mM-1. 0mM)
but AB did. The microtubule gliding was also
observed in the presence of Ca-ATP substituted
for Mg-ATP. The microtubules gliding with IA
was inhibited by 10°®M Vanadate. This value
was 1/100 of that required in AB. The sensi-
tivity of pH were different between AB and IA
From these result, we assumed that IA was one
of the inner arm components in sea urchin
sperm flagellar.

PH 108

MOTILITY OF THE DEMEMBRANATED SALMONID FISH
SPERM FLAGELLA IN THE ABSENCE OF CYCLIC-—
AMP.

M. Okuno. Department of Biology, College
of Arts and Sciences, University of Tokyo,
Tokyo.

The flagellar movement of the demembra-
nated salmonid fish spermatozoa is initiat—
ed by cAMP in the presence of MgATP as re-
ported previously. However, the flagellar
movement was observed in the absence of
cAMP when the concentration of ATP was low.
The motility was maintained even in the
presence of adenylcyclase inhibitor sug-
gesting that the flagella possess an activ-
ity to make the bending wave under some
conditions even in the absence of CAMP.
The present study was focused on the fea-
ture of the flagellar movement in the
absence of CAMP.

The vigorous motility was observed at
the low concentration of MgATP lower than
O.1 mM in the absence of CAMP. When the
demembranated flagella were reactivated in
the absence of cAMP the beat frequency was
a little lower than that in the presence of
cAMP. Both the bend angle and the ampli-
tude of bend were a little larger. Howev-
er, the asymmetry of bend increased drasti-
cally and the three dimensional abnormal
bending waves were often observed. These
results may suggest that cAMP acts in the
control mechanisms which regulate the
metachronal activation of dynein arms
through the flagellar axoneme.

1046 Physiology

PH 109

CHEMOACCUMULATION IN BUBPHARTSMA

vt. Matsuoka, M. Takahashi and K. Taneda
Department of Biology, Faculty of Science
Kochi University, Kochi 780

Blepharisma accumulated in the
vicinity of pellet of alive bacteria
(Enterobacter) which was separated with
cellulose membrane from the cells of
Blepharisma. The cells, however, could
not accumulate in heat-treated pellet.
The crude and heat-treated supernatants
obtained from suspension of the bacteria
also induced conspicuous chemoaccumulation
Oe cae Celis. These results suggest
that the. cells of Blepharisma can detect
certain small molecules produced by alive
bacteria which can pass through the
cellulose membrane. Ninhydrin-positive
substances, which were isolated from
the products of the bacteria by means
o£ TLE, still had an activity to attract
the cells of Blepharisma. The result
indicates that Blepharisma detects certain
NH, compounds, which are secreted products
of bacterial metabolism.

PH 110

RESPONSES TO GRAVITY AND GRAVITAXIS IN
PARAMECIUM.

Akira Murakami, Kenjiro Yoshimura and Keiichi
Takahashi. Zool. Inst., Fac. Sci., Univ. of Tokyo, Tokyo.

The swimming behavior of Paramecium was recorded

and analysed under various gravitational conditions.
Under a video-centrifuge microscope, the paramecia
oriented parallel to the direction of the centrifugal force
with their anterior ends pointing either centripetally or
centrifugally. The threshold of the orientation response
was about 70 g. The time required for 50% of the cells to
orient (orientation time) varied depending on the
centrifugal force, and was about 1 and 2 min for 450 and
120 g, respectively. When paramecia were centrifuged
twice in succession, the orientation time in the second
centrifugation was shorter than that in the first. The
results suggest that the orientation response could be
accounted for by a purely physical mechanism only when
the gravitational acceleration is higher than about 70 g.
The gravitactic behavior observed under 1 g may therefore
involve a more sensitive physiological mechamism. The
presence of a physiological mechanism controlling the
response of paramecia to gravity was suggested by their
response to microgravity which was obtained by using a
droptower (18 m in height). In this experiment, the
swimming behavior of paramecia in a chamber was video-
recorded at 1 g for 30s or more and then the chamber and
the recording apparatus were released from the top of the
tower to fall freely while the recording of the behavior was
continued. Analysis of the records indicates that the
vertical component of the swimming velocity changed on
exposure to microgravity. The magnitude of the change in
swimming velocity was larger than the sinking rate of
paralyzed paramecia determined under 1 g. The results
suggest that a physiological, rather than. a physical,
mechanism is involved in the response to microgravity and
in the gravitaxis under 1 g.

PH 111

MEMBRANE POTENTIAL RESPONSES CONTROLLING
THE CHEMOTACTIC BEHABIOURS TO QUININE

IN THE CILIATE PARAMECIUM CAUDATUM

K. Oami and Y. Naitoh. Inst. Biol. Sci.
Univ. of Tsukuba, Tsukuba.

Membrane potential responses to chemi-
cal stimuli and their relations to
chemotactic behabiour in a ciliate
Paramecium caudatum were investigated.
Specimens swimming in a reference solution
(4 mM K, 1 mM Ca) did not entered into a
quinine-containing (0.5 mM) test solution
adjoining the) ret enwenice isionsuteiomes
Specimens showed conspicuous avoiding
response when they encountered the border
between these two solutions. In the test
solution, a specimen showed a sustained
membrane depolarization superimposed by a
train of depolarizing action potentials.
In response to a quick application of the
test solution, a wild type specimen showed
a sudden membrane depolarization followed
by action potentials, while a CNR specimen
showed a transient depolarization (phasic
response) followed by a sustained
depolarization (tonic response). Amplitude
of these responses increased linearly with
logarhysmic increase in the quinine con-
centration. A current-evoked action
potential in the test solution was larger
in its amplitude and duration than that in
the reference solution. It is concluded
that the negative chemotactic behabiour of

Paramecium to quinine is caused by a

chemoreceptor potential-mediated avoiding
response at the border.

PH 112

CHEMOKINETIC RESPONSE IN AMOEBA PROTEUS.
A.Kihara, K. Ishii, Lab. of Biol., Hosei Univ., Tokyo.

To date, there have been few studies of the chemo-
kinetic properties of amoebae, mostly because of the
difficulty of evaluating their movements. In this study,
we try to evaluate this movement quantitatively, by
using recently developed original image-analyzing
system.

Accordingly, the centroid of the amoeba was calcu-
lated from a two-dimensional projected image at every
sampling. Movement was closely simulated by computer-
tracing centroids. So we tentatively defined the velocity
of the amoeba in terms of the vector which joined
successive pairs of centroids.

The long-term recordings (2-3 hours) of the spontane-
ous movements of amoebae indicated two modes, axial
and non-axial. In the axial phase, amoebae moved with
the elongation of the main pseudopodium, and the
direction changes of velocities were within +30 degrees.
In the non-axial phase, the main axis of movement
disappeared and fluctuations of the periphery were
observed. In this phase, velocity was observed to
decrease, and direction to alter almost randomly.

In solutions of nonanoic acid (at 1x10°, 5x10° Mol.),
velocity in the axial mode decreased and the periods of
the non-axial mode increased with increased concen-
tration. Both effects resulted in decreased motility. But
there were no significant differences in non-axial mode
velocity at either concentration.

From these results it is suggested that the axial mode
velocity and non-axial mode period can be the
important factors in the detection and evaluation of
kinetic response of amoeba.

PH 113

VALENCY OF VANADIUM IN VANADOBIN, VANADIUM-
BINDING SUBSTANCE, EXTRACTED FROM ASCIDIAN
BLOOD CELLS

H. Michibata’, H. Hirose’ and K. Kanamori’, ‘Biol. Inst.
and *Dept. Chem., Fac. Sci., Toyama Univ., Toyama.

Vanadium(III) sulfate, dissolved in acidic solution,
exhibits absorbances at 410 and 610 nm, and
vanadium(IV) oxide sulfate an absorbance at 760 nm
with a shoulder at 625 nm. Vanadobin, extracted from
the blood cells of Ascidia gemmata at pH 2.3 under
anaerobic conditions, shows closely resemble
absorbances at 410, 620 and 756 nm. Changing a
mixing ratio of the two inorganic vanadium solutions,
a simulated spectrum of vanadobin could be obtained
and the vanadobin was found to contain both
vanadium forms in +3 and +4 oxidation states in the
ratio of 45:55 after gel filtration onto Sephadex G-15.
The similar vanadobin could be obtained from _ the
blood cells of all the ascidians; A. ahodori, A.
sydneiensis samea, A. zara, Corella japonica and Ciona
intestinalis, examined in this experiment.

PH 114

VALENCY OF VANADIUM IN THE VANADOCYTES OF
ASCIDIA GEMMATA SEPARATED BY DENSITY-GRADIENT
CENTRIFUGATION

J. Hirata’ and H. Michibata’, ‘Dept. Biol., Tsurumi
Univ., Yokohama and “Biol. Inst., Fac. Sci., Toyama
Univ., Toyama.

Vanadium presents in living organisms in the +5,
+4 and +3 oxidation states. Although ascidians are the
only organisms known to contain high levels of
vanadium in their blood cells, the oxidation state of
this metal is not always known. In the present study,
we determined the amounts of total vanadium and
vanadyl(+4) species contained in each blood cell type
of ascidians, Ascidia gemmata, under anaerobic
conditions, using a combined approach. of cell
fractionation and vanadium measurement by ESR
spectrometry. As the results, it was revealed that the
oxidation state of vanadium in the vanadocyte
remained constant during the process of cell
separation and the separated vanadocytes contained
vanadium in both +3 and +4 oxidation states, with a
ratio of 97.6:2.4.

Physiology

1047

PH 115

VANADOBIN, A VANADIUM-BINDING | SUBSTANCE,
EXTRACTED FROM ASCIDIAN BLOOD CELLS, SHOWS AN
ABILITY TO REDUCE VANADATE(+5) IONS TO
VANADYL(+4) IONS

A. Morita’, H. Michibata' and K. Kanamori’, ‘Biol. Inst.
and Dept. Chem., Fac. Sci., Toyama Univ., Toyama.

Highly concentrated vanadium in ascidian blood
cells is reduced to the +3 and/or +4 oxidation states,
although the vanadium dissolved in seawater seems to
exist as the vanadate(+5) anion. Therefore, in the
blood cells, some reducing agents have been assumed
to be contained. Since the supernatant of homogenate
of the blood cells in Ascidia gemmata, exhibited an
ability to reduce vanadate(V) to vanadyl(I1V) under
anaerobic conditions, vanadobin extracted from the
supernatant was further examined. Consequently, the
present experiments have revealed that vanadobin
showed an ability to reduce vanadate(V) to vanadyl
(IV), using ESR spectrometry and spectrophotometry.

PH 116

NON-INVASIVE MEASUREMENT OF INTRACELLULAR pH
IN THE ASCIDIAN VANADOCYTES BY ESR

Y. Iwata’, J. Hirata’ and H. Michibata’, 'Biol. Inst., Fac.
Sci., Toyama Univ., Toyama and *Dept. Biol., Tsurumi
Univ., Yokohama.

The reported pH values of ascidian vanadocytes
vary considerable. One of the reasons for the
variation in reported pH values may _ be _ that
measurements of pH were made without fractionation
of the total population of cells. Pure subpopulation of
the vanadocytes, that was separated by density-
gradient centrifugation, was submitted for
measurment of pH by ESR under anaerobic
conditions. Consequently, the vanadocytes of three
ascidian species, Ascidia gemmata, A. ahodori and A.
sydneiensis samea, showed low pH values of 2.4, 2.7
and 4.2, respectively. Results obtained by ESR
spectrometry confirmed the values of pH obtained
with the microelectrode under anaerobic conditions.
Therefore, it is very possible that the vanadium ions
are sequestered in an organic complex under the
acidic conditions in the vacuole of the vanadocytes.

1048

PH 117

MONOCLONAL ANTIBODY SPECIFIC TO SIGNET ' RING
CELLS, THE VANADOCYTES OF TUNICATE, ASCIDIA
SYDNEIENSIS SAMEA

T. Uyamal, T. Nishikata’, N. Satoh’ and H. Michibata’,
‘Biol. Inst., Fac. Sci., Toyama Univ., Toyama and Dept.
Zool., Fac. Sci., Kyoto Univ., Kyoto.

Signet ring cells have been newly identified as
the vanadocytes (vanadium-containing blood cells).
We raised monoclonal antibodies to the signet ring
cells of Ascidia sydneiensis samea for use in the
identification of vanadocytes. A hybridoma cell line
which secretes a monoclonal antibody specific to the
signet ring cells have been’ established. The
monoclonal antibody reacted with the cytoplasm and
vacuole of cell of the same type but did not react with
the other cell types of cell. Moreover, this antibody
was also shown to be suitable for identification of the
vanadocytes in two further species. Immunoblotting
analysis showed that the antigen recognized by the
antibody was a single polypeptide of about 45 kDa.

PH 118

UPTAKE OF L- AND D-ALANINE BY A BRACKTISH-
WATER BIVALVE IN RELATION TO SALINITY.

O. Matsushima. Zool. Inst., Fac. of Sci.,
Hiroshima Univ., Hiroshima.

The brackish-water bivalve, Corbicula
japonica, contains high concentrations of
L- and D-ala (20-40 mM each) as
intracellular osmolytes, which fluctuate
remarkably with external salinity.

Uptake of L- and D-ala by C. japonica was
studied in relation to external salinity
and the transport pathways. The animals
acclimated to 250 mOsm (1/4 SW) took up
L-ala at external concentration as low as
10 pM and showed much higher uptake rates
than the freshwatey-acclimated animals.
Radioactivity of C-ala was accumyfated
largely ag @pbILIlLS, -IsMeIo Ose loyorela C-D-
and C-L-ala into isolated gills was
inhibited by O&-neutral amino acids

including C-ala (D-ala was more sensitive

to the inhibition than L-ala), but not
by &-acidic or$-amino acids. Addition of
excess D- and L-ala increased half-
saturation constant (Kt) for L- and D-ala
influx, respectively, without influencing
the maximal rates (Jmax). The results
suggest that the transport mechanism may
play an important role in maintaining the
steep concentration gradients of ala
between environment and intracellular
fluid and that D- and L-ala are
transported via the same pathway(s), but
the pathway(s) is more accessible to L-ala
than to D-ala.

Physiology

PH 119

TRANSPORT OF GLUCOSE IN MALARIA-INFECTED
ERYTHROCYTES.

K. Tanabe and A. Izumo. Lab.) Of "Bol
Osaka Institute of Technology, Osaka.

Transport of D-glucose in Plasmodium
yoelii-infected mouse erythrocytes were
studied using 2-deoxy-D-glucose (2DOG).
Infected red cells increased the 2DOG
uptake. Kinetic analyses measuring the
initial rates of 2DOG uptake showed two
components in infected erythrocytes: a
simple difussion system and a carrier-
mediated system. The carrier was common
for D-glucose and 2DOG and had a Km for
2D0G of 0.18mM. In contrast, the mouse
erythrocyte Carrier had a Km of 10 mM.
Much of ~H-2D0G preloaded in infected
erythrcoytes was localized in the para-
site compartment and a 15-fold concen-
tration gradient of 2DOG was created
across the parasite plasma membrane.
These and experiments with metabolic in-
hibitors indicated that the simple dif-
fusion system localizes in the membrane
of the host erythrocytes and the carrier
in the parasite plasma membrane. The
glucose carrier was much less sensitive
to cytochalasin B than that of the host
cells. Our study thus suggests that D-
glucose gains entry into the cytosol
of infected erythrocytes via the simple
diffusion system and an uptake against
the concentration gradient takes place
at the parasite plasma membrane.

PH 120

GLUT1 GLUCOSE TRANSPORTER IS PRESENT IN
BLOOD-TISSUE BARRIERS.

K. Takata’, T.Kasahara?, M.Kasahara?,
O.Ezaki?, H.Hirano?. *Dept. Anat., Kyorin
Univ. Sch. Med., Mitaka, Tokyo, 7Lab.
Biophys. Sch. Med., Teikyo Univ. Hachioji,
Tokyo, ?Div. Clin. Nutri., Natl. Inst.
Health Nutri., Tokyo.

Facilitated diffusion glucose
transporters are plasma membrane proteins
that transport glucose down its chemical
gradient. Five isoforms of facilitated
diffusion glucose transporters (GLUT1 -
GLUT5) have been so far identified in
animal cells. We examined whether glucose
transporters are present in blood-tissue
barriers in the rat. By immunoblotting
with antibodies against GLUT1, we found
that GLUT1 is rich in tissues with
blood-tissue barriers (brain, retina,
Ciliary body/iris, and placenta).

Immunof luorescence microscopy of semithin
frozen sections revealed that GLUT1 is
localized in plasma membranes of
microvessel endothelial cells in brain,
iris, inner retina, and peripheral nerve.
GLUT1 is also concentrated in the plasma
membranes of epithelial cells in choroid
plexus, pigment epithelium, ciliary body,
and perineurium. In placenta, plasma
membrane of syncytiotrophoblast is rich in
GLUT1. GLUT1 localized in these limiting
plasma membranes of blood-tissue barriers
may play a vital role in permeation of
glucose through the barriers.

lial il

EFFECTS OF EEL ATRIAL NATRIURETIC PEPTIDE
ON ION TRANSPORT ACROSS THE SEAWATER EEL
INTESTINE.

M.Ando', and Y.Takei?. ‘Lab. of Physiol.
Fac. Integrated arts and Sci., Hiroshima
Univ., Hiroshima and 2Dept. of Physiol.,
Kitasato Univ., School of Med., Sagami-
hara.

To elucidate how ion and water trans-—
port is regulated in the seawater eel
intestine, effects of eel atrial natri-—
uretic peptide (eANP) were examined. When
the eANP(1-27) was added to the serosal
fluid, serosa-negative transepithelial
potential difference (PD), short-circuit
current (Isc) decreased dose-dependently.
The tissue resistance tended to increase
after treatment with high dosage. Similar
effects were observed after treatment with
eANP(3-27) or eANP(5-27), indicating that
the N-terminal Ser-Lys-~Ser-Ser residue
does not contribute to its action. Human
ANP and rat ANP also decreased PD and Isc
dose-dependently, though 100 fold higher
concentration was required,. indicating
that active structure is the S~-S bond and
common 17 amino acid residues and remain-—
ing 6 amino acids contribute to affinity
for the eANP receptor. Among these 6
amino acids, 12th amino acid can be omit-
ted, since hANP and rANP has the same
effects. Interestingly, rANP(5-27) had
lower affinity for the eANP receptor than
rANP(5-28), indicating that C-terminal
Arg-Tyr of rANP is one of the sites at-—
tributable to the affinity.

PH 122

REGULATION OF NACL AND WATER TRANSPORT
ACROSS THE SEAWATER EEL INTESTINE: EFFECTS
OF NORADRENALINE.

K.Kondo and M.Ando. Lab. of Physiol.,
Fac. of Integrated Arts and Sci., Hiro-
shima Univ., Hiroshima.

NaCl and water absorption across the
seawater eel intestine was significantly
inhibited by 0.1 pM eel atrial natriuretic
peptide (eANP), 1 pM methacholine (MCh),
an agonist of acetylcholine (ACh), or 10

pM 5-HT. However, no stimulators of NaCcl
and water absorption has not been report-
ed. Using a special preparation which has

low transepithelial potential difference
(PD), short-circuit current (Isc) and
tissue resistance (Rt), we have chanced to
find out that 10 pM noradrenaline (NA)
enhances the serosa-negative PD, Isc and
Rt, whereas its effects are too small to
detect in usual preparations. Similar
effects of NA were obtained after pre-
treatment with | pM MCh and 10 pM 5-HT,
accompanied by an increase in water ab-—
sorption. In the presence of both MCh and
5-HT, NA receptor was characterized. The
effects of NA were mimicked by clonidine
(0.1 pM) or guanabenz (0.1 pM), di -ago-
nists, but not by phenylephrine (1 pM), dad,
-agonist, nor isoproterenol (10 pM), ~ -
agonist, and completely blocked by yohim-
bine (10 pM), dz-blocker, indicating ex-
istence of di-type receptor in the sea-
water eel intestine. The di: -type receptor
seems to antagonize the inhibitory effects
of ACh and 5-HT.

Physiology 1049

PH 123

INCREASE OF RETINALS AND RETINOLS BY

ESTROGEN ADMINISTRATION AND RET INAL-—

VITELLOGENIN COMPLEX IN THE BLOOD OF
XENOPUS LAEVIS.

1 2

T. Irie , M. Azuma”, and MT. Seki.

losaka Meijo Women's College, Osaka and
2Dept. Health Sei., Osaka Kyoiku Univ. Osaka.

Administration of 8B -estradiol to male
X. laevis induced extreme increase of
retinals and retinols in the blood plasma,
along with the induction of vitellogenin.
The vitellogenin fraction obtained by
Mg-EDTA precipitation possessed most
retinals in the blood but little retinol.
The vitellogenin fraction was homogeneous
in SDS-PAGE. These results reveal that
blood retinals are bound to vitellogenin
in X. laevis.

The retinal, / retinalg ratio of

vitellogenin-bound retinals was 0. 52-0. 88.
These values are slightly lower than
those of lipovitellin-bound retinals in
Xenopus eggs. One mole of retinals was
present in 557-1180 moles of vitellogenin.
The retinal/protein ratio on the
vitellogenin was less than one thirtieth
of the value on egg lipovitellin. The
composition of retinal isomers bound to
vitellogenin was different from that bound
to lipovitellin. These results suggest
that egg retinals originate only partly
from retinals bound to blood
vitel logenin.

PH 124

ALLORECOGNITION IN A MARINE SPONGE, HALI-
CHONDRIA JAPONICA (DEMOSPONGI A)

Y.Saito, Shimoda Mar. Res. Ctr., Univ. of
Tsukuba, Shimoda.

The allorecognition ability of Hali-
chondria japonica was examined by using
cut pieces contact assay. When a cut
piece of an individual was brought into
contact with another piece of the same
individual, the two pieces always fused
with each other compatibly. When a cut
piece of an individual was brought into
contact with a cut piece of another indi-
vidual, in most cases rejection reactions
occurred between those two pieces.

The processes of the rejection reac-
tions were studied histologically. Within
two hours after contact, the two pieces
began to fuse with each other 1ike as
compatible pairs. Sequentially, mesohyl
cells of both pieces began to migrate to
the fused area and gradually accumulated
there. About one day after contact, the
number of accumulated cells became nearly
its maximum, and at this time there were
many cells that phagocytosed other cells
among the accumulated cells. Two days
after contact, the mesohyl cells fell into
line on either side of the plane of con-
tact and the boundary became clear.

These results showed the presence of
allorecognition ability in H. japonica,
and suggested that mesohyl cells might
play the leading part of the allorecogni-
tion in this species.

1050

PH 125

THE EFFECT OF THE BODY SURFACE ‘MUCUS
AND CARBOHYDRATE ON PHAGOCYTOSIS IN THE
LAND SLUG, INCILARIA FRUHSTORFERL.

K. Yamaguchi, Te aveemiee and A Sindnomanal

liab. of Med. Sci. and 2Dept. of Anat.,
Dokkyo Univ. Sch. of Med., Tochigi.

The body surface of land slug is
covered with rich mucus which may serve
in preventing the moisture evaporation,
in helping smooth movements and in
protecting the body from mechanical
injures. And some biochemical function
may involved in the mucus. As previously
reported by us, macrophage-like cells
in | the hemolymph of the Wand slug
phagocytose SRBC and latex beads in vivo
and in vitro. In this study, an opsonizing
effect on the mucus for various intervals
and an inhibiting effect of carbohydrate
(GalNAc ) on phagocytic activity were
observed. At 1, 3 and 6 hrs after the
injection of SRBC treated with mucin
fraction, several SRBC were phagocytosed
by the macrophage-like cells, however,
ebe 3) Gis) © Joves} GUEEGIG Tela synsiecieslom! Cie
non-treated SRBC, very few hemocytes
phagocytosed only one SRBC. And GalNAc
inhibited the phagocytosis significantly.
These results suggest that body surface
mucus of the land slug contains opsonizing
factors and lectin-like activity.

PH 126

AGGLUTINATING ACTIVITY OF THE BODY SURFACE
MUCUS AND THE HEMOLYMPH OF THE LAND SLUG,
INCILARIA FRUHSTORFERI.

E.Furuta?, K. Yamaguchi and A.Shimozawa?
Inept. of Anat. and OE. of Med.
Sci.,Dokkyo Univ. Sch. of Med., Tochigi.

Most invertebrate agglutinins are
believed to be lectins and some molluscan
lectins have been shown to have opsonic
activity when tested against relatively
small particles. Body surface mucus of
the land slug was diluted with 5 vol
of water, stirred overnight and centri-
fuged at 10,000xg for 2 hrs at 4°C to
remove insoluble materials(WSF). The
collected hemolymph were centrifuged
(100xg, 10 min) to remove hemocytes and
extraneous debris(SH). WSF agglutinated
the human A and B type erythrocytes and
WSF induced hemagglutaination was, how-
ever, specifically inhibited by GalNAc
(70uM) . WSF also caused hemolysis of
B type HRBC 12 hrs after agglutination.
Studies on the effect of WSF against
human hepatoma cells are in progress.
All these WSF activities were completely
aineroicaieneeyol loyie Joveveticaline; Gig BOG sz@% SO)
min. SH agglutinated A and B type HRBC,
but GalNAc, GlcNAc and NeuNAc effectively
inhibited the hemagglutination. These
results suggest that the mucus and the
hemolymph contain the lectin-like activ-
Vie

Physiology

PH 127

STICKY SUBSTANCE IN THE EARTHWORM COELOMO-
CYESE
F.Shishikura, S.Ohtake, T.Abe, and) Ke danaica
Dept. of Biol., -Nihon Unive ‘Sches ofeiMede
Tokyo.

Coelomocytes of earthworms, Pheretima
hilgendorfi, were separated from the coe-
lomic fluid ejected from the worm's dorsal
pores into a beaker containing anticoagu-
lant such as sodium citrate (10%) and EDTA
(Ora Se The cells were divided into 5
types, including 2 kinds of amebocytes,
chloragogen cells, linocytes and peritoneal
cells under a differential interference
microscope. When the packed cells were
resuspended with hypotonic solution which
increased the disruption of cells, the cell
suspension instantly become sticky coagu-

lum. Moreover, the components of the
packed cells were extracted with 20% of
acetic acid solution, The soluble compo-

nents was converted into a sticky form by
dialyzing against water at neutral pH. The
sticky substance is therefore located
within coelomocytes and it appears that the
conversion does not require extracellular
factors. The protein components of the
coelomocyte lysate in acetic acid was shown
to be 4 major bands on SDS-PAGE. The
highest molecular weight protein seems to
be responsible for the formation of coagu-
lum. The protein was purified by a repeat-
ed gel stiltrat tony )sShowsmeemam appa yen
molecular weight to be more than 1 x 10° in
native form and 160,000 in reduced form.

PH 128

THE INITIAL PHASE OF ENCAPSULATION IN THE
BRAINLESS PUPA OF SAMIA CYNTHIA RICINI
S.Takahashi, Dept. of Biol., Nara Women's
Univ., Nara.

The initial phase of encapsulation was
examined in 70-day-old brainless pupae of
Samia’ (cynithilal “Eilcinay silicone wos sayais
used as a foreign object. When small
amount of the oil (<10pl) was injected
into the unanesthetized pupa at 25C, the
oil balls were encapsulated within 48hr.
However, when the specimen being
anesthetized with CO and/or chilled on
ice, was received the oil injection, the
oil was hardly encapsulated. In specimens
experimented at 25C). dineludanamlones mort
sham operations, circulating hemocytes
converted from spindle to lamellate and
increased from 180 to 900 in number during

24hr after injection. This occurrence
showed that hemocytes responded to an
integumentary injury. In all the

specimens, it could be observed that the
fat-body always clung to the surface of

the oil and around them, a number of
hemocytes aggregated. It was suggested
that the fat-body might secrete

lectin-like substance near the foreign
object, toward which hemocytes aggregated.
A smooth layer which first formed around
the oil surface, became the innermost
layer of the capsule. On this layer,
network structure was formed. Lamellar
hemocyes clustered on the network.

Physiology 1051

PH 129

STRUCTURE OF HEMOCYTES OF THE JAPANESE
HORSESHOE CRAB TACHYPLEUS TRIDENTATUS.
Y.Toh, A.Mizutani, F.Tokunaga, and
S.Ilwanaga. Dept. of Biol., Fac. of Sci.
Kyushu Univ. Fukuoka

Structure of hemocytes in the normal
state and during blood coagulation was
examined in the Japanese horseshoe crab.
The hemocyte contains numerous dense
granules classed into two major types; L-
and D-granules. The L-granules are larger
(up to 1.5 um thick) and less electron
dense than the D-granules (less than 0.5
um thick). Immunohistological data showed
that the L-granules contained three
clotting factors (proclotting enzyme,
factor C and coagulogen) and one
antimicrobial factor (anti-LPS factor),
whereas the D-granules contained the other
antimicrobial factor (tachyplesin). The
granular hemocyte possesses a single well
developed Golgi complex. Materials to be
included in both types of granules are
assumed to pass through the same Golgi
complex, and they appear to be sorted in
the Golgi cisternae on the trans side or
during package as Golgi vacuoles.
Precursor granules of L-granules and D-
granules appear as those with tubular
contents and those with condensed dense
materials, respectively. After treatment
with endotoxin, the L-granules were
released more rapidly than were the D-
granules, though almost all granules were
finally exocytosed. These data are
discussed in relation to the self defense
mechanisms of the horseshoe crab.

PH 130

THE PHAGOCYTIC ACTIVITY OF SMALL GRANULAR
AMEBOCYTE OF HALOCYNTHIA RORETZI

S.Ohtake, T.Abe, F.Shishikura, and K.Tanaka
Dept. of Biol., Nihon Univ. Sch. of Med.,
Tokyo.

The phagocytic ability of hemocytes of
HepsOmcitezt and imgestible tangets were
studied in vitro. Freshly collected hemo-
lymph was incubated at 23 °C with latex
peadsiGEB) wot 1 55 (25 mein diam; ~¢lutar-
aldehyde-treated SRBC and Escherichia coli
suspended in Pantin's artificial sea water
(HEPES, pH 7.2)(ASW) and examined by an
electron microscopy. Small granular amebo-
cyte (SG) actively ingested all kinds of
the particles excepting 25 pm-LB which were
encapsulated by SGs spreading very thin.
Large granular amebocytes ingested only 1
pm-LB. Their activity was, however, weaker
Ehanwehiat Or SCs we could = not find! any,
phagocytosed particles in other hemocytes
during 60 min incubation. Further experi-
ments were done for searching factors
effecting phagocytosis on washed hemocytes
with 1 pm-LB in various incubation media.
Cells were smeared on a slide glass and the
number of SG was counted under an optical
microscope after May-Griinward-Giemsa stain.
The percentage of SG ingesting particles
at 5, 10, 20 min of incubation was signifi-
cantly higher by 6-23 % in Ca*Mg free acid
ASW (0.54 mM EDTA, MES, pH 5.6) than in
fresh hemolymph plasma, in neutral ASW or
in neutral ASW containing 1/10 volume of
hemolymph plasma or tunic-treated plasma.

PH 131

THE STUDY ON THE HEMOCYTES OF ASCIDIAN
HALOCYNTHIA RORETZI

T.Sawadal, Y.Fujikural, S.Tomonaga2 and T.
Fukumotol, lyamaguchi Univ.Sch.Med.,2Sch.

Allied Health Sci. Yamaguchi Univ., Ube.

We proposed the standard condition to
distinguish the hemocytes of Halocynthia
roretzi, (putting on slide glass for 15min
in 5mM EGTA and 1/2M NaCl at pH 5.8-6.0,in
which hemocytes were distinguishable under
phase contrast microscope).In this condit-
ion, we Classified the hemocytes into ten
types; large amoebocyte (30-40%:vacuolated
cell in Fuke 1979), vesicular cell (15-
27%), granular cell (2-7%) ,minute granular
cell (12-23%), macrophage (7-12%: fine
granular amoebocyte in Fuke 1979), morula
cell (5-10%), large basophilic cell (1-23),
large acidophilic cell ( 1%), single
granular cell (1-3%), lymphocyte (1-5%).

Large amoebocytes (LA), irregularly
shaped small cells before they extended on
matrix, were characteristic as encapsulati-
ng cells in their rapid extension on
foreign substances, little mobility and
ability to make syncytium. only LA and
macrophages exhibited active phagocytosis.
Macrophages were more active in movement
and phagocytosis than LA.

Minute granular cells, round shaped
cells with a tendency to make pure
aggregation, were identifiable with mono-
clonal antibody UB15 (Sawada et. al. 1989).
They turned into fibroblast-like cells
after 24h incubation and formed a seat on
which other hemocytes associated.

lelal gy

THE PRIMITIVE COMPLEMENT SYSTEM IN AGNATHA :
CHARACTERIZATION OF ZYMOSAN-BINDING PROTEINS
OF HAGFISH SERUM

T. Fujii’, A. Sekizawa’, & S. Tomonaga®

‘Hiroshima Women’s University, Hiroshima, *Seitoku University,

Matsudo & “Yamaguchi University, Ube

The zymosan-binding proteins of hagfish serum (HZBP) were
analyzed by immunoelectrophoresis (IEP) and SDS-PAGE. Zymosan
particles were treated with normal serum of hagfish, Eptatretus
burgeri, for 1 hr at 10. The serum-treated zymosan was washed with
PBS followed by incubation with 1% SDS in PBS (pH7.5). The HZBP
were effectively eluted from the washed zymosan by subsequent
treatment with an alkaline buffer (pH 11.0) containing 0.1M hydra-
zine. Upon IEP, the greater part of the eluted proteins formed a single
precipitin line in the slow (-globulin region against the rabbit anti-
whole hagfish serum as well as the anti-HZBP. SDS-PAGE analysis of
eluted proteins revealed an intense major band (HX) migrating with a
MW of 170K. After reduction, the HX migrated as two proteins with
MWs of 98K and 70K. The HX, however, was not eluted from zymosan
incubated with serum which had been pretreated with methylamine,
EDTA, or heated to 42°C for 30min. These results suggest that there is
an alternative-like pathway of complement, and that the HX is regarded

as representing the C3 of the hagfish.

1052

PH 133

ANALYSIS OF TRANSPLANTATION TOLERANCE
INDUCED IN METAMORPHOSING XENOPUS TADPOLES.
S.Tochinai, Zoological Institute, Faculty of Science,
Hokkaido University, Sapporo

The J-strain (JJ) clawed frog, Xenopus
laevis, can easily be rendered tolerant against
allogeneic or semixenogeneic (X. laevis x X. borea-
lis; JB) adult skins grafted perimetamorphically,
through a specific suppression. The mechanism of
suppression has been analyzed as follows. When
immunocompetent JJ tadpoles were thymectomized
(Txd) before or simultaneously with JB skin graft-
ing, tolerance was not induced. On the other
hand, if thymectomy was made on tadpoles 2 weeks
after or later, perfect tolerance was induced.
When immunocompetent Txd JJ tadpoles were
grafted with stage-matched JJ thymuses and the
JB skin at the same time, tolerance was induced.
The effect of thymus-graft was progressively
decreased thereafter, until 3 weeks after the skin
grafting when it was no longer observable. These
experiments strongly suggest that the thymus
plays a key role in producing a suppressor cell
population. If the cells must enter the thymus for
differentiation and/or maturation, the next ques-—
tion is whether the recognition of extrathymic
tolerizing antigen is prethymic or postthymic.
Since many attempts to detect the antigen-primed
suppressor cells or their precursors in the thymus
have so far failed, prethymic recognition seems
unlikely. The present hypothesis for suppressor
cell development is: Pre-suppressor cells are
produced in the metamorphosing tadpole thymus.
The thymus-emigrant immature cells recognize
extrathymic tolerizing antigens resulting in sup-
pressor cell maturation. Without appropriate
stimulation, they will abort in a very short period.

PH 134

HYMENOLEPTS NANA:CROSS-RESISTANCE AGAINST
ONCOSPHERE CHALLENGE IN MICE PREVIOUSLY
GIVEN VIABLE OR NON-VIABLE ONCOSPHERES OF
H. NANA,H. DIMINUTA,H.MICROSTOMA AND
TAENIA TAENIAEFORMIS.
J.Sasaki,K.Onitake,T.Takami and iNoIREO)c
Dep.of Biol.,Fac.of Sci.,Ymagata Univ.,
Yamagata and Dep.of Parasit.,and Pathol.,
Gifu Univ.School of Med.,Gifu.

When mice previously given oral inocula-
tion with viable oncospheres of the het-
erologous cestode species (Hymenolepis
and the homologous one (H.nana) were chal-
lenged with oncospheres of H.nana four
days after the primary inoculation, they
showed strong and complete resistance to
H.nana challenge, respectively. However,
the resistance was not evoked in mice
given nono-viable oncospheres of all ces-
tode species examined. Congenitally athy-
mic nude mice given viable oncospheres did
not show sny resitance of H.nana either.
Microscopic observations revealed that
oncospheres of H.diminuta, which require
beetles as the intermediate host like H.
microstoma previously reported, could in-
vade the mouse intestinal tissue. There-
fore, it is strongly suggested that the
strong cross resistanceto H.nana in mice,
induced by oncospheres of all heterologous
cestode species, is thymus-dependent and
due to the oncospheral invasions into the
intestinal tissue of mice.

Physiology

PH 135

SKIN GRAFT PERSISTENCE BETWEEN LITTER-
MATES OF INBREEDING JAPANESE HOUSE MICE,
T.A.Nomaguchi and Y.Sakurai. Dept. of Cell
Biol., Tokyo Metropol. Inst. of Gerontol.,
Tokyo.

Transplantation of skin grafts between
littermates of each generation(F6-F11) in
the inbreeding population of the mice, Mus
musculus molossinus, was performed. Pee

In the first experiment, ventral skin
grafts with white hairs from one donor were
transplanted to dorsal place of recipients
with agouti hairs in same littermates.
The recipients of 50% rejected the graft
within 3 weeks for F6 and F8, 4 weeks for
F9, 5 weeks for F7 and 6 weeks for F10.
The grafts were all rejected within 8 weeks
in littermates of the 6th generation, but
several members of the recipients accepted
the donor more than 10 weeks in F7-F10.
The persistent ratios at 10 weeks after
transplantation were 23.8% for F7, 14.3%
for F8, 24.5% for F9 and 25.0% for F10.

In the second experiment, the ‘cross
transplantation’ between littermates in F9-
F11 was performed, using the littermates
derived from members of the previous gene-
ration which were persisted the grafts for
3 weeks in the first experiment. Persist-
ence of the grafts was observed whether MHC
locus of littermates is identical or not.
The littermates consisted of hetero- and
homozygous MHC haplotypes in F9-F10, but
the transplantation succeeded in 8 litters
from 2 groups of inbreeding line in F11.

PH 136

Neonatal Tolerance Induction in The Thymus: Intra-
thymic Distribution of Tolerance-inducing Cells
and Micro-anatomical Site for Clonal Deletion.

M. Hosono, S. Ideyama and Y. Katsura.

Dept. of Immunol.,Chest Disease Research Institute,
Kyoto University, Kyoto.

The thymus is a central organ in the immune
system, where T lymphocytes are generated and
mature. This organ is thought to play an important
role for prevention of auto-reactive cell develop-
ment(self-tolerance). It is well known that allo-
geneic murine lymphocyte-stimulating antigens of
Mlsa, in association with major histocompatibility
complex (MHC)-class II antigens, preferentially
stimulate a large proportion of T cells with V§6
chains of T cell receptors (TcR). These T cells
could be detected by a-V86 monoclonal antibodies
coupled with appropriate indicator reagents.

When newborn mice less than 24 hr after birth
were intravenously injected with peritoneal cavity
cells of adult Mlsa mice, unresponsiveness to the
antigen was induced in the thymus of one-week-old
recipients. The unresponsive state was accompanied
with clonal deletion in medulla-distributing,
mature V86-positive thymic lymphocytes. In immuno-
histological analyses on seria: cryostat sections
of thymuses it was shown that medulla-distribution
of non-lymphocytes bearing donor type MHC-class II
antigens was correlated with induction of intra-
thymic tolerance. Similar phenomena were observed
in naturally occurine self-tolerance in the thymus
of Mlsa mice. Therefore thymic medulla is supposed
to be a site where negative selection process 1s
taken place in the developing thymus.

Physiology, Genetics 1053

PE 137

GOAT C-REACTIVE PROTEIN: ITS PURIFICATION
AND BIOCHEMICAL PROPERTIES.

W.Nunomura and T.Higashi. Nippon Bio-Test
Laboratories Inc. Kokubunji, Tokyo.

C-reactive protein (CRP), which is one
of the typical acute phase reactants in
some animals, was highly purified from goat
serum using at first phosphorylcholine-
agarose column followed by DEAE-Sephacel
(10 mM Tris-HCl buffer, pH8.0, 0-50 mM NaCcl
gradient) and Sephacryl S-200 gel filt-
ration. The purified goat CRP consisted of
five identical subunits each with a
molecular weight of 24kDa determined by
FPLC (Superose 6) and SDS-PAGE. In the
Cellogel electrophoresis, goat CRP moved in
the y-region. In the isoelectric focussing,
uSing 1.2% agarose, four bands of pI 7.35,
7.38, 7.43 and 7.49 appeared for goat CRP.

Specific antibody against goat CRP was
raised in rabbits. The antiserum formed a
Single precipitin line with goat serum as
well as purified goat CRP. This rabbit
antiserum also reacted with human CRP.

The serum level of CRP, as detected by
single radial immunodiffusion method, in
normal male goat (2-year-old) was 101.9+
34.5 ug/ml (n=21) and that during the acute
inflammation was 320.0+20.0 ug/ml (n=16).

GE 1

GENETIC STUDIES ON THE LIFE HISTORY TRAIT
IN BOMBYX MORI (L.) : INHERITANCE OF GROWTH RAT
A.Murakami, Natl.Inst.of Genetics, Mishima.

Domestic silkworm(B. mori)has an average
total life span ofca.2 months from fertili-
Zation to imaginal death after oviposition,
excepting a duration of embryonic hiberna-
tion periods. The life span of Bombyx gene-
rally depends on the difference of stocks
and indicating that the trait is under the
control of genetic factors. The larval and
moth periods are markedly variable to the
stocks in comparion with the embryonic and
pupal periods. No marked sexual difference
in the growth rate is usually observed from
fertilization to mounting, while the male
pupa shows a slightly but significantly higher
growth rate than the female: the male moth
has a tendency to early(ca. one day) emerge
previous to the female. The phenomenon is
thought to be under the control of Y chromo-
some functions rather than other chromosomes.
However, a reverse case was detected in an
F_hybrid of a cross between a tropical and
short life span stock Cambodge male and the
other stock(e.g. J106) female: the female
moth emerged at least two days earlier than
the male, while no sexual difference was
observed in the reciprocal cross. These
findings clearly suggest the stock Cambodge
is possessed of an X-linked recessive pre
(precocity) gene in the pupal growth process,
whereas the Jjl106 and other stocks holda gene
pre Also,it is very logical to suppose that
the Y chromosome restrains the pre-allelic
gene functions.

GE

COMPARATIVE STUDY OF CHROMOSOMAL RADIO-
SENSITIVITY IN MAMMALIAN SPERMATOZOA.

Y. Kamiguchi, H. Tateno and K. Mikamo,
Dept. of Biol. Sci., Asahikawa Med. Col.,
Asahikawa.

We studied the effects of X-irradiation
on sperm chromosomes of the golden hamster
(GH) and the Chinese hamster (CH), and
compared these results with previous data
on human spermatozoa. Mature male hamsters
were irradiated with 0.45 to 3.63 Gy of X-
rays on their lower abdomen and mated with
normal females within 5 days after the
exposure. Chromosome slides were prepared
at first cleavage metaphase, and 1,383 and
594 sperm-derived chromosome complements
were analyzed in GH and CH, respectively.

The results are as follows: (1) Inci-
dences of spermatozoa with X-ray-induced
structural chromosome aberrations increased
linearly with increase of dosage both in GH
and CH, reaching 45.0 % at 1.82 Gy in GH
and 52.1 % at 3.63 Gy in CH. These inci-
dences were much lower than that in human
(G260 8 ais 1o82 Gy) (2) In GH, the inci-
dence of breakage-type aberrations was far
higher than that of exchange-type aberra-
tions, as was previously found in human,
whereas these two types of aberrations
occurred at similar incidences in CH. (3)
Breakage-type aberrations showed linear
dose-dependent increase both in GH and CH.
Dose-dependent increase of the exchange-
type aberrations was linear in GH, while it
was quadratic in CH.

GE 3

SOME CHARACTERISTICS OF RADIATION- OR ENU-
INDUCED MUTATIONS AT THE b LOCUS OF THE
MEDAKA, ORYZIAS LATIPES.

A. Shimada and A. Shima. Lab. of Radiat.
Bi@ls , “OO, Jbnsics 5 Kee, Or Seino, Witty. OF
Tokyo, Tokyo.

In the course of developing radiation- or
ENU-induced specific-locus test system using
the Medaka, we have obtained to date 52
induced viable mutants at the b locus. Out
of these primary mutants, 4 females and 4
males were crossed with various combina-
tions. From all 16 crosses tested, PF
offspring with eyes like pearl were
segregated with a ratio of about 1/4. The
pearl-eyed embryos recovered from some
crosses were accompanied by dwarfism of
various severity. We tentatively named the
phenotypes of these fish ‘pearl eye', ‘pe’.
We found that the 'pe' fish were the homo-
zygotes of the induced mutations at the b
locus from the allelism tests and other
confirmation crosses.

Both embryos and adults of the 'pe' fish
have no melanotic melanophores on the body,
although slight melanization could be found
on the eyes. However, both the body and the
eyes of the 'pe' embryos showed positive
tyrosinase reaction. In eyes, the extent of
melanizaion was lower in the 'pe' fish than
in bP/bP fish, indicating that the 'pe' fish
show the lowest extent of melanization among
homozygotes of any other alleles of the b
locus.

1054 Genetics

GE 4

THE PRODUCTION OF GAMMA-RAY INDUCED
GYNOGENETIC DIPLOIDS IN THE MEDAKA, ORYZIAS
LATIPES

K. Naruse and A. Shima. Zool. Inst., Fac.
of Sci., Univ. of Tokyo, Tokyo.

We produced the gynogenetic diploid
Medaka by using gamma-ray irradiation to
sperm from O. celebensis and by the
suppression of second meiotic division. The
irradiation doses for sperm were 475, 713 or
950 Gy. The hatchabilities of the
gynogenetic embryos were ranging from 1.6 to
3.7 % . About 3 % of 'gynogenetic' embryos
carried melanotic melanophores. 4 out of 38
fish thus obtained were male. We have not
yet observed these phenomena in the
ultraviolet-light induced gynogenetic fish.
These results suggest some parts of paternal
genome contribute to development in the
gynogenetic Medaka induced by gamma-rays. To
confirm this result, we used the slot bolt
analysis using O. celebensis genome as
probe. DNA from 8 gynogenetic embryos or fry
out of 22 hybridized to O. celebensis
genome. From these results, we concluded
that the gynogenesis induced by gamma-ray
irradiation was not complete and these
"gyhnogenetic' Medaka were considered to be
transgenic fish. We examined the trans-
mission of O. celebensis DNA of gynogenetic
fish to next generation (7 Fy fish). By
PAGE of muscle proteins and slot bolt
analysis, we could not yet detect the
transmission of O. celebensis DNA to next
generation.

GE 5

CHROMOSOME MOVEMENT DURING MATURATION
DIVISION IN THE ISOLATED MEIOTIC APPARATUS
OF TUBIFEX HATTAI (OLIGOCHAETA,

TUBIFICIDAE)

M. Matsumoto and M. Anzawa. Dept. of Biol.,
Fac. of Sci., Yamagata Univ., Yamagata

yolky eggs of the freshwater oligochaete
Tt. hattai were freed from their cocoons in
Lehmannts solution. The vitelline membrane
and egg cortex of the fresh eggs were
ruptured with a pair of fine-tipped forceps
and yolk granules of the eggs dispersed in
a petri dish containing 2M glycerol. The
opaque core of the egg gradually emerged
within 2-3 minutes of the dispersion was
transfered to an isolation medium. During
the 3-5 hours in the isolation medium, the
egg core displayed fine structural features
composed of a spindle and the asters of
meiotic apparatus. These meiotic apparatus
were glued on a slide glass, dried and
stained with acetic orcein after glutar-
aldehyde fixation overnight. During ana-
phase I, if our procedures were performed
at 22-2)°c, formation of the cleavage fure
row of the polar body began during migra-
tion of the half-bivalent into the polar
body. However, some chromosomes still
remained in the narrow area between the
polar body and the egg cell. Furthermore,
at a higher temperature, the separation of
the first polar body from the egg was
completed before the termination of chromo-
some segregation. That is, some chromosomes
that should have migrated into the polar
body resided in the egg cell.

GE 6

SMALL CHROMOSOME FOUND IN TEMNOPLEURUS
TOREUMATICUS

K. Saotome, Yokohama City Institute of
Health, Yokohama.

When chromosome preparation was made
by air-drying method from Temnopleurus
toreumaticus embryos, two kinds of meta-
phase plates were observed. One had 42
chromosomes and the other had 43 chromo-
somes. The latter always carried one small
chromosome besides larger 42 chromosomes.

To clarify the origin of this small
chromosome, the following cases were
analyzed;1)the frequency in appearance of
small chromosome in various pairs of male
and female, 2)male chromosome from andro-
merogon, and 3)female chromosome from
hybrid (T. toreumaticus@% X Anthocidaris
crassispina @ 3

The results obtained were as follows:
1)the frequency in appearance of small
chromosome changed depending on the dif-
ference of male in various pairs, 2)there
were two types of metaphase plates in the
male chromosome from andro-merogon, of
which number was haploid until blastula
stage, one had 21 chromosomes and the
other had 22 chromosomes containing one
small chromosome, and 3)in female chromo-
some from hybrid, only one kind of meta-
phase plate without small chromosome was
observed.

These results indicate that small
chromosome is derived from male.

GE 7

C-BLOCK HETEROMORPHISM AND INVERSION POLYMORPHISM IN
THE CHROMOSOMES OF THE JAPANESE WATER SHREW, CHIMARRO-
GALE HIMALAYICA PLATYCEPHALA.

Y. Obara and J. Izumi. Dept. Biols, Fac of cl.
Hirosaki Univ., Hirosaki.

The No. | homologue of the Japanese water shrew
(Chimarrogale himalayica platycephala) carries quite
large C-bands (C-blocks) on its short arms which are
heteromorphic each other in size (Obara 1985). But, the
extent of their variation as well as their biological
significance are still unknown. The authors examined
the variation, combination and fluorescent pattern of
the heteromorphic C-blocks of 45 water shrew specimens
collected from four mountain streams (Nakanogawa river,
Orakumaegawa river, Okawa river and Oirasegawa river)
with different headwaters in Aomori Prefecture.

The results are presented together with inversion
polymorphism of the Y chromosomes which could be de-
tected by the C-band staining. The C-block heteromor-
phism was considered to reflect the substantial char-
acter of C-heterochromatin, and the Y chromosome poly-
morphism seemed to be genetic variation which has been
fixed in the Okawa-population.

GE 8

THE CHROMOSOME ANALYSES OF
BUERGERI WITH BANDING METHODS.

K. Sekiya. Depite or mBaoill Fac.
Niigata Wineiavaens

The diploid chromosome
Buergeruia buergeri (Amphibia,
26. The No.7 chromosomes
of this Species were
chromosomes, because
chromosomes were
satellite regions (Ohta,1986).

The male cells of this species
collected in Niigata Pref. have 1 or 2
nucleoli and female cells have only 1
nucleolus, just as those of the species
from the West of Japan did. The nucleolus
number of the cultivated cells of this
Species is the same as the above. This
phenomenon indicates that the
determination of the nucleolus number of
the species is controlled by the genetic
difference between male and female on the
cell-level also.

As a result of analyses with DAPI
staining and Q-band staining, there was no
clear difference in the regions except
satellite of No. 7 chromosome.

On an individual, which seemed
male externally and whose gonad was female
type, the nucleolus number was 1 or 2. And
one homologous sat-chromosome pair was in
each mitosis of the individual. This
individual is thought to be genetically
male.

Oe Sele,
Niigata
number of
Anura) is

(sat-chromosomes)
considered aS sex
the female sat-
heteromorphic as for

to be

GES

RESTRICTION SITE ANALYSIS OF RIBOSOMAL
DNA IN FIVE SPECIES OF VOLES LIVING IN

JAPAN 1 1 9
Heotiziksiean Se iakiuisad: K.Tsuchiya ,
S.Wakana -Moriwaki M.

Sakaizumni , -lmse>o ME@s SEalo5 Wns JHai)kea
Univ. School of Med., Tokyo, Miyazaki
Med. College, Central Inst. Exp. Ani-
mals, Natl. Inst. Genet., Mishima. The
Tokyo Metropol. Inst. Med. Sci.,

A survey of rDNA restriction-site
variation for five species of voles
living in Japan was carried out in order
to examine genetic relationship between
them. Analysis of the rDNA spacers of
Eothe: nomys smithi, E. andersoni, Cle-
hin HoOMOmyS aE O Camus) Greens) Walnid) CR
rutilus , using 10 different restriction
enzymes and cloned mouse rDNA probes,
revealed that they have a species-—
specific set of rDNA repeating units
(repetype). Based on the restriction-
Site data, sequence divergence among the
major repetypes was estimated. The
values for EK. smithi vs. E. andersoni,
GC. rufocanus vs. @. rex, EB. smithi vs.
C. rufocanus, and E. smithi vs. C.
rHeLliSs were 2.0%, 2.04, S584, ama 653%,
respectively. Phylogenetic relation-
ships based on restriction site data
indicates that these five species are
Gla ssa feveds amit opt hinieciesse nompsins Cl) En
smithi and E. andersona; (2) C. rufoca=
mus ancl Co mexs (GS wutedilue.

BUERGERIA

Genetics

1055

GE 10

Genetic differenciation of Japanese mole speceis, Mogera
K. Tsuchiya’ ,M.Sakaizumi , ?S.Wakana$ ,H.Suzuki*,& K.Moriwaki®
"Miyazaki Med.College,*The Tokyo Metropol. Inst. Med. Si.,
3Central Inst. Exp. Animals,*The Jikei Univ.School of Mad.,
° Natl. Inst. Genet.

We studied 33 individuals of Japanese mole speceis, Mogera
from Japan.
choramosomal karyotype, serum protein, mitochondrialDNA,
ribosomalDNA,and twelve biochemical markers.
results showed that Japanese mole speceis, Mogera could be
classified three different species groups.

Mogera tokudae group: The species group was distributed in
Sado Is.and the part of Niigata Pref.Genetic markers
between both areas ware a few different.

Mogera wogura : The species group was distributed in
Honsyu main Is. The variability at the choromosomal
karyotype, mitochondrialDNA,and ribosomalDNA levels were
shown in the species. suggesting that long time had passed
after distribution.

Mogera kobeae : The species group was widely distributed
in the west of Honsyu main Is., Shikoku Is.and Kyusyu Is.
A few polymorphisms of these markers levels were probably
the result of recent distribution in Japan from Korea.

Each individual was examined for

These

GEi lal

THE GENETIC DIVERGENCE OF THE NEW VARIANT
OF THE MOUSE £8 -GLOBIN GENE Hbb’, IN
NORTHWESTERN CHINA.

Te Kawashima’, N. Miyashita’, :
Shiroishi!, K. Moriwaki', S. Wang. ‘Natl.
ING, Oe, BEING. 5. WES, “Abains — IDNSES Cie
BiOils IPieoOals , Clarins.

We studied the allelic frequency of
the hemoglobin beta chain gene (Hbb) of
wild mice, Mus musculus, mostly in Asia.
In the process of the survey Hbb
polymorphism in main land China, we found
a new allele of Hbb, named Hbb*. It is
distributed in a rather wider area in
northwestern China. From the
electrophoretic pattern on cellulose

acetate membrane, Hbb* was seemed to be an
intragenic recombination between Hbb* and
Hbb®. Southern blot analysis of bl and b2
genes in Hbb*? and other Hbb haplotypes
were carried out with Hbb® DNA _ probes
using nine restriction enzyme. Based on
the difference iN eSivrEalce tones Sites),
sequence divergence was estimated. It was
revealed that the distance between ._Hbb®
and Hbb*® was 2.64% diversity, and the
distance between Hbb* and Hbb® was only
0.011% diversity.

1056

GE 12

Macrogeographic variation in mitochondrial DNA of
Japanese voles, Hothenomys .-

S.Wakana', H.Suzuki2,M.Sakaizumi?,K. Tsuchiya" ,&K.Moriwaki>
‘Central Inst. Exp. Animals,?The Jikei Univ.School of Med.,
The Tokyo Metropol. Inst. Mad. Sci.,*Miyazaki Mad. College,
5 Natl. Inst. Geret.

Mitochondrial DNA (mtDNA)from 36 Japanese
voles (Fothenomys)collected in 11 localitis was analyzed
for restriction fragment lenght polymorphism(RFLPs) with
eight restriction endonucleases. There was evidence of
extensive mtDNA sequence heterogeneity among vole
populations. Twelve mtDNA different haplotype were found
and four common mtDNA composite were shared among these

haplotype. Estimates of sequence divergence among these
population were high(0.12~0.233). Therefore, we
comparied the sequences which was 990bp of mtDNA amplified
by PCR in order to clear the relationship of the four
vole mtDNA types. These sequences include the D-loop
region,which was very hypervariable region ( ex.
insertion &deletion). We could not clear the relationships
of these four vole groups based on mtDNA D-loop sequences,

GE 13

ORIGIN OF THE TRIPLOID "“GINBUNA", CARASSIUS
AURATUS LANGSDORFT.

Y.Shimizu', T. Oshiro', F. Takashima', and M.
Sakaizumi=. 'Dept. Aqua. Biosci., Tokyo Univ. Fish.,
Tokyo. “Dept. Lab. Ani. Sci., Tokyo Metropol. Inst.
Med. Sci., Tokyo.

Evolutionary relationship between the triploid
Funa("Ginbuna") and other diploid Funa species
remains poorly understood. We examined the DNA
content of erythrocytes and several isozyme
patterns of Funa collected in Japan and Korea.

The results are as follows: 1. Three patterns of
AMY isozyme(A, AB and B) were observed. All
Japanese diploid Funa had the A, while Korean
Funa and Japaneses gold fish had the B. Majority
of the triploid Funa had the AB-pattern. 2. Most of
the triploid Funa in the Pacific coast of eastern
Japan had a characteristic band of the PGM
isozyme. The Kinbuna (Carassius auratus subsp.)
had also the same band. 3. The Funa in the Sea of
Japan coast of the Tohoku District, and the Funa
in the Pacific coast of eastern Japan had
characteristic patterns of the muscle _ LDH,
respectively.

The Heterozygous(AB) pattern of the AMY isozyme
suggested the hybrid origin of the triploid Funa.
The geographic distribution of three types of the
triploid Funa demonstrated that the diploid Funa
living in each area had participated to the
emergence of the triploid Funa living there.

GE 14

GENETIC STRUCTURE OF THE SOUTHERN POPULATION OF
THE MEDAKA ORYZIAS LATIPES .

M. Sakaizumi’ and K. Tsuchiya*. ‘Tokyo Metropol.
Inst. Med. Sci., Tokyo and “7Miyazaki Med.
Coll..Mlyazaki.

An extensive survey using various genetic
characters have revealed that there are four
major biochemical groups in Oryzias latipes. They
are named the Northern Population, the Southern
Population, the East Korean Population and the
China-West Korean Population, respectively.

The Southern Population, distributed in southern
Japan and the Pacific coast of eastern Japan, is
the most polymorphic and many variations of the
protein loci and mtDNA shows region-specific
distribution. Most of them are found in Kyushu
and its adjacent area. A limited range of
polymorphism is observed in eastern Japan
eastward from Kii Peninsula, suggesting that the
Southern Population produced many variations in
south-western Japan and recently expanded its
distribution range toward eastern Japan.

The polymorphism found in the Southern
Population can be classified into two categories;
the variations only found in the Southern
Population and that distributed at high frequency
in the China-West Korean Population. This fact
suggests that the Southern Population have
retained a large population size and that gene
exchange have occurred after separation of the
Southern Populatioon and the China-West Korean
Population.

GE 15

GEOGRAPHIC DISTRIBUTION OF SA1 ANTIGEN IN

Meee einen ale OF ORYZIAS LA IPES
Matsuzaki M. Sakaizum an A.

eis .Zool. Inst., Fac. of aa Univ. of

Tokyo, Tokyo, “The Tokyo Metropol. Inst. of

Med. Sci., Tokyo.

We produced a monoclonal antibody m11C-1
in the course of a survey of alloantigens in
inbred strains of the O. latipes. To study
geographic distribution of the corresponding
antigen, we examined 2 or 4 individuals from
each of 143 locations in Japan, Korea and
China. Cells from both head-kidneys and a
spleen of a Medaka were smeared on a slide
glass, then treated sequentially with m11C-
1, HRP conjugated anti-mouse Ig antibody and
3-amino-9-ethylcarbazole. No antigen-
positive fish was found in the Northern
Population, the East Korean Population and
the China-West Korean Population. On the
other hand, the majority of fish collected
from the Southern Population were antigen-
positive. Thus we named this antigen SA1
(Southern population-specific Antigen 1). In
the Southern Population, fish without the
SA1 antigen was found mainly in the region
from the Kii Channel to Oosumi Peninsula
through the Seto Inland Sea area. This
distribution pattern almost fits with that
of ACP© and also of LDH-A® allele, which
shows a close genetic relationship among SAl
negative fish in the Southern Population.
The SAl antigen was detected also in other 2
closely related species, O. luzonensis and
QO. curvinotus, suggesting that the origin of
the SA-1 antigen can be traced back to a
common ancestor of these 3 species.

GE 16

AN ATTEMPT TO CONSTRUCT A GENETIC MAP OF
MEDAKA, Oryzias latipes

Y.KURIHARA!, M. SAKAI

Y .HYODO-TAGUCHI Div.of Biol.,
Natl.Inst.Rad.Sci., Chiba, Div.of

Lab.Anim.Sci. The Tokyo Met.Inst.Med.Sci.,
Tokyo.

ZUMI2 AND

We developed a simple and efficient method
to construct a cytogenetic map of Medaka,
Oryzias latipes, by using two inbred
strains of the Medaka which were derived
from genetically distinct populations.
Fourty nine genomic DNA clones were
obtained from HO5 male, and each clone
was subjected to subsequent analysis.
RFLP between HNI and HO4C, derived from
northern and southern population
respectively, was examined in ten
different restriction endonucleases. The
length variations were observed in 10 out
of 11 DNA clones examined untill now,
showing these genomic DNA clones could be
used for gene mapping in the Medaka. RFLP
analysis shown by the genomic DNA clones
are powerful to construct a genetic map
in the Medaka.

GE,

MUTANTS OF LEUCOPHORES, VL-1,VL-2, VL-3 AND
MO, IN THE MEDAKA.

H.Tomita. Lab. of Freshwater Fish Stocks.
Fac. of Sci., Nagoya Univ., Nagoya.

Mutants of leucophores in the medaka are
divided to six groups. (1) Leucophores
increase in number and in size. (2) Leuco-
phores decrease in number. @)P Watsaiiolke
leucophores are absent. (4) Leucophores
are at dispersed condition. (5) Leuco-
phores distribute at variegated condition.
(6) Leucophores are white in color at em-
bryonic and larval stages(normal ones are
orange).

The group(2) is consisted of two types.
One type(a) is that a few leucophores
develop at embryonic and larval stages.
Adult fish also have a few leucophores.

The other type(b) is that leucophores de-
velop at embryonic and larval stages. Then
leucophores disappeare gradually at two
weeks after hatching. Most leucophores are
invisible. Adult fish have a few leuco-
phores.

The vl-1, vl-2 and vl-3 mutants belong to
the group 2(b). They do not distinguish
each other in the phenotype. They are re-
cessive and autosomal, and belong to dif-
feret alleles each other.

Leucophores of the mo mutant show the
same feature as those of the vl-l. vl-2
and vl-3 mutants. But the mo mutant has
pleiotropic action on melanophores (dilution
of black color). This mutant is recessive
and autosomal.

Genetics

1057

GE 18

SEX REVERSAL IN INTERSPECIFIC HYBRIDS
BETWEEN THE EAST KOREAN POPULATION OF
ORYZIAS LATIPES AND O. CURVINOTUS.

S. Hamaguchi and M. Sakaizumi“¢. Dept.
Bole Colt "Geni ssduc. Na wgeaital Uniaw.
Niigata, 2The Tokyo Metropol. Inst. Med.
Sci., Tokyo

_ Oryzias latipes includes 4 subspecies,

the northern Japanese, the southern

Japanese, the Chinese and the east Korean
populations (Sakaizumi 1986, 1987). We
produced interspecific hybrids between
OQ. curvinotus and each of 4 subspecies of
QO. latipes. Male interspecific hybrids
were sterile, but females laid diploid
eggs (Hamaguchi & Sakaizumi 1989;
Sakaizumi and Hamaguchi 1989). In most

these crossings, we could produced both
male and female hybrids. Only when female
O. latipes in the east Korean population
were mated with O. curvinotus, all offspr-
ing were female. We mated female Fl fish
with male O. curvinotus to examine the sex
ratio of their offspring. All eggs from
some Fl females developed into females,
whereas those from other Fl females into
either males or females, suggesting that
the former is XX female and the latter is
sex-reversed XY female. These XY females
could not be distinguished from XxX
females. These results indicate that the
testis-determining gene(s) on Y chromosome
of O. curvinotus cannot be expressed in
the presence of genes of O. latipes which
are specifically expressed in the fish of
the east Korean populations.

GE 19

CONCERNING CHICKEN-QUAIL INTERSPECIFIC HYBRIDS
FREQUENCY OF HYBRID FORMATION AND DEVELOPMENTAL METHODS
A.Nakamura,! K.Nishimura,? T.Kaneko,! 1 Univ.of Shizuoka,
Hamamatsu Col.,Hamamatsu. 2 Dept.of Biol. ,Hamamatsu Univ.
School of Med. , Hamamatsu.

In order to understand the genetic manifestation mech-
anisms of both quail and chickens, both species were exan-
ined for interspecific hybrid production. Quail and chicken
are of differnt genera, but it is known that they are able
to produce hybrids.But the probability of such a hybrid be-
ing produced is very low. We were able to obtain a hybrid
by artificially inseminating chiken semen into Japanese
quail (Coturnix coturnix japonica) hens.Fifteen hybrids(0.7
percent) hatched from 2,117 eggs set. For this research,
quail of white egg lineage was used, and observation of the
embryo development was quite easy.Many eggs had insemina-
tion interrupted, or died, but here the deformations were ob-
served. It was seen that these eggs could not hatch because
the positioning of the embryo was incorrect, the upper beaks
were underdeveloped, and the head positioning was incorrect.
They were distinct from quail especially in their adult
weight of 300-350g. (The adult weight of a quail is about
120-150g).During the course of their development to adults,
the individual hybrids displayed various phenotypes of
both parents. In general, at the time of birth coloration and
voice characteristics resembled on the basis of genetics of
the chicken. Their behavior were quite aggressive as with
wild fowl, unlike either parent.

In particular it was seen that,as apart from chickens,
the phenomenon of the chicken sex-linked barring gene
changed during the course of development.

1058 Biochemistry

BI 1

CHANGES IN PROTEIN PHOSPHORYLATION IN THE BRAIN OF
THE RAT AFTER IN VIVO METHYLMERCURY TREATMENT.
Harutaka YAGAME, Hiroshi SUGANO and Saburo OMATA.
Department of Biochemistry, Faculty of Science,
Niigata University, Niigata.

As the first step to clarify the action of MeHg
(methylmercury) on the process of protein modifi-
cation, we investigated effect of MeHg on endoge-
nous protein phosphorylation in the brain cytosol
fraction. Female Wistar rats were injected with
10 mg/kg/day of MeHg for 7 consecutive days and
were sacrificed on Days 4, 9 and 15. The total
protein phosphorylation activity in the brain
cytosol fraction was’ elevated during the MeHg
intoxication in the presence or absence of Ca** in
the reaction mixture. The phosphorylated protein
species were separated further by 2D-electro-
phoresis and protein spots localized by autoradio-
graphy were cut out and counted. Control proteins
phosphorylated in the presence of adenosine 5’-
Y -thiotriphosphate were added as an internal
standard. The results showed that the effect of
MeHg was not uniform for individual protein
species: phosphorylation of many protein species
was activated to different extents on Day 4,9, or
15, while that of other protein species was
reduced throughout the experimental period. The
patterns of the change differed in the presence or
absence of Ca*~.

BI 2

PARTIAL TRANSPORT OF CALCIUM IN RABBIT
SARCOPASMIC RETICULUM STARTS TO TAKE
PLACE IN THE ABSENCE OF ATP

Go Nakamura, BiGl, LaSit,, PACs Setoyg
Tohoku Univ. Sendai

CALCIUM Cie AMSOirce ACEOSS cle
sarcopasmic reticulum membrane vesicles
is mediated by membrane-bound Ca-ATPase
with ATP hydrolysis by the Ca-ATPase.
The ATPase also mediates passive calcium
transport across the membrane without
ATP. It is unknown whether calcium
passes through the same calcium channel
of the ATPase protein in both the active
and passive transport. However, our
stratege for understanding the mechanism
of active calcium transport is to find
elementary steps in the passive process,
and to examine how ATP acts on each of
the steps. Here, we studied exchange
reaction of calcium bound to the ATPase
E Ow Lice, UMlooune ealerum LOM Arte
equilibrium in the absence of ATP. An
"occluded" state of calcium in the ATPase
protein was observed: Half of the bound
calcium slowly exchanged at a rate about
20-30 times slower than the dissociation
rate of ti/222 Ss. The other half of the
calcium was, however, not occluded, and
rapidly exchanged at the same rate of the
dissociation rate. The calcium in both
state was equally dissociated by EGTA.
The ATPase protein itself seems to
partially translocate calcium in the
protein without ATP.

BI 3

ACTIVATING EFFECT OF LIGHT ON THE
RESPIRATION OF ECHIUROID GAMETES
IN THE PRESENCE OF CO.
E. Tazawal, A. Fujiwara? and I. Yasumasu2
1Biol. Inst., Yokohama City Univ.,
Yokohama, Dept. of Biol., Sch. of Educ.,
Waseda Univ., Tokyo.

It has been found that echiuroid eggs
undergo CO-insensitive respiration which
is activated by white light. In the
present study, effect of light irradiation
on respiration of echiuroid gametes was
examined at wavelengths between 350 and
650 nm. Light irradiation was made by the
Okazaki Large Spectrograph. Light irradi-
ation enhanced the respiratory rate in the
presence of CO in eggs and sperm. The
rate of photo-activated respiration in-
creased in relation to CO concentration.
Action spectra for Photo-activation of
respiration in echiuroid gametes, showing
peaks at 430, 530 and 570 nm, are very
alike in their profile to absorption
spectrum of reduced cytochrome b. The
respiration of sperm in the presence of CO
was inhibited by antimycin A and CN-.
Photo-activation of cytochrome b probably
becomes apparent as an increase in the
respiratory rate in sperm which may have
CO-insensitive cytochrome c oxidase. In
eggs, respiration was rather insensitive
to antimycin A. In echiuroid eggs, elec-
tron equivalent may not be transported
from cytochrome b to cytochrome c in the
presence of CO.

BI 4

PHOTO-ACTIVATION OF CO-INSENSITIVE
RESPIRATION IN SPERM OF SEVERAL
INVERTEBRATE SPECIES.
I. Yasumasu!, E. Tazawa2 and A. Fujiwara!
IDept. of Biol., Sch. of Educ., Waseda
Univ., Tokyo, Biol. Inst., Yokohama City
Univ., Yokohama.

The respiration was inhibited by CO in
sperm of the sea urchin, Anthocidaris,
crassispina, the starfish, Asterina pecti-
nifera, the tunicate, Ciona intestinaris
and the oyster, Crassostrea gigas. Among
these species, sperm of sea urchin and
starfish exhibited an increase of respira-
tory rate by light in the presence of CO.
Action spectra for photo-activation of
respiration in the presence of CO in sperm
of these species showed peaks of
activating effect of light at 430, 530 and
570 nm in the same manner to those
obtained in the gametes of echiuroid.
Under light irradiation: or in the dark,
the respiration of sperm was blocked by
antimycin A, CN- and oligomycin in the
presence and absence of CO. In sperm of
all examined species, cytochrome c oxidase
was blocked by CO and CO-induced blockage
was reversed by light. Strong activation
by light of "CO-cytochrome c reductase"
was found in sperm of sea urchin and
starfish. Photo-induced marked increase in
the rate of reaction by cytochrome c
reductase probably enhances’ respiratory
rate, when CO-blockage of cytochrome c
oxidase is reversed by light.

Biochemistry 1059

BI 5

THE ROLE OF PIGMENT GRANULES IN XANTHOMMATIN
BIOSYNTHESIS IN THE EPIDERMAL CELLS

OF THE SILKWORM

H. Sawada, M. Tsusue, K. Dohke

Biol. Lab. Kitasato Univ. Sagamihara.

In the previous presentation we reported that xanthommatin
localizes in the pigment granules binding with a protein in the
epidermal cells of the silkworm. By SDS-PAGE and GPC, the
molecular weight of the pigment protein was estimated to be
13 kDa.

In the present study we will report that 13 kDa protein sepa-
rated by SDS-PAGE has binding activity with a xanthom-
matin. Biosynthesis of xanthommatin was also investigated.
The homogenate of epidermal tissue was fractionated by su-
crose density gradient centrifugation. When the substrate 3-
hydoxykynurenin was add to each fraction, catalytic activity
to form xanthommatin localized in pigment granules frac-
tions. Therefore, the granules were partially purified and
catalytic activity to form xanthommatin was studied. The Km
value was 2.2 x 10°M and pH optimum of the reaction was 7.5
in potassium phosphate buffer. We suppose that phenoxaz-
inone synthase, the permiability and/or catalytic activity of
the granule membrane and properties of the pigment protein

itself may be involved in xanthommatin biosynthesis.

BI 6

ACTIVATION OF LACCASE-TYPE PROPHENOLOXIDASE
IN THE CUTICLE OF INSECT. VII.HARDENING AND
DARKENING MECHANISM OF LARVAL CUTICLE IN
SILKWORM, BOMBYX MORI.

H. I. Yamazaki. ,Biol.,Lab.,Atomi Gakuen
Women’s Univ. , Saitama.

After larval-pupal ecdysis, newly formed
cuticle does rapidly harden and darken. The
laccase,a type of phenoloxidase is found in
cuticular matrix and supposed that the
enzyme mediates the hardening and darkening
process. And the laccase itself is involved
in the newly tanned matrix. The laccase
activity is already found in newly formed
cuticle as inactive proform bound with mat-
rix and it is able to isolate and activate
by proteinase. The origin of the prolaccase
was investigated by immuno-blotting method
using the antiserum raised from purified
active laccase. Peptides that is reactive
with antilaccase antiserum were detected in
the pre-cuticle. The main component of the
peptide group was the one of 150 KDa in its
molecular size.

The peptide dynamics were investigated in
larval-larval ecdysis between 4th and 5th
istar. The 150 KDa peptide was also detected
as same as the case of larval-pupal ecdysis.
And prolaccase molecule was released by @&-
chymotrypsin, or active laccase was found to
be detectable by the action of trypsin.

The dynamics of peptide concerning with
laccase activation is supposed very similar
between larval-pupal and larval-larval
ecdysis.

BI 7

AN ATTEMPT TO RECONSTITUTE PROPHENOLOXIDASE
CASCADE OF INSECT.

Y. Katsumi and M. Ashida

Biochemical. Lab., The Institute of Low
Temp. Science, Hokkaido Univ., Sapporo.

Plasma fraction of hemolymph obtained
from larvae of the silkworm, Bombyx mori
(5th instar, day 7 to 8) was fractionated
sequentially by ultracentrifugation,
ammonium sulfate and column chromatography
on Sephadex G-150, Phenyl-Toyopearl and
Heparin-Toyopearl. In the last step of
fractionation four fractions, A, B, C and D
were obtained. Fraction A and fraction D
were shown to contain proBAEEase (a
precursor of serine protease which is
claimed to be a component of prophenol-
oxidase cascade) and @-1,3-glucan
recognition protein, respectively. When @-
1,3-glucan and homogeneous 6-1,3-glucan
recognition protein purified according to
Ochiai and Ashida (J. Biol. Chem. 263,
12056-12062 (1988)) were added to fraction
A, no BAEEase activity appeared in the
mixture. However,in the above mixture
supplemented with fraction C, BAEEase
activity appeared.

These results indicate that a component(s)
other than 6-1,3-glucan recognition protein
is involved to activate proBAEEase by @-
1,3-glucan in the insect prophenoloxidase
cascade and that fraction A and fraction C
can be used as starting materials to
isolate component(s) required for
activating proBAEEase by @-1,3-glucan.

BI 8

IMMUNO-CYTOCHEMICAL LOCALIZATION OF
PEPTIDOGLYCAN RECOGNITION PROTEIN (PGRP) IN
THE SILKWORM, Bombyx mori.

K. Kinoshita, P. T. Brey and M. Ashida.

The Institute of low temp. Sci., Hokkaido
Univ., Sapporo.

PGRP (a 19kDa protein) specifically binds
to peptidoglycan, a cell wall component of
bacteria. The lal fol@lab jo\ey ae PGRP to
peptidoglycan has been shown to trigger the
prophenoloxidase cascade which is now
considered to be a intergral part of insect
defence mechanisms.

A monospecific polyclonal antibody
against PGRP (anti-PGRP/IgG) was prepared
to investigate the distribution and the
localization of PGRP immunocytochemically.
Silkworm plasma, hemocyte homogenates
and integument were shown respectively to
have a 19kDa polypeptide cross-reactive to
the anti-PGRP/IgG. Ultra thin sections of
the tissues obtained from day 5 5th instar
silkworms were stained by indirect
immunogold labeling method using the anti-
PGRP/IgG as a primary antibody and examined
under an electron microscope. The presence
of PGRP was demonstrated in granules of
granulocyte, spherules of spherulocyte,
procuticle of the trachea cuticle and body
wall integument. In the sections of
integument, the highest labeling density of
gold colloid was observed at the interface
between procuticle and epidermis. The above
observation was discussed in relation to
the function of PGRP recognizing bacteria
as non-self in hemolymph and cuticle.

1060 Biochemistry

BI9

ACTIVATION OF PROPHENOLOXIDASE IN DROSO-
PHILA. I. POSSIBLE INVOLVEMENT OF PRO-
TEASE (S).

M. Yonemura, K. KasSatani, N. Asada and

ih, Olanaleina, IRE, Wel), , WAS, O12 Hews, OKei—
yama University of Science, Ofayama.

Since the discovery of activation of
prophenoloxidase(ProPO) in Drosophila, ex-
tensive works have been done on the mecha-
nism of the activation. Mitchell and his
coworkers proposed a cascade system culmi-
nating the ProPO activation. However,
details of the activation reaction remain-
ed unsolved. In order to know the invol-
vement of protease(s) in the activation
process, we first studied on the protease
activity in extracts of 3rd instar larvae.
By use of synthetic substrates for trypsin
such as benzoyl-L-arginine ethyl ester and
benzoyl1-DL-arginine-p-nitroanilide and of
specific inhibitors such as p-nitro-
guanidinobenzoate(p-NPGB) and p-amidino-
phenylmethanesulfonylfluoride(p-APMSF) , we
were able to show the serin-protease acti-
vity of trypsin-type in the extracts.
Since the activation of ProPO was inhibit-
ed by p-NPGB and p-APMSF at approx. the
same concentrations as those inhibiting
the protease activity, it is highly pro-
bable that the proteases is involved in
the activation of ProPO in Drosophila.

BI 10

ACTIVATION OF PRO-PHENOLOXIDASE IN DROSO-
PHILA. II. ACTIVATION BY BOMBYX ACTIVATOR
AND &-CHYMOTRYPSIN.

N. Asada, K. Fujimoto and E. Ohnishi. Biol.
Lab., Fac. of Sci., Okayama University of
Science, Okayama.

Phenoloxidase in Drosophila occurs as an
inactive pro-phenoloxidase (ProPO), which
is converted to active enzyme by complex
cascade reactions. Nature of the activa-
tion process remained obscure. Partial
purification of ProPO was achieved from
pupal extract of Drosophila. Nearly 110-
fold purification of this enzyme was at-
tained by ammonium sulfate (AS) fractiona-
tion, Sephacryl (S) and DEAE-cellulose
chromatography. The S-fraction could be
activated by 35-55 % AS-fraction (AMM-1)
prepared from pupae. ProPO-activating
enzyme from cuticles of Bombyx and o-chymo-
trypsin also activated ProPO in S-fraction
of Drosophila. The level of final activity
by both activating agents, however, was
comparatively low, and the activation by
d-chymotrypsin was inhibited by TPCK. Try-
psin, Arg-C and heated AMM-1 did not acti-
vate the ProPO. Experiments of the acti-
vation by o&-chymotrypsin in combination
with AMM-1 revealed that the activated pro-
duct had inefficient PO activity. These
results suggested that the protease invol-
ved in the activation is highly specific,
and it could not be substituted efficiently
by other proteases including Bombyx cuti-
cular activating enzyme. ee

BI 11

PREPARATION AND PROPERTIES OF HIGHLY SPECIFIC
MONOCLONAL ANTIBODY AGAINST SEPIAPTERIN REDUC-
TASE FROM FAT BODY OF SILKWORM, BOMBYX MORI.

T.IINO!, H.SAWADA?, K.DOHKE? AND M.TUSUE?.
‘DEPT. OF GEN. EDUC., NIHON UNIV., TOKYO.
?BIOL. LAB., KITASATO UNIV., SAGAMIHARA.

Sepiapterin reductase (SPR) from fat body of silkworm is quite
similar to that from mammalian sources. However, we have
reported that immunological study showed no apparent homol-
ogy between the two enzymes by use of monoclonal antibody
against SPR from fat body of the silkworm. Unfortunately, the
specificity of the antibody used in the previous work was low.
Recently, we obtained highly specific monoclonal antibody
against 27 KDa subunits of SPR from fat body of wild type silk-
worm. This antibody has strongly and specifically recognized
27 KDa protein in the crude extract from fat body after SDS-
treatment, while it did not recognize SPR from rat erythrocytes
at all. On the other hand, this antibody recognized approx. 80
KDa protein instead of 27 KDa protein in the crude extract from
fat body of mutant lemon of the silkworm after SDS-treatment,
whereas the enzyme activity of SPR was not found in the crude
extract from fat body of the mutant lemon. The result indi-
cates the existence of an inactive protein in the mutant lemon,
which is larger molecular weight and immunologically ho-
mologous with SPR from wild type of the silkworm. Charac-
terization of the inactive protein in the mutant lemon must
await further studies.

BI 12

DYNAMICS OF SEPIAPTERIN REDUCTASE REACTION
IN THE CLOSED SYSTEM ANALYZED BY COMPUTER
S.Katoh', T.Sueoka'! and H.Hikita’?. ‘'/Dept.of
Biochem., Meikai Univ. School of Dentistry,
Sakado, Saitama. *Lab.of Physics, Meikai
Univ., Urayasu, Chiba, Japan

At. least two enzymes, 1.e€. Sepiapterin
reductase (SPR) and 6-lactoyl tetrahydro-
pterin synthase are known to catalyze the
reduction with NADPH the key intermediate 6-
pyruvoyl tetrahydropterin (PPH,) in the bio-
synthesis of tetrahydrobiopterin (BH,) which
is an essential cofactor of aromatic amino
acid hydroxylases. The physiological rela-
tion between the two enzyme reactions are
not clear. We have investigated the kinetic
scheme for SPR in the closed system by
computer fitting in order to elucidate the
reaction mechanism of this enzyme. When PPH,
(A), a diketo compound, was incubated with
SPR and NADPH at pH 3.8, two types of mono-
keto intermediates {C1'-keto type(B) and C2'-
keto type (C)} and final product BH, (D), a
dihydrodiol compound, were observed. The
experimental data were analyzed by approxi-
mating our system with computer program to
the compartmental model for [A], [B], [C] and
[D] which represent the concentration of
each reactant in the reaction mixture. The
results from simulation analyses of data
were non-linear least-squares fitted using a
numerical iterative procedure. Then the time
course curves were plotted and the best-fit
parameters of rate constants (k, - k,o) were
obtained. B and C may be the succesive inter-
mediates, and the steps A~B—-C—D may occur
with SPR in the closed system of reaction,
while 6-lactoyl tetrahydropterin synthase
only catalyzes the step A>B.

Biochemistry 1061

BI 13

CHARACTERIZATION OF PIG INTESTINAL PROLYL AMINOPEPTIDASE
AS LEUCYL AMINOPEPTIDASE

T.Takahashi', K.Takahashi', M.Matsushima®, M.Ichinose?,
K.Miki°. ‘Dept. of Biophys. and Biochem. Fac. of Sci.
“Ist Dept. of Internal Medicine, Fac. of Medicine, Univ.
of Tokyo.

We purified prolyl aminopeptidase to apparent homoge-
neity from pig intestine and human liver using L-Pro-
Leu-Gly-NH. as a specific substrate. The purified enzymes
were metalloenzymes which appeared to consist of six
identical subunit polypeptides with a molecular weight of
about 55,000 each, associated non-covalently. The kinetic
study of the pig enzyme with various substrates gave the
spcificity very similar to that of leucyl aminopeptidase.
Double immunodiffusion test using antiserum raised
against pig intestinal prolyl aminopeptidase showed that
the intestinal enzyme and pig kidney leucy] aminopeptid-
ase were indistinguishable. The amino terminal sequence
of the purified pig intestinal enzyme was the same as
that of the leucyl aminopeptidase. From the results, we
concluded that prolyl aminopeptidase and leucyl amino-
peptidase are the same protein.

BI 14

LOCALIZATION OF CATHEPSIN E IN RAT LIVER.
S.Yonezawa. Dept. of Zoology, Fac. of
Sei s ; Hokkaido Univ., Sapporo.

To strictly define the site of localiza-
tion of cathepsin E in the liver, we pre-
pared three different cell fractions from
rat liver; hepatic parenchymal cells (HPC),
Kupffer cell(KC)-rich and endothelial cell
(EC) -rich fractions. Hepatocytes were iso-
lated by a perfusion-digestion method using
0.05% collagenase-KRB buffer at 37°C. Non-
parenchymal cells, which had been prepared
by a modified pronase-digestion method,were
suspended in RPMI 1640-10% FCS and incubated
ae 37°C oie Lobine OQ eullloyy aclnesw@m Oi Ke ico
dishes. Non-adherent cells(EC-rich fraction)
were collected by centrifugation. More than
90% of adherent cells were Kupffer-like cells,
as judged from examination by SEM and from
phagocytotic activity toward serum-treated
zymosan. Electrophoretic analysis revealed
that EC-rich fraction, but not HPC or KC-
rich fraction, contains cathepsin E activity.
In an attempt to examine the role of cathep-
sin E in EC, we tested a possibility of pro-
duction by cathepsin E of endothelin, a va-
soconstricting peptide hormone, from its
precursor big-endothelin. For this experi-
ment, cathepsin E-digests of big-endothelin-
1 were subjected to HPLC, followed by iden-
tification of separated peptides based on
their amino acid composition. The results
clearly showed that cathepsin E produces
endothelin-1l from big-endothelin-l in a
time-dependent manner, suggesting its parti-
cipation in the endothelin processing in EC.

BI 15

LEAKAGE OF ARYLSULFATASE FROM LYSOSOMES BY
TIN-PROTOPORPHYRIN PLUS PHOTOIRRADIATION.
H. Keino!, S. Mimura2 and S. Kashiwamata!.
lDept. Perinatol., Inst. Dev. Res., Kasugai;
2Dept. Pediat., Nagoya Univ., School of Med.
Nagoya,

The in-vivo administration of a syn-
thetic heme analogue (SnPP) is known to
cause serious responses to photoirradi-
ation in suckling rats. In this study, we
report the effect of SnPP plus photoirra-
diation on the leakage of arylsulfatase
(ASase) from lysosomes. The crude mito-
chondrial fraction from the cerebellum of
18 23-day-old rats was used as a source of
lysosome. Lysosomes were incubated with
SnPP under photoirradiation (blue-white
light) or dark. Immediately after photo-
irradiation the reaction mixture was
centrifuged to exclude intact lysosomes,
and ASase activity in the supernatant was
quantified. ASase activity was increased
with increasing concentratons of SnPP
(0-60 uM). No ASase was leaked from
lysosomes under the dark. The higher the
irradiation energy was (range, 45-7 8W/m2),
the more ASase activity was detected.
The leakage of ASase from lysosomes was
prevented by L-ascorbic acid, one of
radical scavengers. It is known that SnPP
has a long-lived triplet state that is
quenched by molecular oxygen. This
quenching results in the formation of
singlet state oxygen, which may cause the
destruction of lysosomal membranes
eventrally leading to tissue damages.

BI 16

JUVENILE HORMONE ESTERASE ACTIVITY REPRES-—
SIVE PEPTIDE IN THE PARASITIZED ARMYWORM
HEMOLYMPH.

Y.Hayakawa. Biochem Lab.,Inst. of Low Temp.
Sci., Hokkaido Univ., Sapporo.

Last instar larvae of the armyworm,
Pseudaletia separata, parasitized with the
parasitoid wasp, Apantales kariyai, do not
undergo metamorphosis and, ultimately, the
wasp larvae emerge from the host larvae
about 10 days after parasitization. It is
necessary for the parasitoid wasp to per-
turb the armyworm's endocrinological proc-
esses that control normal metamorphosis from
larvae to pupae. This endocrinological dis-
turbance allows the parasitoid to complete
its larval growth before emerging from the
host larvae. The growth-blocking peptide
with repressive activity against JH ester-
ase activity has been proven to exist in
the parasitized host larval plasma. The
detailed structure of this peptide was de-
termined as follows:

10 15
H-Glu-Asn-Phe-Ser-Gly-Gly—Cy s-Val-Ala- Gly— Tyr—-Met-Arg- Thr—-Pro—

s —— $

20 25
Asp-Gly-Arg-Cys-Ly s-Pro-Thr-Phe-Tyr-GIn-OH
Furthermore, hybridization of the peptide
specific DNA probes with the ovarian viral
DNA of the parasitoid wasp was detected.
This observation demonstrates that the gene
encoding this peptide is contained in the
DNA of an Ovarian calyx virus injected into
the host hemocoel during oviposition.

1062 Biochemistry

BI 17

PURIFICATION OF SOLUBLE-TYPE OF VITELLIN
BINDING PROTEIN FROM LOCUSTA MIGRATORIA
K. Yamasaki,

Dept. Biol. Tokyo Metropol. Univ. Tokyo

Vitellin binding protein(VBP) is involved
in the uptake mechanism of vitellogenin
into oocytes. VBP locallizes mainly on the
surface area of oocyte during vitellogene-
sis and is found in membrane fraction. The
VBP is insoluble in aqueous solution. Solu-
ble type of VBP found in vitellin frction
was partially purified and characterized.
And its role was thoght to be vitellin
condenser holding high concentration of
vitellin in mature oocytes. Further purif-
ication of s-VBP was carried out and could
be purified over 200 times compared with
starting material. A simple pulse-field
electrophoresis was employed for analysis
of s-VBP and its vitelline binding

Components of the purified s-VBP were
over gel (OG),a,b,c and front (Fb). Each
fraction was prepared by native PAGE
according with Davis system. Specific
binding for vitellin was clearly shown on
the PAGE analysis and the specific binding
was found to be completly lost after elec-—
tro washing. The inactivated s-VBP(OG) was
found to be active for binding with a,b,c,
and Fb fractions and reactivated by con-
jugation with Fb fraction.

In the first phase of saturation process,
it was found that vitellin molecule re-
leased quantitatively its front fraction
(Fv).

BI 18

EFFECT OF Vn PHOSPHORYLATION ON THE SUSCEPTIBILITY
AGAINST A CYSTEINE PROTEINASE IN THE SILKMOTH EGGS.
S.Y.Takahashi} X.Zhao! and T. Kageyamat 1, Fac.of
Liberal Arts, Yamaguchi Univ. 2, Dept.Biochem. ,
Primate Res. Inst. ,Kyoto Univ.

Silkmoth vitellin (Vn) is an excellent substrate
for G-kinase and the kinase is detected only in
early developing eggs but not in @iapause eggs, Su-
ggesting that G-kinase may be operative and invol-
ved in the phosphorylation of Vn in the developing
eggs. In order to confirm this , 32P-o-phosphoric
acid was injected into the eggs and the phosphory-
lated peptides were analyzed by autoradiography.
From the results, phosphorylation of Vn was obvious
only in the developing eggs, again indicating the
involvement of the kinase in the phosphorylation
of Vn.

A cysteine proteinase purified from silkmoth egg
degrades Vn effectively at pH 3.5 (optimal ph),
whereas,at higher pH (5.0), Vn was resitant to the
hydrolysis and it was not hydrolyzed even tnough
incubated for 24 hr with the proteinase. However,
if, Vn had been phosphorylated by G-kinase prior
to exposure to the proteinase, the proteolytic hy-
drolysis of Vn occurred readily at pH 5.0.

The facts suggest that the phosphorylated form
of Vn is much more subject to hydrolysis by the
proteinase than its non-phosphorylated form and
the G-kinase affected on the susceptibility against
the cysteine proteinase in the eggs.

From these results, we may conclude that phos-
phorylation of Vn,by G-kinase, seems likely to be
the trigger of the cysteine proteinase digestion
during the egg development.

BI 19

PURIFICATION AND COMPARISON OF PROPERTIES
OF VITELLINE COAT-LYSINS OF THE GENUS
TEGULA.

K.Haino-Fukushima and M.Shimoirisa, Tokyo
Metropolitan Univ., Setagaya-ku, Tokyo 158.

The vitelline coat—lysin (VCL) “of “a
marine Molluscan genus, Tegula is released
from acrosomal vesicles of sperm during
acrosome reaction and can lyse the vite-
lline coats (VC) of the same species. As
lysin action is extreamely species-specific,
the fertilization of an oocyte by sperm of
a different species seems to be prevented
by a barrier of VC surrounding the oocyte.
The VCL of T.pfeifferi was purified and its
primary and presumed secondary structures
determined in our previous studies.

In the present study, we tried to purify
two VCLs of Tegula, T.rustica and T.lisch-
kei, and compare properties of these lysins
such molecular weight, isoelectric point
and their ability to react with anti-
T.pfeifferi VCL antigens. The results
showed that the two VCLs had very similar
properties to those of T.pfeifferi VCL,such
as molecular weights of 15,000, isoelectric
points of 10.5, positive reactivity to
anti-T.pfeifferi VCL antiserum and similar
amino acid compositions. Moreover, pre-
liminary data of amino acid sequencing of
the two VCLs showed that homologous
sequences to T.pfeifferi were in both a
third of the N-terminus and the C-terminus
with an unhomologous sequence in the
central remainder.

BI 20

CLONING OF mRNA SEQUENCE FOR THE
EGG-MEMBRANE LYSIN OF TEGULA PFEIFFERI
H. Nakano, K. Haino-Fukushima, S. Izumi
and S. Tomino

Dept. Biol. Tokyo Metropol. Univ., Tokyo

A cDNA expression library was const-
ructed from the testicular poly(A)RNA of
Tegula pfeifferi, and a clone bearing the
mRNA sequence for egg-membrane lysin was
isolated by screening the library with an
antibody probe. The cDNA contained an open
reading frame encoding 163 amino acid
residues. Alignment of the deduced amino
acid sequence with primary structure of
lysin indicated the occurrence of 22
residues of signal sequence at the amino
terminal region of nascent peptide. A
consensus polyadenylation signal AATAAA was
present in the mRNA 92 nt downstream of
termination codon TAG. The deduced amino
acid sequences at the amino- and carboxyl-
proximal domains are virtually identical to
those determined for lysin by chemical
method. However, the deduced sequence at
the central domain (position 50 to 110) is
entirely different from the peptide
sequence; the former is larger by 22
residues.

Northern blot analysis demonstrated
that the lysin mRNA is present in the
mature testis but not in the ovary of T.
pfeifferi. The 650 nt mRNA highly homo-
logous to the T. pfeifferi lysin mRNA was
also detected in the testis of T. rustica.

Biochemistry 1063

BI 21

SECONDARY STRUCTURE OF EXOGASTRULA-INDUCING

PEPTIDE D (EGIP-D) PURIFIED FROM EMBRYOS OF
THE SEA URCHIN, ANTHOCIDARIS CRASSISPINA.
T.Suyemitsu and K.Ishihara. Dept. of Regu-
lation Biology, Fac. of Science, Saitama
Univ., Urawa.

The positions of the disulfide bonds in

exogastrula-inducing peptide D purified
from embryos of the sea urchin, Anthoci-
daris crassispina, were determined. EGIP-D
was digested sequentially with acid pro-
tease, thermolysin and aspartyl endopepti-
dase. The peptides containing disulfide
bonds were separated by RP-HPLC and detec-
ted by the SBD-F method. The three cystine-
containing peptides purified in this way
were each oxidized with performic acid and
further separated by RP-HPLC, yielding two
peaks in each case. From the analyses of
the amino acid composition and the sequen-
ces of the cysteic acid peptides, the posi-
tions of the disulfide bonds in EGIP-D were
identified as Cys(6)-Cys(19), Cys(13)-Cys
(33) and Cys(35)-Cys(48). The positions of
the disulfide bonds in EGIP-D are identical
to those in epidermal growth factor (EGF),
suggesting that EGIP-D is the EGF homolog
in the sea urchin.

BI 22

IDENTIFICATION OF A NOVEL AMINO ACID,
L-ORTHO-BROMOPHENYLALANINE, IN SPERM-
ACTIVATING PEPTIDES FROM THE EGG JELLY

OF THE SEA URCHIN Tripneustes gratilla

ke Yoshino!, M. Yamaguchi'l, N. Suzukil,

ala =r,nnYeaSimontsshas aeMeSuhatzasmand
WEG Ikalieelals « 1; Noto Marine Lab., Kanazawa
Wms, Ueldsibiea, BQ IMnSe5 iEic IMcOKESiiN IRSA ;
Osaka Univ, Suita, 3; Dept. of Chem., Fac.
of Sci. Kanazawa Univ., Kanazawa.

In this study, we isolated SAP-I deriv-
atives which contain a modified amino acid
residue from the egg jelly of the sea ur-
chin Tripneustes gratilla. The accurate
mass measurement of the peptides by fast
atom bombardment mass spectrometry (FAB-
MS) indicated that the modified amino acid
was bromophenylalanine (Br-Phe) which has
not reported as natural products. In order
to verify the presence of Br-Phe in the
peptides and to determine its isometric
structure (o-, m- or p-), we carried out
400 MHz !H-NMR spectroscopy, amino acid
analysis and RP-HPLC. As a result, we
concluded that the amino acid was o-Br-
Phe. To determine its optical isomeric
structure we analyzed o-Br-Phe derivatized
with FDAA by RP-HPLC. The results indi-
cated that o-Br-Phe in the peptides was L-
type. Furthermore, FAB-MS and amino acid
analysis suggested that some SAPs contain
m- and p-Br-Phe. The primary structures of
the Br-Phe-containing SAPs were determined
by B/E linked scan analysis in FAB-MS.

BI 23

THE STRUCTURES OF TWO NOVEL CALCIUM-
BINDING PROTEINS (TCBP-25, TCBP-23) IN
TETRAHYMENA.

1. Takemasa!, T. Kobayashi2, T. Takagi2, K. Konishi2, and Y.

Watanabel.
1 Inst. Biol. Sci. Univ. of Tsukuba, Ibaraki. 2 Biol. Inst., Fac. of
Sci., Univ. of Tohoku, Sendai.

In the course of studying the molecular basis of
calcium regulation in Tetrahymena, we became
aware of the existence of three kinds of calcium-
binding proteins (CaBPs). In order to ascertain their
primary structures, we cloned and analysed the
cDNAs. One is calmodulin, and the others are
proteins which cannot be found in any database. We
named them TCBP-25 and TCBP-23 (TCBP stands for
Tetrahymena Calcium-Binding Protein, and 25 or 23
indicates the relative molecular mass which was
calculated from the primary structure). Although
they are four domain EF-hand type CaBPs, they have
the common feature that the distance between
domain II and III is unusualy long for calmodulin
family proteins. To learn the relationship between
them, we studied the homology of the intra- or
intermoleculer domains. As a result, it is suggested
that they have evolved from the same ancestral
calcium-binding domain gene through repeated
duplication.

BI 24

MOLECULAR STRUCTURE OF CYANOPROTEINS OF
THE BEAN BUG, RIPTORTUS CLAVATUS
K.Miura!,y.Chinzeil T. Shinoda” E.Na 3
: AD 'GA : ; ; .Nagao
and H. Numata?. Ipept. Med. Zool., School
of Medicine, Mie Uniyv., Tsu. Natl. Res.
Inst. Veg., Age-gun. “Biochem. Lab., Inst.
Low Temp.Sci., Hokkaido Univ., Sapporo.
4Dept. Biol., Fac. Sci., Osaka City Univ.,
Osaka.

In the hemolymph of the bean bug there
are 4 types of biliverdine-associated
cyanoproteins (GP=1 25854) which are
separated by native PAGE. In the egg only
one type of CP (CP,,,), shown to be iden-
GILCAN CWO C2=i1, 1S demonstrated. By using
chemical cross-linking with DMS these CPs
were revealed to have hexameric structures
consisting of subunits with identical M.W.
of 76 kD. IEF analysis under denaturating
conditions showed that CPag was composed
of only a-subunit and CP-4 Sly b-subunit.
Subunit b had more acidic pI than subunit
a. 2-dimensional peptide mapping with
125;_jabelled CPs revealed that CP-2 and
CP-3 were hybrid molecules of a- and b-
subunits. From these results the molecu-
lar structures of CP were proposed as
follows: CP-1 (CPagg)=ag; CP-2=a,bo; CP-
3=aob,; CP-4=b¢. Electron microscopy,
immunological properties, spot-scanning of
2-D map, and native pIs of CPs_ support
these models.

1064 Biochemistry

BI 25

APLYSIA MYOGLOBIN: IDENTIFICATION OF TWO
KINDS OF CARBOXYL GROUPS INVOLVED IN ITS
AUTOXIDATION REACTION

A. Matsuoka and K. Shikama. Biol. Inst.,
Tohoku Univ. Sendai

Unlike mammalian myoglobins, Aplysia
myoglobin contains only a single
histidine residue that corresponds to the
heme-binding proximal one, consequently
lacking the usual distal one. When
compared with sperm whale oxymyoglobin
( MbO> ) as a reference, Aplysia MbO, is
extremely susceptible to autoxidation,
and its pH dependence is also unusual.
From the kinetic and thermodynamic
studies, we have revealed that two kinds
of carboxyl groups, AH with pk,=4.3 and
BH with pK5=6.1 at 25 °C, are responsible
for the autoxidation reaction of Aplysia
myoglobin.

Recently we have found an unusual
spectral increase at the Soret-peak of
Aplysia metMb in the acidic pH range, and
deduced two kinds of carboxyl groups with
pK,=4.2 and pK5j=5.7 from its titration
behavior. When Aplysia metMb was
substituted with dimethylprotohemin, on
the other hand, it shows only a single
dissociation process with pK = 4.2. From
these results, we conclude that the
carboxyl group AH comes from the globin
moiety and the other one BH corresponds
to one of the propionic acid side chains
of the heme.

BI 26

INTER-SYNGENIC AND INTER-STOCK VARIATIONS OF THE
ACIDIC AND BASIC HEMOGLOBIN COMPORNENTS FROM
PARAMECIUM CAUDATUM
er Haran, ils Ochiai sand oles Usuikae 'Depart.of
iOS 5) WECS wii Seoitcemel emacie, OF Piolo Collcor

Gen Educ., Niigata Univ., Niigata
Pe a ee ee aE ee, AS Ry ee
Paramecium caudatum contains a major hemoglobin
compornent (Hb;,)), which occupies 73-90 Z of the
whole Hb. The Hb;9 shows the same mobility on PAGE
and SDS-PAGE, irrespective of the stock and syngen
examined. However, an IEF analysis showed that the
Hbj 0 from some stocks is resolvable into two,
although the divided two fall into a narrow range
of pI showing 3.88 + 0.06. The presence of two
Hb;9 types (A and B) was verified more clearly by
an analysis using Mono P column for HPLC, and we
found that each of the stocks examined takes one
type among A, Band A + 8B. On the other hand,
distribution of the Hbj) type did not show direct
correlation to a difference of the syngen,locality
as well as the mating type of the stocks examined.
Basic Hb compornents (bHbs) are ammounted to 6-
13 74 sin P. caudatum, and by IEF they give one or
two bands corresponding to 9.8-10.5 in pI, accord-
ing to the stock examined. bHbs were resolved at
least to four types through HPLC analysis using a
Cosmoge1l—CM column. Distribution of these types,
however was not syngen-dependent as seen in Hbj9o.

BI 27

LAMPRAY SERUM ALBUMIN
Y. Yazawa, Dept. of Nutritional Physiol., Hokkido Univ.
of Education at Asahikawa, Asahikawa, Hokkaido.

Lamprey albumin was isolated and characterized its
molecular properties. The molecular weight was deter-
mined to be 170K from SDS-PAGE. It was composed of
about 1500 amino acids and N-terminal amino acid was
Asp. Rabbit antiserum to bovine serum albumin or carp
albumin cross-reacted with lamprey albumin by immuno-
blotting. Three components were obtained by tryptic
digestion of lamprey albumin and each molecular weig-
ht was 95K, 47K, and 30K. Only 95K component cross-
reacted with rabbit antiserun of carp albumin and it
was purified with Sephadex G-200 gel filtration and
DEAE-Toyopear! column. 95K component was composed of
about 840 amino acids and the star diagram was simil-
ar to that of lamprey albumin. It was compared with
the molecular properties of human, bovine, carp, and
lamprey albumin.

BI 28

EFFECT OF HGDGF ON THE GROWTH OF CORNEA
STROMAL CELLS.

Y.Yokoyama!, K.Kano!, K.Kaji2, K.shirama3;,
Y.Seyama!. 1 Dept. of Physiol. Chem. and
Nutrition, Fac. of Med, Univ. of Tokyo, 2
Tokyo Metropolitan. Inst. of Gerontology,
3 Dept. of Anatomy, Tokyo Medical College.

Harderian gland derived growth factor
(HGDGF) stimulated the growth of several
cell lines, such as BHK, SC-1 and BALB/3T3
but did not have any effect on the growth
of EL-4, Vero or KB cells. To investigate
the physiological role of HGDGF, the
effects of HGDGF and/or various growth
factors on the growth of cultured guinea
pig cornea stromal cells were examined.
HGDGF increased incorporation of 3H-
thymidine by 20-50 % compared to the
control (5% FCS). The combination of HGDGF
with FGF or PDGF increased the growth over
that of either growth factor alone,
increasing the incorporation of 3H-
thymidine by 180 and 190%, compared to the
control, respectively. The combination of
HGDGF with TGFB decreased the growth
stimulating activity to 60% of control and
EGF had no effect on the activity of HGDGF.
The growth stimulating activity of HGDGF
was inhibited by suramin in the different
manner from that of FGF, indicating the
presence of a specific HGDGF receptor.
These facts suggest that HGDGF binds the
specific cell-surface receptors and may
play a role in repairing the injured parts
of the cornea.

Biochemistry

BI 29

ROLE OF SATELLITE CELL ON THE COMPENSATORY
HYPERTROPHY IN RAT SKELETAL MUSCLE
S.Yamada',T.Kitajima,.Y.Atomi’ .N.Sakai’ M.Fujimaki’
‘University of Tokyo, Fuji Rebio Research Labolatories, Mitsui
Pharmaceupicall Inc.’Public Welfare Institute of Scietific
Research Fundation

"We teported in the [988 annual zoological meeting that skeletal
muscle satellite cells are the source of newly formed growth
fibers in hypertrophied skeletal muscle. In addition, basic
fibroblast growth factor(bFGF) is localized to the fiber
endomysial space in control and hypertrophied muscle, but the
telative concentration of FGF in hypertrophied muscleappears to
be increased. The source of FGF in hypertrophied skeletal
muscle is demonstrated in the present study. When a
conditioned medium(CM) is added to myogenic cell cultures
growing in minimal essential medium, there is a marked
stimulation of myogenic and fibroblast cell replication for the
first 30 hours, followed by fusion and myotube formation. With
CM there is also stimulation of myosin synthesis and myosin
accumulation after myotube formation. Without CM these
cultures do not grow or differentiate. It is clear. therefore, that
CM promotes growth both of fibroblasts and myoblasts in these
cultures. In vivo, rat soleus hypertrophy was induced by
removal of the gastrocnemius with and without prior (several
week) injection of Marcaine into the soleus. Marcaine will
deplete satellite cell populations. Without Marcaine there was
70% soleus hypertrophy compared to control contralateral
muscle, and only marginal hypertrophy when Marcaine was
injected prior to surgery. These result suggest satellite cell
involvement in hypertrophy.

BI 30

MOLECULAR INTERACTION BETWEEN INITIATORIN
AND SW-AT, AN ENDOPEPTIDASE INHIBITOR RE-
LATED TO THE REGULATION OF MOTILITY OF
APYRENE SPERMATOZOA OF THE SILKWORM.
MemOsanaiiu §S. Nagackaseand) Is, Sasalkals:
1Dept. Biol., Tokyo Metrop. Inst. Geront.;
PHaceeAgr= 7 TokyorUnivie Aga. Lech.) buch;
3Fac. Agr., Nagoya Univ., Nagoya.

Motile apyrene sperm of Bombyx mori in
the d. deferens that are just released
from the testis loose motility in the
ampulla d. deferentis and v. seminalis. On
ejaculation, they become motile again by
initiatorin, an prostatic endopeptidase,
in the spermatophore. On a hypothesis that
a system of initiatorin and its inhibitor
might regulate the motility, a trypsin
inhibitor, sw-AT (silkworm antitrypsin)
purified from the silkworm hemolymph was
examined immunologically. The testis and
g. prostatica, g. pellucida, g. lacteola
showed immunoreactivity against the anti-
body of sw-AT. Castration did not affect
the immunoreactivity of each organ. From
the reduction of activity of initiatorin
with time and the formation of peptide
fragments derived from degradation of a
complex of SW-AT and an endopeptidase, it
appeared that the above hypothesis might
be valid. However, sw-AT did not inhibit
the enzyme activity of initiatorin, nor
the motility of apyrene sperm activated by
the g. prostatica. Since this antitrypsin
strongly reduced the sperm motility
activated by trypsin, it might not conju-
gate with initiatorin.

1065

BI 31

ISOLATION OF MORPHOLOGICAL MUTANTS FROM
C6 GLIOMA CELLS AND CHARACTERIZATION OF
LAMININ DISTRIBUTION IN THEM.
T.Kobayashi. Department of Biochemistry,
Jikei University School of Medicine,
Nishi-Shinbashi, Minato-ku, Tokyo 105.

C6 glioma cells have been excellent to
study the neural cell development because
of their dramatic differentiation to
astrocytes in vitro. I have now isolated
3 mutants from them, which were altered in
morphologies of both cells and colonies,
formed long processes at confluency and
were named "spread", "linear" and
"dispersed", respectively. The "spread"
cells had decreased cell adhesiveness,
much longer processes and less numbers of
branches than the original C6 cells. The
"linear" cells were bipolar and formed
linearly arranged colonies. The "dis-
persed" cells made no colonies because
their daughter cells left each other. Each
of the mutants formed its characteristic
colonies after cells were dissociated and
reassembled. All these data suggested
that alteration in cell adhesiveness was
one of the reasons of these morphological
mutations. Actually, the “linear" cells
and the "dispersed" cells contained much
less amount of laminin than the original

cells. Distribution of laminin on the
"linear" cells was restricted at the cell
ends. Furthermore, laminin increased the

percentage of multipolar cells of the
"linear" cells and the "dispersed" cells.

BI 32

EFFECTS OF THE DIVARENT CATIONS ON THE
CELL MODELS OF THE CILIATE, SPIROSTOMUM.
H. Ishida and A. Matsuno. Dept. of
Biol., Fac. of Sci., Shimane Univ.,
Matsue

A large heterotrichous ciliate
Spirostomum ambiquum is well known for
its own characteristic twisting
contraction. This contraction is easily
triggered by giving various kinds of
external stimuli. The cell model was
prepared by treating with Triton X-100
and EGTA of external medium. The
contraction of this model was induced by
increasing the free Ca2+ concentrations
even in the absence of Mg-ATP, but the
contracted model was induced to extend
by addition of Mg-ATP without Ca2*,

When the level of free Ca2@t
concentrations of the external medium
rose to over 10-/’ M, the cell model
contracted. This cell model was induced
to contract by increasing the free
concentrations of the other divalent
cations such as Sr2+, Ba2+, but the
threshold value of them was higher than
that of Ca2*. The cell model was not
caused the contraction by exitence of
monovalent cations such as Li*.

On the other hand, at the presence of
10-8M to 1073M Mg2+ in external medium,
the contraction of cell model was not
found. But Mg2*+ which coexisted with
other divalent cations shifted the
threshold value higher of other cations.

1066 Biochemistry

BI 33

CILIARY ADENYLATE KINASE AND ARGININE
KINASE IN SOME PROTOZOA.

K. Numa, T. Sawada and M. Noguchi.
IDEjores Cre Wao@llo, MAG, O12 Seal , WoOye_ime:
Univ., Toyama.

Energy for the motility of flagella and
cilia is considered to be supplied by ATP.
The mechanism for supplying ATP to the
whole length of cilia and flagella has
been considered to be supported mainly
diffusion of ATP from cell body, and ATP
regenerationg system such as adenylate
kinase and creatine kinase contribute to
keeping sufficient level of ATP. On the
other hand, phosphocreatine shuttle has
been proposed as alternative system for
energy supplying in sperm flagella.

We have used reactivation systems of
cilia to understand the mechanism of
ciliary movement. To make a reactivation
system of cilia including ATP supplying
system for understanding the mechanism
totally, we reinvestigated adenylate
kinase and some phosphagen metabolising
enzyme of cilia from a couple of protozoa.
Adenylate kinase activity was detected in
all samples of cilia especially in
axonemal fractions. Arginine kinase
activity was detected in Paramecium and T.

pyriformis, but not in T. thermophila and
Chlamydomonas. Arginine kinase shuttle

might contribute to supplying ATP in some
ciliates. e

BI 34

ASSEMBLY OF ACTIN FILAMENTS DURING
CLEAVAGE FURROW FORMATION IN THE SEA
URCHIN EGG. II.

I. Mabuchi, Dept. of Biol., Coll. of Arts
and Sci., University of Tokyo, Tokyo.

I investigated in detail the formation
of the contractile ring by fluorescent
staining of actin filaments in fixed eggs
with rhodamine-phalloidin. No sign of the
formation was observed until late
anaphase. At early telophase when the
chromosomal vesicles began to fuse each
other at the two poles, the density of the
actin filaments at the equator increased.
These filaments could not be distinguished
from the bulk cortical actin filaments in
appearance. The equatorial filaments
seemed to associate with each other to
form tight bundles as cytokinesis
progressed. Involvement of myosin
ELLAMOMES wm EMG ECOMmEractEr le eiame
formation was investigated using an
inhibitor for myosin light chain kinase,
ML-9. O. 15 mM ML-9 blocked cleavage
while nuclear division was not affected.
The contractile ring was not formed in
these eggs. Rather the region where the
contractile ring should have been formed
lacked the actin filaments leaving a dark
belt. These results suggest that the
myosin phosphorylation is relevant to the
contractile ring formation and that a
dynamic interaction of the actin filaments
with the plasma membrane may occur in its
process.

BI 35

LOCALIZATION OF TETRAHYMENA PROFILIN.
M.Edamatsu!, M.Hirono!,2, and Y.Watanabe!.
lInst. of Biol. Sci. Univ. of Tsukuba,
Tsukuba and 2Cell Bank, RIKEN Inst. of
Phys. and Chem. Res., Tsukuba.

We have succeeded in isolating of
Tetrahymena profilin and have shown its
inhibitory activity of actin polymeriza-—
tion. In this study, we examined localiza-—
tion of Tetrahymena profilin by indirect
immunofluorescence using the specific
antiserum. Tetrahymena profilin was local-
ized in furrow region of a dividing cell
as well as in cytoplasm. Although the
profilin in cytoplasm was soluble, the
profilin localized in furrow region was
associated with cytoskeletal structures.
In addition, the localization of profilin
in division furrow coincided well with
that of contractile ring microfilaments.
From the detailed observation of double-
staining immunofluorescence with anti-
actin and anti-profilin, actin filament
formation was shown to precede profilin
deposit in a very early stage of cytokine-
sis and profilin deposit was shown to be
much evident in the late stage of cytoki-
nesis. From the results, we speculate that
profilin may play a role for the disap-
pearance mechanisms of contractile ring
microfilaments.

BI 36

EXPRESSION OF TETRAHYMENA ACTIN GENE IN
MAMMALIAN CULTURED CELLS.

M. Hirono!,2, K. Kurita!, T. Ohno!’ and Y.
Watanabe2. 1!Cell Bank, RIKEN Inst. of Phys.
Chem. Res., Tsukuba, 2Inst. of Biol. Sci.,
Univ. of Tsukuba, Tsukuba.

We previously reported that Tetrahymena
actin has some essential properties such as
ability to copolymerize with skeletal
muscle actin, but does not have some other
properties common to ubiquitous actins.
This leads to the possibility that if this
Tetrahymena actin were expressed in a
higher animal cell, it would be incorporat—
ed into microfilaments, but would inhibit
actin-mediated biological phenomena through
failures in the interactions between copol-
ymerized microfilaments and actin-binding
protein(s). To verify this possibility,
Tetrahymena actin gene on a expression
vector was constructed and transfected into
COS-1 cells. The expressed Tetrahymena
actin was specifically detected by indirect
immunofluorescence microscopy using anti-
Tetrahymena actin antiserum and distin-
guished from intrinsic actin which was
stained with rhodamine-phalloidin. The
expressed actin was indeed copolymerized
with intrinsic actin and localized only at
the tips of actin bundles which exist in
microspikes. We also investigated the ef-
fect of the expression of Tetrahymena actin
on cell growth. Although the expression was
transient, it was suggested that the ex-
pression greatly influences cytokinesis.

Biochemistry 1067

BI 37

DRASTIC CHANGE IN THE LEVEL OF ACTIN mRNA
IN THE COURSE OF SYNCHRONOUS DIVISION IN
TETRAHYMENA. ;

M. Kimura!, M. Hirono!2, T. Takemasa!, and
Y. Watanabe!. lInst. of Biol. Sci., Univ.
of Tsukuba, Tsukuba and 2 Cell Bank, RIKEN
Inst. of Phys. and Chem. Res, Tsukuba.

In order to investigate the
relationships between cell division and
actin mRNA level, we quantified actin mRNA
level in the course of synchronous
division in Tetrahymena induced by
intermittent heat treatment. The level of
actin mRNA markedly decreased from the end
of the heat treatment (EHT) to 45 min
after EHT, and then dramatically increased
towards synchronous division at 75 min
after EHT. Effects of actinomycin D on
both synchronous division and actin mRNA
level indicated that exhibition of
Synchronous division was always
accompanied by the increase in actin mRNA
level after 45 min. Results of nuclear
run-off experiment showed that the rate of
actin mRNA synthesis increased after 45
min and the rate of actin mRNA decay
decreased from 0 to 60 min, so that actin
mRNA must have been effectively
accumulated at the pre-division stage.
From these results, the increase in actin
mRNA level starting from 45 min is
speculated to be prerequisite for the
oncoming synchronous division.

BI 38
DESMIN ASSEMBLY IN THE PRESENCE OF ACTIN
S.Aihara and K.Ohashi. DENES Cie isso 57

PAC. O18 SCiop Clmalloe Uimatys p Claaloel.

Desmin, a component of the intermediate
filaments of muscular cells, was purified
from chicken gizzard smooth muscle. When
actin heavily contaminated the desmin
sample, the assembly of desmin was inhibited.
Artificially denatured actin was added to
the desmin protofilaments (2mg/ml) dis-
persed in a solution containing 10 mM Tris-
Hels pHwoomat the watvwo Of Mis! and then
the mixture was diluted 20 fold with 0.15 M
NaCl and 25 mM sodium phosphate buffer, pH
6.8 (assembly buffer). Desmin did not as-
sembled as usual and a few short filaments
were observed. Desmin (0.02 mg/ml) could
assemble into 10 nm filaments in the pres-
ence of G-actin (0.01 mg/ml). It seemed
that G-actin had no effect to the desmin
assembly. When the mixture of desmin (0.1
mg/ml) and G-actin (0.1 mg/ml) was incu-
bated for 2 h in the assembly buffer,
shorter and a smaller amount of desmin fil-
ament than control were observed and actin
filaments scacely existed. Then actin (0.1
mg/ml) was polymerized in a solution con-
taining 0.1 M KCl, 0.3 mM MgCl. and 10 mM
sodium phosphate buffer, pH 7.4 with desmin
(0.1 mg/ml). Actin polymerization was
heavily inhibited and few desmin filaments
were observed.

These resuits suggested that oligomeric
actin and desmin inhibited their filament
formations with each other.

BI 39

BINDING OF Z-PROTEIN TO ACTIN AND DESMIN
FILAMENTS.
H.Nakagawa and K.ohashi. DEINE, Oe isloils,
WAGs OF Seis, Codloe wins, Clatiog.

Z-protein 1S a component of the Z-line
of skeletal muscle. Freshly purified Z-
protein was added to a F-actin (0.1 mg/ml)
solution containing 0.1 M KCl, 1 mM MgCl,,
and 20 mM sodium phosphate buffer, pH 7.2
at the ratio of 1:5 (w/w) and incubated for
5 h at 10% and centrifuged at 15,000 rpm
for 15 min. Z-protein was coprecipitated
with a part of F-actin. The electron micro-
scopic observation of the mixture showed
that Z-protein bundled and entangled F-actin
into large aggregates. The same aggregates
were formed when G-actin (0.1 mg/ml) was
polymerized in the presence of Z-protein.

When desmin (0.1 mg/ml) was assembled
with Z-protein (0.1 mg/ml) in a solution of
0.15 M NaCl and 25 mM sodium phosphate
buffer, pH 6.8 (assembly buffer), larger
and more dense aggregates of 10 nm filaments
than control were formed. Then desmin (
0.01 mg/ml) was suspended in the assembly

buffer. In this case, few filaments were
formed. In the presence of Z-protein (0.01
mg/ml), desmin of the same concentration

was assembled into short 10 nm filaments.
Even when the concentration of Z-protein
was 0.001 mg/ml, the assembly of some
filaments seemed to begin. These results
suggested that Z-protein would accelerate
the desmin filament formation, decrease the
critical concentration and entangled desmin
filaments.

BI 40

210KD ACTIN-BINDING PROTEIN FROM LOWER
EUKARYOTE, PHYSARUM POLYCEPHALUM.

R. Ishikawa’, T. Okagaki“~, S- Higashi-
Fujime~, and K. Kohama Dept.
Pharmacology, Gunma Univ. School of Med.,
Maebashi, “Physical Sci. Lab., Nihon Univ.
at Narashino, Funabashi, and ~Dept. Mol.
BLOGS, wacwaliesy O12 SGEatop WEC|Os/El WialalN7o
Nagoya

Ss eS SSS Ln EL,
We have purified and characterized the

actin-bundling protein from lower
eukaryote, Physarum polycephalum
plasmodium with apparent Mr of 210,000 on
SDS-PAGE. It binds to calmodulin-agarose
in a Ca** dependent manner, and binds to
F-actin at the stoichiometry of 1:10 in
Ca*t-calmodulin dependent manner. This
210 kDa protein forms F-actin-bundles in
the presence of reducing agent, suggesting
that 210 kDa protein has at least two
actin-binding sites. The bundling
activity of 210 kDa protein is inhibited
by Ca“*t-calmodulin. From EM observation,
210 kDa protein is elongated molecule
which bends at central position. Stokes
CAGaLws Cie DIO IDE jore@ie@stin a WSO. 7c
Immunofluorecence study shows that 210 kDa
protein” clo—Tocalizels iwith actin
stressfiber in the thin-spread Physarum
plasmodium. These results suggest that
210 kDa protein may play an important role
on the appearance-disappearance of actin-
bundles in the contraction-relaxation
cycle of Physarum plasmodium. This is the
first report that lower eukaryote has
caldesmon-like protein.

1068 Biochemistry

BI 41

NON-MUSCLE CALDESMON DURING MITOSIS IN RAT
EMBRYO FIBROBLAST CELLS: RELATIONSHIP AMONG
ITS PHOSPHORYLATION LEVEL, MICROFILAMENT-
BINDING AND LOCALIZATION.

N.Hosoyat* H.Hosoya* S.Yamashiro4% IN 6
Matsumura% and H.Mohri*t *Dept. Biol.,
Univ. Tokyo, Tokyo 153 and“Dept. Biochen.,
Rutgers Univ., NJ 08854-1059, U.S.A.

We have found that caldesmon (83K) is

phosphorylated and dissociates from
microfilaments (MFs) in mitotic-arrested
rat embryo fibroblast cells. We here ex-
amined the relationship among phosphoryla-
tion level, MF-binding, and localization
of 83K during mitotic process.

At prometaphase, 83K was localized dif-
fusely within the cytoplasm. During cyto-
kinesis, 83K was not accumulated in the
contractile ring. As the stress fibers
reappeared in the daughter cells, 83K
began to be localized along them. 83K was
absent in the MF fraction obtained from
prometaphase cells. It began to bind to
MFs during cleavage furrow formation and
the amount of 83K bound to MFs increased
with the progress of mitosis. Finally we
examined the phosphorylation level of 83K
during mitosis. 83K was still highly
Phosphorylated at the beginning of
celavage furrow formation. Dephosphory-
lated form of 83K became dominant at the
end of cytokinesis, and 83K remained all
dephosphorylated after cytokinesis.

These results suggest that phosphoryla-
tion of 83K is a major mechanism modulat-
ing MF organization during mitosis.

BI 42

DIFFERENTIAL DISTRIBUTION OF COFILIN IN
SLOW AND FAST FIBERS OF CHICKEN SKELETAL
MUSCLES.

M. Tsukahara, H. Abe and T. Obinata. Dept.
of Biol., Chiba Univ., Chiba.

Cofilin, a pH-dependent modulator for
actin assembly, is expressed in chicken
skeletal muscles, although the amount
decreases during development. Im elaalS
sreuihy, Ie@eelibsipzetialoin Gre COrwalllaiio sia) e\elulie
chicken muscles were examined with an
anti-cofilin antibody (MAB-22). When whole
muscle lysates were displayed on SDS-PAGE

and reacted with MAB-22, cofilin was
detected clearly in smooth and_ skeletal
muscles. Among skeletal muscles, cofilin
was more abundant in slow (ALD) muscle

than in fast (pectoralis) muscle. In order
tO CGllasesssy elo} COreailalin Glisieiealloyoiest@ye) — abigl
individual cells, serial cryosections of
skeletal muscles were prepared. The cells
in the sections were classified into three
types based on binding of the antibodies
to fast (MF-1) and slow (ALD-66) C-
proteins; i.e., Type 1: MF-1(-) and ALD-66
(+), Type 2: MF-1 (+) and ALD-66 (-) and
Type 3: MF-1 (+) and ALD-66 (+). The type
1 corresponds to slow fiber, while the
type 2 and 3 do to fast fiber. When the
sections were reacted with MAB-22, the
staining was clear in the type-1 cells but
aloe, all ilo ieyfoa D eucvel 3) atslloeiesy, Ane
staining was associated with the I-band
regions of sarcomeres. We conclude that
cofilin is enriched in slow fibers.

BI 43

INTRACELLULAR LOCALIZATION OF COFILIN IN
CULTURED MUSCLE CELLS.

H. Abe, R. Nagaoka, and T. Obinata. Dept.
of Biol., Fac. of sci., Chiba Univ., Chiba
260.

In embryonic chicken skeletal muscle,
considerable amount of actin remains
unpolymerized in the cytoplasm by the
action of low Mr-actin binding proteins.
Among these proteins, cofilin is of
particullar interest, since it binds to
both G- and F-actin and regulates actin
polymerization in a pH-dependent manner.
In this study, we examined the
localization of cofilin in cultured muscle
cells immunocytochemically by using a
monoclonal antibody specific for cofilin
(MAB-22). Cofilin is mostly diffused in
the cytoplasm, but occasionally associated
with Sarcomeric structures. Dimethyl-
sulfoxide-treatment induced actin-cofilin
rods in most nuclei of myotubes, while
heat-shock produced vesicles containing
both cofilin and actin in the extrudes on
the surface of myotubes. The intranuclear
rods and the vesicles were also generated
when myotubes were maintained in the
medium containing cycloheximide.
Cytoplasmic actin-cofilin rods were often
observed abundantly in degenerating
myotubes with vacuoles. The dynamic
translocation of cofilin under these
stresses suggests that cofilin is involved
in redistribution and turn-over of actin
molecules in muscle cells.

BI 44

DIFFFERENTIAL EXPRESSION OF COFILIN AND
ADF DURING DEVELOPMENT OF CHICKEN SKELETAL
MUSCLE.

K. Saiga, H. Abe, and T. Obinata. Dept.
Of "Biol, Face Of (Sci), a ChibamUnanzey i ecntod

Cofilin and ADF are the low-Mr actin-
regulatory proteins which are functionally
distinct but exhibit high structural
homology (70 % identity in amino acid
sequence) (Abe et al, Biochemistry 29,
7420-7425, 1990). They are involved in
the regulation of actin assembly in
embryonic skeletal muscle. In this study,
we examined the expression of ADF and
cofilin in skeletal (SKM), cardiac (CM),
smooth (SMM) muscles and brain during
development by a Northern blotting method
with the cDNA probes for ADF and cofilin.
While the ADF mRNA bands (0.9 & 2.1 Kb)
were most remarkable in brain through
developmental ages, the same message was
also abundantly detected in SMM, CM and
SKM of chicken embryo, but declined as the
muscles developed. In adult SKM, poly(A+)
RNA of about 4 Kb which cross-hybridized
with the probes for ADF appeared speci-
fically. The protein encoded by this
message is unknown. Cofilin mRNA (1.5 Kb)
was detected in SMM, CM and SKM of both
embryo and adult. Additional three faint
bands of 4-5 Kb cross-hybridized with the
cofilin cDNA in both muscles and brain of
all developmental ages. The amounts of
cofilin message among adult tissues were
in the order SMM > SKM > brain > CM.

Biochemistry 1069

BI 45

EXPRESSION OF COFILIN IN AVIAN AND
MAMMALIAN MUSCLE TISSUES.

Ne Minami’, “K. Yabusaki!, H. | Abel; 2.
Totsuka2 and T. Obinatal. 1Dept. of Biol.,
Bac Of Sei auCha bal Univ. ,.yeChilbal,, (2Dept.
of Physiol., Aichi Pref. Colony, Kasugai.

Cofilin is an actin-binding protein
which was discovered in mammalian non-
muscle tissues. It is involved in the
regulation of actin assembly in embryonic
skeletal muscle. It was described that
cofilin is absent from skeletal muscle
(Yonezawa et al., 1987), but recently,
cofilin was detected in adult chicken
muscles. In this study, we re-examined the
distribution of cofilin in mammalian and
avian muscles by immunoblotting with the
antibody to chicken cofilin (MAB-22).
When tissue lysates were displayed on SDS-
PAGE and reacted with MAB-22, cofilin was
detected in smooth and skeletal muscles of
chicken and mammals and in non-muscle
tissues, as well. The cofilin spots of
mammalian brain and skeletal muscle did
not co-migrate on 2D-PAGE suggesting both
may be different isoforms. Cofilin was
enriched in slow fibers both in mammals
and chicken. In dystrophic (dy) mouse
muscle, the amount of cofilin increased
Significantly. When the cultures of mouse
skeletal muscle cells were treated with
10% DMSO, actin-cofilin-rods were formed
in nuclei. From these results, we conclude
that cofilin is expressed in both chicken
and mammalian skeletal muscle cells.

BI 46

CHANGES IN MYOSIN AND TROPONIN T (TNT)
TSOFORMS DURING METAMORPHOSIS OF FLOUNDER
T. Obinata!, Y. Itohl, H. Takano-Ohmuro2 ,
Keagavamano>) |, Yn sinuis..+Dept., jf | Biol,
ChabassUmiv. )Chalba...,, 2 Tokyo. Metro. inst.
Med. Sci., Tokyo, 3Natl. Res. Inst.
Aquaculture, Mie.

It is well known that many isoforms
exist in myofibrillar proteins of avian
and mammalian skeletal muscles and their
expression changes during development, but
little is known as to the isoforms in
fish. Here, we report that myosin and
troponin T (TNT) isoforms in flounder
skeletal muscle change markedly during
metamorphosis. By PPi-PAGE, a single
myosin isozyme in adult but three bands in
larva were detected. The relative
proportion of the three isozymes varied
with the progression of metamorphosis. In
accord with this change, changes in myosin
heavy (MHC) and light chains (MLC) were
detected; MHCs in larva and adult differed
in size and peptide map. Larva and adult
contained different regulatory MLCs. A
major catalytic MLC did not change during
development, although a minor larva-
specific MLC was detected. TNT variants
with approximate Mr of 42 kDa and 34 kDa,
respectively, were identified in adult
with anti-TNT antibody (NT-302); the
former is a basic protein as TNTs in the
other animals, but the latter is an acidic
protein. The amount of the higher Mr-TNT
variant decreased during development.

BI 47

MYOSIN HEAVY CHAIN ISOFORMS IN ADULT AND
EMBRYONIC CHICKEN GIZZARD SMOOTH MUSCLES
AND POSSIBLE POST-TRANSLATIONAL
MODIFICATIONS ON THE ISOFORMS.
H.Takano-Ohmuro', Y.Kasuya’, Y.Katsuragawa
M. Yanagisawa’, Y.Okamoto*, T.Masaki*. ‘The
Tokyo Metro.Inst.Med.Sci.,Tokyo, *Inst.
Basic Med.Sci., Univ. of Tsukuba, Ibaraki
305, *Juntendo Univ. Tokyo 113.

It has been reported that there are
two myosin heavy chains (MHCs) in various
adult vertebrate smooth muscles (SMs)
molecular weights of these proteins on SDS
PAGE are 204 KDa (MHC,) and 200 KDa (MHCz2),
Yanagisawa et al. (1987) showed that the
gizzard (G) SM MHCs were encoded by a
single-copy gene, while Nagai et al.
(1989) reported that two MHCs in rabbit
uterus were generated by different mRNAs
which are made from respective cDNA.

We found that embryonic GSM MHCs were
immunochemically different from adult
GSM MHCs. However, the reactivites of
embryonic GSM MHCs became simillar to
that of the adult MHCs after incubation
with an adult GSM cytosol fraction which
did not contain any myosin. Sl nuclease
Mapping with fragments of embryonic G
MHC cDNA as probes revealed that MHC
mRNAS in adult and embryonic GSMs appeared
to be identical. Furthermore, chemically
methylated MHC2 was co-migrated with
MHC, on SDS PAGE.

These results strongly suggest that
MHC isoforms in chicken GSMs are produced
due to post-translational modifications.

BI 48
SCAFFOLDING STRUCURE OF MYOFIBRILS: DESMIN

‘NET AND EFFECT OF PHOSPHORYLATION.

M. Kurodal, A. Matsuno! and H. Ohmuro2:
IDept. of Biol., Fac. of Sci., Shimane
Univ., Matsue and 2Tokyo Metropol. Inst.
for Med. Sci., Tokyo

Extraction of strated muscle with
dilute lactic acid gave scaffolding struc-
ture of myofibrils, in which Z-disks were
arrayed as a ladder-like orientation. Even
after extraction of major body of contrac-
tile proteins by the acid, Z-disks of the
scaffolding structure did not leave apart
but retained its original disposition.
Electron microscopy of this scaffolding
structure networks of short 100-nm fila-
ments interconnected adjacent Z-disks.
Immunofluoresence and immunoelectron mic-—
roscopy with anti-desmin indicated that
desmin existed not only at Z-disks but
also in the longitudinal spaces between
adjacent myofibrils. From these studies,
we concluded that network of short desmin
filaments constructed basic unit in the
scaffolding structure.Phosphorylation bv
either A- or C-kinase induced drastic
changes in the scaffolding strcture.
Mushroom-like projections periodically
emerged from Z-disks. The mushroom with
diameter of 180 nm in its head portion at-
tached to Z-disk through a stem of 38-nm
wide. We speculated that disintegration of
the desmin net by the phosphorylation
resulted in clumping of desmin and other
components of scaffoldng structure.

1070 Biochemistry

BI 49

ON THE DISTRIBUTION OF TROPOMYOSIN ISOFORMS
IN BIVALVE MUSCLES.

T.Ishimoda-Takagi and R.Inoue. Dept. of
Biol., Tokyo Gakugei Univ., Tokyo.

Our previous observations have shown that
two major tropomyosin (TM) isoforms (referr-
ed to as a and 8 components) were involved
in bivalve muscles, and that distribution
pattern of the TM isoforms in muscle tissues
was varied among various bivalves, although
similar patterns were observed in related
species. These results were so complicated
that we could not elucidate the phylogenetic
correlation in the distribution of TM iso-
forms. However, assuming that various types
of distribution of TM isoforms were derived
from undiscovered prototype which contained
both of the TM isoforms in all muscle tis-
sues, we could interpret the feature of dis-
tribution of TM isoforms in bivalve muscles
without phylogenetic inconsistency. To dem-
onstrate this possibility, it was required
to find out the species which involved both
of the TM isoforms in all muscles. Thus, we
investigated the distribution of TM isoforms
in muscle tissues of the oyster, Crassostrea
gtgas, since our preliminary observation had
shown a possibility that both translucent
and opaque portions of the adductor muscle
of this species contained two TM isoforms.
We showed in the present study that all oys-
ter muscle tissues examined contained two TM
isoforms corresponding to a and 8 components
of the TM isoforms, although ratio of the TM
isoforms was not constant among various mus-—
cle tissues.

BI 50

STRUCTURAL ANALYSIS OF THE CLONED cDNAs OF
ASCIDIAN SMOOTH MUSCLE TROPONIN T

T. Hama, S. Goto, T. Obinata, and T. Endo

Dept. of Biol., Fac. of Sci., Chiba Univ., Chiba

Troponin T (Tn T) is a tropomyosin-binding subunit of
troponin complex that suppresses striated muscle contraction in
the absence of Ca**. We have shown that the body wall muscle
of ascidian Halocynthia roretzi is a smooth muscle but that its
contraction is regulated by the troponin complex. Different from
striated muscle troponins, the ascidian protein is a Ca*+-
dependent activator of the contraction. This activating property
is attributable to the Tn T. Tn T in various chordate animals
including ascidian is composed of multiple isoforms. At least in
higher vertebrates, these isoforms in a tissue are generated from
a single gene by alternative splicing of pre-mRNA. Thus, it is
of importance to examine the functional domains of ascidian Tn
T and which mechanism is responsible for the generation of the
multiple ascidian isoforms. For these purposes, we constructed
a cDNA library of ascidian smooth muscle in Agtl1 and
screened it with a monoclonal antibody to chicken skeletal
muscle Tn T (NT302). Three cDNA clones with similar sizes
(~1.0 kb) were obtained, and the complete sequence of one of
them was determined. This comprized 968 nt and contained one
open reading frame starting at one end and a poly(A) tail at
another end. The amino acid sequence deduced from the
nucleotide sequence showed ~50% homology to rabbit skeletal
and cardiac muscle Tn Ts. The ascidian cDNA and the rat
skeletal muscle Tn T cDNA pTnT15 did not crosshybridize to
mouse striated muscle Tn T and ascidian Tn T mRNAs,
respectively, on RNA blot hybridization. Considering the
difference between the molecular weight calculated from the
deduced amino acid sequence (24,788) and that of the ascidian
protein on SDS-PAGE (33 kDa), the cDNA clones lack the 5’
upstream portion containing a translation initiation site.

BI 51

EXPRESSION OF CHICKEN SKELETAL MUSCLE a-
ACTININ cDNA IN CULTURED MYOBLASTS AND NON-
MUSCLE CELLS

Y. Tokuue, S. Goto, and T. Endo

Dept. of Biol., Fac. of Sci., Chiba Univ., Chiba

a-Actinin, an actin-crosslinking dimer protein, is located in
Z-bands of striated muscles, in dense bodies and dense plaques
of smooth muscles, and in microfilament bundles of nonmuscle
cells. Nonmuscle a-actinin exhibits Ca?+-sensitivity for the
binding to F-actin, whereas striated and smooth muscle a-
actinins do not. The amino acid sequence of skeletal muscle a-
actinin possesses only ~80% homology to that of smooth muscle
and nonmuscle proteins. We have addressed whether the
differences in structure and function of various a-actinin
molecules are responsible for the tissue-specific localization.
The nucleotide sequence of an @-actinin cDNA cloned from a
cDNA library of 12-d old chick embryo skeletal muscle was
very Similar to that of an adult skeletal muscle a-actinin cDNA.
This cDNA was ligated into the expression vector pcDL-
SRa296. Mouse skeletal muscle cell line C2 myoblasts and
fibroblast cell line C3H10T1/2, which express nonmuscle type
Q-actinin but not skeletal muscle type isoform, were transfected
with the construct. Immunofluorescence microscopy by use of
an antibody specific for skeletal muscle a-actinin located the
expressed exogenous Q.-actinin in microfilament bundles, focal
contacts, ruffling membranes, and contractile rings. Thus, the
protein coincided with the endogenous nonmuscle type isoform.
These results suggest that skeletal muscle and nonmuscle a-
actinins function as the same protein in a cell despite their
differences in structure and Ca?+-sensitivity. Remarkably,
microfilament bundles were extinguished in the cells very highly
expressing the exogenous protein. Transfection with cDNAs
deleted in their 3’ portions indicated that the proteins that cannot
form dimers are not able to localize in microfilament bundles.

BI 52

DOES @-ACTININ BIND THE FREE END OF AN
ACTIN FILAMENT IN MYOFIBRILS? i

M. Itoh, S. Watanabe and K. Maruyama.
Dept.of Biol. > rac. ‘Sca., 7Ghilbal Unaweenchrbar

We have reported that @-actinin is iden-
tical with Cap Z protein that bind the
barbed end of an actin filament and is
localized in the Z line of chicken breast
muscle myofibrils (Maruyama et al.,J. Biol.
Chem. ,265,8712-8715(1990)). Funatsu et al.
(J. Biochem. ,103,61-71(1988)) showed that
S8-actinin capped the free (pointed) end of
an actin filament in myofibrils. Therefore
we reexamined this problem.

In the present study, 4 pM @-actinin
completely stopped the elongation of actin
filaments in the socalled IZI brushes
prepared from chicken breast muscle in the
presence of added 1 pM G-actin. However,
it was observed that the growth of actin
filaments in the IZI brushes did occur on
the addition of 3-4 pM G-actin. Thus, in
the presence of excess amount of f-actinin
actin elongation was inhibited.

Furthermore, gold-labeled antibodies to
the subunits (@I and MII) did not bind the
free ends of the actin filaments in IZI
brushes in disagreement with the results
of Funatsu et al.(1988). Hence, it appears
that @-actinin does not cap the free ends
of the actin filaments in myofibrils.

a

Biochemistry 1071

BI 53

DYSTROPHIN FROM RABBIT SKELETAL MUSCLE.
O. Satol, K. Maruyamal and Y. Nonomura2.
ipepesoLyBiol.);-hac.Scei., Chiba UntivA,,
Chiba and 2Dept. of Pharmacol., Fac.Med.,
Univ. Tokyo.

In a previous report ( Murayama et el.,
Proc. Jap. Acad.,66B,96-99(1990)), it was
indicated that rabbit skeletal muscle
dystrophin was a dumpbell-shaped molecule,
100-120 nm long and 3 nm wide.

The present work has improved the puri-
fication procedure of dystrophin: rabbit
skeletal muscle myofibrils were extracted
with 1 % Triton X-100 salt solution and
dystrophin was precipitated with polyethy-
lene glycol 6000. The crude sample was
subjected to hydroxylapatite, WGA-agarose,
and DE-52 column chromatography.

Rotary shadowed images of highly puri-
fied dystrophin were flexible rods with
attached globules, 2-3 nm in diameter. The
flexible rods, about 100 nm long, were very
Similar to erythrocyte spectrin dimers und-
er the same conditions. The dumpbell stru-
cture we previously described was very few
in population. It might be possible that
the dumpbell structure was broken down into
rods and globules during procedures for
purification and also for electron micro-
scopy. Further studies are in progress
using gold labeled antibodies to dystrophin.

BI 54

MOLECULAR SIZES OF INVERTEBRATE MUSCLE
CONNECTIN (MINI-TITIN) ARE GREATLY VARIABLE.
T.Matsuura, Y.Nakauchi, S.Kimura and K.
Maruyama. Dept. of Biol., Fac. Sci., Chiba
Unive Chabal.

We have reported that the molecular mass
of native connectin isolated from crayfish
tail muscle was 1200 kDa (D.H.Hu et al., J.
Muscle Res. Cell Motil.,in press). However,
according to Nave and Weber(J.Cell.Sci.,95,
535(1990)), the MM of connectin from locust
flight muscle was 700 kDa. A similar value
was obtained for other invertebrates,
earthworm and Ascaris. They called "mini-
titin",distinct from vertebrate skeletal
muscle connectin (titin): 3000 kDa.

We have done immunoblot tests with
various invertebrate muscles using a mono-
clonal antibody to chicken connectin (4C9)}
The MMs of the proteins reactive with 4C9
were as follows:600 kDa(polychaete,Annelida
Pecten,Mollusca; seacucumber, Echinodermata)
900 kDa(earthworm,Annelida); 1200 kDa (
locust,honeybee and cockroach,Arthropoda)
and 2000 kDa(acidian,Prochordata).

Thus, a great variety of the MMs has
been observed among invertebrate connectin.
This is not surprising in view of repeated
sequences of the two types of immuno-
globulin superfamily motifs found in C.
elegans twitchin and chicken skeletal
muscle connectin.

BI 55
TETRAHYMENA 14-NM FILAMENT-FORMING PROTEIN

FORMS BEGET BUNDLES IN MITOCHONDRIA
O.Numata,, T.Yasuda‘“~, T.Tomiyoshil, We
Kurasawa’, J.Chiba~, and Y.Watanabe
IMNSiES OL BWiOils Ses. - iene of Tsukuba,
Tsukuba, Ibaraki 305; Dep Ol elechs
National Inst. of Health, Shinagawa-ku,

TOkyOMl4is Dep sot eBoy SCA sand) etechs.

Science Univ. of Tokyo, Yamazaki, Noda 278.

We previously reported cloning and
sequencing of the 49K protein gene. Its
primary structure not only exhibits low
homologies with vertebrate vimentin and
desmin, but exhibits a high sequence
identity (51.5%) with porcine heart citrate

synthase. The 49K protein actually
possesses citrate synthase activity, and is
detected in mitochondria. These results

suggest that the 49K protein has dual
functions as both a respiratory enzyme and
a structural protein in the cytoskeleton.

We further investigated immunoelectron
microscopical localization of the 49K
protein in mitochondria with anti-49K
protein antiserum. We obtained the
results that the 49K protein formed
filament bundles in mitochondrial matrix
and that size of each filament was
approximately 16-nm. The results support
our nortion that polymerized 49K protein
(14-nm filament) functions as citrate
synthase in mitochondria.

BI 56
TETRAHYMENA 14-nm FILAMENT - FORMING
PROTEIN (49K PROTEIN) HAS CITRATE

SYNTHASE ACTIVITY
I.Takagi, O.Numata, Y.Watanabe. Institute
of Biological sciences., University of
Tsukuba, Ibaraki.

We have previously reported the cloning
and sequencing of the Tetrahymena 49K
protein gene and found that its primary
sequence exhibited a high sequence
identity (51.5%) with porcine citrate
synthase.

In this study, we purified 49K protein

and assayed its citrate synthase
activity. The results showed that 49K
protein actually possessed citrate

synthase activity.

Next, we prepared mitochondrial extract
from Tetrahymena and compared its citrate
synthase activity of its extract with
that of 49K protein. The enzyme activi-
ties of mitochondrial extract and puri-
fied 49K protein showed nearly the same
characteristics: (1)They showed the same
jos! Coes, (Kjole 7/GO=755))6 ((2)) Me™ joel Hosp
these enzyme activities were similarly
activated by ATP. (3) These enzyme activ-
ities were inhibited by palmitoyl-CoA at
the same degree.

In addition, the monoclonal antibodies
against 49K protein reacted with 49KDa
protein in mitochondrial fraction. These
results suggest that mitochondrial
citrate synthase is identical with 49K
protein.

1072 Biochemistry

BI 57

ACTIVATION OF 14S DYNEIN ATPase FROM —
TETRAHYMENA CILIA BY MICROTUBULES.

T. Shimizu', S. Ohashi’, S. Hisanaga?, and M.M.
Pratt’. ‘Res. Inst. Polym. Text., Tsukuba, Ibaraki, @Lab.
Devel. Cell Biol., Tokyo Inst. Tech., Meguro, Tokyo, and
3Dept. Anat., Univ. Miami, Miami, FL 33101 U.S.A.

Cilia of Tetrahymena yield two kinds of dynein; the
major form is 22S outer arm dynein and the minor form is
14S dynein. The latter has been shown not only to
translocate microtubules when adsorbed onto a glass
surface, but also to generate torque to rotate
microtubules (Vale & Toyoshima, 1988: Cel/, 52: 459).
Here we investigated the activation of 14S dynein ATPase
by microtubules.

MAP-free microtubules enhanced the ATPase activity
of 14S dynein in a concentration dependent manner.

The magnitude of enhancement was 2.5-3.5 fold at 10
mg/ml microtubules. At 17 mg/ml, the ATPase activity
continued to increase. 14S dynein caused bundling of
microtubules but no periodic decoration of the
microtubule wall was observed. The bundling was ATP
sensitive and, upon ATP treatment, most of 14S dynein
was released from microtubules.

Treatment of the microtubule-14S dynein complex with
EDC, a zero-length crosslinker, caused a 5-10 fold
activation of the 14S dynein ATPase. The bundling was
no longer sensitive to ATP. Most of the ATPase activity
was recovered in the pellet upon air-centrifugation even in
the presence of ATP. SDS-PAGE of EDC-treated
samples revealed new polypeptide bands larger than the
14S dynein heavy chain.

BI 58

CROSS-LINKING OF THE W/IC1l COMPLEX ISOLATED
FROM SFA URCHIN SPERM FLAGELLA WITH DIMETYL
SUBERIMIDATE. K. Inaba. Misaki Marine
Biological Station, Univ. of Tokyo, Miura.

Outer-arm dynein from sea urchin sperm
flagella can be separated into three
fractions ; the & chain, the $/inter-
mednate: chain (Te) Wavand ithe e277 sie
Treatment of the @/ICl complex with dimethyl
suberimidate (DMS) produced two major cross-
linked products whose molecular masses were
higher than the @ chain. The lower molecular
mass product ($-ICl) reacted with both anti-
8 and anti-ICl antibodies. The higher mole-
cular mass one, however, was less reactive
with anti-@ antibody. Treatment of the £-
ICl with methylamine resulted in the
cleavage of it into the $ chain and ICl. The
ATPase activity of the @/ICl complex was
elevated with the concentration of DMS up to
1.2 mM, and then gradually decreased with
increasing the DMS concentration. To
localize the ICl-binding site on the @ chain
the DMS-treated /ICl fraction was subjected
to vanadate(Vi)-dependent photocleavage (V1l-
cleavage). The free @ chain was reduced to
23 % of the initial amount after the
irradiation for 180 min, whereas the cross-
linked @-ICl was to 67 %. Examination of
the Vi-inhibition of the DMS-treated $/IC1
ATPase suggested that the low efficiency
of Vl-cleavage was due to the low
SenSitivilty OL, this product to Va.

BI 59

INTERACTIONS OF FLAGELLAR DYNEINS ISOLATED
FROM SEA URCHIN SPERM WITH MICROTUBULES.

E. Yokota and I. Mabuchi. Department of
Biology, College of Arts and Sciences,
University of Tokyo, Komaba, Meguro-ku,
lokyo153.- = SS ee

We have isolated C/A dynein from sea
urchin sperm flagellar axonemes, which is
composed of C and A heavy chains and some
additional lower mol. wt. chains. This
dynein has bieen cc onisaidieme dat cam pea
component of inner arms. We further
investegated thie) vin te ractaloms) oases
dynein with microtubules polymerized from
calf brain tubulin.

C/A dynein adsorbed onto the glass
surface was able to translocate the
microtubule toward its plus end. The
velocity of microtubule movement induced by
C/A dynein in the presence of 1 mM ATP was
4.0 to 4.5 um/sec at 25 °C. C/A dynein was
capable of binding and) bitndiamcosrhe
microtubules even in the presence of ATP.
Furthermore, ATPase activity of C/A dynein
was enhanced up to several-fold by the
microtubules at concentration as low as 1
mg/ml. On the other hand, 21S dynein which
composes outer arms, did not bind to
microtubules in the presence of ATP and its
ATPase activity was not activated by the
microtubules. From these results, it is
concluded that the affinity to microtubules
of the inner arm dynein is different from
that of the outer arm dynein.

BI 60

SLIDING MOVEMENT OF MICROTUBULES FROM
THE BUNDLE OF DYNEIN MICROTUBULES COMPLEX
Y.Mimori & T.Miki-Noumura

Dept of Biology,Ochanomizu University
Tokyo 112

We studied sliding behavior of microtu-
bules(MTs) from the bundles of dynein
MTs complex, using Tstrahymena ciliary
dyneins, 14S and 22S, and singlet MTs
re-assembled from porcine brain PC-
tubulin.

Bundle formation waS done in _ two
ways.After dynein was added to polymerized
MTs, dynein-MTs comp lex bound side
by side to form bundle,which were designa-
ted here as "MTs-D bundle".On the other
hand, after dynein were added to the
unpolymerizedtubulin solution at 0OT,the
solution was incubated at 37C. £=The
bundle was named as _ Tub.-D _ bundle.
Addition of 1mM ATP induced the MTs-
14SD bundle to separate into individual
MTs, without sliding movement. When
lmM ATP was added to MTs-22SD bundle,the
MTs slid out unevenly from the bundle(2~8
um/s:lmM ATP). On the other hand,
MTs evenly slid out from Tub.-D bundle
GQS528inm/s byel4S pe2e bye22oNre

Analysis with SDS-PAGE indicate that
dynein binding was ATP sensitive and
insensitive at a ratio 1:1 in MTs-D
bundle, although dynein in Tub.-D bundle
bound only in ATP insensitive manner.This
suggest that difference of sliding
behavior may be due to the dynein binding
property.

Biochemistry 1073

BI 61

CIRCULAR SLIDING MOVEMENT IN VITRO OF
DOUBLET MICROTUBULES INDUCED BY DYNEINS
OF TETRAHYMENA CILIA

M.Mori & T.Miki-Noumura

Doublet microtubules (DMts ) of
Tetrahymena cilia was reported to have
the appearance of a circular arc in
solution (Takasaki & Miki-Noumura, 1985).
Using ciliary dyneins, 14S, 22S and the
DMts, we employed an in vitro assay

system for sliding movement of DMts
along dyneins coating glass’ surface,
to characterize the motile properties
of 14S and 22S dyneins.

Addition of ATP induced a circular
sliding movement of DMts by 14S dynein.
Although 14S dynein was reported to
induce rotation of singlet Mts, rotation
of DMts was not observed. Velocity was
found to be 1.8pm/sec at pCa 9 and
QO pm/sec at pCa 6 in the presence of
O.1 mM ATP. The 22S dynein did not
induce sliding movement without ' the
presence of Nonidet P-40. A _ maximal
velocity of 3.4 ym/sec was observed at
50 uM ATP, while the movement’ was
strongly inhibited in the presence of
higher concentration of above 0.5 mM
ATP. Based on the ratio of ATP-
insensitive and sensitive binding of
dyneins to DMts analyzed with SDS-PAGE,
we are now trying to study the
difference of sliding behavior induced
by 14S and 22S dyneins.

BI 62

CHARACTERIZATION OF MULTIPLE 13S DYNEINS IN
SEA URCHIN FLAGELLAR AXONEMES

Y. Y. Toyoshima and T. Miki-Noumura, Dept. Biology,
Ochanomizu Univ. Tokyo

Sea urchin sperm flagellar dyneins consist of multiple
heavy chains which presumably have different motile
properties. We reported previously that 13S dynein
preparations were fractionated into 3 major peaks on the
mono Q column. The two peaks containing A and D
bands showed rotational as well as translational move-
ments. They also exhibited markedly different and
unexpected ATP dependence.

To isolate A and D band dyneins, hydroxyapatite col-
umn chromatography was tried. The 1st and 2nd dynein
peak consist mostly of D band, devoid of A band dynein;
the third dynein peak contains A band. Consistent with
Our previous results, the 1st and 2nd peaks supported the
translational and rotational movement below 0.2 mM ATP
but stuck the microtubules above that concentration; the
third peak supported only translational movement above
0.1 mM ATP but dissociated the microtubules below that
concentration. The first peak showed microtubule-acti-
vated ATPase activity whereas the third peak showed little
activation. In solution, D band dyneins bound to micro-
tubules in the absence and presence of ATP, whereas few
A band dynein in the third peak bound to microtubules.
These different properties may have distinct roles in coor-
dinating sliding movement in flagellar and ciliary beating.

BI 63

CLONING OF DYNEIN B-HEAVY CHAIN cDNA USING A
EXPRESSION VECTOR.

K. Ogawa. Dept. of Cell Biology, National Institute for Basic
Biology, Okazaki.

We constructed a cDNA library to poly(A)+RNA from the sea
urchin unfertilized eggs using Agtl1 expression vector and
screened it by affinity-purified anti-sperm dynein heavy chain
antibody. Six positive recombinants were obtained and their
epitopes were mapped on 6-heavy chain. Two recombinants
(J296 and A103) were extensively studied in the present work.
Northern blots probed with labeled EcoRI fragments of both
recombinants resulted in the hybridization to poly(A)+RNA of
about 16kb. However, the hybridization signals were smear.
Whole insert cDNAs were sequenced. The J296 cDNA (4.5kb)
contained 3'-non coding region and poly(A)-tailing signal,
indicating that the coding region is the carboxyl terminal portion
of a protein. The A103 cDNA (3.8kb) contained N-terminal
amino acid sequence of f2 component of fragment A
(QQVAPLQANEVAT) (Ogawa 1990) and ATP-binding
consensus sequence GKXXXGKT. Therefore, we concluded
that the A103 cDNA truly codes for the f2 region of dynein 6-
heavy chain.

BI 64

NOVEL MICROTUBULE ASSOCIATED PROTEINS PU-
RIFIED FROM SEA URCHIN EGGS. 1.24K protein.
S.Maekawa!, M.Toriyama2, M.Mishima! and H.
Sakai!l. IDept. of Biophys. and Biochem.,
Fac. of Sci., Univ. of Tokyo, and 2Dept. of
Biol., Fac. of Liveral Arts, Univ. of
Shizuoka, Shizuoka.

Chromatographic separation of sea urchin
egg proteins on a hydrophobic column enabl-
ed us to find a new protein that tightly
bound to the column and was eluted under an
alkaline condition. This protein, which
was purified through several column chroma-
tographies, was a basic and dimeric protein,
and showed an apparent molecular weight of
24,000 in SDS-PAGE(24K protein). Immuno-
cytochemical studies using both polyclonal
and monoclonal antibodies against the 24K
protein revealed that this protein was
localized to the mitotic apparatus. Bio-
chemical studies using the purified protein
showed its direct binding to microtubules
reconstituted from tubulin prepared from
starfish sperm outer fibers. A dose depen-
dent activity of this protein to promote
tubulin polymerization was also observed.
Immunoblotting analysis showed that this
protein is recovered in a microtubule frac-
tion obtained from a crude extract using
taxol. From these results, we concluded
that the 24K protein is a newly identified
microtubule associated protein in sea ur-
chin eggs.

1074 Biochemistry

BI 65

NOVEL MICROTUBULE ASSOCIATED PROTEINS
PURIFIED FROM SEA URCHIN EGGS. |. 34K
PROTEIN.

M.Toriyama!, S.Maekawa? and H.Sakai’.
A DeptinvOt (4B Os ehaG.m OF ailaiberal AGES -
Univ. of Shizuoka, Shizuoka and ?Dept. of
Biophys. and Biochem., Fac. of Sci., Univ.
of Tokyo, Tokyo.

Besides the 24K microtubule-associated
protein, a protein having an apparent
molecular weight of 34,000 in SDS-PAGE was
also purified from the alkaline eluate of
the hydrophobic column (34K protein). This
protein waS a dimeric protein and had
alkaline isoelectric point. The 34K
protein bound strongly onto assembled
microtubules but showed no promoting
activity of microtubule assembly. Western
blotting analysis using affinity purified
antibody aganist the 34K protien showed
its presence in the crude extract and in a
microtubule protein fraction prepared in
the presence of taxol. The cellular
content of this protein was relatively
small compared to the 24K protein. Cell
staining experiment showed that the 34K
protein localized to the mitotic apparatus
of metaphase eggs. Observation using a
confocal microscope (Bio-Rad MRC-500)
revealed its co-localization with tubulin
more clearly.

BI 66

INTERACTION OF TUBULIN WITH MICROTUBULE
ORGANIZING GRANULES IN THE CENTROSOME OF
SEA URCHIN EGGS.

K.Ohta, S.Hosoda, S.Endo, and H.Sakai.
Dept. of Biophys. and Biochem., Fac. of
Sci., The Univ. of Tokyo, Tokyo.

The centrosome is a microtubule organiz-
ing center which governs the microtubule
organization in vivo. Little is known about
the interaction of tubulin with the centro-
some. In order to examine this point, we
developed a new method to analyze the inter-
action between tubulin and reconstructed
microtubule organizing granules (MTOGs) which
is capable of nucleating microtubules in
vitro in the same manner as does the centro-
some isolated from mammalian cultured cell.
After incubated with tubulin, reconstructed
MTOGs/tubulin complexes were cross-linked
by dithiobis (succimidylpropionate) , sedi-
mented, and dissolved by adding 2-mercapt-
ethanol. Bound tubulin was quantified by
densitometer after SDS-PAGE. Dissociation
constant of tubulin in the reaction of the
complex formation was 1.1 uM. The binding
was generally noncooperative. At the
concentration of tubulin reacting satura-
tion, molar ratio of tubulin to the 51 kDa
protein, which is a major component of
MTOGs, was 1:3.8.

BI 67

IDENTIFICATION OF PROTEINS THAT BIND TO THE
51-kDa PROTEIN IN THE MITOTIC APPARATUS.
S.Hosoda, K.Ohta, and H.Sakai. Dept. of
Biophys. and Biochem., Fac. of Sci., The
Univ. of Tokyo, Tokyo.

The 51-kDa protein is localized to the
centrosome and spindle, possibly playing
an essential role with other minor compo-
nents in the formation of mitotic apparatus.
In order to identify such minor components
which interact with the 51-kDa protein, we
prepared an affinity column using mono-
clonal antibody, HP-1. The complex of the
51-kDa protein and minor components which
bound to the column was eluted by high ionic
strength solution containing 50% glycerol.
We identified 80-kDa doublet and 45-kDa
polypeptides as protein components that bind
to the 51-kDa proteins. Indirect immuno-
fluorescence staining with anti-80-kDa poly-
peptide antibody showed that the 80-kDa
polypeptide was localized to the centrosome
and the spindle. On the other hand, the
45-kDa polypeptide co-migrates with apoly-
peptide that was co-eluted with the 51-kDa
protein from GTP affinity column, suggest-
ing that these two polypeptides are identi-
cal. Furthermore, the 45-kDa polypeptide
was suggested to be a protein which regu-
lates GTP binding activity of the 51-kDa
protein.

BI 68

STUDIES ON TUBULINS INVOLVED IN CILIA AND
FLAGELLAR AXONEMES: III. CROSS-LINKING
BETWEEN PROTEINS AND TUBULINS.

K. Nakamura and E. Masuyama

Department of Living Sciences, Hiroshima
Women's University, Ujina, Hiroshima.

Tubulin, a major axonemal protein, con-
sists of two subunits (qand@tubulins).
Each tubulin subunit is a heterogeneous
protein composed of a mixture of iso-
tubulins. In order to examine the biologi-
cal implication of isotubulins in ciliary
movements of Tetrahymena, studies were
made with three kinds of cross-linking
agents; EDC (1-ethy1-3-(3-dimethylamino-
propyl) carbodiimide, 4PDS (4,4'-dithiodi-
pyridine), and OIBA (o-iodosobenzoate).
Among these, OIBA is the most useful agent
for the present study since it induces
disulfide bonds between tubulins and some
polypeptides and the cross-linkings are
reversibly reduced by 2-mercaptoethanol.
Two dimensional electrophoresis in which
axonemes treated with OIBA was first elec-
trophoresed and then run after treatment
with 2-mercaptoethanol showed the presence
of the tubulin-protein interactions. It
was further found that dynein crosslinked
with tubulin in the presence of OIBA to
some extent. Identification of tubulin
isoforms involved in these crosslinkings
is now under investigation.

Biochemistry 1075

BI 69

DETECTION OF THE HYDROPHOBIC INTERACTION
BETWEEN MICROTUBULE-ASSOCIATED PROTEINS
AND MICROTUBULES.

35 Kotanil, G. Kawai“, Ss. Yokoyama’, and
H. Murofushi®. Inept. Biochem. Eng. &
Sci., Kyushu Inst. Tech., Iizuka, 2Dept.
Material Eng., Yokohama Natl. Univ., Yoko-
hama, and 3Dept. Biophys. & Biochem., Fac.
Sci., Univ. Tokyo, Tokyo.

A synthetic peptide of 23 amino acid
residues (tubulin-binding peptide; TBP)
was active in binding to tubulin and
inducing its assembly (Aizawa et al., J.
Biol. Chem. 264, 5884-5890). The TBP-
tubulin interaction was analyzed by proton
nuclear magnetic resonance spectroscopy as
a simplified model for MAP-microtubule
interactions. Strong binding of two Val
residues of TBP to tubulin was detected.
Among aromatic residues of tubulin, Tyr
residue(s) was suggested to be involved in
the binding with TBP. Evidently, hydro-
phobic interactions between Val and Tyr
residues are important for TBP-tubulin
binding. Hydrophobic interactions have
not been taken into account previously in
the widely-accepted electrostatic model
for MAP-microtubules interaction.

BI 70

INFLUENCE OF SYMBIONTS ON THE GROWTH OF PEA
APHIDS MAINTAINED ON ARTIFICIAL DIETS.

T. Sasaki and H. Ishikawa.

HOO, IMSeg, WAS Oie Sess, Wiely74 Oi IWeli<xo)

In an effort to clarify what roles the
intracellular symbionts play in the nitro-
gen metabolism of aphids, symbiotic and
aposymbiotic pea aphids, Acyrthosiphon
pisum, were maintained on two kinds of
artificial diets with different amino acid
composition. Diet A was designed so that it
had the similar amino acid composition to
phloem sap. On this diet, 92% of symbiotic
aphids grew to adults, while 77% of aposym-
biotic aphids failed to grow. The amino
acid composition of diet B was after that
of the total tissue protein of the aphid.
More than 90% of aphids, whether symbiotic
Or aposymbiotic, developed well on this
diet. While symbiotic aphids were fertile
on the both diets, aposymbiotic ones pro-
duced progeny only on diet B. As far as_ we
know, this is the first report that apo-
symbiotic aphids are capable of producing
progeny. The capability of amino,acid syn-
thesis was investigated with N-labeled
glutamine. Examining the proteinic amino
acids after 24h incorporation on diet B, we
found heavy nitrogen in the following amino
acids in both symbiotic and aposymbiotic
aphids: Asp and/or Asn, Glu and/or Gln,
Seis Gillis Aare anG 9PrOn Has eAisG),= Vale, illic
and/or Leu, Phe, and Lys were labeled ex-
clusively in symbiotic aphids, suggesting
that these amino acids were produced with
the aid of the symbionts.

BI 71

CLONING AND SEQUENCE ANALYSIS OF THE 18S
RIBOSOMAL RNA GENE OF APHIDS.

O.Y.Kwon, K.Ogino, H.Fujiwara and
H.Ishikawa. Zoological Institute, Faculty
of Science, University of Tokyo, Tokyo.

An EMBL4 recombinant phage which encodes
one of the full length of the aphid
ribosomal RNA gene (rDNA) has been isolated
from the aphid genomic library. In order
to analyze the structure of aphid 18S rDNA,
EcORI-xbalI, EcoRI and xbaI-BamHI fragments
were subcloned into the plasmid Bluescript.
The sites of initiation and termination of
the aphid rDNA were estimated by S1 mapping
and comparison of nucleotide sequence of
different organisms previously examined.

Determination of the complete nucleotide
sequence of the aphid 18S rDNA revealed
that it is 2,586bp long with a G+C content
of 58%, very close to that of rat (55.6%),
frog (53.8%), and yeast (45%) and about 803%
homologous to them in sequence.

We are now trying to construct the
secondary-strucutre model of eukaryotic 18S
rRNA according to the model of eukaryotic
18s RNA established previously.

BI 72

MOLECULAR STRUCTURE AND PROPERTY OF
SYMBIONIN, A MAJOR PRODUCT OF AN APHID
ENDOSYMBIONT IN VIVO.

K.Kakeda, H.Ishikawa. JZool.Inst., Fac.of
Sci., Univ.of Tokyo, Tokyo.

An intercellularsymbiont harbored by the

pea aphid synthesizes virtually only one
protein species, symbionin that is supposed
to play a key role in the symbiosis with
the aphid. By analysis of its N-terminal 40
amino acid sequence, molecular mass, native
form and pl, symbionin has been revealed to
be very Similar to the E. coli groEL
protein, a heat-shock protein and is
supposed to be a member of "chaperonin"
family.

To further characterize symbionin, we
improved the method of its purification
considering its high molecular weight using
sucrose density gradient centrifugation. In
this method, we were successful in
purifying symbionin much faster with a
yield 20 times more than before. Further
determination of the amino acid sequence of
symbionin revealed that part of the
molecule is not evolutionarily concerved.
No protein of the aphid or its isolated
endosymbiont was cross-reactive to
polyclonal rabbit antibodies raised against
the groES protein, which, in E. coli, co-
operates with the groEL protein. Other
chaperonins, the Rubisco subunit-binding
protein of chloroplast and mitchondrial
hsp60 are not shown to cooperate with
groES-like proteins, either. Like other
chaperonins, symbionin was shown to have
ATPase activity.

1076 Biochemistry

BI 73

STRUCTURE OF THE GENE ENCODING SYMBIONIN,
AN APHID ENDOSYMBIONT-SPECIFIC PROTEIN.
C.Ohtaka and H.Ishikawa. Zool. Inst. Fac.
of Sci., Univ. of Tokyo, Tokyo.

We have previously reported that
symbinin is a homolog of the groEL protein
which is a major heat-shock protein of
E. coli, Although synthesis of the groEL
protein is augmented by heat-shock, the
symbionin synthesis in vivo is stopped by
heat-shock.

To elucidate the molecular mechanism of
this responce and the function of symbionin,
we analyzed the gene encoding symbionin
using DNA fragments from the groE operon
of E. coli as probes. By genomic southern
hybridization of the symbiont DNA and
screening of its genomic library, we have
detected and subsequently cloned two DNA
fragments that together encompass’ the
entire symbionin-coding region and its
flanking sequences. So far, we have been
successful in determing the nucleotide
sequence of the 3'-half of the coding
region.

According to the result, similarity of
symbionin to groEL protein at nucleotide
level was about 74%. Compared with the gro-
EL gene, the symbionin gene was characte-
ristic in its high A/T content. Most of the
synonymous base substitutions were also
biased toward the increase in A/T, suggest
ing that the codon usage in the symbionin
gene had been modified.

BI 74

HEAT SHOCK PROTEIN SYNTHESIZED BY PEA APHID
ENDOSYMBIONTS IN VITRO.

M.Morioka & H.Ishikawa. Zoological Inst.,
Fac. of Science, Univ. of Tokyo, Tokyo.

We have proposed the hypothesis that the
endosymbiont harboured by the aphid
mycetocyte is under "stress condition" of a
certain kind, based on the finding that the
symbiont selectively synthesizes symbionin
in vivo, which is evolutionarily related to
one of the heat shock proteins of E. coli,
groEL .

In the present study, we demonstrated
that the symbionts are also capable of
synthesis of symbionin in vitro at normal
temperature. Upon the temperature shift-up,
syntheses of, at least, three heat-shock
proteins (Mr: 63kDa, 43kDa and 72kDa)
become evident within several minutes. One
of them, named HSP63, had an apparent
subunit molecular mass of 63,000 and
determined by SDS-PAGE, sedimented with an
apparent molecular mass of 800,000 on
sucrose density gradient centrifugation,
and crossreacted with the polyclonal
antibody raised against purified symbionin,
suggesting its close similarity to
symbionin. Further evidence suggested that
symbionin is phosphorylated, probably by
autophosphorylation, under both in vitro
and in vivo coditions. It is likely that
the symbionts, when exposed to the heat-
stress, synthesize phosphorylated symbionin.
Studies on the biological role of the auto-
phosphorylation of symbionin are now in
progress.

* E.Hara et al. Symbiosis, 1990(in press)

BI 75

VITELLOGENIN SYNTHESIS AND EFFECTS OF
JUVENLE HORMONE IN THE TICK ORNITHODOROS
PARKERI .

DeMar Taylor, Ken Miura, Katsuhiko Ando
and Yasuo Chinzei, Dept. Med. Zool.,
School of Medicine, Mie Univ., Tsu.

This study was undertaken to character-
ize vitellogenenin (Vg) and determine the
role of juvenile hormone (JH) in the
regulation of vitellogenesis in the tick
Ornithodoros parkeri. Ticks usually
require a blood meal to induce vitellogen-
esis. However, we have shown that a
pyrethroid, cypermethrin (CyM), can stimu-
late Vg synthesis in unfed females. Vg
properties, concentration and synthesis
were analysed by SDS-PAGE, Spot-scanning
and fluorography using S-methionine.
Although unfed females show high titers of
Vg in the hemolymph, this is not due to
new synthesis. Vg synthesis stimulated by
engorgement increases beginning on day 2
after engorgement and reaches a maximum
level on day 8. Vg is synthesized in the
fat body, excreted into the hemolymph and
then processed and incorporated into the
Ovaries as vitellin. JH I, II and III,
methoprene (JHA), and CyM were topically
applied to unfed females and Vg synthesis
analyzed on day 5 by fluorography. JH and
JHA did not stimulate Vg synthesis. CyM
stimulated Vg synthesis but not ovarian
development. These results indicate that
JH does not function in the regulation of
vitellogenesis in this species.

BI 76

LARVAL CUTICLE PROTEIN OF THE SILKWORM
BOMBYX MORI STRUCTURE AND EXPRESSION.
K.Shofuda, H.Nakato, S.Izumi and S.Tomino
Dept. Biol. Tokyo Metropol. Univ., Tokyo

Biosynthesis of a larval cuticle
protein, termed LCP3 of Bombyx mori was
studied. Immunoblot analysis with an
anti-LCP3 antibody revealed that LCP3
first appears in epidermis of larvae 10
hr after hatch and is detectable
throughout the larval stages. It
disappears during the pupal stage, but
becomes weakly detectable again in adult
cuticle.

A cDNA expression library was con-
structed from the epidermal RNA of the 5th
instar larvae and 3 clones encoding LCP3
mRNA sequence were isolated. The result of
RNA blot analysis reflected well the
developmental profile of LCP3 in cuticle,
indicating that the biosynthesis of LCP3
is regulated at the level of mRNA. Topical
application of JH to the 5th instar larvae
inhibited the stage-dependent decline of
the LCP3-mRNA in epidermal cells, inplying
that JH acts as an inducer of the LCP3
gene transcription.

Primary structure deduced from nucleo-
tide sequence of LCP3 cDNA was highly
homologous to LCPs of M. sexta and D.
melanogaster. However, little homology was
observed between LCP3 and pupal cuticle
protein of B. mori.

Biochemistry 1077

BUT

ISOLATION OF cDNA FOR SEA URCHIN COLLAGENS.
M.Tomita, K. Shimizu, So Izumi, Si

Tomino, Tokyo Metropol. Univ., Tokyo.

K. Yoshizato, Hiroshima Univ., Hiroshima.
Biochemical nature of collagens of sea

urchin has been poorly understood.
Collagens of sea urchin were isolated from
the test of sea urchins, Hemicentrotus
purcherrimus and Strongylocentrotus

purpuratus, and used for analysis of
a-chains and for cloning of their cDNA to
compare with the known characters of
Astenosoma ijimai(sea urchin) and mammals.
Collagens of both species of sea urchin
showed similar patterns of a-chain
components on SDS-PAGE which are quite
different from patterns of A.ijimai and

mammals. In vitro culture of H.
pulcherrimus test provided evidence that
collagens are synthesized in test. cDNA

expression library in <Agtll vecter was
constructed from poly(A)*RNA of the H.
pulcherrimus test. Screening of the
library with an antibody against the
denatured H.pulcherrimus collagens yielded
seven positive clones. Among clones’ two
clones, denoted as tHpcol-3 and ’tHpcol-7,
were characterized in detail. These
clones containd the 2.4kbp and 2.6kbp
cDNAs, respectevly. The 8s-galactosidase/
collagen fusion proteins were sensitive to
bacterial collagenase. RNA blot analysis
using tAHpcol-7 as a probe demonstrated the
occurrence of collagen-mRNAs complementary
to the %AHpcol-7 cDNA in the test of sea
urchin.

BI 78

SYNTHESIS AND GENE EXPRESSION OF THE MIDGUT
ENZYMES DURING EMBRYONIC DEVELOPMENT OF THE
SILKWORM, . Bombyx mori 1 1
T.Suzuki~, M.Azuma“, M.Itoh and M.Eguchi

1:Dept.Appl.Biol.,Kyoto Inst.Tech. ,Kyoto

:Fac.Agri.,Tottori Univ. ,Tottori

In relation to the formation of midgut
tissue during embryonic development of the
silkworm, we studied the appearance and
change of larval type-alkaline proteases
(BA-p; trypsin-like, 32K and Ca-p;
chymotrypsin-like, 31K) and alkaline
phosphatases (m-ALP; 58k and s-ALP; 61K).
In the course of our study, we found two
embryo specific proteins, 36K BA-p and 81K
s-ALP. The 36K BA-p was observed in the
abdominal ganglion only in 8 day embryo.
The in vitro translation using RNAs from
the eggs(8 day) and hatched larvae showed
the synthesis of 42K protein for the 32K
and 36K BA-p, suggesting that the post
translational processing is different in
the midgut and abdominai ganglion. The 81K
s-ALP is conceived to be a precursor of the
active 61K s-ALP from the in vitro
translation and the test with s-ALP
deficient mutant. Subsequently, we
performed the cDNA cloning of these midgut
enzymes to examine the gene regulation
during embryonic development. We obtained
several clones including the insert of 1.6-
2.1Kb of m-ALP cDNA. Northern dot
hybridization analysis showed that the
expression of m-ALP gene begins at least at
1-2 days before hatching corresponding with
the appearance of the enzyme protein.

BI 79

GENE EXPRESSION OF aB-CRYSTALLIN mRNA
IN RAT SKELETAL MUSCLE AND THE
INFLUENCING FACTORS
Y, ATOMI and S. YAMADA Dept. of Sports Sciences, College
of Arts and Sciences, Univ. of Tokyo, Tokyo

We identified the specific protein component of 22-kDa,
which decreased in atrophied soleus muscle by rat hindlimb
suspension model (HS) by one- and two- dimentional

polyacrylamide gel electrophoresis, with the B chain of a-

crystallin. Although a-crystallin had been thought to be
expressed in only eye lens, we have found that the B chain
expressed in nonlenticular tissues, especially in slow muscle and

heart. The decrease of aB-crystallin in atrophied muscle was

marked in slow twitch muscle. cDNA of aB-crystallin was
cloned from rat heart cDNA library, about having 0.7kb
corresponding the coding region and 3' uncodiung resion. By

northern blot analysis, the expression of MRNA of aB-crystallin
in these muscles, and the variability in the time course of
atrophy, and the effect of passive stretch and denervation were

studied. a@B-Crystallin mRNA was expressed more in slow
muscle than fast muscles. After 3 days of HS the mRNA markly
decreased and kept to be depressed through 4 weeks of HS.
With HS scietic nerve denervation mRNA in soleus muscle less
expressed compared with that of control, while that in fast
muscle with denervation markedly increased. The passive
stretch of the muscles with tape under HS increased the mRNA
in both muscle type and both with and without denervation. The
decrease of mRNA in atrophied soleus in HS model seems to be
specifically regulated.

BI 80

STRUCTURE AND DEVELOPMENT OF PEPSINOGENS
T.Kageyama’ , K.Tanabe“~, and O.Koiwai
Primate Res.Inst., Kyoto Univ., Inuyama,
and “Aichi Cancer Center Res.Inst., Nagoya

Purification and molecular cloning of
pepsinogens were performed at various deve-
lopmental stages of rabbit and monkey. In
rabbit, pepsinogens were classified into
several components. Fetal- and newborn-
specific zymogens, pepsinogens F and M,were
demonstrated. They were replaced at a later
time by adult-specific pepsinogens. In mon-
key, five pepsinogens were isolated and
they were classified into pepsinogens I,III
-1, III-2, and III-3, and progastricsin.

In contrast to rabbit, fetal-specific pep-
sinogen was not evidenced and pepsinogens
in group III expressed sequentially during
development. The predominant zymogen at
fetal and newborn stages was pepsinogen
III-1 and that at a later time pepsinogen
ITII-3. Structure of each monkey and rabbit
pepsinogen was elucidated by molecular clo-
ning of cDNA. Each was composed of a leader
region (signal peptide), a pro-gegion(acti-
vation segment), and a pepsin region. Two
models of the expression of pepsinogen genes
during development were proposed, which
focused on the expressional switching from
fetal to adult pepsinogens and the sequen-
tial expression of adult pepsinogens.

1078 Biochemistry

BI 81

STRUCTRE AND BEHAVIOR OF CHROMOSOMAL
FRAGMENTS IN MOTTLED MOSAIC STRAINS FOR
LARVAL BODY MARKING OF BOMBYX MORI.

H. Fujiwara, M. Yanagawa and H. Ishikawa.
Zoological Institute, Faculty of Science,
University of Tokyo, Tokyo.

It was reported previously that a
mottled striped strain (pSm) has a
chromosomal fragment 2.5 mb long carrying
the larval body marking gene. When studied
in two other mottled mosaic strains,
mottled zebra (Zem) and mottled new
striped (Sm), chromosomal frgaments
different from one found in pSm were
JaBeeereac lov Wwe Amc SoOwcMmerin
hybridization with specific probes,
chorion and sp-2 gene respectively. The
fragment in Sm is about 2.5 mb long and Ze
fragment is about 6 mb. Existence of
various chromosomal fragments induced by
X-ray irradiation, suggests that each
chromosome of Bombyx mori possesses an
exceptional type of centromere, a
"diffuse" centromere.

From the results of RFLP analysis on pS
fragment, it was shown that a fragmented
chromosome does not recombinate with a
original set of homologous chromosomes.
RFLP analysis also showed that mosaic
patterns of right and left on larval skin
of several half mosaics, are not agreement
with patterns of the presence of the
frgments in two posterior silkglands. This
result implies that original cells of
Ssilkglands differentiate earlier than
epidermis.

BI 82

CLONING AND SEQUENCING OF A cDNA ENCODING
THE MAJOR HEMOGLOBIN COMPONENT OF
PARAMECIUM CAUDATUM

M. Mukai!, K. Yamauchi!, T. Ochiai?.and I.
Usuki? --- !Dept. of Biol., Fac. of Sci.,
Shizuoka Univ., Shizuoka, 2Dept. of Biol.,
WAOo Ose Moro, eyoxcl Jeo Orie JNO 5 Coll
of Gen. Educ., Niigata Univ., Niigata

cDNAs encoding the major hemoglobin
(Hb) component of Paramecium caudatum was
isolated and sequenced. A 32-mer
oligonucleotide synthesized on the bases
of the amino acid sequence data (Iwaasa et
al., 1989) was used to screen a Paramecium
cDNA library. Three positive clones with
cDNA showing 0.4, 0.6 and 0.9 kilobase
pairs (kbp) were obtained, so they were
subcloned into pUC plasmid vector.
Nucleotide sequencing revealed that the
0.4 kbp cDNA encoded a full length of
Paramecium globin. The nucleotide
sequence contained four TAA codons in
reading frame. This codon is known to
encode Gln in ciliates (Caron & Meyer,
1985; Preer et al., 1985), although it
acts as one of the termination codons in
Many organisms. Comparison of the
deduced amino acid sequence with those of
other Hbs from various species showed
little overall homology, suggesting that
Paramecium Hbs diverged from ancestor at
very early period. In addition to this,
DNA blot hybridization suggested the
presence of a gene family for Hbs in
Paramecium caudatum.

BI 83

TREHALOSE, THE INSECT BLOOD SUGAR, INHIBITS
LOADING OF DIACYLGLYCEROL BY LIPOPHORIN
FROM THE FAT BODY IN LOCUSTS.

P.Y. Lum and H. Chino. Biochem. Lab., Inst.
of Low Temp. Sci., Hokkaido Univ. Sapporo.

During the course of in vitro sEudy sto
demonstrate the transformation of high-
density lipophorin (HDLp) to low-density
lipophorin (LDLp) by incubating hemolymph
with fat body in the presence of AKH, we
noticed that LDLp sometimes were formed
even in the absence of AKH. This
observation suggests the possibility that
a low molecular weight factor may be
present in hemolymph, which inhibits the
uptake of diacylglycerol (DG) by lipophorin
from the fat body, because LDLp has never
been found in vivo without AKH and the
hemolymph used in the in vitro experiment
was dialysed against Ringer prior to use.
Previous data on metabolism during flight
led us to deduce that the above mentioned
factor could be the insect blood sugar,
trehalose itself. Here, we report that
trehalose indeed inhibits DG loading by
lipophorin. Trehalose inhibited the DG
uptake by about 40-50 % at various
physiological concentrations (50-100 mM).
Trehalose was observed to have no direct
effect on the Neviel of DG ant aewbocyas,
Suggesting that trehalose affects the
DreOeCases Of DEG LOAaCAMG. Another
disaccharide, sucrose, exhibited no such an
inhibitory effect at all, suggesting the
inhibition may be specific for trehalose.

BI 84

COMPOSITION AND STRUCTURE OF COLORADO

POTATO BEETLE LIPOPHORIN

@. Katagiri, and, C.ALDa dem ho tame

Inst.) Of Low Memp. (Seal hokkan don ymenvger,

Sapponol, and (Dept. Emtomolle AG ta mcunlae
Univ., Wageningen, Netherlands.

Lipophorin, the high density lipoprotein
of insect haemolymph, has been purified
from the Colorado potato beetle,
Leptinotarsa decemlineata. It contains
about 50% protein and 50% lipids. The
protein moiety is composed of two
apoproteins, apolipophorin I (Mr 250000)
and apolipophorin II (Mr 78000), as in
other species. The lipid composition is
rather unusual compared with other
insects. It contains high amounts of
hydrocarbons and small amounts of
diacylglycerol. However, it can take up
considerable amounts of diacylglycerol
when, incubated —in) vitro meoge Enema iwasen
OIG WShts fat body. Thus, its low
diacylglycerol content is probably related
to the peculiar energy supply of the
flight muscles in this beetle.

Using small angle X-ray scattering
analysis, the molecular structure of this
lipophorin was studied. We propose that
Colorado potato beetle lipophorin is
composed of three radially symmetric
layers.

Biochemistry 1079

BI 85

FURTHER CHARACTERIZATION OF LOCUST LOW
DENSITY LIPOPHORIN INDUCED BY ADIPOKINETIC
HORMONE.

E. Nagao and H. Chino. Biochem. Lab., Inst.
of Low Temp. Sci., Hokkaido Univ. Sapporo.

In order to resolve some basic questions
concerning the nature of adipokinetic
hormone(AKH)-induced low density lipophorin
(LDLp), lipophorin was fractionated after
density gradient ultracentrifugation and
CACIN BeAGeELOM WAS Eide ec is ©a6
diacylglycerol(DG) content, apolipophorin-
III (apoLp-III), and also subjected to
electron microscopy (EM). The contents
OE IDE AidCl EjOO MH IILIL sbyol IID)Ijjo) IESE 1eaLOinS}
increased significantly as the density of
the fraction decreases, whereas the DG
content of high density lipophorin (HDLp)
remains almost constant irrespective of the
IE IGE SALONA However, EM observation showed
that LDLp particles in each fraction were
extremely heterogeneous in size with,
whereas HDLp particles were very
homogeneous in size irrespective of the
fraction. A theoretical analysis based on
all experimental data strongly supports the
hypothesis that the heterogeneity in the
size observed for LDLp particles does not
reflect different densities, but rather,
the heterogeneity is resulting from
intermolecular fusion between LDLp
particles of the same density. We believe
that such a heterogeneity in size is the
most distinct characteristic of locust
LDLp.

BI 86

RECONSTITUTION OF INSECT LIPOPHORIN
-APOLIPOPHORIN II IN LIPOSOME-

T. Hiraoka and C. Katagiri

Inst. of Low Temp. Sci., Hokkaido Univ.,
Sapporo

Lipophorin, a major insect hemolymph
lipoprotein, serves to transport diacyl-
glycerol, an energy source, from fat body
Ona aght = musicilker: Lipophorin has two
apoproteins, apolipophorin I (apoLp-I,
MW=250k) and apolipophorin II (apoLp-II,
MW=85k). An anti-apoLp-II monoclonal
antibody has been reported to inhibit
diacylglycerol uptake from fat body by
lipophorin, which suggests that apoLp-II
contributes to diacylglycerol loading by
lipophorin. In the present study, we
isolated apoLp-II from 6M guanidinium
chloride-treated lipophorin by gel filtra-
tion. The isolated apoLp-II is water-
soluble and free from lipids. The purified
apoLp-II was then incorporated into lipo-
some of dipalmitoylphosphatidylcholine by
ultrasonication. This reconstituted lipo-
phorin made of apoLp-II and phospholipids
was further examined by electron microsco-
py and circular dichroism spectropolarime-
try. The uptake of diacylglycerol from fat
body by reconstituted lipophorin was also
demonstrated.

BI 87
SERUM LIPOPROTEINS IN THE CRAYFISH,

PACIFASTACUS TROWBRIDGI1.

Y. Hirayama, N. Hokawa, H. Hase. Dept. of
Chemistry, Hokkaido Univ. of Education,
Kushiro, Kushiro.

Serum lipoproteins of the crayfish
were prepared by precipitation at low
ionic strength. This precipitation
reaction was specific to the serum
lipoproteins. Using this’ fraction we
studied the subunit structure and _ the
lipid composition of the crayfish
lipoproteins. This fraction contained
about 50% lipid. The major class was
Phospholipids (88-89% of the total). Small
amount of free sterol (6-8%) and
triglycerides (2-3%) and diglycerides(l-
2%), together with traces of hydrocarbons
were also detected. More than five
lipoprotein species could be resolved by
polyacrylamide gel electrophoresis. One
lipoprotein band was limited to female.
Other lipoproteins were common to both
sexes. Up to five apolipoproteins were
detected by SDS polyacrylamide gel
electrophoresis. Their molecular weight
were calculated to be 470, 430, 380, 310,
170kDa. The most major lipoprotein
contained mainly 170k apolipoprotein and
the second major lipoprotein contained
mainly 170k, 310k and 380k apolipoprotein
and the other lipoproteins contained all]
apolipoproteins

BI 88

CHANGES IN FATTY ACID COMPOSITION OF HAMSTER
CAUDA EPIDIDYMAL SPERMATOZOA DURING
CAPACITATION AND ACROSOME REACTION.
M. Awano, A. Kawaguchi and H. Mohri
Capacitation of spermatozoa in the female
reproductive tract is an indispensable
process for fertilization in mammals. This
process must occur prior to the acrosome
reaction. Hamster cauda epididymal spermato-
zOa were incubated for 2~4hr in a capacita-
tion medium(m-TALP-6) in vitro at 37°C.
After the incubation, aliquots were taken
from the sperm suspension for fatty acid
analysis. Fatty acid metyl esters were
extracted with hexane and quantitated on
gas chromatograph equipped with a Thermon
3000 Capillary column and temperature-
programmed from 180-220°C at 2°C/min.
There was no change in fatty acid composi-
aoa Clotiealin~; Dinte Cie alinvelortiswony , Joie lic weeus
4hr incubation, the molar percentage of
ddcosapentaenoic acid(22:5) and docosa-
h@xaenoic acid(22:6) decreased and the
percentage of palmitic acid(16:0) increased.
Similar changes were observed in the
spermatozoa where the acrosome reaction was
induced with 1pM A23187. Electron micro-
scopical observations revealed that the
sperm membranes showed no conspicuous
changes after 2hr incubation, but were
remarkably altered after 4hr incubation
owing to the acrosome reaction. So it
appears that polysaturated fatty acids,
such as 22:5 and 22:6, are localized in the
acrosomal region of sperm head.

1080 Biochemistry

BI 89

ISOLATION AND CHARACTERIZATION OF A MAJOR
S IALOGLYCOPROTEIN OF
BUFO JAPONICUS EGG JELLY
Y.Shimoda!, K.Kitajima’, Y.Inouel, S. Inoue”
IDept. of Biophys. & Biochem., University of
Tokyo, Tokyo; School of Pharmaceut. Sci.,
Showa Univrersity, Tokyo

The presence of jelly surrounding the eggs

of toad is known to essential for fertiliza-
tion and subsequent early development to
occur. However, little is known about the

chemical nature of its components. We have
now isolated a major sialoglycoprotein from
Bufo japonicus egg jelly ( about 50% of the
dry jelly). The glycoprotein is carbohydrate-
rich (about 82%) and polydisperse (MW 1,000K
to 4,000K). Almost all the Thr and Ser
residues, which are accounting for 55 mol% of
the core protein, are O-glycosylated.
Composition, methylation, and 1-NVR analyses
of four different oligosaccharides obtained by
alkaline borohydride treatment (0.1 N NaQH/1 M
NaBH,, 37°C, 50 h) established the following
structures:
Ga1B1-3 (NeuAca2 -6 )GalNAcol ;
Fuca 1 -2Gal1B 1-3 (NeuAca2-6 )GalNAcol ;
GalB1-3GalNAcol ;
Fucal -2Ga1B1-3GalNAcol

BI 90

OCCURRENCE OF LECTIN IN VARIOUS TISSUES OF
THE SILKWORM, BOMBYX, MORI
K. Amanai,, M. Saka3', Se Sakuraj!,

- Ohtaki’', T. Mori~, O. Nikaido“.

Dept. Of Biola, hack Or (SCils, (UNIV. OF
Kanazawa, “Divi. of Radi., Fac. of Phar.,
Univ. of Kanazawa, Kanazawa.

Monoclonal antibodies against the 350kDa
lectin purified from larval hemolymph of
the silkworm, Bombyx mori, were prepared.
The antibodies inhibited the hemagglutinat-
ing activity of hemolymph 350kDa lectin.
Immunohistochemical observations revealed
the presence of lectin in the cuticular
intima of the anterior silk gland of fifth
instar larvae and prepupae. Extracts from
the anterior silk gland showed hemaggluti-
nating activity and exhibited the same
biochemical characteristics as the hemo-
lymph 350kDa lectin. We found immunoreact-
ing substances with different molecular
weight in the epidermal cells by Western
blotting analysis, but not in fat body
cells and in middle and posterior silk
gland. The lectin-like molecules in epider-
mal tissues may play an important role in
histolysis during metamorphosis.

BI 91

THE NATURE OF THE XANTHOMMATIN GRANULE IN THE
BLACK FEMALE OF THE SCORPION FLY, PANORPA JAPON-
ICA.

M. Nakagoshi, S. Takikawa and M. Tsusue.

Biol. Lab. Kitasato Univ. Sagamihara.

The scorpion fly displays a seasonal color dimorphism from
black spring-form to yellow summer-form. We already re-
ported that sepiapterin granules were located in the epidermis
of the fly and sepiapterin levels were higher in the yellow
scorpion fly than in the black one. However, only in black fly,
the color of the intersegmental membrane of the abdomen par-
tially becomes changing from yellow to reddish brown with
growing. This colored compound was isoiated from the integu-
ments of the flies. Based on redox behavior, TLC, UV-spectro-
photometric and HPLC analysis, the pigment was identified
as xanthommatin. Observations with the light microscope
showed that the rugged and reddish brown granules were also
located in the epidermis of the fly. Electron microscopy re-
vealed that the pigment granules were electron-dense and
mosaic organelles which fused with the limiting membrane of
the fibrous and elliptical granules like pterin granules. Cen-
trifugal fractionation with Ficoll showed that most of xanth-
ommatin granules were present in the fractions at the density
range of 7-8%. Sepiapterin was also detected in the same frac-
tions besides xanthommatin. On the contrary, sepiapterin
granules in the yellow fly were mainly present at the light
density range of 4.5-5.5% without binding with other granules.
These facts indicate that the nature of the mosaic chromo-
granules was the mixture of pterin and xanthommatin.

BI 92

INSOLUBLE-BOUND FORM OF 4-ALANINE DURING
THE PUPAL STAGE OF PAPILIO XUTHUS.

W. Ukai and Y. Umebachi. Dept. of Biol.,
Fac. of Sci., Kanazawa Univ., Kanazawa.

The abdominal body wall of P. xuthus was

prepared from the pharate adult during the
pupal stage and was examined for #-alanine.
The body wall was homogenized in 70 % eth-
anol and separated into a soluble fraction
and a insoluble fraction. The soluble fr-
action was analyzed before and after acid
hydrolysis. The former is a free form and
the latter shows a 70 % ethanol soluble-
bound form. The insoluble fraction was
divided into (1) 4% HCl-methanol (MeOH)
soluble fraction, (2) 1 N NaOH soluble fr-
action, and (3) residue. #@-Alanine of the-
se three fractions was determined after
hydrolysis. In addition, glucosamine (af-
ter hydrolysis) and protein were determin-
ed. The quantity of protein in NaOH frac-
tion is over 100 times larger than in the
HC1-MeOH fraction. The protein decreases
from the 6th to the 8th day. The free fo-
rm and 70% ethanol soluble-bound form of
@-alanine reach a Maximum at the 9th day
and then decrease. In accord with the de-
crease, 8-alanine in both HCl-MeOH fracti-
on and NaOH fraction increases and then
decreases a little. Residual #-alanine
gradually increases up to emergence. The
f-alanine of HCl1-MeOH fraction per protein
is over 100 times larger than in the NaOH
fraction. Glucosamine increases earlier

than #-alanine and rapidly from 6th day.

Biochemistry 1081

BI 93

ON THE ISOMERS OF PAPILIOCHROME II.
Y. Umebachi. Dept. of Biol., Fac. of Sci.,
Kanazawa Univ., Kanazawa.

Papiliochrome II is a main pale yellow
pigment of the wings of Papilio butterfli-
es. The structure was reported to be N@ -
[f-(3-aminopropionylaminomethy1)-3,4-dihy-
droxybenzyl]-L-kynurenine, which consists
of one molecule each of L-kynurenine and
N-f8-alanyldopamine (Rembold and Umebachi,
1984). There exist two isomers, IIa (543%)
and IIb (46%). Both IIa and IIb decompose
to L-kynurenine and N-#-alanylnorepinephri-
ne (NBANE) by being heated in water at 80°
€ for 30 min. The NBANE from IIa and IIb
both show a positive CD curve at 280 nm.
This suggests that the NBANE from Ila and
IIb have the same configuration. The NBA-
NE further decomposes to (+) norepinephrine
(NE) and @-alanine by being heated in 1N
HCl at 100°C for 60 min. By this procedu-
re, NE is racemized. Interestingly, the
IIa (or IIb) is converted to IIa and IIb
ate25°C an 0.1, N-HGlls inthe same procedu—
re, authentic (-)NE is also converted to
(£)NE. Furthermore, the IIa (or IIb) is
Converted sto. a mixture of ila, lib) eeliia,
and IIIb by being heated in 0.1 N HCl at
SO7eetor 50 min. The Tia, Lib, rita, and
IIIb all decompose to L-kynurenine and NBA-
NE by being heated in water at 80°C for 30
min. Although the II and III are probably
isomeric with each other, the stereochemi-
stry of III has remained unsettled.

BI 94

ENZYMES RELATED TO THE BIOSYNTHESIS OF
PAPILIOCHROME II.

Y. Ishizaki and Y. Umebachi. Dept. of
Biol., Fac. of Sci., Kanazawa Univ.,
Kanazawa.

Papiliochrome II is a aple yellow wing-
Pigment of Papilio xuthus, and the struct-
ure has been reported as N®@Y -[,-(3-aminopr-
opionylaminomethy1)-3,4-dihydroxybenzyl ]-
L-kynurenine. In the present report, seve-
ral enzymes related to the biosynthesis of
the pigment have been studied. The prese-
nce of tyrosinase (Tase), dopadecarboxylase
(Dase), and N-8S-alanyldopamine synthetase
(Aase) in wings were confirmed. The Tase
showed a high activity in both physiologi-
cal saline soluble fraction and insoluble
fraction. The Tase of soluble fraction pr-
ecipitates in 45% saturation of ammonium
sulfate. The Dase shows a high activity
in the soluble fraction and precipitates in
65% saturation of ammonium sulfate. The
activity is found throughout the pharate
adult stage, and not a little activity is
also found in other tissues than wings.

The Aase activity is found in the soluble
fraction of both wings and body integument
of pharate adult, but not found in both
haemolymph and fat body. In wings, the ac-
tivity was highest at the 8th day after pu-
pation. All these results suggest that N-
3-alanyldopamine is synthesized in wings

by way of dopa and dopamine from tyrosine
and combines with L-kynurenine to form the
pale yellow pigment.

BI 95

EFFECTS OF SUGARS ON GLUCOSE-DEPENDENT PROTEIN
KINASE IN HUMAN POLYMORPHONUCLEAR LEUKOCYTES.
M. Shibata!,.2, T. Ohoka2, S. Mizuno!, and K.
Suzukil. ‘4Dept. of Antibiotics, National Institute
of Health, Tokyo and 2Dept. of Biology, Fac. of
Sci., Tokyo Metropol. Univ., Tokyo.

Chemoattractants, such as lung giant cell
carcinoma-derived chemotactic protein/interleukin-8
and formyl-methionyl-leucyl-phenylalanine stimulate
Phosphorylation of 64-kDa protein (p64) in 32pP-
labeled human polymorphonuclear leukocytes (PMN)
and glucose is indispensable for the phosphoryl-
ation in a cell-free system2). Moreover, glucose
derivatives and its metabolites have not stimulated
the phosphorylation. Present study demonstrated
that glucose added to the assay mixture using over
30-Da fraction of cytosol as an enzyme source was
not metabolized during 1-min incubation by analysis
of thin layer chromatography. These results suggest
that a protein kinase involved in the phosphoryl-
ation is stimulated specifically by glucose. We
have referred to the kinase as glucose (dextrose)-
dependent protein kinase (D-kinase). The kinase was
purified by 86-fold from cytosol of human PMN.
Purified kinase was stimulated by glucose in a
dose-dependent manner (6 uM for ED50). Glucose de-
creased apparent Km value of D-kinase for ATP but
did not affect Vmax. Glucose-1-phosphate, glucose-
6-phosphate and fructose-6-phosphate (1 or 10 uM)
completely inhibited the kinase activity stimulated
by 30 uM glucose, and fructose-1,6-diphosphate (10
4M) inhibited by 30 to 40%. It may be inferred that
D-Kinase is positively and negatively controlled by
glucose and the hexose-phosphates in vivo.

Suzuki, K. et al. (1989) Biochem. Biophys.
Res. Commun. 163, 1298-1305. 2)Shibata, M. et al.
(1980) Immunol. Lett. 24, 159-164.

BI 96

SURVEY OF INTERLEUKIN-8 GENE IN HUMAN CELLS
AND ITS HOMOLOGOUS SEQUENCES IN ANIMAL CELLS.

'S. Suzukil2, kK. Hottal, H. Tsuchiyal, J.Ishikawal,

H. Mizusawa’, E. Kataoka’, A. Ishida-Okawaral,
S. Tomino?, S. Mizuno!, and K. Suzukil. 1Dept. of
Antibiotics, Natl. Inst. of Health, #Dept. of
Biol., Tokyo Metropol. Univ., and 3Cell Bank, Natl.
Inst. of Hygienic Sciences, Tokyo.

We have characterized polymorphonuclear
leukocyte chemotactic protein (LUCT/IL-8) and its
gene of human lung giant cell carcinoma LU65C
cells. LUCT/IL-8 has AVLPR at the N-terminus that
other known IL-8 proteins lack. Present study was
undertaken to investigate the genetic basis, if
any, for this structure difference. In addition,
DNAs from various animal sources were examined for
their sequences homologous to LUCT/IL-8 gene.
A 1.9kb sequence spanning the coding region of IL-8
gene was amplified PCR method employing two 30mer
oligonucleotide primers. Existence of the sequence
corresponding to AVLPR in the amplified fragments
was examined by Southern hybridization using a
40mer oligonucleotide as the LUCT/IL-8 probe. Of
153 human DNAS examined, 136 DNAs gave the 1.9 kb
amplified fragment which hybridizes with the
LUCT/IL-8 probe. This indicates that the sequence
of IL-8 gene is highly conserved in human cells so
that the difference in N-terminus of IL-8 proteins
may be due to protein-processing. Among 14 prepara-
tions of animal cell DNAs tested, amplification was
observed with those from mouse (1.5kb) and canine
(2.1 and 2.4kb). Southern hybridization showed that
only the 2.4kb canine fragment exhibited a weak
homology with LUCT/IL-8 probe.

1082 Developmental Biology

DB 1

FORMATION OF PRIMORDIAL GERM CELLS
STIMULATED BY LECTIN-AFFINITY GRANULES
DERIVED FROM FROG EMBRYOS AT THE
TAIL-BUD STAGE.

T. Shirane, Fac. of Integrated Arts and
Sci., Hiroshima Univ., Hiroshima.

In Rana nigromaculata, the lectins
Con A and GS-I have a specific affinity
for certain granules (1 to 5 pm in
diameter), which are distributed on the
ventral side of the mid-gut of the
embryos at the tail-bud stage. An
attempt was made to isolate these
granules from a granular fraction
derived from such embryos, using lectin-
coated beads. These granules caused the
proliferation of primordial germ cells
(PGCs) in other individuals at the
operculum-complete stage. In the case of
granules isolated by use of beads coated
with either lectin, all animals at the
operculum-complete stage that were
injected with the granules showed
evidence of proliferation of their PGCs.
The mean number of PGCs in the
experimental group was 32% higher than
that in the control group. This result
suggests that granules with an affinity
for lectins include a determinant of
germ cells.

DB 2

ORIGIN OF GERM CELLS AND EARLY DIFFEREN-
TIATION OF STARFISH GONADS.
C. Inoue and H. Shirai. Ushimado Marine

Laboratory, Okayama University, Okayama.

The cells in gonads after spawning
were easily classified into two types,
the smaller and the larger type. The
larger type, about 15m in diameter, had
nuclei about 7pm with electron-lucent
nuclearplasm and mitochondrial aggrega-
tions (MA) associated with electron-dense
bodies, here called nuage.

Animals 2cm in diameter had gonads
with germ cells (15pm) containing about
50 MA with nuage and nucleus (5-/pm).
Similar germ cells were detected in
genital branches of the haemal sinus.
Though animals about 0.6cm did not form
gonads yet, germ cells (15ym) containing
about 20 MA with nuage and nucleus (5-8pm)
were detected in the aboral haemal sinus
near the hydroporic canal (CH).

In juveniles 10 weeks after insemina-
tion (lmm) and brachiolaria 5 weeks after
insemination, germ cells (/pm) with
nucleus (4m) were detected near the HC
as a cellular aggregation and each cell
contained about 15 and 5 MA with nuage,
respectively.

These results indicate that the germ
cells are first detected as an aggrega-
tion of the primordial germ cells near
the HC, migrate to gonadal primordia
through the haemal sinus, and differen-
tiate into gonia.

DB 3
HISTOGENESIS AND SOMATIC CELL KINETICS IN
THE PRIMORDIAL GONAD OF Hynobius

nigrescens
Masahiko Kumakura and Hisaaki Iwasawa.

Biol. Inst., Niigata Univ., Niigata

The process of gonadal development in
H. nigrescens was observed ultrastructual-
ly, and somatic cell kinetics related to
gonadal histogenesis were examined auto-
radiographically. The genital ridge con-
sisted of coelomic epithelial cells and
Primordial germ cells. The gonadal medulla
was formed by the proliferation of the
cells derived from the coelomic epithelium
within the primordial gonad. Thereafter,
the labelling index of medullary cells de-
creased, and gonadal sex differentiation
occurred directly from the sexually indif-
ferent primordial gonad. In the rudimenta-
ry testes, epithelial cells in the corti-
cal region differentiated into seminifer-
ous tubules, and efferent ducts were form-
ed by the proliferation of the cells in
the medullary mass ( rudimentary testicu-

lar rete ). In the rudimentary ovaries, on
the other hand, the cortex and medulla
separated, the medullary tissue

degenerated, and the ovarian cavities were
formed in the medullary region. These re-
sults indicate that both the cortical and
medullary cells are derived from the co-
elomic epithelium, and that there is no
antagonism in the development of the cor-
tex and medulla in the process of sexual
differentiation. ;

DB 4

ULTRASTRUCTURAL CHANGES OF OVARIAN

MEDULLA CELLS IN THE LIZARD EMBRYO .
(TAKYDROMUS TACHYDROMOIDES).

T. Oka. Dep. of Biol., Tokyo Gakugei
Univ. , Tokyo

The development and differentiation of
the ovaries were examined by light and
electron microscopy in Takydromus embryos
incubated at 280 or room temperature.
During the stages 32-33, ovarian differen-
tiation of the gonad occurred. The
medullary cells showed elliptical shape
and had rough surfaced endoplasmic.
reticulum(RER), mitochondria with lamellar
eristae and occasional lipid droplets.
From the stage 34 to 38, ovarian medulla
became strongly basophilic, suggesting the
intervention of protein synthesis.
Ultrastructurally, the amount of ER
apparently increased. The majority of ER
were rough-surfaced, although several
places detachment of ribosomes were seen
on the surface of it. The typical SER
demonstrated in the steroidogenic cells,
could not be observed. In addition,
mitochondrial cristae were lamellar
instead of being like those of most
steroidogenic cells. Until the stage 40 (
hatching stage), many medullary cells
showed signs of degeneration.

These ultrastructural alteration
observed in medullary cells occurred at
the same time as the regression of the
genital tubercles in female embryos.

Developmental Biology 1083

DB 5

INTERACTION FOR EFFECTS OF ESTRADIOL AND
TAMOXIFEN ON THE SAZE OF THE GONAD IN
DEVEIOPING QUAIL EMBRYOS.

Y. Araki and N. Yamamoto. Dept. of Physiol...
Gifu Univ.. Sch. of Med.. Gifu.

Estradiol benzoate (Es) (0. 02mg). tamo-
xifen (TAM) (0. 2mg) or Es plus TAM was
injected into yolk sac of fertilized eggs
of quails on the zero day stage. The
gonads of the males and the females from
the sixth to the thirteenth day of inecu-
bation were observed. The maximum dimension
of gonads (cortexes (C) and medullae (M))
was determined from the surface areas of
the cross sections. On the thirteenth day.
the areas (10° 2mm?) of the left and right
gonads in the control femals showed 26+2
(Gals amch Wie lee) ancl BO, B (GeO. SO, OF
and M:4.5+0.6) and were bilateral asymmet-
rical. These in TAM-treated females gave
31+3 (C:7+0.8 and M:24+2) and 15+2 (C:1+
0.1 and M:14+2). were near ‘bilateral sym-
metrical. Qn the other hand, these in the
control males showed 14+1 (C€:2+0.2 and M:
12+1) and 12+1 (C:1+0.1 and M:11+1). was
Symmetrical. whereas these in Es-treated
males showed 12+1 (C:6+1 and M:6+0.6) and
4+0.4 (C:1+0.1 and M:3+0.4) became not sym
metrical. But, these in double (Es + TAM)
treated males showed 11+1 (C:3+0.1 and M:9
+1) and 9+1 (C:1+0.1 and M:8+1). returned
to symmetrical size. These results indicate
that on the point of bilateral symmetry
of gonad sizes. Es results in feminization
of the male gonads, but on the contrary
TAM induced masculinization in the female
gonads.

DB 6
MBRPHOLOGY OF THE SEMINIFEROUS TUBULE IN

THE TESTIS OF THE TORTOISE,GEOCLEMYS REE-
VESIS.

M.Chiba. Dept. of Biol., Fukushima Med.
Coll. Fukushima.

The seminiferous tubule in the lizard,
Tachydromus thehydromoides was found to
be tubular,extending from the rete testis
filling the whole testis with tortuous tu-
pkeseas Seen in the mammalian vests. Bur
the seminiferous tubules of the lizard
were single-track, starting from the rete
testis, another end of the tubule forming
a blind end.This survey was undertaken to
find out all af the reptilian seminiferous
tubules were the same.Examining the testis
of the tortoise, Geoclemys reevesii reveal-
ed that the seminiferous tubule was not
tubular. They were flask-shaped. The dia-
meterséf the tubules varied fram 300pm to
50 pm. And the length of the seminiferous
tubules varied from 200 pn to 300 pm.Some
degenerating primary oocytes were observed
in the tubule. Spermatogenesis did not
take place in the testis of an individual
of the carapace length 9 cm yet.Larger
individuals should be used to find out. the
seminiferous tubule of adult condition.
Animals of the order Chelonia are believed
to be direct descendants of cotylosaurs;
EVO Weel alia wl wWeiansieing Wesces Oi Owlaere
reptilian species such as the lizaeds and
snakes, alligators and crocodiles wait for
the elucidations.

DB7

CALMODULIN DISTRIBUTION IN MOUSE TESTIS
OBSERVED ie JUMNUNO = BSC MICROSCOPY.
Mig | Wigueslyyer” - Gig Ikeicerejalieabh Elinyol I< NVEWesal—

ZOO. SnSite,, Hace SC. 7 Hokkaido: Univ.
Sapporo, “Rakuno Gakuen Univ., Ebetsu,
Hokkaido.

A monoclonal antibody raised against
calmodulin from scallop testes reacted with
calmodulins from a variety of animals, but
not those from plants. Indirect immuno-
electron microscopic observations were
performed on sections of LR-Gold embedded
mouse testis using colloidal gold (15 nm in
size)-labeled secondary antibodies. The
detection of calmodulin was possible only
when tissue sections were reacted in the
presence of calcium. The numbers of gold
particles counted on sections revealed that
a Significantly higher label was observed
in the spermatogenic cells at stages
ranging from the pachytene stage of primary
spermatocytes to round spermatids. The
peak density of grains was found when the
first and second divisions of meiosis
occur. In those cells gold particles were
distributed quite uniformly both in nucleus
and cytoplasm, and there was no
preferential association of the particles
to specific organelles. Consistently lower
labels were observed in the spermatogonia
and young primary spermatocytes, as well as
Sertoli cells and other somatic cells in
testis. Mature sperm had no significant
density of gold particles except at the
site along the inner aspect of acrosomal
cap on the nuclear membrane.

DB 8

ISOLATION OF SERTOLI CELL-SERTOLI CELL
JUNCTIONAL COMPLEX FROM RAT TESTES

H. Ueda, T. Nishino and S. Fujimoto

Dept.) of Anat Tb, (Sch. of Med, .Univi-nioL
Occup. and Environ. Health, Kitakyushu.

Sertoli cell-Sertoli cell junctional
complex (SJC), which mainly consists of the
tight junction, forms the blood-testicular
barrier, and divides seminiferous tubules
into basal and adluminal compartments which
contain spermatogonia and spermatogenic
cells after meiosis, respectively. In or-
der to investigate the molecular components
and biochemical nature of SJC, we tried to
isolate SJC using tight junction purifica-
tion procedures from X-ray irradiated rat
testes, since the X-ray irradiation caused
a decrease in the number of spermatogenic
cells and an increase in the volume of SJC.
By counting the number of SJC by electron
microscope, extraction with the anionic
detergent sodium deoxycholate (DOC) and N-
lauroylsacrosin (SRK) left the detergent-
insoluble SJC. The purity of SJC was about
50 and 70% in DOC- and SRK-insolubule
fractions, respectively. Sodium dodecyl-
sulfate polyacrylamide gel analysis showed
the SRK-insoluble junctional complex to be
characterized by major polypeptides at 31
and 33KDa," with minors bands! at 6, 185,
22) Sil, OO and "68iKDar. The present results
were compared with tight junction molecules
from other tissues and indicated that SJC
in rat testes was constituted by unique
molecular components.

1084 Developmental Biology

DB 9
SPERMATOGENESIS IN MOUSE FETAL TESTES
RECONSTITUTED BY HANGING DROP =

GYRATORY CULTURE
C.Kato and M.Noguchi. Dept. Biol., Fac
of Sci., Shizuoka Univ., Shizuoka 422

In order to analyze the interaction
between germ cells and somatic cells in
spermatogenesis, we developed a new
method of making the reaggregates from
mouse fetal testis cells.

In 129/Sv and C3H strains, 15.5
d.p.c. fetal testes were dissociated and
germ cells were separated from somatic
cells after 12hr culture by different
adhesiveness to the culture dish. The
percent of viability of these testis
cells was 97%. Germ cells were reaggre-
gated with somatic cells by hanging
drop- gyratory culture for 24 hr. The
resultant reaggregates were grafted in
adult testes. It was found that the
seminiferous tubules have been reconsti-
tuted in all grafts and that germ cells
ranging from type A spermatogonia to
spermatids have differenciated in 70% of
them.

These results indicate that hanging
drop-gyratory culture is much simple and
useful method for controlling the aggre-
gate size and producing aggregates from
a few fetal testes. :

Supported by Grant-in-Aid of the
Ministry of Education, Science and Cul-
ture(to M.N.).

DB 10

THE FUNCTION OF A GENE INVOLVED IN THE
SPERMATOGENESIS STUDIED IN A P-ELEMENT
INSERTIONAL MUTANT OF DROSOPHYLA
eerie! T.Akiyamal, S.Kobayashil,
a Wd WAREZ and M.Okada!. linst. of Biol.

GH Ss, Wiaeisv es of Tsukuba, Tsukuba,

Mitsubishi Kasei Inst. of Life Sciences,
Machida, Tokyo.

In order to analyse the mechanisms
underlying germ-line development, we
isolated several lines of mutants causing
defects in its germ-line. These mutations
were induced by insertion of a P-element.
From one of these mutants, which was male-
sterile, flanking genomic DNA of the
inserted P-element were cloned by tagging
of the P-element. Northern blot analysis
probed with the cloned DNA revealed that
the transcription of the gene occurred only
in larva, and that the transcripts were
never detected in the mutant larva. We
observed testes of the mutant at larval,
pupal and adult stages, and found that
there were spermatogonia and spermatocytes,
but no spermatids and spermatozoa. The
result suggests that this is a mutation of
a gene required for meiosis. Considering
that meiosis initially starts at the end of
the larval stage but it is repeatedly
initiated through their life: span, these
results suggest that the product of this
Cine Ines “Gl se@lle sin Bomeleste “eaulile eo
construct a system which is regqired for
initiating meiosis.

DB 11

A 16-kDa PROTEIN IS PRESENT IN HUMAN SEMI-
NAL PLASMA THAT COUNTERACTS THE EFFECTS OF
THE ANTIFERTILITY AGENT, GOSSYPOL.
M.NAKAMURA 7 M. IKEDA 7 M.KOMUKAT,, T.SUYE-
MITSU', S.OKINAGA and K.ARAI.. Dept. of
OB/GYN, Teikyo Univ., Tokyo, “Dept. of Pha-
gmacol., Nippon Medical Sch., Tokyo and
Dept. of Fac. Sci., Saitama Univ., Urawa.

Gossypol inhibited lactate dehydrogenase
(LDH) in human spermatozoa in a non-compe-
titive manner. The inhibitory effect of
gossypol on LDH was cancelled by addition
of human serum albumin (hSA), human r-glo-
bulin, BSA or human seminal plasma (hSP).
hSP was at least 10 times more effective
than the other three proteins, when expre-
ssed on a per mg protein basis. Attempts
were made to purify the active fraction
from hSP. The purification steps included
gel filtration, ammonium sulfate precipita-
tion, centrifugal microconcentration and
fast-performance liquid chromatography. A
Single active protein of Mr=16-kDa was
purified to a final yield of 0.18%. The
16-kDa protein was not observed in male
blood plasma. The protein was found to be
heat-stable, and leucine-rich (16% of the
molecule), and has been designated "gossact!
The inhibitory effect of gossypol on the
LDH reaction was completely blocked by
addition of gossact (5 ug/ml): hSP (25 ug/
ml) and hSA (200 ug/ml) were far less pote-
nt in the assay. The role of gossact to
protect LDH from gossypol is discussed.

DB 12

SPECIFIC INCORPORATION OF VITELLOGENIN BY
VITELLOGENIC OOCYTES OF THE RAINBOW TROUT
(SALMO GAIRDNERI) IN VITRO.

N. Shibata, M. Yoshikuni, and Y. Nagahama,
Lab. of Reprod. Biol., Natl. Inst. for Basic
Biol., Okazaki.

We have developed an in vitro assay system

for vitellogenin (VTG) incorporation using
follicle-free oocytes of rainbow trout. VTG
was purified from serum of vitellogenic
females using cation exchange and gel
filtration chromatography. Purified VTG was
labeled with *?P and was added to incubation
medium. The incorporation of labeled VTG was
strongly blocked by the existence of
follicle layers, suggesting that VTG passes
with difficulty across the thecal cell
layer. Subsequent experiments therefore made
use of follicle-free oocytes, to yield the
following findings. I) Adding non-labeled
VTG to the medium inhibited incorporation of
VTG while adding BSA did not, indicating
that VTG was specifically incorporated. II)
Insulin (10ug/ml) significantly increased
VTG incorporation. However, both chum salmon
GTH-I and GTH-II (lug/ml) had no effect.
III) When using the oocytes that matured in
vivo, only 1/15 of VTG (compared to immature
oocytes) was incorporated. Also insulin had
no effect on these mature oocytes: VIG
incorpora-tion has stopped by this stage.

Developmental Biology 1085

DB 13

SUBUNIT ANALYSIS OF YOLK PHOSPHOPROTEINS OF
MEDAKA EGGS. .

M. Murakami, I. Iuchi and K. Yamagami.

Life Sci. Inst., Sophia Univ., Tokyo.

There are four major phosphoproteins
(PP1-PP4) in the yolk of unfertilized eggs
and blastulae of medaka. PP3 and PP4 are
most probably phosvitins (Murakami et
al.,1989). Molecular weight estimation and
subunit analysis of the four proteins were
done by gel filtration chromatography, Fer-
guson plotting on PAGE and two dimensional
electrophoresis. PP1 and PP2 were found to
be heterodimers composed of subunits of
113K and 94K, and subunits of 84K and 72K,
respectively. The subunits of 113K and 72K
contained phosphorus as revealed by
methylgreen staining. PP3 and PP4 were
phosphoprotein monomers with molecular
weights of 38K and 22K, respectively on
SDS-PAGE, although their molecular weights
were estimated at about 84K and 47K,
respectively, by gel filtration chromatog-
raphy (Murakami et al., 1989). Lipid com-
positions of a mixture of PP1 and PP2,
vitellogenin and yolk were compared by ex-
tracting their total lipids with a mixture
OFC hOrsOF OM ea nda meithianolsss((2)1)) ana
analyzing on silica-gel thin layer
chromatography. A great similarity in
lipid composition was found among them, in-
dicating that lipid composition of yolk
material scarcely changes from oogenetic to
early embryonic stage. PP andesPP2 sane
probably lipovitellins of medaka.

DB 14

PNAS PALI eC SYNTHESIS OF MAWOR
GLYCOPROTEINS OF EGG ENVELOPE, ZI-1,-2 AND
3 IN THE ESTROGENIZED MALE MEDAKA

IX MMPS 7 Wa SoG lslenmel 4 ellie al = I.Iuchil,
K.Yamagami Life Sci. TiN Bie Ss 7 Sophia
Univ. ,Tokyo. 2Dept Bi@ils, Ixeile Winsly, Sea

Med., Yokohama

Polyclonal anti-egg envelope glycoprotein
(F1)-IgG can be fractionated into two
antibodies; A-antibody specific for ZI-3
which is one of the major components of egg
envelope, and B-antibody specific for ZI-1
and -2 which are the other major com-
ponents. Intrahepatic formation of ZI-3 is
demonstrated by immunochemical and
biochemical analyses. The liver of spawning
females is known to react also with B-
antibody. In the present study, whether ZI-
1 and -2 are formed ,like ZI-3, in the
estrogenized male of medaka was examined by
immunoblotting and immunohistochemical
methods using B-antibody as a probe. When
the adult male fish were kept in tap water
containing 100 ng/ml of estradiol-17(E,) at
15 , immunoreactive protein bands cor-
responding to ZI-1 and -2 appeared in
liver extracts and blood. Immunohis-
tochemical analysis also showed that the
liver cells of the estrogenized male con-
tained ZI-1 and -2. ZI-1 and -2 were
ClOcESSsAlgle ain Inejoaicie eels 8 Cre WG love
after estrogen treatment and their amount
increased thereafter. Appearance of ZI-1
and -2 was found to be well synchronized
with that of ZI-3.

DB 15

APPEARANCE OF THE EGG JELLY MOLLECULE,
FUCOSE SULFATE GLYCOCONJUGATE AND SPERM-
ACTIVATING PEPTIDE IN THE GROWING OOCYTE
OF THE GROWING OOCYTE OF THE SEA URCHIN
HEMICENTROTUS PULCHERRIMUS
H.Kinoh, T.Shimizu, M.Yamaguchi, N.Suzuki.
Noto Marine Lab., Kanazawa Univ., Ishikawa
The sea urchin egg is surrounded by
jelly coat, consisted of fucose sulfate
glycoconjugate(FSG), sialoglycoprotein and
sperm-activating peptides(SAPs). It has
been reported that FSG and SAP play impor-
tant roles in inducing the acrosome reac-
tion. There has been considerable con-
troversy where these molecule are syn-

thesized. Using antisera against FSG
purified from H. pulcherrimus jelly coat
and synthetic SAP-I, we examined im-

munohistochemically thin sections prepared
from the ovaries collected from Oct. to
Feb.. The cytoplasm of .mid-size oocytes
(40pm in diameter) which were found in the
Ovaries collected in the middle of Dec.
reacted to both anti-FSG and anti-SAP-I
sera. Smaller oocytes and germinal
vesicles themselves did not react with the
antisera. We also found granule-like
stains in the cytoplasm and the secreted
materials around large-size oocytes(60-
80pm in diameter) of the ovaries collected
in early Jan.. These observations suggest
that egg jelly molecules such as FSG and
SAP are synthesized in the growing oocytes
and are secreted from the oocytes to the
extracellular space.

DB 16

Ultrastructural changes during oogenesis
in the holothurian, Polycheira rufescens
J.TSUKAHARA, M.TOMARI and T.KUBOTA

Dept. Biol., Fac. Sci., Kagoshima Univ.
Kagoshima.

Gonadal tubules of P. rufescens consist
of onky two layers, the outer coelomic and
the inner gérminal layers. Several
oogonia or early oocytes are located in
the bottom. of the folding of the inner

layer. They have two types of nuage as
same as that shown in S. californicus
(Smiley, 1988). At the beginning of

vitellogenesis, oocytes reach zygoten
stage including synaptonemal complexes-in
the germinal vesicle. Active
vitellogenetic oocytes have numerous
microvilli and pinosomes on their surface.
They are surrounded wath the thinner basal
lamina. Animal hemisphere of them is
contact with neighbor nutritive cells.
the other knee supkace some Ehe yoocyten is
contact with genital hemal sinus, which
probably conveys nutrients from the
periphery of the tubule to oocytes. Woe
full grown oocytes(about 130-150 pm in
diameter) bear an axial protuberance,
which inserts into the nutririve cell
layer. The surface of the top of
protuberance has many adhering junctions
attaching it to the surface of neighbor
nutrient cells. After the treatment of
ring nerve extract (RNE), the germinal
vesicle of the oocyte is breakdown during
one hour and ovulation occures at 90 min
or more after treatment of RNE.

1086 Developmental Biology

DB 17

OOCYTE GROWTH IN THE CHIMERA FROG BETWEEN
DIPLOID AND TRIPLOID OF XENOPUS LAEVIS,
M.Amano, T.Ueno, Y.Eto and A.Kawahara.
Cell and Developmental Biology Laboratory,
Fac. of Integrated Arts and Sciences,
Hiroshima Univ., Hiroshima,

Oocytes in triploid frog can not grow
to ooxocytes in Xenopus. To study whether
the arrest of growth of triploid oocytes
due to the germ cells or environmental
somatic cells, the chimera frogs were
produced by the exchange of primordial
germ cells between diploid and triploid
laevae. The block was cut out from ventral
side which containing primordial germ
cells at early stage of tail bud. These
frogs were reared 3 to 5 months after
metamorphosis. The ovaries were fixed,
embedded in paraffin and stained with
hematoxylin and eosin for histological
examinations,

In the operated frog between diploid
and diploid larvae, abnormality was not
observed. The chimera diploid frog trans-
planted triploid primordial germ cells
showed the arrest of oocyte growth simi-
lar to triploid frog ovaries, On the other
hand, the ovaries in the chimera triploid
frog with diploid primordial germ cells
contained many grown oocytes.

These results described above suggest
that the arrest of oocyte growth in tri-
ploid frog is caused by the-triploid germ
Gell wWeselse ,

DB 18

IN VITRO SPERMATOGENESIS AND FLAGELLAR
GROWTH FROM THE MEDAKA,ORYZIAS LATIPES,
PRIMARY SPERMATOCYTES.

K.Onitake,A.Katougi and A.Saiki. Dept.of
Biol.,Fac. of Sci.,Yamagata Univ.,Yamagata
Quite recently, we succeeded in in vitro

spermatogenesis from the medaka primary
spermatocytes and reported in part about
the differentiating process of spermatids
(1990). In this experiments, it was showed
that primary spermatocytes in metaphase
completed within lhr and formed pairs of
secondary spermatocytes. Single flagella,
0.5-lum in length, were observed to grow
from secondary spermatocytes within lhr
after telophase I. The duration time of
interphase II is about 3hr. After inter-
phase II, secondary spermatocytes in meta-
phase II completed the second division
within 55min. 4-5hr are required for the
progression from telophase I to telophase
II at 25°C. The secondary spermatocytes
proceeded through metaphase II and telo-
phase II and formed quadruplets of round
spermatids. Soon, a flagellum developed in
each round spermatid. Flagella attained
the length of 15-20um after 24hr's culture
and did not increase after an additional
hour. Chromoanalysis of the cells in sus-
pension culture showed that about 80% of
the examined cells arrested by Colcemid
were in first or in second meiotic meta-
phase, since the chromosome number of
medaka somatic cells is 48(n=24). BRdU(50n
g/ml) did not inhibit the in vitro sper-
matogenesis. BRdU-incorporation in sper-

matids was revealed by immunocytochemistry.

DB 19

HOW IS THE FLAGELLAR LENGTH OF MATURE
SPERM DETERMINED? IV. STABILITY OF
TUBULIN MRNA IN CULTURED AMPHIBIAN
SPERMATIDS.

S. ABE and S. UNO Dept. of Biol. Sci.,
Fac. of Sci., Kumamoto Univ., Kumamoto.

In synchronous cultures of round
spermatids from Xenopus laevis and
newt, we have shown that the period of
flagellar growth coincides with that of
tubulin synthesis (ECR, 1990).

In this study, we compared _ the
stability of tubulin mRNA in semi-
synchronous cultures of round spermatids
from Xenopus and newt. Total cellular
RNA was extracted from round spermatids
on day 0, 3, 6, 9 and 12, and Northern
blotting was performed with radiolabeled
Xenopus @-tubulin cDNA _ probe. The
amount of a@-tubulin mRNA decreased
linearly in both species but decayed
more rapidly in Xenopus than in newt.
The amount of @-tubulin mRNA on day 0
was compared between both species and it
was found that the amount in newt was
approximately 7 fold more than that in
Xenopus. Considering the difference in
their cell volume, the concentration of
a-tubulin mRNA in round spermatids on
day 0O seems almost the same in both
species. The length of poly(A) tract in
newt spermatids (200b) was larger by
100b than that in Xenopus. The length of
poly(A) tract did not change throughout
the culture period in both species.

DB 20

TRANSLATIONAL REGULATION OF SP4 mRNA

DURING XENOPUS SPERMATOGENESIS.

H. Hiyoshi and S. Abe. Dept.of Biol.Sci.,

Fac. of Sci., Kumamoto Univ., Kumamoto.
We have previously isolated a cDNA

clone of SP4 which is one of the 6

sperm-specific basic nuclear proteins

‘from Xenopus. Northern blotting has

shown that SP4 mRNA is present in
primary spermatocytes, which indicates
that SP4 mRNA is stored until round
spermatid stage where it is translated
(Annu. Meeting of Zool. Soc. Japan,
1989).

In this study, we examined the size
change of SP4 mRNA during in vitro
spermatogenesis by Northern blotting
with SP4 cDNA probe. The size of SP4
MRNA in round spermatids on day 0 (710b)
was more heterogeneous and larger by 60b

than that in primary spermatocytes
(650b). Then, by day 6, the size
decreased to 680b. Removal of poly(A)

tract by ribonuclease H resulted in the
same size (560b) for all three stages.
This indicates that the size change of
SP4 mRNA is due to that of poly(A)
tract. It was revealed that the mRNA was
in postpolysomal fraction in primary
spermatocytes and round spermatids on
day O but was in polysomal fraction in
round spermatids on day 6.

Developmental Biology 1087

DB 21

STABILITY OF MANCHETTE (MICROTUBULE
BUNDLES) IN NEWT ELONGATE SPERMATIDS.II.
K. Hashimoto and S. Abe. Dept. of Biol.
Sene5g J9C6 On “Sens, LainemMOem - Winitri7e -
Kumamoto.

We have previously shown that
manchette structure in newt elongate
sPeumstids is resistant to the treatment
of Ca and low temperature.

In this study, elongate spermatids
was separated from round spermatids by

centrifugation in a gradient of
Metrizamide solution. ©
First, tubulin isotypes in Ca =

sensitive fraction in round spermatids
and those in Ca“ -resistant fractions in
elongate spermatids were compared by 2-D
PAGE. The result showed that there was a
little differences,in the mobility of a@-
tubulin between Ca “sensitiye fraction
in round spermatids and Ca“ -resistant
fraction in elopgate spermatids. Second,
tubulins in Ca“ -resistant fraction _in
elongate spermatids and those in Ca _
sensitive fraction in round spermatids
were polymerized by adding taxol and
centrifuged. MAPS were separated by 1M
NaCl and compared between round
spermatids and elongate spermatids. The
results showed that approximately 4
bands (75K-90K) specific to Ca*-
resistant fractions in elongate
spermatids were identified.

DB 22

ORGAN CULTURE OF TESTIS FRAGMENTS FROM
CYNOPS PYRRHOGASTER.
ton So. dn VeuNcl So FNloxso IDSs g wae Ven @l oSyens 4

Fac. of Sci., Kumamoto Univ., Kumamoto.

To examine the role of Sertoli
cells and hormones in spermatogenesis,
organ culture of newt testes fragments

was performed.

Testes containing mostly spermato-
gonia were cut into fragments (1-2 mm in
diameter) and cultured at 22°C on
nucleopore filters which were floating
on medium. Basic medium was L-15 + 10%
fbs. Hormones-supplemented medium
contained insulin, transferrin, FSH,
testosterone, 5ea-dihydrotestosterone,
retinol, a-tocopherol and ascorbate.

In a basic medium, the diameters of
fragments and cysts decreased within 2

weeks. Differentiation to primary
spermatocytes was barely observed after
2 weeks. On the other hand, in a
hormones-supplemented medium, the
diameters of fragments and cysts did not
decrease during 2 weeks and

differentiation to primary spermatocytes
was observed within a week in more. than

70% of the cysts examined. Also, the
differentiation to round spermatids was
seen within 2 weeks. Among the
components in hormones, FSH,

testosterone and 5a-dihydrotestosterone
were found to be most effective in
promoting the differentiation of
spermatogonia to primary spermatocytes.

DB 23
IDENTIFICATION OF STAGE OF SPERMATOGENIC

CELLS IN Bufo aiepouels

Mita,K. , K.Takamune , S.Uno2 and S.Abe2
Vo64OOI, WAS 5 WACS SCilop WOM AlCl Wiss 7
SAVIIOMO>s AIDE Biol, WAC Sito,
Kumamoto Univ., Kumamoto.

The developmental stages were
identified for the spermatogenic cells in
the testis of a toad, Bufo japonicus. The
cells with lobulated nucleus were
identified as spermatogonia because of
their high synthetic activity of DNA as
defined by incorporation of BrdU. The
cells with spherical nucleus of about 9 pm
in diameter were identified as primary
spermatocytes because of the presence of
synaptonemal complexes in the nucleus.
The cells with spherical nucleus of about
5.5 pm in diameter was identified as
spermatids as defined by prominant
development of acrosomal vesicles and
basal bodies of flagella in their
cytoplasm. Countings on tissue sections
of the number of primary spermatocytes and
spermatids in each cluster suggested that
a spermatogonium differentiates into
primary spermatocytes following 6 series
of mitoses, before undergoing meiosis to
yield spermatids. When dissociated
spermatogenic cells were cultured in 70%
L-15 containing 10% fetal calf serum, the
cells with 20 pm in diameter divided 2
times without growth followed by
differentiation of flagella, indicating
that these cells were primary
spermatocytes.

DB 24

STAGES OF TRANSCRIPTION OF SPERM-SPECIFIC
NUCLEAR PROTEINS AND THEIR SYNTHESIS
DURING SPERMATOGENESIS OF ANURANS

Mita, K., K. Takamune, & Ch. Katagiri,
Zool. Inst., Fac. of Sci., Hokkaido univ.,
Sapporo.

The cDNAs encoding the protamines of
Bufo japonicus and the one of the sperm-
specific basic proteins (SP4) of Xenopus
laevis were cloned. Northern blot
analyses using these cDNAs showed that the
genes encoding these proteins were
transcribed only in the testis. Similar
analyses of the spermatogenic cells
separated by the Celsep equipment revealed
that in Bufo, the protamine mRNAs were
first detected at the round spermatids,
whereas in Xenopus the SP4 mRNAs were
present in the primary spermatocytes.
This difference in the stage of expression
of sperm-specific basic proteins was
confirmed also by the in situ
hybridization technique.
Immunohistochemical observations and AUT-
PAGE profiles on spermatogenic cells at
various stages of spermatogenesis
indicated that these proteins were first
detected concomitant with the commencement
of nuclear condensation at the last step
of spermiogenesis. Comparative studies
using these amphibians will be useful to
analyze the mechanisms of both
transcriptional and translational controls
of sperm-specific protein synthesis during
spermatogenesis.

1088 Developmental Biology

DB 25

TIME COURSES OF SPERMATOGENESIS IN
AMPHIBIANS; EFFECT OF TEMPERATURE ON THE
KINETICS OF SPERMATOGENESIS.

Y. Asakawa, T. Kobayashi, and H. Iwasawa.
Biol. inst. , Nagata Unive, Naacgatal.

We studied the time course of
spermatogenesis which is the basic
reference to use in elucidating the
mechanism of spermatogenesis. With
bromodeoxyuridine [BrdU] as the cell
MWMeieh< ie} Talies line OrigyOlaiewl Oil Aint ©
spermatogonia was detected immunohisto-
chemically. Spermatogenic stages of
labeled cells were distinguished, and the
durations of meiotic prophase and
spermiogenesis were determined.

In Xenopus laevis, labeled leptotene
cells were first observed on day 1 after
the injection of BrdU, zygotene on day 2,
pachytene on day 14, diplotene on day 26,
elongated spermatids on day 30, and
labeled spermatozoa appeared on day 34, at
DAC a In Cynops rrhogaster, labeled
leptotene, zygotene, pachytene, diplotene,
elongated spermatids, and spermatozoa
appearedijaias2Z2G Non idayi uly, 27, 6), 28) aeSie,
and 42 after the injection, respectively.
Hid Xo laevis - Ehe time courses were
SOMocIeOCs Ene Wo Gy BACH ein leisy A
Significant difference in the duration was
seen at the step from pachytene to
diplotene: 34 days at 16C, 12 days at 22C,
elinol 2 GlaivyS Gig ZE(ee Furthermore, a
Significant difference was also seen at
Ehe step Exnom round spermatids to
elongated spermatids. These results
indicate that steps from pachytene to
diplotene and from round spermatids to
elongated spermatids are influenced by
temperature in X. laevis.

DB 26

LIGHT CONDITIONS CONCERNING OOCYTE MATURATION AND
SELF-STERILITY OF THE ASCIDIAN, HALOCYNTHIA RORETZI
T. Numakunai! and M. Fuke2, 1Mar. Biol. St., Fac.
of Sci., Tohoku Univ., Asamushi, 2Dept. of Biol.
Fac. of Sci., Kanazawa Univ., Kanazawa

One of the three intraspecific groups of H. ro-
retzi (Type C) begins to release gametes about 4
hours later after exposure to light following dark
adaptation (D-L treatment) at 13°C. The released
eggs don't have the germinal vesicle, and are
absolutely self-sterile. Immediately after D-L
treatment almost eggs in the gonad were immature
and self-fertile. From 1 to 3hours after D-L treat-
ment the mature eggs increased in number rapidly.
The self-sterile immature eggs also increased from
1 to 3hours. The self-sterility of the mature eggs
was about 90% at 1 hour, almost 100% at 2hours
after D-L treatment. There was no mature egg which
was self-—fertile at 3hours. In the gonad, which was
isolated immediately after D-L treatment and left 4
hours, neither the self-sterile immature egg nor
the mature egg was observed. The self-—fertile
immature eggs were polyspermic, and the mature eggs
of which the germinal vesicles were broken down in
sea water after taken out of the gonad were mono-
spermic. In both cases the peri-vitelline space was
not expanded after fertilization. Some of the self-
fertilized eggs which were matured in the gonad did
not expanded the peri-vitelline space after ferti-
lization, and the others did as naturally released
eggs did.

These results suggest that the germinal vesicle
breakdown and the self-sterility are caused at
thesame time after D-L treatment by effect of some
factors other than gonad.

DB 27

ULTRASTRUCTURAL CHANGES OF THE MEDAKA
OOCYTE DURING MATURATION.

S.Nakashima and T.Iwamatsu. Dept. of
Biol., Aichi Univ. of Educ., Kariya:

The oocyte of the medaka Oryzias latipes
acquires full fertilizability three hours
after germinal vesicle breakdown (GVBD)
(Iwamatsu et al., 1976). It has been dem—
onstrated that the acquisition of ferti-
lizability is achieved without a direct
contribution of germinal vesicle material
in oocytes of Oryzias (Iwamatsu, 1966,
1971). This suggests that the acquisition
of fertilizability of oocytes depends on
some physiological and structural changes
in their cytoplasm. Therefore we studied
the ultrastructural changes of the oocyte
during maturation.

Three hours before and at the time of
GVBD, the morphological characteristics of
the cytoplasm are the existence of
annulate lamellae, mitochondria with a
mass of ribosomes. In the region close to
the surface there is a layer containing
thin filaments, small yolk granules, small
electron denseless vesicles and pino-—
cytotic vesicles, but not large cyto-—
plasmic inclusions such as mitochondria,
large yolk granules, and Golgi complexes.
At the time of GVBD two kinds of vesicles
containing minute vesicles were observed.
Changes in these cytoplasmic inclusions
correspond with the time of full acquisi-
tion of fertilizability.

DB 28

INVOLVEMENT OF CYCLIC AMP IN METHYLATION
PROCESS OF 1-METHYLADENINE BIOSYNTHESIS OF
STARFISH OVARIAN FOLLICLE CELLS.

M. Mita. Dept. of Biochem., Teikyo Univ.
Sch. of Med., Tokyo.
Resumption of meiosis in starfish

oocytes is induced by l-methyladenine CHS
MeAde) produced by ovarian follicle cells
under the influence of gonad-stimulating

substance (GSS). With respect to 1-MeAde
production, the effect of GSS on follicle
cells results in the receptor-mediated

formation of cyclic AMP (cAMP). 1-MeAde is
also shown to be newly synthesized by way
of a methylation process. This study was
undertaken to determine whether increase in
intracellular level of cAMP is related to
methylation in 1-MeAde biosynthesis by
isolated follicle cells of the starfish,
Asterina pectinifera. Methionine and
selenomethionine enhanced GSS-induced 1l1-
MeAde production by follicle cells. These
stimulatory effects markedly depend on GSS
concentrations. Methionine and seleno-
methionine also caused enhancement of I1-
MeAde production induced by 3-isobutyl-1l-
methylxanthine (IBMX) which stimulates cAMP
accumulation in follicle cells. Although
in the presence of l-methyladenosine l=
MeAde-R) follicle cells can produce 1-MeAde
without increase in cAMP levels, these
methionine related compounds did not have
any effect on 1-MeAde-R-induced 1-MeAde
production. These results strongly suggest
that cAMP is a agent responsible for
Mediating the methylation process in 1-
MeAde biosynthesis induced by GSS.

Developmental Biology 1089

DB 29

SPECIFIC BINDING OF [*H]17a,208-DIHYDROXY-4
-PREGNEN-3-ONE TO CELL SURFACE OF OOCYTES
OF RAINBOW TROUT. II. Analysis by equili-
brium dialysis and ultracentrifugation

M. Yoshikuni and Y. Nagahama

Ib@ilo’-, @se Ineyoseorel4 Issloils 5 INEweIl, ImMSte5 ieee
Basic Biol, Okazaki.

We reported that plasma membrane fraction

prepared from follicle-free oocytes of
rainbow trout specifically bound [7H]17a,
208-dihydroxy-4-pregnen-3-one ({*H]diol).
Specific binding had been determined by
using charcoal adsorption to separate
bound- and free-forms (B-F separation) of
[*H]diol. The dissociation constant (Kd) was
about 0.1 uM and binding capacity (Bmax)
was 0.2 fmoles per 1 mg of protein. We
have examined other methods to determine
the binding constants more accurately.
Using equilibrium dialysis instead of
charcoal adsorption for B-F separation,
Scatchard analysis showed the presence of 2
different binding sites: a high affinity
binding site with Kd(high) of 18 nM and
Bmax of 2 pmoles per 1 mg protein; and a
low affinity site with Kd(low) of 0.5 uM
and Bmax of 10 pmoles. This suggests that
B-F separation by charcoal adsorption is so
vigorous that specific binding is greatly
decreased by this method. An ultra-
centrifugation method also showed a higher
binding rate compared to charcoal
adsorption, and since this is more
convenient than equilibrium dialysis, it
is the preferred method for routine use in
the binding assay.

DB 30

PURIFICATION OF MPF FROM STARFISH OOCYTES
E. Okumura, S. Hisanaga, K. Tachibana and
T. Kishimoto.

Lab. Cell and Develop. Biol., Tokyo Inst.
Technol., Meguro, Tokyo 152.

We have purified MPF (maturation-
promoting factor) from starfish oocytes
based on its GVBD-inducing activity upon
injection into immature starfish oocytes.
After 1-MeAde addition, crude extracts
were obtained by crushing maturing oocytes
with centrifugation, and were applied to
p13SUC'_sepharose affinity chromatography.
MPF activity was retained on the gel and
then was eluted with p13S¥4°'-containing
buffer. This one step procedure resulted
in a preparation purified 750-fold. A
yield of 60% was obtained. Western
blotting and immunoprecipitation with
antibody against starfish cyclin B and
antibody against the conserved PSTAIR
region of cdc2 protein indicated that cdc2
protein and cyclin B were complexed in
purified MPF. In addition to histone H1,
we found that the highly purified MPF
phosphorylated myelin basic protein and
H-chain of neurofilament as exogeneously
added substrates. In the dephosphorylated
H-chain of neurofilament, changes of
electrophoretic mobility correlated with
the rate of phosphorylation by MPF. Both
the rate of purification and the yield of
the activity showed little difference
between in neurofilament H-chain kinase
activity and in GVBD-inducing activity.

DB 31

PURIFICATION OF MATURATION-PROMOTING
FACTOR (MPF) FROM MATURE CARP OOCYTES.

M. Yamashita, S. Fukada, M. Yoshikuni,

P. Bulet and Y. Nagahama. Lab. of Reprod.
Biol,, Natl. Inst. for Basic Biol., Okaza-
ki, Lab. of Animal Biol., UST Lille Flan-
dres Artois, CNRS, France.

MPF and histone H1 kinase were purified
from the 100,000g supernatant of crushed,
naturally spawned carp oocytes, using 4
cheomat og saan columns: Q-Sepharose Fast-
Flow, P13SU¥°'-affinity Sepharose, Mono S,
and Superose 12. MPF and H1 kinase activi-
ty co-migrated throughout purification,
indicating that they are the same entity.
The final preparation was purified 5,000-
fold with a recovery of 4% for H1 kinase,
and 1,000-fold with a recovery of 1% for
MPF. The purjfied kinase was thermolabile,
cAMP- and Ca“*-independent, and had a wide
pH optimum (8.0 to 10.5 or over). The
kinase utilized ATP, GTP and ATPrS with
apparent Km values of 230, 1000 and 160
uM, respectively. On a Superose 12 gel
filtration column, MPF and H1 kinase
eluted as a single peak with an apparent
molecular mass of 100 kD. SDS-PAGE analy-
sis of the active fractions after Superose
12 showed that they contained 4 peptides

of 33, 34, 46, and 48 kD. The 33 and 34 kD

peptides are cdc2 kinase as confirmed by
anti-cdc2 kinase monoclonal antibody. The
46 and 48 kD peptides are endogenous
substrates (presumptive cyclins) of the
kinase.

DB 32

cDNA cloning for maturation-promoting
alc On (MPT) Rote gomidirirsihen (Caralsis mans,
auratus). x
THERA ee Yenv oH pes Oulu) Mie a amatsinestiay a Mi
Tanaka, and Y. Nagahama.
Reprodugtive biology and Developmental
Biology , National Institute for Basic
Biology, Okazaki, Japan.
Maturation-promoting factor (MPF) con-
sists of cdc2-kinase and cyclin. To
isolate cDNAs for these molecules, a cDNA
library of mature eggs of gold fish was
constructed in phage vector Agt10.
Two types of cyclin clone (A-type and B-
type) were identified by hybridization
with a Xenopus cyclin A cDNA and with
oligonucleotides for conservative
sequences of cyclin B, respectively. The
nucleotide sequence of the longest clone
for each type was determined. The
sequence for cyclin A encodes 374 amino
acids having a homology of 58% for Xenopus
A, 45% for Clam A, and 43% for Drosophila
A. The sequence for cyclin B encodes 356
amino acids having a homology of 67% for
XCHOPUS BIN is oil te LOG BZ, s15ll 6) elon aCclam By,
and 47% for sea urchin B. However,
neither of them cover the entire sequences
of coding region. The deduced amino acid
sequences for goldfish cyclin A and B have
35% homology. No clones for cdc2-kinase
have been identified yet.

1090 Developmental Biology

DB 33

CALCIUM TRANSIENT UPON MATURATION IN STAR-
FISH OOCYTES.

M. Kikuyama and Y. Hiramoto. Biol. Lab. The
Univ. of the Air, Chiba, Japan

Changes in intracellular concentration
of calcium ions ([Ca?*]:i1) were determined
in oocytes of the starfish, Asterina
pectinifera when 1-methyladenine (1-MA) was
applied to the oocyte to induce maturation.
The [Ca2?*]1 was determined from the inten-
sity of luminescence of a high sensitivity
semi-synthetized aequorin presented by Dr.
Shimomura (hep-aequorin) with a photomul-
tiplier tube or a supersensitive video sys-
tem coupled with a microscope. When 1 wM
1-MA dissolved in sea water was applied to
an oocyte loaded with the aequorin, an in-
crease in the luminescence indicating
[Ca2*];1 increase started within 1~ 2 min,
reached a peak about 30 sec after the
start, and then gradually decreased in
about 2 min. When 1-MA was applied locally
at a restricted region of the cell surface,
the luminescence started at the part of the
cell close to the applied region and spread
over the entire cell in 30 sec. The break-
down of germinal vesicle followed by the
extrusion of polar bodies was generally ob-
served after the calcium transient, but it
did not occur if the duration of the ap-
plication of 1-MA was short. This fact
suggests that [Ca2*], increase is not a
condition sufficient to induce whole
processes of maturation of the oocyte.

DB 34

CHROMOSOME CONDENSATION INDUCED BY
PURIFIED MPF IN XENOPUS SPERM.

A. Shimada, E. Okumura and T. Kishimoto.,
Lab. of Cell and Dev. Biol., Tokyo Inst.
OF Technol, LOkyo lor

In lysolecithin-permeabilized Xenopus
sperm nuclei, chromosome condensation was
induced in vitro by cytoplasmic extracts
of metaphase-arrested Xenopus laevis
eggs, but not by MPF (maturation-promoting
factor) alone which was highly purified
from starfish oocytes as a complex of cdc2
protein and cyclin B. To know the
mechanism of MPF action in these
processes, we prepared Sepharose beads
conjugated with a bacterially expressed
sucl protein of Schizosaccharomyces pombe.
Western blotting with antibody against the
conserved PSTAIR region of cdc2 protein
revealed that cdc2 protein was completely
removed from the Xenopus egg extracts
after the incubation with sucl-beads. The
MPF-depleted extracts could induce nuclear
swelling but not chromosome condensation
in the sperm nuclei. Acid extraction
followed by SDS-PAGE analysis suggested
that sperm nuclei-type proteins were
replaced with egg-type histones. When the
swelled nuclei were washed with a buffer
and then added with the purified MPF,
condensation occurred in sperm chromatin.
These results indicate that after the
swelling of sperm nuclei with the flow-
through of sucl-beads, MPF acts directly
on sperm chromatin to cause condensation.

DB 35

DOWN-REGULATION OF MAMMALIAN CDC2
HOMOLOGUE IN VIVO AND IN VITRO.
N. Hashimoto and T. Kishimoto. Tokyo
Inst.Technol., Meguro, Tokyo 152.
Expression of mammalian cdc omo logue
was studied at product and mRNA levels by
means of immunoblot probed with anti-
PSTAIR monoclonal antibody and northern
blot probed with mouse cdc2 cDNA. In 14
to 18 day mouse fetuses, cdc2 protein
(CDC2) was present in testis, brain,
kidney and liver. There was no
significant difference in CDC2 levels
among tissues examined. In adults, CDC2
and cdc2 mRNA were detected only in
testis, ovary and spleen, which contain
undifferentiated stem cells, but not in
brain, kidney and liver. Thus, mouse cdc2
seems to be down-regulated during
differentiation in vivo. In Swiss 3T3
cells, cdc2 was also down-regulated
during G1/G0O transition induced by serum
starvation. In contrast, expression of
cdc2 was stable during in vitro
differentiation of F9 cells and PC12
cells, induced by retinoic acid and NGF,
respectively. For cytological detection
of CDC2, signals probed with anti-PSTAIR
antibody was enhanced by silver staining
following immunohistochemical procedure.
This improved technique revealed that
majority of CDC2 was located in nucleus
of growing Swiss 3T3 cells but in cyto-
plasm of undifferentiated F9 cells. It is
likely that down-regulation of cdc2 is
related to nuclear localization of CDC2.-

DB 36

EGG TRANSPORT ON COELOMIC EPITHELIUM IN
THE FEMALE NEWT, CYNOPS PYRRHOGASTER.
Y.Yamahama and K.Onitake. Dept.of Biol.,
Fac.of Sci., Yamagata Univ., Yamagata.

In 1935, Rugh reported that the coelomic
eggs were transported from the region of
ovary to ostium by the active ciliary
movement of the coelomic epithelium in the
amphibia. In order to analyze the mech-
anisms of egg transport, coelomic, uter-
ine and fertilized eggs were inserted into
the body cavity or put on the coelomic
epithelium of the open body cavity in ma-
ture female newt, and the egg movements of
the latter were recorded by the photograph
or video tape recorder. In this experiment
speed of coelomic egg transport was the
highest of all examined eggs. To charac-—
terize the coelomic egg transport, coelom-
ic eggs were treated with trypsin, period-
ic acid or boiled Steinberg's solution. It
is showed that the periodic treatment re-
duced the speed of coelomic egg transport
by the ciliary movement, while trypsin
treatment accelerated the speed than nor-
mal one. Another experiments were pursued
to know whether the egg transport by the
ciliary movement is depend on the sexually
activity of the female newt or not. Where-
as the ciliary movement could be usually
observed in sexually active female induced
by Gonatropin injection or inactive female
egg transport speed of the former was 2
times higher than the latter. It is sug-
gested that the egg transport speed is co-
incidence with female sexual activity.

-

Developmental Biology 1091

DB 37

OOGENESIS AND FERTILIZATION OF VIVIPAROUS
FISH, POECILIA RETICULATA

R. Hori. Ogaki Women's Junior College,
Ogaki.

The fertilization of viviparous fish is
internal and the development is intra-
follicular. The fry are released after
four-week gestation period. Oocyte devel-
opment may start immediately after the
parturition. Gravid females of the guppy,
Poecilia reticulata were kept in the aquar-
ium. One day or two days after parturition,
Ovaries were dissected out in isotonic
saline solution. The oocytes were carefully
separated and fixed for electron micros-
copical observations. In the chromatin-
nucleolar stage of the oocyte, the shape of
granulosa cell is flat. Microvilli from the
surface of the oocyte contact with granu-
losa cell. In perinucleolar stage, vesicu-
lar membranous structure appears under the
granulosa cells. In oil-droplet stage, the
nucleus of granulosa cell assumes different
irregular forms and the number of mitochon-
dria in granulosa cell increases. High
electron-dense chorion with many perfora-
tions is formed at the basal region of the
vesicular membranous structure. In early
yolk granule stage, the shape of granulosa
cell with rough-surfaced endoplastic
reticula becomes more irregular. Numerous
projections derived from egg cortex pene-
trate the perforation of chorion and spread
in the space between chorion and granulosa
cells. Fully matured egg with two-layered
chorion is different from that of oviparous
fish which has multi-layered chorion.

DB 38

A STUDY ON ARTIFICIAL PARTHENOGENESIS OF
STARFISH EGGS PRODUCED BY PRE-MATURE ACTI-
VATION.
S. Washitani-Nemoto. Biol. Lab., Hitotsu-
bashi Univ., Tokyo.

Starfish oocytes become fertilizable
after germinal vesicle breakdown (GVBD).
It has been reported that production of
artificial parthenogenesis is successful
only when parthenogenetic agents are intro-
duced to maturing oocytes (after GVBD).

Recently, I found that immature oocytes
(GV stage) of the starfish, Asterina pecti-
nifera, can be activated parthenogenetical-
ly by caffeine at:.30 mM or higher concent-
rations, which are much higher than those
required for activation of maturing oocytes
(around 6 mM). When subsequently treated
with l-methyladenine, the caffeine-treated
oocytes resume meiosis. After extruded the
first polar body, they developed partheno-
genetically without forming the second
polar body. These results indicate that
parthenogenetic stimulation given at the
pre-mature stage is maintained until meio-
sis reinitiation, and that failure of
meiosis is required for occurrence of
cleavages as in post-mature activation.
Most of the parthenogenetic embryos were
tetraploids. An extra round of chromosome
duplication without cytokinesis was found
before the first cleavage. Activation by
calcium ionophore did not cause cleavages.

DB 39

FERTILIZATION OF COELOMIC EGGS OF XENOPUS
LAEVIS

Ch.Katagiri and R.Maeda, Zool. Inst., Fac.
Sci. Hokkaido Univ., Sapporo.

In Xenopus laevis, the requirements of
the oviductal contribution for acquisition
of egg fertilizability have been less eluci-
dated, although there is clear molecular
analyses showing the coelomic envelope (CE)
to vitelline envelope (VE) conversion indu-
ced by the pars recta (PR) portion of ovi-
duct ((Bakosmetrals Deven Broil 36 H69
('90)). To establish the system with
which the physiological roles of oviduct be
analyzed in Xenopus, attempts were made to
successfully fertilize coelomic eggs. We
confirmed the unpublished observation by
Jeffery H. Roberts that the aqueous extract
of jelly (JW) from either Xenopus or Bufo
japonicus supplemented with Ficoll or PVP
supports a high rate of fertilization of B-
mercaptan-de jellied (VE-intact) eggs. For
coelomic eggs, incubation of the eggs for 4
hr in the PR extracts (PRE) obtained from
females of Xenopus prestimulated with PMSG
or of Bufo stimulated with pituitaries, was
highly effective in inducing fertilization
when inseminated in the presence of JW sup-
plemented with Ficoll or PVP. The eggs
matured in vitro by progesterone treatment,
however, gave extremely low rate (<10%) of
fertilization under the same conditions,
suggesting a need for improvement of cul-
ture conditions during the progesterone-
induced maturation period.

DB 40

CHANGES IN ELECTROPHORETIC PATTERNS OF OO-
CYTE PROTEINS DURING ACQUISITION OF FERTI-
LIZABILITY, OF MEDAKA OOCYTES.

We INEM nest ,S.Y.Takahashi~ and T.Hirabaya-
Sink 4) Dee, WileEil,_, Waleink Win, Incltceies ,
Kariya, “Dept. Biol.,Fac. Lib. Arts, Yama-
guchi Univ.Yamaguchi and ~Inst. Biol. Sci.,
Univ. of Tsukuba, Tsukuba.

In the medaka (Oryzias latipes) oocytes,
several protein spots became newly detect-
able in the range between the acidic and
slight basic proteins on the two-dimension-
al SDS-electrophoretograms, while a few of
the protein spots disappeared with the pro-
cess of nuclear and the cytoplasmic matur-
ation. In the oocytes that were matured
without the germinal vesicle breakdown (GV-
BD) ,the formey _proteins were also detected.
The uptake of~~S-methionine into these pro-
teins, was not recognized during meiotic
maturation. On the other hand, the acquire-
ment of fertilizability after the GVBD
stage of oocyte maturation was not inter-
ferred with the presence of inhibitors of
protein synthesis. Judging from these re-
sults, the proteins that were required for
the acquisition of fertilizability of the
oocyte may be hardly synthesized during
meiotic maturation. The present study iden-
tified an intimate relationship between
cytoplasmic maturation and stage-specific
changes in the electrophoretic patterns of
proteins of medaka oocytes.

1092 Developmental Biology

DB 41

THE STRUCTURE AND THE ACROSOME REACTION-
INDUCING ACTIVITY OF FUCOSE SULFATE GLYCO-
CONJUGATE OF THE SEA URCHIN HEMICENTROTUS
PULCHERRIMUS.

T. Shimizu, M. Yamaguchi and N. Suzuki.
Noto Marine Lab., Kanazawa Univ., Uchiura.

Fucose sulfate glycoconjugate (FSG)
from sea urchin egg jelly is a major fac-
tor in induction of the acrosome reaction
of spermatozoa, and sperm-activating pep-
tide promotes the reaction as a specific
co-factor of FSG. FSG consists of at least
four components which are associated each
other by disulfide bonds. In the present
study, we separated FSG into three frac-
tions by HPLC gel filtration under reduc-
ing conditions. We examined the activity
of the fractions in induction of the
acrosome reaction and also in increase of
CAMP concentrations in sperm cell. As a
result, a fraction containing fucose had
the both activity, while the other two
fractions which contained only protein
component(s) did not.The fucose-containing
fraction possessed almost the same amount
of protein as fucose by weight. We
prepared protein-free fucose sulfate
poiymer from the fraction by pronase di-
gestion and 8-elimination. The fucose sul-
fate polymer, however, did not have the
both activity. These results indicate that
the fucose-contaning fraction of FSG is
responsible for induction of the acrosome
reaction and suggest that the core protein
plays an important role in the reaction.

DB 42

ACROSOME REACTION OF HAMSTER SPERM IN
CUMULUS MATRIX OF HOMOLOGOUS AND
HETEROLOGOUS EGG.

N. Uto!, Y. Yamamoto’ and 0. Serizawa’.
“Oe Ore Ibo S Eimcl Ainley” Te@is | japyoeies
Animals, Hamamatsu Univ. School of Med.,
Hamamatsu.

Studies on mammalian fertilization have
indicated that the acrosome reaction of the
fertilizing sperm is most probably
initiated on or in the cumulus oophorus and
zona pellucida. Hamster spermatozoa were
induced strongly the acrosome reaction by
zona of hamster egg but not so much by one
of mouse egg. In the present study, we
investigated whether the cumuli of hamster
and mouse are able to induce the acrosome
reaction as effectively as hamster zona by
a video micrographic recording. After
incubation for 2 hr, capacitated hamster
sperm appeared to become enmeshed in the
cumulus matrix or to adhere to cumulus
cells. Except phagocytosed sperm, about
10% of penetrating sperm was induced the
acrosome reaction by both kinds of cumuli

within 30 - 40 min after insemination.
These effects were some degree lower than
one of homologous zona. Some sperm

retained in mouse cumulus reached to
surface of zona and attached to it.
Compared with the case of cumulus free
mouse egg, hamster sperm passed through
cumulus induced the acrosome reaction more
effectively and bind zona more tightly.

DB 43

ACTIVATION OF PROTEASOME IN SEA URCHIN
SPERM BY LYSO-PHOSPHOLIPIDS AND BY SPERM
LIPIDS.
K.Matsumura and K.Aketa. Akkeshi Mar. Biol.
Stn., Hokkaido Univ., Akkeshi Hokkaido.

We have purifie proteasome
(multicatalytic proteinase) from sperm of
the sea urehin,-) SProngywatocenpnorws

intermedius, and suggested that its
chymotrypsin-like activity is involved in
the acrosome reaction (AR) (1988).

Chymotrypsin-like activity of
proteasomes from various eukaryotic cells
is known to be activated by poly-L-lysine
or SDS. Proteasome of our material was not
exceptional in this respect. These
reagents, however, are artificial and can
not be expected to work in vivo. Some fatty
acids are known to activate the proteasomes
from mammalian -ccells. To clarify the
endogenous activator of sperm proteasome,
ef £ Gicttis’ of \ various frank ty, alcuaciss:
phospholipids and lyso-phospholipids on the
activity were investigated, and lyso-
phosphatidyl inositol (LPI) found to be the
most potent activator of all lipids
examined. LPI at 2.5 pg/ml stimulated the
activity above 10-fold. Endogenous lipids
extracted from the egg jelly-treated sperm
also strongly stimulated the activity.
Quinacrine and p-bromophenacyl bromide
inhibited egg jelly-induced AR, but not
ionophore-induced AR. It appears the lyso-
phospholipids produced by phospholipase A
stimulate the chymotrypsin-like ace ae ey oF
sperm proteasome at onset of AR.

DB 44

THE ROLE OF JELLY LAYER ON THE ACROSOME
REACTION IN THE URODELE, CYNOPS PYRRHO-
GASTER Sout of ee
H. Takai and K. Onitake. Dep. of Biol.,
Fac. of Sci., Yamagata Univ., Yamagata.

In the urodele, Cynops pyrrhogaster, an
acrosomal region of the sperm consists of

the acrosomal cap and the perforatorium.
When the oviducal eggs were inseminated,
it was found that all sperm remaining
in the capsular chamber, which is inside
of the innermost jelly layer, caused the
acrosome reaction (AR) and the acrosomal
caps disappeared completely following the
exposure of their perforatoria. Sperm
which penetrated through the isolated full
or part of jelly layers also caused the
AR. It appears that the components of jel-
ly layer may induce the AR of the sperm,
and then, we tried to extract the factors
responsible to the AR. When sperm was in-
cubated with the water-soluble extracts
from oviducal egg jelly layers(jelly ex-
tracts;JE), the AR was induced following
the disruption of acrosomal caps and the
exposure of perforatoria. JE induced the
AR completely within a minute. AR-inducing
activity of JE was reduced to control
level by the heat- or trypsin-treatment.
While dialysis of JE against Cat+-free
Steinberg's salt-solution reduced AR-in-
ducing activity, the addition of Cat* re-
covered it to normal level.

It may be possible that JE factors re-
sponsible to the AR are protein-like and
Cat*-dependent substances.

a il, el ie

I Pm

Developmental Biology 1093

DB 45

ION SPECIFICITY AND PERMEABILITY OF THE

CA‘* CHANNEL THAT IS ACTIVATED BY HIGH

EXTERNAL PED eND THE EGG JE LY,

Y.Sendai ’ and K. Aketa . Akkeshi Mar.
iol. Stat., Hokkaido Univ., Hokkaido.
Noto Mar. Lab., Kanazawa Univ., Ishikawa.

We previously showed that the same Ca2t
Eranspormt ‘system (Gace "channed) of
Hemicentrotus pulcherrimus sperm is
activated by high external pH (pHe 9.0) and
the egg jelly. In the present study, to
know the ipn specificity and permeability
of this Ca“* channel, we examined, using
fura-2, whether divalent ions such as Ba‘“*,
Sr<* and Mn**t, which are known to pass the
Ca“* channel in many other cells, pass
through the Ca“* channel of sperm.

The Ba“* influx occurred under high pHe
Condwecrons whereas the Sr inf lux
scarcely occurred. Similar results were
observed also with the egg jelly
treatment. The Mn“* influx occurred not
only under high pHe but also ynder normal
PHISH Os Addition of Mn~*' to) sperm
suspended into Ca *free ASW (pHe 8.0), like
addatiion of Ca-", caused pHi rise.
Mn2*(>2.5 mM) significantly blocked the
acrosome reaction (AR) by high pHe or the
egg jelly. Ba“* blocked slightly the AR,
whereas Sr“* was no effect.

These results suggest that (1) the Caz
chgnnel of sperm is Ba“*-permeable and
Sr“*-impermeable, (2) the Mn“* influx is
mediated by Ca‘t(Mn“+)/Ht exchange system
and (3) Mn“* may block the Ca“* channel.

DB 46

RECEPTORS FOR SAP-I AND REGULATION
MECHANISM OF cGMP CONCENTRATION IN
HEMICENTROTUS PULCHERRIMUS SPERMATOZOA
T. Harumi, M. Yamaguchi and N. Suzuki,
Noto Marine Lab., Kanazawa Univ., Ishikawa

A sperm-activating peptide I analogue,
GYGG-SAP-I, was chemically synthesized,
radiolabeled with Na!2°r, and used for
experiments of binding and cross-linking
to H. pulcherrimus spermatozoa. Competi-
tive binding experiments showed that Sees
binds exclusively to sperm tails. 257_
GYGG-SAP-I was closs-linked to 63 kDa and
71 kDa proteins in the sperm tails.

To understand the signal transduction
mechanism after SAP-I binding to the
receptors, we examined the SAP-I action on
H. pulcherrimus spermatozoa in high [K*]
sea water in which the activation of sperm
metabolism was repressed. In high [K*] sea
water, SAP-I did not stimulate the respi-
ration rate and motility of the spermato-
zoa. SAP-I did not elevate on sperm cAMP
level under high [K*] conditions, although
it bound to spermatozoa as in normal sea
water and elevated sperm cGMP level higher
than that in normal sea water. These
results suggest that SAP-I receptors are
localized in sperm tails and SAP-I binding
to the receptors activates guanylyl
cyclase activity at first, followed by
elevation of sperm cGMP level.

DB 47

PURIFICATION OF SPERM-ACTIVATING PROTEIN
FROM UNFERTILIZED HERRING._EGGS.

H. OHTAKE™, S. TANIMOTO“, K. _SAKAI®,
N. SHIMIZU~ and M. MORISAWA”~. lpept. of
Physiol., Dokkyo Univ. Sch. Med., Tochigi,
Dept. gt BOs 7 ECGs Ole Seis 5. Olio. Wine
Chiba, “Dept. of Mol. Biol., Keio Univ.
Sch. Med., Tokyo, and “Misaki Marine

Soll se. Sicziibilom., , Wace, Ol Sees Winilys Ore
Tokyo, Kanagawa.

Contrary to the most marine teleosts,

the spermatozoa of the herring, Clupea
palasii, are almost motionless when
diluted into normal sea water. The sperma-
tozoa around egg become active when put
into egg suspension and gather at the
micropyle area of the egg membrane,
suggesting that unfertilized eggs release
sperm-activating substances into their
surroundings. Recently, we found sperm
-activating proteins (HSAP) , in the
supernatant of herring egg suspension. To
understand the molecular mechanisms of the
sperm activation, we attempted to purify
HSAP from the supernatant of unfertilized
egg suspension. The supernatant was
concentrated with ammonium sulfate and
applied successively to Sephadex G-50 gel
filtration, isoelectric focusing and Mono
Q column chromatography. The preparation
of HSAP thus obtained showed a single band
on SDS-polyacrylamide gel electrophoresis.
The apparent molecular weight on SDS

-PAGE was 11,000 and the pI value was 4.9.
The roles of HSAP in the fertilization of
the herring species were discussed.

DB 48

MOTILITY ACTIVATION AND CHEMOTAXIS IN ASCIDIAN
SPERMATOZOA

M. Yoshida and M. Morisawa.

Misaki Marine Biological Station, Fac. of Sci.,
Univ. of Tokyo, Miura-shi.

In ascidians, it has been reported that
spermatozoa exhibit active motility and chemo-
tactic behavior to an egg at fertilization, and
that follicle cells are supposed to attract
spermatozoa (R. L. Miller, 1975).

In the present study, eggs of the ascidians,
Ciona savignyi and C. intestinalis, were sepa-
rated with steel needles into follicle cells,
chorion, and naked eggs, egg cell. Spermatozoa
were added to each fraction and tracks of motile
sperm were observed under the phase contrast
microscope and analyzed with the automated semen
analyzer, CellSoft (CRYO Resources, Ltd.). Our
results suggested that spermatozoa are attracted
neither to follicle cells nor to a chorion, but
to the vegetal pole of an egg cell.

Then, we investigated the characteristics of
the chemoattractant which may be released from
the vegetal pole of an egg cell. Egg sea water
(ESW) was prepared by incubating eggs from an
ascidian in 0.5ml of artificial sea water (ASW)
for about 12hrs at 4°C. In our observations,
ascidian spermatozoa were usually immotile in
ASW, bat became motile by addition of ESW. The
motility inducing activity in ESW remained when
ESW was boiled for 30min or digested by some
proteases, however, the activity was lost when
ESW was dialyzed. These results suggest that the
chemoattractant from ascidian eggs may be a
small molecule which is not protein.

1094 Developmental Biology

DB 49

COILED FILAMENTOUS STRUCTURE IN THE
ACROSOME OF GASTROPODA SPERM

Ye Shiewolysary, —ainida Werle aikvetin. Bouts abies,
Wayo Women's Univ., Chiba.

The presence of the coiled filamentous
structure in the apex of the acrosome was
demonstrated in the sperm from two species
of gastropoda, Sulculus aquatilis and
Turbo cornutus. This filamentous struc-
ture was located right beneath the outer
acrosomal membrane in a shape of the trun-
cated cone, which appears as a periodic
array-bearing band in thin sections.
Negative staining observations of the
sperm from both species after saponin
treatment revealed that the filamentous
SURWetuUre ws composed “of abouwt 12
regularly coiled filaments with a diameter
of 8-12 nm. During the acrosome reaction,
the filamentous structure elongated an-
teriorly from the opening of the acrosome
closely surrounding the acrosomal process.
In S. aquatilis sperm, this coiled
filamentous structure having a slant
height of about 0.334m transformed into
a cylindrical shape 3 times the original
height, whereas in T. cornutus sperm it
elongated 1.5-fold. At high magnifica-
tion, each filament appeared to have pe-
riodic substructure ranging from 17 to 22
nm. As these results are fundamentally
homologous to those of abalone sperm,
coiled filamentous structure appears com-
mon in gastropoda sperm.

DB 50

MORPHOLOGICAL CHANGES IN SPERM

AT FERTILIZATION

A. Hino! and K. H. Kato* .! Department of
Biological Sciences, School of Science,
Kanagawa Univ., Hiratsuka and * Biological
Institute, College of General Education,
Nagoya City University, Nagoya

It has been reported that the sea urchin
sperm undergo the acrosome reaction when
they pass through the egg jelly layer
(Kato. 1987). It has also been reported
that the sea urchin sperm mixed with
glutaraldehyde-fixed eggs were acrosome-
reacted and adhered on the egg surface and
finally became immotile (Hino et.a/.'80).
In such quiescent sperm, the mitochondria
became spherical. The electron micrographs
showed that the matrix of mitochondria was
the condensed conformation in active
swimming sperm and the orthodox confor-
mation in quiescent sperm. The volume of
mitochondria seemed to be increased in
spherical mitochondria. In almost all
cases, the changes in sperm mitochondria
was accompanied by a shift in its position
toward the direction of the centriole.
Such morphological changes in mitochondria
were also observed during cross fertili-
Zation between Pseudocentrotus (egg) and
Hemicentrotus (sperm). In this case, the
sperm not only on the egg surface but also
in the jelly layer showed the mitochon-
drial shift. These morphological changes
in sperm mitochondria at fertilization were
observed not only in the sea urchins, but
also in the starfish, the ascidians and the
echiuroid.

DB 51

FERTLIZING CAPACITY AND LIFE-SPAN OF
SPERMATOZOA SUSPENDED IN L-15 MEDIUM ARE
PROLONGED IN THE MEDAKA, ORYZIAS LATIPES
M. Terui and K. Onitake Dep. of Biol.,

Fac. of Sci., Yamagata Univ., Yamagata.

Isotonic salt solution was used for
insemination routinely in telost arti-
ficial fertilization, but sperm motility
and fertility suspended in it were lost
soon. When the medaka spermatozoa were
suspended and kept in L-15 medium (L-15
sperm suspension) ,it was appeared that
their life-span was prolonged,and they
had motility and fertility over 5 days.
Using the L-15 sperm suspension kept from
1 day to 5 days, an artificial insemi-
nation was attempted. When the medaka
eggs were inseminated by i-day L-i5 sperm
suspension, about 93% of eggs were ferti-—-
lized,and developed and hatched out
normally as same as control. Insemination
by 2-, 3-, 4- and 5-day L-15 sperm sus-—
pension revealed about 69%, 45%, 20% and
10% of normal development of eggs. Scann-
ing electron microcropic observation
showed that the fertilizing process was
normal,and sperm penetrated into micro—
pylal canal and attached on the sperm
entry site of eggs following cortical
reaction. L-15 salt solution only re-
constructed by inorganic salts of L-15
medium could not have the potency of
prolonging the sperm fertility and life-—
span. It is suggested that L-15 medium is
useul to the artifical insemination of
the medaka for substitution of its Ringer.

DB 52

EFFECTS OF TMB-8 OR CTC ON OXYGEN CONSUMP-
TION ENHANCED BY TREATMENTS WITH TPA IN SEA
URCHIN EGGS.

M. K. Kojima, K. Tamaki and S. Nakamura.
Dept. of Biol., Fac. of Sci., Toyama Univ.,
Toyama.

We reported that 07 consumption of unferti-
lized sea urchin eggs is enhanced not only
by treatments with weak bases, such as pro-
caine and NHq4Cl, but also by sub-threshold
stimulation with Ca Ionophore A23187, insuf-
ficient to induce visible cortical changes.
And, further, it was shown that such en-
hanced respiration is cancelled by treat-
ments combined intracellular Ca-inhibitors,
such as TMB-8 and CTC, with A23187, but not
by treatments combined each of the Ca-inhib-
itors with weak bases which induce an in-
crease of intracellular pH.

TPA (phorbol 12-myristate-13-acetate) is
known to stimulate protein kinase C, follow-
ed by Nat/Ht exchange and the resultant cy-
toplasmic alkalization. When eggs are
treated with more than 0.05 pM TPA sea water
for 10 min, the rate of 02 consumption
reaches 3-4 times higher than that of the
control eggs. However, if eggs are exposed
to 5 uM TPA in Na-free sea water, such en-
hancement is not induced. It is interesting
to note that TMB-8 or CTE has no effect on
respiration which has been enhanced by TPA,
as in the case of treatments with weak
bases, whereas if applied to eggs before
TPA-treatment, these Ca-inhibitors can sup-
press the rise of 027 consumption induced by
TPA, as in the case of A23187-treatment.

Developmental Biology 1095

DB 53

THE EFFECT OF GLUCOSE AND PHOSPHATE ON
MOUSE 2-CELL EMBRYOS IN VITRO.

K.Aoki!, M.Nakamura’”, H.Namiki!, S.Okinaga’, K.
Arai”. 1Dept. of Bio., Sch. of Educ., Univ. of Waseda,
Tokyo, *Dept. Obst. and Gyn., Sch. of Med., Univ.

of Teikyo, Tokyo.

The removal of either glucose or phosphate from
the complete TLP-PVA medium significantly improved
the rate of development of non-inbred (ICR) mouse
2-cell embryos to 3- and 4-cell stages. When neither
glucose nor phosphate was present, further improve-
ment in the development of 2-cell embryos beyond
the 2-cell stage was obtained. Two-cell block was
also removed by addition of NaF, an inhibitor of a
glycolytic enzyme, enolase. In addition, the removal
of 2-cell block was further improved when NaF was
present in the medium without glucose and phosphate.
Together with these findings, it is concluded that the
suppression of the glycolysis is probably important
for the removal of 2-cell block in early embryo deve-

lopment of mouse in vitro.

DB 54

FERTILIZATION CHANGES THE SUSCEPTIBILITY
OF SEA URCHIN EGGS TO PHORBOL ESTERS.
S.Kanai, S.Tanaka, K.Takemoto and H.Kuroda
Sugashima Marine Biol. Lab., Sch. of Sci.,
Nagoya Univ., Toba.

Na’ /H’ exchange is the principal mechan-
ism regulating intracellular pH in various
cells. There are many evidences suggesting
that diacyl glycerol (DAG), the other me-
tabolite of phosphatidylinositol bisphos-
phate besides inositol triphosphate, acti-
vates the Na /H exchanger via protein kin-
ase C and elevates pH. . In sea urchin eggs
a permanent rise in pH. during fertilizat-
ion is a signal for the activation of qui-
escent cells. Application of phorbol esters
» Which mimics the action of DAG in other
animal cells, to unfertilized eggs has been
SOWETO melici va considerable pH. rise as
observed in normal fertilization tSwann
and Whitaker, 1985).

We compared the effects of phorbol-12,
13-dibutylate (PDBu) on the acid-re'‘ease
from unfertilized eggs with that of ferti-
lized eggs of Pseudocentrotus depressus
and Lytechinus pictus. The PDBu-induced
acid release continued for more than 30 m
and strongly depgnded on the existence of
extracellular Ca in fertilized eggs,
whereas in unfertilized eggs the acid re-
legse took place transiently and even in
Ca” -free media. These results suggest that
the potentiality of PKC to activate the
Na /H exchanger,in responce to PDBu and
extracellular Ca may be derepressed by
fertilization.

DB 55

A TRANSIENT DEPOLARIZATION OF FERTILIZED
EGGS OF SEA URCHINS.

M.Osawa, K.Takemoto and H.Kuroda.
Sugashima Marine Biol. Lab., Sch. of Sci.,
Nagoya Univ., Toba.

f Fertilized eggs of sea urchins can be
refertilized" after removal of the ferti-
lization membrane and the hyaline layers
(Sugiyama, 1947). We measured the membrane
potentials of Pseudocentrotus depressus,
Hemicentrotus pulcherrimus and Lytechinus
pictus eggs with a conventional microelec-—
trode, and found a transient change in the
membrane potential associated with referti-
lization. With diluted sperm, a small tran-
sient depolarization of fertilized eggs
with a constant configuration appeared re-
peatedly. The eggs inseminated with dense
sperm showed a big transient depolarization
whose peak attained to -20+12mV. We called
this "refertilization potential" (RFP).

The peak value of the RFP was sensitive to
the external Na concentration but was a-
bout 110mV more negative than the equilib-
rium potential of Na . This means that not
only Na but also other ions are responsi-
ble for the RFP. Using DNA-specific fluoro-
chrome Transfer-Technique, we examined the
sperm-fertilized egg continuity preceding
the RFP and found that continuity was not
necessary to generate the RFP,

DB 56

THE ROLES OF IP3 AND G-PROTEINS IN THE
DEFORMATION AND THE POLAR BODY FORMATIONS
AT THE FERTILIZATION OF THE ASCIDIAN EGG

N. Sensuil, K. Suzuki2 and M. Morisawa2. 1
Dept. Anat.1 Fac. of Med.,Univ. of the
Ryukyus, Okinawa and 2 Misaki Mar. Biol.
Sica Winslw5 Cle Wolo, MELEE

At fertilization in ascidian, egg shape
deforms and polar bodies are formed.
Although it is known that Ca2+ triggers the
phenomena, the roles of guanyl nucleotide
binding proteins (G-proteins) and inositol
1,4,5-triphosphate(IP3) are still unclear.
We performed here the microinjection of IP3
and guanyl nucleotide analogues into
unfertilized eggs of the ascidian, Ciona
Savignyi. The deformation and the first
polar body formation were induced by IP3
without insemination, whereas the second
polar body formation did not occur. GTP-}-S
induced the deformation although the shape
was irregular. GDP=G, -S Jinhibijtted| the
deformation and the polar body formations
at fertilization. However, it did not
inhibit the polar body formations induced
by the treatment with Ca2t+ ionophore.
These results suggest that IP3 and G-pro-
teins are involved in the deformation and
the first polar body formation at fertili-
zation in ascidian.

1096 Developmental Biology

DB 57

FLUIDITY OF MEMBRANE LIPIDS OF CORTICES
ISOLATED FROM SAND DOLLAR EGGS.
K.Hirano, Hamamatsu Photonics K.K., .Tukuba.

Membrane fluidity of egg cortices was
investigated by the steady-state fluores-
cence depolarization using the digitized
video microscopy from the membrane probe
hexadecanoylaminofluorescein(HDAF). This
dye has resistance to "flip-flop".
Cortices were isolated from sand dollar,
Clypeaster japonicus, eggs on the glass
surface coated with polylysine.

When eggs were treated with 10 pM HDAF
for 30 min before isolation, plasma mem-
brane was fluorescently labeled but cor-
tical vesicles(CVs) were not. The fluore-
scence anisotropy(r) was increased from
0.256 to 0.285 during cortical exocytosis
induced _by 3 pM free Ca“+, Lowering of
free Ca“*t to 10 nM after exocytosis or
dislodging of CVs from cortices by further
shearing with a jet of buffer had no
effects on r. When cortices were treated
with HDAF after isolation, both plasma
membrane and the membrane of CVs were
labeled. This sample represents the r
value of 0.215. After dislodging CVs with
jet of buffer, r increased to 0.233.
These results suggest that the fluidity of
plasma membrane is decreased after exo-
cytosis by fusion of the membrane of CVs
and that the fluidity of plasma membrane
is lower than that of the outer leaflet
of the CV membrane and is higher than that
of the inner leaflet of the CV membrane.

DB 58

QUANTITATIVE ANALYSIS OF CORTICAL GRANULE
BREAKDOWN IN SEA URCHIN EGGS.

T. Mohri and Y. Hamaguchi. Biological
Laboratory, Faculty of Science, Tokyo
Institute of Technology, Tokyo.

The single cortical granule in sea
urchin eggs increased up to 1.2 times in
diameter and broke down within 40 msec,
when its breakdown was induced by
fertilization, milcroinyjiecting anositol
1,4,5-trisphosphate (IP3) or Ca-EGTA buffer
Solution anto, the ~egg, or, perfusing a
medium containing 1 mM Ca?+ to isolated
COnti ces. These values were almost
constant among three methods to induce
cortical granule breakdown. Upon
fertilization of Clypeaster japonicus eggs,
the cortical granule breakdown propagated
over the egg surface at a speed of 3.3
um/sec in the observing area of the egg
surface. However, even when the cortical
granule breakdown was observed in such a
small area as a rectangle in width of 3.3
um, it took 9 sec and 18 sec for 50% and
90%, respectively, of cortical granules to
break down because the maximal rate of R
breakdown was 7.6%/sec. Moreover, the rate
did not simply reduce with time, but a
shoulder was found during the reducing
phase, which suggests that cortical
granules would be divided into two groups
according to the responsibility to the
breakdown stimulus. The cortical granule
breakdown induced by microinjecting the Ca-
EGTA buffer and IP3 solutions propagated at
68 um/sec and 35 um/sec, respectively.

DB 59

THE VITELLINE COAT GLYCOPROTEINS IN THE
ASCIDIAN, HALOCYNTHIA RORETZI.
S.Takizawa and M.Hoshi. Dept. of Life
Sci., Tokyo Inst. of Technol., Tokyo.

It is suggested in the ascidian, Halo-
cynthia roretzi, that the terminal fucose
residue of a vitelline coat
glycoprotein(s) is important for sperm
binding to the vitelline coat. Although
the structure of N-linked oligosaccha-
rides of vitelline coat glycoproteins are
studied, no information is available on
the chemical structures of the protein
portions. We have studied the structures
of protein portion of vitelline coat
&lycoproteins, especially that of the
major one, 60k glycoprotein.

DB 60

CHANGES IN SDS-PAGE PROFILE OF THE VITEL-
LINE ENVELOPE BEFORE AND AFTER FERTILIZA-
TION IN THE URODELE, CYNOPS PYRRHOGASTER
T.ADACHI and K.ONITAKE Dep. of Biol.,Fac.
of Sci., Yamagata Univ., Yamagata.
To define the role of vitelline envelope
of an urodele amphibian, VES were isolated
from eggs of the coelom, uppermost portion
of the oviduct (Up) and uterus in Cynops
pyrrhogaster. It was demonstrated by SDS-
PAGE that the uterine egg envelope (UE)
lacked 92.5KD, 76KD and 44-45KD proteins
and possessed 80kD and 70KD proteins which
were not appeared in coelomic egg envelope
(CE), while 108kD,95kD, 78kD 51kD and 45kD
proteins were common in CE and UE. It was
also showed that VE from the Up (UpUE) and
UE resembled each other closely. In order
to clarify the role of Up, the coelomic
eggs were treated with the extracts of Up
for 6hr (TCE),and then TCE was removed and
analysed by SDS-PAGE. The electrophoretic
profile of TCE was already of UE but not
of CE. This results are very similar to
that of Anura. We concluded that in the
newt, Cynops pyrrhogaster, CE changed into
UE by the action of Up. It is well known
that Cynops egg lacks cortical granules
and so, it's fertilization is character-
ized by physiolgical polyspermy. We com-
pared UE with the vitelline envelope of
fertilized egg (FE) by SDS-PAGE. After
fertilization, 165KD protein was newly de-
tected in FE. Immunoblotting against anti-
UE and -FE antisera indicated that 165KD
protein is only specific to FE and may be
derived from the fertilized egg cytoplasm.

Developmental Biology 1097

DB 61

POST-ACTIVATION CHANGES OF CHORION PROTEINS
OF UNFERTILIZED RAINBOW TROUT EGG AND HAR-
DENING OF CHORION.

we Iwelal’ 5 Ks Masuda2 and K. Yamagami'|
(1. Life Sci. Inst., Sophia Univ. Tokyo,
2. Dept. Antibiotics, Natl. Inst. Health,
Tokyo)

Chorion or vitelline envelope of unfertil-
ized egg of rainbow trout, Oncorhynchus
mykiss, was found to consist of 4 major
protein components, 110, 64, 56, and 50kDa
proteins, including some glycoproteins.
Solubility of chorion in 1N NaOH or 8-12M
urea decreased during hardening after water
activation, accompanying transformation of
these constituents into high molecular
weight intermediate components (higer than
160K). This transformation process was
retained or conserved in chorions isolated
from wane Sucre I a7 eggs. This process was
dependent of Ca“*t, pH, and sensitive to
high temperature. It was promoted by 2-
mercaptoethanol but inhibited by cadaverine
(one of inhibitors of transglutaminase).
These results suggest participation of a
transglutaminase in formation of cross-
links between constituent proteins during
the chorion hardening. In addition, we
showed that probably a cadaverine-
insensitive hardening change also occurred
in rainbow trout egg chorion.

DB 62

ON "CA?*- HARDENING" OF CHORION ISOLATED
FROM UNFERTILIZED MEDAKA EGGS.

K.Masuda2, I.Iluchi? and K.Yamagami~. *+Dept.
Antibiotics, Natl. Inst. Health, Tokyo,
2Life Sci. Inst., Sophia Univ., Tokyo.

We have reported that there are two
types of hardening of chorion in fish eggs;
"Tonophore-activation hardening" is induced
by activation of unfertilized eggs with
Ca2*-ionophore and "Ca?*-hardening" occurs
in the chorion isolated from unfertilized

eggs in the presence of Ca?*. When the
isolated soft chorion was incubated with
Ca2~*, the chorion turned hard. This was

confirmed also by reduction of solubility
of the chorion in alkaline solution or
reduction of its capability of swelling in
DW. In parallel with the above described
changes, major components of chorion
glycoproteins (ZI-1, 2 and ZI-3) changed
into high molecular weight intermediates
(H-INTs). According to analysis by SDS-
PAGE, this hardening depended on pH and
temperature of the medium; Optimum pH was
5.5 and high temperature over 70 °C was in-
hibitory. This process was inhibited by
iodoacetamide and promoted by 2-
mercaptoethanol. Both cadaverine and
aminotriazole did not inhibit the
hardening-associated formation of H-INTs.
These findings suggested that trans-
glutaminase did not participate in "Ca?~-
hardening", which seems to be different
from “ionophore-activation hardening".

DB 63

ENZYMATIC BASIS RESPONSIBLE FOR THE DEFENSE MBECHA-
NISM OF THE FERTILIZATION ENVELOPE IN FISH EGGS.

S. Kudo, Dept. of Anat., Gunma Univ. Sch. of Med.,
Maebashi.

The kinds of enzymes present in the fertiliza-
tion envelope (FE) extracts from several species of
fish eggs were examined using an agar plate assay
for enzyme activities. The enzyme assay revealed
that the FE extracts may contain enzyme activities
responsible for digesting each of ten kinds of
substrate (barley 8-glucan, lichenan, laminaran,
carboxymethylcellulose, carboxymethylchitin, xylan,
Micrococcus _lysodeikticus dried cells, caseine,
dextran and mannan), forming a halo around each
well. The FE extracts had an antifungal or fungi-
cidal action on the fungus Saprolegnia parasitica.
However, the vitelline envelope extracts had no
antifungal or fungicidal action and revealed little
or no formation of haloes in the agar plates. The
antifungal or fungicidal action was exerted more
efficiently by prolonging the incubation time of
the suspension of the extract and mycelial pieces
of the fungus at 30 C; incubation for 6 h consider-
ably inhibited the growth of hyphae from the fungus
mycelial samples, incubation for 12 h resulted in
poor growth from only a few mycelial pieces even
at 48 h later, and incubation for 24 h resulted in
complete inhibition of hyphal growth from any
mycelial sample, with rare exceptions where poor
hyphal growth occurred from only one among several
samples. In conclusion, it is considered that the
mechanism by which the FE is able to protect the
developing embryo from invaders or pathogens
present in the water surrounding the FE is probably
enzymatic action, whose properties are acquired
through the cortical reaction process.

DB 64

ULTRASTRUCTURE OF THE CORTEX OF THE
SEA URCHIN EMBRYO, RAPID-FREEZED UNDER
MICROWAVE IRRADIATION.

I. Uemura. Dept. of Biol., Tokyo
Metropolitan Univ., Tokyo.

Ultrastructure of the cortex of the
sea urchin, Hemicentrotus pulcher-
rimus, embryo was studied using a new
cryofixation system equipped with a
microwave oven, and liquid N2 as
coolant. Microwave irradiation during
metal-contact rapid-freezing proved
effective for preservation of ultra-
structure of the cytoplasm of the
embryos suspended in normal sea water;
cryofixation of the specimens with an
high water content have been shown to
be difficult when applied with the
metal-contact method. The best results
are obtained with very low specimen
contact (slamming speed < 50 mm/s) for
preserving the overall shape of embryo.

Microfilaments in the microvilli are
clearly seen extending the underlining
cytoplasm of the rapid-freezed and
freeze-substituted (in OsO4) embryos.
Vacuoles lining the cortex are filled
with granular contents. In the inter-
cellular space are seen the electron-
opaque substances. Cytoplasm of the
primary mesenchyme cells are well
preserved even at 75 pm from the
contact surface, suggesting the
presence of anti-freezing substances
in the blastocoel.

1098 Developmental Biology

DB 65

GLOBULAR MT MASS INDUCED BY HEXYLENEGLYCOL
IN SEA URCHIN EGGS BEFORE MITOTIC PHASE.
S.Endo!, M.Toriyama*, K.Ohta!, and H.Sakail.
IDept. of Biophys. and Biochem., Fac. of
SCi, sl hes Unty, 210 eLOKYO;eLLOKY.O), 2Dept. of
Biology, Fac. of Liberal Arts, Shizuoka
Univ., Shizuoka

When the eggs were treated with hexylene-
glycol(HG) after fertilization, the forma-
tion of a large globular structure, called
globular MT mass was induced, which was
strongly stained with anti-tubulin antibody.
It was constructed inside the plasma mem-
brane and appeared to shrink as the treat-
ment was prolonged. The globular MT mass
consisted of many MTs. In eggs treated for
7 min, only short MTs were found. In eggs
treated for 30 min, long MTs were observed
in the inner region of the structure. In
the outer region of the structure, some
short MTs were seen to be nucleated from
MTOG-like granules. When treated for 2 h,
the globular MT mass was margined by aggre-
gates of MTOGs, from which bundles of short
MT elongated toward inside. After the HG
treatment, the 51 kDa protein was shown to
be condensed in a thin layered ring which
seemed to correspond to the outer region of
the globular MT mass. These observations
provided evidences that many MTOGs are
distributed throughout the egg cytoplasm
before mitotic phase and the 51 kDa protein
is one of the components of MTOG.

DB 66

CORRELATION BETWEEN ACCUMULATION OF WGA
RECEPTORS AND FORMATION OF CONTRACTILE RING
IN SEA URCHIN EGGS.

T.Yoshigaki, S.Maekawa, and H.Sakai. Dept.
of Biophys. and Biochem., Fac. O£ Scl.,

The Univ. of Tokyo, Tokyo.

At late anaphase, the accumulation of
WGA receptors occurred coincidentally with
the appearance of microfilaments for the
contractile ring (CR), and the distribution
of accumulated WGA receptors was just super-
imposed to that of the CR. Microfilaments
of the CR and WGA receptors were dispersed
together by treatment with colcemid, so
that the eggs failed to cleave. This
observation suggested that WGA receptors
were directly or indirectly associated with
microfilaments of the CR, and that micro-
tubule cytoskeleton confined the WGA recep-
tors, therefore, microfilaments to a
restricted region to accomplish the forma-
tion of the CR. Cytochalasin B inhibited
the accumulation of WGA receptors and the
formation of the CR. In contrast, the drug
was not capable of dispersing WGA receptors
which had already accumulated in spite of
disappearance of the CR. When an egg had
three asters by polyspermy, microfilaments
of the CR accumulated in a form of the
triangle reverse to the one formed by con-
necting the three astral centers. This
result leads us to propose an idea that
each aster makes WGA receptors to migrate
toward the equatorial surface, even tripol-
ar spindles are formed with three equatori-
al planes.

DB 67

INDUCTION OF CONTRACTILE RING-LIKE
STRUCTURE BY CALYCULIN A IN SEA URCHIN
EGGS.

H. Tosuji!, I. Mabuchi2, K. Miyaji!, Y. Kato3, N. Fusetani3
and T. Nakazawa!.!Dept. of Biol., Fac. of Sci., Toho
Univ., Funabashi, 2Dept. of Biol., Coll. of Arts & Sci.,
Univ. of Tokyo, Tokyo and 3Lab. of Mar. Biochem., Fac. of

Agric., Univ. of Tokyo, Tokyo.

Calyculin A, isolated from the marine sponge
Discodermia calyx, is known to inhibit protein phosphatase
type-1 and type-2A activities. When both fertilized and
unfertilized sea urchin eggs were exposed to calyculin A,
contractile apparatus was formed in the cortical layer and it
constricted the egg like the contractile ring in the cleavage
furrow. This constriction was inhibited by cytochalasin B.
Actin filaments were found to be concentrated in the
contractile apparatus. Myosin was also concentrated in the
apparatus. In addition, WGA receptors were also co-localized
with the actin filaments. The apparatus was formed by
assembly of filamentous structures appeared in the egg cortex.
The assembly of the filamentous structure was postponed in
the presence of either quercetin or staurosporine, which is
known to inhibit protein phosphorylation.

Neither mitotic apparatus nor aster-like structure was
found in the calyculin A-treated eggs. Furthermore, the
constriction was not inhibited by a microtubule inhibitor,
colchicine or griseofulvin. We conclude that the contractile
ring-like structure was constructed in sea urchin eggs
independently of microtubular structures by the effect of
calyculin A.

DB 68

EFFECT OF COLCHICINE ON FURROW FORMATION
IN AMPHIBIAN EGGS.

T.Sawai. Dep. of Biol., Fac. of General Education,
Yamagata Univ., Yamagata.

It is well known that colchicine inhibits cleavage
furrow formation by impairing the growth of micro-
tubules of the mitotic apparatus. Beside, we have
demonstrated that the cleavage furrow was formed
by the interaction of two factors, cytoplasmic factor
(FIC) and cortical one, in amphibian eggs.

The present experiments were purposed to exa-
mine the effect of colchicine on the furrow forma-
tion by the two factors, in newt eggs. heyexs
periments were made in two systems, i.e., ina
furrow-induction system by transplantation of FIC
and in the normal process of furrow progress.

In the first experiment, colchicine solution (5/10
mM - 1/10 pM , 70-100 nl) was injected into one
blastomere of two-cell embryos. These blastomeres
were inhibited the second cleavage in many cases.
In such cases, they were transplanted FIC taken
from another cleaving eggs beneath the cortex of
the animal half, just after the second-cleavage fur-
row appeared on their sister blastomeres. In
operated cells, the furrow induction by FIC did not
occur in almost all cases. In next experiment,
colchicine solution was injected into eggs during
first cleavage, at the site under the cortex of small
initial furrows, an advancing tip of furrows, or
deepening furrows. In each case, colchicine injec-
tion severely prevented the progress of furrows
and resulted in regression of established furrows,
in many cases.

These results suggest that microtubules take
part in the furrow formation by the interaction of
FIC and cortical factor.

Developmental Biology 1099

DB 69

THE SPINDLE FORMATION AND PARTICIPATION OF
MITOTIC STRUCTURES TRANSPLANTED INTO THE
OOCYTE AND ZYGOTE OF THE STARFISH,
ASTERINA PECTINIFERA IN CELL DIVISION
EXCEPT FOR MEIOTIC DIVISION.

T.Saiki and Y.Hamaguchi. Biol. Lab., Fac.
OLmoCHe Lokyoy insite) Of Lechnols,) Tokyo.
A OR AES ee Re ee ee oe oe i a,

When mitotic structures (the nucleus,
aster, or the mitotic apparatus) were
transplanted into starfish oocytes, they
did not functioned in the oocytes during
meiosis. However, after insemination, the
transplanted mitotic structures functioned
during the first cleavage, and the
recipient zygotes cleaved into more than
three blastomeres, whose number increased
by the number of the aster(s)
transplanted. When the recipient oocytes
were activated by the treatment of Ca
ionophore, they did not cleaved, except
for the case that the whole mitotic
apparatus was transplanted. These results
indicate that the mitotic structures have
the ability to participate in cell
division after meiosis but not during
meiosis.

Secondly, we investigated spindle
formation when we changed the distances
from the nucleus to two asters in the cell
into which a nucleus was transplanted, and
in the cell where its own nucleus was
displaced by micromanipulation. In both
cases, the half spindle formed when the
distance between the aster and the nucleus
was 22 um or less.

DB 70

THE EFFECT OF UV MICROBEAM IRRADIATION ON
THE CLEAVAGE OF SEA URCHIN EGGS

M. Ikeda & K. Dan. Dept. of Biol. Keio
Univ. Yokohama.

with an appendix by S. Kyono & M. Yoneda

Spot irradiation (7 zem in diameter)
was carried out in the cytoplasm at a
polar, a subpolar and a equatorial in
reference to the spindle at the first
cleavage. This resulted a delay in the
division at the second cleavage. In the
delay, sister cells cleave at the same
time, when the spot was at a equal dis—
tance to both blastomeres. The irradiated
spot was unilateral to the blastomeres,
however, the blastomere with the ir-—
radiated spot cleaves later than the
other. This allows us to postulate that
UV irradiation forms a cleavage—
arresting substance in the cytoplasm
that diffuses within the cell.

To define the nature of the arresting
substance, its diffusion was compared
with eosine and cytochrome C within an
agar gel ball. By the analysis in the
appendix, the dxf fusion coefficient of

eosin_is *x10-cm /sec., cytochrome C is
2x10 cm /sec. an cleavage arresting
substance is 3x10 cm /sec..

These diffusion coefficient and
molecular weight of eosin(691. 7) and
cytochrome C(12, 384) give 18, 400 as the
molecular weight of cleavagge-arresting
substance.

DB 71

CYCLIC CHANGES IN INTRACELLULAR FREE
CALCIUM LEVELS IN CLEAVING AND CLEAVAGE-
ARRESTED XENOPUS EGGS.

H.Y.Kubota!, Y.Yoshimoto? and Y.Hiramoto?.
UDSoes Oe Aoolls, Weyes Or Sebo, Mvore@ Winns ;
Kyoto, 2Fac. of Culture, Kansai Med. Univ.,
Hirakata, 3Univ. of the Air, Chiba.

The synthetic aequorin (gift from Dr.
Shimomura) solution was injected into
Xenopus eggs at 20-30 min following ferti-
lization and the luminescence emitted from
the egg cortex was monitored with a photo-
multiplier. The aequorin luminescence
increased gradually and became plateau
shortly before the appearance of the first
cleavage furrow. After the furrow was
visible, the luminescence increased
abruptly and reached its peak between the
first and second furrow formation. The
luminescence reached the bottom when the
second cleavage furrow was first visible.
This change was repeated in the following
several round of cleavages. To avoid the
influence of the furrowing on the lumines-
cence, colchicine solution was injected
into the eggs prior to the injection of
the aequorin. The luminescence from these
eggs was also changed cyclically,
suggesting that the calcium concentration
eycled actually in the egg cytoplasm.
Examination of the histological sections
of the eggs suggests that the period of
the increased calcium concentration
corresponds to that of the metaphase, and
decreased calcium to the karyomere-phase
and interphase.

DB 72

THE REACTIVITY OF MONOCLONAL ANTI-TUBULIN
ANTIBODY WITH MECROTUBULES IN VIVQ. 1
M.T.Oka , Y.Hamaguchi and T.Arai“~. Biol.
Lab., Fac. Sci., Tokyo Institute of
Technology, Tokyo, “Depart. of Applied
Biol. Sci., Science Univ. of Tokyo, Noda.

First, staining patterns of the mitotic
apparatus (MA) were examined in the
extracted but unfixed sea urchin egg by
indirect immunofluorescence using 7 mono-
clonal anti-tubulin antibodies. Two of
them (YL1/2 and TUB2.1) stained MA in the
unfixed egg but the others (D2D6 etc.) did
not, although they reacted specifically
with sea urchin egg tubulin and stained MA
of the extracted and methanol-fixed sea
urchin egg. Secondly, the effect of the
antibodies on MA in the living egg was
examined by means of microinjection. YL1/2
and TUB2.1 disassembled the microtubules in
the dividing egg shortly after injection.
On the contrary, D2D6 did not affect them
or slightly disassembled them. Moreover,
fluorescently labeled D2D6 did not appear
to bind to the microtubules. These results
indicate that the antibody which stained
microtubules in the unfixed egg depolymeri-
zed microtubules in the living egg, and,
however, that the antibody which did not
stain them in the unfixed egg did not
affect them in the living egg. Therefore,
it is suggested that the staining ability
of monoclonal anti-tubulin antibody in the
unfixed egg exactly reflects its reactivity
with microtubules in the living egg, but
that in the fixed egg does not.

1100 Developmental Biology

DB 73

DYNAMICS OF THE MIGRATION OF OIL DROPLETS
UPON FERTILIZATION IN THE EGG OF MEDAKA,
ORYZIAS LATIPES WITH A CENTRIFUGE
MICROSCOPE.

M.Narusé and Y.Hamaguchi. Biol. Lab., Fac.
o£ Sci. ,. Tokyominst of. Technol.) erokyo.

After fertilization, oil droplets
evenly dispersed in the cytoplasm of
Medaka eggs, migrate toward the vegetal
pole. Nocodazole, which depolymerizes
microtubules, inhibited the migration of
the oil droplets at 10 uM, but dihydro-
cytochalasin B, an inhibitor of
microfilaments, did not inhibit their
migration. The oii droplets in
nocodazole-treated eggs were forced to
move eaSily toward the centripetal
direction during the centrifugation using
a centrifuge microscope. Microtubule
network was detected in the fertilized
eggs by immunofluorescence. These results
suggest that microtubules generated the
forces which moved the oil droplets. When
the egg was centrifuged as its vegetal
pole was set toward the centrifugal
direction, the oil droplets were pushed
back toward the animal pole, which was in
the centripetal direction. In this case,
the maximum force to push back the oi)
droplets was calculated to be 2.0x10 ” N.
This value is estimated to be the maximum
motive force which made the oil droplets
migrate toward the vegetal pole.

DB 74

THE SPINDLE-AXIS DURING UNEQUAL CELL
DIVISION IN GRASSHOPPER NEUROBLASTS.
K. Kawamura. Lab. of Biol., Rakuno Gakuen
Whelbti7n pl J OVNER ILA Je) SIRe BUC le)
The unequal cytokinesis of grasshopper
neuroblast(NB) are caused by eccentric
location of the spindle body which
maintains a definite polarity. An
electron dense layer(DL) appeared in the
cap cell(CC)-side cortex at very late
prophase, became fairly rude by the
beginning of middle anaphase, and
disappeared completely at middle anaphase.
The spindle-axis always directed toward
the center of the DL. The present study
analyzed the mechanism of maintenance of
spindle direction along the CC-ganglion
cell(GC) axis of the NB by microdissection
technique. Rotated spindle at metaphase
returned to the original axis without
fail, while the recovery rate decreased in
later stages. The ability to recover the
original spindle-axis occurred only while
the DL existed. When the rotated spindle
at the beginning of middle anaphase could
not return accurately to the original
axis, the spindle-axis was corrected to
some extent by the squeezing action of
cortical contraction wave(CW). At the
middle anaphase, a definite area of GC
Side cortex acquires a character to grasp
a spindle pole. The definite polarity of
spindle-axis during cell division of NBs
seems to be maintained by triple assurance
such aS appearance of DL, propagation of
CW, and grasping action of GC side cortex.

DB 75

POLARITY REVEALED IN THE CELL SURFACE AND
CORTEX OF UNEQUALLY DIVIDED NEUROBLASTS.
N. Yamashiki, Biology, Rakuno Gakuen
Univ., Ebetsu, Hokkaido.

The grasshopper neuroblast has a
contact with cap cells (CCs) on one side
of the cell, and a small daughter ganglion
cell (GC) is always produced on the
opposite side throughout many series of
cell division. The regional difference in
ultrastructures was detected in the cell
surface and cortex. During the period
from late prophase till early anaphase,
the CC-side polar area is characterized by
villous projections and an electron-dense
cortical layer, while on GC-side no such
structures are observed. The metaphase
spindle directs to the center of the dense
layer, and it shifts to GC-side after
disappearance of the characteristic polar
structures. When a large daughter cell
with a neuroblast-type nucleus was
produced on GC-side by micromanipulation,
it had no contact with CCs. Two types of
unequal cytokinesis as to the polarity
were revealed in the subsequent division.
The electron-dense materials always
appeared in the cortical layer of one side
where the metaphase spindle was to be
located. These results suggest that
CCs assist the establishment of the
definite polarity in the cell, although
without CC's aid neuroblasts can develop
some polarized structures.

DB 76

DISTRIBUTION AND ROLE OF POLYMERIZED ACTIN
IN THE LAMPREY EGG DURING FERTILIZATION.
W. Kobayashi. Zool. Inst., Fac. Sci.,
Hokkaido Univ., Sapporo.

Using rhodamine phalloidin staining, the
distribution of actin was studied during
fertilization in the eggs of lamprey,
Lampetra japonica. The surface of
unfertilized eggs had numerous weak
punctate fluorescence sites at the animal
pole. Comparison of this pattern with
surface morphology of the animal pole
indicates that filamentous actin is mainly
localized in the microvilli. Upon
insemination of the egg, a fertilizing
sperm fused with egg by the acrosomal
filament when its nucleus remained outside
the egg coat. During perivitelline space
formation, many cytoplasmic processes
appeared at animal pole. When these eggs

were stained for polymerized actin, these
processes showed intense fluorescence,

indicating the accumulation of filamentous
actin. The fertilizing sperm passed
through the egg coat and moved along one
of these processes. The fluorescence
pattern at the incorporation site of sperm
was indistinguishable from that observed
in other processes without sperm.
Cytochalasin B inhibited both the
formation of the processes and the
movement of sperm through the coat. These
results suggest that actin filaments in
the cytoplasmic process are required for
the passage of sperm through the egg coat.

Developmental Biology 1101

DB 77

BIPOLAR ORGANIZATION OF CORTICAL ACTIN IN THE
TUBIFEX EGG. s

Takashi Shimizu, Zool. Inst., Fac. of Science,
Hokkaido University, Sapporo 060, Japan

Organization of cortical actin in Tubtfex eggs
from fertilization through the second polar body
formation (PBF) has been examined in isolated
cortices and whole mounts stained with rhodamine
phalloidin. When eggs are fertilized, they exhibit
intense fluorescence, as a result of microvillar
elongation, all over the surface except a circular
zone (30 pm in diameter) which appears as a ‘black
hole' due to much weaker fluorescence. Thereafter
cortical actin gradually decreases in amount; only
tiny aggregates of actin are seen in the cortex.
During the first PBF, cortical actin appears to be
localized at the animal pole and at grooves formed
by the deformation movement. Shortly after the
PBF, however, no regional difference is detected in
the density of cortical actin. It is not until
the second PBF that cortical actin is organized
bipolarly. That is, cortical actin is distributed
as a gradient increasing from the equator to the
pole region in the animal and vegetal hemispheres
of the egg. When eggs begin the second deformation
movement accompanying the second PBF, streaks of
actin bundles form in the cortex of the equator of
the egg. They run meridionally and link actin
sheets of the animal and vegetal hemispheres,
Suggesting their involvement in groove formation
at the equator of the egg. Polarized distribution
of cortical actin is retained even after the end
of the deformation, while equatorial actin streaks
disappear this time. These observations suggest
that bipolar organization of cortical actin is
generated shortly before the second PBF.

DB 78

UNEVEN DISTRIBUTION OF MITOCHONDRIAL
ACTIVITY IN DROSOPHILA EARLY EMBRYOS.
T.Akiyama and M.Okada. Inst. Biol. Sci.,
Univ. of Tsukuba, Tsukuba _ 305

Vital staining of mitochondria with the
fluorescent dye rhodamine 123 was per-
formed in Drosophila early embryo. Rhoda-
mine 123 accumulates specifically in mito-
chondria of living cells. This accumula-
tion depends on the high transmembrane
potential characteristic of functional
mitochondria. We found that the fluores-
cence of rhodamine 123 was unevenly dis-
tributed and that the distribution pattern
showed developmental changes.

Laser confocal scanning microscopy
demonstrated in early cleavage embryos
that the fluorescence intensity in the
posterior pole plasm was 2.7-fold strong-
er than in the other periplasmic region.
In late cleavage and pole bud stage
embryos, no regional difference was ob-
served in the fluorescence. In syncytial
blastoderm, the fluorescence in pole cells
was 2.4-fold stronger than in the other
periplasmic region. In cellular blasto-
derm, the fluorescence in pole cells was
2.2-fold stronger than in somatic cells.

Transmission electron microscopy of
early embryos revealed that mitochondria
were not concentrated in the region with
strong rhodamine 123 fluorescence. This
suggests that mitochondria with a rela-
tively high respiratory activity are
distributed in the region stained with a
rhodamine 123 strongly.

DB 79

INTERSPECIFIC TRANSPLANT OF CYTOPLASMIC
FACTORS DETERMINING ANTEROPOSTERIOR POLARI-
TY IN DROSOPHILA MELANOGASTER

S.Sugiyama and M.Okada. Inst.of Biological
Sciences, Univ.of Tsukuba, Tsukuba, Ibaraki

The phenotypes of D.melanogaster embryos
with mutations in genes representing the 3
gene classes (anterior,posterior & termi-
nal) which are required for the determina-
tion of anteroposterior (A-P) polarity can
be rescued by the transplant of cytoplasm
from wildtype embryos. As the first step in
learning to what extent the A-P polarity
mechanism known in D.melanogaster is con-
served through evolution, we transplanted
cytoplasm (0.2nl = 2% of D.m. egg volume)
from Drosophila virilis wildtype embryos
into such D. melanogaster mutant embryos.
So far, rescue of anterior gene (bicoid,
19/32=59%) and posterior gene (nanos,
36/44=82%) mutant phenotypes, but not that
of the terminal gene phenotype (torso,
0/97=0%) have been observed. Cytoplasm
(0.05-0.15nl) from embryos of a chironomid
midge (Paratanytarsus parthenogeneticus),
although causing severe developmental
defects rescued the nanos phenotype (9/32=
28%), whereas no positive results have yet
been obtained with bicoid (0/32= 0%) and
torso (0/39=0%). The data suggest that
while the determination of A-P polarity is
functionally conserved to some extent
within the order Diptera, certain differ-
ences may exist even at the Drosophila
genus level.

DB 80

SPLICING OF THE P-ELEMENT IN POLE CELLS
DURING EARLY EMBRYOGENESIS OF DROSOPHILA
MELANOGASTER

S. Kobayashi, T. Kitamura and M. Okada.
Inst. Bolte o Cisse WUmaivs. of Tsukuba,
Tsukuba, Ibaraki.

It has been suggested that the third
intron of the P-element is excised only in
germ line cells. To test directly whether
splicing occurrs in pole cells, the
progenitor of germ line, during early
embryogenesis, we designed a histochemical
assayS to measure the splicing activity.
The lacZ gene, encoding £-gal, and the
heat shock promoter of the hsp 70 gene were
inserted into the downstream of the third
intron and into the promoter region of the

P-element, respectively. Flies were
transformed with the fusion gene. In the
transformant, the cells where the third

intron is spliced out should be visualized
with the staining for 6-gal activity. When
early syncytial blastoderms were heat-
treated,8-gal activity was detected only in
part of pole cells. Furthermore, we showed
that the pole cells with the 8-gal activity
penetrated embryonic gonads and to
differentiate into primordial germ cells.
Taking into account that not all pole cells
can differentiate into germ cells and that
most PGCs in embryonic gonads showed ££ -gal
activity, the results suggest the
possibility that the splicing machinery of
P-element has an important role in the
developmental process of pole cells to germ
cells.

1102 Developmental Biology

DB 81

P-ELEMENT IS SPLICED ALSO IN NERVE
CELLS DURING EARLY EMBRYOGENESIS’ OF
DROSOPHLA MELANOGASTER
T. Kitamura, S. Kobayashi and M. Okada.
ince, vone” Jejskol, Sears > Wha. ae © AUSio <ovoyele
Psukubes:-slbaraki—= a ee

It has been suggested that the third
intron of the P-element is only spliced in
germ line cells. We showed in the
accompanied paper that the splicing
activity can be traced back to pole cells,
the progenitor of germ line. We here show
that somatic cells, particularly nerve
cells, also have the splicing activity in
later stage embryos. The activity was
histochemically detected as -gal specific
staining. To establish the assay system,
the P-element fusion gene was constructed
with the insertion of the heat shock
promoter from the hsp70 gene and of the
lacZ gene into the promoter region and the
downstream of the Ghebad aL aie eC
respectively. In the stages at start of
neuroblast formation through neurogenesis,
in trsnsformants of the fusion gene, -gal
activity was detected in neuroblasts. In
other somatic cells, -gal activity was
also detected but in different stages.
Based on these results we are inclined to
suggest that the splicing activity so far
detectable only in excision of P-element
can participate in the genetic regulation
of differentiation of neuroblasts and other
somatic tissues.

DB 82

LOCALIZATION OF DORSAL AXIS DETERMINANTS
IN AN EARLY EMBRYO OF Xenopus laevis

M. Fujisue, Y. Kobayakawa, M. Yuge, & K.
Yamana, Dept. of Biol. Wes Cie Senlo5g ©
Coll. of Gen. Educ., Kyushu Univ. Fukuoka.

We have shown that cytoplasm injection
from the equatorial region of dorsal vege-
tal cells of 16-cell embryos causes secon-
dary axis in 42% of simultaueous reci-
pients.

We examined the localization of a tran-
splantable axis-determining activity by
injecting cytoplasm into ventral vegetal
cells at the 16-cell stage. At the 16-cell
stage, cytoplasm from the lateral dorsal
vegetal cells or the vegetal pole region
of dorsal cells showed only 5% inducing
activity ( 5% recipients had secondary
axis). At the 2-cell stage, 17% activity
was found in cytoplasm of the vegetal
pole. However at the l-cell stage, any
equatorial region cytoplasm does not have
the activity but cytoplasm from vegetal
pole showed 60% activity. UV irradiated
eggs that developed into axis-deficient
embryos (IAD more than 4) were also exa
mined. At the l-cell stage, as in normal
embryo, activity localized in the vegetal
pole. However "dorsal" cytoplasm of the
16-cell stage showed no activity but the
high activity was found in the vegetal
pole. These results suggest that in normal
development axis-determinants move from
the vegetal pole to the equatorial region
after the 2-cell stage.

DB 83

RELATIONSHIP BETWEEN RIGIDITY OF THE EGG
CYTOPLASM AND TWINNING IN RESPONSE TO
CENTRIFUGATION IN XENOPUS.

M. Wakahara(1) and A.W. Neff(2). (1) Zool.
Insits, Lacs Seis, Hokkarldo, Univer,
Sappaoro, (2), Medi= Sci. Progs,.ocaeaMeden
Indiana Univ., Bloomington, IN, USA.

Centrifugation of fertile Xenopus
eggs at right angles to the animal/vegetal
axis before first cleavage can induce
bifurcation of the embryonic axis
(twinning). However, eggs from individual
spawnings show considerable variation in
twinning in response to centrifugation.
Eggs from individual spawnings also show
considerable variation in the rigidity of
the cytoplasm. Thus, the mean cytoplasmic
rigidity which was measured in inverted
fertile eggs as cytoplasmic immobility
(CIM) was compared to the degree of
twinning (either twinning index - TI or
percentage of eggs that formed twins, %T)
in 20 different spawnings. Fertile eggs
from those same spawnings were embedded in
gelatin and centrifuged either at 15g or
at 30g for 4 min at right angles to their
a/v axis with their sperm entrance sites
facing the center of the centrifuge at 40%
of the time between fertilization and
first cleavage. Each spawning had a
distinctive CIM and a characteristic TI
amd %$T. There was a significant negative
correlation between the mean CIMs and the
TIs (e.g., correlation between CIM and TI
at 15g was r=-0.08). Low CIM eggs produced
more twins.

DB 84

AUTONOMOUS FLUORESCENCE LOCALIZED AT THE
POLAR LOBE OF OYSTER EMBRYO.

T.Deno. Dept. of Biology, Osaka Kyoiku
University, Osaka.

Under ultraviolet illumination, oocytes
washed out from dissected ovaries of the
oyster, Crassostrea gigas and the related
species, Saxostrea echinata were found to
emit a specific yellowish pale blue
autonomous fluorescence in the cortical
region of the vegetal pole. After
insemination, this fluorescence remained at
the same region of the fertilized eggs and
when the foleIesit cleavage began, the
fluorescence was segregated to the vegetal
cortical region of the polar lobe of the
embryos. At the 2-cell stage, the polar
lobe was incorporated into the CD
blastomere and the fluorescence localized
at the vegetal cortical region of the CD
cell. During the second cleavage, the
fluorescence was localized at the vegetal
cortical region of the polar lobe again,
and then incorporated into the D blastomere
with the polar lobe cytoplasm after
completion of the cleavage. However, the
intensity of the fluorescence gradually
became reduced and so it was difficult to
distinguish the specific fluorescence from
background one after third cleavage.

With the centrifugation at 2000X g for
10 min, the cytoplasm of the oocyte of
C. gigas was stratified into three layers
and the autonomous fluorescence was also
moved and localized at the centripetal
region of the most centrifugal layer.

Developmental Biology 1103

DB 85
APPEARANCE AND LOCALIZATION OF TRANSCRIPTS
OF THE MYOPLASMIN-C1: A POLYPEPTIDE

SPECIFIC FOR THE MYOPLASM OF ASCIDIAN EGGS.
T. Nishikata, Dept. of Zoology, Kyoto Univ.

In the ascidian egg, muscle determinants
are thought to be sequestered in the so-
called myoplasm. We established 12
monoclonal antibodies (Mabs) specific for
the myoplasmic components of Ciona
intestinalis eggs. Antigenic polypeptides
recognized by these Mabs were named
"Myoplasmin" after their specific
localization. We characterized a 40-kDa
myoplasmin-Cl as a candidate of the muscle
determinant because the myoplasmin-Cl
antibody suppresses muscle differentiation
when injected into fertilized eggs. We
isolated a myoplasmin-Cl cDNA probe _ by
screening cDNA library cf the ovary using

the specific Mab. Myoplasmin-Cl
transcripts appeared as a single band of
about 3kb by Northern blot analysis. The

in situ hybridization analysis on the
sectioned specimens of Ciona ovary revealed
that the myoplasmin-Cl mRNAs were abundant

in the small oocytes, which also have
considerable amount of myoplasmin-Cl. We
are now trying to isolate full length
myoplasmin-Cl cDNA. Analysis of the

myoplasmin-Cl molecule will provide us a
cue to the molecular understanding of the
muscle determinants in the ascidian egg.

DB 86

MUSCLE CELL DIFFERENTIATION IN BLACK EGG-
FRAGMENTS OF THE ASCIDIAN CIONA SAVIGNYI.
Y.Marikawa, T.Nishikata and N.Satoh.

Dept ef -Zooll., Fac. of Sei., Kyoto Univ.,
Kyoto

The myoplasm of eggs of the ascidian
Ciona savignyi is characterised with black
tone. When dechorionated unfertilized
eggs were centrifuged at 1500g for 15 min,
eggs were divided into centripital black
fragments and centrifugal red fragments.
Female pronucleus moved into black fragment,
Observation with an electron microscopy
demonstrated that black egg-fragments con-
tained clear cytoplasmic components and
electron-lucent particles while red egg-
fragments were filled with yolk particles.
Most of the cortical regions of black frag-
ments were stained with the Myoplasmin-Cl
antibody while the antibody stained small,
centripital region of red fragments.

When inseminated, both egg-fragments
developed into partial embryos. More than
80% of black partial embryos showed muscle
differentiation detected with a specific
antibody and many cells expressed the spec-
ific antigen. On the other hand, muscle
development was noticed in about 50% of
the red partial embryos but a few cells
expressed the antigen. Epidermal develop-
ment was more conspicuous in red partial
embryos than in black partial embryos.

These results suggest that cytoplasmic
determinants for muscle differentiation are
preferentially partitioned into black egg-
fragments in this ascidian species.

DB 87

DIFFERENTIAL LOCALIZATION OF PROTEASOME IN
EGGS DURING ASCIDIAN FERTILIZATIION AND
DEVELOPMENT.

H. Kawahara and H. Yokosawa. Dept. of
Biochem., Fac. of Pharmaceutical Sci.,
Hokkaido Univ., Sapporo.

Proteasome (multicatalytic proteinase)
is a high-molecular weight protein complex
composed of several components and shows
several different protease activities.

We have previously reported that a
chymotrypsin-like enzyme purified from
eggs of the ascidian Halocynthia roretzi,
which is involved in a process leading to
cell cleavage, is a proteasome.

To define the role of proteasome during
ascidian fertilization and development, we
analyzed the precise localization of
proteasome in H. roretzi unfertilized eggs
and the fertilized eggs undergoing the
cleavage, especially, from the two-cell
stage to the four-cell stage, by
fluorescence immunohistochemistry using
monoclonal antibodies raised agaist
protein components of H. roretzi egg
proteasome. The proteasome (or a
component with isoelectric point of 6.3)
was found to be present in the cytoplasm
of the unfertilized eggs and in both the
cytoplasm and the nucleus of the
fertilized eggs at the interphase stage of
the two- or four-cells. Furthermore, it
was found that the proteasome underwent
differential changes in its localization,
dependent on the progression of cell
cleavage cylcle.

DB 88

EXPRESSION OF A MUSCLE ACTIN GENE DURING
ASCIDIAN EMBRYOGENESIS.

T.G.Kugsakabe ; J.Suzukil, H.Saiga@,
N.Satoh!, K.W.Makabe!. 1) Dept. of Zool.,
Pees GiE S@ils Fy MCOEO Winil\yop INoIE@o” ~~ D)) > Sine
Med., Univ. Occu. Env. Health, Kitakyushu.

Differentiation of muscle cells in
ascidian embryos seems to be controlled by
factors localized in the egg cytoplasm. In
an attempt to investigate the molecular
mechanisms of muscle differentiation in
ascidian embryos, we isolated five clones
of muscle-type actin genes from cDNA
library of Halocynthta roretzi tailbud
embryos. Sequence analysis showed that the
coding regions of these clones are well
conserved, while the 3’ noncoding region has
no Similarity to each other.

We examined the expression pattern of
one of these clones, cA4, by in situ
hybridization and northern blot analysis
using a probe derived from 3' noncoding
region of the clone. In situ hybridization
showed that the transcription of cA4 actin
gene is restricted to muscle cells of
tailbud embryos. The cA4 mRNA was not
detected in unfertilized eggs, fertilized
eggs, and cleavage stage embryos. The
transcripts were detected first in gastrula
stage embryos and accumulate until late
tailbud stage. After metamorphosis, the
transcripts disappeared. These results
suggest that the transcription of cA4 actin
gene is activated at early stage of
differentiation of muscle lineage cells.

1104 Developmental Biology

DB 89

EXPRESSION OF A MYOSIN HEAVY-CHAIN GENE IN
THREE TYPES OF MUSCLE CELL IN JHE ASCIDIAN
HALOCYNTHIA RORETZI. I, Araki . H.Sajga?,
T.Nishikata!, K.W.Makabe+ and N.Satoh

Ipept. gf MONG pW G Cie Seto I7OEo Wala qe
Kyoto, <Dept. of Mol Biol Sch. of Med,

Univ. of Occup. Environ. Health, Kitakyushu,

Three types of muscle are formed during
a life histroy of ascidians; unicellular

and striated muscle cells of tadpole larvae

multicellular and nonstriated muscle of
adult body wall and unicellular and nonstr-
iated muscle cells of adult heart. We have
two probes to examine the occurrence of
myosin heavy chain in the muscle tissues.
Mu-2 is a specific monoclonal antibody

which was raised against larval muscle cell,

cM11 is a specific antisense RNA probe
isolated from cDNA library of tailbud em-
bryos with the Mu-2. cM11 corresponds to
a region near the center of the rat myosin
heavy-chain entire sequences.

When occurrence of myosin heavy chain
was examined with Mu-2, positive reactions
with the muscle-specific antibody were
detected in both adult body-wall muscle and
heart muscle. On the other hand, no tran-
scripts were detected in adult body wall-
muscle and heart muscle when examined with
cM11 antisense probe. Since Southern
blotting analysis suggested that there is
one copy of the myosin heavy chain gene per
Halocynthia haploid genome, there is a pos-
sibility that expression of myosin heavy
chain gene is regulated by an alternative
splicing.

DB 90

AN ATTEMPT TO ISOLATE cDNA CLONES SPECIFIC
FOR EPIDERMAL CELLS OF THE ASCIDIAN EMBRYO.
T.D.Ueki, K.W.Makabe and N.Satoh

DNNBG Ol ZOO, WAGs Ol: SCs, [O7Oe© Wino
Kyoto

When divisions of fertilized eggs of the
ascidian Halocynthia roretzi were continu-
ously blocked with cytochalasin B,
cleavage-arrested one-cell embryos were
found to be able to develop features of
epidermal cell differentiation and they did
not produce differentiation markers of
muscle, endoderm, notochord and trunk late-
ral cells. Poly(A)* RNAs were isolated
from cleavage-arrested one-cell embryos and
fertilized eggs, and were analyzed by in
vitro translation system. Analysis of [*S]
methionine-labelled polypeptide products
by two-dimensional gel electrophoresis
showed several distinct mRNAs specific to
cleavage-arrested one-cell embryos.
cDNA library of cleavage-arrested one-cell
embryos and that of fertilized eggs were
prepared. By differential screening of
these libraries, we are now trying to
isolate cDNA probes specific for epidermal
cells of Halocynthia embryos.

DB 91

EXPRESSION OF A MUSCLE MYOSIN HEAVY-CHAIN
GENE IN QUARTER ASCIDIAN EMBRYOS.
S.Fujiwara2, H.Nishida', N.Satoh and
K.W.Makabe. Dept.of Zool., Kyoto Univ.,
Kyoto. '!Dept.of Biol., College of Liberal
Arts, Kobe Univ., Kobe. #Present Address:
Dept of Biol. Kochi Univ., Kochi.

Ascidian tadpole larvae have tail muscle
cells of two distinct lineages. Primary
lineage muscle cells originate from B4.1
blastomere-pair of the 8-cell embryo and
secondary lineage cells originate from A4.1
and b4.2 blastomere-pairs. In this study,
blastomere-pairs isolated from 8-cell embryos
of the ascidian Halocynthia roretzi were
raised as quarter embryos, and the expression
of muscle myosin heavy-chain mRNAS was
examined by whole-mount in situ hybridization.
As was expected, all of the B4.1 quarter
embryos showed strong expression of the
transcript, corresponding to the develop-
mental potential of primary lineage muscle
cells to differentiate autonomously into
muscle. On the contrary, none of the A4.1
and the b4.2 quarter embryos showed
expression of the transcript at all.

This result agrees with the previous data
that secondary lineage prospective muscle
cells in quarter embryos failed to develop
muscle-specific proteins including myosin
heavy-chain. The result suggests that the
specification of secondary lineage muscle
cells is regulated at the transcriptional
level but not at the translational level.

DB 92

TRIALS TO TRANSFER EXOGENOUS GENES INTO
ASCIDIAN EGGS.
A.Hikosaka, N.Satoh and K.W.Makabe.

Dept. "of "Zool. “Hac. (of Sein KyoromUinrnaer,

Kyoto
We are now trying to establish a
transgenic system of ascidians. First, to

test the damages of embryos by DNA
molecules transfered, we microinjected DNA
solutions at various concentration into
dechorionated fertilized eggs of C.

intestinalis, and assayed the rate of

normal development.We found that embryos
injected with 0.01-0.1 ug/ml DNA solutions
could develop without serious damages.

We constructed fusion genes pH3-CAT.
These fusion genes contain the coding
region of bacterial chloramphenicol
acethyl-transferase (CAT) gene, and various
length of 5' upstream region of histone H3
gene of H. roretzi. (This region is
considered to contain regulatory region of
the H3 gene.) We injected these fusion
genes into the cytoplasm of dechorionated
fertilized eggs of H. roretzi. We cultured
the eggs on agar-coated dishes, and assayed
the expression of CAT protein in the
embryos by immunohistochemical staining.

We found that embryos injected with the
longest pH3-CAT gene expressed CAT protein
when examined at blastula and gastrula
stages, while embryos injected with the
shorter pH3-CAT genes did not show the
expression of CAT protein.

Developmental Biology 1105

DB 93

CDNA CLONING AND EXPRESSION OF NA*,K~*-
ATPASE a-SUBUNIT IN EMBRYOS OF THE SEA
URCHIN, Hemicentrotus pulcherrimus.
K.Mitsunaga’?, K.Yamazakit, M.Hatoh2-2, K.
Yamada*, K.AkKasaka#?, H.Shimada® and I.
Yasumasu+. +*Dept. of Biol., Sch. of Educ.,
Waseda Univ., Tokyo, #Applied Plant Lab.,
Japan Tobacco Inc., Yokohama, #Zool. Inst.,
Fac. of Sci., Hiroshima Univ., Hiroshima.

We have reported several cDNAS
identified by screening a sea urchin prism
Agt1® library with an oligonucleotide probe
derived form a conserved amino acid
sequence of the ATP-binding site of cation
transport ATPases. All these clones have
the sequence coding a part of the binding
site of FSBA, an analogue of ATP. However,
some of them have different sequences from
phosphorylation site and FITC-binding site.
In the present study, we obtained the cDNA
of Na~,K~*-ATPase a-subunit and investigated
its expression.

Comparison of deduced amino acid
sequence of this clone with other transport
ATPases allowed identification of the
sequences that may form phosphorylation
site, FITC-binding site, FSBA-binding sites
and ouabain-binding sites.

Northern blot analysis revealed that the
cDNA hybridized to about 4.5 Kb mRNA which
was predominantly expressed in ectoderm
cells. This mRNA existed in small amount
in unfertilized eggs, but markedly
increased after hatching and maximally
expressed at the mesenchyme blastula_ stage
preceding the increase of ouabain-sensitive
Na~,K*-ATPase activity.

DB 94

ANAT.YSTS OF TRANSCRIPTTON REGUI.ATTON
MECHANISM OF THE EARILY HISTONE GENE IN TIE
SEA URCHIN, H . PULCHERR IMUS 3
Y .lizuka~, E.Magda "NG K. Akasaka 4
.Yamada~, H.Shimada”~, .Yasumasu*.

Waseda Uniy.,Tokyo ,“Chugai Pharm. Co.,
Ltd., Tokyo, “Hiroshima Univ. Hiroshima.

In sea urchin embryos, the early histone
genes begin to be transcripted at the 16
cell stage. Their transcription rate
increases and reaches to a peak at the 64-
128 cell stage, gradually decreases’ and
after the blastula stage, the transcription
of early histone gene is replaced by the
late histone gene transcription.

We tried to analyze _ the regulation
mechanism of the early histone gene
transcription. Among proteins extracted
from isolated nuclei of the 64 cell
embryos, two factors were found by gel
shift assay to bind to the 5' flanking
region of the H1 gene. Three factors were
also found to bind to that of H3 gene. On
the basis of the binding competition with
specific and non _ specific 5' flanking
regions, these factors bound to the 5'
flanking region of H3 gene are found to
differ from those bound to that of H1 gene.
These factors probably bind to the specific
sequences in these region of Hi and Hs.
The extracted protein factors from
gastrulae hardly bound to these regions of
H1 and H3. these factors probably exist in
nuclei of embryos in period in which these
genes are highly transcripted. It is
likely that these factors contribute to the
regulation of Hl and H3 gene transcription.

DB 95

Nuclear proteins that bind to the
regulatory region of sea urchin
(Hemicentrotus pulcherrimus) arylsulfatase
gene.

K.Yamada, T.Yamamoto, K.Akasaka and
H.Shimada, Zool. Inst., Fac. Sci.,
Hiroshima Univ., Hiroshima 730 Japan

Embryonic nuclear proteins that
specifically bind to the regulatory
region of arylsulfatase (Ars) gene of sea
urchin, Hemicentrotus pulcherrimus, were
identified by the gel electrophoresis DNA
binding assay and DNaseI footprinting. At
least 9 distinct sites within the Ars
regulatory region formed discrete
complexes with nuclear proteins. Results
of a series of competition studies suggest
that the proteins comprising these
complexes have considerably high
affinities for different regions in the
Ars promoter. Mobility shift experiment
using nuclear proteins from 12-hr blastula
(Ars, active) as well as 24-hr late
mesenchymal blastula (Ars, inactive)
revealed that appearance of some of these
proteins in the embryo was regulated
developmental stage-specifically.

DB 96

CLONING OF A cDNA EXPRESSED DURING MOUSE
8-CELL STAGE

T.Kamiok , M.Matsumotol, M.Nozaki@,
M.Hoshi Dept. of Life Sci., Tokyo Inst.
of Technol., Tokyo. 2Dept. of Microbial
Genet., Research Inst. for Microbial Dis-
eases, Osaka Univ., Osaka.

Mouse hatching enzyme gene was
screened from a cDNA library of mouse 8-
cell stage embryo. Because the enzyme is
suggested to be trypsin-like, it was
screened by plaque hybridization with a
29mer oligonucleotide probe which corre-
sponds to the amino acid consensus sequence
in the active sites of trypsin-like serine
proteases. Three clones were obtained, and
the total RNA isolated from adult mouse
organs were analyzed by the dot blot hybri-
dization with them. One, but not the
others, was hybridized well with ovary RNA.
It was, however, hybridized little with
other organ RNAs. This clone, having an
insert of about 1.4kb long, hybridized also
With about 0.75% of the 8-cell stage cDNA

library. Sequencing by dideoxy method re-
vealed that there were five homologous
regions to the 29mer probe. Homology

search by the EMBL gene bank showed that
this clone had more than 90% homology with
the signal peptide peptidase A (sppA) gene
of E.coli.

Although no direct evidence has yet
obtained that this clone is indeed the gene
of mouse hatching enzyme, it's expression
in early embryos dose not contradict to
such an idea.

1106 Developmental Biology

DB 97

THE INCREASE IN EXPRESSION OF TUMOR SUPPRES-
SOR GENE RB DURING CELL DIFFERENTIATION

S. Goto and T. Endo

Dept. of Biol., Fac. of Sci., Chiba Univ., Chiba

A number of differentiating cell types including nerve,
blood, and skeletal muscle cells are irreversibly arrested at GO
phase of the cell cycle and express characteristic genes during
terminal differentiaiton. Terminally differentiated skeletal
muscle cells fuse to form multinucleated myotubes. When the
cells are differentiated in the presence of EGTA to chelate Ca**,
they express a battery of muscle-specific genes without
accompanying cell fusion. Such biochemically differentiated
cells are reversibly arrested in quiescent state. A tumor
suppressor gene RB (retinoblastoma gene) suppresses the
growth of several tumor cells when it is introduced in the cells.
Unphosphorylated RB protein is regarded as active in growth
suppression because it is unphosphorylated during GO/G1 phase
but phosphorylated during S to M phase. However, the levels
of expression have been considered not to fluctuate during the
cell cycle. Thus, we addressed whether the expression of RB
is involved in the growth arrest during differentiation. RB
mRNA was remarkably induced during terminal differentiation
of mouse skeletal muscle cell line C2 and rat skeletal muscle cell
line L6. The levels of the mRNA also augmented during
biochemical differentiation of L6 cells induced in the presence of
EGTA. In addition, human promyelocytic leukemia cell line
HL-60 differentiated with TPA exhibited increased levels of RB
mRNA. Immunofluorescence microscopy by use of a
monoclonal antibody to RB protein demonstrated that the levels
of the protein also considerably increased during differentiation
of these cells. In contrast, undifferentiated fibroblast cell lines
BALB/c 3T3 and CV-1 induced to be quiescent by serum
deprivation and contact inhibition of growth, respectively, had
constant levels of RB expression. These results suggest that
the increase in RB expression, in addition to dephosphorylation
of RB, is essential for the growth arrest during differentiation.

DB 98

REGULATION OF SKELETAL MUSCLE CELL
DIFFERENTIATION BY NUCLEAR ONCOGENES SV40
large T, c-jun, AND c-fos

T.Endo Dept. of Biol., Fac. of Sci., Chiba Univ., Chiba

Since a number of myogenic determination factors including
MyoD and myogenin bind to the enhancer/promoter elements of
several muscle-specific genes, they are regarded as muscle-
specific transcription factors. Terminally differentiated skeletal
muscle myotubes but not undifferentiated myoblasts
constitutively express a battery of muscle-specific genes. Thus,
the expression and activation of the muscle-specific transcription
factors are essential for muscle cell differentiation. I have
addressed how nuclear oncogenes involved in gene expression
regulate the differentiation induced by these factors. The
C2SVTts cells, which are transfectants of C2 mouse skeletal
muscle cells with SV40 large T antigen gene linked to Zn2*+-
inducible metallothionein gene promoter, were used for these
studies. When large T was induced by the addition of Zn2*, a
nuclear protooncogene c-jun was coordinately induced as
examined by RNA blot hybridization. On the contrary, the
expression of MyoD and myogenin was suppressed, and
terminal differentiation was blocked. A tumor promoter TPA
was added to C2 cells that were in quiescent state by contact
inhibition of growth. This treatment transiently induced mRNAs
of c-jun and c-fos. In contrast, the levels of MyoD and
myogenin mRNAs were temporarily declined. When C2 cells
were transfected with c-jun cDNA fused to the RSV LTR, the
expression of MyoD and myogenin as well as the terminal
differentiation was inhibited. c-Jun/c-Fos heterodimer (AP-1)
has been shown to bind to the TPA-responsive element (TRE) in
TPA-inducible genes and to activate the transcription of these
genes. Thus, the transcription of MyoD and myogenin or
skeletal muscle differentiation, in consequence, is likely to be
suppressed by a product of TRE-containing gene or c-Jun/c-Fos
(AP-1) itself.

DB 99

EXPRESSION AND DISTRIBUTION OF CHICKEN TROPONIN T
ISOFORMS

Y.Yao and T.Hirabayashi, Inst. of Biol. Sci. Univ.
of Tsukuba, Tsukuba, Ibaraki 305, Japan

Troponin T is known to have many isoforms
among subunits of troponin. In order to elucidate
expression and distribution of the isoforms,
23 chicken skeletal muscles, especially wing
muscles, were examined by two-dimensional gel
electrophoresis and immunoblotting tests with
antisera against to chicken troponin T. In
wing muscles, many isoforms including both breast
and leg types were found, but slow type
isoforms was not present. However, variation of
troponin T expression was observed in different
wing muscles and in developing muscles. The
experssion was changed from normal to
developmentally younger styles by denervation.
Close inspection of two-demesional gel
electrophoresis patterns suggested coordination
between isoform types of troponin T and ratioes

of tropomyosin @ and 6 isoforms.

DB 100

IDENTIFICATION OF TROPOMYOSIN ISOFORMS IN
CHICKEN DIGESTIVE ORGANS. ;

L.Xie, J.Miyazaki and T.Hirabayashi. Inst.
of Biol. Sci., Univ. of Tsukuba, Tsukuba.

Two-dimensional gel electrophoresis of 8
digestive organs from 60-d-old chicks was
performed. Judging from the similarity of
the protein patterns, they were classified
into the following 4 types: 1) oesophagus
type, 2) proventriculus type, 3) gizzard
type, and 4) intestine type. In the
representative organs of these types, the
distribution of tropomyosin isoforms was
examined, and 4 high- and 5 low-Mr-type
isoforms were detected in the embryo in
addition to a and § isoforms. In the
adult, however, there were 3 high- and 4
low-Mr-type isoforms as well as a and 8B
isoforms, the former of which was
restricted to the mucous membrane.
Immunoblot analysis indicated that the
high- and low-Mr-type isoforms) )iny the
embryo corresponded with those in the
adult mucous membrane, but the differences
in number and amount of the isoforms were
found between the embryo and the adult
mucous membrane.

Developmental Biology 1107

DB 101

EXPRESSION OF C-PROTEIN ISOFORMS IN
ELECTRICALLY STIMULATED AND MECHANICALLY
STRETCHED SKELETAL MUSCLE CULTURES.

K. Yabusakil, H. Nakae? and T. Obinata! .
Upeyes Oe Boils, “Gms Winsnzo - Chiba,
2Toshiba R & D Center, Kawasaki.

Expression of C-protein isoforms is
closely related to developmental stages of
skeletal muscle; Although a fast isoform
(CE) is an adult-form in pectoralis
muscle, Candace (€c)i- and slow (Cs)
isoforms are expressed transiently in
embryonic or peri/neonatal stages,
respectively. In this study, we examined
how C-protein isoform expression is
affected by muscle contractile activity
with the primary cultures of embryonic
chicken pectoralis muscle cells. In the
myotubes cultured in the absence of Ara-C,
spontaneous contraction became evident at
10-15 d in culture and Cc disappeared from
Majority of myotubes. When tetrodotoxin
(TTX) was added to the medium, myotube
contraction was completely blocked and Cc
reappeared in most myotubes. In the
cultures with Ara-C, contractile activity
of myotubes was low and Cc persisted even
in 3-w cultures. Repetitive mechanical
stretching of myotubes in a Flexercell
system caused thickening of myotubes and
acceleration of the disappearance of Cc.
When myotubes were stimulated electrically
at 10 Hz for 3 w, Fc became scarce, while
decrease in Sc but not Fc was caused by
stimulating at 100 Hz.

DB 102

ANALYSIS OF SKELETAL MUSCLE NUCLEAR
PROTEINS

M.Kirinoki, J.Miyazaki and
T.Hirabayashi, Inst. of Biol. Sci. Univ.
of Tsukuba, Ibaraki 305, Japan

To examine the relation between
cell differentiation and nucleus
function, we isolated nuclei from
chicken skeletal muscles essentially
as described by Mulvihill (1977), and
compared protein constituents of the
nuclear fraction with those of whole
muscle tissue by two-dimensional gel
electrophoresis. Thirty spots were
found to be. specific to the nuclear
fraction or significantly increased
in the nuclear fraction. Three major
spots, MN1, MN2, and MN3, of the 30
spots were cut out and the characters
of proteins in them were analyzed.
Immunoblotting tests with antisera
against the proteins showed _ that
anti-MN1 and anti-MN2 reacted with
proteins in both muscle and liver

nuclear fractions, While anti-MN3
reacted with proteins only in the
muscle nuclear 1EICELE 16 aL id er

immunohistochemical tests, anti-MN1
and anti-MN2 reacted with proteins
restricted to muscle nuclei, and
anti-MN3 reacted with cytoplasmic
proteins.

DB 103

REGIONAL DIFFERENCES OF NUCLEAR PROTEINS OF
NEWT NEURULA.

T.Asao, Biol. Lab.,. Sch. Med., St. Marianna
Univ., Kawasaki.

Embryonic nuclear proteins were studied
from among different regions of newt neuru-
la, i.e. anterior neural plate (AN), poste-
rior neural plate (PN), epidermis (EP) and
yolk cells (YC). The aim of the study was
to detect key proteins related to morpho-
logical change. The qualitative and quanti-
tative comparison of protein amounts was
due to the density of band patterns in SDS-
PAGE, with silver stains. Proteins applied
were from the nuclei equivalent to 20 pg
DNA. Electrophoresis patterns showed sever-
al differences according to embryonic re-
gion. Firstly, EP contained many more pro-
Feimescroupseinel7.050 17.5 and WS. 5ekd) than
AN or PN. These protein groups were hardly
detectable in YC. As reported in the last
annual meeting, they appeared firstly in
the neural stage. Secondly, low molecular
proteins in 6.5 and 12.5 kd were present
commonly in each region, however, the for-
mer appeared more in PN than in AN, and the
latter, more in AN than in PN. In the case
of the proteins, too, EP contained in the
largest amounts of the four. No differences
were detected in the other higher molecular
DeOUGibAS am BI, ZO, AseS, OF 65, G4, OP5 Wil@
kd etc.

DB 104

POLYPLOID NUCLEI IN THE YOLK SYNCYTIAL
LAYER OF THE MEDAKA EMBRYO, Oryzias
latipes.

T. Kageyama. Department of Biology, Kyoto
Prefectural University, Kyoto.

In the yolk sac syncytium of the embryo
of Oryzias latipes, there are nuclei whose
size, the number of nucleoli suggest that
their DNA content is greater than that of
other embryonic nuclei. The onset, and
the subsequent time course, of "“polyploid-
ization" was examined by fluorescent-
microspectrophotometry from the last cycle
of mitotic wave in the late blastula stage
through the hatching of the embryo.
Embryos were fixed with 4% paraformalde-
hyde, pH 7.4, for 4-24 hrs, and, after the
dechorionation, stained with 4',6-diamidi-
no-2-phenylindole (DAPI). The late blas-
tula includes nuclei in the YSL whose DNA
content is diploid (2n), tetraploid (4n),
and octaploid (8n) at the completion of
the last mitosis. Each population is
duplicated once in the DNA content until
the beginning gastrula stage, and once
again or more until the end of epiboly.
Giant nuclei appear to include more DNA
content up to 65-C. The frequency distri-
bution of the DNA content appears to

remain constant afterward. No mitotic
figures have been observed in the YSL
after the gastrula stage. Morphological

examination of the nuclei indicates the
Presence of nuclear fission, but no
nuclear fusion.

1108 Developmental Biology

DB 105
°'P NMR OF PHOSPHOLIPIDS AND ENERGY RICH COMPOUNDS IN

DEVELOPING JAPANESE QUAIL EMBRYOS.
K.Nishimura ' , Y.Fujise ®? , A.Nakamura ° . ! Dept. of
Biol. and ? Dept. of Chem., Hamamatsu Univ. School of

Med., and ° Univ. of Shizuoka, Hamamatsu Col., Hamamatsu.

°' P high resolution NMR spectroscopy provides
dynamic profiles of phosphorous compounds related to
energy metabolism in intact animals. A developing embryo
of _Coturnix coturnix japonica was investigated at 37 C
utilizing a 27¢ hand made saddle type coil attached on
the surface probe of GSX-270 NMR system ( JEOL ). NMR
peaks of phosphomonoesters, inorganic phosphate, phospho-
diesters, creatine phosphate, and phosphates in adenine
nucleotides were monitored as ongoing process. Thus no
later than 8 days, ATP was recognized as measurable peaks,
while creatine phosphate was in 11 days, suggesting a
delayed precesses of development in the mtabolism of the

latter energy rich compound.

DB 106
DEVELOPMENTAL CHANGES IN TREHALASE AND

CELLOBIASE ACTIVITIES OF ARTEMIA SALINA

-EFFECTS OF CYCLOHEXIMIDE AND ACTINOMYCIN D
Z.Nambu and F.Nambu-Akiyama, Dept. of Biol.
Sch. of Nursing and Med. Technol. Univ. of
Occupational and Environmental Health,
Japan. Kitakyushu.

Activities of trehalase and cellobiase
in embryos and larvae of Artemia changed
during development. They began to vary
after hatching, and 40-hr-nauplii showed
5- and 2- fold increase in the activities
of trehalase and cellobiase , respectively.

In order to examine sensitive time to
translation and transcription of the two
enzymes, effects of cycloheximide and
actinomycin D on the enzyme activities were
investigated. Artemia had been exposed to
cycloheximide (50 ug/ml culture medium) or
actinomycin D (10 ug/ml culture medium) for
10 hrs from a scheduled culture time, and
then they were sacrificed for determining
the activities.

The cycloheximide treatment suppressed
the activities of trehalase and cellobiase
rom C2H,5 t@Q BL Inc ema trom 26.5 co 37 5
hr, respectively. Unexpected increase in
the trehalase activity was found between
25) inal AS. S ine waltely wine erentmente o%
cycloheximide. Actinomycin D treatment from
Gy atom Zor Oishinemsesulitedmam HOm-sudecrease
in the two enzyme activities. Contrary to
what we expected, considerable increases in
the two enzyme activities were detected
with the Actinomycin D treatment from 18 to
ZS hin.

DB 107

[ ADP-RIBOSYL]ATION OF PROTEINS IN SEA
URCHIN EMBRYOS.

Y.Kamata-, K ,Mikami-Takei®, S.Furuyal and
I. Yasumasu’. Dept. of Biol. Sch, ..0f
Educ., Waseda Univ., Tokyo, 2Dept. of
Phys., Fac. of Pharm. Sci., Teikyo Univ.,

Kanagawa.

In sea urchin embryo, the ADP-
ribosyltransferase activity in nuclei
increases in the developmental period
between fertilization and the morula stage
followed by its gradual decrease. After
hatching out, the activity increases again
and becomes maximum at the gastrula stage.
On SDS polyacrylamide gel electrophoresis,
{[ADP-ribosyl]ated proteins in nuclei were
found to be those with molecular weights of
120K and 160K, which were found in embryos
at all examined stages. In the pre-hatching
period, highly [ADP-ribosyl]ated proteins
with molecular weights of 30K and 40K were
found in nuclei. The ADP-ribosyltransferase
activity increases gradually in plasma
membrane fraction during early development.
[ADP-ribosyl ]ation of proteins was
considerably augmented by pertussis toxin
in embryos at almost all examined stages
and by cholera toxin especially in post-
hatching period. A protein band at 45K was
considerably made dense by cholera toxin in
membrane fraction at almost all stages. In
the presence of pertussis toxin, a protein
band at 23K was [ADP-ribosyl]Jated in this
fraction.

DB 108

PURIFICATION OF CASEIN II. IN CHROMATIN
FROM SEA URCHIN EMBRYOS.
T. Natsume!, Y. Masuyama? and I.Yasumasu!
Dept. of Biol., Sch. of Educ., Waseda
Univ., Tokyo, “Yamanouchi Pharma., CO.,
LTD., Tsukuba

Previously, it has been reported a
protein kinase, which is not activated by
cyclic nucleotides, and mediates phospho-
rylation of casein more strongly than
Phosvitin, is localized in nuclei in sea
urchin embryos. In 10,000g sedimentable
fraction a protein kinase which phosphory-
late phosvitin more strongly than casein
is also found in embryos. Protein Kinase
in nuclei was activated by polyamines and
blocked by heparin. This enzyme, probably
casein kinase II, increased in its activ-
ity in developmental period between ferti-
lization and the morula stage and then
decreased. After hatching, an increase in
its activity occurred during development
up to the gastrula stage and then was
followed by a gradual decrease. From
nuclei isolated from gastrulae, the enzyme
was extracted by 0.4 M NaCl and purifica-
tion of this enzyme from the NaCl extract
was carried out by affinity chromatography
and then by agarose and mono @Q chromato-
graphy. On SDS-PAGE pattern, protein
bands other than subunits of casein kinase
II were found. Thus, purification of this
enzyme is found to be incomplete. Further
trials to purify this enzyme will be made
in future.

Developmental Biology 1109

DB 109

PROTEIN PHOSPHORYLATION IN PRESUMPTIVE
VEGETALIZED AND ANIMALIZED EMBRYOS’ OF
SEA URCHIN.

K.Kataoka, A.Fujiwara, I.Yasumasu.

Dept. of Biol., Sch. of Educ., Waseda
Univ., Tokyo.

It was found that A23187 induced
animalization of sea urchin embryos and
ruthenium red and tetracaine prevented
embryos from A23187-induced
animalization, when the treatments with
these compounds were performed for 2 hr
in pre-hatching period of development.
These suggest that artificial Ca?~*
signals induce animalization of embryos.
It was also found that W-7 and ML-9
reversed but H7 and HA1004 did not
eancel animalizing effect of A23187 in
pre-hatching period. Artificial Ca?~*

signals seem to activate Ca?*
calmodulin-dependent protein kinase to
induce animalization. Vegetalization

induced by the treatment with Li~ or
tetracaine initiated 3 hr after
fertilization is augmented by W-7 or
ML-9. Phosphorylation of protein,
estimated *?P incorporation into protein
in embryos, was enhanced by A23187 and
was blocked by W-7 and ML-9, as well as
procaine and tetracaine. Egg homogenate
exhibited W-7 sensitive protein kinase
activity in the presence of calmodulin
and HA1004 sensitive activity in the
presence of cAMP.

DB 110

EFFECT OF PULSE TREATMENT WITH A23187 IN
PRE-HATCHING PERIOD ON MORPHOGENESIS IN
SEA URCHIN EMBRYOS AT THE POST-GASTRULA
STAGE.
A. Fujiwara, K. Kataoka and I. Yasumasu.
Dept. of Biol., Sch. of Educ.,
Waseda Univ., Tokyo.

Sea urchin embryos, having been exposed
to A23187 for 2 hr at various stages in
pre-hatching period, became _ essentially
animalized embryos during another culture
in normal sea water up to the pluteus
corresponding stage. Pulse treatment with
A23187 in the presence of tetracaine or
ruthenium red, performed at any times in
pre-hatching period, allowed embryos to
develop into quasi-normal plutei. Proba-
bly, A23187-induced abnormal increase in
[Ca2+*]i, which is blocked by tetracaine
and ruthenium red, alters the programmed
fates of embryo cells to produce animal-
ized embryos. The pulse treatment start-
ing at a time within 2 hr of development
produced embryos having exoguts and large
embryo bodies with well developed pluteus
arms and spicules. Embryos treated for 2
hr starting at a time between 2 and 12 hr
became those having large spherical bodies
and exoguts. Exoguts were quite poor in
thus treated embryos unless the treatment
was initiated at times between 5 and 7 hr
of development. The stages at which the
cell fates are altered by A23187-induced
abnormal increase in [Ca2t]i probably
differ among cell lines.

DB 111

CELL PROLIFERATION IN PRESUMPTIVE
ANIMALIZED OR VEGETALIZED EMBRYOS OF SEA
URCHIN .
H. Mizoguchi, Lab. of Biol., Jun. Col. of
Rissho Univ., Saitama.

In the animalized or vegetalized
embyos of sea urchin, cell proliferation
study was performed to clarify the

cellular mechanism of archenteron
formation. The animalized embryos were
obtained by culture in seawater
containing ZnCls from the time of
fertilization. The vegetalized embryos

were obtained by 3-hr treatment with
chloramphenicol at the 16-32 cell stage.
Quantitative measurements of cell
proliferation was done following 5-
bromodeoxyuridine(BrdU) label, assayed by
immunocytochemistry using monoclonal
antibody against BrdU and streptavidin-
biotin peroxidase system.

Labelling of BrdU in the presumptive
animalized or vegetalized embyos’ was
performed at the equivalent stage to that
of swimming blastula. In normal
blastulae, the number of S phase cells in
vegetal hemisphere was greater than that

in animal hemisphere. However, the
percentage of S phase cells was not
significantly different between

animalized or vegetalized hemisphere in
both presumptive animalized or
vegetalized embryos.

It seems that S phase cells in vegetal
hemisphere of normal swimming blastula
were related to archenteron formation.

DB 112

ARCHENTERON FORMATION ABILITY OF VEGETAL
REGION OF STARFISH EGG.
M.Kiyomoto and H.Shirai. Ushimado Marine
Laborator Univ. of okayama, Okayama.
Animal-vegetal polarity exists in eggs
of the starfish, A.pectinifera, similar
to sea urchin. Archenteron is formed on
the vegetal pole during gastrulation.
When fully grown immature oocytes or
Mature oocytes were bisected across the
animal-vegetal axis and were fertilized,
animal egg fragments (AEF) never
gastrulated whereas vegetal egg fragments
(VEF) gastrulated. Placing a blastomere
near an AEF resulted in a combined embryo.
When an animal blastomere (AB) in the
8-cell stage was cultured with AEF,
gastrulation did not occur. Contrary to
this, when a vegetal blastomere (VB) at
the 8-cell stage was cultured with AEF
gastrulation occurred. Similar reaults
obtained when a VEF or one of its
descendants was combined with an AEF.
In these experiments vegetal part was
stained with rhodamine isothiocyanate
(RITC). RITC-labelled cells detected
always in the archenteron, coelomic
pouch, mesenchyme, and sometimes in the
ectoderm near the blastpore. Combinning
a VB at 8- or 16-cell stage with an AEF,
non-labelled region was not observed in
the archenteron of the resulting embryo.
However, when a VB of 32-cell stage was
used, a part of the archenteron near the
blastopore sometimes lacked the RITC
mark. This suggests that the fate of an
AEF was altered to endoderm by VB.

1110 Developmental Biology

DB 113

MORPHOLOGICAL STUDY ON THE MECHANISM OF DE-
LAMINATION EXHIBITED BY EMBRYO OF PELMATO-
HYDRA ROBUSTA-II.

K.Noda and C.Kanai. Dept. of Ultrast. Res.,
Tokyo, Metropolitan Inst. of Gerontology.

We used the experimental dwarfed (328p)
and normal large (479) eggs for investi-
gating the trigger events of delamination
(Noda and Kanai, 1988). We showed that
columnar shape of the blastmeres is one of
the event which trigger delamination.
unequal division at various degree attribu-
ting to the situation of animal pole. We
used embryos cleaved in almost equal divi-
sion as present materials. When the volume
of blastmeres reduced less than (50p)? by
cleavage, blastmeres are designed to change
to columnar shape. Large group embryos can
change their blastmeres to columnar shape
during their maximum diameter, since they
have thick wall. However, in small group
embryos with their maximum diameter, blast-
meres in the volume less than (50p) * are
cuboidal or even squamous in shape. So,
small embryos must reduce their diameter to
obtain columnar shape.

Thus, the thickness of the wall of cleav-
ing embryos and the volume of blastmeres
are important for changing blastmeres to
columnar shape.

After delamination, outer layer cells
of embryos become more compactly arranged
and embryos revert their diameter to their
original egg size.

DB 114

EFFECTS OF ENVIRONMENTAL FACTORS ON THE
SITE DETARMINATION OF ORGAN
DIFFERENCIATION IN HYDROID, Stylactis sp.
H.Suzuki and Y.Kakinuma

Dept.of Biol., Fac. of Sci.,Kagoshima
Univ.,Kagoshima

To investigate the mechanism of colony
formation in Stylactis sp., we analyzed
the colony growth patterns and the effect
of rearing conditions on the growth pat-
tern by using time lapse video apparatus
and photos with a florescent dye(DAPI).
These colonies were reared at two feeding
conditions (feeding, starving) and 3 tem-
perature conditions (15%7,20%T and 25%).

Kept feeding to initial polyp and incu-
bated at 15%, the interpolyp distance be-
came short. The stolon branching locus was
determined at the convex site of bent
stolon at 159.6+12.5° (meanst+S.E.), and
was independently of rearing conditions.
In the coelenteron of stolons, with peri-
odic pulsations, we could observe ambient
shuttle flow and monocells migrated indi-
vidually with the flow. And the period of
pulsation varied with each rearing condi-
tions. Around the stolon branching loci
and stolon tips, ‘the monocells stayed
longer than at other sites.

Thus, these data suggest that the
external conditions are important factors
that affect the control mechanisms
determining the site of polyp and stolon
differentiation.

DB 115

THE POSITION DETERMINATION OF HEADS ON
HYDRA CELL AGGREGATES.
M.Sato and Y.Sawada. Research Inst. of
Electr. Comunn., Tohoku Univ., Sendai.
Hydra cell aggregates have no head-foot
polarity since the dissociated cells were
mixed up at the time of aggregation, and
the aggregates have to establish their body
axes anew during the regeneration. To know
how the position of a head (i.e. a pole of
body axis) is determined on an aggregate,
we examined the behavior of the cells whose
axial origin was higher than other cells.
(1) A DAPI-labeled small tissue-piece in
various size from upper or middle body was
embedded in a cell-aggregate made from
original middle body. The tissue of higher
axial origin induced one of the first
appearing heads, while the tissue of same
axial origin as the surrounding cells did
not. Thus, not all the undissociated tissue
fragments can act as a trigger for breaking
symmetry. (2) DAPI(+) heads were
dissociated, aggregated and mixed into
DAPI(-) cells originating from lower body.
The distribution of labeled cells was found
random, not uniform nor localized during
the early regeneration. The binding of the
monoclonal antibody TS19, which labels
tentacles of intact hydra, was observed
near some small cell-groups among the
random distribution of DAPI(+) cells,
suggesting that a small group of original
head cells created by chance would nucleate
a new head, probably by a mechanism such as
a reaction-diffusion mechanism.

DB 116

BASAL DISK FORMATION IN HYDRA, P.robusta.
THE BASAL DISK GLAND CELL DIFFERENTIATION
FROM THE PEDUNCLE EPITHELIAL CELLS IN
SMALL TISSUE EXPLANTS

Y. Kobayakawa

Biological laboratory, College of General

Education, Kyushu University 01, Fukuoka.

Ability of the peduncle ectodermal
epithelial cells to differentiate autono-
mously ‘to the basal “disk” (gland cells
(BDGCs) was investigated. Tissue pieces
(a quarter of the whole peduncle) were
explanted from the various parts and were
cultured in isolation. Then the appearan-—
ce of AE03 (monoclonal antibody which
recognize the BDGC specific mucous gra-
nules) positive epithelial cells in each
explant was examined. Almost all explants
from the lowest part of the peduncle be-

came AE03 positive, but explants from the
upper half of the peduncle show few AE03
positivity. However, half of explants

from the same part (upper half of the
peduncle) became AE03 positive when they
were grafted to the middle position of the
gastric region.

These results suggest that in steady
state hydra ectodermal epithelial cells in
the lowest part of the peduncle already
committed to differentiate to the BDGCs,
however, in upper half part of the pedun-
cle ectodermal epithelial cells are on the
way of commitment, yet.

Developmental Biology fetgtal

DB 117

INVERSION OF IN-OUT AXIS OF BLASTURA CELLS
IN SYCON CALCAR-AVIS(CALCAREA, PORIFERA)
A.Yamasaki and Y.Watanabe. Dept. of Biol.,
Fac. of Sci., Ochanomizu Univ., Tokyo.

In some calcareous sponges, it has been
known that in-out axis of blastura cells
are reversed by inversion of the embryo,
To reseach this process we histologically
examined embryonic development of Sycon
calcar-avis.

The oocyte, surrounded by monolayer of
pinacocytes, laid in the mesohyl just
beneath the choanoderm, where the egg was
fertilized and developed into stomoblastula
During cleavage two types of cells emerged:
rounded granular cells with large nuclei
and elongated cells with small nuclei. The
latter formed a tall columner epithelium
and possessed flagella which extended into
the blastocoel. The granular cells did not
arrange as a epithelium, but aggregate at
the pole of blastula facing te the choano-
derm. An opening, appeared among the
granular cell mass, adhered to the neigh-
boring choanocytes at its edge. Then the
blastula turned inside out through the
opening, that reversed in-out axis of the
flagellated cells and each flagellum faced
outward. The granular cells came together
again under the embryo and amphiblastula
with external flagella was formed. At the
same time inversion occured, the blastula
was extruded into the adjacent choanocyte
chamber. The amphiblastula attaches to the
parental choanoderm at posterior granular
cells until it swims out the mother sponge.

DB 118

EXTRACELLULAR MATRIX IN THE FRESH-WATER
SPONGE DEVELOPMENT
N. Fujieda and Y. Watanabe. Dep. of
Biol.,Fac. of Sci., Ochanomizu Univ.,Tokyo.
To investigate the role of extracellular
collagen during the development of sponge,
the morphogenesis after gemmule hatching
was examined with TEM. As material, we used
fresh-water sponge Ephydatia fluviatilis.
It has been known that sponge has_ two
types of collagenous materials:collagen
fibrils and spongin fiber. E. fluviatilis
also had both. The collagen fibrils of this
sponge was rough fibrils,which had
periodical band and projections.Although
the fibrils were seen in all the area of
mesohyl, they were densely distributed
especially in the narrow mesohyl of dermal
membrane.It was suggested that these
fibrils underline the layers of pinacocytes
and support the structure of thin dermal
membrane. spongin in mesohyl was felt-like
texture of condensed microfiber. The felt-
like spongins were deposited where fast
adhesion is required, such as at the joints
of spicules and on the underside of
basopinacocytes attaching to the substrate.
The appearances of microfibers in the"felt"
were varied depending on their positions.
Sponge is the most primitive multicellular
animal with less organized body system and
poor cell junctions. The extracellular
collagens of sponge, are fairly various and
it is suggested that they could play an
important role in the development and
morphogenesis of sponge.

DB 119

The postembryonic erythropoietic site in
larvae of Xenopus as revealed by
phenylhydrazine-induced gnoicr

H. Ohinata and T. Enami*~. Zool. Inst.,
ac. of Sci., Hokkaido Univ., Sappro.
BOZO Research Center, Gotenba.

The ventral blood islands (VBI) of
st.22 Xenopus laevis are known as an
initial hemopoietic site. To determine
the site of postembryonic erythropoiesis,
the tissue distribution of regenerating
erythrocytes was determined with an
anti-larval hemoglobin (LHb) monoclonal
antibody, in the completely anemic larvae
induced by phenylhydrazine (PHZ). Three
days after the dual PHZ-treatment at
st.31 and 40, the LHbt cells were
detected first in the liver and around
the digestive tract, followed by the
occurrence of a fewer LHbt cells in the
blood vessels. The larvae which had been
hepatectomized and cardiectomized before
the PHZ treatment showed a remarkable
reduction of recovery of the LHb* cells 4
days after the treatment, although a few
LHb* cells were present around the
digestive tract. Production of chimeras
with a cytogenetically labeled VBI tissue
clearly demonstrated that all the
regenerating LHbt cells originated in the
VBI. It is concluded from these results
that the larval liver provides the major
Site where the VBI-derived hemopoietic
cells reside and differentiate into
erythrocytes.

DB 120

CHROMOSOMAL CHANGES OF MOUSE EMBRYONAL
CARCINOMA DERIVED F9 CELLS IN THE
INDUCED ENDODERMAL DIFFERENTIATION.
D. Watanabe!, K. Yamada? and T. Matsuzawa!
‘Dept. of Biology Osaka Kyoiku Univ. Osaka
and Dept. of Nutri. and Physiological
Chemistry. Osaka Univ. Med. chool, Osaka.
Mouse embryonal carcinoma derived F9
cells have a pluripotency to differentiate
into the endodermal cells by treatment with
retinoic acid (CRA). We examined the fre-
quency of sister chromatid exchange (SCE)
and DNA methylation in the F9 cell differ-
entiation. SCE frequency was counted on
BrdU incorporated chromosome samples stain-—
ed with Giemsa solution. SCE frequecy per
chromosome is 0.21 in F9 cells. While, that
of RA treated F9 cells increased up to 0.33,
and 5-azacytidine (DNA methylation inhibi-
tor) treated F9 cells was also increased
similarly. These results suggest that SCE
frequency are increased by both F9 cell dif-
ferentiation and DNA demethylation. To know
the relationship between F9 cell differen-
tiation and DNA methylation. In situ detec—
tion of DNA methylation was carried out in
chromosomal level. *H-thymidine labeled
chromosome samples were digested by restric—
tion enzyme Hpa II Ccuts only nonmethylated
CCGG nucleotide sequence) directly. DNA
fragment released by Hpa II digestion was
4.3 %* in F9 cells, but it was increased to
9.3 % in RA treated F9 cells. And, it was
noticed that this digested chromosome sam-—
ples by Hpa II showed a decreased stain-
ability of Giemsa and some of them showed a
band-like feature. These results show that
the chromosomal DNA is demethylated in F9
cell differentiation, and that these methods
may become a useful to know the chromosomal
structure change in the DNA sequence level.

SV Developmental Biology

DB 121

ANALYSIS OF THE DEVELOPMENTAL POTENCY OF
SINGLE CELLS OF THE MOUSE INNER CELL
MASS USING THE STRAIN SPECIFIC ANTIBODY.
T.Harumi', M.Kusakabe=, AND M.Noguchi',
‘Dept. Biol., Fac. of Sci., Shizuoka
Univ., Shizuoka 422 , "Lab. Cell Biol.,
RIKEN, Tsukuba, Ibaraki 305

In order to know the developmental
potency of single cells of inner cell
mass(ICM), the chimeric fetuses were
immunohistochemically examined by using
the mouse strain specific antibody
(anti-CSA). ICM cells were immunosurgi-
cally isolated from Day 4 BALB/c blasto-
cysts. Single one or 2-3 ICM cells were
injected into one Day 4 C3H/He blasto-
cyst. Day 12 chimeric fetuses were fixed
by microwave irradiation, post-fixed
with 95% ethanol + 1% acetic acid, de-
hydrated with 100% ethanol, embedded in
polyester wax and sectioned. Following
incubation by biotinylated anti-CSA ,
the sections were incubated by peroxi-
dase-avidin and stained by DAB. The CSA
negative cells derived from BALB/c ICM
cells were apparently identified in sev-
eral kinds of cells as ganglion, heart,
liver, esophagus, germ cells, and vis-
ceral yolk sac. These data suggest that
single cells in mouse ICM can express
developmental multi-potency in organo-
genesis. Supported by Grant-in-Aid of
the Ministry of Education, Science and
Culture(to M.N.).

DB 122

ON THE TOTIPOTENCY OF DEDIFFERENTIATED
CELLS ORIGINATED FROM COPULATORY APPA-
RATUS IN THE PLANARIAN.

W. Teshirogi', M. Kudoh! and W. Yoshida?
‘Dept. of Biol., Fac. of Sci., Hirosaki
Univ., Hirosaki and ?Dept. of Biol.&
Chem. Eng., Fac. of Eng., Gunma Univ.,
Kiryu.

Following decapitation of Dugesia
japonica ryukyuensis, and subsequent
repeated removal of regenerating head
-blastema, the copulatory apparatus
degenerated after about 40-60 days. In
this case, the copulatory apparatus dis-
integrated into cell clusters, which
further divided into separate neoblast
-like cells which finally transformed
into ordinary parenchymal cells. When
the head of worms with degenerated copu-
latory apparatus was regenerated after
discontinuing the excision of the head
blastema, a new copulatory apparatus was
redifferentiated from the dedifferenti-
ated cells. Furthermore, we have observed
that the dedifferentiated cells also
participate in the formation of regene-
ration blastema in a way Similar to that
of neoblasts. These cells are thus
observed to be totipotent.

DB 123

PREPRATION OF MONOCLONAL ANTIBODY FOR ELUCIDATION
OF PLANARIAN REGENERATION.

H.Kawarada’, K.Takezaki', H.Tanaka-, K.Inoue* and
T.Shinozawa'. '‘Dept.of Biological and Chemical
Engineering., Fac.of Engineering., Univ.of Gunma,
Kiryu. “Dept.of Pharmacology., Fac.of Med., Univ.
of Gunma,Maebashi. *Inst.of Internal Secretion.
Univ.of Gunma, Maebashi.

Homogenate of planarian(Dugesia japonica
japonica)head were injected several times in every
weeks into abdominal cavity of mouse(BALB/C). Four
days after the last immunization, spleen was take
out followed by the fusion of spleen cells with
mieloma cells(6,5,3). Screening of the antibody
synthesizing clone was carried out by fluorescent
antibody(FITC) technique using planarian parafin
section. As a control of fluorescent antibody
reaction, the binding of antiserotonin antibody to
the planarian nervous system was tested. We got
three clones producing antibody which react with
enteric canal and epidermis, all over the body
except nervous and enteric canal inalis, and all
over the body. The preparation of clone producing
antibody which react with planarian blastema or
nerovus system are now under the investigation.

DB 124

INHIBITION OF PLANARIAN REGENERATION
BY MELATONIN

T. Shinozawa and Y. Yoshizawa

Dep. of Biological and Chemical
Engineering, Fac. of Eng., Gunma Univ.,
Kiryu, Gunma. &
Melatonin, a pineal‘hormone in
vertebrates, is known as an inhibitor of
planarian fissioning (Morita & Best, 1984).

We exogeneously applied melatonin and
its three derivatives (serotonin,
N-acetylserotonin, 6-hydroxymelatonin)
to the decapitated planarian, Dugesia

japonica japonica.

The formation of the head of decapitated
planarian was retarded only by melatonin.
However, formation of the eyes
was not disturbed. When tail was cut in
the worm and kept in melatonin solution,
the retardation, of the tail growth, was
seen as in the decapitated planarian
experiment. The effects were specific to
melatonin and reversible.

The analysis of the melatonin
distribution in planarinan by use of
anti-melatonin antibody are now under
the investigation.

(Reference)
Morita, M. & J. B. Best (1984) J. Exp.
Zool. 231: 273-282.

Developmental Biology 1113

DB 125

BUDDING-SPECIFIC LECTIN INDUCED IN EPITHE-
LIAL STEM CELLS IS AN EXTRACELLULAR MATRIX
COMPONENT OF THE TUNICATE, POLYANDROCARPA
MISAKTENSIS.

K.Kawamura,S.Fujiwara and Y.M.Sugino. Dept.
Of BHOl sy a bace LOL, SCi, KOCH sUnawa0, KOCH».
We have examined immunocytochemically the
expression and localization of a calcium-
dependent, galactose-specific 14kD lectin
(TC-14) purified from the budding tunicate,
Polyandrocarpa misakiensis. Lectin-positive
granules first appeared in epithelial stem
cells of the bud primordium. As soon as a
bud entered the developmental phase, the
granules were secreted in the mesenchymal
space, at which the TC-14-positive extra-
cellular matrix (ECM) developed. Mesen-
chymal stem cells, or hemoblasts, were
observed in association with the ECM. They
indeed showed a specific binding affinity
for the TC-14 labeled with FITC. The TC-14
(0.7-0.07nM) did not have any mitogenic
effect on the stem cells. Polyclonal anti-
lectin antibody inhibited the hemoblasts
to spread on the ECM and to aggregate
toward the epithelial stem cells, while it
did not block the homophilic cell adhesion
of hemoblasts. By three days of bud devel-
opment, the ECM almost disappeared from a
bud, consistent with the termination of
hemoblast aggregation. We conclude that the
TC-14 is a budding-specific ECM component
induced in epithelial stem cells and that
it plays a role in morphogenesis by guiding
undifferentiated hemoblasts to the epithe-
lial cells.

DB 126

REGENERATION OF THE TUNIC CUTICLE IN A
COMPOUND ASCIDIAN, II. EARLY PROCESSES AND
INHIBITION OF FILAMENT FORMATION.

Eu. Hirose, Y. Saito & *H. Watanabe.
Shimoda Mar. Res. Ctr., Univ. of Tsukuba,
Shimoda, Shizuoka and *Tokyo Kasei Gakuin
Tsukuba College, Tsukuba.

We studied on the cuticle
regeneration at the cut surface of a
colony, Botrylloides simodensis. The first
Sign of the regeneration is a condensation
of the tunic matrix at the cut surface
within a few minutes after the cutting.

In 10-30 minutes, the condensed tunic
matrix forms electron dense filaments.
These filaments are found at the cut
surface of tunic matrix, but not the other
parts of tunic. These filaments aggregate
to form a continuous layer of cuticle in
several days, as reported previously.

The filament formation was inhibited
by the presence of 0.1% protease type I or
0.3mM leupeptin. In former case, small
fragments of the dense filaments are often
found around the cut surface of the tunic
Matrix. It was also inhibited in some
extent by 0.1% trypsin, 0.1mM PMSF or
0.1mM E-64, but not inhibited by 0.1% BSA
or 0.1% trypsin inhibitor type III-0O.

Some proteins in the tunic and
proteases probably work on the
condensation of the tunic matrix and the
formation of the electron dense filaments.

DB 127

TWO MODES OF GUT REGENERATION IN THE COM-
POUND ASCIDIAN, POLYANDROCARPA MISAKIENSIS.
Y. Taneda and K. Watanabe. Dept. of Biol.,
Fac. of Educ., Yokohama Natl. Univ., Yoko-
hama.

There are two modes in regeneration of
animals. One is epimorphosis or true re-
generation of missing parts. The other is
morphallaxis, in which a part is transform-
ed directly into a new organism. Anterior
fragments without gut of the compound asc-
idian, Polyandrocarpa misakiensis, have a
capacity for morphallactic regeneration.

On the other hand, when a part of the int-
estine remains in the fragment, entire gut
is regenerated from the remaining part of
the intestine like epimorphosis. Thus, we
studied on two modes of gut regeneration.

When the fragments contained no part of
the gut, gut rudiment was given rise to
from the atrial epithelium and developed
into gut. During the development of the
gut rudiment, many mitotic figures were ob-
served. When a part of the intestine re-
mained in the fragment, one end of the int-
estine swelled and developed into stomach
and esophagus. Many cells were observed in
the lumen of the gut before development.
While, mitotic figures were rarely observed
in the specimens. In this case stomach
cells were developed directly from the int-
estine cells. Therefore, the origins of
stomach cells are different between two
modes of regeneration.

DB 128

STUDIES ON THE HISTOLOGICAL DIFFERENCES
BETWEEN TWO COLOR PATTERN DIFFERENT STRAI-
NS IN THE COMPOUND ASCIDIAN, POLYANDROCAR-
PA MISAKIENSIS.

T. Ishii and Y.Taneda. Dept. of Biol., Fac.
of Educ., Yokohama Natl. Univ., Yokohama.

Two color pattern different strains are
known in the compound ascidian, Polyandro-
Carpa misakiensis. One is white spot stra-—
in with large circular white spot on the
dorsal side between branchial and atrial
siphons. The other is spotless strain with
fine white band instead of the circular
spot at the same area. Thin layer chromat-—
ographical studies on the pigments reveal-
ed that there is no qualitative difference
between two strains. However, we could not
find any histological difference between
two strains by ordinary paraffin section,
because the pigments were soluble in alco-
hol. Thus we tried cryostat section and
carbowax section. It has been found by
eryostat section that the differences bet-
ween two strains were based on the distri-
bution of red pigments in the epidermal
layer. The area lacking of red pigments
forms circular spot or fine band on the
dorsal side according to the strain. Thro-
ugh this area, white pigment cells, perha-
ps nephrocytes (one of blood cells) in the
mesenchymal space can be seen. Therefore,
it is concluded that the differences betw-
een two strains may be attributed to the
difference of epidermal layer.

1114 Developmental Biology

DB 129

ANATOMY OF THE LARVAL MUSCULATURE IN THE
DROSOPHILA ABDOMINAL SEGMENTS.

S.So and H.Tsujimura. Lab. Biol. Tokyo Univ.
Agri. Tech., Fuchu, Tokyo.

We examined the anatomy of the larval
musculature of the abdominal segments by
reconstruction of the serial sections. The 2nd
to 7th abdominal segments has the same basic
Plan of the musculature. The 3rd abdominal
segment has 30 pairs of muscles: 9 dorsal
muscles, 14 ventral muscles, and 7 lateral
muscles. The 2nd, 4th, 5th, and 6th segments
has the same muscles as the 3rd one and are
almost completely homologous to the segment.
However, the Ist segment has 28 of them and 2
pairs of specific muscles. This shows that the
segment has a modified basic plan. This may be
due to the specific role of this segment that
the segment unites the thoracic and abdominal
segments. On the other hand the 8th abdominal
segment shows a specific pattern of the
musculature. It has 26 pairs of muscles and
has almost no Similarity with the other
abdominal segments. This agrees the specific
role of this segment, which has the anus and
posterior spiracles and are the terminal
segment of the body.

DB 130

ANATOMY OF THE LARVAL MUSCULATURE IN THE
DROSOPHILA HEAD AND THORACIC SEGMENTS.

H.Tsujimura and S.So. Lab. Biol. Tokyo Univ. Agri.
Tech., Fuchu, Tokyo.

We examined the anatomy of the larval
musculature of the head and thorax by
reconstruction of the serial sections. The 3rd
instar larva was fixed with Bouin's fixative and
embedded with the paraffin. The sections were
cut 8 um thick and Stained with the HE solution.
The structure of the larval body was
microphotographed and the musculature was
reconstructed.

The head region of the larva was invaginated
in the thoracic segments and formed a
pharyngeal structure. There are 38 muscles in
this region: 31 in the structure and 7 between
the structure and body wall. The musculature of
the prothoracic segments is very different from
the other thoracic segments, suggesting that
this segment is greatly specified. The prothorax
has 7 muscles in the segment and, more over, 6
muscles are coming here from more posterior
segments. The mesothorax has 25 muscles: 9
dorsal, 5 lateral and 11 ventral. Six of them
runs into the prothorax beyond the segment
boundary. One muscle connects to the anterior
spiracle and plays the role in retraction of it
into the mesothorax. The metathorax has 25
muscles: 9 dorsal, 5 lateral and 11 ventral.
These plays roles in bending and shortening of
the thoracic segments. The meso- and meta-thorax
are incompletely homologous and have some
specificity.

DB 131

CHARACTERIZATION OF POKKURI, A MUTATION
AFFECTING LONGEVITY AND EYE DEVELOPMENT OF
DROSOPHILA MELANOGASTER.

I.Nihonmatsu, D.Yamamoto, Y.Sano, R.Ueda,
H.Tei, T.Yokokura, S.Togashi, C.Kobayashi,
S.Tsurumura and K.Sato. Mitsubishi Kasei
Institute of Life Sciences, Machida, Tokyo.

Using P-element mutagenesis and functional
screens, we isolated a mutant, pokkuri (pok:
22C), which showed a dramatic reduction in
longevity at the adult stage. pok was also
defective in ommatidial development. Although
the number of ommatidia composing single
compound eyes was normal (#770), single
ommatidia in pok eyes contained 1 to 16 cells
with the mean of 8. This indicates that the
total number of photoreceptors in the compound
eye may be close to that in the wild type. A
monoclonal antibody against ninaE-opsin stain-
ed selectively outer photoreceptors in pok
eyes, indicating that the central and outer
photoreceptors entered the proper development-
al pathways in pok, at least in terms of opsin
specification. It seems that the pok gene is
essential to cell assembly rather than
proliferation in the developing eye. In the
pok eye discs, the cluster arrangement showed
marked irregularities beginning just behind
the morphogenic furrow. Thus the mutant can
be classified as having a defect in formation
of the array of photoreceptor clusters.
Mosaic analysis is currently being undertaken
to determine whether pok expression is
required in all cells or is instead necessary
in a discrete subclass of photoreceptors.

DB 132

STAGE-SPECIFIC PUFF INDUCTION BY DIGITONIN
TREATMENT OF DROSOPHILA SALIVARY, GLANDS
M.Asaoka’ ,M.Myohara“~ and M.Okada!, Trtnst.
Sus Biol.Sci.,Univ. of Tsukuba, Ibaraki and
Natl.Inst.of Seric.Entomol.Sci., Tsukuba.
Salivary gland chromosomes of Drosophila

melanogaster are useful for studying the
gene regulation mechanisms underlying the
developmental processes, because the dif-
ferential gene expression in development is
visualized as regularly ordered alterations
of the puffing pattern in the chromosomes.
Previously, we showed that a mild deter-
gent, digitonin, induced puffing at 30
specific loci (digitonin puffs) in salivary
glands from late third instar larvae (Dev.
Biol. 130:346, 1988). In this study, we
induced digitonin puffs in salivary glands
of two different stages, the late third
instar larva (PS1) and early prepupa
(PS10), to study the relevancy of the
digitonin puff induction to the gene regu-
lation mechanism used in the normal devel-
opment. Out of the 30 loci where puffing
was induced when digitonin was applied at
PS1, 23 loci didn't respond when digitonin
was applied at PS10. The remaining loci,
including 6 heat shock loci, responded to
digitonin at the both stages. No stage-
specific digitonin puff was induced at PS10
or prepupal stage. The results suggest that
the specific puff induction by digitonin is
relevant to the mechanisms controlling the
differential gene expression in the normal
development.

Developmental Biology eas

DB 133

DEVELOPMENT OF THE SEA STAR, LUIDIA SENEGALENSIS
(LAMARCK).

M.Komatsu, C.Oguro and J.M.Lawrence!, Department of

Biology, Faculty of Science, Toyama University,

Toyama, Japan and ‘Department of Biology, University

of South Florida, Florida, U.S.A.

The entire process of development through metamor-

phosis from eggs to juveniles was observed in the 9
armed sea star, L. senegalensis. Adults were col-

lected from the coast of Florida in April and July,
1989. Spawning was induced by intracoelomic injec-—
tion of 1-methyladenine. Larvae were fed cultured

algae. Ova are transparent, 200 ym in average dia-
meter. The first cleavage occurs ca. 1 hr after

insemination at 25°C. Cleavage is total and radial.
Embryo develops into a bipinnaria through a wrinkled

blastula stage. It grows to a full size, 1.5 mm in
length and has 5 pairs of bipinnaria arms 13 days

after insemination. Metamorphosis takes place while

bipinnariae are pelagic. At metamorphosis, the
larval part is abeor ed into the asteroid rudiment.
The larval part disappears completely within 1 day
after the commencement of absorption of the larval
part (18 days after insemination). Juveniles are
550 pm in diameter 1 week later. They have 5 arms,
each of which bears 2 pairs of tube-feet.

This fact shows that the supernumerary arms will

be formed later. The present species has only a bi-

pinnaria larval stage, without a brachiolar stage,
and therefore, undergoes the nonbrachiolarian type

of development like all species of Luidia previously

reported. It is concluded from the results in the
present study that bipinnaria of Luidia is not
always large and complex. This is contrary to the
previously held belief.

DB 134

EFFCTS OF POTASSIUM ION AND RESPIRATORY INHIBITORS ON
THE BARRIER FUNCTION OF SEPTATE JUNCTION IN THE
STARFISH LARVA

Y. Okushima, H. Kaneko, and M. Dan-sohkawa.

Dept. of Biol., Osaka City University, Osaka.

Septate junction (SJ) is a belt-like structure
that binds invertebrate epithelial cells and is
situated near the apical end of the cell contact. It
seals off paracellular diffusion of molecules through
an epithelium. On the other hand, We have reported
in the starfish larvae that the SJ of the body wall
epithlium opens to let through macromolecules
(dexran-150, 1gG) into the blastocoel via the
intercellular spaces under environmental conditions,
such as hypertonicity generated by glycine and
deprivation of Ca2* or Mg2t.

In an effort to find conditions giving the
similar effect, we treated the larvae (early
bipinnaria stage) with sea water containing various
doses of K* (0.025-0.3 M) or respiratory inhibitors,
dinitrophenol CDNP), KCN, or NaN3. All of these
treatments were found to cause permeation of
fluorescence-labelled macromolecules into the
blastocoel. Moreover, addition of K*+ to the sea
water resulted in a stronger reaction than addition
of the same concentration of Nat. The efficacy of
K+, therefore, is based on its own action rather than
hypertonicity caused by its concentration.

Electronmicroscopic studies revealed that the
10-25 septa, which the larval SJ posseses, were more
or less obscured in all the cases, while DNP and
glycine also widened the intercellular space
immediately basal to the SJ .

These results indicate that membrane potential
and cytoplasmic energy participate in the mechanisms
that regulates the barrier function of the SJ.

DB 135

PROBABLE RECEPTORS FOR INSULIN IN CULTURED
MICROMERE DERIVED CELLS OF SEA URCHIN EGGS

S. Kuno, K. Mitsunaga, A. Fujiwara and
I. Yasumasu. Dept. of Biol., Sch. of
Educ., Waseda Univ., Tokyo.

The outgrowth of pseudopodial cables
from cultured micromere-derived cells of
sea urchin embryos is found to be induced
by insulin as well as by horse serum. In
the presence and absence of insulin or
horse serum, the activity of protein tyro-
sine kinase, localized in plasma membrane
fraction, increased in the cultured cells
in initial 10 hr of culture and thereafter
the activity was maintained high. The
Phosphorylation of tyrosine residue in the
cultured cells was very low in its rate
unless the culture media contained insulin
or horse serum. Probably, increase in the
rate of tyrosine residue phosphorylation
is probably due to activation of protein
tyrosine kinase having receptors of
insulin and horse serum in plasma
membrane. Insulin binding occurred in the
cultured cells and bound insulin was
easily replaced by horse serum. Amount of
insulin binding site was calculated to be
equivalent to about 6 pg insulin/mg cell
protein. Kd value for insulin binding was
assumed to be slightly higher than horse
serum. Probably, the cultured cells have
protein tyrosine kinase having receptors
for insulin in their plasma membrane and
the activation of the enzyme finally
causes growth of pseudopodial cables.

DB 136

EARLY INDUCTIVE INTERACTIONS RESPONSIBLE
FOR DEVELOPMENTAL POTENTIAL OF MESOMERES
IN 16-CELL EMBRYOS OF A SAND DOLLAR,
PERONELLA JAPONICA.

S. Amemiya, Misaki Marine Biologcial
Station, University of Tokyo, Kanagawa.

Early inductive interactions
responsible for developmental potential of
mesomeres in 16-cell embryos were examined
by using embryos of a sand dollar,
Peronella japonica, cultured in Ca**-free
sea water (CF) or in normal artificial sea
water (Jamarine U; ASW). The eggs whose
fertilization membrane was removed
mechanically by aspirated through a Pasteur
pipette, or chemically by treated with
trypsin, were cultured in ASW or in CF.
Mesomere pairs of the embryos were isolated
at 16-cell stage, put into 80-well culture
plates coated with 1.2 % agarose,
containing 0.5 ml of culture medium
(100 ug/ml streptomycin and 100 units
penicillin/ml ASW), cultured for 5 days at
25 °c, and examined the developmental
potential. Spicules were formed more
frequently in the mesomere pairs isolated
from the embyos cultured in CF than in
those cultured in ASW, regardless the
fertilization membranes were removed
mechanically or chemically. These results
indicate that the developmental potential
for mesomeres is influenced by inductive
interactions of other blastomeres during
early cleavage stage.

1116 Developmental Biology

DB 137

DEGRADATION AND UTILIZATION OF THE YOLK
DURING DEVELOPMENT OF THE SEA URCHIN.

Yo eVokoitay SBilol.) hab),eAcha® Pres sUmiv.),
Nagoya

Degradation of the protein and the
lipid in yolk granules was studied in Hemi-
centrotus pulcherrimus. Degradation process
of a major yolk glycoprotein (MYG) was ana-
lyzed by immunoblotting and localization of
the degradation products was observed by
immunofluorescence microscopy using mono-
clonal antibodies {mA72, mA94) against 72K
and 94K proteins, which are 2 of degrada-
tion products. Immunoblot analyses suggest
that MYG is cleaved into 114K and 94K pro-
teins and 114K protein is further processed
into 94K protein, since mA72 reacted with
MYG and 72K protein and mA94 reacted with
MYG, 114K and 94K proteins. The proteins
reacting with mA72 and mA94 were distrib-
uted evenly in the cytoplasm of eggs. Al-
though major fluorescence due to mA72 and
mA94 was observed in the cytoplasm of em-
bryonic cells, fluorescence resulted from
mA72 and mA94 was detected also in the
blastocoel of plutei and on the surface of
gastrulae, respectively.

The changes in the protein and the
lipid in yolk granules incubated in acidic
conditions were analyzed. Triacylglycerol,
phosphatidylcholine and phosphatidyletha-
nolamine were lost more than 50% of their
content after the incubation for 24 hr,
however, 70% of cholesterol and 55% of
protein were conserved.

DB 138

EMBRYONIC INDUCTION IN HORSESHOE CRABS.
Ty JOO Dept.of Biol., Fac.of Edu.,
Shizuoka Univ., Shizuoka.

For the understanding of the role of a
cell-mass under the blastopore at the early
gastrula stage, I tried the transplantation
experiment. Intraspecific transplantation
of the cell-mass has been performed in an
Asian horseshoe crab, Tachypleus rotundicauda
and American species, Limulus polyphemus.
Secondary embryos were formed in the egg
transplanted with the cells. This results
suggest that the transplanted cell-mass
either induced secondary individual or
autonomously differentiated as an embryo.

To test those possibilities I carried out
interspecific transplantation of the
cell-mass from an American to Asian horseshoe
crab. The interspecific transplantation
produced secondary embryos, which had the
characteristics of the Asian species. This
rules out the possibility that the secondary
embryos were formed as the results of
development of the transplanted cell-mass.
The cell-mass be regarded as an organizer
when transplanted.

The homogenate of the cell-mass also
induced secondary embryos.

Besides, the homogenate of the cell-mass
of the spider, Chiracanthium japonicum
induced the secondary embryos in horseshoe
crabs.

DB 139

HOMOITOGENETIC INDUCTION OF NEURAL TISSUE
IN CHIMERIC EXPLANTS OF XENOPUS.

Keiji Itoh and Hiroshi Y. Kubota, Dept.
Zool., Fac. Science, Kyoto Univ., Kyoto.

When neural tissue is induced by neural
tissue that has already been induced by
mesoderm, this is called homoiogenetic
induction.

In order to examine whether there is
homoiogenetic induction in Xenopus,
chimeric explants were made between
presumptive anterior neural ectoderm of
X. laevis late gastrulae and presumptive
ectoderm of X. borealis early gastrulae in
two ways. In the first, two pieces of
ectoderm were juxtaposed under a glass
bridge to keep them flat, which is similar
to their relative locations in normal

embryos. In the second, two pieces of
ectoderm were combined to face each other
to give a large contact area. The

explants were cultured until control X.
borealis embryos developed to St. 40.
They were serially sectioned and stained
with neural-tissue-specific (NEU-1) and
epidermis-specific (XEPI-1, XEPI-2)
monoclonal antibodies and quinacrine. The
results showed that X. borealis” cells
located’ next’. to) neural’ cello W@eEromnmx-
laevis were also NEU-1 expressing in many
cases in either type of culture, which
indicates that competent ectoderm of X.
borealis was induced by neural ectoderm of
X. laevis. These experiments suggest that
homoiogenetic induction occurs in the
formation of neural tissue in normal
development of Xenopus.

DB 140

CON A-DORSALIZED VENTRAL MESODERM ACQUIRES
NEURAL INDUCING ACTIVITY IN CYNOPS EMBRYO
M.R.M. Diazl S. Osawal, 1.C. Takahashi2,
K. Takeshima2 and K. Takata3. lDept. of
BioOlo, Wa, OF Sele, Radioisotope Re-
search Center, Nagoya Univ., 3Gen. Educ.,
Doho University, Nagoya.

Ventral mesoderm explants were treated
with 3 uM Con A, aged for certain time
lapses and then sandwiched between ectoderm
explants for 8h. The mesoderm and ectoderm
explants were separated from each other,
cultured and assessed respectively for dor-
sal or neural differentiation.

Con A-treated ventral mesoderm began to
acquire neural inducing activity 20h after
Con A treatment, reached its peak at about
24h and dropped readily. Neural induction
was abolished after 36h and 42h but a sec-
ond minor peak was observed with the 48h
culture-mesoderm. Such a bimodal charac-
teristic of neural induction was also ob-
served with the living organizer. 69% neu-
ral differentiation was observed at St ll,
but it was abolished when the organizer was
aged for 3h and 6h. A second peak was ob-
served with the 10h aged organizer. Ante-
rior as well as posterior neural structures
were observed in both Con A-dorsalized me-
soderm- and organizer-induced tissues.

Oligomannose-type oligosaccharide, a
potent Con A inhibitor, was applied to the
Con A~treated ventral mesoderm at certain
time lapses. Inhibition took place up to
30 min after treatment with Con A, but lh

later, the oligosaccharide could not re-
verse Con A action anymore.

Developmental Biology 17,

DB 141

MESODERM INDUCTION BY ACTIVIN A (=EDF) IN
THE BLASTULA ECTODERM IN XENOPUS LAEVIS.
T. Ariizumi, H. Uchiyama and M. Asashima
Department of Biology, Yokohama City
Univ., 22-2 Seto,Kanazawa-ku, Yokohama 236

We have found previously that activin A
(or erythroid differentiation factor=EDF)
has high mesodermal inducing activity in
the blastula ectoderm of Xenopus laevis.
In this experiment we tested activin A on
the ectoderm to determine whether the cell
differentiation depends on the concentra-
tion used or the length of treatment. When
we treated ectoderm with 50 ng/ml for 3
hours, the percentages of muscle and noto-
chord induced were 56% and 28% re-
spectively. When we treated ectoderm with
50 ng/ml for 3 days, the percentages of
muscle and notochord induced were 35 and
71% respectively. The lorger the treatment
with activin the higher the percentage of
notochord induced. The percentage of
muscle induced at a concentration of 1
ng/ml for 3 hours treatment with activin
was 94%. At the present time we could say
that activin has the highest mesodermal
inducing activity in the TGF-f family.
To determine the differentiation of muscle
and notochord, we used the specific anti-
bodies for immunohistochemistry. In the
areas of tissue differentiation, we de-
tected the specific immunofluorescence.
We also confirmed the presence of myosin
in muscle differentiation and keratin in
notochord differentiation using western
blotting analysis.

DB 142

INDUCTION AND DIFFERENTIATION ANALYSIS USING CUL-
TURED CELL DERIVED FROM XENOPUS LAEVIS
N.Moriya-, Y.Kudo!, K.Itoh* and M.Asashima?

Department of Biology, Yokohama City University,
Yokohama, “Department of Zoology, Kyoto University,
Kyoto.

In the early development of amphibians , mesoderm

induction plays a major role. Some mesoderm-
inducing substances have been discovered, for
example, fibroblast growth factor(FGF), transform-
ing growth factor-beta(TGF-beta) and activin
A(=EDF). The XTC cell line derived from Xenopus
tadpole also secretes mesoderm-inducing factor
which induces muscle, notochord and other mesoder-
mal tissues in isolated ectoderm from
Xenopus(anulan) embyos. We tested the mesoderm-
inducing activity of the XTC cell line on isolated
ectoderm from Cynops pyrrhogaster(urodele) embyos

using the sandwich method. In addition to the
XTC cell line, the XTY and A6 cell lines were also
tested. The XTY cell line is derived from a

tumor of the Xenopus, and the A6 cell line is
derived from Xenopus kidney. The XTC cell line
was found to induce muscle, notochord and other
mesodermal tissues in isolated ectoderm of
Cynops(urodele) blastulas. The A6 cell line also
induced mesodermal tissues in isolated ectoderm
from Cynops embryos at a rate higher than the XTC
cell line. The A6 cell line differed from the
XTC cell line in that the A6 cell line induced not
only mesodermal tissues but also neural tissue at
a high rate. The XTC cell line induced mesoder-
mal tissues in the ectoderm of Xenopus embryos,
but the A6 cell line did not.

DB 143

DETECTION OF ACTIVIN A (=EDF) AND RELATED
PROTEINS IN XENOPUS CELLS . 2
A.Fukui, H.Uchiyama’,,S.Nishimatsu“,

-Ueno“~ and M.Asashima

Department of Biology, Yokohama City
University, Yokohama 336. Institute of
Applied Biochemistry, University of Tsuku-
ba, Tsukuba 305.

' In our laboratory we have cultured sever-

al Xenopus cell lines for several years.
The A6 cell line which was derived from
normal kidney has a very high mesoderm
inducing CL Ay LLY However, some activin
like genes, inhibin B,, BMP-2 (bone mor-
phogenetic protein-2), BMP-4 and Vgr-1
were isolated from the genomic library of
the Xenopus liver. We investigated the
gene expression of these activin like
substances in Xenopus embryos using the
Northern blotting analysis. We were able
to detect the signals of BMP-2 from the
oocyte to the neurula. BMP-4 was recog-
nized from the early gastrula to the
neurula, but not detected in the oocyte,
cleavage and morula _ stages. Although
there is a high percentage of homology
between BMP-2 and BMP-4, they had quite
different gene expression during the Xeno-
fue embryonic development. It had been
ound first in our laboratory that activin
A had high mesoderm inducing activity, so
we expected these activin related proteins
to be correlated to mesoderm induction.
Using the Northern blotting analysis, we
detected a high level of BMP-4 m-RNA in A6
cells. We could not detect the Xenopus
activin m-RNA in the embryos, but we did
detect activin m-RNA signals in the XTC
and xXTY cell lines. The relationships
between the mesoderm inducing activity and
these substances are discussed.

DB 144

EXPERIMENTAL AND IMMUNOLOGICAL CHARACTERI-
ZATION OF NOTOCHORD DURING AMPHIBIAN DEVE-
LOPMENT .

H.Uchiyama!, T.Yamagighi’, K.Shibuya!,
M.Asashima!, and E.Sato“~. Dept. of Biol.,
Yokohama City Univ., Seto 22-2, Yokohama
236, Dept. of Biochem., Hamamatsu School

of Med., Handa-cho 3600, Hamamatsu 431-31.

When early larva of Xenopus’ and Cynops
were bisected transversely with a_ sharp
scalpel, notochord was found to protrude
from the cut ends. The protrusion occurred
only around St. 31 in Xenopus and St. 30 in
Cynops. This phenomenon lead to the assump-
tion that notochord has intrinsic tension
along A-P axis, thus stimulating the elon-
gation of the amphibian larva. We searched
for molecular markers of Cynops notochord
using immunofluorescent techniques. Anti-
fibronectin antibody stained notochord
sheath from the middle neurula stage.
Anti-keratin antibody stained notochord
cells and anti-type II collagen antibody
stained notochord sheath from the early
tailbud stage. We made monoclonal anti-
bodies against Cynops notochord. Not 1, one
of the obtained antibody, stained notochord
cells from the early tailbud stage, and the
staining was most intense at this stage.
The spatial and temporal pattern of stain-
ing was very similar to that of anti-kera-
tin antibody. Anti-type II collagen and
anti-fibronectin antibodies also stained
induced notochord in an explant of Xenopus
ectoderm which was subjected tc mesoderm
induction by activin A (=EDF).

1118 Developmental Biology

DB 145

EXPRESSION OF EMBRYO=SPECIFIC EPIDERMAL
KERATIN GENE IN ANIMAL CAP CELLS CULTURED
WITH MESODERM-INDUCING VEGETAL CELLS IN
XENOPUS LAEVIS EMBRYOS. Koichiro Shiokawal

Gen Fujiic, and Atsuchi Yasuol. lDept. Biol.

Fac. Sci., Kyushu Univ., Fukuoka; Zool.
IMs MAGA Sala G Wisetg BOlaye), “WWoliey/o)q

A Xenopus laevis keratin gene (XK81) which
has been isolated by T. Sargent et al.
(1985) is unique in that it is specifically
expressed in embryonic epidermis and the
expression is negatively controlled, or
Suppressed, when animal cap cells were
mixed with mesoderm-inducing vegetal cells.
To study the regulation of the expression
of the keratin gene at the cellular level,
the keratin gene mRNA level as well as the
levels of alpha-actin and histone H4 gene
mRNAS was examined in dissociated cells of
Xenopus laevis embryos under various con-
ditions.

Since preliminary experiments showed
that the keratin gene in animal cap cells
is suppressed when these cells came into
contact with vegetal cells at the blastula
but not at the gastrula stage, we attempted
here to exactly determine when the keratin
gene expression becomes resistant to the
inhibitory activity of vegetal cells. For
this purpose, we cocultured vegetal cells
and animal cells in various combinations,
Results obtained showed that animal cap
cells acquire the ability to express the
keratin gene autonomously at the late
blastula stage, slightly before the onset
of gastrulation.

DB 146

REGIONAL SPECIFICITY OF THE FEATHER- AND
SCALE-INDUCING DERMISES IN THE CHICKEN
EMBRYO.

Ws WELLING) 7» Sa Tanenura!, M. Saito!, and
K. Kitamura“. TDept. of Biol., Fac. of
Pharmaceut. Sci., Teikyo Univ., Sagamiko,
and “Dept. of Dev. Biol., Mitubishi Kasei
Inst. of Life Sci., Machida.

Amnionic ectoderm of 6.8-day chick
embryos, recombined with 6.8-day dorsal
dermis and cultured on host chorio-
allantoic membrane, developed feather
filaments consisted of a horny sheath and
barb ridges. With several feather
keratin-specific monoclonal antibodies,
these structures in the induced feather
filaments were shown to express horny
sheath- and barb ridge-specific keratin
antigens. In contrast, the amnionic ec-
toderm, recombined and cultured with 13-
15-day shank dermis, developed scales.
These results show that the dermis pos-
sesses definite regional specificity con-
cerning the feather- and scale-inducing
potency. By comparing polypeptide pat-
terns of dorsal and shank dermises by
means of HPLC and successive SDS-PAGE,
several polypeptides specific to either
one of the dermises were obtained.
Whether these region specific polypeptides
are candidates for the feather- or scale-
inducing substances is currently under in-
vestigation.

DB 147

POSSIBLE FUNCTION OF DNA REPLICATION IN
AVIAN SCALE DEVELOPMENT : REPLICATION
PATTERN OF W-REPETITIVE SEQUENCES IN
INDIVIDUAL CELLS .

S. Tanaka, T. Ueda, K. Nakajima and

T. Higashinakagawa, Dept. Develop.
Biol., Mitsubishi

Kasei Inst. Life Sciences, Machida, Tokyo 194.

To study significant function o

replication in embryonic development, methods
of quantitative in situ hybridization were
developed and applied to examine replication
pattern of specified sequences in a genome
DNA. W chromosome specific repetitive
sequences DNA ( Gift from Prof. S. Mizuno) was
biotinylated by random primer reaction for
use as a probe. The cell cycle phases of
respective cells was determined by a system
of multiparametric microfluorometry. The
hybridized DNA _ probe, detected after
immunofluorescence staining, was quatitated
by the system equipped with a scanning stage.
Time of replication of the sequences in scale
dermis was consistently very late in S phase
during a period from day 5 through day 12.
Replication duration were remarkably exteded
only during a quite short period ranging from
day 71/2 through day 8 when DNA tenia on
period also transiently elongated.

DB 148

DIFFERENTIATION OF THE PRECARDIAC MESODERM
OF CHICK EMBRYO
Y.Yamazaki, R.Hirakow. Dept. of Anatomy,
Saitama Medical School, Saitama

In vitro study was made to explore
the interaction between the mesoderm, the
endoderm and the ectoderm extirpated from
the precardiac area in chick embryos at
stage 5. Three types of tissue and two
different liquid media were examined. The
tissues were: 1) the precardiac mesoderm 2)
the precardiac mesoderm and the underlying
endoderm and 3) the precardiac mesoderm and
the overlying ectoderm. Firstly, the
tissues were placed in plastic dishes
filled with Eagle’s MEM. The precardiac
mesoderm in the absence of other tissue
disintegrated into single cells after 1 day
of culture. These cells neither attached to
the dish, nor had beating capacity. When
the precardiac mesoderm was cultured in
association with the endoderm, it formed a
beating tissue attached to the dish, after
1-2 days of culture. Myofibrils were
detected in beating cells by TEM. When the
precardiac mesoderm was cultured in
association with the ectoderm, it did not
dissociate, attaching to the dish, but
pulsation was not observed after 2 days.
Next, the precardiac mesoderm was cultured
in Eagle’s MEM supplemented with 15% horse
serum. It formed a beating tissue attached
to the dish after 1-2 days. This experiment
shows that the endoderm may Support
differentiation of the precardiac mesoderm.

Developmental Biology 1119

DB 149

INDUCTIVE EFFECT OF THE OESOPHAGEAL MESEN-
CHYME OF THE CHICK EMBRYO ON THE DIFFEREN-
TIATION OF THE PROVENTRICULAR ENDODERM.
S.Matsushita. Dept. Biol., Tokyo Women’s
Medical College, Tokyo.

The endoderm of various regions of the
digestive tract of 6-day chick embryos was
cultured in recombination with the oesopha-
geal or other mesenchymes of 6-day embryos,
and it was examined whether the oesophageal
mesenchyme had the mesenchyme-specific
inductive ability on the differentiation of
the digestive-tract endoderm.

Among the intact fragments of various re-
gions of digestive tract cultured for 12 or
18 days, only the oesophageal explants de-
veloped stratified squamous epithelium spe-
cific to the differentiated oesophageal
epithelium. When the endoderm of these re-
gions was cultured in recombination with
oesophageal mesenchyme, stratified squamous
epithelium appeared only in the recombina-
tes with oesophageal or proventricular en-
doderm. The proventricular endoderm cul-
tured in recombination with the mesenchyme
other than that of oesophagus never devel-
oped stratified squamous epithelium, which
denied the possibility that the proventri-
cular endoderm was easy to show oesophagus-
type differentiation under the stimulus of
heterologous mesenchyme. Thus, the present
study showed that the oesophageal mesen-
chyme could elicit oesophagus-type differ-
entiation at least in the proventricular
endoderm.

DB 150

SUPPRESSION OF DIFFERENTIATION OF
DIGESTIVE ORGANS BY INHIBITORS OF
PROTEIN KINASE C IN CHICKEN EMBRYOS.
S.Yasugi' , E.Yasugi? and K.Kano 2
Zool. HINGre > 5 Fac. Sea. and ? Dep.
Physiol. Chen. and Nutrition, Fac.
Med., Univ. Tokyo, Tokyo.

To examine the possible involvement
of protein kinase C (PKC) in ehe
process of morphogenesis and cyto-
differentiation of digestive organs
of chicken embryos, the rudiments of
proventriculus (PV) and small intes-
tine were cultivatated in vitro in
the presence of inhibitors (H7 or
sphingosine) or an activator (SC9) of
PKC. HY hindered the development of
both organs dose-dependently and at
100 pM, both organs almost degenerat-—
ed and expressed no pepsinogen and
sucrase, marker enzymes of PV and
small intestine, respectively. These
influences of H7 were counteracted by
the addition of SC9: Both organs
survived and more or less showed

morphogenesis, although cytodif-—
ferentiation to express pepsinogen
and sucrase was not achieved.
Sphingosine, up to 100 uM (PV) and 10
pM (small intestine), did not show

inhibitory effect on mopohogenesis,
but cytodifferentiaion of both organs
was invariably affected. These
results suggest that PKC plays some
important roles in the differentia-—
tion of embryonic digestive organs.

DB 151

REGIONAL DIFFERENCE OF EXPRESSION AND
METHYLATION OF EMBRYONIC CHICK
PEPSINOGEN GENE

K.Fukuda! , M.Ichinose? and S.Yasugi!
L7G ain sith each Sea eae ands taDepit +
Imes Wiecls 7 Iees Wietla, Wisaiiyc — WwOlAYC),
Tokyo

Embryonic chick pepsinogen(ECPg) is a
marker protein of the proventricular
(Pv) epithelium. During normal devel-
opment expression of ECPg gene is
restricted to Pv epithelium, and it is
never expressed in other organs. It
has been reported that in rat and
human the pepsinogen genes are hypo-
methylated in the stomach, where the
gene is expressed. To know whether DNA
methylation is involved also in ECPg
gene regulation, we analyzed methyl-
ation of ECPg gene in normal embryonic
chick organs using methylation-sensi-

tive restriction enzymes. Since ECPg
gene showed polymorphism among indi-
viduals in Southern hybridization

after digestion with several restric-—
tion enzymes, we selected those indi-
viduals which have the same hybridiza-
tion pattern for analysis of methyl-
ation. The hybridization patterns of
DNA samples from ECPg-expressing and
—-non-expressing organs were identical.
These results suggest that DNA methyl-
ation is not involved in the regula-
tion of ECPg gene expression during
normal development.

DB 152

ACID PROTEINASES OF LARVAL FORE-GUT AND
ADULT STOMACH OF RANA CATESBEIANA.

T. Inokuchi, K.kobayashi and S.Horiuchi
Life Sci. Inst., Sophia Univ.,Tokyo.

Acid proteinases existing in mucosa of
adult stomach of Rana catesbeiana could be
separated into six ( I, II-1, II-2 II-3,
III and IV ) by chromatography. These
enzymes were divided into two types.

The first type ( I to III ) was newly
appearing during metamorphosis. M.W. of
partially purified enzymes II-2 and III
were estimated to be 44,000 and 42,000 by
SDS-PAGE. M.W. of two enzymes decreased to
36,000 after acid treatment. Acid-treated
enzymes showed maximal activity at pH 2.0-
3.0 and their activities were inhibited by
pepstatin. As a result, enzymes II-2 and
III were identified as pepsin-type
proteinases.

The second type ( only IV ) was existing
in larval fore-gut as well as adult stomach
throughout metamorphosis. The enzyme was
tried to purify from adult stomach and
larval fore-gut. M.W. of partially
purified enzyme IV was estimated to be
50,000 in the presence of mercaptoethanol,
but about 100,000 in the absence of
mercaptoethanol by SDS-PAGE. Enzyme IV
seemed to be composed of subunits. They
showed maximal activity at pH 2.0-3.0 and
were bound on Con A-Sepharose. Their
activities were inhibited by pepstatin.
These characteristics were not different
between adult and larval enzyme IVs and
resembled to those of cathepsin E.

1120 Developmental Biology

DB 153

TRANSCRIPTION OF VENTRAL PROSTATE-SPECIFIC
GENE IS INDUCED BY EPITHELIAL-MESENCHYMAL
INTERACTION.
H.Takeda~, N.Suematsu and T.Mizuno?.
Tsukuba Center, RIKEN, Tsukuba, “Dept. of
Biochem. St. Marianna Univ. Sch. Med.,
Kawasaki, and ~Dept. of Pharmaceut. Sci.,
Teikyo Univ., Sagamiko.

The adult rat prostate consists of
ventral, lateral and dorsal lobes. In spite
of their common developmental origin, their
epithelia show the different biochemical
characteristics, for example, in the pro-
teins they secrete. In the present study,
we examined a role of the mesenchyme in the
lobe-specific cytodifferentiation of the
prostatic epithelium. The prostatic steroid
binding protein (PSBP) was used as a spe-
cific differentiation marker for the ven-
tral prostate.

Fetal urogenital sinuses were subdivided
into dorsal and ventral parts and the
reciprocal epithelial-mesenchymal recombi-
nations were carried out. In situ hybridi-
zation analysis with anti-sense RNA probes
demonstrated that the transcription of the
PSBP was induced in the ventral and dorsal
epithelia when they were associated and
cultured with the ventral mesenchyme. In
contrast, the transcription was never ob-
served in the epithelia associated with the
dorsal mesenchyme. These results show that
the ventral lobe-specific’ cytodifferentia-
tion in the prostatic epithelium could be
directed by the ventral mesenchyme through
epithelial-mesenchymal interactions.

DB 154

EGF INHIBITS FORMATION OF THE GLANDULAR
LUMEN OF THE UROPYGIAL GLAND IN QUAIL
EMBRYOS

Y. Fukui Dept. Biol., Tokyo Women's
Medical College, Tokyo

The glandular lumen and the papilla are
formed in the uropygial gland of the quail
embryo at day 10. Both of them are also
formed when the uropygial rudiment of the
8.5-day quail embryo is cultured for 2
days in serum-free medium. However, when
50 ng/ml of EGF is added to the medium,
the rudiment fails to form the glandular
lumen and the uropygial papilla. Tt
remains at the placode stage or only
undergoes epidermal invagination into the
mesenchyme. In many EGF-treated explants
the basal surface of the epidermis is
waved while it iS smooth in control
explants. Failure of lumen and papilla
formation was observed when the rudiments
were treated with EGF for longer than 16
hours at the beginning of culture.
Stratification and cornification of the
epidermis take place when rudiments’7 are
cul tured in the medium containing EGF at
the concentration of 500 ng/ml.

Labelling indices were higher in the
glandular epithelium than in the
neighbouring epidermis in the cultures
treated with 50 ng/ml of EGF as well as in
control cultures. Higher concentration of
EGF increased L.]. in the epithelium but
not in the mesenchyme.

DB 155

TRANSFILTER ANALYSIS OF MESENCHYMAL
INFLUENCE ON BRANCHING MORPHOGENESIS OF
MOUSE SALIVARY EPITHELIUM

Y. Takahashi! and H. Nogawa2 1Z001. Inst.,
Fac. of Sci., Univ. of Tokyo, Hongo,
Tokyo 113 and 2Biol. Lab., Coll. Arts &
Sci., Chiba Univ., Yayoi-cho, Chiba 260

We developed a novel transfilter culture
system in order to analyze the mesenchymal
influence on branching morphogenesis of
mouse fetal salivary epithelium. The
submandibular salivary epithelium from
early 13-day mouse fetuses was covered with
Matrigel, separated from the mesenchyme by
membrane filter and cultivated on
semisolid medium for 3 days.

The epithelium separated from the
mesenchyme by the filter with pore size
0.4, 0.1 or 0.05 um showed active branching
morphogenesis and there was statistically
no significant difference in the number of
lobules among the cases with different
pore sizes.

The ll-day lung mesenchyme supported
branching of the submandibular epithelium
in transfilter culture, while the 13-day
mandibular mesenchyme did not, which
correlated with the results in direct
recombination culture.

These results suggest that the
mesenchymal influence on branching
morphogenesis of the epithelium can be
analyzed into extracellular matrices and
some diffusible factors, and the latter is
organ-specific.

DB 156

LOCALIZATION OF ELASTIN mRNA IN THE WALL OF
ARTERIOLE DURING MOUSE ORAL DEVELOPMENT.
T.Yamaai, Dept. of Oral Anat.II, Okayama
Univ. Dent. Sch., Okayama

The wall of arterioles basically consist
of three tunics, the tunica intima, tunica
media and tunica adventitia. The elastic
fibers are found most abundantly in the
elastica interna, which is the boundary
between the tunica intima and tunica media.
The cell of arterioles which synthesizes the
elestin of the elastica interna is still not
clear.

The localization of the elastin mRNA was
investigated by a Simplified in situ
hybridization method with riboprobes in the
wall of the arteriole during development.
C3H mice were fixed by perfusion with 4%
paraformaldehyde in 0.0/M Na-cacodylate
buffer. In situ hybridization with 35S
labeled riboprobes was carried out on 5 um
thick serial paraffin sections of the upper
jaw at the level of the 1st. molar and then
the sections were made into autoradiograms.

Signals of anti-sense probe _ were
observed only on the endothelial cells in
the tunica intima from 18-day fetus to 4-day
neonate. Whereas the resorcin-fuchsin
positive elastic fibers were observed only
in the elastica interna in same stage. The
signal detected cells began to appear in the
tunica media from 4-day neonate. The signal
intensity became maximum at one week neonate
and weakened thereafter both on the tunica
intima and tunica media. The results
indicate that the elastic fibers in the
elastica interna is synthesized only in the
endothelial cells of arterioles in the stage
of development.

Developmental Biology NI

DB 157

PRELIMINARY OBSERVATION ON’ THE
JUNCTIONS AMONG THE EPIBLASTIC CELLS
OF CHICK EMBRYO DURING THE PRIMITIVE
STREAK FORMATION.

S. Takeuchi(Inst. Biol. Sci., Fac. Sci., Kanagawa
Univ.

In order to know when and how the epiblastic

cells establish epithelial structure during an initial
phase of morphogenetic movement, the subcellular
structure of epiblastic cells were surveyed with a
transmission electron microscope.

The epiblastic cells just prior to the top of
primitive streak(PS) were fixed and ultra-thin-
sectioned in an ordinary procedure. At stage
2(Hamburger and Hamilton's table), the epiblastic
cells in low profile were connected with, perhaps,
primitive Zonula occludens. From St. 3- to 3+
when PS elongated and the epiblastic cells increased
in height, an amorphous mass of electron-dense
substance was recognized junst under Z. occludens,
though not in every cell. Sometimes, fibers were
observed to protrude from the mass. Well defined
desmosomes were also started to be formed.

Based on the position and the state, both the
amorphous mass and the fibrous material might be
considered as F-actin bundles attaching to Z.
adherens. This supposition agrees with the previous
observation that, at St. 3, F-actin bundles began to
be aligned along cell border making a polygon in
the cell surrounding PS (Takeuchi, 1988).

DB 158

BEHAVIOR OF MESODERMAL CELLS EN MASS
AND THEIR CYTOSKELETAL ORGANIZATION

ON PARALLEL LINEAR FIBRONECTIN.
R.Toyoizumi!, K.Shiokawa?, S.Takeuchi!> 'Dept. of
Biol. Sci., Kanagawa Univ., Kanagawa, 2Zool.Inst., Fac. of
Sci.,Univ. of Tokyo, Tokyo.

In order to understand the process how the
mesodermal layer was formed in the chick
embryo, fragments of the mesodermal cells were
explanted onto the substratum on which parallel:
lines of fibronectin(FN) were deposited artifically
and the behavior of them were followed. And, the
distribution patterns of F-actin, q-actinin and
vinculin in these explants were surveyed with
immunofluorescent methods.

On EN lines, the explant became a flat sheet with
an elliptic outline, the major axis of which was
oriented to the direction of FN lines, and moved
along them. The marginal cells protruded filopodia
or lamellipodia which tended to attach to FN lines.
At these attachment sites, vinculin and a-actinin
co-localized both with the terminals of F-actin
bundles and FN lines. The marginal cells. were
proved, thus, to be responsible for the movement
of mesodermal cells en mass in vitro.

Also in ovo, the marginal cells of the mass are
supposed to anchor to and be guided with FN-rich
fibrous ECM on the undersurface of epiblast to
move outwards, and the rest were pulled outwards
in tandem to form the mesodermal cell layer.

DB 159

ULTRASTRUCTURAL CHARACTERIZATION OF BLAS-
TOCOEL WALL CELLS AT THE PRIMARY MESENCHYME
CELL HOMING SITES IN SEA URCHIN EMBRYOS.

Y.Nakajimal H.Katow2.!Dept.Biol.,Keio Univ.
Yokohama. Biol.Lab., Rikkyo Univ., Tokyo.

Histological characterization of blastocoel
wall cells at the primary mesenchyme cell
(PMC) homing sites during termination of
PMC migration was conducted in sea urchins,
Hemicentrotus pulcherrimus, Mespilia
gZlobulus, and Clypeaster japonicus. Trans-
mission electron microscopy indicated that
the sites .were composed of triangular
cells. These cells characteristically con-
tained clusters of mitochondria on the
basal side. Such mitochondria accumulation
occurred only during the period when PMCs
terminate their migration at the sites. Ac-
cording to the time-lapse videomicroscopy,
these cells also undulated their basal sur-
face during termination of PMC migration.
The undulation was inhibited by the treat-
ment with cytochalasin B, indicating that
the undulation involves microfilaments. aT
fact, these cells contained microfilaments
which were located beneath the plasma mem-
brane on the basal side. The bundles of
microtubules were aligned apico-basally.
These cells also contained well developed
rER which was filled with flocculus sub-

stance. Monensin treatment caused enor-
mously inflated rER filled with the sub-
stance. These results suggest that the

sites 1S active in secretion.

DB 160

INHIBITION OF JIN VIVO CELL MIGRATION BY
FIBRONECTIN-RELATED SYNTHETIC PEPTIDE, PRO-
ALA-SER-SER. 2 1
Hits Katow , and S. Sofuku , Biology
Laboratory, Department of Chemistry,
Rikkyo University, Toshima-ku, Tokyo.

Primary mesenchyme cells (PMCs) of sea
urchin blastulae use fibronectin (FN)
during migration in vivo. From our
previous in vitro studies, which were
conducted using the sea urchin,
Pseudocentrotus depressus, it has’ become
obvious that during migration PMC uses not
only Arg-Gly-Asp-Ser (RGDS) amino acid
sequence in the cell binding domain of FN,
but also Pro-Ala-Ser-Ser (PASS) amino acid
sequence in the same domain. PASS resides
immediate carboxyl terminus side of the
RGDS amino acid sequence in FN _ molecule.
Authenticity of the result, however, needed
to be proved in vivo.

In the present study the PASS peptide
was introduced into the blastulae of the

sea urchin Clypeaster Japonicus with
the intention of proving the previous in
vitro results. The introduction study

indicated that PASS inhibits PMC migration
in vivo. An in vitro immunohistochemistry,
which was also conducted in this’ study,
indicated that PASS peptide inhibits FN
attachment to PMC surface.

Thus, it is concluded that the PASS
amino acid sequence is used for FN binding
to PMC surface and for promotion of PMC
migration by FN, whereas RGDS is used only
for promoting PMC migration.

1122 Developmental Biology

DB 161

MIGRATION AND DIFFERENTIATION OF NEURAL
CREST CELLS TO PIGMENT CELLS IN CHICK
EMBRYO.

Y.Marikawal-2, K.Itoh!, T.Sakakura2 and
T.Morita*. ‘Dept. of Biology, College of
Gen. Education, Osaka Univ. Toyonaka,
2Dept. of Zoology, Fac. of Science, Kyoto
Univ. Kyoto and “Lab. of cell Biol. Riken,
Tsukuba.

The factors govern the migration and
differentiation of neural creast cells
into pigment cells have been studied by
the immunohistochemical method and the
cultur of neural creast cells. At the
initial migration stage of neural crest
cells, tenasein and fibronectin were
uniformly distributed in somites. At the
later migratory stage, the distribution of
fibronection did not change, but the
localized distribution of tenascin in the
ventral migratory pathway of neural crest
cells was observed. It was found that many
of differentiating cells into pigment
cells were lately derived from the neural
creast and the number of differentiating
cells into pigment cells have increased as
the retardation of cell migration from the
neural creast by the treatment of
cytochalacin D.

These results suggest that the
determination for the pigment cell may
be depend on the retention period of cells
in the neural creast and the distribution
of extracellular matrix in cell migration
pathways.

DB 162

DEVELOPMENTAL INTERACTION IN THE PIGMEN-
TARY SYSTEM OF THE MOUSE TAIL TIP. EFFECTS
OF COAT COLOR GENES ON THE EXPRESSION OF
TAIL-SPOTTING GENE.

T. Hirobe. Div. of Biol., Natl. Inst. of
Radiol. Sci., Chiba.

The tails of agouti C3H/HeJmsHir mice
are completely pigmented, whereas those of
black C57BL/10JHir animals possess unpig-
mented tips. Genetic analysis indicates
that white tail-tipping is due to an auto-
somal recessive, with incomplete pene-
trance, that segregates independently from
agouti with a maternal influence in the Fy
generation. To analyze the influence of
specific coat color genes on the expres-
sion of tail-spotting in mice, five
congenic lines of C57BL/10JHir with dif-
ferent coat colors were prepared. No
influence was observed on the occurrence
of tail-spotting in agouti (A/A), dilute
(d/d), F, between black and albino (c/c),
and F, bétween black and pink-eyed dilu-
tion toy However, the frequency of
tail-spotting was dramatically decreased
in brown (b/b) mice. These results suggest
that the mutant allele (b) at the brown
locus is involved in determining extent of
pigment areas in the tail tips of mice
through an interaction with the tail-
spotting gene.

DB 163

TEMPORAL CHANGES IN AGGREGATIVE ABILITY OF
CHICK LIMB BUD DISTAL CELLS.

Naoyuki Wada and Hiroyuki Ide, Biol.Inst.,
Tohoku Univ., Sendai.

We assumed that proximo-distal pattern
specification in the chick limb bud may be
caused by the change in the ability of
aggregate formation in the cells of
progress zone(PZ). So we examined the
developmental change of aggregative
ability in PZ cells,using chick-quail
chimeras.

When stage 20 PZ cells and stage 25 PZ
cells were mixed and cultured,stage 20
cells preferentially aggregated and formed
nodule-like structures,although stage 25
cells occupied the internodule spaces.

Next,the PZ cells at each stage were
mixed, reaggregated centrifugecally,
jacketed in the ectodermal hull,and grown
on a host limb bud. Stage 20 cells could
form proximal cartilage structures, but
stage 25 cells could not form the proximal
structures, instead they formed distal
cartilage structures with the younger
cells.

These result suggest that the
aggregative ability of PZ cells is high
at early stages and is lost gradually at
later stages,and this gradual change may
relate to pattern specification along the
proximo-distal axis.

DB 164

THE ANALYSIS OF THE POSITIONAL VALUE OF THE LIMB
BUD ALONG THE ANTERO-POSTERIOR AXIS WITH CHICK-
QUAIL CHIMERA

Koji Uchiyama and Hiroyuki Ide, Biol. Inst.,
Tohoku Univ., Sendai.

To elucidate the difference in positional
value-related property of the progress zone along
the antero-posterior axis, quail wing bud
fragments of the progress zone were grafted into
the different site along the antero-posterior axis.
Developed chimera wing was examined immunohisto-
chemically with A223 antibody. Moreover the
cell lineage of the wing bud mesoderm was traced
by the analysis of chimeric wing bud. When the
presumptive digit3 region of stage20 wing bud
was transplanted into the presumptive digit2 or
digit4 region, it formed the skeletal structures
which corresponds to the position in the host
wing bud. When the same transplantation was
performed at stage25, the grafted presumptive
digit3 region did not form the position-related
host skeletal structures. These results suggest
that the positional values along antero-
posterior axis are not yet fixed at stage20. At
stage25 the values fixed almost completely.

Comparing the cell lineage of the mesoderm
with the expression pattern of AV-1l, the AV-1l
distribution at stagel93 to 24 corresponds
roughly to the tissue proliferation pattern, that
is, the cell population which can express the
AV-1 antigen is already specified at stagel9-20.

Developmental Biology 1123

DB 165

CLONING AND EXPRESSION OF HOMEOBOX CONTAINING
GENE CHOX3. 6.

K. Hara', A. Kuroiwa?, T. Miyabayashi®, N.
Takahashi*, K. Ohsugi’ and H. Ide’. ‘Biol. Inst.
and *Res. Inst. for TB and Cancer, Tohoku Univ..

“Dept. of Mol. Biol., Nagoya Univ.

As shown in the coordinate expression
pattern of Hox5 cluster genes in mouse limb bud
(Dolle et al., 1989), homeobox containing genes
play important roles in the process of limb
pattern formation.

The expression of Hox3.6 in 10.5 day mouse
embryo is observed in the posterior part of
spinal cord and hind limb bud.

We have cloned a chick homeobox containing
gene which corresponds to Hox3.6, termed as
Chox3.6, and examined using in situ hybridization
the expression of Chox3.6 in chick limb bud
morphogenesis. The high expression of Chox3.6
was detected in peripheral region of st.23 limb
bud, especially the antero-proximal and
postero-distal region. We also detected the
expression in apical ectodermal ridge and
peripheral region of dosal half in st.24 limb
bud.

DB 166

DISTRIBUTION OF SYNTHETIC RETINOIDS
APPLIED TO CHICK LIMB BUD: ANALYSIS BY
AUTORADIOGRAPHY.
io WehinbieeS lMe5 Olney plreal | He ICIS, “Eich Nee
Shudo. Biol. Inst., Tohoku Univ., Faculty
of Pharm. Sciences, University of Tokyo.

Retinoic acid (RA) May be a natural
morphogen involved in limb pattern
formation since it induces antero-
posterior duplication and exists 2.5 times
more abundant in the posterior side than
in the remaining anterior part in
developing chick limb bud. It has been
reported that RA applied at the anterior
Margin of wing bud forms an
anteroposterior concentration gradient
across the limb bud (Tickle et al., 1985).
In order to find the region where
exogeneously added RA moves and
distributes, we implanted carrier beads
containing tritiated stable synthetic
retinoids (Am80, Ch55) to the limb bud and
observed the distribution of the labeled
compounds with autoradiography. Strong
radioactivity was found to be accumulated
in the peripheral region of limb bud and
no gradient of the radioactivity was
observed. This accumulation pattern in the
peripheral region agrees with the
distribution of Vitamin A revealed with
anti-Vitamin A Mab. (Tamura et al., 1990).
These results suggest the importance of
peripheral region in RA function and also
the posibility of homogeneous distribution
of RA in the limb bud.

DB 167

IDENTIFICATION OF GTP-BINDING PROTEINS IN
CHICK EMBRYONIC SKIN AND THEIR
PARTICIPATION IN EPIDERMAL MUCOUS
METAPLASIA INDUCED BY RETINOID.
K.Mikami-Takei, H.Endo and A.Obinata.
Dept. Physiol. Chem., Fac. Pharmacol.
Sci., Teikyo Univ., Sagamiko, Kanagawa.
Epidermal mucous metaplasia of cultured
13-day-old chick embryonic skin is induced
by excess retinol ( aes ). Retinol
primarily affects the dermal cells to make
signals that alters epidermal
differentiation toward secretory
epithelium. When retinol-treated cultured
skin was treated with cholera toxin ( CT ),
pertussis toxin ( PT ) or NaF, mucous
metaplasia was stimulated ( 120-140 % )
and the effects were dependent on the
increase of epidermal cellular cAMP
concentrations. In the plasma membrane
fractions, 43 kDa proteins from both
epidermis and dermis were [32P]ADP-
ribosylated by CT. 41 and 39 kDa proteins
were [32P]ADP-ribosylated by PT in the
dermal membrane fraction, while only 41
kDa protein was specifically labeled in
the epidermis. As Bt2cAMP ( 2 mM )
enhances mucous metaplasia several times
( Obinata et al., 1990*), G proteins in the
epidermal cell are coupled to adenylate
cyclase, and there should be signals
that activate or inhibit adenylate cyclase
of epidermis.
*Obinata, Akimoto, Hirano and Endo. Exp.
Cell Res.(1990) in press.

DB 168

CHARACTERIZATION OF CONSTITUENT GLYCEROPHOS-
PHOLIPID FATTY ACIDS IN NEWT FORELIMB AND
TAIL REGENERATING BLASTEMA.
M. Ryuzaki. Dept. of Biol., Kitasato Univ.
School of Medicine, Sagamihara.
Clarification in this study was made of
the relationship between changes in membrane
phospholipids and the regeneration of newt
amputated limb and tail. Characterization
was thus made of constituent fatty acids in
phosphatidylcholine (PC) and phosphatidyl-
ethanolamine (PE) molecules. These phospho-
lipids are present in the forelimb having
ca. 1/2 the forearm (FL) and 1.5 cm tail tip
(T) amputated and the regenerating blastema
of FL (RBFL) and T (RBT) obtained from the
amputee of Triturus pyrrhogaster pyrrhogas-
ter. The major constituent fatty acids at
the 1- and 2-positions in PC were palmitic
(C16:0) (ca.55%), oleic (C18:1) (16%), stearic
(C18:0) (12%), myristic (C14:0) (9%) and ara-
chidonic (C20:4) (4%) acids and in PE, C18:0
(34%), Cl16:0(22%), C18:1(16%), C14:0(148),
C18:2(8%) and C20:4(8%). The percentage of
C16:0 increased ca. 11% the total fatty ac-
ids at position 1 of PC in RBT, and whereas
C14:0, C18:0 and C18:2 decreased 12-5%, C16:
0 and C18:1 increased 17-14% at position 2
from RBT and RBFL, compared to corresponding
T and FL. In RBFL, C18:1 and C18:2 decreased
ca. 10% while C18:0, Cl6:0 and C14:0 in-
creased 20-10% at position 2 of PE. C20:4
could hardly be detected in PC or PE from R
BT and RBFL. Gross positional distribution
patterns, essentially the same as those of
embryos (blastula), showed undifferentiated
cells rather than T and FL.

1124 Developmental Biology

DB 169

5-AZACYTIDINE INHIBITS CLOSURE OF THE
CEPHALIC NEURAL TUBE IN EAT EMBRYOS.
M. Matsuda. Dept. of Embryol., Inst. for
Dev. Res., Aichi.
fe a ——!
The head-fold stage rat embryos (9.5
days of gestation) were cultured for 48 h
in rat serum supplemented with or without
0.2 ug/ml 5-azacytidine. Control embryos
cultured for 33 h (16-somite stage)
completed the fusion of cephalic neural
folds. In contrast, the cephalic neural
plate remaind open in 5-azacytidine-
treated embryos after 48 h of culture.
However, the extent of their development,
which was scored according to the method
of Brown and Fabro (1981), was not
different between control and 5-
azacytidine-treated embryos except the
head region. There was no significant
difference in DNA and protein contents
between control and treated embryos
cultured for 36 h. Embryos were sensitive
to 5-azacytidine during 6-12 h in culture
(3-5-somite stages). Immunocytochemical
observations using 5-methylcytosine-
specific antibody indicated that the
methylation of DNA was depressed by 5-
azacytidine in the median part and apices
of the neural plate and in the area around
the apices. These results suggest that
the methylation of cells in these areas
play an important role in closure of the
cephalic neural tube.

DB 170

A GOLGI STUDY ON THE HISTOGENESIS OF THE
NEOSTRIATAL NEUORNS IN THE CAT.

N. IWAHORI, Dept. of Anat., Fac. of Med.,
Nagasaki Univ., Nagasaki.

The histogenesis of the neostriatal
neurons was studied in the cat using the
rapid Golgi method. The neostriatal neurons
were derived from the striatal elevation
protruding into the ventrolateral region of
the lateral ventricle. Cells seen in the
striatal elevation had irregularly shaped
somata which generated several processes,
and migrated toward the pial surface to .
differentiate into primordial neostriatal
neurons. Primordial neostriatal neurons had
irregularly shaped cell bodies which
extended one thin long process and two to
four short ones. The precursor form of the
neostriatal medium-sized spiny neurons was
first identified as spindle- or irregularly
shaped cells. As the development proceeded,
immature spiny neurons increased in somatic
size as well as in the number and length of
processes. In later developmental stages,
the processes became thick and possessed
numerous spines. The primordium of
neostriatal medium-sized aspiny neurons had
oval or triangular cell bodies, from which
three to six thin processes extended. In
older specimens, the number and length of
processes increased considerably. However,
many of the processes remained thin and had
few spines. Aspiny neurons differentiated
later than spiny neurons, and some showed
immature features even in postnatal stages.

DB 171

GAP JUNCTION mRNA EXPRESSION IN THE NEO-
NATAL RAT BRAIN.

A. Matsumoto’, Y. Arai’, A. Urano’and S.
Hyodo, 'Dept. Anat., Juntendo Univ. Sch.
Med., Tokyo, Ocean Res. Inst., Univ.
Tokyo, Tokyo.

Gap junctions are considered to play an
important role in intercellular communica-—
tion. Possible participation of gap junc-
tions is presumed in the neuronal organi-
zation of developing brain. We studied
the cellular distribution of the mRNA for
gap junction protein in the brain of rats
at postnatal day 2 by using in situ hybri-
dization. A complementary DNA specific
for the mRNA for rat liver gap junction
protein (connexin’32)\Ca ealtth fuomseeee DA.
Goodenough) was applied to in situ hybri-
dization histochemistry on cryostat and
paraffin sections of the middle level of
the brain. Autoradiographic signals for
connexin 32 mRNA were found to distribute
in various regions of the brain such as
frontal cortex, hippocampus, thalamus,
striatum and anterior hypothalamus. These
Signals were localized on neurons, glial
cells and ependymal cells. The level of
expression of connexin 32 mRNA on cryostat
sections was much higher than that on
paraffin sections. These results indicate
that connexin 32 mRNA may be expressed in
neural substrates in the neonatal rat
brain, and that the process of paraffin
embedding may result in degradation of
connexin 32 mRNA in tissues.

DB 172

EFFECTS OF HORSE SERUM ON THE DEVELOPMENT
OF ORGANOTYPIC CULTURE OF NEWBORN MOUSE
CEREBELLUM.

DaSatoma, Depita of) Biol = Colle wo terAnet sm ainG
Sci., Univ. of Tokyo, Tokyo.

Cerebellar culture has been used as a
useful model for studying the development
of central nervous system. Previously, we
reported that under standard conditions
(feeding medium contains 23% horse serum)
biochemical changes are closely related to
morphological development in nearly same
manner as in vivo. Without serum, myelin
formation does not occur. To clarify the
role of serum, we examined several biochem-—
ical changes in the early developmental
stages in the absence of horse serum. In
explants incubated in serum-free medium
(SFM) from first day in vitro(1 DIV),
protein contents decreased rapidly during
first week of incubation, but, the amounts
of y-aminobutyric acid, an inhibitory
neurotransmitter, increased according to
the same pattern observed in serum-contain-—
ing medium(SCM). In explants transferred
to SFM from SCM at 8DIV, cerebrosides, the
major glycolipids of myelin, did not
increse during the incubation period.

In explants transferred to SFM at 15 DIV,
myelin could be observed even at 22 DIV,
indicating that serum is not necessary for
the maintenance of myelin.

From these results, it was supposed that
horse serum contains the factor(s) which
induce oligodendrocyte differentiation.

Developmental Biology 15)

DB 173

A MONOCLONAL ANTIBODY STAINS MOUSE NEURAL
TISSUES.

H. Kitanil,3, K.Ito2, T.Morita2, T.Sakakura! and

Y. Tomooka!

1 RIKEN, Life Science Center, Tsukuba, 2 College of General

Education, Osaka University, Toyonaka, 3 National Institute
of Animal Health, Tsukuba

A monoclonal antibody (MAb) against neural cells of mouse
embryos has been generated as a possible tool to study the
development of mammalian CNS. Rat was immunized by
subcutaneous injections of neural cell homogenate prepared
from the 10th day mouse embryos. After fusion of the rat
spleen cells with mouse myeloma cells, hybridomas were
screened immunohistochemically with polyester wax sections
of the 10 th day mouse embryos. MAb 16F6 strongly stained
almost all the cells of brain, spinal cord and ganglia as well as
nerve fibers in the 10 th day embryo. In the 16 th day embryo,
nerve plexus in the intestine as well as cells in the adrenal
medulla were stained with this MAb. Immunoreactivity with
MAb 16F6 in brain, spinal cord, ganglia, nerve fibers, nerve
plexus and adrenal medulla retained after birth up to adulthood.
Both cell bodies and neurites of the cultured neural cells were
strongly stained with MAb 16F6. Moreover, cultured neural
crest cells became positive with MAb 16F6 when they
differentiated into neurons. Immunofluorescence in these neural
cells seemed to localize in the cytoplasm. Western blot analysis
showed c.a. 260 kD protein band recognized by this MAb. The
antigen recognized by this MAb may possibly be one of the
cytoskelton-associted proteins in the cells of neural lineage,
which appear in these cells from the early stage of their
development . MAb 16F6 could be a useful marker to study
neural cells both in vivo and in vitro.

DB 174

DIFFERENTIAL LOCALIZATION OF NTA ANTIGEN
IN THYMOCYTES AND NERVE CELLS.

M.Kamiya, M.Miyakawa, T.Seki, Y.Arai.
G.Ueda* and T.Shirai*. Dept of Anat. *Dept
of Pathol.,Juntendo Univ. Sch. of Med.,
Tokyo.

NZB mice, which often show immunolo-
gical abnormalities and nervous disorders,
are known to bear many kinds of autoanti-
bodies. Monoclonal NTA (naturally occur-
ring thymocytotoxic autoantibody) estab-
lished from NZB mice was found to be reac-
tive with mouse, rat, chick and human
brain tissues. In order to characterize
the brain and thymocyte antigens, the
molecular weight and the localization of
these two NTA antigens were examined.

SDS-PAGE and immunoblotting showed that
the molecular weight of NTA antigen of
brain tissues was 55kDa, being equivalent
to that of thymocytes. FACS (fluorescence-
activated cell sorter) analysis, however,
revealed that NTA reacted with thymocytes
but not with dissociated brain cells.
Cultured nerve cells were not immunostain-
ed with NTA unless they were fixed. These
results indicate that the neuronal NTA
antigen exists dominantly inside of the
cell, whereas the antigen of thymocytes is
located mainly on the cell surface.

DB 175

IMMUNOHISTOCHEMICAL ANALYSIS OF NTA-
REACTIVE ANTIGEN.
M.Miyakawa, M.kamiya’, Y.Arai’ and T.Shirai*
Depts. Anat’. and Pathol%, Juntendo Univ.
Sch.Med., Hongo, Tokyo.

Natural thymocytotoxic autoantibody
(NTA) is known to react with surface anti-
gen of thymocytes and a part of T cells.
Recently we reported that NTA binds to
adult and embryonic nervous tissue. In
order to know the characteristics of NTA-
reactive antigen in the nervous system
localization of the antigen was studied by
immunohistochemical methods with monoclonal
NTA in BALB/e mice. In the embryo, NTA was
observed to bind to peripheral nerve fibers
and all over the brain and the spinal cord
except cell nuclei. In the neonatal cere-
bellum, developing dendrites of Purkinje
cells showed strong reactivity. However,
the white matter of the spinal cord was not
reactive at postnatal day 25. By the elec-
tronmicroscopic immunohistochemistry NTA-
reactive substance was found to be present
within the dendritic structures of Purkinje
cells in adult cerebellum. In addition,
some cells in the embryonic skeletal mus-
cular tissue showed NTA reactivity, but in
postnatal skeletal muscles immunoreactivity
was no longer found. The presence of NTA-
reactive antigen in the developing cells or
growing processes of the nervous and muscle
tissues seem to suggest the important role
of the antigen in morphogenic differentia-
tion.

DB 176

CHOLINE ACETYLTRANSFERASE DEVELOPMENT IN A
CO-CULURE OF SPINAL CORD AND SKELETAL
MUSCLE CELLS IS INHIBITED BY MYOGENIC
DIFFERENTIATION INHIBITORS.

M.Kengaku!, M.Nakane*, A.Kawata3, T.Deguchi?
and S.Kawashima!. !Zool. Inst., Fac. of
Sci, pUnivEOretokyo, Tokyo, sDeptslorMol..
Neurobiol., Tokyo Met. Inst. for Neurosci.,
Tokyo and #Dept. of Neurol., Tokyo Met.
Neurol. Hosp., Tokyo.

Choline acetyltransferase (ChAT) activ-
ity in spinal cord cells from E14 mouse
embryos is markedly increased when co-
cultured with skeletal muscle cells from
E20 embryos. This enhancement of ChAT
activity was suppressed in the presence of
concanavalin A, N-methyl-1-deoxynojirimycin
or TGF-8& which inhibit myoblast fusion. The
biochemical and morphological differentia-
tion of skeletal muscle cells was also
inhibited by these agents in parallel with
the inhibition of ChAT activity. TGF-8
accumulated fibronectin in the extra-
cellular matrix of skeletal muscle cells
and the inhibition of ChAT activity by TGF-
R was almost fully recovered by the
addition of a competitive inhibitor of
AIeOMOGiC ain Joalinclalime tO Cally Swisrace
receptor. These results indicate that the
development of ChAT activity in motoneurons
is dependent on the differentiation of
their target skeletal muscle cells and is
partly regulated by the expression of
fibronectin in skeletal muscle cells.

1126 Developmental Biology

DB 177

CHANGES IN UBIQUITIN-PROTEIN CONJUGATES BY
MICROTUBULE POISONS IN PC12H CELLS.
K.Takada, M.Matsuda and K.Houi'. Dept. of
Biochem., Dept. of Internal Med(III),
Jikei Univ. Sch. of Med., Tokyo.

Ubiquitin (Ub) makes to conjugate ATP-
dependently with intracellular proteins.
And ubiquitination is necessary for Ub-
dependent proteolysis. Ubiquitinated
cytoskeletal proteins accumulate in many
inclusion bodies of the neurodegenerative
disorders. We analyzed Ub-protein
conjugates in PC12h cells treated with
nerve growth factor (NGF) and microtubule
poisons (vinblastine, colchicine or both)
to study roles of Ub in neuronal
differentiation and degeneration. Ub and
Ub-protein conjugates in PC12h cells were
detected by immunoblot analysis using
anti-Ub monoclonal antibody. Exposure
(>1h) of the microtubule poisons caused an
increase of high molecular weight (HMW)
(>29kDa) Ub-protein conjugates in NGF-
untreated PC12h cells, but not in NGF-
treated (differentiated) PC12h cells in
which the HMW (>40kDa) conjugate levels
were originally higher than that of the
NGF-untreated cells. Results suggest that
Ub may play a part in degradation of
denaturated microtubule proteins in PC12h
cells, and this proteolysis might be realy
accelerated by neuronal differentiation.
However, the accumulation of the
conjugates observed in the neuro-
degenerative diseases may not be induced
by only microtubule destruction.

DB 178

THE DEVELOPMENT OF HATCHING GLAND CELLS AND
THE PARALLEL DISSOLUTION OF FERTILIZATION
COATS. 1 >

N. Yoshizaki” and H. Yamasaki. Dept. of
Bioles) Gifu Unives iGitu andecDepte. of Bao.
Wakayama Med. Coll., Wakayama.

SEM observations of the development of
hatching gland cells (HGCs), cytochemical
detection of their proteolytic activities
and TEM observations of the changes in fer-
tilization coats (FCs) during the hatching
process were made in Xenopus laevis embryos,
The first identifiable HGCs’ were microvilli
bearing cells in the epidermis of stage 20
embryos. The cells formed a Y-shaped mass
at the antero-dorsal portion of the embryos
at stages 26-35. After hatching occurred,
the mass of HGCs gradually dissociated and
individual cells were finally buried among
common epidermal cells. Proteolytic activ-—-
ity, detected by the silver proteinate
method, appeared exclusively in secretory
granules in the apical region of the HGCs.
This activity was heat-labile and dependent
on a pH range of 6.0-8.5. The FC consists
of a vitelline coat (VC) and a fertiliza-
tion (F) layer. A meshwork of filament
bundles in the VC were dissociated by stage
21 and the filaments themselves were dis-
integrated by stage 31 and totally disap-
peared by stage 34. The F layer also show-
ed dissociation by stage 34. These results
Suggest a view that the hatching enzyme
which digests the FC is produced by HGCs.

DB 179

PURIFICATION AND SPECIFICITY OF HATCHING ENZYME OF
THE SEA URCHIN HEMICENTROTUS PULCHERRIMUS.

K. Nomura!, H. Tanaka2, Y. Kikkawa , M. Yamaguchi2
and N. Suzuki@. lDept. of Biochem., Tokyo
Metropol. Inst. of Gerontol., Tokyo and 2Noto
Marine Lab., Kanazawa Univ., Ishikawa.

The hatching enzyme (HE) of the sea urchin, H.
pulcherrimus was purified to electrophoretical
homogeneity by a few sets of isolation procedures.
The purified HE showed the MW of 36k on SDS-PAGE,
and had the activity to dissolve the fertilization
envelope (FE) of ethanol-fixed embryo in seawater,
as well as the proteolytic activity on casein. HE
was completely inhibited by o-phenanthroline,
EDTA, EGTA and ao-macroglobulin, and slightly by
chymostatin, pepstatin and PMSF, but not by other
inhibitors of serine- and thiol-proteases,
indicating that HE is a metalloproteinase. The
Sites of cleavage, by the action of HE, in the
various bioactive peptides, e.g., oxidized insulin
B chain, substance P and neurotensin, were deter-
mined by isolating the product fragment peptides by
reversed-phase HPLC. The results clearly revealed
that HE preferentially cleaved the peptide bonds on
the NHo-side of hydrophobic residues, -Leu, -Ile,
-Tyr and -Phe, in a similar but more limited manner
than thermolysin. Interestingly, the glycine-to-
sarcoSine substitution of residue 9 in substance P
rendered the analogue very poor substrate of HE,
while substance P itself was a good substrate. This
Suggests that HE belongs or is closely related to a
class of stromelysin, a mammalian matrix metallo-
proteinase, degrading fibronectin, laminin, type IV
collagen and proteoglycans. The detailed mechanism
of FE dissolution is still left to be elucidated.

DB 180

PROTEOLYTIC ENZYMES IN THE HATCHING MEDIUM
OF SALMO GAIRDNERI.

S.. YOKOYA. Div. Cell Seaeye Cenke Recwaetabe:,
Fukushima Med. Col., Fukushima.

In order to study the charactor of the
hatching enzymes by electrophoresis, the
enzyme-activity in the bands of electro-
phoresis was demonstrated by using gelatin
and solubilized egg membrane (chorion) as
substrates.

In the hatching medium, in which ferti-
lized eggs were reared untill hatching out,
about 12 and 8 proteolytic active bands
were detected on the gelatin-substrate
zymograms by IEF electrophoresis (pH 3.8-9)
and SDS-PAGE, respectively. In the chorion
substrate zymography, although the enzyma-
tic bands were less resolutive than gelatin
substrate, two bands which moved to high pH
point by IEF, and three of low pH point
were able to digest the solubilized chorion
substrate. In SDS-PAGE zymography, both
gelatin- and chorion-lytic bands were less
in number than those of IEF-zymograms.

These results suggest that the hatching
enzymes of Salmo gairdneri consist of
several kinds of enzymes and some of them
are able to keep activity against SDS
treatment. Moreover, activity of the
hatching enzymes must be assied by using
native substrate so that hatching medium
may contain other proteolytic enzymes.

Developmental Biology 1127

DB 181

CHORIOLYTIC ACTION OF HCE, A COMPONENT OF
HATCHING ENZYME OF MEDAKA 1

O. Tsuchiya { S.. Yasumasu“, I. Iuchi and
K. Yamagami Life sci. Inst., Sophia
Univ o), sLokiyol; WoO, ilmihes, Waco Sets,

Hiroshima Univ., Hiroshima

It is considered that high choriolytic

enzyme (HCE) binds to chorion when it ex-
erts choriolytic action and that the bind-
ing site is situated separately from the
catalytic site in the molecule. A
monoclonal anti-HCE antibody (E72) inhibits
the choriolytic activity probably by af-
fecting the binding site, while it does not
inhibits the catalytic activity. Binding to
chorion seeins to be prerequisite for
choriolytic action by the catalytic site.
HCE which had been incubated with E72 could
not digest solid chorions, but digested
chorion protein solubilized with alkali.
This result strongly suggests that binding
site is necessary for HCE to digest the
chorions only when HCE acts to solid
chorions. By partial cleavage of HCE with
Drolidase, we obtained two fragments (mw,
ca 10K) which were reactive with E72. When
these fragments were incubated with solid
chorions, they were found to bind to the
Cchorions. The »Dinding was inhibited if the
HCE fraginents had been previously incubated
with E72. These HCE fragments seem to con-
tain binding site. Thus these are 'binding
fragments'.

DB 182
Cloning of cDNA of LCE, the hatching enzyme

of medaka, Oryzias latipes.

S. Yasumasu’', K. Yamada, K. Akasaka!, H.
Shimada and K. Yamagami“~. 1: Zool. Inst.,
Hiroshima Univ. 2: Life Sci. Inst., Shophia
Univ.

The hatching enzyme of medaka consists of
two types of proteases (HCE, LCE). A single
cDNA clone was obtained from the cDNA
library of poly(A)*RNA of the 3rd day
embryo, built up in Agt1i1, by screening
with a polyclonal antibody against LCE. In
an RNA blot analysis of the total RNA from
various stages of prehatched embryos, it
was found that the cloned cDNA hybridized
at 1.3Kb and that the band was first detec-
table in day 2 embryos in advance of the
production of LCE. The cloned cDNA was
1087 nucleotide long and contained a single
long open reading frame of 271 amino acids.
The sequence of N-termial 25 amino acids
analyzed from purified LCE was detected in
the open reading frame of the cloned cDNA,
though the sequence was preceded by a
leader sequence of 71 amino acids. It may
be that this leader amino acid sequence
contains a signal peptides. The amino acid
composition of mature LCE deduced from the
reading frame of this cDNA was quite close
to that of purified LCE. We conclude that
this cloned cDNA encodes the complete
sequence of LCE.

DB 183

ANDROGEN-DEPENDENT DIFFERENTIATION OF THE
FIBROCARTILAGE IN THE PENIS OF THE RAT
R.Murakami, K.Miyake, K.Izumi, I.Yamaoka.
Biol. Inst. Yamaguchi Univ., Yamaguchi.

Androgen is necessary for the develop-
ment of the fibrocartilage in the penis of
the rat (distal segment of the os penis).
Androgens act directly on the cartilage
blastema and cause the cell differenti-
ation of the chondrocytes. The chondro-
cytes produce types I, II collagens,
glycosaminoglycan (GAG) and vimentin
during development, and matrix vesicles,
alkaline phosphatase (ALPase) and
calcified matrix during endochondral
ossification which begins after the
fibrocartilage formation.

We examined the effect of testosterone
on the expression of these cytological
characters by in vitro organ culture of
the cartilage blastema. Type I collagen
and vimentin appeared and increased
rapidly after the start of the culture
whether with or without the testosterone
treatment. Type II collagen, GAG, ALPase,
matrix vesicles and calcified matrix were
produced after 10 days of culture only in
the explants treated with testosterone.

These results suggest that there are at
least two steps in the cell differenti-
ation) kof sche tabrocaiztaslage! Of = the os
penis: the first is independent of
androgen and the second is caused by
androgen.

DB 184

EFFECTS OF STEROID HORMONES ON THE MORPHO-
LOGICAL DIFFERENTIATION OF EPITHELIAL CELLS
IN NEWBORN GOLDEN HAMSTER OVIDUCT.

H.Abe and T.Oikawa, Research Institute for
the Functional Peptides and Developmental &
Reproductive Biology Center, Yamagata.

The effects of steroid hormones on the
differentiation of epithelial cells of the
ampullar region in the newborn golden ham-
ster oviduct were investigated by electron
microscopy. The consecutive injection of
steroid hormones induced various ultra-
structural changes in the undifferentiated
epithelial cells. In particular, the cilio-
genesis and the production of secretory
granules were frequently observed in the
undifferentiated epithelial cells of ovi-
ducts treated with estradiol-17£8 (E 2).
Quantitative studies clearly showed that E92
induced the differentiation of both cili-
ated and secretory cells. On the other
hand, consecutive progesterone (P) and 5a-
dihydrotestosterone (DHT) treatments
stimulated the appearance of ciliogenic and
ciliated cells, but not that of secretory
cells. These results indicate that the
inducing effect on secretory cell differ-
entiation was specific for E2, whereas
ciliated cell differentiation may be close-
ly related to P and DHT as well as EQ. It
is suggested that there are differences in
the hormonal effects on the differentia-
tion between ciliated and secretory cells
in the oviductal epithelium of the newborn
golden hamster.

1128 Developmental Biology

DB 185

INDUCTION OF AUTOLYSIS AND TERMINAL DIF-
FERENTIATION BY THYROID HORMONE IN CELLS
OF THE CHORIOALLANTOIC MEMBRANE OF THE
CHICK

Y.Tahara, K.Tokuda, Y.Hikita, Y.Yoshida
and K.Yamazaki Dept. Biol., Osaka Kyoiku
Univ., Osaka

The chorioallantoic membrane (CAM) of
the chick embryo is composed of three
layers of tissue : the chorionic epitheli-
um on the top, the allantoic epithelium
on the bottom and the fibroblastic mesen-
chyme between them. In contrast with the
cells composing the embryo body, cells of
these tissues do not appear to terminally
differentiate throughout the incubation
period. We administered thyroid hormone
(107-105 M or 10%103g/m1 Or WS eimel Ge)) losy
dripping onto the CAM on day 6 or 8 of
incubation and found that the hormone
induces cell death by autolysis and tissue
differentiation followed by enhancement
of cellular activities.

These responses of the CAM remind us of
the metamorphic changes in the anuran
tadpole induced by thyroid hormone. Based
on this and other findings we hypothesize
that the cells of the chick CAM express
developmental program homologous to those
expressed in the cells of amphibian
larvae.

DB 186

A NEW METHOD FOR IDENTIFICATION OF SKEIN
CELLS, LARVAL SPECIFIC SKIN CELLS OF FROG,
WITH FLUORESCENCE MICROSCPY
A.Kawail, K.Yoshizato2, K.Sugimoto3 and
K.Yamashita

Jo Deyteo Cl? I3H@ILS, Bee, Ce Seisp BWelso
Metropolitan Univ., Tokyo

2 ZOO ein Sie hac OLS Clive umeHascoshama
Univ., Hiroshima

3. Dept. of Anat., Nippon Medical School,
Tokyo

Skein cells are epidermal cells present only
in larvae of anura and have the structure like
a skein called the figure of Eberth. It has
been known that the figure of Eberth is
visible only when subjected to fixation for EM
and staining with toluidine blue. We found
that the figure of Eberth could be recognized
by fixation in Bouin's solution, hematoxylin-
eosin staining and observation its strong
fluorescence with a fluorescence microscope.
Using these methods we observed change of the
figure of Eberth from larvae just after
hatching to adults of Rana catesbeiana. It
was seen that the figure of Eberth appeared on
the 3rd day after hatching (25°C), grew
gradually and disappeared before adult.
Microscopical observations suggested the close
association of basement membranes in the
genesis and breakdown of the figure of Eberth.
Moreover we succeeded in isolating the figure
of Eberth with Percoll density gradient.
Because the fraction which contained figure of
Eberth was stained with antibody to human
cytokeratin and because stratum corneum in the
epidermis of adults had also strong
fluorescence, the possibility that the figure
of Eberth is keratin in its nature seems to be
likely.

DB 187

CELL PROLIFERATION RELATING TO TISSUE
REMODELING IN LARVAL EPIDERMIS OF XENOPUS
LAEVIS.

T. Kinoshita and F. Sasaki. Dept. of
Biol., School of Dent. Med., Tsurumi
Univ., Yokohama.

In larval back skins of Xenopus
laevis, it was observed that
incorporation of BrdU was gradually
restricted to a basal layer of the
epidermis during climax of metamorphosis.
In order to study a role of thyroxine in
the restriction of cell proliferation,
back skins were cultured in medium
supplemented with triiodo-thyronine (T3).
In this system, apical layers of
epidermis peeled from pieces of skins
during preparation of culture, and
remaining epidermal cells called "skein"
and "basal" cells spread out within 2
days at 23°C. In culture medium without
T3, incorporation of BrdU was maintained
at a constant level both in skein and
basal cells. However in the medium
containing T3, BrdU were incorporated
exclusively into basal cells, whose
labeling index gradually increased.
Immunocytochemical analysis of
cytokeratin isoforms showed that basal
cells differentiate into adult type of
epidermal cells under an existence of T3.
These results suggest that T3 stops cell
cycles of skein cells and simultaneously
promotes cell division and
differentiation of basal cells.

DB 188

THYROID HORMONE-INDUCED CHANGES IN EPITHE-
LIAL CELLS OF THE ANURAN SMALL INTESTINE
IN VITRO.

A.Ishizuya-Oka and A.Shimozawa. Dept .of
Anat., Dokkyo Univ., Sch.of Med., Tochigi.

We established an organ culture
system appropriate for the analysis of
metamorphosis of the anuran small intes-
tine. A fragment of the intestine behind
the bile duct junction was isolated from
Xenopus laevis tadpoles of NF stage 57.
The fragment was slit open and incubated
at 26°C on the Millipore filter in
Leibovitz-15 medium (X0.6) supplemented
with 10% charcoal-treated FBS. When 3,3;5-
triiodo-L-thyronine (1078 M), insulin (5pg/

ml), and hydrocortisone (0.5yg/ml) were
added to the medium, larval epithelial
cells degenerated, whereas adult ones

proliferated and differentiated just as
during spontaneous metamorphosis. On the
5th day of cultivation, most of the larval
cells died, and adult cells proliferated
rapidly to form islets. By the 10th day,
a simple columnar epithelium showing the
adult-type binding pattern for SBA and
RCA differentiated. Under this condition,
various other parts of the intestine were
cultured. When a fragment without a
typhlosole was cultured, only a larval
cell death was observed. This suggests
that the connective tissue in the typhlo-
sole is responsible for the formation
of the adult epithelium.

Developmental Biology, Endocrinology 1129

DB 189

MORPHOLOGICAL CHANGES IN THE LARVAL
PANCREAS OF THE BULLFROG, Rana
catesbeiana, DURING METAMORPHOSIS: CELL
SUBSTITUTION FROM LARVAL TO ADULT TYPE.
H. Takahama, Biological Institute,
Faculty of Education, Oita University,
Oita

Morphological changes of the larval
pancreas were studied in the bullfrog,
Rana catesbeiana, during metamorphosis.
The wet-weight and length of the whole
Pancreas show steady increases through
stage XVIII (T-K stage) whereupon they
rapidly decrease to stage XXIII. Acinar
structures, which are major components of
the larval pancreas, develop gradually to
stage X. At stage XIII, small vacuoles
are visible in the acinar cells, and
these gradually increase in size through
stage XVIII. At that point, the vacuoles
occupy a large part of the cytoplasm near
the nucleus of the acinar cells. Finally,
degeneration occurs in the acinar cells
but not in other types of the cells such
as the centro-acinar cells and endocrine
cells. The acinar cells lose zymogen
granules and the acinar structure
disappears at stage XX. The acinar
structure, which contains masses of small
cells, is reconstructed at stage XXI. At
stage XXIII, cell division is observed in
the acini.

DB 190

IMMUNOLOGICAL CHARACTERIZATION OF TADPOLE
COLLAGENASE

K. Oofusal 2 and K. Yoshizato“

Devaney DOs abr ,.eDept.. sot. Bolle jpbace.t of
Sci. Tokyo Metropolitan University, Tokyo
Bo Iw@ilo Ceiliy Seis 1hAlijo 5 “OOo IbNSik6 4, ees

of Sci., Hiroshima University, Hiroshima
Collagenase was detected in tissue
homogenates of back skin of adult bullfrog
(Rana catesbeiana), tail of african claw
frog (Xenopus laevis) tadpole and tail of
adult newt (Cynops pyrrhogaster) by using
anti-tadpole skin collagenase (anti-TSC)
antisera. But, no protein band corre-
sponding to the enzyme was shown in liver,
heart, lung, brain and small intestine of
both adult and tadpoles of bullfrog.
Epidermal cells were isolated from bullfrog
tadpole tails and subjected to the western

blotting. Two protein bands (75kd, 39kd)
were recognized by anti-TSC. The 75kd
band could not be shown in tadpole skin
homogenate and culture media. If the band

corresponds to procollagenase, this is the
first demonstration of the presence of
procollagenase in amphibia.

Anti-TSC also reacted on the purified
human fibroblast collagenase about 45 % in-
tensity as compared with the tadpole ones.
Three protein bands (57k; procollagenase
with sugar chain, 52k; procollagenase, 43k;
active collagenase) were observed by the
western blottings. Anti-human fibroblast
collagenase antisera also reacted on tad-
pole skin collagenase (39kd). These
results clearly demonstrate the immunolog-
ical cross-reactivity of collagenases
between mammals and amphibia.

EN 1

COMPARATIVE STUDY OF DIVISION OF LABOR AND
JUVENILE HORMONE (JH) IN HAEMOLYMPH, ON
WORKER IN JAPANESE HONEYBEE (Apis cerana
japonica L.) AND EUROPEAN HONEYBEE (Apis
mellifera L.).

H.Sasagawa!, Y.Kuwahara!, M.Sasaki2 and
T.Kusano3. !Inst. of Appl. Biochem., 3Inst.
of Agr. & Forest., Univ. of Tsukuba, Tsuku-

ba, and 2Fac. of Agr., Tamagawa Univ.,
Tokyo.
In the European honeybee(AM), division

of labor is regulated by increase of haemo-
lymph JH along with age. JH and methoprene
treatment exhibited various behavioral and
physiological changes among workers. Al-
though the Japanese honeybee(ACJ) belongs
to the same genus to AM and share the same
habitat with AM, very little is known ‘about
JH and its function, because of its hard-
ness of rearing. Trial analysis of the
ACJ’s JH extract by micro-HPLC gave the
same retention time of those of standard
JH-III, therefore, haemolymphal JH was
determined on individual ACJ workers,
engaging with 8 kinds of labor, and com-
pared with those of AM. JH quantity in AM
increased in the following order, cleaner <
queen rearing < brood rearing < nectar
forager < pollen forager < and guard, while
in ACJ, day-O worker < larval rearing <
queen rearing < bees attacking Vespa sp. <
guard < raging bee on disturbance < nectar
forager < and pollen forager. As a result,
a parallel relationship was observed be-
tween two species of labors.

EN 2

IN VITRO SECRETION OF ECDYSTEROIDS BY Y-
ORGANS OF THE CRAYFISH, PROCAMBARUS CLARKI.

Hie Sonobel, M. Kanbal, K. Ohtal, M. Ikeda2
and Y. Naya2. 4Dept. of Biol., Fac. of
Sci., Konan Univ., Kobe, 7SUNBOR, OSAKA.

The purpose of the study is tc character-
ize ecdysteroids secreted by Y-organs of P.
clarki. The Y-organs, which have been
activated by excising the eyestalks of the
donors, were cultured in sterile medium.
The ecdysteroids were extracted from the
culture medium and subjected to reverse-
phase HPLC. Each fraction from the column
was analyzed by radioimmunoassay (RIA). Two
different antisera were used for RIA: H-22
with high affinity for ecdysteroids with a
modified side-chain, and S-3 with high
affinity for a modified A-ring.

Our present experiments demonstrated that
the Y-organs of P. clarki secrete primarily
an ecdysteroid that was less polar than
ecdysone together with a small amount of
ecdysone. This ecdysteroid was identified

as 3-dehydroecdysone on the basis of
chromatographic profile, immunological
cross-reactivity and mass spectrometry. We

also found in the eyestalk-deprived animals
that a major ecdysteroid in the hemolymph
was 20-hydroxyecdysone (20-HOE , and the
amount of 20-HOE in the hemolymph increased
in parallel with the elevation of the
secretory activity in the Y-organs. These
results suggest that the 3-dehydroecdysone
secreted by the Y-crgans may be converted
in vivo into 20-HOE via ecdysone.

1130 Endocrinology

EN 3

ANALYSIS OF ECDYSTEROIDS IN OVARIES OF THE
SILKWORM, BOMBYX MORT.

M. Kanba and H. Sonobe. Dept. of Biol.,
Fac. of Sci., Konan Univ., Kobe.

In order to analyze ovarian ecdysteroids
of Bombyx silkworm, two different antisera
were used for radioimmunoassay (RIA): S-3
that detects the difference in the side-
chain of the ecdysteroids, and H-22 that
recognizes the difference in the A-ring of
the ecdysteroids. From analysis by a
combination of reverse-phase HPLC and RIA,
it was demonstrated that a large amount of
unknown ecdysteroid (UKE) elutes between 2-
deoxy-20-hydroxyecdysone and 2-deoxy-
ecdysone in addition to some ecdysteroids
which have been already identified in the
silkworm ovaries. The UKE cross-reacted
Mainly with H-22, but not with S-3. This
result suggests that the UKE is modified in
the side-chain rather than the nuclear
portion of the molecule. It was also
demonstrated by enzymatic hydrolysis and
HPLC-RIA system that the UKE accumulates
largely in the ovaries as a phospholic
ester in addition to the free form. Next,
we tested whether [4-14c]cholesterol which
has been injected into developing pupae is
transformed into the UKE. Contorary to our
expectation, the UKE does not incorporate

4c so much as other ecdysteroids in the
ovaries. This fact suggests that the UKE is
in part derived from phytoecdysteroids in
Mulberry leaves or dietary phytosterols
without conversion to cholesterol.

EN 4

ECDYSTEROID PRODUCTION IN THE OVARY OF
5TH INSTAR BOMBYX LARVAE
O. Parlak, S. Sakurai and T. Ohtaki.
DANE OQ WiOls, Pacuiliesyy Oi Sis ;
Kanazawa Univ., Kanazawa

In 5th instar larva of Bombyx mori,
Ovaries grow gradually until gut purge.
Ecdysteroid content in the ovary was
estimated by radioimmunoassay as 3 to 4
pg/pair ovary on day 0, 1 and 2. At day
3, amount of ecdysteroid in ovary
increased to more than 20 pg/pair, and
then returned to the original level on
day 4, and remained low until day 6.
Ecdysteroid production in isolated
ovaries cultured in Grace's medium
changes in the same manner as observed
in intact ovaries. PTTH or brain
homogenates did not stimulate
ecdysteroid production of cultured
Ovaries dissected from day 2 larvae.
Application sot WH) to) Sth anisitar Varvale
inhibits increased production of
ecdysteroids in the ovaries of day 3
larvae. Hemolymph ecdysteroid titer was
very low until day 3, before the
prothoracic glands became active.
Therefore, the larval ovaries produced
ecdysteroid without stimulation of 20-OH
ecdysone or PTTH, being different from
the pupal ovaries.

EN 5

ISOLATION AND PURIFICATION OF ECDYSONE
RECEPTOR PROTEIN FROM THE FAT BODY OF

BOMBYX LARVAE

Le Tsudats, ¢Si.eSalkciun acl. uke. Yamaguchi and
Tf. Ohtaki li, | Dept. Biol, sEac asc
Kanazawa Univ., 2Inst. Gene Res., Kanazawa
Univ., Kanazawa.

Radioreceptor assay using 3H-
Ponasterone A revealed that the fat body
cells of last instar Bombyx larvae
contained a certain amount of ecdysteroid
receptor. We tried to isolate receptor
protein from the fat body cells as the
original materials. Binding activity was
found in precipitant of cytosol by 502
(NH,)2S0O, saturation. Then the precipi-
tant was applied onto DEAE Sephacell
column and the binding activity was
considerably concentrated by eluting with
O.2M KCl. Fraction with high binding
activity was further applied to 20-OH-
ecdysone conjugated affinity column.
Finally, purification fold came up to one
million times by the steps mentioned
above. Using synthetic oligo nucleotide
of a part of hsp27 and Sgs-4 gene, which
encodes the DNA binding domain and is
thought as common to that of ecdysone
receptor gene, gel shift assay of
semipurified samples was performed. The
results suggested that the samples
contained factors which bind specific to
this sequence. The relationship between
these factors and ecdysone receptor
remains to be studied.

EN 6

THE BRAIN REGULATES THE RELEASE OF DIAPAUSE HORMONE FROM
THE SUBOESOPHAGEAL GANGLION PRIMARILY THROUGH GABAERGIC
NEURONS IN THE SILKWORM, BOMBYX MORI.

K. Hasegawa and I. Shimizu. Res. Sect. of Environ.
Physiol., lab. for Plant Ecol. Studies, Fac. of Sci.,
Kyoto Univ., Kyoto.

Brain-suboesophageal ganglion complexes (Br-SGs) and

suboesophageal ganglia (SGs) were incubated in vitro and
diapause hormone (DH) released was determined by bioassay,
Br-SGs of non-diapause-egg producing pupae (ND-pupae) se-
creted little DH, whereas Br-SGs of diapause-egg produc-
ing pupae (D-pupae) secreted significantly more DH. SGs
of ND- and D-pupae released considerable amount of DH.
The release of DH from Br-SGs of D-pupae and SGs were
inhibited by y-aminobutyric acid (GABA), an inhibitory
neurotransmitter, in a dose-dependent manner. Picrotoxin,
a specific blocker for GABA, caused the release of DH
from Br-SGs of ND-pupae. The distribution of GABA-like
immunoreactivity was examined by staining the serial sec-
tion of the pupal Br-SG with antisera against GABA.
Immunoreactive somata were mainly found in the lateral
portion of protocerebrum and the ventral side of deuto-
cerebrum, tritocerebrum, and SG. Numerous cut ends of
axons stained in the cross-section of circum-oesophageal
Connectives suggested the presence of GABAergic neurons
extending axons from the brain (Br) to SG. These lines

of evidence suggested that Br regulated the release of DH
from SG primarily through GABAergic neurons.

Endocrinology 1131

EN 7

ARCHITECTURE OF NEUROSECRETORY CELLS’ IN
THE BRAIN OF SILKWORM, Bombyx mori.

T. Ichikawa. Dept. of Biol., Fac. of Sci.
Kyushu Univ. Fukuoka 812.

Cerebral neurosecretory cells (NSCs) in
the silkworm have their somata at the
medial or dorso-lateral parts of the
brain. Intracellular injection of Lucifer
Yellow into the NSCs revealed at least six
types of medial cells with specific
dendritic arborizations in the brain. Five
of them projected axons in the contra-
lateral corpus allatum (CA), or the corpus
cardiacum-recurrent nerve (CC-RN) complex.
Remaining one located ventrally projected
an axon into the ipsilateral proctodeal
nerve via the ventral nerve cord. This
descending NSC was’ immuno-histochemically
reactive to the monoclonal antibody
against Bombyx eclosion hormone. Three
morphological types of NSCs were distin-
guished in the dorso-lateral group of
cells. They projected axons in the ipsi-
lateral CC-RN complex or CA or _ contra-
lateral CA. Two pairs of paraldehyde-
fuchsin positive cells (type M2) in the
medial part of the brain had been
considered as NSCs (Bassurmanova & Panov,
1967). However, electrical stimulation of
their somata showed an action potential
with a short duration (2-3 ms) rather than
a typical NSC's action potential with a
long duration (30-50 ms). They send axons
to the contralateral ventral nerve cord.

EN 8

PHOTOPERIODIC DETERMINATION OF THE DEVE-
LOPMENT OF SEASONAL MORPHS AND ADULT RE-
PRODUCTIVE DIAPUSE IN THE ASIAN COMMA
BUTTERFLY, POLYGONIA C-AUREUM L.

K.Endo, S.Ueno and Y.Kakuo. Environ.
Bi@il, als, Pal@lls limes, WAC ysSE@als 4 Gin
guchi Univ., Yamaguchi.

Autumn morph adults of the Asian comma
butterfly, Polygonia c-aureum, developed
under short day conditions at 20°C, en-
tered reproductive diapuse after eclosion.
Large proportions of autumn morph adults,
developed under short days at 25°C, were
judged as being reproductively active,
since they matured eggs within 4 days
after eclosion.

For induction of summer morph develop-
ment and prevention of reproductive dia-
pause in 50 % adults, they required 4 and
5 long-day cycles, respectively. The lar-
vae became photoperiodically sensitive on
the first day of the 2nd instar and the
photoperiodic sensitivity disappeared 4 to
5 days after larval-pupal ecdysis.

The results indicate that P. c-aureum
accumulates photoperiodic information to
determine reproductive activity from the
2nd-instar larval to the pupal stages.
But, larvae seemed to determine seasonal
morph development on the basis of photo-
periodic informations accumulated before
larval-pupal ecdysis which was followed
by the secretion of summer-morph-producing
hormone (36 hr after larval-pupal ecdysis
aie 2O-G)e

EN 9

ON THE TESTIS AND ANDROGENIC GLAND DURING
DEVELOPMENT OF EXTERNAL SEXUAL CHARACTERS

OF THE CRAYFISH, PROCAMBARUS CLARKIT,

-Nishikawa , Y.Taketomi and S.Koga
Dept. of Biol.,,Fac. of Sci., Kumamoto
Univ., Kumamoto. Karatsu Commercial High

School, Karatsu City, Saga.

The external sexual characters of
male crayfish, Procambarus clarkii,
include reversed spines on the ischia of
the third and fourth walking legs, and the
enlarged lst and 2nd copulatory abdominal
appendages. These characters play
important roles in the mating process.

We followed the histological changes
of the testes and androgenic glands during
the development of the abdominal append-
ages and also noted whether the _ sexual
spines were present. Crayfish having a
carapace length (C.L.) less than 25mm
lacked these spines. Almost all, of the
males of C.L. 30mm or longer had these
spines. The development of the abdominal
appendages could be divided into five
Stages, A-E. St.E represents full develop-
ment. All St.E males had these _ spines.
Testes of crayfish with St.A appendages
are immature. Many spermatocytes appear
when the appendages are in stage B.
Spermatogenesis begins at St.C. Iie ae
difficult to distinguish the androgenic
gland when the appendages were in St.A. At
St.B the gland was small, in the shape of
an extended string. These results suggest
that the function of the androgenic gland
iil Seo!) alg) GlakiseSie@imie sie@n) (Elie aly Sie olive

EN 10

CONTROL OF THE TIMING OF EGG-HATCHING IN
THE TERRESTRIAL CRAB, SESARMA.

M. Saigusa. College of Liberal Arts and
Sciences, Okayama University, Okayama.

Egg-hatching of the terrestrial crab
Sesarma haematocheir occurs just before
releasing larvae; the timing which is
actually controlled is that of highly
synchronized egg-hatching occurring in the
female's abdomen. The subject of this study
was how the timing of egg-hatching was
controlled by the female or the embryos.

Female crabs have 8 pairs of plumouse
and non-plumouse setae in the abdomen. Eggs
are attached to the non-plumouse hairs just
like the bunch of grapes, and are incubated
by the female until hatching. One bunch of
eggs was detached from two females whose
eggs were estimated to hatch within several
days. These bunches were exchanged to each
other, and then transplanted between these
females using a thin thread. It was exami-
ned with which female is the hatching of
transplanted egg bunch synchronized. The
results have obviously showed that the
hatching occurred synchronized with the
eggs of the 'nurse' female, not with those
of the 'mother' female. This indicates that
the timing of egg-hatching is controlled by
the female, not by the embryos themselves.

132 Endocrinology

EN 11
MOLTING CONTROL IN THE SHRIMP, CARIDINA

DENTICULATA 2
K.Miura and Y.Taketomi
Ogawa Junior High chool, Ogawa-machi,

Uto Galiey7n Kumamoto. De\pit. Ose Ba-opleey,
Fac. of Sci., Kumamoto Univ., Kumamoto.

The stages of the molt cycle of
Caridina denticulata were determined by
observations Oem t eke) Snpardarty, and
setogenesis of the second antennal _ scale.
This species repeats its molt cycle every
12 to 13 days at 21°C. St.A has an average
Glin Aeon Ot) Ibo) days SecGo Aoshi dayss “Sie.
Do» 2.5) GCAVSS St.D,; 6.0Pdayse Sit. Do, 0.9
day. che” mokt cyeltesas Sdiecdysitc, = wath
short intermolt and long premolt periods.
Crustacean eyestalks have long been known
to be a source of molt-inhibiting hormone.
Bilateral ablation of the eyestalks of the
shrimp was done during St.C, D, and D of
the molt cycle. Control shrimp had “only
one eyestalk removed.

The duration of the molt cycle of
bilateral eyestalkless shrimp was similar
to that of the control shrimp regardless
of what stage in the molt cycle the
eyestalks were ablated. But after the next
noe, alnost Jalil of ithe eyestalkless
shrimp remained in St.C or St.D until
they died. The eyestalkless shrimp
remained in St.C were kept in solution
with 20-hydroxy-ecdysone. They molted
successfully after about ten days.

These results suggest the eyestalks
of these shrimp release a factor that
activates Y organs.

EN

STUDIES ON THE NEUROPEPTIDE-CONTAINING
CELLS OF A FRESHWATER OLIGOCHAETE, TUBIFEX
HATTAT WITH SPECIAL REFERENCE TO
MAINTENANCE OF THE BLOOD GLUCOSE LEVEL.
H.Jaana. Biol. Lab., Asahikawa campus,
Hokkaido Univ. of Education, Asahikawa.

A blood glucose level of T. hattai
was measured by HPLC. The measurement was
performed in the worms which were
decerebrated or were not fed during the
experiment. As” av aresuites it was
ascertained that in both experimental
groups the level of blood glucose was
being maintained in a similar fashion with
that of control animals at least for two
weeks, by the end of which a mouth had
regenerated in the decerebrated animal;
however, by comparison with control
animals, increased blood glucose in
decerebrated animals was observed for the
first three days after the operation. On
the other hand, an immunohistochemical
technique revealed the existence of cells
containing glucagon-like substance(G-
cells) in ventral ganglia of the worm. It
was shown by electron microscopy that the
G-cells of decerebrated animals actively
synthesized and discharged their secretory
granules especially for the first three
days after the operation. Therefore, it
Was suggested that the blood glucose level
increased by secretion of glucagon-like
substance from G-cells of ventral ganglia.

EN 13

HEPARIN-BINDING GROWTH FACTORS FROM HYDRA
K.Hanai , K.Nishikawa and T.Fujisawa’.
Dept.Bio}., Fac. of Sei., Kyushw Univ,
Fukuoka, Dept. of Biochem., Kanazawa Med.
Univ., Uchinada and “Dept. of Develop.
Biol., Natl. Inst. of Genetics, Mishima.

Tentacle-ball formation of Hydra was
induced by S-methylglutathione and de-
pressed by a small amount of growth factors
and other biologically active substances.
Activities similar to heparin-binding
growth factors elevated in the cerebrospi-
nal fluids from rat brain after feeding.
Similar activities were also elevated in
the medium after feeding of Hydra. When
Hydra was excised near the head region, the
animal released a similar activity into the
medium. Heparin-binding proteins of Hydra
were examined to explore these activities.

Freeze-dried polyps were homogenized in
tris.HCl buffer (pH 7.2) with a Waring
blender. The heparin-binding protein-like
depressing activities were extracted from
the precipitate of the homogenate with 0.1

M acetic acid. The extract was applied to
a heparin-Sepharose column after neutral-
ization. The column was eluted with 0.15

M, 0.6 M, 1.0 M, and 2.0 M NaCl. Mitogenic
activities on Balb/c 3T3 cells were detect-
ed in the eluates of 0.6 M, 1.0 M and 2.0 M
NaCl. The 1.0 M and 2.0 M NaCl eluates
were applied to a 5C4-300 column of re-
verse-phase HPLC. The activities were also
detected for HPLC purified proteins with
some inactivation.

EN 14

ACTIVATION AND INACTIVATION OF

MEMBRANE BOUND GUANYLATE CYCLASE AND ITS
ROLE IN SIGNAL TRANSDUCTION OF CRUSTACEAN
HYPERGLYCEMIC HORMONE OF PROCAMBARUS
CLARKII.

M. Ishida, T. Ezure and T. Ohoka. Dept. of
Biol., Fac. of Sci., TOKYO Metropol. Univ.,
Tokyo.

In vitro administration of sinus
gland extract or purified crustacean
hyperglycemic hormone ( CHH ) to isolated
Procambarus heart resulted a significant
increase of intracellular cyclic GMP level,
Purification of Procambarus CHH by HPLC
revealed to be a peptide having a molecular
weight of 8,700 and composed of 73 amino
acid residues.

The effect of CHH was investigated using
particulate fraction pf Procambarus heart.
Binding studies by I-labeled CHH
suggested presence of CHH specific 3
receptor, Kd value being about 4.0 x 10 “M.

Membrane bound guanylate cyclase activity
was stimulated about 3.5_fold in the pre-
sence of enough CHH (10-/M). Effect of CHH
concentration on the activation was co-
ordinated to 1:1 binding curve, and Ka
value for CHH to be 1.0 x 10 “M.

The hormone dependency of the enzyme was
rapidly lost only by incubating at 30°C
for 1 min. Inactivation of hormone
dependency cannot be protected nor restored
by addition of proteinase inhibitors, Ca vi
- Calmodulin, Gpp(NH)p or creatine kinase.

Endocrinology 1133

EN 15 EN 17

THE PRIMARY STRUCTURE OF THE RAT GUANYLYL MOLECULAR CLONINGS OF COMPLEMENTARY

CYCLASE A/ATRIAL NATRIURETIC PEPTIDE DEOXYRIBONUCLEIC ACIDS FOR THE QUAIL

RECEPTOR GENE. LUTEINIZING HORMONE a - AND £8 -SUBUNITS

M.Yamaguchi, L.J.Rutledge and D.L.Garbers. He yAndovand Si) dshit Dept. of Biol) Sche

Howard Hughes Medical Institute, of Edu., Waseda Univ., Tokyo.

Vanderbilt Univ., Nashville, TN, U.S.A. i ee a CALC RE OGa ae A a ae
We have isolated and characterized A quail adenohypophyseal cDNA library

three genomic clones and a genomic frag- was constructed with the A gtll expression

vector. After screening with chicken
pituitary glycoprotein a or LH8& cDNA as
a probe, clone pQA312 and pQL119, which
encoded the entire region of quail a and
LHS precursor molecules, respectively,
were obtained. The quail a@ precursor
molecule deduced from the nucleotide
sequence consisted of 24 residues of the
signal peptide and 96 residues of the
apoprotein. The deduced quail LHS precur-
sor molecule consisted of 47 residues of

ment amplified by PCR, that contain the
rat guanylyl cyclase A/atrial natriuretic
peptide receptor gene. The gene spans
about 17.5kb and includes 22 exons and 21
introns. All the exon-intron junction se-
quences coincide with the GT/AG consensus.
Guanylyl cyclase A consists of at least
four domains: extracellular ligand bind-
ing, transmembrane, kinase-like, and cy-
clase catalytic. Exon 7 encodes the trans-

membrane domain. The kinase-like and cy- the signal peptide and 119 residues of the
clase catalytic domains are encoded by apoprotein. There were two repetitive
exons 8-15 and 16-22, respectively. The 5' sequences (72 and 30 bases long) in the 3'
end of the transcript, estimated by primer untranslated region of the quail LH8

extension and S1 mapping, is 370 nucleo- cDNA. The amino acid sequence homology
tide upstream of the methionine initiation between quail and mammalian a apoprotein
codon. The initiator sequence (-3 to +5) ranged 85-69%, being higher than _ that
has two mismatches when compared to a between quail and mammalian LHB apo-
consensus initiator sequence of CTCANTCT. proteins (48-42%). Homologies of the a

The 5' flanking region contains three po- apoprotein between quail and fish (71-66%)
tential Sp1 binding sites and an inverted and fish and mammal (74-59%) were higher
CCAAT box, but no apparent TATA box. Three than those of the LHS apoprotein between
different short interspersed repetitive quail and fish (47-45%) and fish and
sequences are found within intervening se- Beek eee be Bees Se eae Soe
quences and within the 5' and 3' flanking ew & CONSE TNC

regions of the gene. They fall between the while the LH& is diversified during the

four domains suggestive that these may be evolution of vertebrate gonadotropin.

sites for frequent recombination.

EN 16 EN 18

CLONING AND SEQUENCE ANALYSES OF cDNAS OF GENE STRUCTURE AND EXPRESSION OF PORCINE GLYCOPROTEIN
VASOTOCIN AND ISOTOCIN PRECURSORS FROM HORMONE a-SUBUNIT.

MASU SALMON Y.Kato, T.Ezashi and T. Kato

M. Suzuki, S. Hyodo, and A. Urano.

Ocean Res. Inst., Univ. of Tokyo, Tokyo. Hormone Assay Center, Institute of Endocrinology,

Gunma University, Maebashi.

We determined the nucleotide sequences
of cDNAs encoding the precursors of Pituitary glycoprotein hormones are synthesized in two
neurohypophysial hormones, vasotocin (VT) different type of cell under the control of several extra-
and isotocin (IT), and deduced their
primary structures in the hypothalamus of
the masu salmon Oncorhynchus masou. They

cellular signals. Genomic sequence of the 5'-untranslated
region of glycoprotein common a-subunit revealed the dis-

are homologous to the chum salmon VT-I similarity with the corresponding region of LH@ and FSHB
and IT-I ones, respectively. Each of genes and the presence of unique several segments homologous
them is composed of a signal peptide, to the consensus regulatory elements.

hormone, and unique neurophysin (NP), To elucidate the regulatory mechanism of common a-subunit
which has the highly conserved middle gene expression, primary porcine pituitary cells were
POPE De Zoe WE eine EE ONES ee Je treated with GnRH, phorbol ester (TPA) and forskolin. Common
amino acids longer than the mammalian NPs ‘ : : ‘

at the carboxyl terminus. The extended a-subunit mRNA has increased by those stimuli suggesting
regions have a leucin-rich core segment that two membrane signals are integrated to the common a-
as copeptin in vasopressin precursors. subunit gene. When chimeric gene containing 5'-untranslated
The Southern blot analysis showed the region of common a-subunit gene is transfected into the

presence of two distinct VT genes (VT-I
and VI-II) in masu salmon as well as in
chum salmon, although the VT-II gene may

primary culture and expressed with the stimuli, forskolin
and TPA dependent increase of reportor molecule was observed

be scarecely expressed. A SRA eee whereas GnRH increased slightly. The data suggest that
analysis of evolutional distance between common a-subunit gene is regulated by two different signal
the masu salmon VT and IT precursors transduction pathways and so that the CRE and AP-1 homolo-
suggests an occurrence of gene conversion gous sequence, present in the 5'-flanking region of porcine
between the highly conserved portions in a-subunit gene, respond to those signals.

the middle of NPs. Between masu salmon
and chum salmon, the phylogenical dis-
tance of VT precursors is about 10-fold
larger than that of IT ones.

1134 Endocrinology

EN 19

HORMONAL INDUCTION OF TESTICULAR FSH RECEPTORS:
PITUITARY AND GONADAL HORMONE-DEPENDENT AND
-INDEPENDENT INDUCTION

K. Tsutsui

Zool. Inst., Fac. of Sci., Hiroshima Univ., Hiroshima

The number of FSH receptors increases during testicular
development in mammals. The present study was aimed at
identifying factors inducing developmental changes in
testicular FSH receptors. Male Wistar rats were
hypophysectomized at 9, 16, 23, 30 days or 3 months of age.
These rats were killed 10 days after surgery along with intact
control rats. A group of intact rats were sacrificed on the day
of surgery and designated initial controls. All the
hypophysectomized immature rats showed a higher density of
FSH binding(FSH binding per unit weightJand a lower total
FSH binding (FSH binding per two testes) when compared with
each of the matched intact control rats. Unlike
hypophysectomy at other immature ages, surgery at 9 days of
age was followed by significant increases in testicular weight,
density of FSH binding and total FSH binding when compared
with the initial control rats.When male rats hypophysectomized
at 25 days of age were injected with FSH for 5 days beginning
on the eleventh day after surgery, dose-dependent increases
in the total FSH binding and testicular weight were observed.
Testosterone treatment induced an increase only in the total
FSH binding. Changes in FSH binding were due to changes in
the number of binding sites. Plasma FSH concentrations in all
the post-operative rats were below or at the level of
detectability. It is considered that FSH and testosterone act as
hormonal factors to induce FSH receptors in the developing
testis. Other factors that are independent of pituitary and
sex hormones may also contribute to FSH receptor induction.

EN 20

COMPARISON OF SERUM FOLLICLE-STIMULATING
HORMONE(FSH) AND TESTICULAR FSH RECEPTOR
LEVELS AMONG SOME STRAINS OF MICE.

K. Shiraishi, M. K. Park and S. Kawashima
Zoological Institute, Faculty of Science,
University of Tokyo, Tokyo.

Tsutsui et al.(1985) found the presence of
down-regulation of FSH receptors in normal
conditions in the testis of a mouse
strain(C57BL/6NCrj). Whether the down-
regulation of FSH receptors is unique to
this strain or general among various
strains of mice is not yet decided. As the
first step for this goal, we have used
three strains of mice, C57BL/6NJcl,
BALB/c6NJcl and DBA/2NCrj, and the serum
FSH level and testicular FSH binding during
sexual maturation were studied. Serum FSH
concentration showed different age-related
changes specific to each strains. In
C57BL/6NJcl males, the pattern was similar
to that in C57BL/6NCrj. BALB/c males showed
a peak of FSH on day 31, followed by a
gradual decrease. Changes in the serum FSH
levels in DBA/2N males were quite different
from the other two strains and the levels
were markedly higher. However, on day 70,
the FSH level was not significantly differ-
ent among the three strains. On the other
hand, specific binding of FSH to the tes-
ticular preparations was maximal around day
20 in all strains. Comparison of FSH bind-
ing to the testicular preparations among
the three strains of hypophysectomized mice
is underway, in order to clarify whether
down-regulation occurs generally in mice.

EN 21

LH RECEPTORS IN OVARIAN FOLLICLES
ARTIFICIALLY DEVELOPED BY GONADOTROPIN
ADMINISTRATION IN THE JAPANESE QUAIL
M.Kikuchil, S.Wakabayashi2 and S.Ishiil.
lpept. of Biol., Sch. of Edu., Waseda
Univ., Tokyo and 2Dept. of Biol., *Schesonr
Dent., Nihon Univ., Tokyo.

Six-week-old female quail were
treated with chicken pituitary glyco-
protein by ALZET Osmotic pump (12.5y g/
hr) for 2 weeks or by the pump followed by
5 daily injections of the glycoprotein
(0.5 or 1.5mg/inj). Among 7 females
treated, three laid 8 eggs and 2 chicks
hatched out. The treated ovaries contained
0-19 yellow follicles while normal ovaries
contain 4 or 5 yellow follicles. In the
treated ovaries, the proportion of the
number of F3-size follicles was abnormally
increased, and atretic follicles and
superovulation were observed in abnormal
frequencies. In normal adult females, LH
receptors in theca abruptly decrease to
1/5 when the follicles reached the Fl
stage. In the treated ovaries, LH bindings
to F2- and F3-size follicles were about a
half of the normal level and the decrease
at the Fl stage was not so marked. There
were smaller changes depended on the
follicular size in LH bindings. These
results suggest that LH receptors were not
normally regulated in the hormone-treated
females, although some of the females
ovulated normally and laid normal
fertilized eggs.

EN 22

DYNAMICS OF GROWTH HORMONE RECEPTORS DURING
SEAWATER ADAPTATION IN RAINBOW TROUT.
T.Sakamoto, S.Hyodo and T.Hirano. Ocean
Res. Inst., Univ. of Tokyo, Tokyo.

To determine target organs of the
seawater-adapting actions of growth hormone
(GH) in salmonids, rainbow trout, weighing
about 100 g, was transferred to 80%
seawater, and the binding of radioiodinated
chum salmon GH to membrane preparations of
liver, gill and kidney was measured. The
transfer to seawater evoked a significant
rise in plasma concentrations of GH,
chloride, cortisol, thyroxine and
triiodothyronine. GH binding to the liver
membranes decreased significantly 1 day
after the transfer. Binding in the gill
and kidney was not altered significantly.
MgCl5o treatment was used to remove bound GH
from the receptors. The reduction in the
binding in the liver was suggested be due
to receptor occupancy by increased
endogenous GH. The occupancy of liver GH
binding sites was maximal 4 days after the
transfer. Total (MgClo-treated) binding
sites in liver increased significantly 14
days after the transfer. Scatchard
analysis indicated that receptors altered
in capacity without changes in the binding
affinity. Although these results do not
exclude the possible importance of the GH
receptors in osmoregulatory organs, liver
seems to mediate, at least in part, GH's
seawater-adapting actions in the rainbow
trout.

Endocrinology 1135

EN 23

EFFECT OF GROWTH HORMONE (GH) INJECTION
ON HEPATIC GH RECEPTORS IN THE EEL.
I. Mori, T. Sakamoto and T. Hirano.
Ocean Res. Inst., Univ. of Tokyo, Tokyo.

To examine the control mechanisms of
hepatic growth hormone (CH) receptors in
the eel, we measured total (MgCl5-treated
membranes) and free (untreated membranes)
GH binding sites. Hypophysectomy
decreased both total and free binding
sites significantly 1 week after the
operation. Scatchard analysis indicated
that reduction in GH receptors seems to
be mainly due to a decrease in binding
capacity rather than to the change in
binding affinity. When recombinant eel
GH was injected intramuscularly into the
hypophysectomized eel (2ug/g BW), plasma
GH concentration increased to the maximal
value after 10 h and decreased to the
initial level after 72 h. Free binding
sites decreased to the minimal value 24 h
after the injection, and returned to the
initial level after 72 h. The reduction
in free binding sites seems to be due to
the occupation of binding sites. Total
binding sites increased gradually to
almost 2-fold over those of saline-
injected controls after 5 days, and this
increase was due to the increase in
binding capacity. These results indicate
an important role of GH in inducing and
maintaining its own receptors in the eel.

EN 24

ISOLATION AND CHARACTERIZATION OF GROWTH
HORMONE FROM FLOUNDE

S.Sakata , H.Kawauchi’ and T.Hirano®
Res. and Dev. Dept., Nippon Oil
Company,LTD., Tokyo, “School of Fish.
Sci., Kitasato Univ., Iwate, and ‘Ocean
Res. Inst., Univ. of Tokyo, Tokyo.

A highly purified growth hormone (GH)
was isolated from the pituitary of
flounder (Paralichthys olivaceus) by
extraction with alkaline solution, gel
filtration on Sephadex G-75, and reversed
phase HPLC on an ODS column. Flounder GH
(f1GH) had a molecular weight of 20 kDa in
SDS-PAGE, and an isoelectric point of 7.1
by gel electrofocusing. Amino acid
composition was consistent with other fish
GH. The f1GH was reduced, carboxy-
methylated, and subsequently cleaved with
endopeptidase Asp-N, lysyl endopeptidase
and cyanogen bromide. The intact protein
was also ceaved with o-iodosobenzoic acid.
The resulting fragments were separated
with reversed-phase HPLC and subjected to
sequence analysis by automated gas phase
sequencer. The N-terminal amino acid was
determined using methylamine conversion of
Pyro-glutamate to glutamate. The f1GH is
composed of 173 amino acid residues with
two disulfide lincages and pyro-glutamate
as N-terminal amino acid. The f1GH was a
potent stimulator of growth in juvenile
rainbow trout which received intraperito-
neal injections of the hormone.

EN 25

GROWTH HORMONE LEVELS IN THE PLASMA AND
GROWTH HORMONE-RELEASING ACTIVITY IN THE
HYPOTHALAMUS OF METAMORPHOSING BULLFROG
TADPOLES

M. Nagano!, T.Kobayashi? and S. Kikuyama!.

lDept. of Biol., Sch. of Educ., Waseda
Univ., Tokyo. 2Dept. of Regulat. Biol.,
Fac. of Sci., Saitama Univ., Urawa.

Plasma growth hormone (GH) concentrations
in the bullfrog (Rana catesbeiana)
tadpoles and adults were determined by a
homologous radioimmunoassay. Plasma GH
increased as metamorphosis progressed.
Average concentration was 10 ng/ml at
stage 18 and 47 ng/ml at the end of
metamorphosis (st.25). After metamorphosis
plasma GH levels remained high and
declined as the animals gain their body
weight.

GH-releasing activity of extracts of
hypothalami from tadpoles at various
developmental stages and fully grown
adults was determined in vitro, using the
pituitary from adult bullfrogs. The
extract of tadpole hypothalami exhibited
more potent activity than that of adult
hypothalami. There was no appreciable
difference among the extracts of
hypothalami obtained from tadpoles of
various developmental stages.

EN 26

ADAPTATION OF HUMAN GH-IRMA TO MEASUREMENT OF SERUM
CONCENTRATIONS OF CYNOMOLGUS MONKEY HORMONES
T.Yoshida!, K.Ohtoh* and F.Cho!. !Tsukuba Primate
Center for Medical Science, N.I.H. and “The
Corporation for Production and Research of
Laboratory Primates, Ibaraki.

A newly developed immunoradiometric assay
(IRMA) method for human growth hormone (GH) was
adopted to measure cynomolgus monkey (Macaca
fascicularis) hormones. An extract of monkey
pituitary tissue showed parallelism with the
reference preparation (human GH). Age-related
changes in serum GH concentrations of laboratory
bred monkeys were determined. A screening was
carried out and those animals exhibiting a high GH
level were observed to be in periods of rapidly
increasing body weights. Male animals over eight
years of age and females over six had basal level GH
concentrations, corresponding to a plateau in body
weights.

Serum hormone concentrations of pregnant
monkeys were also measured with this method. Higher
value hormone concentrations corresponded to the
first trimester of pregnancy, while only basal value
hormone concentration was observed during the second
and third trimesters. This hGH-IRMA may thus
crossreact with cynomolgus monkey chorionic
somatomammotropin (=placental lactogen).

We already reported on serum fluctuating
patterns of chorionic gonadotropin (CG) and
prolactin (PRI) in pregnant cynomolgus monkeys.
Serum fluctuating patterns of somatomammotropin in
cynomolgus monkeys and human subjects during
pregnancy are also quite markedly different. Fron
this data it seems that the cynomolgus monkey is not
similar to the human subject in the endocrine
mechanism of maintaining pregnancy.

1136 Endocrinology

EN 27

EFFECTS OF ENVIRONMENTAL CALCIUM CONCENTRA-
TIONS ON PLASMA PROLACTIN LEVELS IN THE
EEL.

E.Arakawa, S.Hasegawa and T.Hirano. Ocean
Res. Inst., Univ. of Tokyo, Tokyo.

Prolactin (PRL), a freshwater-adapting
hormone in euryhaline teleosts, has been
suggested to act by maintaining primarily
plasma Na, and possibly Ca, concentrations.
In this study, we examined the effects of
changes in internal and external Ca concen-
trations on plasma PRL levels in the eel.
Transfer of the eel from seawater to fresh
water resulted in a significant increase in
plasma PRL levels, while transfer from
fresh water to seawater tended to decrease
the plasma PRL. However, there was no
significant change after transfer from
fresh water to fresh water containing 10 mM
CaClo or deionized water, or transfer from
seawater to Ca-free seawater. Plasma PRL
level was significantly lower in the stann-
ifectomized eel than the sham-operated
control 2 weeks after the operation, when a
significant elevation in plasma Ca was
observed. On the other hand, injection of
CaClo or EGTA did not affect plasma PRL
levels, in spite of significant changes in
plasma Ca concentrations. Thus, there
seems to be no apparent correlation between
external as well as internal Ca and plasma
PRL levels, although hypercalcemic action
of PRL cannot be ruled out.

EN 28

PROLACTIN TURNOVER RATE DURING FRESHWATER
ADAPTATION IN MATURE CHUM SALMON.
T.Ogasawaral, T.Sakamoto“, I.Mori“,
M.Suzuki* and T. Hirano*. 1Fac. of SCGilo g
Kanagawa Univ., Hiratsuka, “Ocean Res.
Inst., Univ. of Tokyo, Tokyo.

Returning chum salmon (Oncorhynchus
keta) in northern Honshu Island, Japan,
complete gonadal maturation while in the
bay. Mature fish caught in Otsuchi bay
failed to survive in seawater (SW) for more
than a week, whereas they adapted to fresh
water (FW) efficiently. Plasma prolactin
remained low in the SW fish, indicating
that the increased secretion of prolactin
is not the cause of maladaptation to SW.
When the mature fish caught in the bay were
transferred to FW, an increase in plasma
prolactin concentration was seen only in
females. Metabolic clearance rate (MCR)
and secretion rate (SR) were calculated
from the plasma level of prolactin after
intra-arterial injection of chum salmon
prolactin into chronically cannulated fish.
In both males and females, a significant
increase in MCR was seen after transfer to
FW. In females, SR increased significantly
after 6 days in FW. No such change in SR
was seen in males during adaptation to FW.
Thus, prolactin seems to be important
during FW adaptation in both sexes. It is
also likely that prolactin secretion is
affected with reproduction-related changes
in mature females in FW.

EN 29

DEVELOPMENT OF UTERINE ADENOMYOSIS IN MICE
TREATED WITH PSYCHOTHERAPEUTIC DRUGS

ie Singtripop*, T. Morit, M.K. Park?,

S. Kawashimat, S. Sakamoto”, and T. Katoh?
1Z001.« Inst.,; Fae. Sci. ,; Unive Tokyowand
2Dept. End., Med. Res. Inst., Tokyo Med.
Den. Univ., Bunkyo-ku, Tokyo 113

Adenomyosis is an abnormal growth of en-
dometrial tissues extending into myometri-
um. Our previous studies suggest that in
mice prolactin plays a key role in the de-
velopment of adenomyosis. On the other
hand, it is known that psychotherapeutic
drugs produce a significant rise in serum
level of prolactin. Therefore, the pres-
ent study was designed to examine the gen-
esis of adenomyosis after treatment with
psychotherapeutic drugs.

Female SHN mice were given subcutaneous
injections of 100ug metoclopramide, 200ug
sulpiride or 50ug perphenazine twice daily
between 40 and 90 days of age. Controls
received the vehicle only. All mice were
killed on the day of the last injection.

The results showed that administration
of these psychotherapeutic drugs induced a
prolongation of metestrous plus diestrous
phase of the estrous cycle, and a signifi-
cant increase in the incidence of adeno-
myosis when compared to those of the sa-
line-treated controls. Therefore, our
mouse data give a caution to the long-term
use of these drugs for clinical treatment.

EN 30
SEXUALLY DIMORPHIC DUCT SYSTEM IN _ THE
SUBMANDIBULAR GLAND OF MICE WITH

TESTICULAR FEMINIZATION (Tfm/Y).

K.Sawada and T.Noumura

Dept. of Regulation Biol., Fac. of Sci.,
Saitama Univ., Urawa

The X-linked testicular feminization

(Tfm/Y ) in the mouse is characterized by
androgen insensitivity of the target
cells. Sexually dimorphic development of

the submandibular gland in Tfm/Y mice was
examined morphometrically in comparison
with normal male, normal female and
heterozygous Tfm carrier female mice. At
30 days of age, both the relatively
occupied area (ROA) and the mitotic rates
of the granular convoluted tubules (GCT)
in the gland were greater in normal males
than in normal females and in heterozygous
Tfm females. After this age, the ROA of

male GCT was retained, but the male
mitotic rates gradually declined to the
female levels until 90 days of ages. In

the Tfm/Y mice, both the ROA and the
mitotic rates of GCT held the lower levels
resembled to those of both normal and

heterozygous Tfm females during this
period. In other three regions of the
gland, the acini (A), intercalated ducts
(ID) and excretory striated ducts (SD),

were not observed any difference in either
the ROA or the mitotic rates among normal
male, Tfm/Y, normal female and
heterozygous Tfm female mice.

Endocrinology ILS)

EN Si

GENOMIC EXPRESSION OF ESTROGEN RECEPTOR IN
THE RAT UTERUS TREATED WITH ESTROGEN
NEONATALLY.

S.Hayashi, H.Okamura and H.Ueda, Dept.
Anatomy and Embryology, Tokyo Metropol.
Inst. Neurosci., Fuchu, Tokyo 183.

Ontogeny of estrogen receptor (ER) in the
rat uterus was examined by in situ hybri-
dization method with RNA probe specific to
rat ER-mRNA. In the l-day-old rat, the
signals were detectable in the myometrium,
the endometrium and the epithelial cells,
while they were detectable in the epi-
thelial cells and in the uterine gland but
only vaguely found in the endometrium of
the 21-day-old uterus. Neonatal administ-—
ration of 10 pg estradiol benzoate (EB)
for consecutive 10 days exerted the
Signals to appear only in the epithelium
and the uterine glands. In the adult
uterus, the signals were clearly detected
only in the epithelium and uterine glands
regardless of neonatal EB treatment. Thus,
EB given to the newborn female rats might
have affected via ER system in the endo-
metrium and myometrium cells of the im-
mature uterus during the period when the
ER molecules were still actively produced
in situ and have exerted desensitization
of the uterus against the estrogen treat-
ment at adult. The authors thank Prof.
M.Muramatsu (Univ.Tokyo) for the supply of
rER-cDNA and Dr.A.Kuroiwa (Tohoku Univ)
LOGS enisiertaon) to) a veltorn.

EN 32

EFFECTS OF SEX STEROIDS AND TAMOXIFEN
ON THE SHAPE OF MOUSE PELVIS.

Y. Uesugi, T. Iguchi and N. Takasugi.

Dept. Biol., Yokohama City Univ., Yokohama.

The sexual dimorphism of the innominate bone
was examined in 14 strains of mice. In female
mice of all strains, the pubis (PB) was significantly
longer and thinner than that in the strain-
matched males. In 13 out of 14 strains, the
ischium (IS) in male mice was longer and thicker
than that in the females. In the testicular-
feminized male (Tfm) mice, IS was longer and
thinner than that in the wild-type males, showing
a value similar to that in the wild type females.
The PB of Tfm mice was longer than in the wild
type males. The PB width in Tfm mice showed a
value between those in wild-type males and
females. Gonadectomy at ages of 5, 10, 20, 30
and 60 days in both sexes showed that IS develop
to female-type when sex hormones are absent. By
contrast, peripubertal testicular androgen induced
male type IS. Gonadectomy at 60 days gave
minimum effect on PB, indicating that sexual
dimorphism of PB was determined within 60 days
of age. In addition, shape of PB of both sexes was
markedly affected by neonatally administered
tamoxifen, resulting in the shorter and thinner
bone compared to that in the controls. These
results indicate that 1) sexual dimorphism of
pubic bone is consistently present in 14 mouse
strains, 2) the shape of pubic bone seems to be
determined by sex steroids within 60 days of age,
and 3) neonatal tamoxifen specifically affects the
growth of pubic bone.

EN 33

IN VITRO FERTILIZATION OF OOCYTES FROM
POLYOVULAR FOLLICLES IN MOUSE OVARIES
EXPOSED NEONATALLY TO
DIETHYLSTILBESTROL.

K. Kamiya, T. Iguchi and N. Takasugi. Dept. Biol.,
Yokohama City Univ., Yokohama.

In 35-day-old female mice given 5 daily
injections of 1 ug diethylstilbestrol (DES) from
the day of birth, a significantly higher incidence of
polyovular follicles (PF) was found in the ovaries
than in those of age-matched control mice.
Oocytes were collected from the ovaries of 40-day-
old DES-exposed and control mice given a single
injection of 5 IU PMSG 58h before laparotomy
followed by 5 IU hCG 10 h before laparotomy.
Division of oocytes was examined 24-72 h after in
vitro insemination to ascertain whether
fertilization had occurred. Sixty-nine and 77% of
oocytes from uniovular follicles (UF) of control and
52 and 66% from UF of DES-exposed mice
developed into 2- to 4 -cell stage embryos
following in vitro insemination in C57BL and ICR
mice, respectively. By contrast, only 23 and 47%
of oocytes from PF of DES-exposed C57BL and ICR
mice, respectively, showed the division up to the
4-cell stage 72 h after insemination, indicating a
significantly lower fertilization rate compared to
the oocytes from UF of control and DES-exposed
mice. Without insemination, oocytes taken from
the same pools in these experiments were never
divided during the period of manipulation and
incubation. These results indicate that oocytes
from PF in DES-exposed mice have a significantly
decreased fertilization capacity in vitro.

EN 34

EFFECTS OF NEONATAL EXPOSURE TO
DIETHYLSTILBESTROL ON PROTEIN
EXPRESSION BY VAGINA AND UTERUS IN MICE.
Y. Takamatsu, T. Iguchi and N. Takasugi. Dept.
Biol. Yokohama City Univ., Yokohama.

Neonatal treatment of female mice with
diethylstilbestrol (DES) results im genital tract
abnormalities including ovary-independent
persistent vaginal proliferation and cornification.
Protein profiles were examined in vagina and
uterus from ovariectomized 45-day-old C57BL
mice given 5 daily injections of 2 ug DES
(estrogen-independent=E-IND) and the vehicle
alone (control) from the day of birth, and from
ovariectomized mice given 3 daily injections of 0.1
ug DES from 42 days (estrogen-dependent=E-
DEP). Proteins extracted with lysis buffer were
analyzed by two dimensional polyacrylamide gel
electrophoresis. In comparison with protein
expression (PEx) by the control vagina, 5 new PEx
(Mr=14.1, 20.0, 20.5, 40.0, 40.0 kD) by E-IND
vagina were found; 2 other PEx (Mr=15.0, 28.0
kD) by E-IND vagina were increased; 9 remaining
PExe (Mi—2 0°08 2 Gro 3o: Ol 3070610 o7038.0)
38.0, 63.0 kD) by E-IND vagina were reduced.
Keratin expression by E-IND vagina was identified
by immunoblot analysis. In the uterus, only one
PEx difference was found between E-DEP and E-
IND mice. These results indicate that neonatal
DES exposure induces tissue specific alterations
in the synthesis of a select number of proteins in
mouse vagina and uterus.

1138 Endocrinology

EIN'S5

ESTROGEN RECEPTOR (ER) AND ALKALINE
PHOSPHATASE (ALP) EXPRESSION IN FEMALE
GENITAL TRACTS OF MICE TREATED
NEONATALLY WITH DIETHYLSTILBESTROL.
T. Sato, T. Iguchi, Y. Ohta* and N. Takasugi. Dept.
Biol., Yokohama City Univ.,Yokohama, *Tottri Univ.
The distribution of the ER was investigated im-
munohistochemically in female genital tracts of O-
to 50-day-old mice treated neonatally with 2 ug
DES/day for 0-4 days using an anti-ER monoclonal
antibody. In the oviduct, cervix and vagina, both
epithelial and stromal cells showed positive ER
immunostaining on day O. Uterine stromal cells
also gave a strong ER immunostaining even on day
O. By contrast, the ER in uterine epithelial cells
began to form by day 5. The number of positive
epithelial cells and staining intensity gradually in-
creased until day 10. A single injection of DES on
day O induced ER in uterine epithelial cells 12 h
later. Vaginal epithelial cells undergoing ovary-in-
dependent stratification and cornification in DES-
exposed mice showed ER immunostaining only in
the basal and parabasal cells. ALP isoenzyme pat-
tern was histochemically examined in vagina of
DES-exposed mice. Vaginal epithelium showing
irreversible proliferation had kidney type ALP,
which is heat-unstable and insensitive to L-leucine
and L-phenylalanine. These results indicate that
1) ER in uterine epithelial cells is induced by
neonatal DES exposure, 2) ER is present only in
the basal cells of DES-exposed vaginal epithelium,
3) kidney type ALP is consistently expressed in all
epithelial cells of DES-exposed vagina.

EN 36

GROWTH OF CULTURED VAGINAL EPITHELIAL
CELLS FROM MICE EXPOSED NEONATALLY TO
DIETHYLSTILBESTROL IN RESPONSE TO
VARIOUS GROWTH FACTORS.
S. Ozawa, T. Iguchi, N. Takasugi and H. A. Bern*
Dept. Biol., Yokohama City Univ., Yokohama, *Dept.
Integrative Biol. and Cancer Res. Lab., Univ.
California, Berkeley, CA, U.S.A

Neonatal treatment with high doses of natural
and synthetic estrogens including diethy]l-
stilbestrol (DES) induces ovary-independent
vaginal epithelial stratification and cornification in
female mice. The growth of vaginal epithelial cells
from BALB/cCrgl mice treated neonatally with 2
ug DES daily for 5 days was studied in response to
growth-stimulatory and -inhibitory factors in a
serum-free collagen gel culture system which
sustains the growth of normal vaginal epithelial
cells. Cells taken from control and DES-exposed
mice at 21 days of age showed ca 5-fold increase
during 10 days in a serum-free medium
supplemented with BSA, h-transferrin, h-insulin
and m-EGF (cf. F.-D. Uchima). rb-IGF-I (1-100
ng/ml) failed to stimulate cell growth. Estradiol
(1.8 nM) and h-TGF-8 (0.05-5 ng/ml) consistently
inhibited cell growth in a dose-dependent
manner. TPA (10-6 -10-7 M) also inhibited cell
growth. EGF and insulin stimulated proliferation
of cells from both control and DES-exposed mice;
however, cells from DES-exposed mice showed
reduced growth in response to EGF and increased
growth in response to insulin compared with
control cells. (Supported by NIH grant CA 05388.)

EN 37

DEVELOPMENTAL CHANGES IN MOUSE
PLACENTAL CELLS AT DIFFERENT STAGES OF
PREGNANCY IN VIVO AND IN VITRO .

N. Tani, T. Iguchi, N. Fukatsu, Y. Ohta! and N.
Takasugi. Dept. Biol., Yokohama City Univ.,
Yokohama , and 1Dept. Biol., Fac. General
Education, Tottori Univ., Tottori.

Developmental changes in mouse placentae from
the 6th to the 18th day of pregnancy were studied in
vivo and in vitro. Placental volume increased from
the 6th to the 18th day of pregnancy in vivo; however,
total number of placental cells reached plateau on the
14th day. Decidual (D) cells were predominant in the
placenta on the 6th day. D cells increased in number
from the 6th to the 10th but decreased on the 14th,
whereas the number of trophoblasts (T) increased pro
rata until the 14th day. Two types of placental cells
were distinguished in vitro: small fibroblast-like cells
and large flattened cells containing 2-3 nuclei. Only
the large cells reacted to anti-desmin antibody,
indicating D cells. The small cells showed the
appearance of T. D cells from any gestation days
were unproliferous, becoming regressive with time in
culture. Estradiol (E) and progesterone (P)
stimulated growth of the T only on the 6th and the
10th. By contrast, T from the 14th and the 18th
proliferated in culture without sex steroids. These
steroids did not stimulate the growth of D cells in
culture. The present results indicate that KE and P
were effective in stimulating the growth of T of
placenta from early pregnancy, although uneffective
in modifying the limited life span of D cells in vitro.

EN 38

SEXING OF GIANT PANDAS BY FECAL IMMUNO-
REACTIVE SEX STEROIDS.

K.Kubokawa’ , K.Tanabe*, S.Ishii!. ‘Dept.
Biol., Sch.Edu., Waseda Univ., Tokyo,
“Ueno Zoological Gardens, Tokyo.

Feces were obtained monthly from
October 1987 from Giant Pandas of an adult
male (Fei Fei), adult female (Huan Huan),
and two infants of 4 years old (Tong Tong
or 11) and 2 years old (You You or 12) at
Ueno Zoo. They were kept at -20C until
the analysis. Steroids were extracted by a
modified method of Risler, et al. (1987)
using a mixture of acetone and ethanol,
petroleum ether and ethyl ether as extrac-
tion solvents. From extracted steroids
were separated estradiol (E2) and androgen
(T) by a column chromatography (DEAE A-
PAS) c In radioimmunoassays of E2 and T,
125I-labeled steroids of Amersham and the
first antibodies provided by Dr. K.
Wakabayashi of Gunma University were used.
Concentrations of T and E2 varied over
wide ranges among months. However, con-
centrations of T in the adult male were
always clearly higher than those of the
adult female. Concentrations of E2 were
similar between the adult male and female.
Concentrations of T in I1 were not so high
as the adult male but obviously higher
than the adult female. Concentrations of
T in I2 were lower than those in I1 at
the same age. These results suggest that
the fecal T concentration is useful for
sexing, and that I1 and I2'may be male and
female, respectively.

Endocrinology 1139

EN 39

ROLE OF SEVERAL HORMONES IN THE BODY
WEIGHT REGULATION BY PHOTOPERIODS IN THE
DJUNGARIAN HAMSTER.

A. Masuda and T. Oishi. Dept. of Biol.,
Fac. of Sci., Nara Women's Univ., Nara.

We investigated effects of testosterone
(T), melatonin (M) and thyroid hormones
(T3, T4) on the photoperiod-induced body
weight change in the Djungarian hamster.
Male hamsters with minimum body weight
after 25 weeks in a short photoperiod
(8L16D) were divided into eight groups;
(1) long and (2) short day controls, (3)
castrated control and castrated groups
with (4) testosterone implantation, (5)
testosterone and melatonin implantation,
(6) vehicle implantation, (7/)) T4
implantation and (8) 1T3 implantation.
Seven groups except the one group under
short day were transferred to a long
photoperiod (16L8D) after operation. In
the castrated group, an increase of body
weight was inhibited in spite of an

increase of food intake. Thus,
testosterone seems to attenuate energy
expenditure. Since melatonin inhibited

body growth in the castrated and T-treated
hamsters, it seems to act on their body
weight change through an independent
pathway from the hypothalamo-pituitary-
gonad system, probably by inhibiting their

feeding activity. Thyroid hormones
somewhat stimulated their food intake but
these hormones were not effective in

inducing the body weight change.

EN 40

CHANGES IN PLASMA THYROID HORMONES IN
JAPANESE QUAIL WHOSE BREEDING ACTIVITY
TERMINATED BY SHORT DAYS AND LOW AMBIENT
TEMPERATURE

M. Wada and H. Tsuyoshi

Dept. Gen. Educat., Tokyo Med. Dent.
Univ., Ichikawa, Chiba and Dept. Biol.,
School Educat., Waseda Univ., Shinjuku,
Tokyo.

In the previous report (Zool. Sci. 6,
1195), we suggested that termination of
breeding activity in Japanese quail in-
volved changes in plasma levels of thyroid
hormones especially T increases. To
clarify the changes during the treatment,
Samples were collected every other days
for 2 weeks and plasma levels of LH, Ty
and T3 were measured in 3 groups of mature
male birds; (1) birds kept on long days of
16L8D with 19 C, (2) birds transferred to
Short Wdays Of Sil 6D with! 19 ie; = and (3)
birds transferred to short days of 8L16D
Walia -LAlnvowre iS) CgilAonowie 2) Gc In group
1, plasma LH, Ty and 13 were unchanged
during the experiment period. In group 2,
plasma levels of LH remained at relatively
high levels and T increased after a ini-
tial decrease; T3 showed a slight in-
crease. In group 3, plasma levels of LH
decreased rapidly to the non-breeding
level and T3 increased dramatically while
changes in T, were basically similar to
those in group 2. The results confirm the
previous hypothesis that thyroid hormones
is involved in termination of a breeding
season by short days and low temperature.

EN 41

HORMONAL CONTROL OF COURTSHIP BEHAVIOR AND
SEXUAL ATTRACTANT SECRETION IN THE NEWT,
CYNOPS PYRRHOGASTER.
F.Toyoda' and S.Kikuyama’. ‘Dept. of Physiol.,
Nara Med. Univ., Kashihara and*Dept. of
Biol. Sch. of Educ., Waseda Univ., Tokyo.
We have previously reported that treatment
with gonadotropin (GTH) and prolactin (PRL)
elicites courtship behavior of male newts
and enhances the responsiveness of male and
female newts to the attractant(s) from the
opposite sex. Further studies revealed that
the water in which GTH and PRL-treated
newts had been kept attracted the opposite
sex more intensely than the water in which
GTH-or PRL-treated or saline-injected newts
had been kept. Treatment of castrated males
with testosterone propionate (TP) and PRL,
but not with GTH and PRL, elicited court-
ship behavior. The water in which castrated
males treated with TP and PRL, but not with
GTH and PRL, attracted female newts. Simi-
larly, the water in which ovariectomized
females injected with estradiol and PRL,
but not with GTH and PRL, attracted male
newts. These results indicate that PRL and
gonadal steroids are involved in eliciting
the male courtship behavior and enhancing
secretion of attractant(s) to the opposite
sex. Furthermore, it was revealed that ex-
tracts of well-developed oviduct and abdom-
inal gland of cloaca attract male and fe-
male newts, respectively. The oviduct and
cloacal glands are known to develop res-
ponding to PRL and sex steroids.

EN 42

THE RELATIONSHIP BETWEEN AMPLEXUS BEHAVIOR
AND PLASMA GONADOTROPIN AND SEX STEROID
LEVELS IN MALE TOADS, BUFO JAPONICUS.

M. Itoh and S. Ishii Depita sot Biloley,
Sch. of Edu., Waseda Univ., Tokyo.

We previously reported that there is a
surge of LH in male toads during the
breeding season and also that, under the
experimental condition, amplexing with not
only a female but also a dummy of the
female (a 10 x 6 x 1.5 em DbDlock of
"konnyaku", Japanese food whose component
is mannan) induced the LH surge in male
toads. In this work, we measured plasma
androgen and progesterone levels in addi-
tion to plasma LH and FSH levels in male
toads amplexing with a female (Group A) or
dummy (Group B) under an experimental con-
dition. In Group A and B, males clasped
the female or dummy, and the amplexus
lasted for about 18 hours. The sex steroid
levels decreased slightly at 36 hours in
Group A, B and control solitary males,
while the LH level increased markedly over
3 iO) 1A Inowies (Gtirom Sil s es} wo wWSs5il iss
3.2 ng/ml in Group A and from 2.5 + 0.6 to
13.5 + 3.3 ng/ml in Group B), and the FSH
level less markedly in Group A and B. Only
a slight but significant decrease of the
LH level was observed in the control
group. We conclude that the amplexus has
no effect on the plasma androgen and
progesterone levels in mala toads during
the breeding season, although it induces
the gonadotropin surge.

1140 Endocrinology

EN 43

TIME PROCESS OF OVULATION AND SUCCEEDING
FORMATION OF EGG SACS IN THE SALAMANDER
HYNOBIUS NIGRESCENS.

Masato Hasumi, Hisaaki Iwasawa and
Yoshitaka Nagahama*.

Biol. Inst., Niigata Univ., Niigata, and
~Itelojs One Isisjorctoyol4 lsist@ibo Neel. ibeSiss s@i6
Basic Biol., Okazaki.

Gravid females were trapped with drift
fences just before entering a pond during
the breeding seasons of early spring 1987
and 1990. These females were still ina
preovulatory condition. Ovulation was
caused in these females between 14 and 18
h after a single injection of gonado-
trophic hormone (600 IU hCG) in water at 8
C, because hCG-injected females had not
yet ovulated 14 h after the injection and
were just during ovulation 18 h after the
injection. Most of all ripe ova (a full
clutch) ovulated within the coelom entered
the oviducts and ovisacs 32 h after the
injection. The formation of egg sacs was
completed in the ovisacs at least 50 h
after the injection. Body mass did not
show any significant change among groups
@, 9, 14, and 18 h after the injections:
However it increased extremely 32 h after
the injection and further increased 50 h
after the injection. This increase in
body mass after ovulation was clearly the

result of absorption of water during the
formation of egg sacs.

EN 44

INVOLVEMENT OF CYCLIC NUCLEOTIDES IN THE
GONADOTROPIN-INDUCED SPERMIATION OF RANA
NIGROMACULATA, WITH SPECIAL REFERENCE TO
THE RELEASE OF SPERMATOZOA FROM SERTOLI
CELLS

Tohru Kobayashi and Hisaaki Iwasawa.

Biol. Inst., Niigata Univ., Niigata.

bses) Se ye ere a eee eee
The role of cyclic nucleotides in the
gonadotropin(GTH)-induced spermiation of
Rana nigromaculata was studied by means of
an in vitro spermiation system (Kobayashi
Gig aula MSI) Particular attention was
paid to the role of cyclic nucleotides in
the release of spermatozoa from Sertoli
cells. First, the effects of GTH and
forskolin, an activator of adenylate
cyclase, on intratesticular accumulation
of cAMP were determined. GTH(hCG: 10
IU/ml, bullfrog glycoprotein: 5 yg/ml) and
forskolin(10 uM) stimulated cAMP formation
within thr of incubation. Second, GTH,
forskolin, dibutyryl cAMP, dibutyryl cGMP,
and inhibitors of phosphodiesterase, 3-
isobutyl-1-methyl-xanthine on the in-
cidence of spermiation were investigated.
The action of GTH-induced spermiation was
mimicked by all the test agents known to
raise the cellular level of cAMP, with the
exception of dibutyryl cGMP. These
results provide evidence that action of
GTH which induces the release of sper-
matozoa from Sertoli cells depends on in-
creased intracellular cAMP in testis.

EN 45

EFFECT OF HYPOTHALAMIC PEPTIDES ON THE
PROLIFERATION OF ANTERIOR PITUITARY CELLS
IN VITRO.
;: Shinkai! ’2, Ee Ooka! and At Noumura’.
Dept. of Cell Biol., Tokyo Mefropol.
Inst. of Gerontol., Tokyo and “Dept. of
Regullatiion Biol., Fac. of Sci, Unav., oF
Saitama, Urawa.

The anterior pituitary cells from
adult rats were separated by centrifu-
gation on a Percoll gradient, and
cultured in vitro. Changes in numbers of
mMammotrophs and somatotrophs during
culture period cf 6 days were counted
after immunccytochemical staining. Growth
hormone releasing factor (rat GRF-29)
suppressed the increase in numbers of
mammotrophs in vitro. Cytosine
arabinoside, a mitotic inhibitor,
suppressed the increase of mammotrophs in
control culture, but it had no effect on
the number of mammotrophs cultured with
GRF. These results indicated that GRF
inhibited the proliferation of mammo-
trophs. Somatotrophs, which did not show
the change in number in control culture,
increased in the presence of GRF. The
proliferation of pituitary fibroblasts
was not affected by GRF. Somatostatin had
no effect on the number of acidophils.
These results suggest that GRF secreted
from the hypothalamus regulates the
relative population of acidophils in the
anterior pituitary, through their
opposite response to GRF.

EN 46

INDUCTION OF PEPTIDYLARGININE DEIMINASE
AND PROLACTIN BIOSYNTHESIS INA RAT
PITUITARY CELL LINE

S.Nagatax, T.Ueharax*, K.Watanabe*x,
K.Inoue** and T.Senshu* (*Dept. Biochem.,
Tokyo Metropol. Inst. Gerontol., Tokyo;
*xkInst. Endoc., Gunma Univ., Gunma)

When growth hormone’ (GH)-producing
MtT/S cells are exposed to insulin or IGF-
I in the presence of normal sera, GH
production decreases and prolactin (PRL)
biosynthesis is induced aS evidenced by
western immunoblotting and radiolabeled
amino acid incorporation. Contents,
biosynthetic rate and mRNA level of
peptidylarginine deiminase (PAD) are also
greatly increased showing Similar dose-
dependence and time course. Both IGF-I
and insulin receptors are demonstrated on
the cells by ligand binding assay and
affinity labeling technique. Neither PRL
nor PAD expression is induced by insulin
treatment in the medium containing
steroid-depleted sera, even though the
IGF-I and insulin receptors are expressed
normally. Simultaneous or pre-treatment
with estrogen restored the insulin action.
These results, together with our previous
findings that PAD immunoreactivity is
localized in PRL-producing cells in the
female rat pituitary and that estrogen

stimulates PAD expression, suggest
physiological roles of PAD related to
differentiation or function of PRL-

producing cells.

Endocrinology 1141

EN 47

REVERSE HEMOLYTIC PLAQUE ASSAY OF RAT
SOMATOTROPES.

S. Takahashi, Zool. Inst., Fac. of Sci.,
Hiroshima Univ., Naka-ku, Hiroshima.

The functional heterogeneity of GH
cells, and its age-related change were
studied by the method of reverse
hemolytic plaque assay. GH cell
population was divided into two groups:
one group of large plaque-forming GH
cells secreting more GH and the other
group of small plaque-forming GH cells
secreting less GH. GH-releasing hormone
(GHRH) increased the relative number of
the large plaque-forming GH cells.
Somatostatin increased the relative
number of the small plaque-forming GH
cells. Age-related changes were analyzed
in young and old female Wistar/Tw rats.
In old females, the relative number of
the large plaque-forming GH cells was
less than that in young females. The
average plaque area which reflects the
amount of GH released from individual
cells was smaller in old rats than in
young rats. GHRH increased GH secretion
in both ages. However, the mean plaque
area of GHRH-treated old pituitary cells
was smaller than that of young ones. The
present study showed that irrespective of
ages, GH cell populations consisted of
two groups. The diminished GH secretion
in old rats may be associated with the
decrease in the amount of GH released
from individual cells.

EN 48

SYNTHESES AND RELEASE OF GROWTH HORMONE
AND PROLACTIN IN THE CULTURED TROUT

PITUITARY.
T.Yada, A.Urano and T.Hirano.
Ocean Res. Inst., Univ. of Tokyo, Tokyo.

Pituitaries of rainbow trout
(Oncorhynchus mykiss) released a large
amount of growth hormone (GH) over a
week in serum-free culture, although
release of prolactin (PRL) declined in
the second day of culture. Decreases in
the residual contents of GH in the
cultured pituitaries were shown by both
RIA and light-microscopic immunohisto-
chemistry, while the residuals of PRL
did not alter during the culture. The
in situ hybridization technique showed
that the GH mRNA levels were maintained
over a week. On the contrary, 8-day
culture resulted in a reduction of the
PRL mRNA levels. By electron-
Microscopic immunohistochemistry, a
decrease in the secretory granules and a
dilation of the rough endoplasmic
reticula were observed in GH cells in
the cultured pituitary, although there
was no change in the fine structure of
PRL cells. These findings showed that
the synthesis and release of GH ina
cultured trout pituitary are still
active over a week without controls by
hypothalamic and peripheral hormones.

EN 49

DIFFERENTIAL SYNTHESIS OF GONADOTROPINS IN
FISH PITUITARIES DURING OVARY DEVELOPMENT.
N.Naitol, Y.Nakail, S.Hyodo2, A.Urano2,
N.Okumoto,3 Showa Univ. Sch. Med. Tokyo,
20cean Res. Inst. Univ. of Tokyo, Tokyo,
3Natl. Inst.for Aquaculture, Nikko.

In situ hybridization histochemistry and
indirect immunocytochemistry were used to
examine changes in the expression of sub-
unit genes of GTH I and GTH II (matura-
tional GTH) in female rainbow trout pitui-
taries during ovary development. Results
have indicated that gene expression of GTH-
subunits (a, If & IIf) is specifically
regulated both temporally and spatially
during gametogenesis in trout. GTH If8-and
Iifg-subunit mRNAs appeared separately in
Gwoudistinet cells) (GTH T-. & Ti-celis),;
whereas a-subunit mRNA appeared in both
cells, before appearance of either I8- or
IIf-mRNA. GTH I-cells exhibited coincident
increase in a- and I8-mRNA levels with
onset of vitellogenesis, indicating active
GTH I synthesis, and during vitellogenesis
further increased in number. Pn AGL shi
cells, however, parallel marked increase in
a—- and II8-mRNA levels occurred during late
vitellogenesis and continued thereafter,
indicating activity of GTH II synthesis.
Numerous GTH II-cells maintained elevated
a-and II8-mRNA levels during maturation and
ovulation. However, in GTH I-cells, levels
of a-mRNA began to decline prior to final
ovary maturation, although I8-mRNA remained
at high levels even after ovulation.

EN 50

ORGAN CULTURE STUDY OF GONADOTROPIN-
RELEASING HORMONE EFFECTS ON GONADOTROPHS
OF HYPOPHYSIAL PRIMORDIA FROM FETAL RATS.
A. Kudo, M. K. Park and S. Kawashima
Zoological Institute, Faculty of Science,
University of Tokyo, Tokyo.

We investigated the effects of gonado-
tropin-releasing hormone (GnRH) on the
population of gonadotrophs in cultured
adonohypophysial primordia at different
embryonic stages. Adenohypophysial pri-
mordia derived from rat fetuses at 12.5,
13.5, 14.5 and 17.5 days of gestation were
maintained in vitro until the day equiva-
lent to 21.5 days of gestation. During
the first 24 hours of culture GnRH (107 2m)
was added. The number of LH- and FSH-
immunoreactive cells was counted in ex-
plants. GnRH treatment failed to alter
the total cell number of explants derived
from any embryonic stage, but the ratio of
LH-immunoreactive to total cells signifi-
cantly increased by GnRH treatment in
exp Lanits: Shroom 35 — (8.316% 60.2 39%)" and
14.5-day fetuses(5.0% to 11.7%). The same
treatment was ineffective on FSH cell
population. These results indicate that
the effect of GnRH on the development and
differentiation was different between LH
and FSH cells. Radio-immunoassay of LH
and FSH released in the culture media
showed that LH was detectable only in
cultures of 17.5-day fetuses and FSH, in
all cultures, regardless of fetal ages,
and that any significant effect of GnRH
was not detected.

1142 Endocrinology

EN 51

ANALYSIS OF DYNAMIC LH RELEASE FROM IN-
DIVIDUAL GONADOTROPHS IN MALE RATS BY THE
REVERSE Our ASSAY.

A. Hattori elie Tomoike2, Rye Ohtani! and
- Suzuki

Department of Anatomy, St. Marianna
University School of Medicine, Kawasaki,
“College of Agriculture and Veterinary
Medicine, Nihon University, Fujisawa.

Two modified methods of reverse
hemolytic plaque assay (continuous and
sequential plaque assays) were used to
clarify the dynamics of LH release by in-
dividual gonadotrophs from 60-day-old male
rats. In the sequential assay, only the
pituitary cells attached to the floor of
the chamber, and fresh batches of RBC, an-
tibody and complement were added twice
sequentially for determination of the
release pattern of each gonadotroph.
There were 4 patterns of LH release; 1)
rapid release and then no more release, 2)
continuous release, 3) release after a
time lag, and 4) non-release during in-
cubation time (7 hrs). Daily sequential
plaque assay for LH release from the same
cell showed that a few gonadotrophs con-
tinued the same pattern of the release,
however, most of the cells changed from
day to day. These results demonstrated
that gonadotrophs can change their release
patterns, which suggests’ that each gonado-
troph has a "cellular release cycle".

EN 52

APPEARANCE OF o-MSH AND ACTH IN THE
PITUITARY GLAND OF THE DEVELOPING CHICK
EMBRYO SHOWN BY IMMUNOCYTOCHEMISTRY
AND RADIOIMMUNOASSAY
K. Imai, K. Imai? and H. Hayashi?
IFac. of General Studies and 2Inst. Endocrinol.,
Gunma Univ., Maebashi

We investigated the appearance of oaMSH- and
ACTH-containing cells and both hormones in the
developing chiken pituitary using anti-wMSH and
anti-human ACTH sera by immunocytological and
RIA methods. Cells immunoreactive to these antisera
are detected at first in the differentiating hypophysial
tissue of Day 7 and Day 8 of incubation, respectively.
As pituitary glands grow larger, the number of MSH
and ACTH cells increases gradually. RIA shows that
immunoreactive aMSH and ACTH are demonstrable
from Day 5. Contents of both hormones are very low
at first. ~MSH remains at a low level until hatch and
increase remarkably from 5 days after hatch. On the
contrary, ACTH increases from Day 15 of incubation,
followed by a slight drop at 5 days after hatch. The
original source of immunoreactive hormones found
by RIA before demonstrable in immunocytochemistry
is not clear.

EN 53

COLOCALIZATION OF a-SUBUNIT OF GLYCO-
PROTEIN HORMONES IN PROLACTIN (PRL)
CELLS OF THE BULLFROG PITUITARY GLAND.
S.Tanaka, K.Yamamotol, S.Kikuyama! and K.Kurosumi
Dept. of Morphol., Inst. of Endocrinol., Gunma Univ.,
Maebashi and Dept. of Biol., Sch. of Edu., Waseda Univ.,
Tokyo.

Immunoreactive o-subunit cells in the pituitary invariably
exceed the sum of immunoreactive LHB, FSHB, and TSHB cells
in number through larval life of bullfrog, indicating that the a-
subunit also exists in the pituitary cells other than LH/FSH
cells and TSH cells. The present study demonstrated that there
are a considerable number of cells which were stained with both
@-subunit antibody and bullfrog PRL antiserum when
consecutive sections were immunohistochemically examined.
Absorption and immunoblotting test of these two antibodies
revealed that each of them is specific for the corresponding
antigen. Immunoelectron microscopy, using double-labeling with
gold particles of different sizes for a-subunit and PRL revealed
that almost all of immunoreactive PRL secretory granules were
also positive to @-subunit antibody. The a-subunit
immunolabeling was often observed in the peripheral region of
the secretory granules. In pimozide-treated frogs the plasma
levels of &-subunit and PRL showed a significant increase when
compared with those in the saline-treated ones, suggesting that a
simultaneous release of &-subunit and PRL took place. -

EN 54
SUBCELLULAR LOCALIZATION OF LH AND FSH IN
BULLFROG PITUITARY GONADOTROPHS
F.Mizutani, H.Iwasawa, S.Tanaka! and K.Kurosumi!
Biol. Inst., Fac. of Sci., Niigata Univ., Niigata, 1 Dept. of
Morph., Inst. of Endocrinol., Gunma Univ., Maebashi.

Subcellular localization of LH8, FSHB and a-subunit in the
bullfrog pituitary gonadotrophs were examined by the two face-
double labeling method with gold particles of diffrent sizes
for their specific antibodies. Through sexual maturation, three
types of gonadotroph were found: the 1st cell type was small in
size having only several round secretory granules (SG). The 2nd
was of intermediate size which contained a well-developed Golgi
apparatus and rER in addition to round or oval SG. The 3rd had
numerous SG in the large amount of cytoplasm with less
developed organelles. The 1st type was found predominantly in
larval pituitary (TK St.25). The 2nd and 3rd types were
frequently observed in adult, while the Ist and 2nd were in
young adult. The double-labeling for LH8 and FSHB showed a
differential localization depending upon the cell type and
individual SG. In the 1st type almost all the SG were labeled for
FSHB, but scarcely for LH8. Similarly, almost all the SG in the
2nd were labeled for FSHB, but more than a half of which were
also stained for LHB. In the 3rd, both LHB& and FSHB were
colocalized in almost all the SG, but some SG stored either LHB
or FSHB. Immunoreactivity of @-subunit was always found in
all the SG of these three cell types.

Endocrinology 1143

EN 55

Immunohistochemical study of bullfrog
gonadotropin in the course of sexual
maturation.

T.Numa!, S.Tanaka”*, and H.Iwasawa!
Ast Ol WMS sp Niigata Univ., Niigata,
2Inst. of Endocrinol., Gunma Univ.,

Maebashi

We immunohistochemically studied
gonadotropin productibility in young and
subadult frogs of Rana catesbelana.

Immunostaining was performed by the PAP
method using monoclonal antibodies to
bullfrog LHf, FSHf, and «-subunit. In this
study, frogs which metamorphosed this year
were termed 1-year, and 1- to 3-year male
and female frogs were used as materials.
Regardless of age,sex or season, the rates
for each type of immunostained cell to all
anterior pituitary cells were in the
following order: «-subunit >» FSHB >> LH.
In the course of sexual maturation, each
type of subunit-producing cell increased
in 2- and 3-year male and female frogs.
Regardless of age or sex, no change was
detected in the immunoreactive cells of
each type during autumn.

EN 56

ANNUAL CHANGES OF PLASMA PROLACTIN LEVELS
IN THE NEWT, CYNOPS PYRRHOGASTER

K. Matsudal, S. Tanaka“, K. Yamamotol, and
S. Kikuyamal. lDept. of Biol., Sch. of
Educ., Waseda Univ., Tokyo. 2Inst. of En-
docrinol., Gunma Univ., Gunma.

The homologous radioimmunoassay (RIA)
for newt prolactin (PRL) was applied to
the determination of plasma PRL levels in
newts captured in every month of the year.
In the adult newts of both sexes, plasma
PRL levels were relatively low after the
breeding season in early spring and during
summer and early autumn (5-70 ng/ml). In
the male, the levels rose markedly in
March (205.4 + 49.4 ng/ml), while in the
female, the levels were elevated
moderately in February (95.5 + 10.9 ng/ml)
and November (107.8 + 8.5 ng/ml). PRL in
the male is presumed to be necessary for
the development of the cloacal glands,
tail fin, Mauthner's neuron, and courtship
behavior. In the female, PRL may induce
oviducal development. Immunohistochemical
study revealed that pituitary PRL cells in
the spring newt were stained weakly, while
those in the summer animal were stained
rather strongly. In the PRL cells of
spring newt, but not of the summer animal,
well-developed Golgi apparatus was ob-
served, suggesting that the secretory ac-
tivity is higher in spring than in summer.

EN 57

ESTABLISHMENT AND APPLICATION OF TIME-
RESOLVED FLUOROIMMUNOASSAY (TR-FIA) FOR THE
MEASUREMENT OF NEWT GONADOTROPIN (GTH)

A. lwasawa, S. Tanaka and K. Wakabayashi

Gunma University Institute of Endocrinology, Maebashi.

Plasma- and pituitary gonadotropin of newts was measured for
the first time via cross-reaction with a TR-FIA system for bullfrog
LH. TR-FIA is a newly developed, non-isotopic immunoassay, in
which europium, a lanthanide element, is used as a label. The
present assay system is based on the "sandwich-binding" principle;
monoclonal anti-bullfrog LHB (BL4B11) immobilized in microtiter
wells is used as a Capturing antibody, and europium-labeled
polyclonal anti-bullfrog LH as a detecting antibody. Dilution curves
of newt pituitary homogenates and high GTH sera obtained from
mammalian GnRH-treated newts were parallel to the standard
curve below 4 ng/ml of bullfrog LH. To determine the electric
nature of newt GTH measurable by this assay system, pituitary
homogenate of male newts collected in July was subjected to an
isoelectric focusing analysis. Nine components with pl's ranging
from 9.8 to 5.2 were observed by TR-FIA. All of them were also
detectable by RRA using Xenopus testicular homogenates as the

receptor preparation and 125) rat FSH as the radioligand. The
RRA is thought to be specific for LH-like GTH. Pooled plasma of
adult newts collected monthly throughout one year was assayed
for GTH by TR-FIA. Significant increases were observed in June
(0.39 ng/ml) for males and in May (0.79 ng/ml) for females, which
correspond to the period of spermatogenetic recrudescence and
ovulation, respectively, showing the usefulness of this assay system
for the further investigation on reproduction of newts.

EN 58

HOMOLOGOUS RADIOIMMUNOASSAY OF BULLFROG
JOINING PEPTIDE

S. Iwamuro!, M. Nomizu?, S. Tanaka3, H.
Hayashi, and S. Kikuyama!. !Dept. Biol.
Sch. Educ., Waseda Univ., Tokyo, 2?Pharm.
Lab., Kirin Brewery, Maebashi, and 3Inst.
Endocrinol., Gunma Univ., Maebashi.

A double antibody RIA for bullfrog (Rana
catesbeiana) joining peptide (fJP), which
belongs to POMC-related peptides, was es-
tablished. The antiserum was produced by
immunizing a rabbit with synthetic fJP con-
jJucated with BSA. Synthetic fJP was used
for radioiodination and as a reference
standard. The intra- and inter-assay coef-
ficient of variations were 3.7% and 4.8%,
respectively. Sensitivity of the assay was
0.054 + 0.021 ng/ml. Several dilutions of
plasma as well as homogenates of anterior
(PD) and neurointermediate lobes (PIN)
of bullfrog and R. ornativentris yielded
dose response curves which pararelled to
the standard curve. Plasma from totally
hyopophysectmized bullfrogs contained no
measurable JP, but from distalobectmized
bullfrogs contained twice as much as those
of intact ones. Plasma and homogenates of
PD and PIN of Bufo japonicus and Xenopus
laevis gave inhibition curves which did not
pararell the standard. Purified frog NPP
and several human POMC-related peptides did
not react in this assay. Both PD and PIN
cultured separately released JP con-
siderably, the amounts of JP from PD being
1/2 of those from PIN. Co-cultured PD and
PIN released JP more than the sum of JP
from PD and PIN cultured separately.

1144 Endocrinology

EN S9

LOCALIZATION OF PITUITARY PRIMORDIUM IN
EARLY EMBRYOS OF BUFO JAPONICUS.

K. Kawamura and S. Kikuyama. Department
of Biology, Shool of Education, Waseda
University, Nishi-Waseda, Shinjuku-ku,
Tokyo 169, Japan.

The aim of the present study is to
localize pituitary primordium in the open
neurulae of Bufo japonicus. Tissue pieces
taken from wild-type eggs were
transplanted to albino eggs, and the
developmental fate of the transplants was
traced by using melanin granules as a
natural marker. It was revealed that the
primordial pituitary cells are situated in
the anterior part of the neural ridge
(ventral neural ridge, VNR) on the open
neurulae, and that the presumptive
infundibular neurons are localized just
posteriorly to the VNR. The VNR performs
a morphogenetic movement antero-ventrally
before forming a conical cell mass under
the diencephalic floor. The conical cell
mass, which corresponds to the RathkKe's
pouch in mammals, establishes a connection
with the infundibulum and finally
differentiates into pars distalis, pars
intermedia and pars tuberalis. It was
also shown that the presumptive pituitary
cells acquire a self-differentiating
potency between the neurula stage and the
tail-bud stage.

EN 60

CHANGES OF VASCULAR ROUTE AFTER INDUCED
ANTERIOR PITUITARY HEMORRHAGE IN MICE.

I. KOSHIMIZU and Y. KOBAYASHI, Dept. of
Biol., Fac. of Sci., Okayama Univ.,
Okayama.

Acute and intense hemorrhage into _ the
anterior pituitary occurs’ by intraperito-
neal injection of 9% NaCl at a dose of
0.03 ml/g body weight. After 3 days of
hemorrhage, the middle portion of the
anterior pituitary was occupied by
extensive necrosis that was replaced by
granuration tissue-like structure after 7
days of hemorrhage. On day 4-7, a part of
the necrosis occasionally protruded into
the pituitary lumen. Scanning electron
microscopy revealed that conglomeration of
thin capillaries sprouted up from the
middorsal portion of the anterior
pituitary into the pituitary lumen. This
glomerular form of the vascular’ route
corresponds to the histological extrusion
of the necrosis and it seems to reflect
the restoration of the necrosis after
hemorrhage.

EN 61

EFFECTS OF BLOOD COAGULANT ON INDUCED
PITUITARY HEMORRHAGE IN MICE

T.Nakano and Y.Kobayashi, Dept. of Biol.,
Fac. of Sci., Okayama Univ., Okayama.

Hemorrhage into the anterior pituitary
occurs immediately after ip injection of
9% NaCl at a dose of 0.03ml/g B.W. After
hemorrhage, erythrocytes of extravasation
do not coagulate in the anterior lobe but
move preferentially to the pituitary
lumen. In the present study, blood coagu-
lant (anti-plasmin, 4mg/g B.W./day) was
applied on day 1-9 after hemorrhage, and
the fine structural changes were examined
during restoration of the gland.

Extensive necrotic area waS replaced by
granulation tissue by day 7. The necrotic
region was frequently observed in the
pituitary lumen extruding from the dorsal
part of the anterior lobe forming a mush-
room-like outline. On day 9, there were a
number of non-granulated cells in the
rostral region of the anterior lobe.
These agranular cells seems to be undif-
ferentiated and / or reserved cells that
may participate in the restoration of the
gland after hemorrhage.

EN 62

SEXUAL DIMORPHISM IN THE MOUSE BRAIN AND
MODIFICATION BY TESTICULAR FEMINIZATION
MUTATION.

S. Tsukada and T. Noumura.
Dept. of Resullat.) (Bi oles: ce Oram colar.
Saitama Univ., Urawa.

Nissl-stained sections of the brains
from 90-day-old male and female mice of a
colony carrying testicular feminization
mutation (Tfm) were examined for possible
sex differences in the sizes of hypothala-
mic nuclei, including the paraventricular
nucleus (PVN), the suprachiasmatic nucleus
(SCN), and the medial portion of the ante-
rior hypothalamic area (AHm). The X-linked
coat markers blotchy (Blo) and tabby (Ta)
were used for identification of the X-chro-
mosomes present in an animal. Animals used
in this study were normal and mutant males
(+++/Y, Tfm+Blo/Y) and females (+++/+++,
srl ise // saves Tfm+Blo/+++).

There were much differences among them
in terms of body weight but no significant
differences in terms of brain weight. The
volumes of PVN were larger in normal males
than in normal females and intermediate
in Tfm males between them, while those of
SCN were larger in Tfm males than in Tfm
carrier females.

Endocrinology

EN 63

EFFECTS OF ORCHIDECTOMY ON LATERALITY OF
DIENCEPHALIC NEURONS IN MALE MICE.

Y.Inase and T.Machida

Dept. of Regulation Biol., Fac. of Sci.,
Saitama Univ., Urawa.

Male mice of the CD-1 strain were
castrated unilaterally or bilaterally at
60 days of age. One week after
operation, changes in the size of neuronal
nuclei and the number of LHRH _ immuno-
reactive cells in discrete regions of the
diencephalon, i.e. preoptic area (POA) and
diagonal band of Broca (DBB), were
examined.

Neuronal nuclei of these regions were
Significantly larger in the right side of
the brain than in the left. side. The
left-right difference was maintained
following removal of the testis at either
the left, right, or both side(s).

Similarly, the number of LHRH cells
were greater inthe right side of the
brain than in the left side and removal of
the testis exerted no alterations on this

asymmetry.
On the other hand, in both sides of the
brain, either bilateral or unilateral

orchidectomy reduced the number of LHRH
cells in the order of sham-castrated,

left-side hemicastrated, right-side
hemicastrated and bilateral castrated
mice.

The results showed a left-right dif-
ference in the neuroendocrine regulatory
system of gonadal function in male mice.

EN 64

EFFECT OF SEX STEROIDS ON THE SURVIVAL OF
CULTURED NEURONS DERIVED FROM THE NEONATAL
RAT PREOPTIC AREA.

K. Takagi and S. Kawashima

Zoological Institute, Faculty of Science,
University of Tokyo, Tokyo.

It is well established that some nuclei
of the rat brain are sexually dimorphic in
the volume and neuronal cell density. For
example, sexually dimorphic nucleus of the
preoptic area is about seven-fold larger in
the adult male rats than in the adult
female(Gorski et al.,1978). Such a sexual
dimorphism is believed to be caused by the
exposure to sex steroids like testosterone
and estradiol-178(E5) during the perinatal
period. Using serum-free cell culture
system, the cells of neonatal rat preoptic
area were cultured, and the effects of
three kinds of sex steroids, testosterone,
E5, and 5a-dihydrotestosterone(DHT), on
the survival of neurons were examined in
the present study. The survival of neurons
was significantly increased by 1.0 mM
testosterone at 14th day of culture(12 days
of treatment). Eo slightly increased the
survival at the concentration of 1.0 mM.
DHT was not effective on the survival at
any concentration. To examine the effect of
testosterone on the non-neuronal cell
growth, protein content of cultures was
measured. Protein content was significantly
increased by 10.0 nM testosterone at 8th
day of culture, but this effect was not
observed at later stages of culture.

EN 65

LORDOSIS IN MALE RATS: EFFECT OF DORSAL
RAPHE NUCLEUS LESIONS

M. Kakeyama and K. Yamanouchi
Neuroendocrinology, Dept. of Basic Human
Sciences, Sch. of Human Sciences, Waseda
University, Tokorozawa

The dorsal raphe nucleus was destroyed
by a radio frequency lesion generator in
castrated male rats (DRL group). In
addition, sham operated male control
(SHAM) and non brain surgery male and
female (MALE CONTROL and FEMALE CONTROL,
respectively) groups were made. Five weeks
after the surgery, all animals were
implanted with two 3 cm Silastic tubings
(1.57:3.18, ID:0D) containing estradiol
(E2). Behavioral tests were started 2 days
after implantation of E2 and carried out
every other day for 16 days (8 tests). The
mean LQs of all groups in the first test
were very low. In the FEMALE CONTROL
group, LQs increased to maximum level in
the 4th test. In contrast, MALE CONTROL
and SHAM males showed low scores of LQ
throughout the tests. On the other hand,
all of DRL males showed lordosis and the
mean LQ was significantly higher than
those of MALE CONTROL and SHAM groups, but
lower than that of FEMALE CONTROL group
in all tests except the first test. These
results suggest that the dorsal raphe
nucleus may participate in an inhibitory
system of lordosis behavior in male rats.

EN 66

SUPPRESSION OF SEXUAL BEHAVIORS BY RAPHE
OBSCURUS NUCLEUS LESIONS IN MALE RATS:
EFFECT OF P-CHLOROPHENYLALANINE (PCPA)
K. Yamanouchi and M. Kakeyama
Neuroendocrinology, Dept. of Basic Human
Sciences, Sch. of Human Sciences, Waseda
University, Tokorozawa.

The effect of PCPA on sexual behaviors
in male rats with raphe obscurus nucleus
lesions was examined. The raphe obscurus
nucleus was destroyed by a radiofrequency
lesion generator (ROBL) in castrated males.
Non-brain surgery castrated control (CONT)
group was also made. All male rats were
implanted subcutaneously with Silastic
tubings containing testosterone (T).
Behavior tests were carried out on day 7,14,
and 21 after implantation of T. In the second
test, half of ROBL- and CONT-males were
treated with 100 mg/kg PCPA daily for 4 days.
Male rats with ROBL showed low levels of
mounting and intromission frequencies, when
compared to those of other groups. None of
them indicated ejaculation pattern. On the
other hand, PCPA potentiated sexual activity
in both ROBL and CONT groups. These results
suggest that the raphe obscurus nucleus may
play an important role in regulation of
sexual behaviors in male rats. In addition,
PCPA may act on the raphe nucleus other than
the raphe obscurus nucleus and facilitate
sexual activity.

1145

1146 Endocrinology

EN 67

INTRACRANIAL INJECTION OF 5,7-—DIHYDROXY-
TRYPTAMINE LOWERED TESTOSTERONE THRESHOLD
FOR INDUCTION OF RAT COPULATORY BEHAVIOR.
Y¥.Kondo, and Y.Arai. Dept. of Anat., Jun-
tendo Univ. Sch. of Med., Tokyo.

Since systemic administration of p-
chlorophenylalanine facilitates male copu-
latory behavior, it is suggested that
serotonergic system exerts an inhibitory
influence on copulation in male rats. In
this study, the effect of 5,7-dihydroxy-
tryptamine (5,7-DHT, neurotoxin of seroto-
nin) injection into the preoptic area (POA)
or the corticomedial amygdala (CMA) on
copulatory behavior were examined in cas-
trated male rats. Three weeks after the
intracranial injection of 5,7-DHT or saline
(CONT), all animals received subcutaneous
implantation of a silastic tube (Dow-Corn-
ing, #602-285, 5cm) containing testosterone
(Ty. Two observations of copulatory behav-
ior with an estrous female were carried out
every 10 days after T implantation. In
control animals, mounts and intromissions
were not seen in the 1st and 2nd observa-
tions. In contrast, animals with POA
injection showed significantly higher mount
frequency than animals of CONT. All of
these animals showed both mounts and intro-
missions in the 2nd observation. Four out
of 5 animals with CMA injection also showed
both mounts and intromissions, although the
differences to those of CONT were not
significant. These may indicate that
serotonin terminals in both POA and CMA
inhibit copulatory behavior in male rats.

EN 68

THE APPEARANCE OF LHRH NEURON IN THE BRAIN
OF RAINBOW TROUT, SALMO GAIRDNERI, DURING
EMBRYOLOGICAL DEVELOPMENT

T.Saga!, y.Oota! and M.Nozaki%. !Biol.
WASiEs5 PAC. Ol SEit.5. Sladguolxe Wma ys oq
Shizuoka. “Primate Res. Inst., Kyoto

Univ., Inuyama.

Localization of LHRH immunoreactive
neurons in the brain of rainbow trout has
been studied only in the adult. In this
study, we investigated the appearance and
distribution pattern oof the LHRH immu-
noreactive neurons in the rainbow trout
brain during early development. Further-
more, correlation between the development
of GTH-I cells and gonadal differentiation
was examined. LHRH immunoreactive neurons
have not appeared in the brain of the
larval trout till stage 32. At stage 32,
LHRH immunoreactive neurons first appeared
and distributed throughout the brain but
not so much. Although most of the LHRH
immunoreactive neurons ran from the olfac-
tory bulb via the hypothalamus towards the
pituitary, distinct penetration to the
pituitary was not observed in this stage.

Our previous investigation suggested
the close correlation between the develop-
ment of GTH-I cells and gonadal differen-
LEAL Gl1G aL Oia ¢ The present study demonstrate
the coincident the appearance stage of
LHRH immunoreactive neurons and GTH-I
cells. These results suggest that the
hypothalamo-pituitary-gonad axis may not
be established in larval rainbow trout
till stage 32.

EN 69

LOCALIZATION OF NEURONS INVOLVED IN THE
BLEACHING REACTION IN THE TOAD, BUFO
JAPONICUS.

Z. Kato, K. Kawamura & S. Kikuyama, Dept.
Biol., Schl. Educ., Waseda Univ., Nishi-
Waseda, Shinjuku-ku, Tokyo 169, JAPAN.

i

Background response in amphibians is
accomplished through a regulation of
circulating levels of aMSH. The
secretion of a MSH is under inhibitory
control of hypothalamus. There is ample
evidence that the inhibitory signal is
mediated by dopaminergic neurons. In this
species, dopaminergic neurons are situated
in the preoptic recess organ (PRO), the
paraventricular organ (PVO) and the dorsal
infundibular nucleus (NID). In order to
determine which hypothalamic nucleus is
responsible for the inhibition of a MSH
secretion at the time of white-adaptation,
we made transsections at various levels
between the PRO and the hypophysis, and
observed the possible disappearance of
bleaching reaction. It was suggested that
the dopaminergic axons which originate in
PRO and pass through the dorso-lateral
parts of the optic chiasm are necessary
for white-adaptation. This result
coincides with the fact that the
background responsiveness and the
dopaminergic neurons in PRO develop
concomitantly at metamorphic climax, while
the dopaminergic neurons in PVO and NID
develop in the premetamorphic period.

EN 70

HIBERNATION-RELATED PROTEINS IDENTIFIED IN
PLASMA OF CHIPMUNKS

N. Kondo! and J.Kondo@. IMitsubishi kasei
Inst. of Life Sci., Machida, Tokyo and
Mitsubishi Kasei Corp. Res. Center,
Yokohama, Kanagawa.

In a previous study, we found that
chipmunks, mammalian hibernator, have four
types of proteins related to hibernation
(HPs). HPs (20, 25, 27 and 55kD M.W.)
begin reducing from blood before
hibernation and disappear during
hibernation. These reappear in blood as
hibernation cease. In analysis of amino
acid sequence, 55kD protein (HP-55) was
highly homologous with a,-antitrypsin,
while 20, 25 and 27kD proteins (HP-20, 25
and 27) were not homologous with any known
proteins. These novel HPs were homologous
to each other and contained a collagen-like
sequence in their N-terminal regions (about
40 residues). In plasma from other
hibernator in rodents, the 13-lined ground
squirrels, the proteins corresponding to
novel HPs were detected but not in plasma
from nonhibernators in same order, formosan
tree squirrels and rats. The proteins
corresponding to HP-55 existed in plasma
from all animals tested. The present novel
proteins were specific for hibernation.
This suggest the possibility that
hibernation is genetically controlled.

Endocrinology 1147

EN 71

A NOVEL NATRIURETIC PEPTIDE ISOLATED FROM
EEL CARDIAC VENTRICLE

Y. Takeil, A. Takahashi2 and T. Watanabe.
1Dept. Physiol. and 2Mol. Biol., Kitasato
Univ. Sch. Med., Sagamihara, and 3Peptide
Inst. Ince., Minoh.

From 2500 eel ventricles (225 g), a new
species of natriuretic peptide which
possesses all spectra of actions known to
be characteristic to atrial and brain
natriuretic peptide (ANP and BNP), was
isolated from eel cardiac ventricles, and
have named ventricular natriuretic peptide
(VNP). The basic structure of eel VNP is
quite similar to ANP and BNP so far
identified, but it has a uniquely long C-
terminal ‘'tail' that extends from the
second half cystine. Thus eel VNP appears
to be a novel type of natriuretic peptide
that has not been found in mammals. Eel
VNP is, like eel ANP and BNP, much more
potent than human ANP with respect to the
vasodepressor activity in the homologous
animal (eel). With respect to the
natriuretic activity in the rat, however,
eel VNP was much more potent than eel ANP
and BNP, thus is almost equipotent to
human ANP. Since VNP is secreted in a
larger amount than ANP into the
circulation of eels, VNP as well as ANP
may be involved in the excellent
osmoregulatory mechanisms of this
catadromous fish. We are now examining
whether or not VNP is present throughout
vertebrate classes.

EN WZ

TWO NOVEL PEPTIDES ISOLATED FROM THE IN-
TESTINES OF THE EEL.

T.Uesaka’, T.Ikeda’, I.Kubota*, Y.Muneoka’
and M.Ando’. ‘Lab. of Physiol., Fac. of
Integrated Arts and Sci., Hiroshima Univ.,
Hiroshma and ?Suntory Bio-Pharma. Tech.

Center, Chiyoda-cho, Gunma.

Acetone extract of the intestines ex-
cised from 200 eels was forced through a
C-18 cartridges. The retained material
was applied to a reversed-phase column.
Each fraction was bioassayed on phasic
sponse to repetitive electrical pulses (15
V, 3 ms, 10 Hz, for 5 s). The active
fractions were then subjected to a cation
exchange column. Two peaks of potenti-
ating activity on the phasic contraction
of the ABRM. These peaks were further
purified each other through several HPLC
purification steps. The structure of the
purified substances were determined by
chemical analysis to be as follows:

EICP-1 H-Gly-Phe-Trp-—Asn-Lys-—OH

EICP-2 H—-Phe-Pro-Ser-Ile-Val-Gly-Arg

—-Pro-Arg-OH

The peptides do not appear to be members
of any other previously identified peptide
family. Both of these new peptides also
induced tonic contraction of the in-
testinal strips (longitudinal muscle) iso-
lated from the eel. These peptides were
termed EICP(eel intestine-contracting
peptide)-l and -2,respectively.

EN 73

CENTRAL NEUROTRANSMITTERS AND OSMOTIC
VASOPRESSIN (VP) SECRETION.

K. Yamaguchi, Dept. of Physiol., Niigata
Univ. Sch. Med., Niigata.

Effects of intraventricular (ivt) injec-
tions (10 pl) of antagonists on plasma VP
(pVP) responses to agonists applied ivt (10

1) and osmotic stimuli were examined in
conscious rats. VP was measured by RIA.
Dopamine (DA), an alpha-adrenergic agonist
phenylephrine (PHE) (150 or 750 nmol) and
a cholinergic agonist carbachol (CB; 1.4
nmol ) augmented pVP 1.5 and 5 min later.
A beta-agonist isoproterenol (150 nmol) was
without effect. The effect of 750 nmol DA
was prevented by either 150 nmol haloperi-
dol (HAL), SCH 23390 (SCH; a D1 receptor
antagonist) or sulpiride (SUL; a D2 antago-
nist), given 10 or 40 min before. The VP
response to 750 nmol PHE was blocked by an
alpha-antagonist phenoxybenzamine (POB; 150
nmol) given 10 min before. The CB-induced
VP response was abolished not by a nicotin-
ic blocker hexamethonium (HEX), but by a
muscarinic blocker atropine (ATR)(28 nmol).
The increase in pVP caused by a hypertonic
solution (990 mOsm/kg; 10 pl) applied ivt
was not affected by POB, HEX or ATR,
whereas it was blocked by HAL. SCH, SUL and
HAL (150 nmol), when given 10 min before
the intraperitoneal injection (2 ml1/100g)
of 600 mM saline, inhibited rises in pVP 15
or 30 min after the osmotic load. These
results suggest that osmotic VP secretion
may be mediated or modulated by periven-
tricular dopaminergic neurons.

EN 74

LOCALIZATION OF VASOTOCIN-LIKE
IMMUNOREACTIVE CELLS AND FIBERS IN
TELENCEPHALON AND DIENCEPHALON OF ZEBRA
FINCHES (Poephila guttata) AND BENGALESE
FINCHES (Lonchura domestica).

T.Kimura’, K.OkKanoya’’, and K.Aoki’*. “Life
Science Institute, Sophia University. *Bird
Control Laboratory, National Agriculture
Research Center, Tsukuba.

We examined the localization of
vasotocin-like immunoreactive (ir-VT)
cells and fibers in the brain of adult
Zebra and Bengalese finches with immuno-
histochemical techniques. Cell bodies were
located in the superchiasmatic nucleus,
preoptic area, paraventricular nucleus
(PVN), superoptic nucleus (SON), and
ventro-lateral of hypothalamus. Fibers
emerging from the PVN run toward lateral
hypothalamus and redirected toward ventral
hypothalamus along the lateral forebrain
bundle. Axons from the PVN and SON formed
tight bundles of the hypothalamo-
hypophysial tract in the lateral
hypothalamus and terminated into the
medium eminence. Some fibers from the PVN
and SON run toward telencephalic region.
The overall organization of the ir-VT
system in Zebra finch and Bengalese finch
hypothalamus was similar to that described
in the canary. In telencephalon, Ir-VT
fibers were located around vocal control
nuclei, RA (Robust nucleus of arch-
striatum) and HVC (Higher vocal center) of
Bengalese finches and RA of Zebra finches.

1148 Endocrinology

EN 75

ANGIOTENSIN II AND DRINKING INVERTEBRATES.
H. Kobayashi. Res. Lab., Zenyaku Kogyo,
Co., Ltd. Nerima, Tokyo.

Drinking was observed for l hr after
single ip injection of angiotensin II (ANG
II). Three of 11 mammalian species (Mongo-

lian gerbil, mouse, hamster) originating
in dry area and three herbivorous’ species
(hare, rabbit, guinea pig) did not

respond to ANG II by drinking. Five other
species (flying fox, rat, cat, dog, chip-
munk) responded to ANG II (10-50pg /100g).
Most of 42 avian species investigated
responded to ANG II (1-10ng/100g).
However, 4 species originating in dry area

such as-cockatiel, some species of parrots
and 3 carnivorous species (shikra,
kestrel, owl) did not respond. In

reptiles, 2 of 7 lacertilian species and 3
of 5 ophidian species did not respond.
Relationship between their habitats or
habits and responsiveness to ANG II is not
clear at present. In amphibians, none of
3 uroderan and 8 anuran species
responded. In fishes, those exclusively
inhabiting sea water (11 species) or fresh
water (10 species) did not respond. One
species of hagfish, dogfish and shark each
was not responsive. However, teleosts
inhabiting brackish water (7 species) and
teleosts surviving in diluted sea water (6
species) were responsive’to ANG II (1-10
ug/100g). Thus, responsiveness of animals
to ANG II in drinking varies with their
habitats or habits.

EN 76

IMMUNOREACTIVE ENDOTHELIN IN BLOOD PLASMA
OF NON-MAMMALIAN VERTEBRATES.
H.Uemura!, T.Hirohama!, T.Aoto!, M. Naruse2
and K.Naruse2. '!Biol. Lab., Kanagawa Dent.
Coll., Yokosuka, 2Dept. of Med., Tokyo
Women's Medical Coll., Tokyo.

Immunoreactive endothelin (ir-ET) in
plasma was studied in a representative
animal from each class of non-mammalian
vertebrates. Blood samples were collected
individually from the unanesthetized
animal. Plasma ET was extracted by Amprep
C2 column (Amersham) and was determined by
radioimmunoassay of ET-1. Plasma levels of
ir-ET were 3.3 + 0.6 pg/ml(n=8) in Coturnix
ohewusy¢ HleyerouaiCcei, Oo a OSG jae//imil (a=s))) alin
Trionyx sinensis japonica, 6.4 + 0.8 pg/ml
(n=5) in Rana catesbeiana, 9.6 + 1.4 pg/ml
(n=/) en Cy prinusmecanrpioy 3)604I 08s poyaml
(n=3) in Conger myriaster, 4.3 + 0.9 pg/ml
(ni=5)) ene histalkis mus cylultamancl m0 jw. iu Oy
pg/ml (n=6) in Eptatretus burberi. The
dilution curves of the plasma extracts from
each species almost paralleled the standard
CUISViC Ne ROG Elite Analysis of the carp
plasma extracts by reverse-high performance
liquid chromatography, using a column of
Capcell Pack 5C;g-SG 3000 (Shiseido),
revealed that ir-ET consisted of three
components, the dominant peak being located
at the elution position of synthetic ET-1.
The present results demonstrate clearly
that the ET-1-like substance circulates in
blood, suggesting the hormonal function in
non-mammalian vertebrates.

ENG?

CO-LOCALIZATION OF IMMUNOREACTIVE ENDO-
THELIN AND NEUROHYPOPHYSEAL HORMONES IN
THE RAT NEUROHYPOPHYSIS.

S.Nakamura, Y.Kasuya, T.Aoto and H.Uemura
Biol. Lab. Kanagawa Dental Coll., Yokosuka.

Localization of immunoreactive endo-
thelin (ir-ET) and arginine vasopressin
(ixr-AVP) in the rat neurohypophysis was
investigated immunohisto- and immunocyto-
chemically by employing the antisera
raised against endothelin-1 and AVP. Ir-ET
was detected in the neurohypophysis by
avidin-biotin-peroxidase complex method.
A double immunogold technique revealed
three types of neurosecretory granules in
the axons: the one containing ir-ET, the
one containing ir-AVP and the one
containing both ir-ET and ir-AVP. In the
axons containing ir-ET and ir-AVP, AVP
immunoreactivity was abolished by the pre-
absorption test completely with arginine
vasotocin but slightly by that with
oxytocin. The number of neurosecretory
granules decreased following 3-day dehyd-
ration, although all of the three types of
granules persisted. These results suggest
that ET and AVP coexist in the same
secretory granules and that the two
peptides may be secreted by dehydration.

EN 78

COEXISTENCE OF TWO NATRIURETIC PEPTIDES IN
THE HEARTS OF VERTEBRATES.
T. Hirohama, Y. Kasuya, H. Uemura, and
T. Aoto. Biological Laboratory, Kanagawa
Dental College, Yokosuka.

Localization of atrial natriuretic pep-
tide (ANP)- and brain natriuretic peptide
(BNP)-like immunoreactivity was studied in
the heart of the rat and certain lower
vertebrates by use of the avidin-biotin-
peroxidase complex (ABC) method and the
post-embedding double protein A-gold
technique, employing the antisera absorbed
by heterologous antigens: human ANP anti-
serum by synthetic porcine BNP, and
porcine BNP antiserum by synthetic human
ANP or rat ANP. All antisera and antigens
were obtained from Peptide Institute
(Osaka). Immunoreactive ANP (ir-ANP) and
BNP (ir-BNP) were shown by the ABC method
to coexist in most cardiocytes of the
atrial wall and some ventricular cardio-
cytes in the newt, Cynops pyrrhogaster.
The double protein A-gold technique,
applied to the atria of the rat and the
atria and ventricles of newt, Rana
catesbeiana and Cyprius carpio revealed
that both ir-ANP and ir-BNP are localized
within the same secretory granules in the
cardiocytes. These results suggest that
two kinds of natriuretic peptides are
secreted simultaneously and they may have
coordinated physiological roles.

Endocrinology 1149

EN 79

FMRFAMIDE-LIKE IMMUNOREACTIVITY IN THE
BRAIN OF THE CLOUDY DOGFISH.

A-@insisas S.Okal and VHOnnas Nese. of

Biol., Nippon Dent. Univ., Niigata and

2sado Mar. Biol. Stat., Fac. of Sci.,
Niigata Univ., Niigata

Distribution of a substance related to
the molluscan cardioexcitatory tetrapeptide
FMRFamide in the brain of the cloudy dog-
fish, Scyliorhinus torazame, was examined
by immunocytochemistry. Extensive
FMRFamide-like immunoreactivity was demon-
strated in various regions of the dogfish
brain, except in the cerebellar corpus.
Immunoreactive perikarya were located
chiefly in the proximal ganglion of the
terminal nerve, the preoptic area, the
hypothalamic periventricular gray consist-—
ing of the nucleus medius hypothalamicus,
the nucleus lateralis tuberis, and the
nucleus lobi lateralis. Some of the immuno-
reactive cells in the hypothalamus were
identified as cerebrospinal fluid-
contacting neurons. Immunoreactive fibers
derived from the terminal nerve were
pursued towards the hypothalamus. Labelled
varicose fibers were abundant in the basal
part of the telencephalon and the hypo-
thalamus. Some of the labelled fibers
terminated in the pars neurointermedia
through the median eminence. These results
suggest that in the dogfish, FMRFamide-
like substance may participate in the
regulation of the adenohypophysial
function.

EN 80

DISTRIBUTION OF FMRFamide-LIKE IMMUNO-
REACTIVITY IN THE BRAIN OF SEVERAL SPECIES
OF VERTEBRATES.

K. Fujii and H. Kobayashi, Research
Laboratory, Zenyaku Kogyo Co., Ltd., Tokyo.

Distribution of FMRFamide-like immuno-
reactivity in the brain of several species
of vertebrates (hagfish, lamprey, goldfish,
eel, newt, quail) was studied by means of
the PAP method. Cells showing FMRFamide-
like immunoreactivity were localized in the
hypothalamus of each species especially
behind the group of AF-positive cells.
Fibers showing FMRFamide-like immuno-
reactivity were found in the ventral
telencephalon, thalamus, hypothalamus,
midbrain, medulla oblongata, and spinal
cord, in general. These fibers were also
found in the median eminence of all the
species except for the hagfish. In the
goldfish and the eel, fibers with
FMRFamide-like immunoreactivity existed in
the saccus vasculosus. Immunoreactive
fibers were abundant, especially in the
cerebellum of the newt. FMRFamide-like
peptide may play a role as neurohormone,
neurotransmitter or neuromodulator in the
vertebrate brain.

EN 81

LOCALIZATION OF ENDOTHELIN-LIKE IMUNOREAC-
TIVITY IN LOWER VERTEBRATES AND INVER-
TEBRATES

-Kasuya , H.Kobayashi2 and H.Uemura!
Biological laboratory, Kanagawa Dental
College, Kanagawa, Research Laboratory,

Zenyaku Kogyo Co., Ltd., Tokyo

The localization of endothelin (ET)-
like immunoreactivity was found in two
species of vertebrates and six species of
invertebrates, by employing the avidin-
biotin complex (ABC) method with antiserum
against synthetic ET-1. Specificity of the
reaction was tested by the preabsorption
test. In the central nervous system, im-
munoreactive perikarya and/or nerve fibers
were observed in the nereid Neanthes
diversicolor, mussel Mytilus edulis, slug
Limax marginatus, ramshorn snail Indo-
planorbis exustus, cricket Gryllus
bimaculatus, and sea-squirt Ciona intesti-
nalis. In the medaka, Oryzias latipes,
immunoreactive ET was found in the
neurohypophyseal and urophyseal systems.
Immunoreactive cells were also found in
the mucous iqiand sof, sehewstiug 7 3 nw chic
adenohypophysis of the lamprey, Lampetra
japonica, and in the branchia and head
kidney of the medaka. These results indi-
cate that ET-like substances are dis-
tributed in the lower vertebrates and a
wide range of invertebrates, and provide
evidence for the case that ET found in
mammals has a long evolutionary history.

EN 82

IMMNOHTSTOCHEMICAL ANALYSIS OF THE EARLY DEVELOPMENT OF THE
CAUDAL NEUROSECRETORY SYSTEM IN THE CHUM SALMON

S.Oka! ,Y.Honma? ,T. Iwanaga*,and T.Fujita?. 'Dept. of Biol.,
Nippon Dent.Univ. Niigata, *Sado Mar.Biol.Stat.,Fac.of Sci.,
Niigata Univ. ,Niigata,*Dept.of Anat. ,Niigata Univ.School of
Med. ,Niigata.

Because little is known of the development of the caudal
neurosecretory system(Sano et al.,1959,1962; Honma and Tamura,
1967), specific antisera raised against urotensins(U) I andII,
were used for immunohistochemical examminations of artificially
reared embryos, larvae, and juveniles of chum salmon
(Oncorhynchus keta). Immunoreactivities for UI and II were
first demonstrated in 14 day-old, post-hatching larvae,
indicating labelled, spinal cord perikarya and fibers, from
which the future caudal neurosecretory system would develop.
However, histologically distinct neurohemal organs had not
developed by this stage. Thereafter, immunoreactive perikarya
and fibers gradually increased, and a typical urophysis,
comparable to that of adults appeard 5 months after hatching.
At this time weak immunoreactivities for UI were also demon-
strated in the neurohypophysis.

1150 Endocrinology

EN 83

EFFECTS OF CALCEMIC HORMONES ON IONIZED
CALCIUM IN THE BULLFROG,

C. Oguro, T. Matsuba and Y. Sasayama
Dept. of Biol., Fac. of Sci., Toyama

Univ. , Toyama.
Much has been remained to be clarified

on the role of calcitonin (CT) on serum Ca
regulation especially on ionized Ca in
lower vertebrates. In the present study,
ultimobranchialectomy (UBX), partial para-
thyroidectomy (PPTX) and their combimation
(UBX-PPTX) were made in order to know the
effects of these treatment on serum total
and ionized Ca concentration in the bull-
frogs. Furthermore, effects of salmon CT
or ultimobranchial extracts combined with
Operations above noted, were examined.

In Exp. |. Blood were taken from the is-
chiadic artery through a cannula before
UBX, PPTX or UBX+PPTX and 14 days after
the operation. In Exp. I. Blood was taken
before UBX or UBX+PPTX and then ultimo-
branchial gland extract (equivalent to 2
individuals), salmon CT (300 MRC mU) or
muscle extract as a control was adminis-—
trated. One, 3 and 5 hr after the adminis—
tration, serum total Ca, ionized Ca, total
Na, ionized Na and pH were determined. In
Exp. I, total serum Ca concentration was
decreased in PPTX and UBX+PPTX groups. On
the other hand, ionized Ca was decreased
only in UBX+PPTX group. In Exp. I, serum
total and ionized Ca concentrations were
decreased in UBX and UBX+PPTX groups re-
ceived salmon CT or ultimobranchial gland
extract, but not by muscle: extract.

It is coneluded that ultimobranchial
gland, hence CT, plays a role on serum to-
tal Ca concentration, may be through ion-
ized Ca control.

EN 84

EFFECTS OF CALCITONIN ON THE HEPATIC CELL
IN THE BULLFROG.

C. Oguro, S. Asano, Y. Sasayama and H.
Yoshizawa?.

Dept. of Biol., Fae. of Sci., Toyama
Univ., Toyama. ‘Dept. of Oral Histol.,
Matsumoto Dental Coll., Shiojiri.

It has been known that target sites of
calcitonin(CT) are the bone, kidney and
intestine in mammals. In young animals, ad-
ministration of CT induces hypocalcemia as
a systemic effect. Furthermore, it was re-
ported that mitochondria in the hepatic
cell are involving in this hypocalcemia in
rats. In the present study, effects of CT
on the respiration of the liver and hepat-
ic mitochondria have been studied using
the bullfrog. Hepatic lobe was homogenized
and mitochondria were isolated. Mitochon-
dria were incubated with a solution(0. 1M
sucrose, 20mM KCl, 3mM NazHPO., 3mM MgClz.,
5mM Tris-HCl buffer) activity was checked
by adding dinitropenol and sodium succi-
nate. Respiration of mitochondria was en-
hanced by adding either of 125, 150, 175
or 200 nmol Ca in parallel to the increase
of Ca added. No effects were observed in
the respiration with synthetic salmon CT
(300mU) or extracts of an ultimobranchial
gland. CT in these concentrations gives no
effect on the respiration in the skeletal
muscle or pieces of liver. On the other
hand, the respiration decreases by 40%
with very high concentration of CT(3X107’M
/mg mitochondrial protein), as was found
in rats.

It may be concluded from these results
that CT is not a major factor controling
the respiration of mitochondria in the
bullfrog liver.

EN 85

THE SERUM MINERAL CONCENTRATIONS IN THE
EELS SUFFERING FROM SPINAL CURVATURE

Ve Sasayamal, Ve Nosel, (Ge Ogurol and
M. NakataZ: Irac. of Scei-,; Toyama Univ.,
Toyama, 2Nisshin Flour Milling Co., Ltd.

It is reported that about 5 &% indivi-
duals of the cultured eels fall into a

disease of spinal curvature. However,
the cause has not been clarified so far.
We examined the serum mineral levels of
the diseased eels, and compared with those

of normal fish. Furthermore, histology
of the ultimobranchial glands and the
corpuscles of Stannius, which are the

main endocrine organs for the hypocalcemic
hormones waS examined. The samplings were
performed 4 times at May 30, July 4, August
20 and September 11, in 1990. There were
no differences in Ca, Pi and Na levels
between the diseased eels and normal ones.
However, Mg value was significantly higher
on May, but significantly lower on August
and September in the former. Serum K levels
were always higher in the former, although
the differences were statistically not

significant. The implications of these
phenomena are not clear at present. On
the other hand, no obvious difference

in the histology of the ultimobranchial
gland and corpuscles of Stannius was recog-
nized between those 2 groups. This result
suggests a possibility that those endocrine
glands are not involuving in the develop-
ment of the spinal curvature.

EN 86

EFFECTS OF THE STANNIUS CORPUSCLE HOMOGE-
NATES ON THE SERUM MINERAL LEVELS IN RATS
Y. Sasayama, K. Ukawa and C. Oguro

Fac. of Sci. Toyama Univ., Toyama

The corpuscles of Stannius (CS) which
secrete hypocalcemic hormone, are present
only in the holostean and teleostean fish-
es. It has been reported by some authors
that CS homogenates bring about calcemic
effect also in frogs, birds and rats.
On the other hand, recently, it was known
that the CS shows’ immunohistochemically
the positive reaction to the antiserum
for the parathyroid hormone-related protein
(PTHrP) which is responsible for humoral
hypercalcemia of malignancy. Therefore,
it is suggested that the calcemic action
evoked by the CS homogenates in other
animal groups may be due to PTHrP. In
the present study, effects of CS homoge-
nates of the fresh-water fishes (goldfish
and eel) and sea-water fish (wrasse) on
the serum Ca, Mg, Pi, Na and K levels
in rats were examined. Initially, it was
confirmed that CS homogenates were effec-
tive on lowering the serum Ca level, when
administered to the goldfish, implying
that the calcemic factor is certainly
present in the homogenates. In the rats,
however, there were no significant changes
in those mineral levels. These results
suggest that some conditions may be neces-
sary, for CS homogenates to exhibit the
calcemic action in rats.

Endocrinology WISI

EN 87

CHANGES OF CALCIUM METABOLISM DURING AGING
OF WISTAR/TW FEMALE RATS.

Y.Kobayashi!, K.Shimoura!, Y.Tanabe* and
K.Hattori!. ‘Dept. of Pharmacol. and ?Cent.
Res. Lab., Shimane Med. Univ., Izumo.

The Wistar/Tw strain female rats show
polyuria after 20 months of age (Mo).
Morphological changes in renal glomeruli
were observed. An interaction between
renal function and calcium metabolism is
well known. In the present study, urinary
and serum calcium and phosphate concen-
tration were measured in rats at 4, 7, 10,
13, 15, 18 and 21 Mo. Urine was collected

by aluminum metabolic cages. Urinary
volume increased 2-fold at 21 Mo compared
Waiithe that vat) 7) 9Mo). Urinary protein

concentration began to increase after 13
Mo. and it was significant at 18 Mo.

Urine concentrating rate began to decrease
after 10 Mo and it was significant at 21
Mo. Urinary calcium concentrations were
increased significantly at 13 Mo and it
showed 5-fold at 21 Mo. Daily excretion
of urinary calcium increased with aging and

it was 10-fold at 21 Mo. Serum calcium
concentration increased slightly, but
Significantly. Urinary and serum phos-

phate concentrations were not changed
during aging, but daily excretion of
urinary phosphate was increased at 21 Mo.
In conclusion, changes in calcium metabo-
lism during aging of Wistar/Tw female rats
were observed.

EN 88

EFFECT OF MASTOPARAN ON SIGNAL-
TRANSDUCTION COUPLES TO PHOSPHOLIPASE C
IN RAT PAROTID CELL

A. Tanimura Dept. of Dent. Pharmacol.,
Sch. of Dentistry, Higashi-Nippon-Gakuen
Univ., Ishikari-Tobetsu, Hokkaido

eS

The effects of mastoparan on the Ca**
mobilization and phosphoinositide
hydrolysis in dispersed rat parotid cells
were examined. At concentrations higher
than 10pM, mastoparay rapidly increased
intrg¢ellular Ca concentration
( [CE M lst) atyoy 1 dose-dependent Manner, and
the increased [Ca Ji did not decline for
at least 10 min. In the absence of
extracellular Ca’, mastoparan evoked a
rapid and transient increase in [Ca i.
The accumulation of inositol bisphosphate
(IP,) and inositol trisphoshate (IP.,)
were induced by mastoparan in both intatt
and saponin permeabilized cells. The
preincubation (5 min) of permeabilized
cells with GDP-B-S (100pM), an inhibitor
for GTP-binding protein (G-protein),
decreased mastoparan-induced accumula-
tion of IP, and IP3. These results
suggest that the mastoparan-induced
phosphginositide hydrolysis resulting in
the Ca~*-mobilization is mediated though
the activation of G-protein couples to
phospholipase C.

EN 89

SOME FACTORS INFLUENCING ON PLASMA MAGNESIUM LEVEL IN
SNAKES.

M.Yoshiharal, M.Uchiyamal, T.Murakamil, H.Yoshizawa2,
and C.Oguro3. IpPept. of Oral Physiol., The Nippon Den-
tal Univ., Niigata, 2Dept. of Oral Histol., Matsumoto
Dental Coll., Shiojiri, 3Dept. of Biol., Toyama Univ.,
Toyama.

It has been known that magnesium is important in
biological systems such as intervention in enzymatic
systems, membrane role and function in neuromuscular
excitability. In mammals, parathyroid hormone, 1,25-
dihydroxy-vitamin D and calcitonin seem to be major
hormones involving in magnesium metabolism. However,
very few report is available on the role of hormones in
reptilian magnesium metabolism.

In the present study, changes of plasma magnesium
levels during ontogenetic development and effects of
parathyroidectomy and ultimobranchialectomy on plasma
magnesium level were examined in the rat snake, Elaphe
quadrivirgata.

Plasma magnesium level was 2.0 mg/100 ml two weeks
before hatching and this level was maintained to three
days after hatching. Then the magnesium level began to
increase and reached 4.1 mg/100 ml two weeks after
hatching. This is same to that in adult snakes. Effects
of parathyroidectomy and ultimobranchialectomy on
plasma magnesium level were examined immediately after
hatching. Two weeks after the removal of the
parathyroid gland, plasma magnesium level was sig-
nificantly lower than that in the sham-operated hatchl-
ings. On the other hand, plasma magnesium level in the
ultimobranchialectomized hatchlings did not show sig-
nificant changes.

EN 90

STIMULATION OF °°S-SULFATE UPTAKE BY
INSULIN-LIKE GROWTH FACTOR I (IGF-I) AND
INSULIN IN CULTURED CARTILAGE OF THE EEL.
C. Duan and T. Hirano, Ocean Res. Inst.,
Univ. Tokyo, Nakano, Tokyo 164.

We have shown that sulfate uptake by
branchial cartilage of the eel (Anguilla
japonica) is under indirect stimulatory
control by growth hormone via somatomedin-
like plasma factor(s). In this study,
effects of IGF-I, II and insulin on sulfate
uptake were examined. Addition of bovine
IGF-I to the culture medium significantly
stimulated the uptake in a dose-dependent
manner (2.5 to 250 ng/ml). Rat IGF-II was
much less potent than bovine IGF-I.
Addition of bovine or human insulin, on the
other hand, stimulated the sulfate uptake
dose-dependently from 1 to 250 ng/ml.
Salmon insulin was equally effective as
Mammalian insulins. To further examine the
possible role of insulin, homologous eel
insulin was isolated and characterized. Its
amino acid sequence shows much greater
homology to other insulins than to IGF-Is.
Nanogram concentrations of eel insulin,
like bovine IGF-I and insulin, also
stimulated the sulfate uptake by eel
cartilage. These results suggest that not
only IGF-I but also insulin are likely to
be involved in the regulation of cartilage
matrix synthesis in the eel.

1152 Endocrinology

EN 91

Does bombyxin have the insulin-like
physiological function in Bombyx mori?
M.Masumura, H.Saegusa, A.Mizoguchi,
H.Ishizaki, H.Nagasawa, A.Suzuki.
Dept. of Biol., Sch. of Sci., Nagoya Univ.,
Nagoya, Dept. of Agr. Chem., Fac. of Agr.,
Univ. of Tokyo., Tokyo,

Bombyxin, a brain peptide of the silkmoth
Bombyx mori, has the prothoracicotropic
activity on the saturniid moth Samia
Cynith va naicaniy, seb uUltuei Sup niy,SHsoog ac ap
function in Bombyx is still unclarified.
Because of the considerable homology in the
amino acid sequence with vertebrate
insulin, we attempted to study the role of
bombyxin on the sugar metabolism. First, we
investigated the quantitative changes of
bombyxin in Bombyx larval hemolymph and
brain during fasting and re-feeding. On
fasting, bombyxin accumulated in the brain
and its hemolymph titer fell. After re-
feeding, bombyxin was released from the
brain in the first hour. The changes of
the hemolymph glucose level during fasting
and re-feeding correlated to those of
bombyxin. As glucose injection stimulated
the release of bombyxin, it is thought that
glucose regulates the release of bombyxin.
Further, we investigated the effect of
bombyxin on the hemolymph sugar level.
Injection of bombyxin lowered the trehalose
level but did not the glucose level. The
hemolymph trehalose levels of bombyxin-
injected larvae and pupae decreased
approximately by 20% compared to those of
control animals.

EN 92

ANTIMETAMORPHIC ACTIONS OF PROLACTIN AND
SEX STEROIDS IN THE FLOUNDER.

de Jesus’, Y. Inui“ and T. Hiranot.
Ocegn Res. Inst., Univ. of Tokyo, Tokyo,
and “Natl. Res. Inst. of Aquaculture,
Mie.

The effects of ovine prolactin
(OPRL) and sex steroids, testosterone and
estradiol, on the metamorphosis of the
Japanese flounder, Paralichthys oliva-
ceus, were investigated. All three hor-
mones attenuated the stimulatory effects
of T3 (0.001 pg/ml) on the resorption of
the isolated dorsal fin rays in vitro,
when administered together, while having
no effect when given alone. Five intra-
muscular injections of OPRL or either sex
steroid (0.05 ypg/fish, every 3 days) into
metamorphosing flounder larvae delayed
the resorption of the dorsal fin rays
relative to the saline-injected control.
The degree of eye migration, however, was
not significantly affected. Thyroid
hormone levels in fish sampled at the end
of the experiment showed no significant
difference among the T4 levels of the
various groups. On the other hand, T3
level in oPRL treated fish was signifi-
cantly higher than in the control fish,
whereas the T3 levels of the estradiol-
or testosterone-injected fish tended to
be lower. It is not known whether PRL
and the sex steroids are acting through
similar or divergent mechanisms.

ENG9S

DEVELOPMENT OF THYROID STIMULATING HORMONE
BIOASSAY USING LOWER JAW OF MEDAKA.
M.Tagawal, D.K.Okimoto2, Y.Koide?,
E.G.Grau2 and T.Hirano!. !Ocean Res.
Inst., Univ. of Tokyo, Tokyo. #Hawaii
Inst. Marine Biol., Univ. of Hawaii,
Honolulu, %School of Fish. Sci., Kitasato

Although thyroid hormones are known
to be involved in some aspects of fish
development, there are few reports on con-
trol mechanisms of the hormone secretion,
mainly due to the absence of homologous
thyroid stimulating hormone (TSH) in
fishes. In order to purify fish TSH, an
in vitro bioassay method was established
using medaka (Oryzias latipes). Test
samples were added to the incubation media
of the lower jaw, where thyroid follicles
are located. The medium was collected
after 3 hours, and released thyroxine (T4)
was measured using specific radioimmuno-
assay. The response of the lower jaw of
mature females was greater than those of
immature fish and mature males. A linear
log-dose relationship was observed between
0.3-3 mIU/ml of bovine TSH [specific ac-
tivity (SA) = 0.5-1.0 IU/mg]. The effects
were parabolic, the higher doses being
less effective than the lower doses.
Gonadotropin I, II, growth hormone and
prolactin purified from chum salmon
pituitary did not stimulate T4 release. A
crude ethanol extract of the chum salmon
pituitary revealed potent T4-releasing ac-
tivity (SA = 0.003 IU/mg).

EN 94

EFFECT OF T4 ON THE ACTIVITY ON LAND OF
THE MUDSKIPPER, PERIOPHTHALMUS CANTONENSIS
K. IWATA, Y.IKEBE,J.HAYASHI and M. YAMAZAKI
Biol.Labo. Fac.of Edu. Wakayama Univ.,
Wakayama

In the present study, to elucidate the
role of thyroxin in the activity on land of
the mudskipper in natural conditions, the
landing activities were compared using the
fish captured monthly during breeding sea-
son (May to August) under 15L-9D cycles at
25°C. The results from the preliminary
survey revealed that the height of follicle
epithelium of thyroid gland showed the max-
imum at early breeding season (June) and
gradually decreased in late breeding season
(August). Before breeding season (early in
May), the landing activity (frequency of in
and out of water) was quite low, and the
mean sojourn time on land was very long (3
hrs.). In the middle of breeding season,
the landing activity of both sexes greatly
increased, and the mean sojourn time on
land became very short (14 min.). When the
fish was immersed in Thiourea(30 ppm) solu-
tion for 10 days, the landing activity
markedly decreased, and the mean sojourn
time on land increased. The landing acti-
vVity oof the fish’ Mimmersed = inten. @Z™ ppm)
solution for 10 days was not much defferent
from that of control, but showed a clear
diurnal change. These results suggest that
thyroxin causes an increase in locomotor
activity.

Endocrinology 1153

EN 95

INITIATION OF THYROGLOBULIN SYNTHESIS IN
THE DEVELOPMENT OF BULLFROG THYROID
K. Fujikura’!, Y. Ohmiya2 and S. Suzuki!
1Tnst. of Endocrinol., Gunma Univ.,
Maebashi and #Inst. of Osaka Bioscience,
Suita, Osaka.

By immunoblotting analysis using an
anti-bullfrog thyroglobulin (TG) antibody
the biosynthesis of TG was examined during
thyroid development of the bullfrog, Rana
catesbeiana. Thyroid anlages or glands
from developmental stage 22 to early
stages of metamorphosing tadpoles were
collected and applied to 5% polyacrylamide
gel electrophoresis. After transferring
the gel to a polyvinylidene difluoride
membrane, the samples were treated with
anti-TG antibody and 1257T-protein A.
Autoradiography showed that TG was found
first in the thyroid anlages of stage 24.
At this stage two lobes were formed from a
heart-shaped thyroid anlage and PAS-
positive materials were observed in the
intracellular spaces of these lobes.
Thyroidal cells containing the
intracellular materials were also stained
immunohistochemically by the peroxidase
antiperoxidase method. These results seem
to show that synthesis of TG as a
prothyroid hormone begins at stage 24
Simultaneously with the formation of the
intracellular spaces in the thyroid
anlage.

EN 96

SERUM THYROGLOBULIN IN METAMORPHOSING
TADPOLES
S. Suzuki, K. Fujikura and Y. Shimoda
Inst. of Endocrinol., Gunma Univ.,
Maebashi.

Serum thyroglobulin (TG) in metamorpho-
sing tadpoles and adult frogs of Rana
catesbeiana was measured by
radioimmunoassay, using purified frog-TG
and its antisera. The cross-reactivities
of the antisera with proteins of thyroid

glands and sera were examined by
immunoblotting on polyacrylamide gels.
Only one immunoreactive component,

corresponding to TG, was found in both
frog and tadpole sera. Serum TG increased
at the prometamorphic stage and reached
the highest level (477 ng/ml) at the
metamorphic climax. Thereafter, the
circulating TG decreased _ slowly. This
decrease was lowest in froglets. Serum TG
in adult frogs was 267 ng/ml. Such
changes in circulating TG during metamor-
phosis were similar to those in thyroid
hormones. These results suggest that the
serum TG level is closely correlated with
the secretion of thyroid hormones’ in
tadpoles and adult frogs.

EN 97

BULLFROG THYROGLOBULIN. PARTIAL AMINO ACID
SEQUENCE AND ONE POSSIBLE Ty RELEASING SITE.

H. Hayashi!, Y. Ohmiya2 and S. Suzuki. Department of
1Protein Chemistry and 3Comparative Endocrinology, Institute
of Endocrinology, Gunma University, Maebashi and 2Osaka
Bioscience Institute, Suita, Osaka

The thyroid hormone (T4) is synthesized from specific two di-
iodotyrosine (DIT) residues in the thyroglobulin (TG) molecule.
The "donor" DIT lose the di-iodophenoxy side chain group, and
convert into dGehydroalanine residue. At present, several "donor"
and “acceptor” tyrosine residues are identified or proposed in
mammalian TG, since the amino acid sequence of human, bovine
and rat TG are deduced from the DNA base sequence.

We isolated TG from bullfrog tadpole thyroid. The molecular
weight (660KD) and amino acid composition are rather similar to
those of mammalian TG. The frog TG was reduced and
carboxymethylated, then digested with lysyl endopeptidase,
which is known to be highly specific to cut the carboxyl! side of
lysine residue. Soluble fraction of the digest was first separated
by Mono Q ion exchange chromatography. The acidic peptides
were further separated by reverse-phase HPLC. The amino acid
sequences of 5 peptides (total 117 residues) were determined.
Four peptides have lysine at the C-termini as expected according
to the enzyme specificity. However, one peptide (27 residues)
has isoleucine at the C-terminus. We compared those with the
deduced sequence of bovine TG. The peptide which has
isoleucine at C-terminus seems to correspond with 103-129 of
bovine TG (60% homology). Tyrosine at 130 in bovine TG is
proposed to be one of the "donor" site. Present study shows
that this tyrosine should be “acceptor" on the assumption that
residue 130 in frog TG is also tyrosine. If this tyrosine residue is
“donor" site, the peptide bonds in both sides of the resultant
dehydroalanine should remain in the process of T4 release.

EN 98

EFFECT OF IMMUNONEUTRALIZATION OF BULLFROG
TSH ON THE T4 RELEASE
M.Sakail, S.Kikuyama!, Y.Hanaoka2,
S.Tanaka2 and H.Hayashi2

lDept. of Biol., Sch. of Educ., Waseda
Univ., Tokyo. 2Inst. of Endocrinol., Gunma
Univ., Maebashi

Thyroid-stimulating hormone (TSH) was
purified from the pituitary glands of
bullfrogs, Rana catesbeiana, by ethanol
precipitation, hydrophobic interaction and
anion exchange chromatography, and
affinity chromatography employing
monoclonal antibodies against bullfrog LH§
and FSH8. The TSH stimulated the in vitro
T4 release from the thyroid glands from
prometamorphic bullfrog tadpoles in a dose
dependent manner. Subunits (a and 8) of
TSH separated by reverse-phase HPLC after
dissociation had no T4-releasing activity.
TSH incubated with monoclonal antibody
(IgG) against a subunit of bullfrog
gonadotropin nullified T4-releasing
activity while TSH incubated with IgG from
normal mouse serum retained the activity.

1154 Endocrinology, Cytology and Morphology

EN 99

EFFECT OF PROLACTIN ON THE LUNG MATURATION
IN THE BULLFROG TADPOLE

M. Ohkawa and S. Kikuyama

Dept. of Biol., Sch. of Educ., Waseda
Univ., Tokyo 169.

The anuran lung begins to function at
advanced stages of metamorphosis. We have
previously reported that the amount of
phosphatidylcholine (PC), the main con-
stituent of surfactant lipids, in the lung
of bullfrog (Rana catesbeiana) tadpoles
markedly increases during climax. Experi-
ments were conducted in order to see
whether prolactin (PRL) and growth hormecne
(GH) of which levels are known to rise
during climax are involved in the increase
of PC in the lung. Stage 18 tadpoles
received injections of 25 ul antiserum
against bullfrog PRL (PRL/AS), antiserum
against bullfrog GH (GH/AS) or normal rab-
bit serum (NRS) every other day. When the
animals reached stage 24, the lung tissue
was subjected to phospholipid analysis by
extraction with chloroform and methanol
and separation by HPTLC. Average PC con-
tents of the lung from PRL/AS-treated tad-
poles and GH/AS-treated animals were 38
and 80 % of the value for NRS-injected
tadpoles. The results indicate that PRL,
and possibly GH, contribute to the lung
maturation occuring during metamorphic
climax. ,

EN 100

IMMUNOCYTOCHEMICAL STUDY OF TSH CELLS
BEFORE, DURING, AND AFTER METAMORPHOSIS IN
HYNOBIUS NIGRESCENS AND H. RETARDATUS
Kaoru Yamashita and Hisaaki Iwasawa

Biol. Inst., Niigata Univ., Niigata

TSH producing cells of Hynobius nigres-
cens and H. retardatus were identified by
the peroxidase-antiperoxidase method using
anti-human TSH#serum. Immunoreactive TSH
cells appeared at first at Usui and
Hamazaki's stage 45 (forlimb-buds stage).
With the advance of larval growth,
immunoreactive TSH cells increased in
number in the ventral region of the pars
distalis. At stage 63, the amount of
immunoreactive TSH cells in all of the
pars distalis was 18% in H. nigrescens and
27% in H. retardatus-. At stage 64, the
beginning of metamorphosis, the immuno-
reaction was weaker in the central region
than in the peripheral region, and
immunoreactive TSH cells decreased in
number. At stage 66, the immunoreaction
was very weak. One month after metamor-
phosis, some immunoreactive TSH cells with
intense immunoreactivity appeared again,
and their number increased rapidly
thereafter.

CM 1

TERRESTRIAL DISTRIBUTION OF ACANTHAMOEBA
IN TOKYO.

T. Tsuruhara*, Y. Tomiyama*, Y. Hirukawa*, & K.
Ishii**

*Dept. Biology, Tokyo Gakugei Univ.,

**Lab. Biology, Hosei Univ.

Five cases of Acanthamoeba keratitis were recently
reported from soft contact lens wearer in Japan. This is
the first survey investigation of terrestrial distribution
of Acanthamoeba in Japan.

Eighty samples of the surface soil collected randomly
from campus of Tokyo Gakugei Univ. were inoculated
on NN-agar plates spreaded heat-treated Bacillus
subtilis, then were incubated at 30C for about 2 weeks
at which time they were examined for the presence of
cysts.

The cysts of Acanthamoeba was able to identified from
43 cases among 80 samples, made up as follows:

38 polyphagids, 4 astronyxids, and 1 culbertsonids.

Farthermore, comparative studies on fine structures

of their ostioles will be discussed.

CM 2

STUDIES OF FOUR SPECIES OF COCHLIOPODIUM
(TESTACIA) ISOLATED IN YAMAGUCHI PREFEC-—
TURE.

I. Yamaoka and K. Kitanaka. Biol. Inst.,
Fac. Sci., Yamaguchi Univ., Yamaguchi.

According to Bark (1973), there are
fourteen described species of
Cochliopodium. In 1881, Nagatani et al.
reported one species of Cochliopodium in
Japan, after then in 1986, Yamaoka and
Kunihiro isolated different species of it
in Yamaguchi city. In 1989 four species
of Cochliopodium were collected at
Yamaguchi prefecture. Two species of them
were identified as the same species
described above. Each four species was
difficult to classify in the locomotion
form and the characteristics in the
light microscopic morphology. In the
electron microscope 2 species of them
were Similar as the species described by
Bark (1973); C. bilimbosum and C. sp.3.

In the classification of amoeba
(Cochliopodium) the electron microscopic
morphology is of importance. Further, it
may be pointed out that the characteris-
tic scale formation is controlled geneti-
cally and contains some of the key to
resolve the problem in the self assembly
of molecules.

Cytology and Morphology 1155

CM 3

HYDROGENOSOME IN THE TERMITE INTESTINAL
PROTOZOA, TRICHONYMPHA AGILIS.

S. Koga, R. Murakami and I. Yamaoka.

Biol. Inst., Fac. Sci., Yamaguchi Univ.,
Yamaguchi.

T. agilis is the second largest type of
intestinal protozoa in the termite.
Yamaoka et al. (1984) reported that T.
agilis had the slightly oxygen consumptive
ability. In spite of it mitochondria could
not be observed in the cytoplasm of the
protozoa. Many granules which have similar
size to mitochondria were observed in the
Supra-nuclear region. Some cytochemical
reactions in the granule was examined, the

results were as follows; succinic
dehydrogenase "SDHase" ( +), cytochrom
oxydase (- ), Ca-ATPase (-), acid
phosphatase (- ), catarase (- ) and

hydrogenase (+ ). In the case of SDHase
the reaction products were detected in
both experiment and control groups, but
lower in the control. In the electron
microscope the envelope of the granule was
composed double unit membrane and was
adherent to each other. The invagination
like the mitochondrial cristae didn't be
observed.

The results suggest that the granules
may be a kind of the hydrogenosome.

CM 4

CONTROL OF MICRONUCLEAR DIVISION DURING
CONJUGATION OF PARAMECIUM CAUDATUM
K.Mikami, Res.Ins.for Sci.Educ., Miyagi
Univ. of Educ., Sendai.

When the macronucleus (mac) was removed
from a cell of a pair by micromanipulation
2 -2.5 hrs of conjugation (27 C), the germ
micronucleus (mic) in the operated cell
skipped over the crescent stage (meiotic
prophase) and divided twice (Mikami, 1988).
The result shows that the mac of its mate
provides a genetic factor which is trans-
ferable between mating cells and induces
mic division. We can say that the nuclear-
division-inducing factor acts even on a mic
which has not passed through a meiotic
prophase. In order to examine the
occurrence of a somatic nuclear division at
meiotic stage, mac and mic were removed
from a cell of a conjugating pair at a late
crescent stage and the mic was transplanted
from a cell at vegetative G1 phase to the
nucleus-free cell. Ten cells were trans-
planted with the mic. The mics disappered
in 4 cells but remained in 6 of them. In 5
of the latter 6 cells, mics divided by the
time when the mic finished two divisions of
meiosis in its mate with the mac. All of
the mics did not pass through the crescent
stage. The result shows that the cytoplasm
at the meiotic nuclear division induces
mitosis of the transplanted mic. The
intrinsic mechanism of nuclear division of
meiosis seems to be the same as that during
a vegetative phase.

CM 5

STUDIES ON THE DISPERSION OF MACRONUCLEAR
INCLUSION BODIES IN PARAMECIUM CAUDATUM.
Y. Tsukii! and K. Yazaki?. !Laboratory of Biology, Hosei
University, Chiyoda-ku, Tokyo 102, 2Tokyo Metropolitan
Institute of Medical Science, Bunkyo-ku, Tokyo 113.

Some strains of P. caudatum contain macronuclear in-
clusion bodies morphologically distinct from bacteria. Nu-
clear transplantation experiments demonstrated that they
are a new type of macronuclear endosymbionts.

Cells in culture medium mostly contain one or two large
inclusion bodies and some smaller ones within their macro-
nuclei. However, when the cells are transferred to a solu-
tion of 1 mM CaCl) and 1 mM Tris-HCl (pH 7.0), large in-
clusion bodies gradually break into numerous small ones
dispersing throughout macronuclei. They become smaller
and smaller in sizes, and finally disappear under photo-
microscope. SDS-PAGE of the macronuclei in which in-
clusion bodies had disappeared did not show a marked de-
crease of the protein components of inclusion bodies, sug-
gesting that inclusion bodies only disperse within a
macronucleus.

Under electron microscope, inclusion body consists of
tightly-packed microfibrils about 22-24 nm thick, and
chromatin-like fibers are also present within it. When cells
were transferred to the dispersion-inducing solution men-
tioned above, inclusion bodies swelled and their boundary
with the macronuclear chromatins became indistinguisha-
ble. Fibrillar structures gradually disintegrated in the mac-
ronucleus as the dispersion process proceeded.

CM 6
OBSERVATIONS ON THE MACRONUCLEUS IN THE
CILIATE EUPLOTES AEDICULATUS BY SCANNING

ELECTRON MICROSCOPE.
T. SUENAGA, Zool. Inst., Fac. of Sci., Hiroshima
Univ., Hiroshima

The feature of macronucleus of the ciliate
Euplotes aediculatus morphologically change during
asexual and sexual reproduction. In the present
paper, three-dimensional change of the macronu-
cleus (MA) was investigated by use of a scanning
electron microscope (SEM). After the cells were
lysed in the solution containing 0.5% Triton X-100,
the Ma were transferred to the washing solution
with pipette, and then they were fixed with
Parducz’s fixative. The specimens were dehydrat-
ed through a graded series of ethanol, critically
point dried, sputter coated with platinum and
examined. At Gl phase the Ma looked like C-
shape in ventral view, while the posterior part of
Ma slightly bent towards ventral surface in lateral
view. When the Ma were fractured, spherical
bodies of about 0.8 wm in diameter were visible.
They seems to be corresponded to the granules in
the Ma stained with acetic carmine. The surface
structure indicating the replication band (RB) at S
Phase was not detected on SEM specimens. At D
Phase the Ma was shortened to like a cucumber-
shape with smooth surface. The macronuclear
anlage in exconjugants could not be isolated by
the present method. But newly formed Ma and
old Ma fragments were successively isolated and
examined. The surfaces of these Ma resembled
that of the Ma at Gl phase.

1156 Cytology and Morphology

CM 7

ULTRASTRUCTURAL CHANGES IN NUCLEAR |
DIFFERENTIATION IN PARAMECIUM TETRAURELIA

Y.Yashima. Dept. of Biol.,Sch. of Lib.
Arts and Sci., Iwate Med.Univ., Morioka.

Ultrastructural changes associated with
nuclear differentiation were observed for
the conjugation induced between st.5l1 O
and E from 5-60 min after separation of
the conjugants using TEM. At telophase of
the second postzygotic mitosis, the nuclei
were found midway between their eventual
antero-posterior positions attained by 30
min after separation. The anterior became
micronuclei and the posterior became
macronuclear anlagen. At this stage the
diameters of the micronucleus and the
macronuclear anglage were the same (3 pm),
and the condensed chromatin in the
micronucleus resembled that in the
micronucleus in the vegetative stage; it
was not found in the macronuclear anlage.
Fine fibrillar material of low density
was present in the macronuclear anlage.
There were many disc-shaped and membranous
structures in the cytoplasm of the
posterior region around the nucleus,and
there were a small number of these
structures in the cytoplasm of the
anterior region.

These results suggest that nuclear
differentiation occurs in the 15-min
period between telophase in the second
postzygotic mitosis and the next resting
stage.

CM 8
REPLICATION BAND IN EUPLOTES
QCTOCARINATUS.

K. Sato. Dept.of Biol. Naruto Univ. of
Educ., Naruto.

In the S phase macronucleus of Euplotes
octocarinatus, replication band (RB) in
which DNA synthesis occurs, appears at
both tips of the rod-like nucleus, mi-
grates into the center, meets each other
and finally disappears. To examine the
appearance of RB and its movement, a part
of macronucleus or RB was removed with a
micropipette in E. octocarinatus. Removal
of tip or a part of the nuclei in the
cells of various G1 phases showed that RB
had a tendency to appear at the parts
which were nearer to both tips of the
macronucleus. While, if a RB was removed
from early or mid- S phase cells, the
remaining RB passed the center region
where two RBs usually meet each other, and
reached to the regions in which the anoth-
er RB was removed, or met the third RB
which appeared at the new broken end of
the nuclei. When the third RB appeared,
the RB didn't appear in the regions where
RB had already migrated. If these oper-
ated cells were isolated into fresh cul-
ture medium, most of them grew into
clones, and had normal macronuclei. These
suggest that the factors for the appear-
ance of RB are more in regions nearer to
the tips, that the signals for RB may
exist through the S phase of macronucleus,
and no special signals for the direction
of RB movement in this Euplotes.

CM 9

RECONSTRUCTION OF THE GOLGI COMPLEX IN AN
AMOEBA, COCHLIOPODIUM SP. (TESTACIA).

K. Ito, R. Murakami and I. Yamaoka

Biol. Inst., Fac. Sci., Yamaguchi Univ.,
Yamaguchi.

An amoeba Cochliopodium sp. (Testacia)
has a well developed Golgi complex in the
supra-nuclear region. It plays to form
the scales that cover the free surface of
the cell body. When the amoeba do to
encyst, the breakdown of the Golgi
complex takes place and whole cisternae
disappear, then the Golgi attachment
which is a characteristic structure in
this amoeba remains near the nucleus.
when the excyst was induced by the E.
coli extract, the Golgi cisternae were
reconstructed along the Golgi attachment.
Their processes were observed within
180 min after induction of excyst as
follows; adhesion of one cisterna to the
Golgi attachment, after then the number
of cisterna increase and complete the
reconstruction takes place at the inside
Wrapped up loosely by a paired Golgi
attachment. With the completion of the
stack the Golgi attachment side asso-
ciated with the nucleus. This result
suggests that the Golgi attachment may
plays an important role in the reconstruc-
tion of the Golgi complex.

CM 10

GROWTH OF THE CILIATE EUPLOTES ENCYSTICUS IN
BACTERIZED INORGANIC MEDIUM.

T. Takahashi and F. Yonezawa. Zool. Inst., Fac. of
Sci., Hiroshima Univ., Hiroshima

It has been reported that the encystment-—
inducing substance (EIS) is released into the cul-
ture medium by the bactivorous ciliate Euplotes
encysticus (Yonezawa, 1986). We are now trying
to isolate and purify the EIS. So far this ciliate
had been cultured with bacterized lettuce infusion
which is one of the most suitable culture medium
for bactivorous ciliates. This medium contains a
brownish unknown substance. Recently, we found
that the EIS could not be separated from the
brownish substance by gel filtration or ion ex-
change chromatography. Thus we tried to culture
the ciliate in bacterized inorganic medium. The
purpose of the present work is to determine the
optimum culture conditions for the growth of the
ciliate in bacterized inorganic medium. Osterhout
solution was selected out of 7 inorganic media
examined because it hardly affected the encystment
of the ciliate. The optimum inoculating size of
bacteria as food organisms was also examined.
When the ciliate was cultured with 50-fold diluted
Osterhout solution including about 1.0x10® bacte-
ria/ml at 23°C, the most stable growth of the cili-
ate was observed.

Cytology and Morphology 1157

CM 11

ENCYSTMENT-INDUCING SUBSTANCE IN CELL-FREE
FLUID FROM THE CULTURE OF THE CILIATE
EUPLOTES ENCYSTICUS GROWN IN BACTERIZED
INORGANIC MEDIUM.

F. Yonezawa and T. Takahashi. Zool. Inst.,
Fac. of Sci., Hiroshima Univ., Hiroshima.

It has been reported that encystment-
inducing substance (EIS) which is re-
leased into culture medium by the ciliate
Euplotes encysticus seems to be low mol-
eular weight protein (Yonezawa, 1989). So
far the ciliate was cultured with bacteri-
zed lettuce infusion for 4 days, and then
cell-free fluid (CFF) was obtained by re-
moving cells from the culture. This CFF
contains a brownish substance derived
from lettuce infusion. Recently, we found
that EIS could not be separated from the
brownish substance by ion exchange chroma-
tography. Thus, we tried to culture the
Ciliate in bacterized 50-fold diluted
Osterhout solution, and to isolate the EIS
from the obtained CFF. 80 liters of CFF
were filtered through sterilizing membrane
filters (0.45 wm pore size), and then was
treated by a batchwise with DEAE-Sepharose
CL-6B. The ion exchanger was then poured
into a column. Elution was done at 4 ©
by using a linear gradient of 0-0.5 M
NaCl. Fractions of 2.8 ml were collected.
When the A,,, of the eluate was monitored,
5 peaks appeared. The third peak contained
EIS. When the fractions of this peak were
analysed on SDS-PAGE, more than 30 bands
were detected. The problem as to which
band is EIS remains to be examined in
the future.

CM 12

ORAL DEVELOPMENT IN LH CELLS OF THE CILIATE
GLAUCOMA SCINTILLANS.

K. Matsumoto and M. Suhama. Zool. Inst.,
Fac. of Sci., Hiroshima Univ., Hiroshima

The left-handed form(LH) of Glaucoma
scintillans with the reversion of large
scale asymmetry of arrangement of con-
structive structures possesses anomalous
oral apparatuses (OAs) showing various
orientations. We investigated how these
abnormal OAs arose. The orientation of
OAs was judged by the tilt of membranelles
within OA. Of the mature OAs examined,
one third showed the orientation fundamen-
tally corresponding to that of OAs of
normal right-handed (RH) cells which
rotated 180° counterclockwise. In one
fifth the orientation appeared to agree
with that of normal OAs which rotated
270°. In one tenth the orientation was
mirror-image of that of normal OAs. In
one twentieth the orientation resembled
that of normal OAs. In the remaining OAs
the orientation was ambiguous. In addi-
tion, the developmental process of oral
primordia (OPs) in cell division and oral
replacement of LH cells were examined.
Many OPs at the late stage of development
possessed membranelles arranged in the
mirror-image configuration of normal OPs
of RH cell. The results suggest that OPs
rotate counterclockwise to various extent
at the final stage of development.

CM 13

REGULATION OF MIRROR-IMAGE SYMMETRY DOU-
BLETS DURING CONJUGANT FUSION IN’ THE
CILIATE STYLONYCHIA PUSTULATA.
J. Yano and M. Suhama. Zool. Inst., Fac.
of Sci., Hiroshima Univ., Hiroshima.
In mirror-image doublets of Stylonychia
ustulata, the left component possesses
ciliary organelles arranged in the normal
pattern, while the right component has
those arranged in mirror-image pattern.
In conjugant fusion between a CF singlet
cell having conjugant fusion inducible

factor and a mirror-image doublet, the
singlet fused with the right component of
the doublet side by side. Most of the

ciliary organelles in the fused area were
degraded. The oral apparatus(ROA) of the
original right component moved into dif-
ferent regions of the original doublet
during cell fusion. The ROA had an influ-
ence on the development of oral primordi-
um(OP) and other ciliary organelles in the
first cell division of the fused cell.
First, when the ROA was located in an
expected area of development of OP, forma-
tion of OP was prevented in that area.
Second, when the ROA came close to the OA
of the original left component, a long OP
and excess primordia of frontal-ventral-
transverse cirri were formed in the middle
region of original doublet. Third, when
the ROA was located in the middle region
of the fused cell, OPs and cirral primord-
ia were formed in both original CF singlet
and left component.

CM 14

REVERSIBILITY OF EXCYSTMENT IN A HYPOTRICH
CILIATE, HISTRICULUS CAVICOLA.

T. Nakamura, T. Yamashita and T. Matsusaka.
Dept. of Biol. Sci., Fac. of Sci., Kumamoto
Univ., Kumamoto.

The cysts, which were in the excystment
medium (EM) within 30 min and then trans-
ferred into Osterhout's solution (OS), did
not excyst. Preincubation in the EM longer
than 30 min resulted in some excystment and
the excystment rate increased as the pre-
incubation time increased. The preincuba-
tion in the EM longer than 90 min resulted
in normal excystment. During first 30 min
preincubation in the EM, the cysts formed
many kinetosomes, but disassembled them
after transfer into the OS, if the preincu-
bation was shorter than 30 min. This
result suggests that the kinetosome forma-
tion during excystment is a reversible
process.

The cysts, which were in the EM within
30 min and then transferred into the EM
containing 5 ug/ml cycloheximide, did not
excyst, but formed many kinetosomes during
30 min incubation in the medium as they did
in the normal EM. The result suggests
independency of kinetosome formation from
protein synthesis. When the preincubation
in the normal EM was between 40 to 70 min,
some cysts attained stage 3 of excystment
but did not excyst. Preincubation in the
normal EM longer than 115 min resulted in
normal excystment. These results suggest
that protein synthesis is essential for the
initiation and continuation of excystment.

1158 Cytology and Morphology

CM 15

STUDY ON THE ENDOPARASITE OF ANURA
3. PREVALENCE OF HELMINTH IN HYLA ARBOREA
JAPONICA.
Y. Sasaki and N.Makino. Dept. of Biology,
Tokyo Med. Coll.,Tokyo.

By the last time,we reported on prevalence of
helminth in Rana brevipoda porosa and Rana japonica
at Inzai, Chiba prefecture. Parasites were detected
from these hosts,were general helminth to frogs.But
a difference of prevalence of parasites was marked-
ly between R.b.p.and R.j.. This time, Hyla arborea
japonica was observed at the same field and com-
pared with two hosts of Rana.Species and numbers of
parasites in H.a.j.were fewer than in Rana.It seems
that the reason is caused by young hosts. Fourty
four male were collected and fourty twe female were
collected from June to August,1990. Parasites were
not seen in the lung and the urinary bladder. Nema-
todes;Oswaldocruzia sp.was observed in the duodenum
only one male. And interestingly cestodes;Sparganum
mansoni was observed in the coelom only one. This
Warm Was not seen in Rana since 1987.Nematodes; Cos-
mocerca japonica was detected low rate in the rec-
tum, but about R.b.p.and R.j.,rate of C.j.was shown
high prevalence. It seems that this difference is
influenced by ecological environment. However life
cycle of C.j.remains unexplained. At present they
are being raised under controled condition in the
laboratory.

CM 16

PHENOBARBITAL INJECTION INLEECHE,HIRUDO
MEDICINALIS ON BOTRYOIDAL TISSUE

H.Inamura

Dept. of Biol.,Tokyo Medical Coll.,Tokyo
The study of the structural and functinal
relationship within the three types granul-
ated botryoidal cells of the leeches is
continued by electron microscopy.

Previous studies was shown that nucleus of
a fed individual was surrounded with rough
endplasmic reticulum after the stavation
for 1 year,and on the basal parts of the
cells,endocytosis consisted of protuberanc-
es of the cells,on the apical parts of ones
,exocytosis were vascular lumen directly.
This study show that the cells fine struct-
ure in phenobarbital treated leeches did
through the use of the stavation leeches
for 1 year.The leeches received 10yg per
100mg by once injection in body.The medici-
ne dissolved in 0.65%saline water at conce-
ntratoin of 0.1%.Control leeches received
0.65%saline water injectoin.By the third
day,a lot of high density smooth endplasmic
reticulum were accumulation on the apical
parts of the cells of treatment.By the
fourth day,oil droplets(Llgranules) were a
lot on the basal parts of ones of treatment
-Both endocytosis and exocytosis shown
active.After treatment, glycogen particles
consisted of small types compared with fed
individual.The Golgi complex is to apical
parts of the cells.

It is supposed that the botryoidal cells
are related to regulation in metabolism.

CM 17

OBSERVATIONS OF THE PHARYNGEAL REGENERA-
TION OF THE LAND PLANARIANS, BIPALIUM
KEWENSE AND B.MULTILINEATUM.

Y.Shirasawa and N.Makino. Dept. of Biol.,
Tokyo Med. Coll., Tokyo.

The pharyngeal regeneration in the tran-
sected small pieces of two species of land
planarians have been studied morphologi-
cally and histologically. They often fis-
sion on the post-pharyngeal regions. The
pieces always regenerate well and form new
heads and pharynges within two weeks. Also
on cutting, the transected pieces show
Similar high abilities of regeneration,
with no regard to their body levels. The
histological preparations stained by
Gomori's aldehyde-fuchsin and H-E stains
revealed that many basophilic cells accu-
mulate in the parenchyma between the in-
testine and the ventral nerve cord(v.n.c.)
on the seventh day. Soon after the accumu-
lation of these cells, a‘slit’is construc-
ted in the region.In this stage, the con-
siderable disintegratidén of the intestinal
wall near the region is occur, and the
composed cells of the intestinal wall be-
come cordlike and extrude into the paren-
chyma. And then, the transformation of the
intestinal cells into undifferentiate ones
takes place. Many remarkable AF-positive
granules were recognized in the parenchyma
during the accumulation of the cells. And
the neoblasts in the parenchyma seemed to
migrate along the v.n.c. into the region
where the new pharynx was to be formed.

CM 18

ULTRASTRUCTURAL STUDIES ON THE SECONDARY
GILL FORMATION IN BRANCHIURA SOWERBYI.
M.Shirasawa and N.Makino. Dept. of Biol.,
Tokyo Med. Coll., Tokyo.

During the posterior regeneration of Bra-
nchiura sowerbyi, gills temporarily appear
on the operated region by the transforma-
tion of segments which originally lack in
gill structures. The process of the trans-
formation, called secondary gill formation,
has been investigated by electronmicrosco-
py- The dorsal and ventral gills of the
intact worm consist of outer simple epi-
dermis with microvilli and inner capilla-
ies surrounded by connective tissue conta-
ining contractile filaments. The gills
from secondary formation have, fundamenta-
lly, the same structure. There is no rema-
rkable difference in cytological figure of
epidermis between the gill and other body
wall. In the operated worm, primary change
of the body wall were recognized on the
most posterior region at six hours after.
Small vessels just under the epidermis jut
out on one side and push up the epidermal
cells by the endothelium. Secondarily, in
twelve hours, the rudiments of gills are
constructed by the growth of the activated
epidermal cells and the propagation of the
endothelial cells of undifferentiated type.
The completion of gill structures was ob-
served, in good condition, on the second
day. The collapse of the fine structure of
once-formed gills was already proceeding
on the seventh day.

Cytology and Morphology 1159

CM 19

MECHANISMS DETERMINING THE NUMBER OF EYES
IN POLYCELIS. 1.RELATION BETWEEN THE NUMBER
OF EYES AND THE: BODY LENGTH.

M.Aikawa and A.Shimozawa

Dept. of Anat., Dokkyo Univ. Sch. of Med.,
Tochigi.

The. freshwater planarian, Polycelis, has
hundreds of eyes in the head. We investi-
gated the mechanisms determining the number

of eyes in Polycelis sapporo (non-fission—
ing) and P.auriculata (fissioning).

P.sapporo was reared and investigated at
7-8°C after hatching. A positive correla-—
tion existed between the number of eyes and
the body length. Starvation decreased the
number of eyes. A similar correlation
existed in P.auriculata. A rate of eye
formation in the regenerating heads of
P.auriculata was higher in those with
larger tail-pieces than in those with
smaller tail-pieces. As a head regenerated
from a tail piece or as the body size
increased with feeding, the number of eyes
increased. These results suggest that the
number of eyes of an individual Polycelis
is determined by the body length.

The fine structure of an eye in
P.auriculata was investigated. The eye is
usually constituted of one photoreceptor
cell and one pigmented eyecup cell. Ina
decapitated head piece, the degenerating
eyes became apparent since 4 weeks after
cutting. The degenerating eyes were
sometimes surrounded by the cells with many
small round vacuoles.

CM 20

SCALE PLATE FORMATION AND ITS CALCIFICA-
TION IN EARLY STAGES OF SCALE REGENERATION
IN THE FLOUNDER, PARALICHTHYS OLIVACEUS.
S. Kikuchi’, A. Shimozawa2and H. Nakamura?
Kominato Lab., Fac. of Sci., Chiba Univ.,
Chiba and "Dept. of Anat., Dokkyo Univ.
School of Med., Tochigi.

Ctenoid and cycloid scales were plucked
in small areas of each side of young
adult flounders and the operated areas
of the skin were observed histologically;
staining with methyl green-pyronin for
detection of RNA and with Alizarin red
for observation of calcium deposition
against the scale plate. On days 1 to
2 after the operation, a thin layer of
pyronin-positive cells appeared in the
healing dermis. Within 5 days, a scale
plate was formed between 2 layers of
scleroblasts; superficial scleroblasts
(SSs) and deep scleroblasts (DSs). Small
deposits of calcium were observed in
the plate and only a few circuli had
formed on the scale margin. On day 7,
calcium deposition became apparent through-
out the upper surface of the plate. By
this time, some characteristics of 2
types of scales (ctenoid and cycloid)
became distinguishable. SSs were hardly
stained with pyronin, while DSs and mar-
ginal scleroblasts (MSs) were pyronin-
positive. This suggests that DSs_ and
MSs mainly participate in the formation
of collagen layers of the plates.
Obsevations on later stages of the regenera-
tion are in progress.

CM 21

BUDDING ZONE AND STEM CELL DYNAMICS IN THE
PROCESS OF STOLONIAL BUDDING OF THE COLO-
NIAL ASCIDIAN, PEROPHORA JAPONICA.
S.Kohguchi,Y.M.Sugino and K.Kawamura. Dept.
of Biol., Fac. of Sci., Kochi Univ., Kochi
We have examined histologically the stem
cell dynamics in the process of stolonial
budding of the colonial ascidian, Pero-
phora japonica. At most area except the
distal tip, the epidermal stolon was di-
vided vertically by a single-layered wall
of the septum that has long been consid-
ered as an epithelial stem cell line for
budding. However, there was no case in
which squamous septum cells participated in
bud formation. In contrast, the septum took
a tube-like structure at the distal area of
the stolon. It did not make direct contact
with the epidermis but was surrounded by
the well-developed extracellular matrix.
The tip of the tube was indented in outline.
The nucleolus and endoplasmic reticulum of
most cells were poorly developed. The mi-
totic index (MI) was less than 1%. The tube
varied in length (10-2200um) and sometimes
contained the budding zone. Cells of the
budding zone took a cylindrical shape and
showed a high activity of mitosis (MI, ca.
5%). Ultrastructural studies showed that
they were characterized by a large nucleo-
lus and by the increased number and size

of intercellular space. We conclude that
the tube-like septum is the source of stem
cells for stolonial budding and that those
cells form the budding zone periodically,
depending on the length of the tube.

CM 22

ULTRASTRACTURAL CHANGES OF THE MIDGUT
EPITHELIAL CELLS WITH FEEDING AND DIGESTION IN
NYMPHAL HEAMAPHYSALIS LONGICORNIS.

T. Morit, K. Koh? and S. Shiraishi’. *Zool., Lab., Fac.
Agr., Kyushu Univ., Fukuoka and *Lab. of Animal Sci.,

Fac. Agr., Kagawa Univ., Kida-gun.

The midgut epithelial cells in nymphs fed on
laboratory rabbits were examined during feeding and
after detachment. The midgut epithelium at the unfed
stage consisted of inactive digestive cells containing
such nutritive substances as protein, lipid and
glycogen. As feeding proceeded, the cells became
active in intracellular digestion. At the middle of the
feeding stage, the spent digestive cells derived from
the active digestive cells began to be replaced by the
regenerated inactive digestive cells. After detachment,
the pinocytotic activity of the above cells increased
greatly, and the digestive activity to some extent: as a
result, many large endosomes were formed by fusion of
numerous pinosomes. Thereafter, endosomes decreased
in exchange for an increase of hematin granules. On
day 7 after detachment, the regenerated inactive
digestive cells of the ‘nutritional reserve’ type
appeared adjacent to the spent digestive cells which
almost exhausted all endosomes, and_ replaced
completely the spent cells at day 3 after molting.

1160 Cytology and Morphology

CM 23

PSEUDOPOD FORMATION IN PLANARIAN INTESTI-
NAL PHAGOCYTIC CELLS: A SCANNING ELECTRON
MICROSCOPIC STUDY.

S. Ishii. Div. Cell Sci., Cent. Res. Lab.,
Fukushima Med. Col,, Fukushima.

It is known that the intestine of pla-
narians carries out phagocytosis and in-
tracellular digestion. Many morphological
studies have been published, but little is
known about the ingestion phase of phago-
cytosis on account of difficulties of ob-
servation in sections. This study reports
three-dimensional morphology of pseudonods
which develop in the intestinal phagocytic
cells as they are cultured and activated
AT VAneasO!.

Small tissue fragments were excised from
starved specimens of Dugesia japonica and
cultured in HEPES-buffered Holtfreter for
4 hrs. Cultures were then incubated with
1/3 fetal calf serum (FCS) or FCS-suspend-
ed polystyrene latex beads of different
diameters (0.6, 6, and 27 mm) for 0.5 to 2
hrs, After fixation and preparation for
SEM they were examined.

When incubated with FCS, individual cells
formed one or two large cupnp-like pseudo-
pods. When incubated with latex beads,
pseudopods formed varied in shape accord-
ing to the size of the particles, reveal-
ing multiple folds or funnels. Observed
morphology of pseudopods and modes of
phagocytosis were fundamentally similar to
those of other groups of phagocytes of
vertebrate and invertebrate animals.

CM 24

IS THE STERNAL GILL OF A FRESHWATER
AMPHIPOD A SIMPLE RESPIRATORY ORGAN?
AN ULTRASTRUCTURAL COMPARISON WITH THE
COXAL GILL UNDER VARIOUS SALINITIES.

S.Kikuchi,M.Matsumasa and Y.Yashima
Dept.of Biol., Sch.of Liberal Arts & Sci.,
Iwate Medical University, Morioka.

A freshwater amphipod, Sternomoera sp.,
has sternal gills in addition to coxal
gills. We found that these two kinds of
gills clearly differ from each other in
structure and function.

The epithelium of the sternal gill, an
extension of the sternal epithelium, is
extremely thick (15-20um) and character-
ized by an abundance of mitochondria and
well-developed lamellar infoldings of the
basal and the lateral cell membrane, which
involve the greater part of the cell sur-
face and the intercellular space. Consid-
ering these characteristics, the sternal
gill should be regarded as a transporting
as well as a respiratory organ.

The coxal gill, on the other hand, has
a thin(4-5y m) epithelium which is char-
acterized by extreme infoldings of the
apical cell membrane and a huge number of
large mitochondria situated among them.
This apical infolding system exhibits a
strong activity of Na*/K*-ATPase and its
size becomes markedly reduced by accli-
mation to the hypertonic ambience. These
facts strongly indicate that the coxal
gills are involved in the absorption of
ions in freshwater environment.

CM 25

DEPENDENCE OF SELECTIVE NUCLEAR STAINING OF
eels Gp ON TBs! PROTEINS.

M.¥ar , Y.Miyake’, T.Suzuki’ and J.Matsu-
no Dept. of Anat., St, Marianna Univ.
Sch. of Med.,Kawasaki and Coll, of VAgzuic.
and Vent. Med., Nihon Univ., Fujisawa.

Molecular species and their properties

involved in the hematoxylin-staining of
cellular nuclei still remain obscure. Using
biochemical technique, we have demonstrated
that hematoxylin stains nuclear proteins
rather than DNA in nuclear staining.

Hematoxylin bound some of purified pro-
teins available in the market, the proteins
were not necessarily basic. Many of nuclear
proteins had a affinity with hematoxylin,
but the majority of cytoplasmic proteins
did not. Staining intensity of nuclei after
enzymatic extraction of DNA decreased a
little, however, hematoxylin-stainable
nuclear proteins had been gone from nuclei.
Then, we measured the amount of remaining
DNA and proteins in digesting nuclei on
prescribed times. The amount of hematoxylin
bound to remaining nuclei had a closely
correlation to that of proteins, and the
reducing rate was distinguished from DNA's
one. In the same manner as above except the
enzyme was proteinase, the amount of hema-
toxylin reduced to be accompanied by pro-
tein reducing.

These results show that selective nuclear
staining of hematoxylin depends on proper-
ties of nuclear proteins rather than DNA,
and suggest that hematoxylin binding sites
may be nuclear transport signals in nuclear
proteins.

CM 26

ADRENAL TISSUES OF TOOTHED SMELT AND SURF
SMELT AND THEIR COLD ADAPTATION.

M.Ogawa - and M.Fukuchi

1: Dept.of Biology, Saitama Univ., Urawa,
Saitama, and 2: National Inst. Polar Res.,
Itabashi, Tokyo.

The adrenocortical and chromaffin cells
of the toothed smelt, Osmerus dentex and
the surf smelt, Hypomesus japonicus were
investigated by light microscopy.

They were collected in March (winter
fish) and July (summer fish) from lagoon
Saroma-Lake of Hokkaido. The head kidney
including the adrenal tissue was fixed with
Zenker-formol, embedded in paraffin and
sectioned at 6um using routine procedures.
Sections were stained with hematoxylin and
eosin.

The adrenal tissue is located around the
posterior cardinal veins in the head kidney.
Chromaffin cell mass is completely separat-—
ed from the adrenocortical cells. In both
fishes, the cross-sectional areas of adre-
nocortical cell nuclei in winter fish were
similar to those of summer fish with no
significant differences. Those of chromaff-
in cell nuclei also showed no seasonal
difference between the winter and summer
fish. No seasonal difference in the kidney
was observed in either the toothed smelt or
the surf smelt.

The present results suggest that both
the adrenocortical and the chromaffin cells
are probably not related directly to cold
adaptation in these fishes.

Cytology and Morphology 1161

CM 27

EFFECT OF ELECTRIC PULSE ON MORPHOLOGICAL
DIFFERENTIATION OF RAT PHEOCHROMOCYTOMA
CELLS (PC12)

H. Nakae and Y. Ishimori,

TOSHIBA R & D Center, Kawasaki.

We examined the effect of the electric
pulse on the morphological differentiation
of PC12 for the purpose of the application
to the control of the neural network
formation from the external field.

The electric pulse was loaded as
follows. (1) Its shape was designed
resembling action potential with a
programable pulse generator. (2) It was
loaded as short bursts (4 pulses in 35 ms)
at 5 Hz (i.e. one burst every 200 ms).
This is the same pattern as._the stimula-
tion to be known to elicite LTP in rat
hippocampus. (3) Its peak voltage was set
at 2.5 V/cm with an amplifier. (4) It was
applied to the culture medium containing
100 mM vitamin C through platinum elec-
trodes. We preliminarily ascertained that
vitamin C facilitated the differentiation
of PC12 in the presence of NGF.

As a result, cells extending neurites
appeared first after the stimulation for
9 days. Stimulating for 11 days, we
observed the neurite outgrowth clearly in
comparison with the control. This phe-
nomenon was less obvious in the medium
without vitamin C.

In conclusion, we successfully found
out the potentiality that the electric
pulse stimulation could induce the
morphological differentiation of PCl12.

CM 28

HOG Ne MOnmnC eR EAS 2, sO RUE Nee Ne SION) His
JUXTAGLOMERULAR CELLS.

R.Yoshimura, T.Migita and Y.Endo. Dept.of
Appl.Biol., Kyoto Inst.Tech., Kyoto.

The juxtaglomerular(JG) ceils are lined
around the afferent arterioles adjacent
the glomerules of the kidney. This type
of cells has a lot of large granules in
their cytoplasm. It is thought that the
renin-containing granules are released by
Chmcew sources .Ohms CamuUlie ssympatbhetic

nerve, Nat in blood, and/or Nat in urine,

There has been few morphological
demonstration of their release. The
direction of release and the route of
renin into the blood stream remain to be
eucidated, because the endothelium of the
arteriole is not permeable type. In order
to visualize the exocytotic images of
renin-granules, we have tried the
perfusion of high K* and Ca?* Ringer
solution and the fixation of tannic acid-
containing solution. The exocytotic
images of renin-granules are
chraracteristic omega figures. Those
images are found in the surface facing
other JG cells. Beside of them, narrow
and empty canals are found in similar site
which are thought to be membrane-recycling
images after the exocytosis. We think
that the released renin may flow between
JG cells anc enter the blood stream
through capillaries around JG apparatus.

CM 29

BIOCHEMICAL IDENTIFICATION OF
INTRACELLULAR LOCALIZATION OF LAMININ
IN THE ANTERIOR PITUITARY OF RATS

T.Kikuta, H.Namiki
Dept. of Biol.,Sch. of Educ.,Waseda Univ.,
Tokyo

Laminin, one of the major components of
the basement membrane, was first detected
in the gonadtrophs of the male rat
anterior pituitary by Tougard et al in
1985, and thereafter evidence has been
reported by other persons. But results of
identification of its positive cells and
its intracellular site lack unity. We
(1988) identified the laminin positive
cells as LH cells and showed the
difference of behavior of positive cells
between male and female rat at puberty.

We presently separated the secretory
granules from rat anterior pituitaries and
demonstrated the existence of laminin in
the granule, through immunoblotting
methods using both polyclonal and
monoclonal antibodies.

Among electrophoresed bands of the
granule proteins, B-chains and some other
fragments were positively stained, but not
A-chain, by antiserum for mouse EHS
sarcoma laminin. On the other hand,
monoclonal antibody against human placenta
laminin react neither A-chain nor B-
chains.

We have thus tentatively concluded that
only B-chain(s) exists in the rat LH
granule.

CM 30

THYMIC ACTIVITY OF THE MEDAKA, ORYZIAS
LATIPES, IN RELATION TO THE ACTIVITIES
OF PINEAL AND INTERRENAL GLANDS.

E. Tamura!, Ye Honma* and A. Chiba>.

Niigata Kita Senior High School, Niigata

2Sado Mare Biolmm State. .Hacw sot Scie.
Niigata Univ., Niigata,~Dept. of Biol.,
Nippon Dental Univ., Niigata.

The thymus of the wild Medaka inhabiting
in Niigata district showed a clearcut
seasonal changes in parallel with the annu-
al fluctuation of the daylength. In addi-
tion, a remarkable involution of thymus
occurred in the fish kept in a complete
dark room. These findings would intimate
a possible relationship between the thymic
activity and neuroendocrine system in the
Medaka. To clarify this problem, the thymus
of melatonin administered fish was examined
histologically. As a result, a remarkable
involution was observed. Pinealectomized
fish showed a drastic increase in the amount
of thymic cells, whereas the decrease in the
cell size of interrenal gland was detected.
On the contrary, thymic involution and
interrenal hypertrophy occurred in the
blinded fish. ACTH and HC treatments in-
duced a remarkable increase in the volume
of pineal gland of the fish. These results
suggested that the activities of pineal and
interrenal glands may affect the thymic
function of the Medaka.

1162 Cytology and Morphology

CM 31

DEVELOPMENT OF THE LYMPHOID FOLLICLE IN
THE BURSA OF FABRICIUS OF THE CHICK EMBRYO
N. Shiojiri and M. Takahashi, Dept. Biol.,
Fac. Sci., Shizuoka Univ., Shizuoka.
Lymphoid follicle formation in the bursa
requires close association of bursal epi-
thelium and mesenchyme. We do not know how
constituent cells of the bursa interact

with one another to form lymphoid follicles.

The origin of tuft epithelial cells is
debatable. The present study was undertaken
to reveal the tissue interaction and the
cell lineage in the bursal development.
Chick and quail embryos were used.
Tissue sections of the bursa were stained
with panoptic method, fluorescent lectins
and phalloidin, and anti-keratin antiserum.
Chimaeric bursa was constructed by trans-—
planting 7-day quail bursa onto the chorio-
allantoic membrane of the chick embryo.
Haematopoietic cells containing abundant
F-actin in the cytoplasm colonized in the
bursa at 7-8 days, and invaded the epithe-
lium at 10-11 days. The basal lamina dis-
appeared in the places where the invasion
occurred. The bursal epithelium appeared
passive to the invasion first. Lymphoid
follicles developed in the epithelial part
where haematopoietic cells invaded. The
result suggests that the invasion of haema-
topoietic cells in the epithelium is impor-
tant for lymphoid follicle development.
Keratin-positive tuft epithelial cells
developed from the endodermal epithelium in

the chimaeric bursa.

CM 32

IMMUNOCYTOCHEMISTRY OF PIT CELLS AND
STON ATU AES Ea slte lati ye mL lee ealem aM peat | C mY aul lcm N nlHelsy
DIGESTIVE SYSTEM OF MICE.

H.Kubo, T.Imanishi and Y.Endo. Dept.of
A\pipileB oles) sksyioie on inisitrwhelchiwsm Kiyo OF.

The pit cells locate inside of the
sinusoids of the liver, but they have a
tendency to adhere the liver, because they
remain after the perfusion. Recent
immunological studies described these
cells as the granular lymphocytes having
functions of natural killar cells. The
intraepithelial lymphocytes(IEL) in the
small and large intestine are also thought
to have natural killar activities. In
QiPceice OF) Clarity BN Orv ain Amel
relationship of them, we invesitigated the
both cell types in the digestive system
of mice by immunocytochemistry using
antibodies agaist Thy-1.2, Lyt-2, L3T4, Ia,
Mace. Cryostat sections fixed with 44
paraformaldehyde and 0.2% picric acid
were immunostained by ABC method. About
a half of pit cells were immunopositive
for Thy-1.2, but IEL were negative in all
parts of intestine as far as we examined.
Other antibodies failed to stain the both
cell types, although many lymphocytes and
macrophages were positive in the lamina
propria and submucosa. These results
suggested that the pit cells and IEL may
not belong to same functional ‘group' or
‘stage' of natural killar cells.

CM 33

ARTIFICIAL BREEDING AND HISTOLOGICAL
OBSERVATION OF GONADS OF MONOPTERUS ALBUS.
S.Matsumoto!, Y.H.Wang? and Y.Endo?2. !
Unebi JrsHigh Scho, Nara, ee wep eon
Appl.Biol., Kyoto Insti Teeh., Kyoto:

Monopterus.albdius, 15)-Ome. Gc uateme
representative species of sex changeable
teleosts. IM CNQCloie “ae @llaiawiry wine
cytological and endocrinological bases of
their sex change, we have tried their
artificial breeding including the egg-
laying and hatching. We improved two
types of the breeding instruments and the
hatching instruments. We succeeded in
the observation of behavior of their egg-
laying and hatching in the our breeding
instruments. We found that their eggs
were laid within the bubbling nest which
was produced by male after the oviposition,
The cumulative injections of testosteron
(10 pe/fish, 1 week interval, 4 times)
caused the partial degeneration of
oocytes and the start of spermatogenesis
in the relatively young fishes, 20-28 cm
in body length, which are generally female
in natural condition. This suggested
that the changes of steroid hormones may
be necessary to the cytological changes of
gonads, although the steroids may not be
a primary key on their sex change.

CM 34

POSTEMBRYONIC DEVELOPMENT OF OVARY IN
EUDIGRAPHIS TAKAKUWAI NIGRICANS
(PSELAPHOGNATHA, DIPLOPODA, ARTHROPODA) .
K.Yahata and T.Makioka. Inst. of Biol.
Sci., Univ. of Tsukuba, Tsukuba.

Some arachnid-like characteristics have
been mentioned in ovarian structure of
some diplopods despite their systematic
position as a mandibulate. In Eudigraphis
takakuwai nigricans, we studied structure
of the ovary through the postembryonic
stages.

In the 2nd larva, a young germ cell-mass
with some interstitial cells (the primary
germarium) was found between the gut and
the ventral nerve cord, entirely sur-
rounded with a thin gonadal epitherium.
Young oocytes occurred first, in the 4th
larva, in the lateral sides of the primary
germarium, surrounded with several inter-
stitial cells, which developed into
folliculated oocytes. During the sub-
sequent stages, with elongation of the
Ovary, some interstitial cells and oocytes
were transferred to form paired secondary
germaria along the ventromedian line of
the ovary. In the 7th larva, the ovarian
lumen was rapidly extended, and larger
oocytes drifted out to the ovarian lumen,
moored to their own secondary germaria
with unicellular connections made from the
interstitial cells. Adult females usually
had 10 pairs of the secondary germaria.

The ovarian structure of the present
species was similar to those of many
mandibulates, not of chelicerates.

Cytology and Morphology 1163

CM 35

POSTEMBRYONIC DEVELOPMENT OF OVARY IN
PROPALLENE LONGICEPS (ARTHROPODA,
PYCNOGONIDA) .—

K.Miyazaki and T.Makioka. Inst. of Biol.
Sci., Univ. of Tsukuba, Ibaraki.

Propallene longiceps has 9 juvenile

stages (Nakamura, 1981). In the 4th stage,
a pair of the youngest gonads appeared
above the gut in the 2nd pedigerous seg-
ment. The paired gonads extended forward
and backward, and temporarily completely
united to become unpaired during 5th and
the 6th stages, and then separated again in
the 7th stage. In the 8th stage, the
posterior ends of paired gonads united with
each other in the last (the 4th) pedigerous
segment, forming an almost U-shaped trunk
gonad as in many other pycnogonids. Howev-
er, it stopped further development and
disappeared during the 8th and 9th stages.
In the 7th stage, the paired gonads
branched toward the walking legs. Each
branch entered the corresponding leg up to
the femoral segment in female, where the
branch developed into a tubular pedal ovary
with many oocytes. Larger oocytes protrud-
ed outward from the ventral ovarian wall

with cellular stalks. The adult ovaries
were uniquely located in the femoral seg-
ments only. In the 2nd coxal segment of

each walking leg, in the 7th stage, an
oviduct branched from the ovary and extend-
ed toward an area of the ventral epidermis.
The epidermis was invaginated to form a
female genital pore, but the oviduct and
the genital pore were united in the adult
stage.

CM 36

MEIOSIS-REINITIATION-INDUCING SUBSTANCE
OF STARFISH OOCYTES FOUND IN PARAMECIUM.
Y. Katsu and M. Fujishima. Biol. Inst.,
Fac. of Sci., Yamaguchi Univ., Yamaguchi.

Ciliate P. caudatum extracts induced
reinitiation of meiosis (maturation) of
starfish Asterina pectinifera oocytes if
the extracts were added to external medium
of the oocytes. We examined the nature of
the activity and found followings. (1)
the activity is present in the suppernatant
of homogenates centrifuged at 109,000 g
for 1 hr. (2) the activity remains even
if the extracts are boiled. (3) the
activity is not lost by treatments with
trypsin, proteinase K, &-glucosidase, #-
galactosidase, neuraminidase, lipase and
RNase. (4) the extracts do not induce the
maturation if it is injected into the
oocytes. (5) time needed for induction of
germinal vesicle breakdown (GVBD) is about
20 min as well as the induction with 1-
methyladenine (1-MeAde). (6) the activity
appears in the same fraction with 1-MeAde
by gel filtration (TSK3000SW). (7) the
activity appears in 2 different fractions
by cation exchange column (Mono S). One
of them is the same fraction with 1-MeAde.
(8) Similar activity is found in
Blepharisma japonicum and Tetrahymena
thermophila extracts.

CM 37

VIDEO-MICROS COPIC ANALYSIS OF THE COELOMOCYTES OF
HALOCYNTHIA RORETZI

Marina Dan~-Sohkawa

Dept. of Biol., Osaka City University, Osaka.

H. roretzi is reported to possess six to nine
species of coelomocytes, which function together to
defend the interior environment of the animal.

In this study, we have transferred these cells
into a culture condition and observed their behavior
by time-lapse video-microscopy. By this method, we
were able to distinguish six to seven cell species
on the basis of their morphological and behavioral
features; namely, hyaline amoebocytes, granulated
amoebocytes, morula cells, giant cells, lymphocytes,
and two other species which we could not identify in
referece to the literature.

Among the five identified cell species, only the
first three responded to sheep red blood cells by
phagocytising them. Morula cells also responded
often by releasing their vesicular inclusions
immediately before or after phagocytosis.

When starfish phagocytes were added to a
coelomocyte population, here again the first three
species responded by fiercely attacking the invador.
Granulated amoebocytes wrestled with the invador by
casting their broad lamellipodia over it, hyaline
amoebocytes were the quickest to run to the site of
the battle and morula cells released their
inclusions at the site. Giant cells sometime came
straight to the site, but usually only lingered for
a while and left. Lymphocytes seemed quite
indifferent of the battle.

Analysis of the route by which these cells came
to the site of the battle strongly indicates the
possibility of chemotaxis, most likely towards two
different chemical substances.

CM 38

AN ULTRASTRUCTURAL STUDY OF PHAGOCYTE FUSION IN
STARFISH

S. Towa, M. Dan-Sohkawa, and H. Kaneko.

Dept. of Biol., Osaka City University, Osaka.

Phagocytes of the starfish, Asterina pectinifera,
adhere with each other to form an aggregate when
taken out from the coelomic cavity. These cells
spread out and fuse into a large syncytium when they
are innoculated on a substratum, such as a glass
coverslip or a plastic dish, in the form of an
aggregate. They were never seen to fuse ina
floating aggregate no matter how long they were kept
in a suspension culture.

The process of fusion of phagocytes was studied in
both spreading and floating aggregates with a
transmission electronmicroscope. There were two
structures found only in the spreading aggregate.

One is a structure resembling the adherence junction.
This structure consisted of closely opposing cel]
membrane of adjacent cells with some electron dence
material extending across the intercellular space in
a ladder fashion. The parallel membranes were also
lined by electron dence material(s). This adherence-
junction-like structure(AJLS) was frequently found to
be continuous with regions of plasma membrane which
were severely disturbed in morphology and extended
for quite a long distance. Structures were also
found in these regions which seemed to be fragments
of the AJLS. We suggest that such regions are sites
of fusion of a vast area of cell membranes which are
bound together by the AJLS. Another structure
characteristic to the spreading aggregate, although
not very frequently found, is some 2-20 vesicles
alined in the cytoplasmic region linking two cells.
Simillar role of vesicles, however, has been reported
as a fusing site in other cell types.

1164 Cytology and Morphology

CM 39

MORPHOLOGY AND EMBRYOLOGY OF THE
INFUSORIFORM LARVA OF DICYEMIDS

( MESOZOA ).

H. Furuya, K. Tsuneki and Y. Koshida.
Dept. of Biol., Coll. of Gen. Educ.,
Osaka Univ., Toyonaka.

During the course of the study for the
last two years, we found 14 species of
dicyemids in the kidneys of 7 species of
octopuses and cuttlefishes. Four of
these dicyemid species had been
described previously by various authors,
but ten species had not been described.
Dicyema sp.1, one of the dicyemids in
the kidney of Octopus vulgaris, was used
for the study of development of the
infusoriform larva. Embryos’ developing
in the axial cell of rhombogens’ were
observed under the oil-immersion system
of 2000-fold magnification. Each mitosis
was checked and the cell-lineage was
pursued. The early cleavage pattern was
spiral. At the 16-cell_ stage, four
germinal cells differentiated. At the
24-cell stage, the cleavage pattern
changed to the bilateral type. Cells in
the vegetal half of the embryo formed
the ventral surface and internal part of
the infusoriform. Cells in the animal
half formed the dorsal surface. The
completed infusoriform larva was
composed of 37 cells. The. cell-lineage
of these 37 cells was shown in a table.

CM 40

THE FINE STRUCTURE OF PIGMENT GRANULES IN
CHROMATOPHORES OF ARMADILLIDIUM VULGARE.
S.Negishi and Y.Hasegawa. Dept. of Biol.,

Keio Univ., Yokohama.

The ontogeny and fine structure of
pigment granules are studied in chromato-
phores of the isopod, Armadillidium
vulgare in taking into account the
comparison with vertebrate organelles.

Pigment granules of the animal appear at
the biginning of 2nd larval instar in the
chromatophore dendrite under the epidermal
cell layer. The organelles of spherical
and electron dense body are surrounded by
a thin membrane. The pigment granules have
an .average size of 407nm at 2nd instar

larvae, while the adult ones have 626nm.
It is shown that the dimension of pigment
granules is gradually increased from

larvae to adult in averaging them for each
larval instar. Among the mature pigment

granules, a lot of fibrillar or lamellar
structured organelles are observed in the
2nd instar larvae. These immature

organelles are characterized by the fuzzy
grains located on filaments of a diameter
6-8nm. The invasion of small pigment
granules by a large vesicle is commonly
observed in the chromatophores from 2nd to
4th instar larvae. Moreover the fusion of
immature and mature pigment granules is
remarked in the adult chromatophores. Both
invasion and fusion of pigment granules
may be involved in producing the large
size of organelles.

CM 41

A FREEZE-FRACTURE STUDY OF THE SINUSOIDAL
ENDOTHELIAL CELL IN THE GOLDEN HAMSTER
LIVER.

K.Taira. Dept. of Anat., Gunma Univ. Sch.
of Med., Maebashi.

Sinusoidal endothelial cells in the
fasted-refed golden hamster liver were
studied in thin sections and freeze-
fracture replicas. In thin sections,
trabeculated cellules were preferentially
found in the basal part of the cytoplasm
of endothelial cells. Trabeculated
cellules were made up of vesiculated
spaces and anastomosing networks of
slender cytosol. Some cellules appeared
to be composed of a few vesicles. On the
freeze-fractured P-face plasmalemma,
these cellules often appeared as oval or
round sinkages with prickles of fractured
slender cytosol. Correspondingly, these
cellules often appeared as oval or round
hillocks with small pits of fractured
slender cytosol on the E-face plasmalemma.
Ridges on the P face and corresponding
furrows with lined intramembranous
particles (IMPs) on the E face were seen
spanning between some fractured slender
cytosol. A few prickles encircled with
IMPs were sometimes seen on the E-face
plasmalemma. Moreover, IMP-circles were
seen singly or in association with the
sinkages. Some fractured cellules joined
with the sieve plates.

Trabeculated cellules seem to be
responsible for the sieve plate formation.

CM 42

EFFECTS OF UV IRRADIATION ON THE GROWTH

OF PAPILLOMAS OF JAPANESE NEWT,

Cynops pyrrhogaster. 3 1
-Oka~,M.Asashima“~,K.Kishi~ and T.Shiroya~.
Dept..of Biol.,Sch. of Med.,Toho Univ.,

Tokyo,“Dept. of Biol.,Yokohama City Univ.,

Yokohama,~Dept.of Anat.,Sch. of Med., Toho

Univ. ,Tokyo.

Spontaneous skin papillomas in the Japa-
nese newt, Cynops pyrrhogaster have been
reported. Progression and regression of
the tumor was dependent on the temperature
which controlled the size of the tumor.
Under a controlled temperature of 13°C, we
performed daily UV (254 nm) irradiation on
papillomas for 6 weeks. The size of the
irradiated papillomas began to decrease
from the 4th week. There were statistical-
ly significant differences between the
irradiated group and the control group at
the 4th, 5th, and 6th week. We examined
the histological changes in papillomas
before and after UV radiation. The epi-
dermis of normal skin consisted of 3-5
layers but the epidermis of papillomas
consisted of 10-30 layers before irradia-
tion. In the process of UV irradiation,
the number of cell layers of the papillo-
mas decreased and some epidermal cells
invaded the dermal layer. Moreover,
extra-cellular spaces were observed in the
irradiated areas. Extra-cellular spaces
were not reported in papilloma regression
induced by low and high temperature treat-
ment.

Cytology and Morphology 1165

CM 43

FETAL WOUND HEALING. THE SIGNIFICANCE OF
TYPE III COLLAGEN IN THE MIGRATION OF FETAL
MESENCHYMAL CELLS. Y. Motobayashi and S.
Ihara , Dept. Biochem., and ‘Develop. Biol.
Lab., Dept. Plastic Surg., Kitasato Univ.
Sch. Med. Sagamihara, Kanagawa 228

n open wound in the day-16 fetal rat
(Sprague-Dawley) skin rapidly closed but
the closing rate slowed down within another
2 days. The mesenchyme rather than the epi-
dermis was cotributory to the wound closure
(Ihara et al., Development, in press). No
difference, however, were noted between the
day-16 and day-18 mesenchymal cells in
cell-to-substratum attachment, spreading,
mitotic activity and contractibility of
type 1 collagen lattices. - In this report,
we studied the effects of type III collagen
and the peptide GRGDS on collagen-lattice
contraction and the migration of cells into
the lattice. This peptide had no effect on
the contraction and migration. On the
other hand, the addition of type III
collagen to the lattice apparently reduced
the rate of the lattice-contraction by both
the day-16 and day-18 cells, and however,
enhanced in a concentration-dependent
Manner the migration into the lattice of
the day-16 cells but not the day-18 cells.
The ratios of type III to type I collagen
in the living skin were determined to be
1.5 and 0.67 for the day-16 and the day-18
skin, respectively. These results suggest
that type III collagen plays a critical
role in the movement of mesenchymal cells
through 3-dimensional milieu in the rapid
wound closure in the fetal skin.

CM 44

DISTRIBUTION AND LOCALIZATION OF B-AMYLOID
PROTEIN PRECURSOR (B-APP) AND DEPOSITION
OF B-APP ON THE BASEMEMENT MEMBRANE IN
CULTURED VASCULAR CELLS.

K.Yamamoto , M.Yamamoto , T.Shinkai and

T.Ooyama
Dept. Cell Biol., Tokyo Metropol. Inst.
Gerontol., Tokyo, Div. Immunol., Tokyo

Metropol. Geriatric Hospital.

The distribution and localization of B-
amyloid protein precursor (B-APP) in cul-
tured vascular cells was investigated using
three types of antibodies to the synthetic
predictive peptides (extracellular domain;
274-286, B-protein domain; 597-620, C-
terminal domain; 681-695) of B-APP. Mon-
oclonal antibody toward the f§-protein
domain stained the surface membrane of the
cultured human endothelial cells as well as
cytoplasmic organelles. Immunoblotting
revealed that the plasma membrane-associ-
ated B-APP consisted of 105-130 kDa prot-
eins and that the mitochondria-microsome-
associated B-APP consisted of 30-67 kDa
proteins. Electron dense B-APP immunore-
activity was demonstrated in the plasma
membrane as well as in the cytoplasma.
Cytoplasmic B-APP looked as microbody st-
ructure. B-APP immunoreactivity was also
demonstrated on the basement membrane fo-
rmed in the dishes. These findings suggest
that the abnormal production or secretion
of B-APP in the vascular wall cells could
be a cause of the formation of amyloid
angiopathy in the Alzheimer's disease.

CM 45

MIGRATION OF HUMAN SKIN FIBROBLASTS:
EFFECTS OF GROWTH FACTORS AND MATRIX
MOLECULES ON CELL MIGRATION.

H. Kondo, R. Matsuda and Y. Yonezawa, Dept.
Biol., Tokyo Metropol. Inst. Gerontol.,
Tokyo 173.

It has been recently reported that, in
addition to extracellular matrix, growth
factors also regulate wound healing and in
vitro cell migration. Since we found that
human skin fibroblasts derived from adult
donor migrated more slowly in serum-free
medium than in serum-Supplemented medium,
We tried to find out a factor(s) in serum
which stimulates fibroblast migration. Cell
Migration was determined by a slight
modification of Stenn's method (1980),
which was devised as an in vitro model
system of wound healing. Both suramin,
which inhibits the binding of growth
factors to cells by competing with growth
factors, and monensin, which inhibits the
secretion of extracellular matrix, inhibit-
ed cell migration. Then, we examined the
effects of growth factors and matrix
molecules on cell migration. Platelet-
derived growth factor (PDGF), basic fibro-
blast growth factor (bFGF), acidic fibro-
blast growth factor (aFGF) and transforming
growth factor-f8 (TGF-8) were added to the
cultures and negative results were obtain-
ed. Next, collagen, fibronectin and heparin
were examined. No stimulatory effects were
observed. However, the combination of PDGF
and collagen stimulated cell migration.

CM 46

A HUMAN GLIOMA CELL LINE, SK-MG-1, CAN GROW
IN PROTEIN- AND HORMONE-FREE MEDIUM

K. Kaji (Tokyo Metro. Inst. Geront.), K.
Matsuoka (Tokyo Metro. Toshima Hosp.), R.
Hirai (Tokyo Metro. Inst. Med. Sci.) & S.
Kobayashi (Tokyo Metro. Geriat. Hosp.)

We were investigating the cell growth
control mechanisms of human glioma cell
lines. For these studies, serum-free

culture methods are thought to be powerful
because of the escape from complexity in
serun. In this study, we found that SK-—MG-—
1 cells derived from a human malignant
astrocytoma could grow in MCDB-104 basal
culture medium (no addition of any growth
factors and hormones). SK-MG-1 cells were
subcultivated serially more than 18 times
in MCDB-104, so far. We tested the
possibility whether this cell line produces
autocrine and paracrine growth factor(s).
The conditioned medium of SK-MG-1 cells
strongly stimulated the cell growth of
human umbilical vein endothelial cells and
somewhat stimulated the growth of human
fibroblasts. But, surprisingly, the
conditioned medium inhibited the growth of
its own SK-MG-1 cells. These activities
were found weakly in conditioned medium of
logarithmically growing cells and strongly
in that at a confluent state. These data
showed existing evidence for paracrine
growth control by the cell line. The
molecular characteristics of the factor(s)
responsible to these activities are to be
investigated.

1166 Cytology and Morphology

CM 47

PROLIFERATION OF RAT GLANDULAR STOMACH
EPITHELIAL CELLS IN PRIMARY SERUM-FREE
CULTURE.

H. Fukamachi, Zool. Inst., Fac. of Sci.,
Univ. of Tokyo, Tokyo.

We have previously reported that rat
duodenal epithelial cells proliferate ina
serum-free culture. Here we report that
rat glandular stomach epithelial cells
also proliferate in serum-free conditions.

Glandular stomach tissues were
dissected from 16.5-day fetuses of F344/Du
rats, and epithelial and mesenchymal
tissues were separated by the aid of
collagenase. Epithelial cells were
cultured on collagen gels in Ham’s F-12
supplemented with bovine serum albumin
(1mg/ml), and were found to proliferate
when appropriate growth factors were added
to the medium. Their maximal
proliferation was observed when epidermal
growth factor (20ng/ml), insulin
(10meg/ml), cholera toxin (500ng/m1),
transferrin (10mcg/ml), and bovine
pituitary extract (100mcg/ml) were added
into the mediun.

These results indicate that glandular
stomach and duodenal epithelial cells are
quite different in their responsiveness to
growth factors/hormones, and that this
culture system would be useful in
analyzing the mechanism of epithelial
proliferation and differentiation in the
glandular stomach.

CM 48

CELL GROWTH FACTOR AND INHIBITOR OF
MAMMALIAN CULTURE CELLS FROM SEA URCHIN
Takahashi, N.!, Sato, N.?, Hayakawa, Y.?,
Ohnishi, R.*, Nishikawa, K.*, Takahashi, M.*
and Kikuchi, K.” (‘Marine Biomed. Inst.,
Sapporo Med. Coll., Higashirishiri, Hokkaido
097-01, *Pathology, Sapporo Med. Coll.,
Sapporo 060, *The Inst. of Low Temp. Sci.,
Hokkaido Univ., Sapporo 060, and
*Shiraimatsu Shinyaku Co., Ltd., Minakuchi,
Shiga 528.)

The purpose of this study was to assess
the contribution of marine organisms to
medical science. Bioassay of the cell
growth factor and inhibitor was conducted by
mesuring *H-thymidine uptake by the
following cells. a fibloblast cell (WEB)
from a rat fetus, a human neuroblastoma
(IMR-32) , and EJ ras transformant (W14)
from WFB. The growth factor, of the order
of ng, caused *H-thymidine uptake by these
cells to increase by twice as much. This
factor had a molecular weight of 3,800 and
was also capable of inducing contraction of
the smooth muscles of sea urchin gonad and
rat uterus. In the cell growth inhibitor
within sea urchin intestine, three molecular
species were found present, the smallest
molecule of which having a molecular weight
or 27000).

CM 49

BIOCHEMICAL AND CELL BIOLOGICAL STUDY ON
SING ele OSA Si OF FIBROBLASTS
K. Shimizu!>* and K. Yoshizato“.
Develop. Biol. Lab., Dept. of Biol., Fac.
of Sci., Tokyo Metropolitan Univ., Tokyo!
and Mol; Cell Sci. Mabie Zool, sinstenemace
of Sci., Hiroshima Univ., Hiroshima‘.
Organ-specificity expressed by fibro-
blast(FB)s were studied in terms of mor-
phology, physiology and function. FBs iso-
lated from skin, lung or heart in 3-day-old
C57BL/6 mice by digestion with collagenase
and hyaluronidase were cultured in Dulbec-
co’s modified Eagle’s medium with 10%
fetal calf serum. Skin fibroblast(S-FB)s
were small and a rugby ball in shape; lung
fibroblast(L-FB)s, spindle shape; heart
fibroblast(H-FB)s, spindle shape with
several cytoplasmic processes in confluent
state. Low density culture and 3H-thymidine
incorporation tests revealed that S-FBs
proliferated at the highest rate among the
three, following by L-FBs and H-FBs in this
sequence. Amounts of collagen secretion and
strength of collagen gel contraction by FBs
were reversely proportional to the cell
proliferation rate mentioned above. After
confluent FBS were incubated with S=
methyonine, labeled proteins were analyzed
with SDS-PAGE or 2D-PAGE. Some proteins
were synthesized in an organ-specific
manner by the three types of FBs, respec-
tively. From the observation in this study
we conclude that FBs have organ-specific
characters and that collagen seems to be
involved in this organ-specificity of FBs.

CM 50

APPLICATION OF A RECENTLY DEVELOPED CELL
GROWTH ESTIMATION METHOD TO THE STUDY ON
THE CONTACT INHIBITION OF GROWTH IN
CULTUTRED C3H10T1/2 CELLS.

T.Yamaguchi.., _H.Shigematsu~., Y.Nu ada2
and H.Manki“, Fac. Gen. Educ. and “Fac.
Sci., Ehime University., Matsuyama.

Fixed monolayer of cultured mouse C3H
10T1/2 cells was stained with Methylene
Blue at pH 8.5. After washing the mono-
layer, the dye was eluted with an
acidified ethanol solution as described by
Oliver. ets ad. 1918.9) \(di.) Cenvieks caro Zi oaieoa
A linear correlation was confirmed between
the absorbance at 650 or 660 nm of the
eluted solution and the cell number. The
technique was applied to assay the inhibi-
tory effect of the cellular fractions pre-
pared by homogenization and successive
centrifugations on the cell proliferation
of a sparsely seeded (ca. 20,000 cells per
5-cm dish) growing culture. The increase
of cell number during a period from 1 to5
days after the initiation of culture was
reduced by the presence of a cell membrane
fraction which was sedimented by a 100,000
x g (60 min) centrifugation of the super-
natant subjected to a centrifugation at
20,000 x g for 20 min. A dose-dependent
Manner of the inhibitory effect suggests
that when added to growing 10T1/2 cells,
the cell membrane fraction can mimic the
effect of incresing cell density on the
cell growth.

Cytology and Morphology 1167

CM 51

EFFECT OF LOSS OF MECHANICAL DISTENSION ON
COLLAGEN DEGRADATION IN THE MOUSE UTERUS.
K.Shimizu and M.Hokano. Depat.of Anat.
Tokyo Med. Coll.,Tokyo.

The loss of mechanical distension from
the uterine wall is a principal factor in
starting the collagen degradation in the
uterus. To discover how the loss of
mechanical distension from the uterine wall
affects the removal of collagen bundles,
we investigated the distribution of
collagen bundles in the endometrium with
the picro-sirius red polarization method.

The animals used were. female mice of
the IVCS strain. At 8 weeks of age, mice
were mated. Group 1:On day 17 of pregnancy,
all fetal materials were removed from one
uterine horn. Group 2:At 7 weeks of age,
the right oviduct ligated with silk thread.
After 1 week, thy were mated. Group 3:0On
day 18 of pregnancy, both ovaries were
removed. Sections of 3um thick were stained
in picrosirius red solution. The stained
sections were viewed with polarizing light.

Artificial loss of mechanical distension
by removal of fetuses was followed by edema
in the endometrium. No endometrial edema
was found in the non-pregnant uterine horn.
Ovariectomy did not prevent the appearance
of endometrial edema on the day of
parturition. These histological results
indicate that the collagen degradation
following the loss of mechanical distension
of the uterine wall is attributable to
formation of edema in the endometrium.

CM 52

THE ROLE OF THE GLYCOSAMINOGLYCANS IN

THE DERMAL CHROMATOPHORE UNIT FORMATION
IN THE FROG, HYLA ARBOREA.

M. Yasutomi. Lab. of Biol. Aichi Medical
College, Aichi.

In amphibians, the dermal chromatophore
unit (DCU) is formed during metamorphic
stage and the animals acquire the ability
of rapid color change. Obviously, this
unit formation involves the migration of
pigment cells. The factors governing
the pigment cell migration were studied.
Glycosaminoglycans (GAG) such as hyaluro-
nate (HA) and chondroitin sulphate (CS)
proteoglycan were seen in the migrating
pathways. The GAG was identified electron
microscopically using ruthenium red stain-
ing. Collagen fibers in the dermis were
compactly arranged in the young tadpole.
However, the arrangement became loose at
metamorphic stage and many open spaces
appeared among collagen fibers. HA micro-
fibrils were distributed in the spaces
and associated with the collagen fibers.
HA has been shown to open spaces in the
collagen matrix (Pratt, 1975; Tucker &
Erickson, 1985). These suggest that HA
microfibrils make open spaces among
collagen fibers and enable the pigment
cells to migrate to subepidermal region
by squeezing through collagen matrix.

The DCU was formed in the subepidermal
region and surrounded by the networks of
HA microfibrils and 30-50 nm CS proteo-
glycan aggregates.

CM 53

IN SITU FRACTIONATED NUCLEI WITH
TRANSCRIPTIONAL ACTIVITY.
N.Koseki and K.Yoshizato. Mol. Cell Sci.
Lab., Fac. of Sci., Hiroshima Univ.,
Hiroshima

It is known that the change in
cytoskeletons alters bilogical activity of
cells. We introduced in situ fractionation

as one method to elucidate the
interactions between nuclei and
cytoskeletons. We previously reported

that in situ fractionated cultured human
skin fibroblasts obtained by treating with
buffer containing 0.5% Triton X-100
contain nuclei and cytoskeletons on the
surface of the culture dish in almost
intact oraganization. In the present
study, we examined the biological activity
in these nuclei after in situ
fractionation. Incorporation of [?H]-UTP
as one of the biological activity into RNA
was determined in medium containing 30mM
Tris-HCl, 150mM KCl, 5mM MgClze, 0O.05mM
EDTA, 20% glycerol, 0.5mM NTPs (ATP, GTP
and CTP) and 37kBq/ml of [?H]-UTP. The in
situ fractionated nuclei showed a
Significant activity of the RNA synthesis
with higher incorporation at pH 8.0 than

at pH 6.8. The incorporation was
inhibited by actynomicine D. It is
cocluded that these nuclei are
transcriptionally active and a useful
cell-free system to investigate the

relationship between nuclear activity and
cytoskeletal organization.

CM 54

ParGF, PARAMECIUM GROWTH FACTOR: ITS EX-
TENDED SIGNIFICANCE 2 3
Y.Takagi ,, K.Nimura , N.Nishi' , H.Tanabe

,and K.Kaji .

2dept. of Biol., Nara Women's Univ., Nara,
3Dept. of Endoc., Kagawa Med. Sch., Miki,
qcovt. Indust. Res. Inst., Takamatsu, and
Tokyo Metropol. Inst. Gerontol., Tokyo.

We have recently purified a 17 kDa prote-
in from the conditioned medium of the jumyo
mutant of P. tetraurelia and named it Para-
mecium growth factor (ParGF).

We questioned if other species of Parame-
cium might excrete ParGF or its equivalent:
cell-free culture media of P. caudatum and
P. multimicronucleatum were concentrated
100-fold by ultrafiltration with a mem-
brane which cuts off molecules less than 10
kDa and then the samples were tested for
the capacity to restore the reduced fission
rate of the jumyo mutant. The answer was
LVeSu foimce Slsmmlot: s bothwmatang: ty pesm™ OF
these two species.

We revealed that the purified ParGF pro-
moted an incorporation of [° H] thymidine in
BALB/c 3T3 fibroblasts. This indicates the
presence of receptors for ParGF or its hom-
ologue in mouse cells.

We studied if mammalian growth factors
such as EGF, FGF, PDGF, insulin and TGF
beta might be effective to restore the re-
duced fission rate of the jumyo mutant.
They were ineffective under the conditions
we have studied. However, we need further
study before conclusion, because fetal bo-
vine serum (FBS) was somewhat effective.

1168 Cytology and Morphology

CM 55

RELATION BETWEEN COMPOSITIONS OF CILIARY
MEMBRANE PROTEINS AND MATING REACTIVITY
IN PARAMECIUM CAUDATUM.

M. Fujishima and M. Hori. Biol. Inst.,
Fac. of Sci., Yamaguchi Univ., Yamaguchi,

Cilia of P. caudatum were isolated
by MnClo method from mating-reactive
mature cells of early stationary phase
of growth, mating-nonreactive immature
cells of stationary phase of growth,
mating-nonreactive proliferating mature
cells, and mating-nonreactive mature cells
of late stationary phase of growth,
respectively. Then, their ciliary
membrane vesicles were prepared by urea-
EDTA method and these polypeptide
compositions were compared by two
dimensional SDS polyacrylamide gel
electrophoresis and silver staining. We
found 4 polypeptides (pI 5.6-5.9, MW 60-
90 kD) which existed only in mature
cells of both mating types. All of 4
polypeptides were detected in mating-
nonreactive proliferating cells and cells
of late stationary phase of growth, but
one of the polypeptides (pI 5.6, MW 90
kD) was missing in mating-reactive mature
cells in the both mating types. These
results suggest that some or all of the
4 polypeptides may be involved in the
mating-type substances of both mating
types, and that the disappearance of one
specific polypeptide from the mating
substances may be correlated to form
active site of the mating-type substances.

CM 56

MONOCLONAL ANTIBODY AGAINST VENTARAL
SURFACE CILIA OF PARAMECIUM CAUDATUM.

Y. Inoue and M. Fujishima. Biol. Inst.,
Fac. of Sci., Yamaguchi Univ., Yamaguchi.

Mating reactive cilia isolated from
mating type odd of P. caudatum syngen 12
were injected into mice and got a
monoclonal antibody (VSC-1) against
ventral surface cilia. Indirect immuno-
fluorescence microscopy showed that the
antigens were present at the cilia of
mature cells of log phase of growth,
mating-reactive mature cells of stationary
phase of growth and immature cells of
stationary phase of growth, of both mating
types of syngen 12. The antigens were
also detected in both mating types of
Syngens 2, 3 and 6, but mating type odd
of syngen 5 did not show the fluorescence.
Other Paramecium species and Tetrahymena
examined did not show the fluorescence. The
fluorescence was detected only in fixed
cells, but not in living cells. The
antibody did not inhibit the mating
reactivity of living cells. These suggest
that the antigen determinants are not
present at the ‘outer ciliary membranes.
Immunoblot of native-PAGE showed that
molecular weight of the antigen was more
than 200 kD. But, the immunoblot of
native-PAGE showed that antigen was a
protein of about 61 kD, and that the
antigen could be detected only in even-
type. Relation of the antigen with the
mating type substances are discussed.

CM 57

MONOCLONAL ANTIBODIES AGAINST NUCLEAR
ANTIGENS IN PARAMECIUM CAUDATUM.

A. Yanagi. Dept. of Biotechnology, Fac. of Science and
Engineering, Senshu Univ. of Ishinomaki, Miyagi.

Ciliates have two kinds of nuclei, germinal nucleus
(micronucleus) and somatic nucleus (macronucleus). They
differentiate from an old micronucleus during conjugation
and are structurally and functionally different. To know the
molecular differences of these nuclei, I obtained two
monoclonal antibodies against micronuclear and
macronuclear antigens in Paramecium caudatum. Observing
immunocytochemically, the micronuclear antigen recognized
by one antibody (93A) appeared in a part of the micronucleus
in interphase cells. In the dividing nuclei, the antigen
appeared in the spindle, mainly in its poles. These results
suggest that the micronuclear antigen may be a component of
the centrosome and mitotic spindle. On the other hand, the
macronuclear antigen recognized by the other antibody (36B)
was dispersed in the macronucleus. But in late stationary
phase, aggregation of the antigen was observed in some
stock (27aG3), while in some other stock (c103) this was
not observed. These results suggest that the macronuclear
antigen is a component of the nucleoli. During the
development of macronuclear anlagen, the micronuclear
antigen persisted in a new micronucleus but disappeared in
newly developing macronuclear anlagen. At the beginning
of the anlagen development, the macronuclear antigen was
not observed in the macronuclear anlagen but after a while it
appeared in all of them. These two nuclear antigens may be
good markers of the differentiation of germinal and somatic
nuclei.

CM 58

CONSTRUCTION OF MONOCLONAL ANTIBODY LIBRARY
AGAINST BRAIN SPECIFIC PROTEINS.

M. Sugano, H. Nakae, and Y. Ishimori,
TOSHIBA R & D Center, Kawasaki.

We have been constructing a monoclonal
antibody library against central nervous
system (CNS) -specific proteins in order to
investigate properties of nervous cells.

We prepared monoclonal antibodies (Mabs)
against partial-purified fractions mainly
containing cell membranes of an adult rat
brain. Two species of Mabs, RB2-4 and
RB3-2, were obtained. These Mabs only
reacted with cerebrum, cerebellum, pons,
medulla oblongata, and spinal cord by means
of immunoblotting. Their antigens' charac-
teristics elucidated by the analyses of
SDS-PAGE or 2DEP, were as follows.

(1) RB3-2 antigen: MW ca. 110KDa, acidic
protein. (2) RB2-4 antigen: MW ca. 40KDa,
acidic protein. As the first step for
understanding the functional significance
of these antigens, we investigated their
expression in development using immuno-
blotting. Mab RB3-2 continuously reacted
with cerebrum and cerebellum since 12-day
embryo. On the other hand, Mab RB2-4
reacted with postnatal cerebrum and
cerebellum. And immunohistochemical
observations showed that the RB2-4 antigen
was localized on the molecular layer and
the granular layer in cerebellum of 11-day
rat.

From these results, we concluded that
two Mabs obtained here were specific to
CNS and Mab RB2-4 antigen was expressed
only after birth.

Cytology and Morphology 1169

CM 59

IMMUNOCYTOCHEMICAL CHARACTERIZATION OF THE
DIFFERENTIATION OF NEURAL CREST CELLS INTO
MELANOPHORES AND ITS RELATION TO MIGRATORY
BEHAVIORS IN GOLDFISH Carassius auratus AND
MEDAKA Oryzias latipes.

T. Akiyama , J._Ma sumoto!, M. Nakamura!
and K. Kitamura“. ‘Dept. gf Biology, Keio
University, Yokohama and “Dept. of
Developmental Biology, Mitsubishi-Kasei
Institute of Life Sciences, Machida.

The differentiation of neural crest cells
into melanophores and their migratory path-
ways in goldfish and medaka embryos and
larvae are studied by means of immunocyto-
chemistry using HNK-1 and MEBL-1 anti-
bodies, which respectively recognize the
epitopes on the neural crest cells in
migratory phase and melanoblasts. Paraffin-
or cryosections are stained by a biotin-
streptavidin-FITC system. The results
indicate that (1) HNK-1 positive cells
appear first in the outer ridge of the
neural tube and then mostly in the ventral
route while in limited numbers in the
dorsolateral, (2) the antigenicity to MEBL-
1 appears in the cells located in the same
routes from the neural crest in goldfish at
the stage slightly later than the appear-
ance of the reactivity to HNK-1, and is
maintained until the onset of melanogene-
sis, (3) fully melanized melanophores show
little reactivity to these two antibodies.
Suggestions are obtained that the major
population of melanoblasts migrates along
the routes taken by pioneering melano-
phores.

CM 60

ORGANIZATION OF THE CEREBRAL GANGLION OF
THE COLONIAL ASCIDIAN, POLYANDROCARPA
MISAKIENSIS.

H.Koyama, Dept. Anat., Sch. Med., Yokohama
City University, Yokohama.

Light and electron microscopical
studies were made on the cerebral ganglion
and its associated structures of the
styelid colonial ascidian, Polyandrocarca
misakiensis. The lateral and ventral side
of the ganglion is surrounded by blood
sinuses. The ganglion is covered with a
fibrous sheath, through which many nerve
fibers run. The ganglion is composed of a
cellular cortex and a fibrous medulla. The
cortex consists of several layers of large

and small neurons. Some neurons are also
scattered within the medulla. Many neurons
are monopolar, and some are bi- or

multipolar. The Gyegs ol of the large
neurons is dense with extensive rough ER,
free ribosomes, mitochondria, dictyosomes,
large dense bodies, and many vesicular
structures. Some neurons contain
neurosecretory granule-like structures. The
nerve fibers occupying the medulla contain
mictotubules, vesicles of various sizes,
and granules. Since the arrangement of the
simple squamous endothelial cells of the
sinuses is quite loose, some exchange of
substances may be possible between the
ganglion and the blood. Some neurons extend
their processes directly into the lumen of
the sinuses. Most synapses are formed
between the nerve fibers, but soma-to-
soma, and soma-to-neurite synapses are
also seen. The neuronal somata and their
fibers are covered with loose extracellular
matrices.

CM 61

SUBSTANCE P IN THE OPTIC NERVE OF THE
LAMPREY.

pe Kadota, 7 RCo Gorts, oR. Kishida! and T.
Kusunoki. Dept. Anat., Yokohama City U.
Sch. of Med., Yokohama and lyamaguchi U.
Sch. of Med., Ube.

The structures containing substance P
in the brain with their enveloping
structures, including the optic nerve and
eyes, of Lampetra japonica were studied by
the PAP technique with rat monoclonal
antibody against substance P. Substance
P-like immunoreactive nerve fibers (SPFs)
were uniformly distributed in the optic
nerve, not branching but running parallel
walthy ist On entering the retina, most
SPFs entered the inner plexiform layer and
the rest entered the ganglion cell layer.
A large number of SPFs were distributed in
the ganglion cell layer and a medium to
small number were distributed in the inner
plexiform layer and inner nuclear layer.
There were no substance P-containing
ganglion cells, amacrine cells, or other
cells in the retina. It appears that
these SPFS are retinopetal fibers.
Substance P-containing cell bodies were
distributed in the midbrain tegmentum. A
small number of SPFs entered the optic
nerve from the hypothalamus and
contralateral side at the optic chiasma.
The retinopetal fibers may come from both
sides of the midbrain and from some other
source.

CM 62

COMPARATIVE AND MORPHOLOGICAL STUDIES OF
THE ACCESSORY CILIARY GANGLION IN MAMMALS.
S. Kuchiiwa. Dept. of Anat., Hirosaki
Univ. Sch. of Med. Hirosaki.

The orbits of 16 mammalian species were
stained with silver-impregnation or
acetylcholinesterase method and dissected
under a dissecting microscope with special
attention to the presence and location of
the accessory ciliary ganglion. In the
cat the ganglion was studied with the
light microscope by histological and
hodological techniques.

The accessory ciliary ganglion was
present in all 16 species, although the
number and degree of development varied
greatly from species to species. The
ganglion was located on the short ciliary
nerve and it never had a root derived
directly from the inferior trunk of the
oculomotor nerve. In most species a
communicating branch from the trigeminal
nerve fusing with the ganglion was
observed. In the cat, after injection of
horseradish peroxidase-wheat germ
agglutinin conjugates into the rostral
midbrain including the oculomotor nucleus,
anterogradely labeled terminals were
observed in the accessory ciliary ganglion
and many labeled fibers were found in the
communicating branch. This indicates that
there must be two efferent pathways for
the oculomotor parasympathetic functions,
one passing through the main ganglion, the
other running through the trigeminal nerve
to reach the accessory ciliary ganglion.

1170 Cytology and Morphology

CM 63

INNERVATION OF MONGOLIAN GERBIL HEPATIC PO-
RTAL SYSTEM.

T.Amano, N.Yamashita, N.TSujita, T.Tagawa
Dept. of Anat., Sch. of Med., Fukuoka

The neuro innervation of Mongolian Gerbil
hepatic poltal system was separated to the
distal(portal.v) and proximal (sup.mesenter-
ic.v, inf.mesenteric.v, splenogastric.v)
and investigated by catecholamine fluoresc-—
ene, cholinesterase histochemistry and imm-
unohistochemical methods.

Sympathetic nerve fibers was generally di-

stributed densely however density was grea-"

ter in the distal portion.

Parasympathetic nerve fibers was distribu-
ted sparsely.

SP,NKA and CGRP-like immunoreactive (LI) fi-
bers was generally distributed, and distal
portion was more dense.

The coexistence of both SP and NKA, SP and
CGRP-LI fibers were confirmed.

As similary to parasympathic nerve fibers,
there was no Significant difference between
distal and proximal distribution of VIP-LI
fibers.

NPY-LI fibers was more densely distributed
than the other 4 peptide neuro fibers, esp-
ecially in the proximal portion.

Several thick neurobundls were frequently
scene parallel to the vessel axis among
many thin neuro fibers.

CM 64

A HISTOLOGICAL STUDY ON THE NERVE INNERVA-
TION OF HUMAN TEETH.

H. Ishihara. Biolog. Lab., Fac. of Sci. &
Engineer., Aoyama Gakuin Univ., Tokyo.

The morphology of nerve innervation and
nerve endings of human teeth was investi-
gated through the application of improved
techniques devised by the present author.

The nerve plexus was found in the pulp
which consisted of both myelinated and un-
myelinated nerve fibers. Generally unmy-
elinated nerve fibers were located in the
bundles among the myelinated nerve fibers.
Fine nerve fibers branching from plexuses
were observed running parallel to capil-
laries or winding around them. The nerve
fibers from the plexus of Raschkow running
among the odontoblasts change their direc-
tion to reach the predentin surface, where
they were divided into numerous branches
forming a marginal plexus. In the pre-
dentin layer, nerve fibers interlace; some
run parallel with the tubules and with
Tomes'fibers; some loop back toward the
pulp, and larger ones run transversely in
the grooves. Sometimes at the junction
of two or more branches, cell-like ele-
ments or bodies which look like nerve

cells were observed. Further at the
swelling points of nerve fibers non-stain-
ed center was frequently found. The

presence of minute black beans, or fusi-
form vesicular swelling, occurring along
the fibers were served as a distinct sigh
of terminal nerve fibers.

CM 65

THE RELATIONS BETWEEN THE BEHAVIOR OF
NEUROSECRETORY CELLS AND THE MATURATION
OF THE SLUG (LIMAX MARGINATUS).

N. Seo and N. Makino, Dept. of Biol.,
Tokyo Med. Coll. , Tokyo.

We have studied the mechanism of the
maturation in the slug (Limax marginatus).
We investigated the relations between
the behavior of neurosecretory cells
(NSCs) in the cerebral ganglion and
the suboesophageal ganglion and the
maturation of the hermaphroditic gland.

NSCs were stained with Ardehyde-
Fuchsin, and these were counted during
a year. Two types of NSCs were grouped
roughly by the size, under 10um in
diameter NSCs (T,-NSCs) and over 15ym™ in
diameter NSCs (T.-NSCs). (1) The hatching
larva has no NSCs in the cerebral ganglion
and the suboesophageal ganglion. (2) The
number of T,-NSCs of the cerebral ganglion
and the suboesophageal ganglion increases
as the slug grows, and a maximum number
reaches at the oviposition stage. Iie
decreases rapidly after some ovipositions.
(3) The number of T,-NSCs of the
Ssuboesophageal ganglion takes a maximum
at the inmature (bigining spermatogenesis)
stage, and a maximum number keeps till the
Oviposition stage. Two groups of T,-NSCs
of the visceral ganglion and the pedal
ganglion appear at only the maturation
period of male cells.

We think that these activities of
T,z-NSCs of the suboesophageal ganglion
infleuence the differentiation of male
cells and the following spermatogenesis.

CM 66

THREE-DIMENSIONAL VASCULAR ARCHITECTURE OF
AMPHIBIAN CAROTID LABYRINTHS

T. Kusakabe. Dept. of Anatomy, Yokohama
City Univ. Sch. of Med., Yokohama.

The three-dimensional structure of the
carotid labyrinth in five species of anu-
rans(Rana nigromaculata, Rana catesbeiana,
Bufo japonicus, Hyla arborea, and Xenopus
laevis), and three species of caudates
(Cynopus pyrrhogaster, Hynobius nebulosus,
and Ambystoma mexicanum) were compared
using corrosion castings and scanning
electron microscopy. Anuran carotid
labyrinths are spherical in shape and are
classified into two groups according to the
origin of the external and internal carotid
arteries. One group, which includes Rana,
Hyla, and Bufo, is characterized by the
presence of a vascular ring at the proximal
end and some vascular routes at the distal
end of the labyrinth. The external and
internal carotid arteries originate from
these structures. In.the other group,
which includes only Xenopus, the external
and internal carotid arteries originate
from the central chamber and from within
the vascular maze, respectively. The
carotid labyrinth in Cynops and Hynobius is
oblong in shape. The fundamental organi-
zation in salamanders is similar to that
in anurans. There is no specialized
swelling in Ambystoma. The present findings
suggest that most amphibian carotid labyr-
inths have the appropriate architecture for
controlling vascular tone.

Cytology and Morphology 1171

CM 67

HISTOLOGICAL OBSERVATIONS AND COMPUTED
THREE DIMENSIONAL RECONSTRUCTION OF THE
TRACHEA OF HYNOBIUS NEBULOSUS TOKYOENSIS.
Y.Kikuchi, T.Gomi, A.Kimura and kK.Kishi.
Dept. of Anat., Sch. of Med., Toho Univ.,
Tokyo.

The trachea of Hynobius nebulosus
tokyoensis was seen to branch out from the
ventral side of the esophagus, and in this
area the trachea was slit-like in shape.
The trachea had a vertical ellipsoid shape
in the cranial area, and a crosswise
ellipsoid shape in the caudal area. The
trachea branched into the right and left
lungs on the dorsal side of the heart. A
pair of cartilages existed on both sides of
the trachea. The cartilages were half-moon
shaped in the most developed area. In the
cranial area, the sphincter surrounding the
cartilages was well-developed. Tracheal
epithelium was mainly composed of ciliated
cells and goplet cells. Furthermore, brush
cells possessed long microvilli and
structures seeming to be neuroepithelial
bodies were observed. Moreover, in the
boundary area between the trachea and
lungs, epithelial cells were located
between the adjacent well-developed
capillaries in the dorsal area. These
structures were similar to those of the
respiratory portion of the lungs.

CM 68
MORPHOLOGICAL STUDIES ON THE ONTOGENY OF
THE RAT LUNG. (1) LIGHT AND ELECTRON

MICROSCOPIC OBSERVATIONS OF THE FETAL LUNG.
T.Gomi, A.Kimura, Y.Kikuchi, Y.Kitazawa ,
-Tsuchiya and K.Kishi. Dept. of Anat.,
Pathol., Sch. of Med., Toho Univ., Tokyo.

We observed the fetal rat lung by both
light and electron microscope. The
development of the fetal rat lungs were
roughly classified histologically into
three periods: 1) Pseudoglandular period:
The developing lung somewhat resembled a
gland. The epithelial cells of the
primordial system of the lung were columnar
in shape, and contained many glycogen
granules. The shape of nuclei were usually
oval. Capillaries were scattered in the
mesenchymal tissue. 2) Canalicular period:
The glandular structure changed into
irregular canalicular structure. The lumina
of the bronchi and bronchioles have become
much larger, and the lung tissue was highly
vascular. The epithelial cells were low-
columnar or cuboidal in shape, and
capillaries existed near the epithelial
cells. 3) Terminal Sac period (primitive
alveolar period): .Many terminal sacs
developed, and the epithelial cells of the
terminal sacs were very thin. Capillaries
began to bulge into the sacs. Epithelial
cells consisting of the terminal sacs were
seen to differentiate into two cell types:
Type I and Type II. Many osmiophilic
lamellated bodies were observed in Type II
cell the last gestational day.

CM 69
MORPHOLOGICAL STUDIES ON THE ONTOGENY OF
THE RAT LUNG. (2) BINARY IMAGE PROCESSING

ANALYSIS OF THE DEVELOPMENT OF THE FETAL
AND NEONATAL RAT AIR-CONDUCTING SYSTEM.
A.Kimura, T.Gomi, Y.Kikuchi and K.Kishi.
Dept. of Anat., Sch. of Med., Toho Univ.,
Tokyo.

We analyzed the development of the
fetal and neonatal rat pulmonal air-
conducting system (ACS), employing computed
image processing analyzer (SPICCA-II
system). Light microscopic photography
images were input to the computer by video
TV camera and, image data was binaly
produced with the use of an image
processor. The development of the fetal and
neonatal rat lung was roughly classified
into four periods: 1) Pseudoglandular
period (15.4-15.6%; ratio of ACS): The
developing lung somewhat resembled a gland.
The bronchial divisions were beginning to
differentiate into the ACS. 2) Canalicular
period (24.7-25.3%): The lumina of the
bronchi and bronchioles have become much
larger. 3) Terminal Sac period (35.5-38.83%,
primitive alveolar period): Many terminal
sacs have developed. 4) Alveolar period
(47.6-59.1%, neonatal): The terminal sac
walls have become increasingly thinner, and
the alveoli have expanded. The ratio of the
ACS to interstitial tissue increased in
accordance with gestational and neonatal
days. Moreover, there was a correlation
(p<0.05) between the ratio of the ACS and
body or lung weight.

CM 70

FILAMENTOUS STRUCTURE CONNECTING TO J-RODS
IN THE OBLIQUELY-STRIATED MUSCLE OF THE
NEANTHES SP.

Y.Kawamura and A.Matuno. Dept. of Biol.,
Fac. of Sci., Shimane Univ., Matsue.

The longitudinal muscle of Neanthes sp.
(Annelida, Polychaeta) is composed of
obliquely-striated muscle cells. Two types
of myofilaments, thick and thin , are
arranged obliquely in the cell. The cell,
consequently, shows obliquely-striated
pattern of sarcomere-like structure. J-rods
are situated in the center of the I-bands,
and thin myofilaments connect to the rods.
In a cross section of a cell, filamentous
bundles of about 0.1 um in diameter are
composed of 8-10 nm filaments.These bundtes
appear beneath the cell membrane to make
mesh-work and connect tightly to J-rods at
the knobs of the mesh. In a contracted
state, the bundles seem to across A-bands.
On the other hand, in a stretched state,
they are strained longitudinally. These
structural changes between the two states,
Suggest that the filamentous bundle keeps
the J-rods from disordered arrangement
during contraction and relaxation. Therefor
the characteristic arrangement of J-rods
may be kept by these bundles.

1172 Cytology and Morphology

CM 71

ULTRASTRUCTURE OF ADDUCTORS OF THE HEART
SHELL.

A.Matsuno!, Y.Kawamura! and T.Yamasu?’.
1Dept. of Biol., Fac. of Sci., Shimane
Univ., Matsue, *Div.Gen.Educ., Univ.,
Ryukyus, Okinawa.

The heart shell, Corculum cardissa, is
a peculiar bivalve mollusk that shows
heart shaped outline when viewed from the

end. An anterior and a posterior adductors

are situated in the shell. The anterior
adductor is smaller and shorter than the
one of the posterior. The two adductors
have a main large part constructed by

obliquely-striated muscle cells and a small

part of smooth muscle cells distributed in
the periphery of each adductor. Smooth
muscle cells in the anterior adductor
contain thick myofilaments measuring about
50 nm in diameter, and ones in the poste-
rior contain about 80 nm thick myo-
filaments.

It is generally believed that "catch"
muscle has very thick myofilaments sup-
ported by the paramyosin-core. In our
observations, heart shell had 80 nm thick
myofilaments in the posterior adductor and
had not such thick ones in the anterior
adductors. These observation suggest the
posterior adductor may contract "catch"
mode and the anterior one does not show
"catch" contraction. :

CM 72

TRANSVERSE INTERMEDIATE FILAMENTS CONNECTING
ADJACENT MYOFIBRILS IN HONEN BEE GEGHi MUSCLE -

K. Terakado and K. Maruyama’. Dept. of Regulation

Biol., Fac. of Sci., Saitama Univ., Urawa and ‘Dept.

of Biol., Fac. of Sci., Chiba Univ., Chiba.

An extensive network of transverse and oblique
filamentous bridges was revealed by transmission
electron microscopy in honeybee flight muscle.
These filaments possessed a diameter of about 10nm
and ran in small bundles. Observation of these
intermediate-sized filaments was very easier as
compared with those of vertebrate skeletal muscle,
apparently due to a conspicuous development of the
intermediate filament network. The bundles of
transverse or oblique filaments connected adjacent
myofibrils by forming Z to Z bridges. In cross
section, the transverse intermediate filaments
often attached perpendicularly (not encircled as in
vertebrate striated muscles) to myofibrils at the
level of Z-discs. The transverse bundles connected
also peripheral myofibrils to the cell surface by
linking the Z-discs to the subsarcolemmal dense
patches. Presence of longitudinal intermediate
filaments that run parallel with myofibrils and
connect peripheries of successive Z-discs was not
ascertained in the present study. These results
indicate that the intermediate filaments occur much
in quantity and are well organized in a network
which may play an important role in maintaining
the arrangement and organization of contractile
and other structures in honeybee flight muscle.

CM 73

THE ULTRASTRUCTURE OF THE SEA URCHIN EGG
CYTOMATRIX REVEALED BY QUICK-FREEZE DEEP-
ETCH TECHN TOQUE.

S.Hisangga , S.Endo“, N. irokawa?>,
H.Sakai“~ and J.Pudles”. ‘Lab. of Cell &
Dev. Biol., Fac. of Lifesci. & Lifetech.,
Tokyo Insti. Tech., Dept, of Biophys. &
Biochem., Face. of Sci., ~Deptialol Andes
Cell Biol., Fac. of Med., Univ. of Tokyo
and “Lab. de Str. et Reactivite des Prot.,
CNRS, Univ. de Paris-Sud.

Ultrastructure of the sea urchin egg
cytomatrix, defined as the detergent-
resistant cytoskeleton, was studied by the
method of quick-freeze, deep-etch electron
microscopy. Two types of cytoskeletal
organizations, (1) cortical actin
filaments and underlying cytoplasm, were
observed. With respects to cortical
actin, our results confirmed several
recent observations. The cytoplasm was
composed of the complex network of fine
filaments interconnecting the aggregates
of small globular materials, yolk granules
and masses of meshwork. The ground
network consisted of several sizes of
filaments, 6 nm to 13 nm in diameter. The
meshwork was a novel structure which has

not been previously reported. The

meshwork was composed of homogeneous
filament with a diameter of 8-9 nm. Each
filament formed junctions with other
filaments at both ends, forming an
assembly of polygons. The cytomatrix did
not changed it's ultrastructure after
fertilization.

CM 74

CORNEAL NIPPLE FORMATION IN THE COMPOUND
EYE OF THE ERI-SILKWORM
R.Furuta and S.Takahashi, Dept. of Biol.,
Nara Women's Univ. Nara.

The morphogenesis of corneal nipple
structures in the eri-silkworm Samia
cynthia ricini was studied with electron
microscopy. Pupae were kept at 25C. During
the first 5-7 days after pupation, a large
number of microvilli arose from the
corneagenous cell surface. On the 8th day
after pupation, each microvillus secreted
a patch of cuticle distal to the tip.
Patches bulged out markedly. In the
developing microvilli, longitudinally
orientated microfilaments were observed.
The protruding patches, connecting each
other at the edges, became 'nipples' of
the facet.

To examine the effect of cytochalasin B
(CB) on the nipple formation, pupae were
received each an injection of 5 ug of CB
at various stages. When the adult
emerged, nipples occurred in a wide
variety of hights, irregullary arranged.
TEM observation suggested that the
abnormal development of nipples were
caused by irregular protruding of
microvilli. There must be a strict
one-to-one relationship between the
patches and the microvillar tips. The roll
of microfillaments must be important for
the normal protruding of microvilli which
is closely related to the nipple
development.

Cytology and Morphology 1173

CM 75

GAP JUNCTIONS OF THE CULTURED SMOOTH
MUSCLE CELL LINE.

Y.Endo! and Y.Sasaki*. 'Dept.of Appl.Biol.,
Kyoto Inst.Tech., Kyoto, *Life Sci.Center,
Asahi Chemical Industry, Miyazaki.

In general, each of smooth muscle
cells has not supply of nerve fibers, but
those cells are communicated with each
other by their gap junctions. In order
to investigate the formation process and
distribution pattern of gap junctions, we
used the cell line (SM-3) established from
the rabbit aorta.

Existence of gap junctions was demon-
strated by microinjection of fluorescent
dye (hucifer yellow) solution. ~The
pattern and number of intercellular
junctions per one cell were estimated by
the scanning electron microscope images.
The micro-injected dye was rapidly
transferred into adjacent cells. The
intercellular communication by gap
junctions was affected with the addition
of dibutyryl cAMP(10 #and 10 4M) into the
culture medium. The numbers of Lucifer
yellow-transferred cells increased
Significantly by dbcAMP. But the number
of intercellular junctions was not changed
by dbcAMP. These results indicate that
dbcAMP may act the gap junctions
intracellularly to strengthen their
ability of intercelluar communication.

CM 76

CO-CULTURE OF PC-12 CELLS AND SMOOTH
MUSCLE CELL LINE.

T.Maruyama'!, T.Yoneda', K.Watanabe',
Yesenidonw wand YoSasiak a2. 'Dept.of
Appin Kvotco bnist..lrech.>, Kyoto’,
?Life Sci.Center, Asahi Chemical Industry,
Miyazaki.

In general, the autonomic nerves do not
have a synaptic contact with smooth muscle
cells. The terminal portion of the
autonomic nerves consists of a continuous
network of unmyelinated nerve fascicles.
In order to study their innervation
mechanism, we have tried to construct a
model system in vitro. We chose PC-12
cells as neural elements, because this
Cen sie ist ne iss) sna belies to, ibe. eas ly
differentiated to a neural form by the
addition of nerve growth factor. SM-3
cell line was estabished from the rabbit
aorta by Sasaki et al.(1989) and they are
differentiated into the smooth muscle
cell type during the growth arrest stage.
After the co-culture, the relation
between them was examined with
immunocytochemistry using antibody against
neurofilament protein and scanning
electron microscopy. The neurites of PC-
12 cells appeared to approach to the
smooth muscle cells. But it is unkwown
whether the neurites are differentiated
into axons or dendrites.

CM 77

ULTRASTRUCTURE OF THE LATERAL-LINE ORGANS
IN THE CRAB-EATING FROG LARVAE.
M.Uchiyama!, S.Iwasaki* and H.Yoshizawa?.
'Dept. of Oral Physiol. and 2Dept. of
Anatony, The Nippon Dental Univ., Niigata
and 7Dept. of Oral Histol., Matumoto Dental
GOIN. 5 Slesi@ yates

Neuromast structure in Rana cancrivora
larvae was studied by scanning and trans-
mission electron microscopy. Neuromast
units, each being composed of two or three
neuromasts, are arranged in several lines
in the head, body and tail regions. The
structure of neuromasts in these regions is
basically identical. The neuromast is conm-
posed of sensory, sustentacular and mantle
cells. The top of each neuromast has a
hillock-like appearance, and is surrounded
by four to six epidermal cells with tight
intercellular junctions. Long kinocilia and
Many stereocilia occur in the apex of the
neuromasts and are surrounded by numerous
microvilli. Numerous granules are present
on the apical portions of the mantle and
the sustentacular cells. Four or five man-
tle cells are connected closely with each
other to form the shell of the neuromast.
Large intercellular spaces occur between
the mantle cells and the cells of the inner
epidermal layers, and between the cells of
the inner epidermal layer. Two types of
nerve endings, afferent and efferent, are
observed. Morphologically, the neuromasts
of R. cancrivora larvae resemble those of
generalized pond anurans although larvae of
this species inhabit brackish water.

CM 78

ULTRACYTOCHEMICAL DETECTION OF Ca-ATPase:
ITS BIOSYNTHESIS IN THE ROUGH ENDOPLASMIC
RETICULUM AND TRANSPORT THROUGH THE GOLGI
APPARATUS TO THE PLASMA MEMBRANE IN TASTE
BUD DURING EARLY POSTNATAL PERIOD IN
MOUSE VALLATE PAPILLA.

M. Kudoh. Department of Biology,
Fukushima Medical College, Fukushima

Ca-ATPase in the taste bud cells of
mouse vallate papilla was detected by the
electron microscopic cytochemical
procedure. In mature taste bud, the
enzyme was intensively located on the
plasma membrane of various cell types and
around the periphery of nerve fibers, but
not in subcellular organelles such as
mitochondria, the rough endoplasmic
reticulum (rER) and the Golgi apparatus.
These findings indicate that plasma
membrane Ca-ATPase plays the primary role
in determining the cytosolic Ca ion
concentration of taste bud cells and
nerves at a very low range. In the
gustatory epithelial cells during early
postnatal period, on the other hand, Ca-
ATPase was apparently visualized in the
lumen and cisternae of the rER, the Golgi
apparatus and small vesicles, which were
presumably a Ca-ATPase transporting
machinery. These results suggest that
Ca-ATPase synthesized in the rER passes
through the Golgi apparatus to the plasma
membrane, mediated by transport vesicles.

1174 Cytology and Morphology, Behavior Biology and Ecology

CM 79

FINE STRUCTURE OF THE DORSAL EPITHELIUM OF
THE TONGUE OF THE FRESHWATER’ TURTLE,
GEOCLEMYS REEVESII
S. Iwasaki and K. Kobayashi, Dept. Anat.,
Sch. Dent. Niigata, Nippon Dent. Univ.,
Niigata

Scale-like lingual papillae were observed
all over the dorsal surface of the tongue of
Geoclemys reevesii, by scanning electron
microscopy. The Surface of the papillae was
composed of compactly distributed semi-
spherical bulges of cells. Microvilli were
widely distributed over the surface of
cells. Histological examination revealed
that the connective tissue penetrated deep
into the center of papillae and that the
epithelium was of the stratified columnar
type. Under the transmission electron
microscope, the cells of the basal and the
deep intermediate layers of the epithelium
were somewhat round. A large nucleus was
located in the central area of each cell. In
the cytoplasm, mitochondria, endoplasmic
reticulum and free ribosomes were observed.
The cell membrane formed numerous processes.
In the shallow intermediate layer, two types
of cell were observed. One type contained
numerous fine granules, other than
mitochondria, ribosomes and_ endoplasmic
reticulum, in its cytoplasm. The other type
of cell contained highly electron-dense
granules. In the surface layer, two types of
cell were recognized. One was the typical
mucous cell. The other contained fine,
electron-lucent granules. The latter was
located on the free-surface side covering
the mucous cell, and had microvilli on its
free surface.

BE 1

MORPHOLOGICAL VARIATION AND CANNIBALISM
OF THE SALAMANDER LARVAE IN HYNOBIUS
NIGRESCENS

Hisashi Takahashi. Dept. of Biol., Fac.
of Sci., Kanazawa Univ., Ishikawa

Interpopulation variations in larval
morphology of Hynobius nigrescens, and
the relationship between morphological
variation and cannibalism were studied in
field sampling and laboratory experi-
ments. Body length, head width and
stomach contents of larvae taken from 9
various larval density ponds were deter-
mined. Adjusted head width was signifi-
cantly differed among populations.
Larvae in high density pond had a contin-
uous polymorphism of larval morphology
with large head width (LHW) larvae.
Stomach contents of larvae were less full
in populations had LHW larvae than in
ones had no LHW larvae. An experiment in
the laboratory was conducted to evaluate
the relationship between food and density
of conspecifics, and expression of the
variation in head width. In larvae
derived from eggs of the population had
LHW larvae in natural condition, LHW
larvae were shown in high density of
larvae and low food level blocks. In
larvae derived from population had no LHW
larvae in the field, LHW larvae were not
shown in every experimental designs. LHW
larvae systematically eaten smaller head
width larvae.

BE 2

SPAWNING BEHAVIOR OF THE SALAMANDER, HYNOBIUS
RETARDATUS.

T. Sato. Obihiro Centennial City Museum,
Obihiro.

The spawning behavior of the salamander,

Hynobius retardatus, obtained from the foot of

Hidaka Moutains, Hokkaido was examined in the
laboratory. It was confirmed that this species
had external fertilization in water. Although
there was no contact and no apparent mutual
stimulation between the pair till egg
deposition actually begins until now, several
courtship behaviors were confirmed in this
examination. The spawning behavior of the
females was as follows: @ seeking a twig to
attach her egg sacs, @ grasping the twig with
her limbs,@ rubbing her cloaca over the twig,@
raising her tail, and® attaching her egg sacs.
The courtship and spawning behavior of the
males was as follows: @ pursuing the female, @
confirming the female’ s approache to other
males,@ looking at her cloaca, @ placing his
snout near her cloaca, © scratching around her
vent, @® bestride the back of the female, @
grasping the base of the egg sacs with his
hindlimbs to assist in the delivery of the eggs,
@® brisk male activity around the female, and @
the end of fertilization.

BE 3

COPULATION BEHAVIOUR AND FECUNDITY IN THE
ASIAN TADPOLE SHRIMP (Triops granarius)

Y. Okuno. Dep. of Biol. Osaka Kyoiku Univ.,
Osaka

When two Asian tadpole shrimps
encountered each other, one of them
interwound itself with the other. A_ total
of 93 encounters in a rice field were
observed during the present study. Among
them, 49 cases were observed between males
and 44 between the opposite sexes, but none
between females. The expected values when
individuals meet randomly were calculated
and compared to the observed ones.
Supposed that only the male initiates the

interwinding behaviour, there is no
Significant difference between the observed
and expected values. Observation on the

video film proved that the success of
coitus was due to the female's acceptance
of this behaviour.

For the estimation of fecundity, one
female and two males reared in ae small
tank. A method for collecting eggs with
more than 98% probability was devised.
Eggs were collected and counted every two
days. Maximum records of eggs produced per
day and during her lifetime were 406 and
3741, respectively. Although Takahashi
('80) estimated production at 29 eggs per
day with a total lifetime production of 410
eggs, the results of this study were
approximately tenfold.

Behavior Biology and Ecology LS

BE 4 BE 6
ENTRAINMENT OF SWIMMING ACTIVITY RHYTHM IN CLOCK MUTANTS FROM NATURAL POPULATION OF
THE CUMACEAN DIMORPHOSTYLIS ASIATICA DROSOPHILA MELANOGASTER IN YAMAGUCHL
(CRUSTACEA) BY LIGHT AND HYDROSTATIC PRES- T. Murata, A. Matsumoto, K. Tomioka and Y.
SURE CYCLES. Chiba. Biol. Inst., Yamaguchi Univ.,
T. Akiyama. Ushimado Marine Laboratory, Yamaguchi.
Univ. of Okayama, Okayama. ooo
24 hr light cycles and 12.5 hr hydros- Locomotor activity of the fruit fly
tatic pressure cycle entrained swimming ac- Drosophila melanogaster is under circadian
tivity rhythm in Dimorphostylis asiatica. control. Mutations affecting this rhythm
Light-entrained activity occurred just have been isolated by using mutagen. Among
after light off. In some cases unentrained them, period gene on the X chromosome has
rhythmic component longer or shorter than been extensively studied at a molecular
24 hr also occurred during the dark period level.
along with the light-entrained component. We investigated the periodicity pheno-
About 0.3 atm amplitude of hydrostatic type of D. melanogaster by recording the
pressure cycle which simulated 3 m locomotor activity of individual flies in
amplitude of tidal fluctuation entrained iso-female or inbred lines established
12.5 or 25 hr period rhythmic pattern in from a natural population of Yamaguchi.
which the activities occurred on pressure As the result, two strains of clock
decreasing phase. Few unentrained rhythmic mutants (RitsuA and RitsuB) were obtained
component occurred under the condition. 6 in the inbred lines. RitsuA and RitsuB
One lZ hr shift, of light: ‘cycle -and 6 “hr show average period lengths of 27.3+ 0.8
shift of pressure cycle induced phase (SD)hr and 27.74+1.4hr under constant
shifts in the rhythmicity which accompanied darkness, respectively, which are three
transients, suggesting that the both sig- hours longer than that of the standard-
nals are synchronizers of endogenous timing type strain Canton-S(CS). There is no sig-
system in the animals. When animals were nificant difference in length of period
subjected to light and pressure cycles length between thies strains . The major
concurrently, the activity occurred on the gene(s) involved in lengthening period are
pressure decreasing phase during the dark located on the second chromosome. This was
period. Some animals exhibited rhythmic clarified from a chromosomal analysis
pattern which composed of light-entrained using strain synthesized by chromosomal
and pressure-entrained rhythmic components substitution between RitsuA and CS.

Simultaneously.

BE 5 BE 7

AUTOMATIC ANALYSIS OF THE FEEDING BEHAVIOR CHRONOBIOLOGICAL MECHANISM OF CIRCABIDIAN
IN DROSOPHILA AND LONG TIME AUTOCORRELATION. RHYTHM. 1 1
I. Shimada, Y. Minesaki, Y. Hayakawa, Y. Y.Chiba ry N.Toyofuku 5 -Matsumoto ,
Kawazoe and H. Hara. Tohoku Univ., Sendsi. K.Tomioka and Y.Ikeda“. ‘Biol. gnst.,

Yamaguchi Univ., Yamaguchi, Zeria

We developed the automatic analyzing SRO eae tenes

System of the feeding behavior in Drosophi-
Wap whachs ls, composed of the X=¥ disp llace=
ment detector (IV540, Hoei Co.). With this
ee aiytiinde conpenentay feo ane clrcadian
: : F and the other is circabidian (abou

ee ieee ene Sit Seonerae h)(Ikeda and Chiba 1982). The circabidian

Sapte : : rhythm has been known also in man (Jouvet
different angles of observation, most C(t) et. al. 1974) and dmsect (cockroach, Page

were resulted in inverse power-law. We 1 z : : A Gil 1984).
have already suggested the fractal property Too uC MN GOES Ee See ee a Re, ae
of ane feeding ES DAuAOe TT ee ese circadian components as in the above-
SoG ee ae mentioned rat, making it possible to
sults will _ ort the isrevetetsrte3 of the specucake Chat; Phen shontes .components
fractal EBSA behavior in nature BIDE ENCE tee) HORI CMO Leger Velo
& : 1984). In a comparative study with
mosquitoes of Culex pipiens complex, we
obtained a result agreeable roughly to
this speculation. A strain of C.p.
molestus from Ebira, Nagasaki revealed
two circadian components, period of which
is stable in one but is lengthened by
increasing light intensity in the other.
On the other hand, the circabidian rhythm
was sometimes seen under LL; activity
peak was much lowered or totally
disappeared every other day. However,
this speculation should be changed for
the better, because the lowering of
circadian peak often continued for
consecutive two days.

Locomotor activity of rat given d-
amphetamine chronically shows three

1176 Behavior Biology and Ecology

BE 8

CHOICE LIGHT/DARK AND STRIDULATORY ACTIVI-
TY IN THE CRICKET, GRYLLUS BIMACULATUS.
I. Nakatani. Department of Biology, Facul-
ty of Science, Yamagata University, Yama-
gata.

The choice between light and darkness
by male crickets, Gryllus bimaculatus, was
examined under conditions in which the
cricket was able to choice between light
and darkness for 12 h each day (CD12:12)
or at all times (CA). Under CD12:12, the
choice was irregularity from the last-
instar nymph through adult. The mean
percentage of daily choices of darkness
per 12 h was 57.9 + 0.7, and the cricket
spent mean 79.0% of each day in darkness.
The stridulation occurred in synchrony
with the phase when the electricity for
the fluorescent tube was disconnected.
Under CA, there was no rhythmicity in the
choice between light and darkness until

the beginning of stridulation. The mean
rate of daily choices of darkness
decreased after the beginning of

stridulation from 51.4 + 1.9% to 35.6 +
1.3%. The stridulation occurred at lights-
on and its rhythm exhibited free-running
periodicity. Two circadian pacemakers can
be proposed for stridulatory activity, the
first operates at lights-on under CA and
the other at darkness under CD12:12. In
the field, crickets may go in and out from
their shelters irregularly, Iie thus
activity is considered without reference
to temperature cycles.

BESS

MUTUAL INTERACTION BETWEEN NYMPHAL OPTIC
LOBE CIRCADIAN PACEMAKERS AFFECTS ADULT
FREERUNNING PERIOD IN THE CRICKET, GRYLLUS
BIMACULATUS.

iKGu wonton Bunch Yq Sinstiog 4 alOd6. WmsSige,
Yamaguchi Univ., Yamaguchi.

The freerunning period (FP) of the
cricket,Gryllus bimaculatus, was found to
vary not only dependently on the lighting
clondastaon but la lsio wa thet hie muiteual
interaction between optic lobe (OL)
circadian pacemakers. We examined the FP of
four groups (G1, G2A, G2B, G3) of male
crickets, receiving, as 7th instar,
unilateral removal of OL (G1) or compound
eye (CE) (G2), or removal of unilateral OL
and the contralateral CE (G3), and being
reared under constant light (LL). Hence,
pacemaker in G1 and on intact side of G2
animals was exposed to LL, while one in G3
and on operated side of G2 animals to
constant darkness (DD). G2 animals was
further subjected to a unilateral OL
removal on either intact (G2A) or operated
side (G2B) just before an assay of
locomotor activity. After the imaginal
molt, their locomotor activity was recorded
under constant darkness at least for 3
weeks. FPs of G1 (24.58 +0.51, n=15) and G3
(2B168s 019 pen=6))" signastcant ly. distrened
QGaACin CBieiieic, alinclaGeresame elne alee
information from the CE substantially
affects the FP. FPs of G2A (23.97 +0.64
n=7) and G2B animals (24.06 +0.45, n=16)
were intermediate between G1's and G2's FPs
and significantly different from both of
them. These results indicate that the FP
was also affected by mutual interaction
between bilateral OL pacemakers.

BE 10
SEASONAL CHANGE IN THE LOCOMOTOR ACTIVITY
RHYTHM OF LOACHES: EFFECTS OF

ENVIRONMENTAL FACTORS-II.
M. Naruse and T. Oishi, Dept. of Biol.,
Fac. of Sci., Nara Women's Univ., Nara.

We previously reported seasonal change
in the locomotor activity rhythm of
loaches in an outdoor water tank. In
winter, the locomotor activity rhythm
showed a typical diurnal pattern, but it
became arrhythmic (active during both day
and night) in summer. Analyses of the
relation between environmental factors and
the seasonal change of locomotor activity
rhythm revealed the importance of light
(photoperiod) and temperature.

To elucidate the role of environmental
factors on the locomotor activity rhythm
of loaches, we performed indoor
experiments under artificially controlled
environmental conditions using individual
loaches. Either light-dark cycle (LD) or
continuous darkness (DD) was combined with
constant or cyclic temperature. Under LD
conditions, the activity mainly showed
nocturnal patterns, but it tended to show
diurnal patterns under low temperature
(both constant and cyclic temperatures).
Diurnal activity pattern became more

prominent under cyclic temperature
conditions than under constant temperature
conditions. Since distinct activity

rhythms were observed under DD and cyclic
temperature conditions, cyclic temperature
must be an important factor to induce the
locomotor activity rhythm of loaches.

BE Wt

INFLUENCE OF ENUCLEATION ON THE LOCOMOTOR
ACTIVITY RHYTHM OF Xenopus laevis.

H. Fujisawa and T. Oishi. Dept. of Biol.,
Fac. of Sci., Nara Women's Univ., Nara.

We recorded the locomotor activity
rhythm of intact and blind adult Xenopus
laevis under light-dark cycles (LD12:12)
with three different light intensities
(2010 =4100;, > =25 and” OS 02) ais) cuaasinc
photophase and under constant darkness
(DD). All the intact frogs displayed
typical nocturnal activity rhythms under
LD with each light intensity. About 803%
of the blind frogs displayed nocturnal
activity rhythms under LD with high and
intermediate intensities, but the
percentage decreased to about 60% under LD
with low intensity. The results suggest
that extraretinal photoreception is
involved in the entrainment of the
circadian rhythm to LD, although the
sensitivity to light in blind frogs seems
to be less than in intact frogs. Under LD
with high intensity, about 60% of the
blind frogs showed a temporal increase of
activity responding to light-on, but this
response was not recorded under
intermediate and low light intensities.
Since intact animals did not show the
response, the eyes might have inhibited
such response. Free-running rhythms (7 )
under DD were shown in about 90% of both
intact and blind frogs and the (S$ of blind
animals were shorter than those of intact
ones, suggesting that the eyes have some
role in the circadian clock(s) system.

Behavior Biology and Ecology C7,

BE 12

FEEDING BEHAVIOR AND LATERAL LINE SYSTEM
IN AXOLOTL, AMBYSTOMA MEXICANUM, II.
EFFECTS OF BLOCKING THE MECHANOSENSORY
LATERAL LINE SYSTEM BY COBALT IONS.
H.-A.Takeuchi and H.Namba. Dept. of Biol.,
Fac. of Sci., Shizuoka Univ., Shizuoka.

The previous study suggested that
mechanosensory, electrosensory and cutane-
ous Signals mediated by anterior lateral
line nerves were involved in the release
of feeding behavior in axolotl (Takeuchi &
Namba, 1989). In present study, we phar-
macologically blocked the lateral line
system by cobalt ions that selectively and
reversibly suppress the mechanosensitivity
of the lateral line organs, and quantita-
tively analyzed the effects on the feeding
behavior evoked by vibrational stimuli
(frequency: 5 Hz). The feeding responses
were Significantly reduced in the animals
exposed to 0.05-1.0mM Co2+ solution for
one hour. The animals showed no sign of
general behavioral abnormality during Co?2+
treatment and recovered their normal level
of feeding responses 2-6 days after Co2+
treatment, indicating that the inhibiting
effects of Co2+ on feeding behavior were
selective and reversible. The strength of
inhibiting effects and the length of
recovery periods tended to depend upon the
Co2+ concentration of the exposure solu-
tion. The present results suggest that
the mechanosensitivity of the lateral line
system are important for the release of
feeding behavior in axolotl.

BE 13

ACOUSTICAL AND PERCEPTUAL STRUCTURES OF
DISTANCE CALLS IN BENGALESE FINCHES.
K.Okanoya’’, T.Kimura’, and K.Aoki’. ‘Life
Science Institute, Sophia University. *Bird
Control Laboratory, National Agriculture
Research Center, Tsukuba.

Species-specific distance calls were
recorded from twelve female and twelve
male Bengalese finches (Lonchura
domestica). Acoustic parameters such as
bandwidths, peak frequencies, frequency
and amplitude modulation rates were
obtained from sound spectrograms.
Differences between the sexes were
significant on most of these parameters.

We next examined perceptual salience of
these parameters using operant condition-
ing techniques and a multi-dimensional
scaling (MDS) procedure. Eight calls from
both sexes were selected and reaction time
matrices for detecting differences among
male calls and among female calls were
constructed separately.

The MDS analysis revealed that birds
were attending to the bandwidth and pitch
of the calls when listening to the female
calls. For the male calls, birds were
attending to the ending frequency, the
frequency of maximum amplitude, and the
frequency of the second maximum amplitude.

These results suggest that Bengalese
finch distance calls are not only sexually
dimorphic in acoustic dimensions, but also
analyzed differently in perceptual
dimensions. (Work supported by JSPS)

BE 14

DIFFERENCE OF THE PERIODS OF CIRCADIAN
ECLOSION RHYTHM BETWEEN TWO STRAINS OF
THE HOUSE.FLY, MUSCA DOMESTICA.

M. Aizawa and H. Yoshino . Kurihara
gricultural High School, Miyagi,
Furukawa High School, Miyagi.

We previously reported (1986) that when
strains of house fly were grown at temper-
atures between 19° - 28°, the period of
eclosion rhythm was different between the
strains P-9 and wbp, showing their inter-
mediate in their Fl. To exclude the effect
of temperatures upon the above results,
the period of eclosion rhythm was analysed
this time under a constant temperature and
using populations of different instars,
since the eclosion maxima appear periodic-
ally when the periods of larvae under DD
were changed.

Eggs were harvested for 4 days under
the conditions of constant temperature and
a fixed harvest time, and grown ina
culture bottle. The one bottle becomes a
culture of mixture of different instar
larvae. In the beginning 7 days they were
kept in LD 12:12 and then kept in DD. In
this condition, individuals in the bottle
were under the DD of different periods
between 0 to 4 days. During the DD and
the period of pupa, the flies measure the
time with their endogenous rhythm and
eclose at a certain time 2 weeks after
their hatching. The results showed that
the time of the eclosion maximum appears
periodically and the periods agree with
the previous results.

BE 15

SETTLEMENT PATTERN OF A MARINE VORTICELLA
H. Horikami, K. Ishii, Laboratory of Biology, Hosei
University, Ciyoda-Ku, Tokyo

Freshwater Vorticella can be conveniently classi-
fied into two types, grouping(G) and solitary(S) type,
being based on a mode of their settlement on a sub-
strate. The former settles in a group and the latter
solitarily at random. According to the criterion, V.
campanula, V. convallaria, V. monilata and V. vestita
belong to the G type, while V. octava andV. microsto-
ma to the S type.

Furthermore, specific cell's profile is found in
each type. A ratio of the body width(BW) or the body
length(BL) to the peristomal width(PW) in the G cells
is smaller than it in the S(G type, BW/PW=0.7~0.9,
BL/PW=0.9~1.2; S type, BW/PW=0.9~1.5, BL/
PW=1.8~2.4). And a ratio of the BL to the posterior
cilliary writh width(PCW) in the G's telotrochs is also
smaller than it in the S(G type, BL/PCW=1.3~2.3; S
type, BL/PCW=3.2~3.9)

The cell's proportion and the settlement pattern in
a marine species were investigated. Average values
of the BW/PW and the BL/PW in the adult cells are
1.2+0.1 and 1.9+0.2(n=103) in each. The telotroch's
BL/PCW is 2.9+0.3(n=129). These values compare
with those of the S type. However, the telotrochs
were attracted to the settled young cell and settled
in a group.

The relationship between the cell's profile and the
settlement pattern in freshwater Vorticella was not
found in a marine species. The interpretation of
these results was discussed.

1178 Behavior Biology and Ecology

BE 16

LIFE CYCLE OF PARAMECIUM BURSARIA SYNGEN 1
IN NATURE.

ils Kosaka. Zool. Inst., Fac. of Se. ,
Hiroshima Univ., Hiroshima.

Studies were done on the natural population

density of Paramecium bursaria, as well as
on the stages in the life cycle to which
individuals collected belonged. Collections
of paramecia from two streams in Hiroshima
Prefecture were carried out once every 20
days for over one year. A total of 413 indi-
viduals were collected from 26 times of col-
lection in Mikumarikyo stream, while in
Momijidani-gawa stream, a total of 83 indi-
viduals were collected on 23 collection
dates. Individuals in nature were not main-
tained at a steady density; rather there
were great changes in numbers depending on
months through a year. It seems that conju-
gation occured in nature from April to June
in Mikumarikyo stream and from May to June
in Momijidani-gawa stream. The appearance
of individuals with mating ability might be
closely related to the increase in the num-
ber of individuals. Sexual reproduction
probably occurred near the peak of the
population density. Of 413 individuals from
Mikumarikyo stream, 309 individuals(74%)
were immature, and 55(13%) and 49(12%) were
adolescent and mature, respectively. Of 83
individuals from Momijidani-gawa stream, 44
individuals(52%) were in the immaturity pe-
RLOdsa and, a2lC25%) mands al8C2ZilA ye “weres sine the
adolescent and mature periods, respectively.

BE 17

ON THE STUDY TO THE LIFE-CYCLE OF A RARE
"KOUGAIBIRU": BIPALIUM PENNSYLVANICUM.

N. Makino and Y. Shirasawa,Dept.of Biol.,
Tokyo Med. Coll.,Tokyo.

In Tama-Kyuryo,Mogusaen in Hino-City,
senior author found first several worms in
1983. Authors reported some morphological
structures of them, especially the sexual
organ and the karyotype(1984-1986).

After that,the same species were collect
ed in Pennsylvania,USA and Dr.R.Ogren name-
ed the new species,Bipalium pennsylvanicum.
From 1983,authors have cultured the worms
in the laboratory,keeping 18 C. In the
natural environment,the worms oviposition
in April to May,after the copulation.Then,
they die out with one year after the hatch-
ing.In the laboratory,similarly the worms
die after the oviposition,having 500mg, but
in the laboratory,a few worms lived over a
half year after the oviposition. This
species often do the copulation in breed-
ing season and they oviposition several
times. Some worms were born in the
laboratory in 1985, did the copulation
after one year from the hatching,but they
did not oviposition,when they were two
years old the copulation was observed and
they ovipositioned after several days,and
the larva hatched out after about thirty
days. Those parent worms died after four
years from hatching(1985). When the worms
were born their weight were 4-15mg,after 1
year they became 500mg,and 2 years(1000mg)
,3 years(1500mg) and 4 years(2000mg) ,and
then they died breaking the bodies. In
addition, this species propagate with only
bisexual reproduction and never asexual
(fission) reproduction.

BE 18

COLD-HARDINESS OF THE WINGED ANTARCTIC
MIDGE PAROCHLUS STEINENII IN THE SUMMER
SEASON AT KING GEORGE ISLAND.
K.Shimadal,Y.Ohyama2 and C.X.Pan3.
lInst.Low Temp.Sci.,Hokkaido Univ. ,Sapporo
2Nat.Inst.Polar Res.,Tokyo and 3Inst.Zool.
Acad.Sinica,China.

We joined the 6th Chinese National Ant-
arctic Research Expedition and stayed at
the Great Wall Station (63°13'S,58°58'W) on
King George Island from December,1989 to
March,1990. During the stay we collected
adult and immature P.steinenii (Diptera:
Chironomidae) from shallow streams and
lakes near the station. The collected ani-
mals were examined for their supercooling
Capacity and lower lethal temperatures.The
individuals were attached to a fine thermo-
couple for measurement of the supercooling
points or confined to a vial with and/or
without 1ml of water for assessment of the
lower lethal temperatures. The specimens
were covered with a Thermos flask and cool-
ed in a freezer at 0.1-0.2°C/min.The super-
cooling points of the larvae, pupae and
adults ranged from -5 to -11 (mean+SD=-7.4
+170), from =8 to —23 (-16.344.6) and from
-3 to =23°C (-15.345.6°C) ,respectively.The
median lower lethal temperatures of the
larvae,pupae,adults and eggs suspended in
water were about -3,-9,-7 and -5°C, respec-
tively. From the ecological point Of view
it seems that P.steinenii except adults on
the ground become fatal at temperatures
above their supercooling points due to the
occurrence of inoculative freezing.

BE 19

PHOTOPERIODIC CONTROL OF ALARY DIMORPHISM
AND OVIPOSITION IN A WATER STRIDER,
GERRIS PALUDUM INSULARIS. ;

Te Harada! and K. Taneda2. 1Dept. of
Biol., Fac. of Sci., Osaka City Univ.,
Osaka and 2Dept. Of Biol, hacen Ot oCHltar,
Kochi Univ., Kochi.

Absolute long-day photoperiods in the
nymph stage induced high frequency of
brachypters (>65%) and oviposition (>80%
of paired females). On the other hand,
absolute short-day photoperiods induced
low frequency of brachypters (<35%) and
did not evoke oviposition at all during 60
days after the last ecdysis. The critical
photophase for wing-morphs is between
13.5h and 14.0h, whereas it for oviposi-
tion is between 12.0h and 13.5h.

Stepwise decrease in photophase (by
15 min. every 3-5 days) from 16.5h to
14.5h in the nymph stage had no effects on
alary dimorphism and oviposition.

However, stepwise decrease in photophase
as shown above from 14.5h to 13.5h close
to the critical photophase delayed the
onset of oviposition by about 10 days.

In the field, the adults of the 2nd gen-
eration of offspring (nymph stage: mainly
the latter half of Aug.) lay eggs, al-
though most of them are macropters (Harada
and Taneda, 1989). The longer critical
photophase for wing-morphs than it for
oviposition seems to be a factor of the
above-described phenomena in the adults of
the 2nd generation of offspring.

Behavior Biology and Ecology 1179

BE 20

ADULT ECLOSION RHYTHM IN MAMESTRA BRASSICAE
(LEPIDOPTERA: NOCTUIDAE).

Y-Watari-~;, T,Arai;, Y Nakajima! and M.
Suzuki-. Lab. of Biotech.,Asiya Univ.,
Asiya and ?Kaminoyama Yogo School, Kamino-
yama.

In M. brassicae, the adult eclosion was
observed during the hour close to dawn
under natural conditions. There are no
differences in the pattern of emergence
time between the nondiapause and summer-
diapause pupae. We examined the effects of
light-dark cycle and thermoperiod on the
time measuring mechanisms of the adult
eclosion. When the pupae were exposed to
the photoperiod or the thermoperiod, the
eclosion time was affected by them. In 12L
:12D at various temperature levels(15, 20,
25 and 30C), the eclosion occurred close to
the light off. Subjected to the thermo-
period each consisting of 12h warm phase
and 12h cool phase in DD or LL, however,
the eclosion peak appeared at around the
temperature fall in various temperature
eyelles(20 156. 125-206. 25:15C2 30:25, 30):
2GC rand 30215C).

When a thermoperiod and a photoperiod
were combined in various phase relation-
ships, the emergence was controlled by
both cycles. The adult eclosion rhythm
(about 24h) was initiated by the transfer
EOD etereom 232 D) on thes liZth days at ver
PUupaLTtone bul NOt by teransher town “ihe
results show that the eclosion timing of
M. brassicae may be controlled by
circadian rhythm.

BE 21

BIOCHEMICAL STUDIES ON THE CASTE
DIFFERENTIATION OF JAPANESE DAMP-WOOD
Beet

K.Ogino , ¥-Hjrono?! G.Shinjo? and
Heishwkawas. ZOO. Inst., Fac. Sci. , 2Dept.
of Biol. Cgll. Arts & Sci., Univ. of Tokyo,
Tokyo and ~Sumitomo Chemicals Co., Ltd.,
Osaka.

The caste differentiation of termites is
regulated by juvenile hormone(JH) titer in
the haemolymph. When the titer falls below
a critical level, the termite
differentiates into a reproductive. When
the titer exceeds another critical level,
then it molts to a presoldier. Although it
has been known that the differentiation of
workers into presoldiers is induced by JH
analogues(JHA) such as methoprene, its
molecular mechanism is yet to be known.

In this study, we induced workers and
nymphs of Japanese damp-wood termite
Hodotermopsis japonica into presoldiers
with a new JHA, S31183. The workers and
nymphs that had fed on the filter paper
containing $31183 differentiated into
presoldiers at relatively high level, 70-
80%, within 4 weeks. Comparison of 2D-
electrophoresis patterns of different
castes revealed that presoldiers and
soldiers have a specific protein whose
molecular weight is about 24kDa. It was
shown that its synthesis as early as 6days
before the molt to presoldiers when the
bodies of termites turn white signifying
the beginning of molt.

BEY2

SOLDIER AND MALE OF A SOLDIER-PRODUCING APHID,
COLOPHINA ARMA, LACK ENDOSYMBIONT.

T. Fukatsu and H. Ishikawa. Zoological Inst., Fac.
of Sci., Univ. of Tokyo, Tokyo.

In an attempt to better understand the aphid
sociality, we studied its anatomical, physiological
and biochemical aspects for the first time, with
reference to the endosymbiotic system using
antiserum against symbionin, an aphid endosymbiont-
specific protein. We studied Colophina arma, an
eusocial aphid with 1st instar soldier. The major
results obtained by westernblot analysis and
immunohistochemical studies were as follows;

1) The soldier and the male lack endosymbiont.

2) In the soldier, not only the symbiotic but the
reproductive system degenerated and was replaced by
well-developed fat body cells.

The endosymbiont in aphid is transmitted to the
next generation through maternal inheritance. Thus,
the lack of endosymbiont in the male may be due to
the symbiont's selfish strategy so as not to infect
the host that is useless in terms of its trans-
mission. The same will be true for its lack in the
soldier because the caste dies in the 1st instar
without producing progeny.

Well-developed fat body of the soldier seems of
great physiological importance. The soldier cannot
feed phloem sap with its very short mouthpart which
has been acquired through adaptation to attacking
the predator. At the same time, however, it has to
live long enough to defend its colony. The well-
developed fat body with much reserve substances may
provide one way to satisfy the self-contradictory
requirement 'Live long but never feed!'.

BE 23

A STATISTICAL ANALYSIS OF SEASONAL FLUCTU-
ATIONS OF POPULATION SIZES IN Drosophila
POPULATIONS NEAR HUMAN HABITATION AT
HIMEJI CITY

K. Mikasa Dept. Biology, Sch. of Dent.
Meikai Univ. Sakado

Monthly collections were carried out at
Himeji, from June to November of 1984 and
from May to November of 1985. Twelve
collection sites are near human habita-
tion. In grand total, 8454 flies belong
to genus Drosophila were collected and
classified as 22 species. D. lutescens,

D. melanogaster, D. simulans, D. auraria,
and D. immigrans were dominant species
in those samples. Dey .ComacainaleD).

busckii, D. suzukii, D. nigromaculata,
and D. bizonata were subdominant species.
Multiple regression analysis was conducted
with respect to the monthly total of
collected flies in each species over the
monthly records of temperature, precipita-
tion, and humidity. Significant linear
multiple regression equation was detected
in D. auraria, D. immigrans, and DE
busckii. Coefficients of linear multiple
regression over precipitation and humidity
were statistically significant in D.

auraria and these over precipitation in D.
busckii and D. immigrans. None of signif-
icant coefficients of linear multiple
regression over temperature was detected
in the other species. The former three
species feed on decaying leaf subsidiari-

lives

1180 Behavior Biology and Ecology

BE 24

FOOD CACHING BEHAVIOR IN SMALL JAPANESE FIELD
MICE (Apodemus argenteus).

S.Koyama and M.Daumae. Dept. of Psychol., Tokyo
Woman's Christian Univ., Tokyo, Dept. of Biol.,
Coll. of Arts and Sci., Univ. of Tokyo, Tokyo.

Food caching, the movement of potential food
items from one location to another before con-
suming them at later time, is often observed in
the small Japanese field mice (Apodemus argente-
us). We assessed caching tendency in A. ar-
genteus under laboratorial condition. Adult mice
cached about 50 % of peanuts provided as food,
while young mice (25 - 30 days of age) cached
about 22 %. These facts suggests the occurrence
of caching behavior is related to maturation.
Under the intermittent feeding condition, A.
argenteus did not show larger caching tenclency
than under the continuous feeding condition. In
order to assess the ovmership of cached food, we
tested their ability to search their cached
foods or other's cached foods. The field mice
were separately cagecl in large cage to let them
cache the food. 3 days later after removing the
mice from the cached cage, hungered mice were
put into cached cage. In its own cage, 90 3 of
mice found the cached peanuts to eat within 10
minutes, but only 27 % of mice could get peanuts
in the cage in which other individual cached.
After olfactory block by the injection of 4 3%
ZnsO, solution into nostrils, 75 % of mice
failed to search their own cached fools. These
results suggest that A. argenteus has some
methods to find the location of cache sites
which may related to their olfactory sense.

BE 25

PREFERENCE OF FEMALE MICE FOR MALE ODOR IN
THE LIGHT PERIOD

K. Ninomiya and T. Kimura. Dept. of Biol.,
Coll. of Arts and Sci., Univ. of Tokyo,
Tokyo

Female mice prefer to huddle together
in the light period. In order to study the
relationship between this huddling tenden-
cy and preference for male odor in the
light period, we estimated the preference
of sexually experienced females for male
urine odor with or without the presence of
other females which were kept behind a
wire-mesh barrier. Resting positions of
test females were videotaped for 14 hrs
and analyzed.

They showed no preference for male odor
when put into the test arena alone, but
when other females (also experienced) were
present behind the barrier, test females
(both estrous and diestrous) showed clear
preference for male odor. This effect of
live females could not be reproduced by
female urine odor. Interestingly, naive
females showed no preference even in the
test with the presence of other females.
Finally we tested the preference of expe-
rienced females when other naive females
were present behind the barrier. They
showed no preference for male odor and
diestrous test females rather avoided it.

It is likely that in the light period,
preference of female mice for male odor
depends on the presence of other females
and that the behavior of these other fe-
males influence the presence.

BE 26

REGULATION OF PATERNAL BEHAVIOR IN MICE
F.Matsumoto and T.Kimura. Dept.of Biol.,
Coll.of Arts and Sci.,Univ.of Tokyo,Tokyo.

Male mice which had copulated and co-
habited with the mate until her delivery
showed various paternal behavior patterns
with much higher frequency than naive
males. In order to determine which exter-
nal factor is crucial for males to be
paternal, effects of copulation-experience
and those of cohabitation with a pregnant
female was separated and analyzed.

In males which were isolated just after
copulation, incidence of paternal behavior
starts to increase as soon as 4 days after
copulation, and reached to significantly
higher level as compared with naive males
at 13 to 15 days, but the incidence fluc-
tuated before and after this period. It is
possible that the experience of copulation
alone can induce the paternal behavior,
but the effect is rather unstable.

In the next experiment, sexually naive
males were individually housed for 5 days
with a female but separated from her by a
wire-mesh barrier. Paternal behavior of
males were not enhanced by cohabitation
with either non-pregnant, early pregnant
or late pregnant females. However, if the
delivery occurred during the period of
cohabitation, males showed high incidence
of paternal behavior thereafter.

It is likely that the encounter with
delivery rather than cohabitation is the
important factor for the initiation of
paternal behavior.

BE 27

RELATIONS BETWEEN AGGRESSIVENESS OF MALE
MICE AND A STRANGE MALE ODOR

S. Hayashi. Dept.Biol., Fac.Educ.,.
Kagoshima Univ., Kagoshima.

To investigate aggression of male mice
they were bred under different conditions.

Some male mice were reared with a
castrated male until 104 days of age and
then they were isolated for 14 days ina
cage stained by strange intact males. The
males which had sniffed at an intact male
until 104 days of age were subordinate not
only to males which had not experienced
such an odor but also to males which had
sniffed at a preputialectomized male al-
though they showed similar aggressiveness
toward a castrated male.

The other males were reared with intact
males until 104 days of age and then they
had been isolated for 7 days. The mice
which were isolated in a cage stained by
an intact male dominated males which had
stayed in the original cages although they
showed similar aggressiveness toward a
castrated male, too.

On the other hand, males which had been
isolated in a cage stained by an intact
male attacked on a castrated male more
frequently than males which had been
isolated in a cage stained by a preputia-
lectomized male. However, there was no
dominant-subordinate relationship between
them. It seems that there is a discrepan-
cy between social dominance and aggres-
siveness.

Behavior Biology and Ecology, Taxonomy and Systematics 1181

BE 28

EFFECTS OF DIFFERENT PERIODS OF ISOLATION
ON OPEN FIELD BEHAVIOR IN MALE MICE.
M.Kawamura, Y.Wada, T.Machida and Ts
Noumura. Dept. of Regulation Biol., Fac.
of Sci., Saitama Univ., Urawa.

Manipulation of social environments re-
sults in significant modifications of
physiological and behavioral responses
in rodents. We already reported a positive
correlation of measures of the open field
behavior with housing density in mice.
In the present experiments, CD-1 male
mMiLGe @O~ Ail; 23, 35, 56 cine 824 Clee Olle
were housed singly for 0, 4, 16 or 64
days and effects of social isolation on
measures of the open field behavior were
investigated. In mice of 0 day isolation
group, age-associated increase in loco-
motor activity was observed from 21 to
84 days of age. Single housing caused
an increased locomotion in 21- and 28-
day-old animal, while in mice of 35 to
84 days of age isolation resulted in a
decrease in locomotion. Measures of emo-
tional reactivity, i. e., scores of groom-
ing and defecation, generally decreased
following single housing. Daily scores
of these behavioral measures became more
variable and unstable following prolonged
isolation. Social isolation of male mice
thus exerted an intense influence on loco-
motor activity and emotional reactivity.

TS 1

SPECIES DIFFERENTIATION OF THE RICEFISH IN
THAILAND: KARYOTYPE EVOLUTION AND GEO-
GRAPHICAL DISTRIBUTION OF THE THAI-MEDAKA,
ORYZIAS MINUTILLUS. “

2

W. Magt on!, N. Nadee“, T.Higashitani~, K.

Takata~ and H. Uwa~. Fac. of Sci., Srinak-
harinwirot Univ., Bangkok, Wines UNSiws Oi
ci. and Tech. Res., Bangkok, Thailand,
Fac. of Sci., Shinshu Univ., Matsumoto.

The Karyotype of QO. minutillus was
observed with specimens collected from 18
localities in Thailand. Specimens from the
South; Phuket, Songkhla, Surat Thani and
Chumphon, and the Northeast; Bua Yai,
Phimai and Phayakkhapum Phisai, had 2n, 42
acrocentrics (2n=42, 42A, NF=42, NORs-
Atm) . Specimens from the Central; Bang-
kok, Chai Nat, Suphan Buri, Ratchaburi,
and Saraburi, and the North; Chiang Mai
and Chiang Rai, were characteristic by
having large metacentrics and NORs-
chromosomes of submetacentric type re-
gardless of the difference in chromosome
numbers (2n=34, 32, 30 or 28; 8M+2SM+24A,
10M+2SM+20A, 12M+2SM+16A, or 14M+2SM+12A;
NF=44, NORs-SMsa). Specimens from the
East; Chachoengsao, Nadee and Rayong,
were intermediate: they have no or one
pair of large metacentrics but NROs were
located on submetacentrics (2n=42 or 40
2SM+40A or 2M+2SM+36A, NF=44, NORs-SMsa)

The Karyotype with 42 acrocentrics is
basic for O. minutillus (Ashida and Uwa,
1987). Thus, geographical polymorphisms
of karyotypes seem to be caused by peri-
centric inversion and centric fusion.

TS 2

SPECIES DIFFERENTIATION OF THE RICEFISH IN
THAILAND: GENETIC DIFFERENTIATION OF
ORYZIAS MINUTILLUS.

K.Takata!, M.Hoshino!, N.Nadee*, W.Magtoon?
and H.Uwa!. !Dept. of Biol., Fac. of Sci.,
Shinshu Univ., Matsumoto, “Thai. Inst. of
Sci. and Tech. Res., Bangkok and *Dept. of
Biol., Fac. of Sci., Srinakharinwirot
Univ. at Bangkhen, Bangkok, Thailand.

An allozyme survey in 6 populations of
Oryzias minutillus from Thailand having
different chromosome numbers was conduct-
ed.

The 6 populations of O. minutillus were
divided into three major groups by cluster
analysis based on 22 loci. The first group
was consisted of the populations from
Chiang Mai, Chiang Rai and Bangkok, pos-
sessing smaller chromosome numbers (2n<34)
than other two major groups (2n=42). The
second group included the populations from
Chumphon and Phuket. The population from
Bua Yai was classified into the third
group. The second group of O. minutillus
having 42 chromosomes clustered with the
first group possessing smaller chromosome
numbers (2n<34) before it fused with the
third group with 42 chromosomes in the
dendrogram.

The genetic relationship estimated by
the current allozyme analysis supports
karyotype evolution that O. minutillus
populations having smaller chromosome
numbers evolved form a population having
42 chromosomes.

TS 3

GENETIC DISTANCE AND PROTEIN POLYMORPHISM IN TWO
SEA-URCHIN SPECIES OF THE ORDER ARBACIOIDA AND
IMPLICATIONS FOR THEIR EVOLUTION.

N. Matsuoka and Y. Nakamura. Dept. of Biol.,
Fac. of Sci., Hirosaki Univ., Hirosaki.

As a basic study to elucidate the evolutionary
relationship between the sea-urchins of the order
Arbacioida and Echinoida, enzyme variation and
genetic differentiation were investigated in two
sea-urchin species, Glyptocidaris crenularis of
the family Phymosomatidae and Stomopneustes
variolaris of the family Stomopneustidae, of the
order Arbacioida by electrophoretic analyses of 17
enzymes. The average heterozygosity per locus (H)
was in the range of 3.2-7.2% with a mean of 4.7%.
The degree of genetic variation was comparable to
that observed in many other echinoderm species
from shallow water such as G. crenularis and
S. variolaris, but considerably lower than that
from deep-sea. Of 17 enzymes assayed, the glucose
metabolizing enzymes of more strictly functional
constraint were on average less variable than the
non-glucose metabolizing enzymes. These results
on enzyme variation were not contradictory to the
neutral theory. The Nei’s genetic distance
between G. crenularis and S. variolaris of the
different families was 1.417, and comparable to
the D values between two closely related families
of the order Echinoida and those between sea-stars
of confamilial different genera. Although
these two species of the order Arbacioida have
been considered as sea-urchins having the old
evolutionary origins from morphological and
paleontological standpoints, the divergence time
of two species estimated from the genetic distance
was the relatively recent geological period
(the Miocene) of about seven million years ago.

1182 Taxonomy and Systematics

TS 4

ALLOZYMIC DIFFERENTIATION BETWEEN THE TWO
ETHOLOGICALLY DISTINCT TYPES IN THE FIREFLY,
HOTARIA PARVULA.

H.Suzuki’, Y.Sato®, S.Fujiyama® and N.Ohba’.
1Olympus Optical Co.LTD.,Hachioji, *Shinshu Univ.
Matsumoto and °Yokosuka City Mus., Yokosuka.

The firefly Hotaria parvula is separated into
two types on the bases of the difference in their
mating behavior, body size and geographic distri-
bution. Namely, the male of each type emittes
flashes at intervals of 1 and 0.5 seconds respec-
tively in searching flight for females. Body size
of the former type is larger(1 cm) than the latter

(0.6 cm). Further more, their both distributed
in Honshu but the small type is only found in
the west of Japan. We evaluated the degree of

genetic differentiation between them by the method
of electrophoresis of 14 different enzyme systems.
The average value of Nei’s genetic distance (D)
calculated from 22 loci for inter-typic pairwise
comparison of ten local populations was 0.11.
The dendrogram constructed with D values revealed
that small type populations formed one cluster
and large type populations were genetically dif-
ferentiated from each populations. Considering
other ethological, morphorogical and ecological
data, this biochemical result obtained here con-
firms that Hotaria parvula is separated into
two distinct species.

TS

BIOCHEMICAL SYSTEMATICS ON THE GENUS
CHAENOGOBIUS DEDUCED FROM ALLOZYME
POLYMORPHISM.

T.Aizawa,M.Hatsumi and K.Wakahama.
Dept.of Biol.,Fac.of Sci.,Shimane Univ.,
Matsue.

Allelic frequencies and genetic var-
iability based on 10 enzymes and 1 non-
enzymic protein were compared among six
species of the Goby,Chaenogobius,by use-
ing electrophoretic methods. The genetic
distances which estimated from allozyme
variation of 15 presumptive loci in six
species ranged from 0.213 to 2.013. This
indicates that genetic differentiation
proceeds intragenus. Especially,the de-
grees of genetic differentiations of
three species of similar external shape,
C.castaneus, C.laevis, and C.sp.3(tenta-
tive name "Shinziko-haze”), were almost
identical. The dendrogram from the ge-
netic distances indicates that these
species are roughly divided into two
genetically differentiated groups ;
small-type and large-type. And small-
type contains more genetic variability
than large-type.

TS 6

INTRASPECIFIC DIFFERENTIATION OF RANA
RUGOSA ELUCIDATED BY ELECTROPHORETIC
ANALYSES.

M. Nishioka’, M. Sumida’, Y. Kodama’ and M.
Ryuzakiz ‘Lab. for Amphibian Biol., Fac. of
Sci., Hiroshima Univ., Hiroshima and Dept.
Of Biol... 2Kitasato.. Unig School of

Medicine, Sagamihara.

In order to elucidate the genetic differ-
entiation of Rana rugosa in Japan, 14 kinds
of enzymes extracted from skeletal muscles
and livers and two blood proteins were ana-
lyzed by starch-gel electrophoresis in 45/7
frogs of the 29 populations. The results
showed that these enzymes and proteins were
controlled by 25 loci. At these loci, there
were 1 ~ 18, 5.4 on the average, phenotypes
produced. by, 1-> 7, 3.3 son. theweaverace:
alleles. The fixation indexes were 0~1.000,
0.323 on the average. Examination of the
geographic distribution of alleles at the
25 loci showed that there were distinct
differences between the eastern and western
groups in the alleles at the Pep-A, LDH-B
and Hb-II loci. In the 29 populations, the
rates of heterozygotes and polymorphic loci
were 9.4% and 30.8% on the average, respec-
tively, while the mean number of alleles
per locus was 1.40 on the average. Nei's
genetic distances among the 29 populations
were 0.003 ~ 0.474. When the phylogenetic
tree is constructed from genetic distances
by the UPGMA method, it is supposed that R.
rugosa has evolved into the eastern and
western groups, and then the eastern group
was roughly divided into three subgroups.

TS7

MORPHOLOGY OF A HYNOBIUS SP. WITH TRANSPARENT AND
SPINDLE TYPE EGG SAC FROM THE CENTRAL JAPAN.
H.Nambu. Toyama Science Museum, TOYAMA.

Morphological features of Hynobius sp. from
mountainous area of the central Japan reported by
Nambu(1990) which has tansparent and spindle-type
egg sac were compared with those of H. nigrescens
from lowland of Toyama Pref. and H.sadoensis from
Sado Island with white and spidle-type egg sac.

Snout vent length of the present species is
shorter:the present species, n=13, 66. 7+0.93(58. 7-
72.1);H. nigrescens, n=19, 74.3+0. 96(66. 1-83. 2) ;H.
sadoensis, n=13, 73. 7+0. 54(70. 5-77.8);shown by
means+SE and ranges in mm.The ratio of vent lengt
to head length of the present species is larger:
the present species, n=13, 2.72(2. 49-2. 89) ;H. nigre-
scens, 2. 93(2. 52-3. 39) ;H. sadoensis, n=13, 2.92(2. 70-
2.99);shown by median and ranges. Number of vomer
teeth of the present species are fewer:the present
species, n=13,39.1+1. 48(31-48), H. nigrescens, n=14,
55.6+1. 70(46-65), H. sadoensis, n=14, 47.14+1.11(40-
54). Skull of the present species differ from H.
nigrescens and H.sadoensis by following features:
thicker premaxillary and maxillary, larger pre-
frontal bone and slender vomer. Results of the
present observation indicate:1) Some of the fea-
ture of the skull and body of the present species
are differnt from the two species compared in this
study. 2) Features of the skull of the presen
species resemble H.sadoensis rather than H. nigre-
scens. The present species is related to H.sado-
ensis, although it may be new to science.

Taxonomy and Systematics 1183

TS 8

SYSTEMATIC POSITIONS OF THE THREE CONGLOBATING
FAMILIES OF TERRESTIAL ISOPOD CRUSTACEANS
N. Nunomura, Toyama Science Museum, Toyama

The systematic positions of the three conglo-
bate families, the Tylidae, the Armadillidae and
the Armadillidiidae are discussed.

The Tylidae has so many characteristic and
enigmatic morphological features, that its
systematic position has been puzzling. The Tylidae
has some "derived features” common to all the
members belonging to the suborder Oniscidea, such
as reduction of both antennae, reduction of
endopod of male 2nd pleopod, reduction of maxilla,
and presence of water conducting system of 7th
‘peraeopod. But some characteristic features such
as valve-like uropod, extreme reduction of
antennule, short and swollen peraeopods. These
chracteristic features show the relation to the
suborder Valvifera. The Tylidae might have been
derived from the ancestor common to Valvifera.

The remaining two families, the Armadillidae
and the Armadillidiidae have been considered to be
much related each other. But the two families have
quite different number of pseudotracheae, unlike
structures for conglobation, dissimilar shape of
cephalon and telson, and their disparate geo-
graphical distribution. The Armadillidiidae seems
to be rather related to the Porcellionidae, and
the Armadillidae to the Trachelipidae. It may be
reasonable that the two families should be placed
to rather far positions.

TS9

"TO-RAN-SEN MOCHIWATARI CHOJU NO ZU":
NEWLY-FOUND MATERIALS FOR THE STUDIES
ON IMPORTED ANIMALS IN JAPAN.
N.Isono! and Y.Uchida*. ‘Dept. of Biol.,
Keio Univ., Yokohama and ?Dept. of Gen.
Education, Surugadai Univ., Hanno.

"To-Ran-Sen Mochiwatari Choju no
Zu" (Mita Libr., Keio Univ.) is a collec-
tion of pictures of foreign animals
brought to Nagasaki in the Edo Period.
In those days, when curious animals
arrived from abroad, the Takagi family
serving as trade officials had the duty
to inform the Shogunate their arrivals
with accurrate pictures. At the same
time, the family always made their copies
with detailed notes, which were finally
compiled to the above collection.

hig COnsAsts.of. 225. pictures in 5
vols.; "Birds" 84 in 2 vols., "Beasts"
Sopini la vole. sHOESeSE 25) ane de vol.
"Dogs" 81 in 1 vol.. The present study
has shown that it involves 20 spp. of
mammals, 57 spp. of birds, 2 spp. of
reptiles, and 2 spp. of fishes. Some
of them have not so far been reported
to be brought in Japan in those days;
e.g., Nycticebus coucang, Presbytis
cristata, Hylobates moloch, Tragulus
javanicus, Muntiacus reevesi, Tragopan
temminckii, Numida meleagris, Electus
roratus, Amazona barbadensis, Anthra-
coceros coronatus, Pitta moluccensis.
These new findings will contribute great-
ly to the studies on the history of
imported animals in Japan.

TS 10

HETEROTARDIGRADES IN JAPAN.
K.Utsugi. Dept. of Biol.,Tokyo Women's Med.
Coll., Tokyo.

Japanese heterotardigrades were identi-
fied and classified from the specimens
which were obtained from mosses and lichens
collected from about 1400 stations of main
cities and several mountainous areas in
Japan during 1983 and 1989.

25 species of heterotardigrades were found
in all samples including 20 species of
genus Echiniscus(E.), 4 species of genus
Pseudechiniscus(P.) and one species of
genus Cornechiniscus(Co.).

Among these tardigrades, 6 species; E.

japonicus, E. baius, E. crassispinosus, E.
fischeri, E. viridissimus and P. facettalis
were found to be widely distributed in many
urban and mountainous areas.

On the other hand, 15 species; E. canad-
ensis, E. elegans, E. kerguelensis, E. lat-
erospinosus, E. limai, E. merokensis f.sue-
cica, E. reticularis, E. rugospinosus, E.
spiniger, E. tesselatus, E. testudo quadri-
Psy. Vairginecus, sis Splish. barekes
and Co. lobatus could only be found at one
or two stations.

The remaining 4 species; E. dreyfusi, E.
lapponicus, P. ramazzotti and P. suillus
could be found at 4 to 8 stations of
several cities and mountainous areas.

It is expected that more kinds of tardi-
grades will be found in Japan by continuing
to examine broader areas.

TS i
BRYOZOANS FROM VON STEBOLD COLLECTION.
9: nS Meiwetearcd = and N. Kanekose

Zool. inst. hac. Sci, Hokkaido Unuatv.
and “Geol. Dept., Fac. Sci., Tohoku Univ.

The huge collection of Japanese
organisms collected by Dr. von Siebold
about 170 years ago is now deposited in
Rijksmuseum van Natuurlijke Historie,
Leiden, the Netherland. One of the authors
(S.F.M.) found 13 bryozoan specimens from
the collection during his short stay in
the museum. Among them, we identified six
species as follows: Ellisina levata,
Hippopetraliella magna, Adeonella
Platalea, Reteporella denticulata,
Todictyum sanguineum, and Iodictyum sp.
nov. Of these specimens, JIodictyum sp.
nov. and two other unidentified specimens
bore labels with species names and the
author’s name (de Haan) and each
designated as ’Holotype’. Since these
holotypes have never been published by de
Haan but cited three times in other
literature without description, definition
or indication, their names are judged to
be unavailable under the International
Code of Zoological Nomenclature (Art.12).
For lIodictyum sp. nov., therefore, a new
name should be given when published.

1184 Taxonomy and Systematics

TS 12

A GREGARINE FROM THE LARVA OF PROTOHERMES
GRANDIS THUNBERG

K. Hoshide and N. Tokiwa Biol. Lab. Fac.
of Educ. Yamaguchi Univ. ,Yamaguchi

Only three species of gregarines have been
reported from Neuroptera insects until
now. But no gregarines reported from
Japanese Neuroptera. During the study on
the infection of gregarines to the aquatic
insects, we found that Protohermes grandis
larva is infected with gregarine. The
gamont and gametocysts were found in the
mid intestine of the host. The gamont is
solitaly and two indibiduals come together
just before the gametocyst formation.
They attach to gather head-to-head. The
Shape of the gamont is cylindrical, the
widest at just behind the septum and grad-
ually tapered to the posterior end. The
nucleus is spherical and 21um in diameter.
It has three nucleoli and is at the mid-
dle of the deutomerite. The gametocyst is
spherical and 150 um in diameter. The
oocyst is barrel-shaped and discharged by
simple rupture. The average size and the
ratio of gamont showed the floowing TL=
162.2, LP=26.5, LD=135.7, WP=46.1, WD=46.7.
LP:TL=1:6.1, WP:WD= 1:1. The distribution
of the body size of gamonts are rich in
variety in all seasons. The lifecycle of
this gregarine is comparatively short and
repeat itself in a year. The fine struc-
ture of the gamont is studyed by TEM. A
strange reticulum-like structure is found
at near the septum but the function of
it is unknown.

TS 13

A NEW TYPE OF THE HOST-PARASITE RELATION-
SHIP IN THE ORTHONECTID MESOZOANS.

Kea whajaikae Dept. (Of mbtol. a Nahon Univ.
Sch. of Med., Itabashi, Tokyo.

Morphological features of the second
species of the genus Rhopalura from the
Pacific are discussed. The host of the
species is an unknown turbellarian of the
genus Nematoplana (Proseriata, Nematoplani-
dae) which was collected in September 1988
from Jackson Beach of San Juan Island,
Washington, USA. The female is somewhat
cylindrical, tapering anteriorly and poste-
riorly, and 72 to 92 pm long by 23 pm wide.
The somatoderm consists of about 30 annuli;
a cilia-free smaller annlus alternates with
a ciliated larger one which includes dark
and fine granules. Within the body about
20 oocytes are arranged in a column. The
male, uncertain in shape, is found to be
embedded in the parenchyma of the host,
measuring 8 pm in diameter. The general
features of the species remind of those in
Rhopalura leptoplanae and Rh. linei. The
species in question, however, is distin-
guishable from any other Rhopalura-species
not only in the arrangement of the
granulous annuli of the somatoderm. It is
also characterized as the smallest species
in the family Rhopaluridae. Comparison
with the other host species of the ortho-
nectids reveals that the Turbellarians have
a broader tolerance for the orthonectids
than the hosts of the other phyla.

TS 14

REVISION OF ALPHEUS BREVIROSTRIS (OLIVIER,
1811)

Y. Miya. Fac. of Liberal Arts, Nagasaki
Univ., Nagasaki.

A. brevirostris has been known only by
the holotype (MNHN) from Australia. In
the present study the 7 following speci-
mens referable to this species are found:
Male (55mm long), WM242-65, northwestern
Australia. Male (56.5mm), RMNH, Irian
Jaya. Holotype (fragments), ANSP CA246,
Manila and 2 others (43, 43.4mm) from
Philippines, USNM 240215, 240216 of A.
dispar Randall, 1840. Types, male (50.3
mm) and female (63.2mm), ZMUA, of Alpheus
brevirostris angustodigitus De Man, ror
East coast of Kalimantan. Although DM &
AH Banner (1982) suspected Alpheus brevi-
rostris as an aberrant form oF Al

heus
distinguendus De Man, 1909=A. digitalis
De Haan, 1844, this study and my long
examination of A. digitalis from Japan
make me doubt the reality of their ee
cion. A. brevirostris seems to be a good
species. It can be distinguished by hav-
ing a transverse groove on the palm of
large chela; the spoon-shaped small chela
of the male with the medial face of palm
convex and sloping to the dactylar artic-
ulation; the sharp rostral carina reach-
ing at most the anterior third of cara-
pace; and the scaphocerite generally very
slender and curved laterally.

SHS

VALIDITY OF SAKAINA YOKOYAI (GLASSELL) (CRUSTA-

CEA: PINNOTHERIDAE) FROM NORTHERN JAPAN.
K. Konishil, S. Maruyama? and M. Takeda’.

1Natn. Res. Inst. Aquacult., Mie, 2Hokkaido
Wakkanai Fish. Exp. Stn., Hokkaido and 3Dept.
Zool., Natn. Sci. Mus., Tokyo.

The genus Sakaina Seréne, 1964 is represented
by the following four Japanese and one Korean
species: S. yokoyai (Glassell, 1933) from off
Shukunobe, Mutsu Bay, S. asiatica (Sakai, 1933)
from Izu Shimoda and Kii Nagashima, S. japonica
Seréne, 1964, the type species, ranging from the
Japan Sea coast of Hokkaido to the west coast of
Kyushu along the Pacific side of Honshu, S.
incisa Sakai, 1969 from Manazuru and Kii Minabe,
and S. koreensis Kim et Sakai, 1972 from Impo
Dolsan-do Island, Korea. Among them, S. yokoyai
has been known only by a female specimen which
was described and figured by Yokoya (1928) under
the name of Parapinnixa affinis Holmes and later
renamed by Glassell (1933). During the recent
field survey along the Okhotsk coast of Hokkaido
we collected many specimens of both sexes
referable to this species. The general shape of
the carapace, chelipeds and ambulatory legs is
very close to that of S. japonica, but the
carapace is proportionally narrower in both
sexes (CW/CL=1.45 in male, 1.60 in female), and
the terminal segment of male abdomen is strongly
widened distally, with weakly concave distal
margin, differing from that of S. japonica which
is oblong, with truncated distal margin.

Taxonomy and Systematics 1185

TS 16

ON THE SACOGLOSSAN OPISTHOBRANCH, PLAKOBRANCHUS OCEL-
LATUS COLLECTED FROM OKINAWAJIMA ISLAND.
T. Yamasu!,A. Adachi*,.'Dept. Biol., Div.Gen.Educ.,
Univ. Ryukyus, *Dept. Biol.,Fac.Sci., Univ. Ryukyus.
Specimens of Plakobranchus ocellatus (Opisthobran-
chus, Sacoglossa), have been collected from several!
localities in the Okinawajima Island, from 1988 to
1990. They were examined for ecological and morpho-
logical features, spawning habits and embryonic deve-
lopment in field and laboratory. Some of specimens,
collected from Komesu, a southern part of Okinawajima,
differed not only in color but also in spawning habits
from those of the others, major in number. The former
(Type 2),living on rocky area, showed a planktotro-
phic nature in development while the latter (Type 1),
living on sandy area, showed a lecithotrophic one.
Spawning season differed also between two types, namely
April-late August in Type 1, and September-January in
Type 2 respectively. Type 1 lays an egg mass consisting
of larger egg capsules with an larger egg while Type 2
produces an egg mass containing smaller egg capsules
and eggs. The third and forth types of specimens(Types
3 and 4) were found, furthermore, from Bise, Motobu Pe-
ninsula, and from Komesu. They differ in color to each
other and also from the first and second types. The
result of this study indicates re-examination should be
necessary on four synonyms for Plakobranchus ocel latus
Hasselt 1824 since specimens collected from the Okinawa-
Jima, at least, are considered to be two different spe-
cies. More precise study for other two types remained
to be investigated.

TS 17

RECORD OF A PTEROBRANCHIAN HEMICHORDATE
CEPHALODISCUS (CEPHALODISCUS ) S126 FOUND
MEDIOLITTORALLY IN OKINAWA, SOUTHERN JAPAN.
H. Uchida’! and T. Nishikawa2 1Sabiura Mar.
Park Res. St., Wakayama and 2Biol. Lab.,
Coll. of Gen. Edu., Nagoya Univ., Nagoya.

The material, found in crevice of dead
coral boulder from mediolittoral reef flat
on Kuroshima Is. by H. U. (Feb., 1989).
Later search failed to find any there (T.
N. July, 1990). It consists of 3 fragments
of coenecium inhabited by 5 clusters of
zooids. Living zooids colored orange ex-
cept the trunk, where brown viscera seen
through transparent body-wall sprinkled
with white dots in the largest zooids; a
white opaque spot near tip of each arm.
The coenecium, probably part of a network
of creeping tubes, has a continuous’ lumen
up to 1mm in diameter, is devoid of iso-
lated spines, has erect branches 1-2 mm
long, terminating in ostia with 3-4 spines
up to 1 mm long. The largest zooid in each
eluster has a trunk 500-650 pm long, 5
pairs of tentaculated arms, up to 4 young
zooids, and one undifferentiated bud. All
zooids were immature. This material is un-
like the so far known two Japanese species
of pterobranchs, C.(Idiothecia) levinseni
Harmer and Atubaria heterolopha Sato. Al-
though similar to C. (C.) gracilis Harmer
recorded from Borneo and Bermuda, the lack
of isolated spines may be of taxonomic im-
portance. T. N. acknowledges grant from
Kato Ryutaro Foundation.

TS 18

FREE-LIVING NEMATODES COLLECTED FROM AN ANAEROBIC
HABITAT AT A SEA MOUNT IN THE EASTERN PACIFIC
Y. Shirayama, Ocean Res. Inst.. Univ. of Tokyo
Tokyo.

Results of the taxonomic study are reported
for the free-living deep-sea nematodes collected
from the “Volcano 7” sea mount in the eastern
Pacific using a deep-sea submersible “Alvin’. The
environment at the foot area of this sea mount was
completely anaerobic. Only xenophyophores could
be found as macrobenthos there. Nematodes were
exclusively dominant in meiofauna, but due to the
extreme condition, their diversity was low, namely
only four orders, seven families, ten genera, and
ten species could be classified. The Order Chroma-
dorida was the most abandant, and accounted for as
high as 75% of the total nematodes. Within it, a
species belonging to the family Selachinematidae
accounted for 40% of the total. The unique mor-
phology of its mandible, i.e. a simple shape with-
out any branchings but with fine’ longitudinal
striations, requires to establish a new genus to
address this new species. Other new species of the
genera Bolbolaimus and Spiliphera were secondary
abundant. The genera Syringolaimus, Thalassoa-
laimus and Southerniella were included.

TSS

SUMMER ROTIFERA FROM SOUTHWEST ISLANDS
OF JAPAN*

M.Sudzuki,Biol.Lab.,Nihon Daigaku Univ.
Omiya,Saitama.

From such islands of Okinawa-Ken as
Hateruma,Kohama,Ishigaki,Kume besides
Mainland of the Ryukyus ca.90 taxa have
been detected.Of these,(1) one each or
2 (ois 7 genera(Brachionus» , Anuraeopsis®),
Lecane®, Monostyla®, Lepadella/)), Testudin-—
ella®), Cephalodella®) new to science, (2)
PBrachionus b.jirovci,Monostyla bifurca,
Colurella dicentra, C.oblonga,Monommata
dentata, Taphrocampha serenula, Dicranop-—

horus grandis,Testudinellap sculpturata
Floscularia peduculata new record for

the oriental region,(3) Dipleuchlanis
macrodactyla,Cephalodella misgurnus,Br.
a.orientalis,Trichocerca gracilis, Syn-
chaeta stylata,Testudinella p.dendrade-

na, na,Ptygura melicerca, Dicranophorus eral —
pleuchlanoides, Anchitestudinella mekog-—
ensis second record in the oriental re-
gion,(4) Keratella v.procurva,Epiphanes

macrourus, ,Monostyla furcata,Lepadella
rhomboides, Euchlanis meneta,E.oropha,

Mytilina Mytilina bisulcata,Diurella dixon-—nutt-
alli,D.ruttneri +(1)+(2)+(3) -Synchaeta
SHEN Ea decmmnrecerd lain Japan.1) Species
like pterodinoides,2) species between
fissa & navicula,3) species like nodosa
4) 2 spp.like bulla,one with triangular
lorica,the other with special toe base,
5S)en Spe likes Gr ipiteray, 6) spiwi th) narrow
neck,7) species like mucronata.* Nihon
Daigaku Research Grand for '90(D90-003)

1186

ANNOUNCEMENTS

THE 62ND ANNUAL MEETING OF THE ZOOLOGICAL
SOCIETY OF JAPAN

The 62nd Annual Meeting of the Zoological Society of Japan will be held at Okayama
University from October 13 to 15, 1991. Further information and application forms will be sent
to the domestic members in the April issue of ‘Biological Science News’ (No. 2). The deadline
for application is July 10, 1991.

For application from abroad, please contact:

Professor T. YAMAGUCHI

Organizing Committee of the 62nd Annual Meeting
of the Zoological Society of Japan

Zoological Institute, Faculty of Science,

Okayama University,

Okayama 700, Japan.

Phone: 0862(52)1111.

ZOOLOGICAL SCIENCE AWARD

Annual awards for the best original papers have been established through the donation of
Narishige Scientific Instrument Laboratory, Tokyo. The sum of about 500,000 yen will be
awarded annually at the Annual Meeting of the Zoological Society of Japan to a few papers
published in ZOOLOGICAL SCIENCE during the preceding calendar year. Every original
papers published in this journal will automatically be candidates for the award. The aim of the
award is to encourage contributions to this journal. Selection Committee for the award will be
organized every year.

ZOOLOGICAL SCIENCE AWARD 1990 was given to the following four papers.

Matsuoka, N. and H. Suzuki: Electophoretic study on the phylogenetic relationship among
six species of sea-urchins of the family Echinometridae found in the Japanese waters.
Vol. 6, No. 3: 589-598 (1989)

Kimura, K., T. Tanimura and T. Shimozawa: Mosaic fate mapping of the behavioral and the
muscular defects induced by a Drosophila mutation, abnormal probosis extension reflex C
(aperC). Vol. 6, No. 4: 659-666 (1989)

Okajima, A., K. Kumagai and M. Watanabe: The involvement of interoceptive che-
mosensory activity in the nervous regulation of the prothoracic gland in a moth, Mamestra
brassicae. Vol. 6, No. 5: 859-866 (1989)

Takada, K., M. Itoh, H. Nishio and S. Ishii: Purification of toad (Bufo japonicus)
gonadotropins and development of their homologous radioimmunoassays. Vol. 6, No. 5:
963-974 (1989)

ACKNOWLEDGMENTS

The Zoological Society of Japan and the Editorial Committee of Zoological Science are
grateful to Narishige Scientific Instrument Laboratory Co., Ltd. (President Mr. Eiichi Narishige)
for the finantial aids for publication of Zoological Science, Volume 7 and constant support for
Zoological Science Awards for 1985-1990.

1187

AUTHOR INDEX

A
PIE. JBITROTS US eee ee ne JUS SIt7/
PING ORPANKATO® chess be basen deat is She ok 141
PNOKIMICCIII Renae st shies eT Bae), AB 973
PMC OSHI22 2 Oy st /s5 sh 5002 bes se AE, Ol 973
ATES 5 16 UCTS) C015 0) ena aA 691
sada: sNobuhiko. «2.25: :2ccce cries c lee. 79
PMUOMIONOASUTO: 255655808500. 747, 779, 913
B
Bemdoward Av .sctisiisscse sec 3 TA. 731
Branconee Deneck. ..63 ccc ded scs cay PPI olf 815
C
Gamnevali, Oliana: «2: .:...2...442. BAER OS, 265
Gite we SMiteleeg a4 2.665 dss ceca PRIA. at 187
(Chitty JES NES) ae ae ee ec 419
Chibaspvoshihiko 4 ..2200 0.42.3. RL OS® 895
@GoopermeBdwin EB. .:.6c..csei ee Ze,
Gronshaw-> James... 524 2.20.6. S22 AE 723
Gunning Obert: ...4.52...00 2052. VES. .BROe 265
D
demesussbvelymoG. oc ucts. cence e cece tees 93
I DSTI) Sie WANG LITE HN ee 821
Dukelow, W. Richard ....... 3804.32 85, 871
E
Be emyrg Maly waite veer eho 2a AIEEE APE: 731
EndowKatsulikos : s46<.24585%%.. A A! 697
Pverand Beverley Ani. 2.245% 60 cuts led oe 815
EzashieshoshihikoO: 4... .556 040646 ITAL 38 879
F
emmMan dO Av. VAs. 4 <6 bets odin Sd 419
We UCHAN Gg. rte aios tarsi an wan IRE NY ct ELIS 195
OOF Lal aay ae eee ce eee tee ee ees 47
Bn TieINO MIKO os at oe hides co coetiwin eg Obert 763

BujmmotosSunaOgea 3. en. 5.2 = -eeeescraee 681, 691
Hujimoto: -YasulhitOs yan 2 soe. «ides 697
Eujishimay Masahiro 75... 5..20..- speck 849
BukurgareViasusinth pes ses cas so she 6 eas Mio ale ws He 303
UU OS INTO Beecgs voy Os eee oc recites os alae 691
Bukumitsur akashi 20... tans cot sies hee we raed 79
Bukumoto;.Makoto’ . 35.05. .c6s cs ek he oe oe 989
Burukawa-Y aSuO. soso s ccs. ss eeieer eee. asyse BU,
G
GobbertisAnna:...2.. 2 8... gaeedsod. Sree 265
Gotoh: MoOshiOn 26h a eo oo oe. Re 1
Grunz, Jost. 3... oso s cs eee ean 657
Giycrem Marka ieee nodose piet at 485
H
Harada. Kaori oc 00 oo ee ae ee OE ae 763
Harada, Masashis (0 3). 6. os SA oe 469
Iaayashi;, SUSUMU. 021.55... ss ee « ere 889
Pecins JRObent Pc. ccc cc obo ees TIE eRe 939
J DULIRN RCO) bya) cel 8 ee I Tn An Sore Aare ae 999
inary Hidematsu.. 6 o.55. 66. eed eee: 767
Pdirat hoOshiakw io. kis ones ASI es 879
PliranOr WakeShi : 4 so. oo AOA BRE 681
Bliran@: Retsuyia soul on rec RS ae 93, 967
Peat a ITO oe cos ores dha os io sous oe ARORA OM RO »)
Piirayama Akita soc. e a... oN: oe 955
FLO MGOS MCI ooo olde eacenieis esds Ea a ee 229
J (OVA ERR, Pe er nse eae fs Looe so Regt DOM,
lolmess William Nic oe ees 723
OK. Katsuhiko. jioci0s acess SA 377
PLOTSt— WO Korte oii ladle on cyeaetee ARBOR «AR 783
Hosokawa. Keicht), \...:5.,.4..0 ..2enld se: cel. 195
I
Woae Me tSUTO a eke i ckotnce cate eat cot line opm 401
euc ines RatSe Musee et ctad stance Catal es owe sk 541
INP ere NODE I Bey aoe ne agai cane oe eke os Spy
PANU E YASUO) § Facies eee es SRE ee 93
MORNE USIC ons ety ier ciracicecies ARR S: 29, 187
Irisawas SatOkKO) 32.0 ken be tol... anes 541

1188

Ishihara, Katsutoshi ................... Tica ee 831
Ishihara, ‘Vakaaka ah osc.) Gteretery vist. 763
Tshint SUSUMU ic icae oe ee aos 97, 249, 705
Ishikawa, Tomoichita te eto ee ee 153
Ishizaki; Hironori-.232 ee ee Oe eon 47
Ito; Tatsunori’. 62228 sh ee en 485
Ttow, Tomio’ s:$. 2s ee eee eae 287
J
Jarosz, Geis See ete op OE EOS 85
Joshima, -Hisamasa ©. ..<.2...4...0.0%.4..5884 763
K
Kamishima, Yoshihisa ............. 63, 371, 649
Kam, Syuhein® ..csc) ine os lk OO 821
Kanayamas Mumetoshit *<..-5--2.-2 sete ee oe 649
Wann: Mal. se eke cessed ara h oven AO Rm 537
Nato; Takako e.cei santos! POY ee ita AT 879
IAtOs YUKA Oct rete etna co mye sn Setar es eae 879
Kawashima, Selichiro ....................... 821
Kenned ye Jin e. cp hs ton akeck out teen. ats 871
NikwehiesKOkiGhit sce teiotee : Geena. oR 861
Kokuyama, Sakae. aici rnctt nn «ieee wee 967
IKGIMUCAMIS JIGO: ie aren sce oe ME: SOE 401
Kishida. -Yoshikazu' .5..c0 26.0 cin. 2b. cane 841
Kobayashiz Fideshite-2. eae ere ESI)
Kobayashi, Makito))22:em--en ena a TS
Kobayashi, Makoto ................... 377, 801
Kobayashi, Yasuo iii.8) ocr eee ae 281
Koike,-Mitsuhiro &.... 2...) Ee 613
Kojima, Yoshivukii <uetcsss. Ste ie 849
Nondo, -Akihiro t4.0:.ivn din ee A ee 861
Kondo, Hiroshi: %. Ac. ee Pee 39, 47
Koolman, Jansases.iscenkei Lk Se See een 563
Noshimizu, Ichitow.......e5.002 a eee 771
KumiagalsNainjit... 6:5 cteejonk aie MSR ee 97
Kurabuchigshingos xercgae one oe eee 841
Kurosumi, Kazumasa .....................-- 427
Kusakabe, Tatsumitin. 252500026. s tee 201
Kuwdsawas Kiyoakio ec esto unesee Se 999
L
Bane sdeWWVe David time Rictprhlges id hoe ee Me 29
aubier yen Clenyeal. Aye. a ate RG 319

Lawrence Jonni eas... eae Be 17

Branes Xing Geis 2.05 hoc cone ee 133, 459
Lim, Rita W.. Lo 0s0 hiss dic sn eee 187
bin) Fine oo ce hs nck we ee 593
Low, Wai Poo...) 0)... Se eee 29
M
Ma, Y. Keon 0 eee eee eee 147
MeLachlan, John:A..... 52.2.2...) eee IBY
Maeno, Naomi .:...........55225- 6 401
Marion, J. O. -60..5....5 eee 537),
Martin, Len: ............25 so eee 871
Masaki; Takakatsu '.: = .: 2325-5. eee 697
Masuda, Tatsuya .:....5.9:25- eee 287
Meyer, Wilfred ....:2. 7.) 22 eee we
Michibata, Hitoshi’ .:....%. 3.25.83 55
Mills, Karen T. ...).2..4. 223 eee IBN
Mitani, Hiroshi °.2.:.7.5....225.. eee 395
Miura) Tomoyukil) ..... 7.202) sig)
Miwa, Satoshi) «.......s:.2-3ep eee 93
Mizoguchi, Akira. ¢o0 22 es eee eee 47
Mizuno, Takeo .... 2.5.0.0 2 ase eee 159
Mizutani, Akiko... :).. 0 52.5...5) eee 331
Mizutam, Makoto . °°). =. ae 593
Mugiya, Yasulo’...........52 5... eee ZS
Mukai, Hideo... 52>. ee eee 947
Munakata, Meiyo ..2.. 2.5... eee Pere 985
Muneoka,,. Yojiro ®\...3 :: 4.2.00 asaeoeee neers 801
Murakami, Toshiki) .<<..4:.5. 430-eeeeeeeee 73, 105
Muraoka, Shiro: ....5: i ake eee eee 763
N
Nagahara, Kazumi --..). 2.) d:eeeoee eee 849
Nagasawa, Hiromichi +: -2) jse- see eee 47
Nakai, Yasumitsw:. «.)..3445:3- ese eee SI)
Nakamura, Masahisa «. >... 3-3 eee 973
Nakashima, Hiroshi... 2. 3.1.4.5. 2epeeee ce Sy.
Namiki, Hideo) os :acecc02502). See 973
Nassel, Dick-R:. |. o...:..... 5 ee See ee 223
Nuida,. Akiyoshi ............. 2. sae oe Bye)
Niwa, Michiko- .........4:2 eee 209
Noji, Shinobu. ..........4 ee eee Ooh
Nomaguhi, Takashi A. 7.45.5) seen ore 39
Novara, Caludio. ... 2.00305. sce eee 265
Nunomura, Wataru .....-3.. $:0.54-. beeen 767

O
Weasawatra, Tsuyoshi, ... |. 2:/jferes 3s lee ee 967
emi MIKIO: 3.5. occ ces. a EGTA ae ae 783
Were Chan: «occ. occ eR E ee 623, 967
WitkcuwonNorhide: jci66 5. oe cs woe ORE 979
HON OM AKASH «60:5 5 oo sn oss 0 HR hs Se Spa
SEMOMIOWNOFO), .o5ce.c.o. oss ss oc ae ae 861
POAT SHIKAPOELISASI 53. wus ois oc ee oes 2 739
ROA Ae ATAU 2 oon oie fons ov TASK Oe osc 281
Mea aS oo once ee oe se I tes Sit
SMHOCO- KIYOSHI J. .- = ac «> Saeed wee ee 605
Okara SHOICHT 2 oo. soos nies bones Sb oaks 973
COREG. ANTOVCVICE. aa a 783
CORSE, YC Ge 783
Wonka WAKOUO: - 62. goles cede scenes cede 409
GoraeeyvOsnmhikoO: 5h nls. coco otibered. Spils 7/7/ik
P
PAMUW PE SACOND fo. u cc net hiecs ve beg ss teens 547
EaneyesubhashiG. 2.2.2... 6. ccc dee e esse ees 547
eM PON ONG ee Se cece es oe Bee sek eens Doi,
iP rern@ Sea BS ae ee ee 419
Blisersayatsbrika M2... sec cee cae enet's 335
Polzonetti-Magni, Alberta M. ............... 265
POY, (CS. TS ee a 871
R
[RSIS ISIC 6 goa 5 cea nara 147
ESBS EG) LIC | ea ne 723
[POSE Ay E07 1a Ras ee a en oer ee nae ar ern 671
RMI AKINMIASASIN 6. 5 ais eine ds oe enc eet oe 409
S
SEC. ZANE) 6 [Eee eae ee ZAP. 683i
Sakapanis SHON Fs 25s. 5 ks oie ss we keene on 985
Sasayamias NUICH: ccs. 2 5 oa hae eee Poe new 967
SEQUIN CN ZO lomo ccSiiaeee see es ties Gee wares 195
St LOMe VI AKIN foc Pe AN coe oe Ce Soe 395
SALOMINOMYUKME ee. Kies. saaelag se ededsi esses 861
SENECA U1 2 ee ee 469
Seki Such NOCH isso. feel Seine s Shawe S 287
SIPEIOVIE SGI) |G oe Sees ea gee oe tae 127
Shikatani} INOmkaZu).. 26 os ssbb is ccc vsle ses « 477

Shai ay AMNITO ie 2 ces heeds ae ee Shee cece 395

Shumazur Takeshi 2: o .ctoc5 8: SpA Sees ok: 933
Shimazas (@niekOe <0) 50nd eA Secs: 895
SHTOGAMSE ltr a a ee ost: oe RE Zi
Shiokawa=KOIchtOrs cs: os iw Gee ee 195
ShinaishteSatoshitascta nek ccd seat. es 141
Shirayamarzowmosnihisass: 4: <2: ies cae ec ws 939
Sivasubramanian, Pakkirisamy .............. 223
Stace ysuNOmMaANME SN cr sons oes ees a! DS
SUV EMMES IMU AKASH, 3.502: dette jase 831
SucmmurasMakoto: 2.5.22 55< 5 weer 747, 779
SUPATI CSP A\L C1110) 0 (eee Rs Pe ca 47
SUZUKaGINOTIO tees ced se ahah eee gece 355)
Suzukaelomonikon:<.c..cdsic. « hheindese eee 1
Suzukie- Yoshitaka). 2... 6... Weddace TAT AS Obs;
T
dRachiee@ iikashiy nes ein os ia ee 235
Rachie SUIMIC AH eye a oe ee 235
Ghacke sIboOth ane re 86 os ao ee 657
Magawa, Masatom 22252 oh. none ese se 93
NagawaePakasuke e286 foie cntes che es een 605
Makcadar Vasu aks si: cick ace Sass oe eee awk 779
MakahashivWazvuhide: -2 3... 2enmst dee nak en 229
MakahashieeMasaki 2682). 8 ee ao os eas 861
Makahashie Nobuaki a0). sea cee ee 861
RakeasugitNODORUM i: ae5 caste aoe eee 541
akayanapieouji 2. oh .cs<e aes eee ones 747
iRakedareMiasatsune We. . ceeiosen oi. eee 477
MN AKC IY OSIMIO MES sf csisie oes ee Rech he ee one 435
MamamakroNobuaki: @ 22.652 s5c6 dae ceed: 385
ATT ATIOMPMES UNO Mees Seis ee eae 763
(Manabe eksutoOmUy 0050s eo ae eee 443
Mamalksave NK ics rc shu cee ee ee 671
Manakas Shigeyasa: je 6 tie cence ee eee 427
Maniguchitevautakiay 2 ics vate faeces ek 427
MAS AKAREVO KO Kes penile es Nai eee 97, 249
Mate dar tlie kilt. Sree) os oo yoke tee Bale: 593
ihiedemann: ElemzZ 4). ---- ee WA
RodasMasanoner sss. do. Gee 297
Roh Yoshino. os 5624.0 k ee 3315593, 961
NomiOkaweWenplepeeee a. Sone owe SO 895
TomookaMYasuhito 242450225... oe. obs Wh
Mone RaAwass VASULO) 2 24. Vea. sa os ce 229, 831
TSUKISCRW AUN aie eee so cee oe oe et eek 923
U

1190

Uchida, Teru Aki or.si25.4500. eee Se ae 141
Uchiyama, Miroru ............. 73, 105, 623, 967
Wedas Hiroshi) 29.45 <4 80, er eke 681, 691
Wk Motomichi a eee, eee 895
Wyamay Maro eee cet Sooctes ees See ee es Bere 5)
WwW
Wakasugin-Chikasiieis pis c Ae. arenes 623
WakasuciNoborlenn taaniactee sent ae eee 209
Waku: Yoshio. e . Be 613
Wassersugs Richard isto eae 57
Watabes Hide=akite io. (eee 133, 459
Watanabe, Hiroshi Kis... 00. bo Ie oe 767
WatanabexTakushiox.: sic 2.50 oo cueteor en aed 435
Wie SOC ert cunt ads eee eee 297)
Y

Yamada, Chifumi

Yamamoto, Yutaka: . 6... i. ...neeeee 895
Yamianoucht, Hiroshi *....:.... ..:ceeeeeeeeeee 705
Yamanouchi, Korehito ...../ieses eee 907
Yanagimachi, Ryuzo: ..:.....-+...-e. see 639
Yang, Cheng-Hsiung. ..........2. - ee eee 639
Yasugl Sadao... ....40%5 000 ¢ =. eee 159
Yonezawa, Yumiko ....34......<. see 39
Yoshida,. Naoyuki -............:.:0 eee eee 613
Yoshi; Makoto: .......2...5.% 255 sae 895
Yoshimura, Katsuki ........... . see eee fix)
Yoshizaki, Norio ...4:......2s5285 eee 581
Yoshizawa, Hideki. .«...... 2... 2a IOLG23
Yunxia, Ta |... ica nnn snc iein See eee 127
Z
Zaitsu, Makoto. ..........060cnn « See 895
Zerani, Massimo: _:...5: 05 222 eee eee 265
Zhang, Wen Xs 2. ..s 4.5. ee ee eee 133, 459

1191

AUTHOR INDEX

(Abstracts)
ASTON CZROMME 2G eth ooo ea cadens os we 1107
A INS AO Ka MMII ORS fo Sica si oveicce cc SALE A cee 1114
PMta PMLA OS IT sfoyt fears tee roy Genre cts 1068, 1069 Asashima, Makoto ................... 1117, 1164
PM pare OV UKG ys hys yep). coos Ape taie nit MOI a mipye WNsinda, Masaaki. jo. cercest.tiod oaisoed eens 1059
POS AOS ANIME FIP ore rs sn aries rnialls Clase MMS Oh 1039 AshleyA@hristopher C..2......«.......eeeeie 1039
AOE) SUING TS 01 een sin eer LOSGHAOST . AtomimunkOnes obese HERERO 1077
PRWEB AKGCVANKG, «05... cirri enic ee il ee eds HOSOHMOSHY. ~ “AGOMIY OMIKO RG o.oo sco asorene. PES ok 1065
PAC NANA Fale re ere tein Sd areas _ Bee AMISS) PAUSUCHIE YASUO) oii ceacesjsioi sins ACR 1118
PEMA e WAKEHILOs «0.515060 SH d. ER LOM: Atwanos Mim Omir epi. sie cicene oon EES 1079
PE MIPMMINAKUUD Tine pederes tle a fo cn ores nseaw «EERE DINE NOZS. ANZ IRAtSUGRE footie ceeiroaiengo eee 1031
ANT INFOLIITR 0): ate ae a re ee Ree be 2 1034)". AzumvaliMasadakity 5. oc ccieccccee setae. ee eee 1077
ANAL, SIT ae ee ane en et 1067 - Azuma Masamin ee 5c) eeeec tocar cs ee 1031, 1049
JESU SENET JN CXS (0) eA 1159
EAIVAINWEN, APES 01 a eo od 1182
Puimawas MAK OO: 0.5.06... :cioan oh OS Ee SA GT, B
PieaSAKa IKOJI)e: 2.02.2. ee nha bees IOS.“ sBalbarpyoshichikia) 2. .,.0e/sccwieccce eee 1024
ANRC), | XC) oe es Ae ee er 109210938) — BandargRouichi cise -cou2yccdunt RIA 1028
Akiyama; Takahiro .................. 1084 WHOm- * Benn, Howard An. oo. celeste 1019, 1138
/\IRTINVES100 oe 016 E10) (ge iulysy * Bie yac aul Wee oy eiesdoee se clangeed neces eecetonnt 1059
PM ANIA) NOVOKO 0... 5). oa scaieyeeas oh Pelee ae ile HGS “BuletuPhilppea. is. cnnes co Xoo oc sctversene 1089
/NTIMANUE NS 1 REVAL NCO ne me A 1080
PATO PE UNTO TU. irae ove oie ceva ane, RSENS ae 1086
PMTAUADIMOPPAUSU KAS A swesgalecs cases ve spose ve wove sores « hucleroBle 1170 c
PAIMEMIV A SHOMAN, «5.0.0, 0:6 <6 acs aw caaelel eS Sy | “@haeny Shigenthy Xcess use cack ew See 1040
/ENTCIO), LE [TCO)010) ES Be Ee cae MBSE Giibaw Akita isso 8 cha ssw ee sesocanck 1149, 1161
/ENTIGIOS | SEVIS) O01 <0 en a ec hoger “Chiba, Chikatumi, ... 0.0 obese... ee 1041
NTO S ING REE 6 Ue see 1049 SNAG e = Clb a YO! ccs Sanin ok anineend ge See BAe 1071
PRONZAN AGVAIWAKO! 5. csi be ee eee ca bee baa es LOSAS | “ChibamMotoy ost. sci: ico AEE 1083
PMOKOPBRCC III acer. os oe css sae ak DS LOSS) <ChibarsYoshihiko, <2 55.04. .-2.cesteee: 1G Se6
POKES OSIM ajsieieyes eee ol SEM «Sele ELeE li47ecninn: Chino Haro: .<). o..36.s sao bee 1078, 1079
PMO PEABO TINO YI 242 ie cess ose gay ye asojand eis eavee es CRIA oes MASY | WMINZEI VV ASUO oot ea ea yeas ee RE 1063, 1076
JENIN SSN KOSS CVE ore ean GP LOSA ROIS ‘Cho, Pumiaki eo. 266s wiiece co SURO LEE 1135
NTA AKAO ARI. tocol otocs ee eae S «EUR INOOS “Coxe Os Alsetii sh cee ke ee Bs SOIREE, 1042
PREAMP CES UO 58 oki ac cosine ie aches RE Soe 1179
PNTAIRDY ASUIMNASA: . c.cc ea ca eae ee 1124, 1125, 1146
Acaitervel [2h ee Tema 1136 ,
PER Aclep SAL Ope gs eseacie is ciclo on os kihdeuod AMOR: IHO4)) DankiKatsuima obec nce RRR c 1099
ANTRUSS SQOLSO! a ee es Ren ROPE STAR ERAS Se Me 1083 Dan-Sohkawa, Marina ............... LES ailiG3
ANTAIZUIM PAK ASW ecd wes obese Soe dio) DawmaciMasasht: . oo. cceciclccs LSM oo: 1180
PERTARCA WA ADIN ECTIEATEO soso isis dire soso one oucne.one ooh OR « ODS We suchiblakCO) coi ceie ico uwlew wre aie AME oe IVD
PRTAIMOLORBCAKUY Ai ces oseie ore aioe ee eie PRAISES HONS Sdewesuss Evelyn, Give 3)... 2.) eee REE 1152
Sada Nobuko. occ eicecac snc c ue teed. BE 1060s”, derKort= Constantine A. D. ... e022 2..23 1078
EXSAKAW DP NCO NK Oper cess ocd) SOA BA AR REM HOSSZ, DenoFMakwuyay as oc vec cieco edo OIG. ole BG 1102
NSA OMS AtONU hee eo .ci8 2k cence See ROME SOAR S09 — Diaz Manian. Mec rc. 2... cee Bate 1116

1192

Dohkes Nemypiroi ke iu.csncwsecanioeace 1059, 1060
Duany:Cunmun swe cate ne 1151
E
Ebihara,,Masayukia pence erent ee fee 1026
Edamatsu, Masaki: sv. oinnenc05 Mee Re 1066
Edman. Pak: ..:ccccecnnncuice Jae ae 1039
Beouchivbisuke iyo conse eee a 1028, 1031
Bouchie Masalvatstn ery sarreratccpcrten nce Sa ete 1077
nam, NOmononmeneyincceretes eee ee 1111
Endo ThnOy Os hiverdea acerenrren:.c, OGRE ee 1123
BNGdoOx Katsuhioegck pocvhersictoose sk ee ee 1131
FE MdOsseitae wi weet cre sect cece cis Re ae 1084
EndosSachikomeetctnccncscss 1072, 1074, 1098
ESMGOM BAKES Ib 5, carer topes eedee ssc ecacscns 1070, 1106
Endo. Vasulisa = ee 0 ca LAGI G2 73
Shel Dantes ote eas nee 1044
ESOP AUMIIK Oe ia eonecaer os as ai oe Se. 1086
EZakiiOSamuU yee eee ck eee Ree. 1048
Ezashi, Toshihiko .......... i SRO 1133
FE 7UITEs MOMONODUS sure ir tosses. Se 1132
F
ujredas INOTMKO) rag nae ne sti i 2 os heen 1111
Ui Geinige aacketa ea eee oe ie Cima 1118
PIL KOM ascii can cau an erent whe Saget t 1149
FUT INVOZO 9 eGeonie en an eye nane anit 1019, 1020
Pups Pamotsuc 3 &,..5oe eco el ese os) 1051
PujikuraKeik0! sey .tecten ote eacponee: Seek 1153
Bujikuras Yoshihisa: -.ser-cece sere ee 1051
Fujimakiy Masato. «.....:2.-s220). cee ee 1065
Fujimotos Katsuyuki... io: nee: Se ee ss 1038
Fujimotos Keng oi. den Renita ee 1060
Fujimoto, Masaaki. gcc ccce en cee Ae 1019
Fa fiMOlO 5 SUAMO) cnr iain meee ce een ee 1083
uyisawas Hiromi) 925: 2k)s. (oes Cea 1176
eujisawae m@oshitakialas 2) sone eee 1132
Fujisawa VukO 5 Hes eae anon 1037, 1038
UISE YUCAK Ay copys denen eonind 10 eee 1108
Fujishima, Masahiro ................. 1163, 1168
ujisue, Me guint i inco.icn dane MARA, et 1102
atta UN EO}. 18.2 e-aieatoe ere k oo as Ne ne 1149
Fujiwara, AkiKO: 34 checkin od. 1058, 1109, 1115
Fujiwara, Shigeki, <2. .A4.00. uae 1104, 1113
RojiwaraslanuhikO)..ce.c-- cence oe 1075, 1078

Fujiyama, SHIZU0. creda ccnelarehe ee Spat 1182

Bukada«Sachikos, sx.) 0544202 1089
Rukada Yoshitaka, s'..2......- 1030, 1031, 1032
Fukamachi}. Hiroshi 3.........)..:..292 eee 1166
Fukatsu,; Nobuko: <..........¢.<4: eee 1138
Fukatsu; Vakemd. ..:: ..:...22. eee 1179
Fuke,;,Masako ........\..c.2..52).. eee 1088
Fukuchi,: Mitsuo >.......:....:.2.-5..5 eee 1160
Fukuda, Kimiko. :2:..........:....5. eee 1119
Fukui, Akimasa ........0...0...0.0000 4 eee AAA:
Fukui Yuriko «osc e oes ce cee nce eee 1120
Fukumoto, Tetsuo. ......... 2) eee 1051
Fukushi, Tsukasa... .....¢..2:.2..02000 eee 1022
Funayama, Tomoo .........:.7- sass 1045
Furudate, Hirouki'......... 232.222 eee 1026
Furuta, Emiko | ......c...3ccic-ep eee 1050
Furuta, Rika «......5.cceeneeeeeee eee 1172
Furuya, Hidetaka .........2....- eee 1164
Furuya, Shigehisa ..........5..ca54 eee 1108
Fusetani, Nobuhiro: ..-. 222222) eee 1098
G
Garbers, David L..............2. see 1133
Gibbons, Jan R.............. 5... 423 1044
Gleadall, Tan 3.03...0.05...cc 0. eee 1032
Gomi, Toshiaki ......00:......c..0c0 Jee 1171
Gots, R.C.. ..6 sect oeeeeee eee 1169
Goto, Atsushi :.....5..0..¢:0.ee8 Seo 1 wie 1040
Goto, Susumu. ...0)..c.6-eeeee eee 1070, 1106
Grau, E. Gordon. «......4...400.) eee 1152
H
Haino-Fukushima, Kazu ................... 1062
Hama, Taro .\..c:.50.i.4 coon eee eee 1070
Hamaguchi, Satoshi... 22 sesecee een 1057
Hamaguchi, Yukihisa .......... 1043, 1099, 1100
Hamazaki, Tatuo. ...6:.:00... 0000000 eee 1085
Hanai, Kazumitsu «02... .nc.cce Se 1132
Hanaoka, Youichi. ..«........c..2e25-eeeeeee 1153
Hiara,, Hiroaki . ....0.00.cc...<cceee eee eee WS)
Hara, Katsuhiko) ....).:.....4.cne eee eee 1123
Hara, Masayuki .. ............. oc. 2 See eee 1160
Hara, Reiko: ...:.c..00c0.eeee eee 1031, 1032
Hara, Tomiyuki ..............<ceeeee seer 1031, 1032
Harada, Tetsuo. ..::52..¢046.c0 eae 1178
Hariyama, Takahiko ....... 02.590 1032
Harumi, Bakashit ......2.:s0% 0000.00 4. ee eee 12

PACH MATSUO IGN. .cchesnoiies te me 1093
et ASC op Fel ATO LUIAT eS 1200. .hs:sposaro i AA OE 1079
BAS Said KOU iss (2i.cieic,.jensne SES SINR 1130
IaSeSaAWal SANA mart... :saiseciec eins Sk eas 1136
ASS CEES QUITS 7 een meee teeth eae, 1026
BAAS AWA ANAT KO) ccc) soe ich salves Hoteles See 1164
ELAS TTT, oy) C0) 6) Ta eee 1087
Flashimoto, Naohiro: ....:....00/:.......:.8. 1090
[EI SISCOPETIT IN BS 0 ee ee 1140
aatanakea USUMEO: «...)osc:cich ws cicjone de betas ese 1042
ELE TINS nM IG POY 1 ie rae a eee eas ee 1105
PitsmmeMachiko: ©... 35 ss in Oe 1182
PACMAN ACSTIMIKO » 0 s50006: veces cio viene o's 06 0 SUNS wane 1142
AO GIPMICCISUIKG 8.555 oes ws te dts Oe 1151
iavakawa- OV OIC... ess ks enes sens 1061, 1166
PiavakcawaseVOSMIMOM. -...... 56 -.c000000+ 5 00. 1175
ayasai Hiroaki... 3). 2.2... 6... (142 1143 1155
EASASMI GS MITOSM. «oi. gioco eevee ee od 1020
LEWES TINS I 0) een 1152
ReeeaAS MPL MITES on ove ones vd ee eee Ie y7/
rap SSMI SUSUMU fossiei. oe ene he ee a eee’ 1180
Welet tka WHGIMO Ree... 6.6 8.065 52k os Rilo oes 1041
FARA SOM iliac os 5 A ete ee 1027
eeasinireOKUMIKO 2 5.6. cnn oss SEAR ae 1053
Eieasinoujime, Sugie:.......... 2.1 ideas deh 1067
Eieashinakagawa; Toru ...... 2240.02 1118
Elicasintam, Tadashi... ..:..........2855.44 500005 1181
IBC CCS) 18 Ge 110) re a ee ae 1060
EID VC) ee aR Dl rg 1128
ARGOS AAW ANKAT A 4 5 5: <ic:s20 yornos on LS Ee ed ST 1104
RoW KAN ANGE 5.0.5 os oe cechs cei c woos SMR wo EL 1094
ALT), NEYO) 0 2 2 oe tne be am ae 1041
Mirabayashi, Tamio........... 2: 1091, 1160, 1107
PART APMC ATU PS Bs ae, oc once s TSR ae 1064
MTR CIKO, 09 2 055k 5b ins FREER ES 1165
PIETER OSIIAKITSS | ..0. 5, oc cso s CEE. SER 1089
PeteAKOMABRE III 88 F 2 «.ci.0)s.. CR ee 1118
firamoto, Y UKIO! cs.....2.<...5.. 5 eRe 1090, 1099
FATTO Pa AIK OS I 25 fos, sicko exstacors. odes 4 1048
Pat ANOMINGM AICHE 2. 3.6.5.0). o es oo kOe 1096
Faimano;, Tetsuya). 225... 1134, 1135, 1136, 1141

1iSieed5S2
loka Wlisuy OSs ac /.<tnee 2 eee 2 Se 1079
eae TKO PEE sis oo2 5 keds ER 1047
ReARCAN ATTN OVA Da 66.5, 515. oceias, sin. ass Se 1079
BOTA CAS ZAG F985 5 oes csiseccastlns SRA: SA 1038
PATO DE AMNOMONISA WIE 5 ocacce corel SES 1122
liro amas ONEUos5.i os ..c ss 1148

Rirokawa Nobutaka 3.5243. -Geta Bo Sex
Hirono, Masafumi
Hirono, Yoshiyuki
Hirose, Euichi
Hirose, Hisayoshi
Hirota, Kiyonori

Hirukawa, Yukiko

Hisada, Mitsuhiko

Hisanaga, Shin-ichi
Eniyoshieironobu %:. ....53<<.2%... 2 sae
Hokano, Masami
Hokawa, Naoki
Hlontmartayvoshiharw ....5:...0:5 04.005"
Hoshi, Motonori

Hoshide, Kazumi
Hoshino Masaliani, (..4.....> 0 eee
PIOSOGaSMZUKO ye 5 see e. eee
Hosonos Masamicht . 5.3... 2. ee
Hosoya, Hiroshi

Hosoya, Natsumi
el Obire Manabe ys. esc otc. becca
PLOT IINC Ile tess 9884 od Sa ies 4 oe
Horikami, Hideki
Horiuchi, Shiro
Hotta, Kunimoto
FIOUISIKOUIIC Ee? rh |. oocss Sac Nee LS
Hyodo, Susumu ..........
Hyodo-Taguchi, Yasuo

ee er)
ry
i

ee ]
2
|
eee e eee ee eee eee eee

eee eee eee eee eee ee eee ee we

eeece eee eee eee eee ee eee ee eee
eee eee eee ee eee ee ee ee ee eee eee

eee ee eee eee ee eee eee ee ee eee

eevee ee eee eee ee eee eee

Kelnikawar sloshiow< .a55% cde tio oe eee ee
Ichinose, Masao
Ide, Hiroyuki
Ideyama, Shin
Reahlic (SUN Og isc ro ese aero ie oe INS
KeuchimRaisenerera oc. .s.: sue ae
Ihara Setsumosuke =... c5 0. seus. oe eae
lino, Teruhiko
lizuka, Yasumoto
Ikebe, Yuko
NkedaryMakOtolirn sss A big icacics ook. oe
Ikeda, Mariko
ike are Md ORE 2.22 Ar, od ss oe. on eek
IkedaymMetsuya -2 32. <:
Ikeda, Yugo
Imai, Katsutoshi
Ta INC TKO eGR AR 555. AR. eR EE

eee e eee ee ee ee eee eee re

eecee eee ee ee eee eee oe ee ee ee ee esos

eee e eee eee eee eee ee eee ese ee eee

eceeee eee eee ee ee eee eee ee eee ee

ese es eee eee eee ee ee eee ee we ew ee

1023,
1072, 1089,

1149,
1096,

1061,
1122,

1172

1024
1172

1079
1161
1105
1184

1126

1057

1131
1119
1128
1052
1021

1194

Imamoto, Yasushi
Imanishi? Tount\es..0s6< ee on Mt ADK
Inaba, Kazuo
Inamura: iroko:22 fon eae sn eee ee
Inase, Yoshinori
Inokuchi, Tomofumi
Inoue, Chiemi
NOUS RGN! seer hy ate UE,
Inoue, Rie
Inoue, Sadako
Inoue, Yasuo

Inoue, Yasutaka
Inui, Yasuo
Irie, Toshiaki
Ishidashidekd 2 o.oo weed oF es t. ROOS
Ishida, Makoto
Ishida-Okawara, Akiko
ishthiaras Katsutoshi ...<..... 0-2/0, Se
Ishihara, Hiroko
Ishii, Keiichi
KshnNaokatas sis seer ec dre cso ROOM,
Ishi, Saburo
Ishi, Susumu
Ishi, Teruhisa

Ishijima, Sumio
Ishikawa, Hajime
Ishikawa, Jun
Ishikawa, Ryoki
Ishimoda-Takagi, Tadashi ..................
Ishimori, Yoshio

Ishizaki, Hironori
Ishizaki, Yumi
ishizuka-Oka,*Atsuko: : 24554565 ee eee
Isono, Naohide
MOMWAIKIFA: ORM ties db eae doko d em
MtOMIKOIG. Stes ss ad saga Waele
MCONMINAZUO RIG Nats ate eho SO
Itoh, Keiji
MohtiMasanobDutie, 95 6045 08¢00 85 eee,
Itoh, Masanori
HtohneMiasayard: is53095sGancsesonaean eee
Itoh, Yuriko
Itow, Tomio
Iuchi, Ichiro
Iwahori, Noriaki
Iwakiri, Minoru
Iwama, Akifumi
Iwamatsu, Takashi

eee e eee ee ee we ee eee ee hh

Ce

Cc ed

eee eee ee ee ew ee ee ee ee ee

eee ee ee eee eee ee ee ee ee ee ee ee

cecece eee eee eee ew ee ee eee ee we ee ew eo

eecece eee eee ee ee ee ew eee ee we oe ew ew ow

cece e eee ee ee ee ee ew eo ee owe ew oo ow

CC 2

i i

eeceece ee eee ee ee eo ew ee ee eee ee

eoocee eee ee ee ee ew ww wo

ecceceecec eee eee ee ee we ee ew eo eo ew wo we wo

coe vee eee eee eo

ecece eee ee ee eo ee we eo we ee ee ee ew we

eceee ec ee ose ee we eee ee ee eee ee

ecece ee ee

CC ey

eoceeseece ee see cee ee ee ee ee ew ee

coecece eee ec ee ee ee ee eo ee eee ee

ecee eee ee ee © © eo ew we ew ew ew ee ew ew ee 8 8

ecceece ee eee eee eo ee we ee ee ee eo

eeeee eee ee ec ee eo we eo ee eee ee eo

ecee ee cee ee ee ee ee we ee ee ee wee 8

eoecee eee eee ee ee we oe ee ee ewe ee ee ee

i

eeceeee eee eee eee ee ee

i ?

eeceeeceee eee ec se ee ee ew eo ew ee ee ew

eeceeere eee eee ee eee ee ew ee ew

sec ee eee ee ee ee ee wo 8

1168

1049

1170

1039
1160

1113
1043

1122

1116

1124

1033

Iwamuro; Shawichi.................... See eee 1143
Iwanaga, Sadaaki.....:...2:5.....41 SR 1051
Iwanaga, Toshihiko ...............) Gene 1149
Iwasa, Tatstlo? «........0..)..0..222 800 ee 1030
Iwasaki, Shin-ichi ..............<.cehe epee 1173, 1174
Iwasawa, Atsushi ....../:......... See pee ee 1143
Iwasawa, Hisaaki ........ 1082, 1088, 1140, 1142

1143, 1154
Iwata, Katsuya ...........0.2:%% eRe nearer dA52
Iwata, Yukimari..........:.......433eeRep eee 1047
Izumi, Kazunar®t: ........3.4.4.4552....2 eee 1127
Izumi; Susumu.......-45..2e6es 1062, 1067, 1077
Izumi, Jun . 2 00s scent eee 1054
Izumo; Akihisa ©.........:.:..2.63 Sheen 1048

J
Jaana, Haruo |: 2.0.3 .4.5455 55. See 32
Jip ZarSi oo. Se tee eee eee 1087
K

Kabasawa, Hiroshi. ..........254:42.55 eee BtKl022
Kadota, Tetsuo... 3.0.52 05> 2 es eee ee 1169
Kageura, Hiroshi ...... ..: e265 eee 1033
Kageyama, Takashi: :.:2: 24222 eee 1062
Kageyama, Tetsuo: .;...::.>ssaee eee 1077, 1107
Keay; Kazuzhiko) (255255 -seeeee 1164, 1165, 1167
Kakeda, Keiko 2: {2.22:<52.055325555Reeeeee 1075
Kakeyama; Masaki .....:. 222225: eee 1145
Kakinuma,, Yoshiko........2.55.....58 eee 1110
Kakuo,, Yoshiki. .......<. .4:0qos sae eee 1131
Kamata; Yasuyuki........5....3ene- eee 1108
Kamatani, Yoshimi..-2...-45-255 0-2 1038
Kamiguchi, Yujiro .....225-25ss5ssseePenee 1053
Kamioka, Takeshi... .....:¢:55554+ eee 1105
Kamishima, Yoshihisa ..................... 1020
Kamiya, Kiyonobu..:.-..+..72. Geena 1137
Kamiya, Mie...) 2... 0252093400 0=eeeeeeee 25
Kanai, Shigeaki..... . /:0.:0.355- 200 1095
Kanai} Chieko: 0.2600): 5.0.99. eee 1110
Kanamori; Kan’. . 22.5. 32.333 eee 1047
Kanba, Mari): s..0<s.6s000500500eeeeee 1129, 1130
Kaneko, Miroyuki. 33245.442440452 ee 1115, 1163
Kaneko, Naotomo: ...5.4.s25244. 08 eee 1183
Kaneko, Tomoko « «0.2.06 40.1625 2G ee 1057
Kaneko, Yuko) 42000. osese005- eee 1025, 1026
Kano, Kazutaka ... «dee. 56s. Ee 1064, 1119

Kasahana> Michihiro, 3.0.5.6... <ooeee. . HOA: “KEIO iTOOM ae: 2k 32d ete crobces 1061
Keasabaradcl OSHIKO ss is: .j.)2)0.s:0.ss:c.cs RAE 1OASY - engakuryMinckomiry. 2c. oe cies eee IDS
Kasatani Kayo soc0..scc cls nacician AUR ls. 2 MOGOE pKallhatiaeA ieee See oss. 2s Seopa Wes otaceeig 1046
Kashiwamata,,Shigeo ........ idee es: HOGI 5. {kakkawal Yamato. 255 eee rdaeeilSs Soto 1126
ease, YOSMILOSII . oic.5.0)0:055. 3002 Soe eee 1069: . *KoikuchiPWokichie. < .o. Gh aeese lieder 1166
BRAUN AE OKO o.ccos os sisrsisvovoies SERRE leh A Selo aketkerchibs Miyake Pees ses. 2 5 oeecua P= poids 1025
Rataein Chia 425.5... lc. ED HOSS: LOSTeelOONs —aKakrehig Motoshi, : os. .6 65:0 nc eee heres 1134
React iG Dit) hfe. 5 5.éceys.accec RRR 1O78eN079) . KikuchisShinzicht . 50). 255. oeigasee eee 1159
iM ECTOLGTE G1 <0) tee en ee oe ee 1OSME: Dikuichik: Susumu =... 2k. eee ae 1160
PeeerOkera ICC ml segs 5 cases, cys AOR M09... “iikuchi= Yasuhiro... 2.0 oS) oc cee oe. Dig
Kenayama, Heizaburo . .:...........i1Gcee ie ..0e hOZOY © KakutasMWoshitemtnes .n. o. 5. a. ceeeent oe seen 1161
2S. Chie 1 nna coro Romney 1OS4S -akikuyamass Munehiro,- 2.22 5... 22> eeeeee 1090
REO MICO ICH Tass) cts nse tis SERRE Se OAS 1094. Kikuyama, Sakae ........ LS es WSS INI, US)
COMMN UK O sls to essen ced. ROS, Le 1133 1144, 1146, 1153, 1154
I GAOPPNYATUK TO) 5.2 a5. bool uns ROR Oe, Hee MiGSe AKetmura sy AWntkO gy. 2 3s arc ees a os Gal
RSAHOMIVAUKO 0 SS os sc s MRE ees OOS Akimunray Masashi 2060... . 2 ce. eee ee 1067
RAO A CTIKG ON eo eo5 acs iSa. 015 OPE 4Gy > AheimunassSumilkOms ose na: eee ee 1071
GAO ESE tSUKO Ne icine. cc wae Rs 1060" iKomurnasiladashie...... 0... ote. 47a
EEMOMEMIOMOV ASU 5... 5,..01sis.<.004:-06 SSE + ASR iGo, ghimuragMakeji— 22.5. co. 3 fons Soke: ees 1189
Rao VATSUSIN oo aces .c. «sae assc SEL A LOSGW RMON INO AKI es see oc ie ae Ve a ak 1085
TOW AEG S KI cssc906) «k,n SI, oe ii AinoshitaKWunmonleso.. . 3 oo. A ss. 1059
FEAST COS HIN AOA cia 5. oe ance SSR es SSE 68). Kaneshitasslsutomul. >>... -.. weiss oe 1128
CAS TITAM I YCOL CIN oct. 56-8 clave Ss Oe: 1059" sKinmokaeMasashit....5.5. 6.525. +. ete ae es 1107
KMatsurasayVoshimoto ............<...0 ae) boa HOSP -RishiedKayOshiter es: han to aie 1164, 1171
Katsuragawa, YOshinari ...........:2..9.. 0.84: HOGOE PRShiicl atINe tea ieee aco: ere stn cts ba SR 1169
Kayraeuchit Akihiko 1... o...6.0eecee ee tae: hOZOy. WKaishigamimy Akiko... o.io.0 5.0.50... Seas 1032
Keanna oS NKITA oo ee ewe oe ee woes LOSGH. WKashimotor: Makeo 3.0.4. hk ose. Se 1089, 1090
eawalvanastirOyuki .. x22... ses. be cewe sce ce POSE gheissschi bon 2 yan... oo oo: cae oes eee 1038
MCA AlN KOE Soo oc )ss =o ocsso.6 ole SNS TRIOS MOS aiitaieie touyanee te 6 hoo a oe epee: 1063, 1125
| SERVES CCUCTIRY is 6 See en LOfS2 AWitajimasiKen 2.408) 5 2 oak ee ibe ge: 1080
eae BKK Ores 5 fo oa asin cccy ene 2H. RBS: LOMO AKatayumia, Vakashi-. .. -...... J ss55 bane ee 1065
EAWalkcamil: YOSIO) Sc.tec ccs Soca bees cde eles 10295 “iKitamuras Kunio se . 086 os sk 1118, 1169
awamiuna, Kazio: 2... 6.60.52 ee ogee s 113. 11595" Kitamura omiichino .........2..6s TO AO?
Keawamina KeNaya ....encs sce decceetea sss OOF KWetanakasKazuyoshi =...) .4.-...2¢eeee se 1154
Kawamiuna.Kosuke .... 0.0.0.0... 5650: LA 4eaGe itanie WaInOShiy ee... 522s eg). eee! 1125
Reanwammuitras Mathis, .e ce oo SUS oe clanleles Se (kettarikeumilkOma.. cs secs cass os once 1019
Kawamiunas cay uukiy 4. .....2.... choke Ligdenlig2- sKitazawa. Yoshiaki) ..%.2.2..: hte aes IGA
Rawanishi-.b mikOe. se ekieh ae ees MOBAGS KG O RGU Ie eases ai crea ce «seve 1027, 1030, 1032
Kawaradas THayime (oo ,6 1-2 «deste oe MIs *Kiyomoto; Masato) 2 2.24.22 4c... Saee a 1109
|S ENWIGISET ONG, J 210) 1 ee nea ee 1022 Kobayakawa, Yoshitaka ............. 1102, 1110
Keawashima: Seuchino =... 1023.41125: 113441136 Kobayashi; Chizuy.............4..¢00e-8. os: 1114

44S = WKobayashivkiideshi . 4. 22....... . se 1148, 1149
Kawashima Tsuyoshi”. ...../46 2.2004 4. 22k L055" aKobayashiniKam 2.) oso. nos cee 1174
Re awcal P NK IO) 255.095, <5 bsayezessnso.2 REI Ss SOE 1255 > Wobayashinken=ichiro) -.4.....h ness. 6k 1119
KeawataypAtsuk Optic auc. <6. 3 2 je 103i sKobayashiyMakoto ..............des: 1037, 1038
KawauchirsiroOshiiee <.. 2... <i. <0 Mae S55) “KobayashinSatoru) ............... 1184, 1101, 1102
Kawazoes Yoshiyuki. ............ ceases . ayee: ible Shobay ashi, Shue. ..0 5: «4. cen: seen: - 1165

KAY AG AMS CII 25 hes cc a.cy. 3.01. SRRSRN OS «Se 1030% Kobayashi lakaaki «,.. 2.0.0.6. c9ae ee. 1065

1196

Kobayashi; Takakazur *+ 322 eee 1040
Kobayashi, Detstyal tscc.8.0122 ee ee eee 1135
KRobayashitelohnure.:..' a eee 1088, 1140
Kobayashi.eRomoyoshit.. ..s2.-2550 aes 1063
Kobayashi Watanu: «2s 1100
Nobayashin Wasuo nico baste fats A 1144
Kobayashin uta <s +2 e020 ears hae 1151
KodamasYumikos 2005... eee Ae 1181
Kogan Shing: +32 sovie 1 cca neers eres oes SH
KoganSiisenue i204 cat eeoe ease ene eee 1155
INopaiil cia cf. oar roo ae 1021
INohsiKatsukiank aust ee 1159
NohamaKazuhitow cs feck. eee. Sa 1067
KoheuchiGShiecomr ye... iiceten ed ceo nde IS
Koidex Voshiow ssn eee ts. VAST ee 1152
Koiwal, OSamit! =... hones os MOR 1077
KojimasDaisuke ar fecal ee eee 1032
Noyinias Manabuvker 3 745. j055 2 1044, 1094
Kokames Noichitesy 4523.35.53 6ce 0 eet 1032
KomiatsuneA katara isch oe oe SR 1027
Komatsus Micko«.c0se0 55. DES Ae 1115
Komulkai Masayuka:ssccyce. Ps 1084
Ondo s Daisuke: s.ccs ese he ee 1040
ondohhinoshi ssh ni oe cad cS 1165
NOnG oun 1s8 Ghetne ck et A: 1146
Kondowkyokoinrs snaked ee 1049
INondoMNoriakt)+s..4008.9 4554 eee fh 1146
Nondos Yasuhiko escaped eae Soe 1146
Kondoh*Yasulhino: }s29556 say cane ee 1026
Nonishr Kazuhiko: (49.03. cece eee 1063
Konishi, Kooichias.<. accadosces ER HI 1184
Nosakar Toshikazu t.::c...18200 000. ee ee 1178
NosekisNorimasaintss.< 20 ee 1167
KMoshidas Yutaka +s 2.2008). Ae 1164
Noshimizuy [Chilo :<.16 0 6264000.0ee ee ee 1144
NOtanies SUSUMU +..4.0..5 eit od. PR 1075
Koyanrawibhromichisy1.05.2 ee eee 1169
Koyama Sachiko! y.s..0.caitic. accrue Le he 1180
KRubo:- Fidekt eins sacra dee eee 1162
KubokawaiKakoni: 30. 22 eee Set 1138
Kubota iroshi Yio. 08 .00.. Bee. 1099, 1116
Kubota Ichin0! <05.15 14.44.0955 08: 1035, 1037, 1038
INUbDOta I ChHITOU cscs ees A RT 1147
Kubotasliomoyukii..d420 se See ee 1185
uchiiwat Satoshi ss40.0.c4, sR ee 1169
Kudo mA kihtkOe vcs dts >. HAIR RY. 1141
KudoxShigehait )4.c.0.)cnre cine AO ee 1097
udor Yuko Ae sae AOS AS Sc IRE JUILZ/

Kudo}: Yosinhisa’ 2 )0........0... 210 eae 1045
Kudoh, Masato si io:s.i0.0k0, 2 i G4
Kudoh Michiel iia... sc0400 ee 1173
Kumakura, Masahiko ......... o9230e) Sipe 1082
Kuno; Shinn ich “.\.... 5.0.0.0... eee 1115
Kuramoto, Taketeru .s.20..4.045.945-eeeeee 1037
Kurasawa, Yasuhiro. .........../ 05 0eeeeen eee 1071
Kurihara, Yasuyuki .............2.2 eee eee 1057
Kurita, Kaoru’ sna... 5s0ca0c eee 1066
Kuroda, Hideyo) ........:..025.40350540- 4 eee 1095
Kuroda, Masaaki. .............. 91206353 5ps eae 1069
Kuroiwa;/Atsushii - ........-.0.0050000 0g ee 1123
Kurokawa, Makoto .......s\...5454.08h Gee 1036
Kuroki, Yoshihiro: ©. «......003.55.425 1037, 1038
Kurosumi, Kazumasa ...... 4... 2oseeeeee 1142
Kusakabe, Moriaki ...............054505. eee 1112
Kusakabe, Takehiro G. ......22022 1103
Kusakabe, Tatsumi... ..........5.5.550 eee 1170
Kusano, - Tadashi»... 005... 005 eee 1129
Kusano, Tatsuro «.....0....05..00.0 ee 1025
Kusunoki, Toyokazu —.5.225-55-2 eee 1169
Kuwahara, Yasumasa .:..-....55. eee seeeeee liz
Kuwasawa, Kiyoaki .................. 1033, 1036
Kuwata, Osamwi....:......<...0.5 eee 1032
Kwon, O-Yu wo. ...0... 1075
L
Lawrence, John’M.. .........55..9. eee 1115
Lou; YH. Seo. Sk eee 1089
Tum), :Pek -Y e¢ 38.0)8335.45.50)5 eee 1018
M
MabuehitIssett 233735525 1066, 1072, 1098
Machemer, Hans: ...\......4.-5...05 eee 1043
Machida, Takeo oi...2).0...0. 2 ee 1145, 1181
Machiyama, Etsuko...........0..2 3 1027
Maeda; Shihoko.: «.................. eee 1028
Maeda, Ry twin... .cs.s2c:0c ee 1091
Maekawa, Shohei .............. 1073, 1074, 1098
Magtoon; Wichian. .. ~........c.-eeeeeeeeeeeee 1181
Makabe, Kazuhiro W. ............... 1103, 1104
Makino: Naoyalss...5-ee reer 1158, 1170, 1178
Makioka, Toshiki.....-:....nt05 eee 1162, 1163
Manki; Hirohito......0... 4S eee 1166
Manikawa;,, Yuusuke........):.... 0.0 110350122

Maruyama, Koscak ............ 1070, 1071, 1072

Maniyamas Shuka:s22f22ivs0. 5.2897. PS 1184
Maniyama,Ratsuyale<s5¢:e.s2 2 Pet S) gee 1173
IMasakitahomol: 2 ese) hs2¢s; 54. Baleee 2a 1069
Wasuda Atsuko thes... chose Te 1139
Masuda; Hiroko -..:. 2 6...22¢.650.640: 1027, 1030
MasudaniKaonl >: 22.33: coz os: IM PS 1097
ivMasumuntas Makoto-.::: 2.552068 1152
MasuyamaEtsuko:...s.efce. SMEG I ee 1074
Wasuyamia; Yoichi: :: 25 ..566.200/ 892 1108
Matsuba,oshihiro < os. <: e626 oe eG 2G 1150
Matsubara, Kaord: .: 63s. non PRG 1029
Matsuda KMOuneL «<2: ess. 22 SPE OSS, BOPP 1143
Matsuda, Makoto? .i:223 cece cencs Pe ee 1126
Matsuda’ Motokol! ofs5 . ccs oes bP es 5 O! 1124
WRAESTIGAOINGH YE lee hes bh. Sas ess HEAR Ae 1165
Matsumasa, Masatoshi ..................... 1160
Matsumoto, Akira -: 2... 6.55006 000)0 6. 1124, 1175
Matsumoto; Hiroko. 2:2 65:62 528.2 0. A, 1032
Matsumoto JiTo »s)6iccssccecreu, WES 1169
Matsumotos Kiyoshi ..:....25254.12 9.5 SY.
Matsumoto, Masami ....................... 1054
Matsumoto, Midori .:::5:. 6.2.6.) 000i sk. 1105
WratsumlOlOsSeljl sees isciaee cdc ddh Ge 1162
Matsumura) Fumio ....<.s55566. 8005 2. 80S: 1068
Matsumura, Kiyotaka ...................065 1092
Matsumura, SHIM <2 66 6 sce eas ccee Ua OM 1036
Matsuno, Akira: . oc. ..:55666 605: 1021, 1065, 1069
TA 2

Matsuno, Jun-ichi ..2......6¢65265 20 1160
MiatswokacAriKt iio hec8 testa ct sD 1064
WratSUOK Am NOs fci0 be ie ts see as AUS 1165
Matsuoka, Norimasa ...................000. 1181
Matsuoka, Tatsuomi ................. 1022, 1046
Matsusaka, Tadao: - ccccccs oso ccc BUEN 1157
Matsushima, Masahsi ...................... 1061
Matsushima, Osamu... 6.6.6... so. oe 1048
Watsushimas Toshiya sc f..2.e0885.50529.%. 1024
Matsushita, Susumu ........................ 1119
Wiratswuuraeetsu santos eo ds ces BANTIMES 1071
Watsuzaki, Takahi os... cece 28a) Lue. 1056
Matsuzawa. Petsuno . feces c er bi de, TASS, 1111
MatzumotoeEUmiO (sesh c2es.055. Hae. 1180
Mawatari, Shunsuke F. .................... 1183
Wiazd amit. -23 od eyondd eile be IME ALS EERE 1105
Michibata, Hitoshi ............. 1045, 1047, 1048
Michinomae, Masanao ......... 1027, 1029, 1030
1032

Micilar WOKOMRAS 2 rca sece oie SRA SRI, 1161

Mikami, Kazuyuki
Mikami-Takei, Kaori.................
MikamoiWazuyar. 2224500 052 wishtinwar Seis
Mikasa Kem jineen? 465257 sects sis Re 2s
Miki, Kazumasa
Miki-Noumura, Taiko ................
Mimori, Yuko
Mimura, Keiichi
Mimura, Shunji
Mimura, Toshiko
Minakata, Hiroyuki
Minami, Narihiro
Minesaki, Yoshiaki
Mishima, Masanori
MitasMasatoshile.< <6. <oo255 cn it A ee
Mita, Kouichi
Mitsunaga, Keiko
Miura, Katsuhiro
Miura, Ken
MiyvanMasuliko nee ce. oss 4st dae:
Miyabayashi, Tomoyuki ....................
Miyagaw ds FUMIO: . 4253528 #009
Miyagishi, Makoto
Miyaji, Kazuyuki

Miyakawa, Momoko
Miyake SKazushiger: 4s... 4.-- asec ee
Miyake, Tadashi
Miyake asulninoys aac e500) aun aaa he,
Miyako s umikorse yaa sa tenttanatae ee
Miyashita, Nobumoto)........-..-eseeee
Miyazaki, Junichi

Miyazaki, Katsumi
Mizoguchi, Akira

Mizoguchimbiayime, 2.22525 62. i: ae Seren
Mizukami, Atsuo
MizunosSatoshie<) 3 20.6365) AAAS
MizZUNOs MAKE Os soe 0h oo. 3. MS
Mizusawa, Hiroshi
Mizutani, Akiko
Mizutani, Fumihiko
Mogami, Yoshihiro
MohnesAideor. 5s i55 22 ss5 2.
Mohnitalastuiita, oc es.cs one irc BREESE
Mori, Ikue
Mori, Mariko
Mori, Takao

Mori, Takayuki
Mori OShOF nea i oss. o le ee

eecese ere ee ee eee ee ee ew eo ee ew eo

eee e cee eee ee eee oe ee eo eee ee oe

eocee eee ee ceo ee se ee ee ee oe ee eo eo oe

eceeee ee ee ee eee ee we ee we ee eo

eeceeec ee eee ee eee ee ee ee oe eo

eeeee see ee ee ee eo oe

cece e eee ee ee ee ee ee we ee ee oe

eecee eee eee eee ee we ee ew ee 8

eoeeeeec eee eee ee ee ee ee eo

Ce

cece eee ee ee ee we ee we ewe ee ee

eoecece eee oe eee ee eo ee we oe oe we

eee e ee eee ee ee oe ew ew eo ww ow 8

cece eee ee eee ee ee eee ee wo eo eo

eeceeec eee eee ee eo ee ew ee ee

eceece eee ee ee ee ee ee we we ee

ecee ee ee eee ee ee ee ew eo oe oe

eeoeee eee ee ee eee ee we ee we

cece ee eee ee ee ee ee ew eo eo wo

ese eee eo ee eee ee ee oe ee we ww ee

eeceeeec eee eee eee ee oe ee eee eee eee

eocee eee ee ee eo we oe ee we we ee ee ee

1155

1061

1042

1069

1087

12

1184

1081

1043

1096

1073

1198

Moniokai Mizuew itis «.«+ tee oo Meee ae 1076
Morisawa, Masaaki ............ 1045, 1093, 1095
Morishita, Fumihiro .................. 1019, 1021
Monta cA kemileeescutess: . incisal ee 1047
Monta loshiteris... 5.2.2: 5s0 eee 1122, 1125
Monta. sYukitomo: | 6:.../2.:5) ees Dee 1025
Montya,.Mepumit) .. i252). 3e angen oA .a8 1083
Momyars NaOmt cs.s. cass eh eon MOOR. cts 1117
Momniwaka Kazuo .: 2. o5s0. soonne ee 1055, 1056
MorodomeoAkihiro -s5.26..52-. oe. 1032
Motobayashi, Yumikow sa. eee aoe 1165
Motokawas Matsu@ os3s.c200 3. 7 Se Ce 1041
Muka Masanonliticcsccd5c ls... Jao eee 1078
Muneoka, Yojiro ........ 1035, 1037, 1038, 1147
MurakamivAkio. 252.302 5556.. ). Say 1053
Murakami Akita (0 cco 2 5635 bin ons 1046
Murakami, Masahiro:.......... 40:02). 258 1085
Murakami, Motohiko ...................... 1026
Murakami, Ryotaro ............ 1127 S556
Murakami, Toshiki........... se: 1151
MuramotoxAtsuko:.<2...a sea ae 1025
Miunaoka. Akkikomses: 23 oc. 605 5. oc Re Le 1040
MiinatasKeinpires gales xtiecia ed aac Ries A 1085
Murata; Makehide:. 20.255... seen: 1175
Murofushi,-Hiromu'.: 52... 29s. ee. 1075
Miyoharay Maroko. 2.42.0... .. Beene oe 1114
N
INadeesNivieeshi i... 2 ee. ee 1181
Nagahama, Yoshitaka.......... 1084, 1089, 1140
Nava, Takatoshiva:.,......... 2... ae 1035
Nagano; Masatoshi: : 2. 420... 2. Susan eee 1B
Na Ba ORE TiIKO) a bce jek... 2 ot ioe 1063, 1079
Nagaoka iRic vsice <2 is 32 soc Sp 1068
Nagaokay Sumihar.o. 5. ..,-2.2-. dete ee 1065
Nagasawa, Hiromichi ...................... 1152
Nagashimas@hieko..... -2.2.... -HipGath). Ae 1025
Nagata tSabuno 2. oi. ct kee see ee 1140
Nagatani. Elanumiey, . 222 22-20 Meee bse 1029
Nagayamas, Loshiki 74. °...)0.0 .onielaey ie 1023
INaito#Nobukol! 25.5. <.0026..8: See. 1141
INaitoh= Yutakay, 0608... kee ce ee 1043, 1046
INakacrerinOkiiyas. co sope tees s 1107, 1161, 1168
Nakagawa, iroyuka- 2.2. .....0. 0. 22 see 1067
NakagoshinMotok@. ... a.3.3. 0.2... ARABS. 1080
Nakai, Yasumitsuy cs: .jo0eec os os ARORA 1141

Nakajima iKumko 2262. c.co cal on SOS, 1118

Nakajima, Yoko... 23.4... 5.5. 1121
Nakajima; Yuuko.: .........4. .60:n5ee L179
Nakamura, Akira... ......csas.<.n ee 1057, 1108
Nakamura, Hiroaki) 2............eeeee 1159
Nakaniura; Jum)... ...-....0..052.0 eee 1058
Nakamura, Ken-ichi: .{........). eee 1074
Nakamura, Masahisa................. 1084, 1095
Nakamura)Mizuho: -........ 333533 eee 1169
Nakamura, Shogo ..........5..;850e 1044, 1094
Nakamura;;Soichi ......:..).. gcse 1044
Nakamura; Sumo .;... 5. ... 3.) 66g 1148
Nakamura, Takahiro ..:.....2.. 333350 157;
Nakamura, Yuri... .......... 6365 1181
Nakane, Masaki ...............43e eee 1125
Nakano, Hdrumin.2::....095..2.5) 05 1062
Nakano, Tomoya ......... lagieechd.peper 1144
Nakashima, ‘Seiko ....5........... 335 1088
Nakashima, Eiji.«.........:...gae0e eee 1029
Nakata, Makoto.’ ..............05¢55 ee 1150
Nakata, Yukiko»... 2.0...) 449 1044
Nakatani; [Samu ss). |... 2. 3 eee 1176
Nakato; Hiroshi. ::.....s. ...5.... SRB 1076
Nakauchi,, Yuji od... ..:...5.4)44 eee 1071
Nakazawa, Tohru... .. 25... ...243 1044, 1098
Namba, Hisaaki /........) (3:2 eee nga
Nambu, Jiro’. ..3.20.0. .sec55 eee eee 1108
Nambu; Hisao:!.............2.. 3 eee 1182
Nambu-Akiyama, Fumiko ............. ay ALOS
Namiki, Hideo... 6. :.)..ccs000 eee 1095, 1161
Narita; Kinya ..:...2.. 40a 1030, 1032
Naruse; Kiyoko.....:....:.5.225.50 ene 1148
Naruse, Kiyoshi\........:.2..43e:eeneeeeeee 1054
Naruse;/Masakazu .......... eee eee 1100
Naruse, Mayumi ....:....... 5. 2Reeee eee 1176
Naruse, Mitsuhide ......... 2a eeeeeeeeee 1148
Natsume; Tsugitaka........2: eeeeeee eee 1108
Naya, YOkKO)j08. 52. os. 5.4 eee 29
Neff, Anton W.. .is.3......... .1See eee eee 1102
Negishi, Sumiko: .............2.0:a4eeie eee 1164
Nihonmatsu, Itsuko .......... iis ls ee 1114
Nida, Akiyoshi: :. .........:. ..42teeet eee 1033
Nikaido, Osamu 2). 0.0... 4 eee 1080
Nimura, Kumiko... &\.......5. 2 3Gepeee eee 1167
Ninomiya; Kyoko. ..............G00e00e eee 1180
Nishi, Haruhisal :. ......05 +... ee ee 1020
NishisNozomiul ):........ ...-2aeee hee eeeeee 1167
Nishida sHiroki\estlos. 6 « cncccc deco eee 1104
Nishikata, Takahito ............ 1103, 1104, 1148

Nishikawa, Katsuzo ......................0. 1132
Nishikawa, Kazuyoshi ....................-. 1166
Nishikawa SVWUfl 220s escae. MUR 1131
Nishikawa: Tenuaki . 6:52 2.25 32895 1185
Nishimatsu, Shin-Ichiro .................... 1117
Nishimoto; Wohrue .. os sscc5 cc. OE 1022
Nishimura, Kenji: 2: cc: 022.2008 504. 1057, 1108
Nishino, Tomoko /. 5.0... 66692 8. Me 1083
INishtokasMidoOri .:.o6626e.5 esc. ces cs eek 1182
INGE OWI ee se bs ee ee ES 1110
INogawa, Hiroyuki -....:.5:52.5..000 0k 1120
INosucht,,Motoko:. .. 262... 82200: 1084, 1112
Noguchi, Munenori .................. 1043, 1066
Nomaguchi, Takashi A......:...20......6.0. 1052
INomuizus Motoyoshi ...:..2.2.25.0 0000.00! 1143
Nomoto, Kyosuke ................... 1037, 1038
INOMUMARMONI +2 oscavscdac in cacecde ved e een 1126
Nonomura, Yoshiaki ....................... 1071
INDSCMNGASUNITO 26) c2isckaebeteass JERE. 1150
Noumura, Tetsuo ........ 1136, 1140, 1144, 1181
Nozaki, Masumi ..................... 1105, 1146
NuksadassYOshiyuki ....224.0.:00.05. 005% 1166
INGTIOMPINOUICHL © 2 scene ce esi ones Pek Set 1066
INMManOSMIAKG 2465 65s. cess cies cae dew dblee 1143
Numakunai, Takaharu ..................... 1088
Numata, Hidehar......::....2.. 4.0. 0ei0e: 1063
INmianataw@SA@mUs:<s aes scssdsdecsaceds Jeedne 1071
Nunomuray Noboru. ....:.)...0608.62 220. 1183
Nunomuras Wataru ......525.665..00 009908. 1053
O
OannieCa7 NOM: sie scc ceeds ees pce eee 1046
@bararevoshitakayesis. ss cccee. oc YO. 1054
@batasiShuichies : cesses ec ce RN 1027
@bikatiMasatakaws .. os. esss a cewek 1021
@binatayAkikOgs; 765.222 os. DAP 1123
Obinata, Takashi ........ 1068, 1069, 1070, 1107
@ehrai. Gakehiko ...6..620:.5 5.00282 1064, 1078
Olas hoje Paes oi 84S 25 i acs SUR 1045
Ogasawara; Usuyoshi :...........)0)00 0.4.3. 1136
WeawatyK AZO rs is is ines seas ANPP S.. 1073
OcawaMizuho. i ss. eesn a. BRM 1160
pmo Kimihiro: 22.2.6 sees eee OS LOWS, lS
OsuroyChitanw:. 24... 22<5 0.40485 1115, 1150, 1151
OhashisazvyO: 3. os .452245 US 1067
@OhashiesShimichis «2444026644424 40 i. RS, S 1072
OhbasNobuyoshiy . ec ces oo. hes ol 1182

@hinatan Binoshis os eos se eens
Ohkawa, Muneyoshi
OhmiyassYoshihino;. 4. 432.65. 6a... =a:
OHMuUTOsEMNOM Be oe sass oe
Ohnishi, Eiji
Ohnishi, Ken
@hnishi WMRatswkO. 3 03) se ee
Ohno, Tadao
Ohokam@adakazZzuy: .. sesso. woes soos ee
OhsueiAojumess ose. fone men's 3
Ohta, Kunihiro
OhtacYasuliko. csosos ns co.cc, Ee
Ohtaka, Chiaki

Ohtake, Hideki

Ohtake, Shinichi
OhtakigWetsuya . 205 2..050 06.20. ee
@htaniRitsukonc. 65. 6c ae
Ohtoh, Keiko
Ohtsuki, Hisasi
Ohyamamvoshikunivas.Go. ss 2.55 ese ek
Oikawa Mane ak .:...osccs teh Oe ED:
Oishi, Noboru
Oishi, Tadashi
Oiwar Kaz unites. ss 6.5) oo
Oka KaZuyuktece. cies ae a ee
Oka, Mikako T.
Okcapes lum aaerr es a ta decry ates ee
Okeamloshinkatey ior cult he oe, ee oy he,
Oka, Yoshitaka
Okada, Jiro
Okada, Masukichi
Okada VoshinOne sco ok ee oe Se
Okagaki, Tsuyoshi
Okamoto, Ken-ichi
Okamotow Mole. oe <oecao05 5 SS ee
Okamura, Hiroaki
Okanowloshiywhkiys : 6 ssn4\ik AG:
OkanoyarwKazo aac oe ae
Okimoto Darren. 5205) bs. SA
OkinagarShoichi...2 sac c se ot ee
Okuma, Jiro
Okumotoy Naoto 2.5 2s, oc. on
Okumura, Eiichi
OkunowMakoton.. 2655.45 ooo et
Okuno Yoshinobu )).............40 ae.
Okushimay WaSuO 2655 aos so ce A
Omata, Saburo
Onitake, Kazuo ..........

TY

i Y

eee eee ee ee ee we eo we we oe ee wee woe ew

eeceece ee eee ee ee we we ewe we ew ee ww ew 8

ec ee ee eee ee ew ee 8

Ce i CC

cece eee ee ee eee ee oe ee eee ee ee

cece eee eee ee ee we we oe we ee ew oo eo ow Oo

ecee ee ee eee ee we eo ee eee we we we ew

Ce |

cece eee ee ee ee ew ew ew eo ew we

eecerecee eco eee ee ee ee we ee oe ee ee ee ow

eee eo ee

Ce

eee eee ee ee ee ee ew ew ew ew we

ecece ee ee eee ee oe we ee ee wow

eee eee eee ee eee ew ew we we ew ewe oe ew eo

eecee ee ee ee © oe ew ew we ee

eececee eee ee ee ee oe ew ee ee we oe eo ow

1050,
1080,

1147,

1084,

1089,
1044,

123

1093
1051
1130

1040

1036

7
1152
1095
1026
1141
1090
1045

1058

1200

1094, 1096
@Oofusa,.K eMac. 5 vi s0.80 ce ae LL
Ookax HMiroshisyigwis’s cave ee 1140
Ookia, SadakOne. ox. i keke os oe es ae 1022
Oota:;, Yoshihiko \..6.05..00 oe sac ee 1146
OovamaryMoshino “.2.4cadyawees ess eek oe 1165
Osanai SMinOnu seb avsace 14 Meee 1065
Osawa wMasakies 32 coseseca nas ces Re 1095
OSawabtSVOZO iis s:0'd sate aes 64.0) ee Or 1116
Oshimias. NOLMK Oi Walesa oh os ee oc ee 1019
O shim VokOrn iisans css oso 5) oA 1041
@shito,, Makashiy cco cs + isco 5 ER oe 1056
OlAMK OIC gai es Sow ca ee 1129
Otani Masumi. 4.3.02 o:: ordkes : Ae 1030
Ozaki WO 4.0 si ees ks i 1029, 1031, 1035
Ozaki) Mamiko) ........0555-..: 1029, 1031, 1035
Ozawa SatOshiiitey. gen. aso. ke ac 1138
P
Pana @enXtaneye ens eT, 1178
Rarike IMimaKee bso s-asces alone: 1134, 1136, 1141
PanlalesOsman, e026 oS oes oxen ass SAS 1130
BantnidoecJuliamy, «leis bee os ans ceo IR 1032
Pratt. Melanie: Mis... cs.. face 3. SER 1072
Rudles Juli @yoseiess oe hore aes bk SEERA 1172
R
Rutledge Laura Jos . 2. os.5e sec ee 1133
Ryuzaka Masashi tee:.........saeeee 1123, 1182
S
Sacousa, Pironaos 2.3... ees est. ee ee 1152
Saga, misuyoshii 3.0 10. .2.25... ARE ee 1146
Saigasprlidetoshic o55.5....... esse 1103, 1104
Saigatahan 2 joke es alse so ee 1068
Saigusa, Masayuki ...........0: #20. ee 1131
Saiki Atsusiicins ¢o sche. ld Le we 1086
Saka shomoakie .3 00.06 Seed ae Oe 1099
Saito gMiwakO e <a. ecco OE OO: 1118
Saito. Wakehiko, os 3.6 6. eds EE 1025, 1026
Saito, PWaASUNOM 6 6222 fee ke es AS 1049, 1113
Sakai, Hikolichi .......... 1073, 1074, 1098, 1172
Saka wWONSUKE whence ccac ass. « AMR a ae 1093
Sakai dVlakOtOue coos lon oe os RUE 1153
Saka UNG A is AEE 1080

Sakai,.Nobuhiko :.........:...:os5ee eee 1065
Sakai, Yoshic: 5..)...: i... 40 1094
Sakaizumi, Mitsuru ............ 1055, 1056, 1057
Sakakura, Veruyo! ......2...3...deeee 1122,1:25
Sakamoto, Katsuhiko .... 4caqto5;e ee oeeee 1029
Sakamoto, Shinobu °.........2. gasp eee 1136
Sakamoto, Tetsuya............. 1134, 1135, 1136
Sakata, Shusaku....>..:..2:.:2..3oseee 1135
Sakurat, SHO vss. jd. eee ee 1080, 1130
Sakurai, Susumu .......:22.-2 25050 1055
Sakurai, Youko .....:......5.... epee 1052
Samejima, Michikazu ..........sseseene ogee 1025
Sano, Yumiko: ........22... ) ene 1027, 1114
Saotome, Kyoko ..........A; asd pee 1054
Sasagawa, Hiromi ...........- epee M129
Sasaki;:Fumie =...........:20.. eee 1128
Sasaki; Junko >... ....... 0225) 3 eee 1052
Sasaki, Masami_............ bese ee 1129
Sasaki, Takuji........:..:....555 epee 1065
Sasaki; Tetsuhiko 42... 5.2. 2 eee ee 1075
Sasaki; Yasuharu ..;..........0. Geese 1173
Sasaki, Yuri... ... 05. -..0.2 os eee 1158
Sasayama, Yuichi ............. «poe =gli'50
Sato,-Enjl g5s6 e066 5.0425 2 eee 1117
Sato, Kanako .......:. 25: ae ee 1114
Sato; Katsuyuki......:.....2. «eens 1156
Sato;.Mika ..:..2.5...2.25 supe eee 1110
Sato; Motoaki.... . ccc. i022 oe ‘gaeree 1023
Sato, Noriyuki .............2. Wea 1166
Sato, Osamu... ...-25. 520-3 1071
Sato;. Takanori .......: .: 5.0... 1174
Sato; Tomomi.........4.70.555- eee 1138
Sato, Yasushi ........<.....50:55 dee 1182
Satoh, Noriyuki). .2:2...5.3 eer 1048, 1103, 1104
Satomi, Daisuke... ......... eee 1124
Sawada, Hiroshi .............:. See 1059, 1060
Sawada, Kazuhiko ...............03g oe 1136
Sawada,;Masami va0i))....... 0 seleeeeen eee 1132
Sawada: Taketoshi. ..-.......2. a3) 1066
Sawada, Tomoo.:........:...5050 seen 1051
Sawada, Yasuji .....:...... J2aseeen eee 1110
Sawai, Tsuyoshi ............2 2,55: 0e eee 1098
Seidou, Masatugul .2.... 2... ses seee 1027, 1032
Seki. Takaharu |... ..5....0.04 4: eee eee 1049
Seki, Tatsunori/ ......%...2 5.4.0. age 1125
Sekiguchi, Tatsuhiko .......2.2.2 seeseeeee 1033
Sekiya, Kunio)... .....2:.20:/. 0205 oto eee 1055

Sekiya, Nomko) ws... 002.2... .. 2a 1031

SemizZawasAVaiins. sic 2ed.odevee secs LAYS. 1051
Sendatewutaka were: haves chs. PA bE 1093
Senshwmlatsuors. tes 5.62. / RR i ee 1140
Sensuibs NODON «sc3 6666 600 SUES 1095
SeOm NAOMI es sr fs Fs So hd5 ooo oo EV 1170
Senzawa OSamuess.+5)a2sec.few his. wae 1092
SeoyaimawiOuUsuke: sis. 22222 44432) MERGE 1064
Sinibatay: Michio: <2 4022620055 SER VSR 1181
ShibataeNaokb << es eeesccs 3) GUO NS 1084
Shibuyas Kazunori. :2csis6.003 ee! UEY/
SmibuyasmMlatsuakt:. . sis. essen... 8 ole be 1034
Shichida,, Yoshinori .... <<<. Soe 5.4.. 2083 1032
Shigematsu; Hiroshi ....:.2.0.2. 680i. 1166
Shikeamra elie ss khei cd eeea se ey RAMS. 1064
Shimas-Akihiro’ x. 3 ..2.0056 65635 1053, 1054, 1056
Shimada Akemi. :.i..:.240.2. S92 Ae 1021
Sima@aswArata. sc. 2s5ss0cce5cs ee om 1090
Shimada Atsuko «<2 ..c5 scene EUR 1053
Shimada, Hiraku ........2:ces2e5 e040: 1105, 1127
Sinmadare TehiiO ej... 652 keke. ee 1175
Shimiadax Kimo: «2635.05 scenes es IN Lot 1178
Shimazaka“YOujl® 2... .2..5..5.4. 8A oe 1031
ShimizuIsamu «2. ss .525...0.3. 000 1029, 1030
Shimizu,-Katsuhiko..............86004 1166, 1077
Shimizu vKiyosht. 2.05.2... e200. s A 1167
Shimizu, Nobuyoshi .....................05. 1093
SHUMIZUE SMUMI Ie be. cen ane eee enw a | Lieee Om 1024
Shimizu} Makashi'............. 00000 1072, 1101
Simmizu. Vekesht .. 0... ....002.00 oh% 1085, 1092
SinMIztiey OSHIAKE «2.025 eee a wees Like 1025
SIOUITTUPATES DCE Oe se 1056
Shimoda, Yasushi ............. 00.5005 1080, 1153
Shimoimsals Masaki .............0.00ce5e- 1062
Shimonishi, Yasutsugu ..................05. 1063
Shimoura, Keiko .......66.05% Cio bu 1151
Shimozawa, Atsumi............ 1050, 1128, 1159
Shiney oj, Chikako «0.0.0.0... EOLA 3} 1044
SLOT Op, (GeO) CO eae eee 1179
Shimkat; Madashi..........-.4........ 22 1140, 1165
Shinozawar waka +. 5..cee sevice. SII 1112
Shiojini,,Nobuyoshi. ........... /fe ne Te. 1162
Shiokawa, Koichiiro ................. 1118, 1121
Shhimar, TOKO. cis. ..5:. 5. eveee oeleb eves wes 1082, 1109
Shinaie MoshikaZay 40.) OL, 15125
Shimaishi; (AKIO. tose. IE ABS 1021
Shmais MRsOChbe eres neh sorcchncee ens AU 1134
SHimaiShrs| Sats hy vccccssp. sok evsesr toes «ea. 1159

Shiramias Kazuhiko e050, A, 1064

Shimane sMenukivou.., 25. ssa. Soe. Ge’. < 1082
ShirasawaniMikioot ooo) hese. ote oak 1158
Shirasawa Yasuko. 22) su. ss 1158, 1178
Shirayamaoshihisa 35450-0504. .0.5+5 2c 1185
Shinloday Tetsuro. ses es oo eee 1063
ShiroyaseEsusioges... 2... ...,s cee aee’ 1164
Shiroyas MOkOe eo 142.3 ceed oe 1094
Shishikura; Fumio: o..).255.... 2. adh: 1050, 1051
ShofudawWemehi ooo. cio soos. epee 1076
Shudo#Koutchity2 ir oe a 1123
SimpsomaAleceWe Me coe os so ee 1039
SingtnpopsdappawanNl 24-2 2 ...5.9.5. seek. 1136
SOMPSENIVOM feo kee ks i oe SA Se 1114
Sofuoku, Shosuke ......... I ses Bae VAIL
Sonobe, Hanuyukt: ..2..:....5:-heeas 1129, 1130
Suematsus|Naova so)... 5. ...5-shenee es eee 1120
Sucnacae Rakahito:.o.2.<. 5. aged ie. eee 1155
SUCOK PERU: os joc s eos ee alee: 1060
SuganoMAinOsntes 22. . oo oss ose ae 1058
Sugano yMatsukO: 623 oe. 2c. 2. Shee etek: 1168
Sei, HanuOy. 35.5..22.22 0. 1039, 1040, 1041, 1042
SusieNaokOghs oo. ceeie os ce A ee 1034
SUSIMOLO; Keifip es vaso. 5.. os eee ee 1128
Suginosyiasuo Me: 4... ../. Reese 1138, 1159
SusivamavAkiko oo oo. fc 5 1 ee ae 1035
SueiyamayShin.. 2.55. 255-4. eee Lee 1101
Suhama, Mikio 2.00.05. 5 2. sane jaan 1157
SubaravMasahiko:.... = J.c....... tee. deere: 1063
Sumiday Masayuki 2) 5.)..4\0. 54> tenner teens. 1182
Sunayama ny oshiki. <3 /.... .xeeseeee . saa: 1033
Suyematsuslakashi <- 3). ......aceeaek 1063, 1084
Suztki AKinOf 1345.55 js. eee a ee: IBZ
Suzuka irofumiiy 6.0 oo. uo ee oes: 1182
Suizatka, alisashi), ai0c5.0 5 3 ste, SR ae 1110
SUZUKI TAIMOSHI 6.5.5 .c os oie win AR 1055, 1156
SuzukiMsunkO ted ecics ccc LAME 1103
SUZUKI KatUNOR 5s ckc sc oio.s SACRO cee 1095
SUZUKIM AZO sos Deter Ae eee 1081
Suzuka Masakazu ance... ieee. 1133, 1136
SuzukinMasartvite::: i. surmised. betes ce 1179
SUZUKI MINORU 2 ihaee sc siiud econo MA eee 1185
SUVAUGY INGO) 1 5s DAR erne c 1063, 1085, 1092

1093, 1126
SUZAIKGINO FY Ons cheat nec ajcpetasrnecs oR Oe 1134
Suzuki ShimtarOy sc. cco ase 1153
Suzuki Shoichiiss. irene ues ERR oe 1081
Suzukis Suechikea a5 soo ey ed, bk 1041
Suzuki, MakeOws 3 hoe cok. ccau pega. oak 1077

1202

Su7kt.: Pakuroisc\sccend saaans ee 1142, 1160
T
FachibanasKazunons 25265 3 ae Bt 1089
Pagawa,Masatomo'.22 Sass 225s 3. SERS oe 1152
Tagawa. Takasuke -sss.5 ee sy. A 1170
‘Naharayutaka-2: <<< feces ess, Oe ae 1128
Raita, KazvOwwe ss. 8-282 oes ess 1164
‘Pajikay:Ken-Icht\a. see se2 3 so 2scs | Sea 1184
Takada, Koji. jas sveeeses sss 20) he RO 1126
Wakagr, Tous 2 4ev ete sss DOM ae 1071
Pakapis Nenchi seen sd oicder ieee. ee 1145
Takagr, Bakashioe cares fens os eee 1042, 1063
Pakagire*YOshiomil es: 2225! fess 1S ee oe 1167
Pakahamea:., Hidekts 2.22 2..'5:.. 2. Se 1129
Pakahashi Akayoshi-. o.. ic .3 22 eae. 1147
PVakahashijPhisashi 3.5 + <2 ss2 52 sx. SIS 1174
akahashileWettehis s 5.46% = 3265.2 1039, 1044, 1046
PakahashijeKenji 222s c225c4%. PANO 1061
Pakahashi,-Kyohslchi" <2 2.2... .. 2. 2162S 1026
Makahashit: MakOtoes.<s.022005 2s. 2. ee 1046
Makahashiy MaSakies 22> ss. 222.4. SS ee 1166
Takahashi. Masatoshie 22.22... ee ea 1162
Pakahashi sMikio: s22255425.01 4. Rees SE 1024
@Makahashit Naoki) ¢252 2.200452. (Oe Bay 1162
Takahashi; Nobuaki=<3..22 2.3. .2. ee 1166
Wakahashi,, SOM: ¢4+ cree coke ener AOSONae2
Makahashis; SUMIO\ 87.5 cen ee. Se 1141
Takahashi, Susumu Y. ............... 1062, 1091
Nakahashr Tadao 4.255. +... Cae S65 tiS7
Wakahashi,,Vadashi-C. .2........0c dae 1116
Wakahashi, Takayuki <2... <... eee 1061
Makealnashity. VAI. 25 cccc han work «seks we Ree 1120
Wakahatal Masakazur <).05.4455 0. RE 1024
Pakale PluirOyuki 0.24 3 beckon ceca REE 1092
Takamatsu... Yoshiki << \.2... 020.5 Stee 1137
Makami- Makeshiviw tte. 0058 ck eee: 1052
Makamune-. Mazutumills.. 2c. enon eee 1087
Makano-Ohmuras Hiromi! . 2... eee ee 1069
Gakao, Toshifumil vh.0ki5. We OE 1063
Makashima: (Hum1O: i.e. ee. e 1056
akasupreNobom 2s .so56. ee 1137-1438
Makatar rNeISWKE <<) akdacoktotdane Dele: 1181
Malka IC MZ Or 0.5. 4toteacacessstaantuts ee RN oe 1116
MA Kata s KGW AKA: &.5.3. copsecocser rolsdextaeos A 1048
akeda. Hlinovuky 2652.00.38 es SRO 1120
Wakada. Masatsuners..45.(alcn tc 1184

‘Pane; ‘YOSHiOe" .. 2.23.2 oc eee 1049, 1147
Takemasa. Tohru... -:..<: 522: <nseee 1063, 1067
Takemoto, Kazuliisa ...........2) use 1095
Takeshima, Kazuhito: .:.....225)2 55a 1116
Waketomit: YOKO! 2... 2 eee Ak leg 132
Takeuchi, Hiro-aki . 4. -...... 2 177
Takeuchi, Shigeo”... .........22-- meee 1121
Takezaki, Kazuhisa: 222-2. -peeee 1112
Takikawa, Shinichiro ............-e=eee 1080
Takizawa, Satoshi |... ..).0)..2< eee 1096
Tamaki, Keishi. 0.222. >... eee 1094
Tameyasu, Tsukasa’. ......-2 fee eee 1039
Tamotsu) Satoshi. .......-... eee 1025
Tamura, Eimitsu.... .. 2.0.32 eee 1161
Pamuray Kojrte: . 0.150000 eee 1123
Tanabe, Hiroyuki ........ <4. 5... 1167
Tanabe, Kazushi ....0..,44s eee 1077
Tanabe; Kaztyuki........3..... 5 1084
Tanabe} KohkKi «35.60.00s5055) eee 1138
Tanabe, Youko"” .o)2..5.55s553 eee 1151
Tanabe, Yumi’... <.25223 54535 eee 1029
Tanaka; Hideaki . <2..:...5;355-3ee eee M2
Tanaka;'Hiroshi’...:....-..:.45)5..ee eS
Tanaka» Kosuke -. ......... eee 1036
Wanaka; Kunio :;.::::; 2s. 5ssssseee 1050, 1051
Tanaka, Minoru. ...::.;:.2.5 43 1089
Panaka, Seiji : :is2:c20.357592 eee 1095
NanakayShiceyasu;.- 2-25-29 1142, 1143, 1153
Tanakas Shoji ....0:<::s: 222) asec eee 1118
Waneday Koji s3 cass see ese ener 1022, 1046, 1178
‘Taneda; Yasuho. ..::::2:<¢.254.43eeeee lS)
Tanemura, Shigeji.........:-:. 45 eee 1118
Tami, Naoko. .....:.%...2.:2.249eeeeeee 1138
Tanimoto, Satoshi. ....... .26:33eeee eee 1093
Tanimura, Akihiko. ........-... eee blSit
‘TatenoS Hiroyuki .........:.cae9ee eee 1053
Taylor, DeMar =..........5..ccee eee 1076
Mazaka, Kenro! ......)..04..c8 eee 1041, 1134
Tazawa, Bigoro.............22.¢.2= eee 1058
Tet; Hajime 2... nek eee eee 1114
Terakado; Kiyoshi ............:--eeeeeeeeeeeee 1172
Terakita,/Akihisa. ..............<: G4: eeeEEeD eee 1031
Meruty Masami: )06.0...)..c.0 eee ee 1094
Teshirogi, Wataru .........c..<..-- eee ele?
Mochinat, SHIM. j.¢..c0..:.<.04 ee 1052
Nogashi,, Shin 0... s.cndenigee eeeeee 1114
Toh, Yoshihiro. ........0:.66.4.403 40. eee 1051

Tokiwa, Nobuyuki ................ coe 1184

Mokudas Kazuhiko’ secchrccr. MPO ORE 1128
Mokunagsa Puminon>s<652:. 2 e R 1051
Tokunaga, Fumio ........ 1029, 1030, 1032, 1035
iGkinties YUKON! Piet esces isc BEE BO 1070
Momarn Masakazu’)... f2.4 050 SEP 1085
onminaga’:Takashti:2.c23. 020.0 QPS 1043
sRomno, Shito-') 2.2 os... 0: 1062, 1076, 1077, 1081
(omrokaik enji' 2 fksscn.e Ee 7S alt76
MommitariHdeO: 00. ed ae ee eA ee 1057
Momita, Masahiro: ....22....00. 5. 2, 1076
Tomiyama, Yasushi ........... MOTT ARES 1154
Momuyoshi> Padashi: . 22.50.20 Kee 1071
Momonagas SUSUMU! <5 256050505. 2e ee etek es 1051
Momotkes Hideki (06 ......0.0 2 1142
Montookas Yasuhiro... ....... 6008.2 0000.05 1125
Monyama; Masaru’: .:.........! 1073, 1074, 1098
MasmyieelditOakd <2 224.523. SERIA Pee 1098
Momaka Tsuyoshi: .ii...4 280 VPN 1069
Movoday FumIyO:.. 2:23. va..cec bore sss 1139
Hioyotuku; Narufumi: :.2..2 20S ae 7S
fovoharas JUN. <5 ...525 45.05 RE 6 ORE 1020
MOvoIzZUM= RYUJl i.e. 236. Ee 121
Movoshima,-Y OKO: ¥ o.3...2..0.08 9 Oe 1073
MOWa SMUICHE fois... ss sc. ben Coe eo. 1163
SMCs ItaPUM : 6 eect ts Ce Pe 1081
WsuchiyaHiromasa «2.222... 45. .Fe es. LIS JA
Msuchiya> Kimiyuki -.......22 50705." 1055, 1056
MsuchiyaOkmort 2. 2.25. .02 Ne US 1127
MiseehiyaseNCiZOn fen vie beh hon buss Mel oe eae 1039
SUIGAPMICCO os eee ec 5p ok. ee, A 1130
Dsapimura, Hidenobu: -..........2.2..220).: 1114
MsmpitasNaotaka:.si.. 4.2 o.¢..e5s e535 RS 1170
Msukadas Shun-ichi ...2........5. 9:22) 5.0 1144
Misukadas Yutaka ....2.244%.4 AS, oe 1025
Msukahara, JUNZO: csi. csse. essa) PAE 1085
sukahara, Masatoshi ............. 80220. 1068
Msukalhtara. eYaSuo.s<28 2.458.825 SIE SS 1032
“DISUOLSTUL, CUO SG av ete ene ele ae ISS
Msunekio Kazuhiko. 2242506. cee ee eR 1164
Msunugal. LOT s.4. 6.24. e ee ee Se 1034
Wsuruhara,: Takashi+ .....2.4... AOU SSS. 1154
MSUEUMUTA.-SEtSUs is... .e.5.0%254 NE 1114
MStISUEsMIOLOO ..5:.54225440 460% 1059, 1060, 1080
MsutsuiKazuyosht <¢..c2.. . Hees 1134
isuyoshinkirohisa +. + .242. PRT: oP 1139
U

Wchida.sHirovomile et aioe. ne eet 1185
WichidarmVWasunoneeers .c3.s, so. Beek. sae 1183
Uchivyamawhidehor ii. en. eee... See: LILY
Wehiyaniasblisatoshive 52.5 eee e..Se a: 1030
Wehiyamia shoe: Said EDS Re oes I VZ2
Uchiyama Minoru... .....eeesaake HiSiesiiTs
WedarEGemiiOme 6 fac cke eis cc Men esas 1125
Wedarehinokoperr as... sss Fe ND. Oe 137
Wied amino shies wises nos ios 5 ES... SO: 1083
Wear, eee cae ca een kus MRA 1O27 4
Ucdarmalkayilki ee ccc: nc RY «Bas 1118
Wehana- lake mil | 420.cknke cos EIR RE Sse 1140
WekamatsuyaD 2 sale es ust Ia 1104
WemunaHlaniko: aes. ose. 1148, 1149
LOWEST OUTCRY. IS ie ee el nee Sn Aen 1097
Wienom NAOtOR Meee kee cee cine dell Se aex IG
WenowSatosie icc oe oe Be ot Oe a. 1131
Wenor mi atsuy Aer eri wate dara ul Me aeieute es 1086
WesakaMtloshthino: =). 50606: nha ned. she 1147
WiesugieNCASUO) scete cinta us pecans wanes aes 1137
keane WWiatanut soos osc ksals ox cd ae 1080
Wikawawikenichit.<:.caescseetos ts RR ee 1150
Umebachi, Yoshishige ....... 2222.7: 1080, 1081
WmonShulhsert ac 5 ae ee cs Baten 1086, 1087
WranowAkihisa, (06.0 cee cee: 1124, 1133, 1141
WO Stikie Mtane. cx Aes cet cece oo ee 1064, 1078
WW SuktimaeitOs seen bocca cho on 3 Ee 1027
itok Nomhiko otc fe icndon «5/5 Se Va 1092
WS STaZUO oes ace ee cciewud ws 6 eae eee 1183
Ww prirOShi cs ck cee, oo ee ee 1181
Wiyamnas LanOy cosa needa cscs Meee aie 1048
Ww
Wradar Masaru tc. dar p cee ae 1139
Wada, INaoyulkiiit. nnn n.d eee 1122
Wradas SRigeOs fj cciccn cr tee atte kee 1045
Wadae yak ani sa atcukion sv sanit eee eee 1181
Wakabayashit Katsumi {22 4.-..-.+-esss00-- 1143
Wakabayashis Shuichi: 22. 25; 6 f24.5 eee 1134
Wakahania.*Ken-ichi........ <u). Aaa 1182
Wakahara.; Masami. ....2...¢<.2 00 5... Ree 1102
Wakanayshigeharn: 5.25... 4.00 ose. 1055, 1056
Wakatabi, Hoshitsugu ..................... 1019
Walter vAlinediEs cncckec. BAS Ot 1031
Wane. Chen cal luaio.geciee sh cc ees ae 1055
Wraney YuRRISUC sien ceca k ete ee eel ye 1062

Washio, Hiroshi

1204

Washitani-Nemoto, Setsuko ................ 1091
Watanabe, Daisuke. <c sccunwas. RRS. 1111
Watanabe. Miroshiy .. 225-4065... 0R Ree eee 1113
Watanabe, Kanade s.ce:05 351 eee. Bee 173
Watanabe.dkazutadar..c..22.0.2) Ee 1140
Watanabey Kunithitoy 22... >... 20524. 6 1113
Watanabes Satoshi, 60.23 cock 0.40 Se 1070
WatanadbesShim) cc icc edn tn 1044
Watanabe, DPakushi. ... =. .22..2..2. dee 1147
Watanabe. VOKOnr dicks ic oeccenas ces ae. 1111
Watanabe, Yoshio ....... 1063, 1066, 1067, 1071
WatarieVasuhikois sic oc. ee Te 1179
Walliamsmiicd Pek ee i cain OS BE. 1031
xX
NIM TIM «Pee «6 ci ots os oe 1106
Y

Wabusaki sKatsumie.:....4.<2.... 88 1069, 1107
Néadam ak ashitertas. 26 oidscn os coe SR 1141
Nagamewarutaka . 3... .. eaten ase 1058
ia Sip IO ICIS A ARE oe 26 ak eine oo PORE 1083
Wahatanhensuke 3.20.25 jocccoh io Sees 1162
WamadiMelOmeoichifo) 4060. akon SER 1120
Mamaday Ka7zmie t;.,.css.0a0.5eeee 1105, 1127
iantaday Kazuyial oc eee ave ace se eee 1111
Mamaday Koen .2: 2c. Sauetc ase 1019, 1021
Yamaday Shigeru. ¥... 2 2.ce6s256 cee 1065, 1077
Mamadas<Lakenort (32.2 25.....60. 5 eee 1039
Yamagami, Kenjiro ............ 1085, 1097, 1127
Wamlacishi poinoshit.... 2. .1.20405 cae 1037
Wanrasishiwloshiyukiqe. 2-25-40. asso 1117
MamiaguchitKazGit..... 2js.cccci< 0 eee 1130
Yamaguehi; Kenchiro).. 25,002. cee 1050
Wamaguchi Kenichi: . 5 5,.2<2..<0-c60 Sane 1147
Yamaguchi, Masaaki ................. 1063, 1085

1092, 1093, 1126, 1133
Wamaguchis Takeo @ ees: Soh snares 1166
Yamaguchi, Tsuneo ............ 1024, 1026, 1033
Wamahamas Yuma 2.55 0......2. seen see 1090
Wamamoto,, Daisuke :..0....5.44552e 1027, 1114
Yamamoto, Kazutoshi ............... 1142, 1143
Hamamoto; Kiyotaka 227 .)-45- as sostack 1165
MammamOtos Matas. ioo2)s.2. sun SESS ee 1165
Yamamoto; Naoyukt 2.6, ..e6...0 feeeo see 1023

WamamotoxNOstko vis ses Sos dc AS ee 1083

Yamamoto; Takashi .......:...7stieeeee 1105
Yamamoto, Tomoko .... +... séateeeeeoeee 1021
Yamamoto; Yoshimi) :.....:.....55 poe 1092
Yamamoto; Yujii >. .....4 2552.02. eee 1033
Yamana, Kiyotaka ........2-. sqsseeeeeeee 1102
Yamano,.Keitsuke ../......... 93 1069
Yamanobe, Tomoyo}: .....:2.---- sees 1042
Yamanouchi; Korehito ... 2335.53 1145
Yamaoka sikwo 25.5... 1127, LISaeeS ee LiS6
Yamasaki, Akiko. ....:. 3). 222. eee (aT
Yamasaki; Hisashi... ....<. oe} 1126
Yamiasaki, Kenji: ...:...co: eee peeeeeeeee 1062
Yamashiki, Naoko... .2:..: -..93eeeeee 1110
Yamashiro; Shigeko ...-25.. 2544 1068
Yamashita; Kaorw ......:2) eee 1154
YamashitasKazu0: .-.- 2 eee 1128, 1130
Yamashita; Masakane ...:.-22.2.233 eee 1089
Yamashita, Nobuya:..... 7:25.) see 1170
Yamashita, Satoru. ...).:.2..-6.-- eee 1033
Yamashita, Tomohiro ..-23) so. 1157
Yamasuy Terufumi. 2555-45 1172, 1185
Yamauchi, Kiyoshi: ..°......... 4 eee OTS
Yamazaki, Hiroko l..........4 3232 ee 1059
Yamazaki, Katsuya ....:::-.05.-2 eee 1128
Yamazaki; Ken. ...:..,:.55.) eee 1105
Yamazaki, Mika ......:.)...933—eeeee eee 152
Yamazaki, Yoshihito’: .... 5. .349ee ee 1118
Yanagawa, Masayoshi <=. ..-..: sss ee: 1078
Yanagi, Akira ...... 22.5 /as9es2 sR Ree eeee 1168
Yanagisawa, Masashi\-.252 =. 2444s 1069
Yano; Junjii.2...6.3.:.2 35 2ee eee Sy
Vaoy Yao ous.) hs se o5es ee 1106
Yashima;: Yoichi ...........0s005 see 1160, 1156
Yasuda, Tomoyoshi ......2.5-.s:eeree eee 1071
Yasugi; Etsuko':::......2..2 550 -ee eee 1119
Yasugi, Sadao'.........+.. . 4a 1119
Wasumasus Ikuo) <7). 35550 eeeee 1058, 1105, 1108

1109, 1115
Yasumasu, Shigeki .......:..25. 65ers 127
Yasutomi; Masumi ...........4050s4s5eaoeeee 1167
Yazaki, Kazumori ...........5. .3e54 epee 1155
Wazawa, oORm:: 05. 3..0009 eee 1033, 1036
Wazawaiy Y oi¢ht! .S.......0..0c50000- eee 1064
Yokokura, Takakazu ..........0.42eeee e 1114
YWokosawa, Hideyoshi ......... 2: eee eee eee 1103
Yokosuka» Michio. .3...0. 4.0305 eee 1033
YokotasEBtsuO: .0..5 dco cows eee 1072
Yokota; WIKIO.) 6.09520 cn eee 1116

WokoyarSsachihikotganc cc. 00). 05 0a. ciceryae 1126
Wokoyama Shigeyuki... 0.025... 2. gene 1075
Wokoyamay YOshikOm ic). 0.3.5... 00s 2 1064
WonedasMoshimicht)) oy. Jai. . tees. 1173
Nonemunra, Masayoshingiin sia5.2n i... 1060
NWonezZawa BuUMIAKI sche. ube ek ISO, USS 7/
Miaone ZaWwals SAlOSHI ils. oc 6 oe bee les Halen 1061
WoneZawa. YUMIKO; 00... eben ee ein oe ee 1165
NOShIGawKOZI x. : hints bie sc. A a PROS 1128
NWoshidasManabuls inci... sce. PRB LS 1093
Yoshida, Masayuki............ ORM. Aston? 1037
Woshidakbakasht. (i... ce wenee lt. cca if35
NoshiGaaeWatan 2 sin. 6s cede eae e ee deers ney,
Noshigaki) MomoOnom 3:2 .0b6 eee 1098
MoshilanalaZo: nc... RS. cee 1032
Noshihanra, Masayoshi-. s.. . 02. 7iihie ee 2 ee fsa

Yoshikuni, Michiyasu
Yoshimoto, Yasuaki
Yoshimura, Kenjiro
Yoshimura, Ryoichi
Yoshino seinoshi Boxee he le ME IG,
YoshinowMen-ienit .. 0... eR. tee. dix
Yoshizaki, Norio

Yoshizato, Kats
Yoshizawa, Hid

Yoshizawa, Toru

Yoshizawa, Yas
Yuge, Masahiro

Zhao, Xiaofan

1205

Dae con te rl er ae 1099

eoceee ee ee ee ee wo ew ow ow

eceee eee ee ee ee we ew we ee we oe

1126
12S 29" 1665 1 167
L150) tetisie 1173
1030, 1031, 1032

eececee eee eee eo eee eee ee oe

utoshi ....
eki

uhiro

Ce rd

Ce eS ey

1206

ACKNOWLEDGMENTS

The editors express their gratitude to the following reviewers, who evaluated papers for
ZOOLOGICAL SCIENCE Volume 6 and Volume 7. Without their cooperation, the journal
could not function.

Aida, Katsumi
Akai, Hiroshi
Amemiya, Shonan
Ando, Hiroshi
Ando, Masaaki
Aoki, Kiyoshi
Aoki, Jun-ichi
Aoto, Tomoji
Arai, Yasumasa
Asashima, Makoto

Baba, Keiji
Baba, Shoji A.

Chiba, Akira
Chiba, Yoshihiko
Chino, Haruo
Chinzei, Yasuo

Eguchi, Eisuke
Eguchi, Goro
Ehara, Shozo

Fujii, Ryozo
Fukuda, Yasushi
Fukushi, Tsukasa
Furuta, Emiko

Hattori, Atsuhiko
Hayashi, Susumu
Hayashida, Kazuo
Hidaka, Tohru
Hidaka, Toshitaka
Higashi, Kazuyoshi
Hiramoto, Yukio
Hirano, Tetsuya
Hirashima, Yoshiharu
Hisada, Mitsuhiko
Hiwatashi, Koichi
Honma, Yoshiharu

Hori, Hiroshi

Hori, Isao

Hoshi, Motonori
Hosono, Masamichi

Ichikawa, Masumi
Iguchi, Taisen
Imai, Katsutoshi
Imamura, Taiji
Ishibashi, Takaaki
Ishii, Kiyoshi
Ishi, Susumu
Ishikawa, Hajime
Ishizaki, Hironori
Ito, Tatsunori
Iwamatsu, Takashi
Iwasaki, Shin-ichi
Iwasawa, Hisaaki
Iwata, Katsuya

Kamegai, Shunya
Katagiri, Chiaki
Katakura, Haruo
Kawamura, Kazuo
Kawamura, Kenya
Kawashima, Seiichiro
Kikuchi, Yoshiaki
Kikuyama, Sakae
Kimura, Takeji
Kishida, Yoshikazu
Kishimoto, Takeo
Kobayashi, Hideshi
Kobayashi, Makoto
Kobayashi, Michiyori
Kobayashi, Yasuo
Kobayashi, Yuta
Kojima, Manabu K.
Komatsu, Miéko
Kondo, Hiroshi
Konishi, Kooichi

Koshida, Yutaka
Kuroda, Yukiaki
Kurokawa, Haruo
Kuwasawa, Kiyoaki

Machida, Masaaki
Machida, Takeo
Maeki, Kodo
Maruyama, Koscak
Matsui, Masafumi
Matsumoto, Jiro
Matsumoto, Akira
Matsuno, Akira
Matsusaka, Tadao
Matsuzaki, Takashi

Mawatari, Shunsuke F.

Michibata, Hitoshi
Mohri, Hideo
Moriya, Tsuneo

Murakami, Yoshiteru

Nagahama, Yoshitaka

Nagata, Saburo
Nakamura, Masahisa
Nakamura, Shogo
Nakasone, Yukio
Nakatani, Isamu
Namiki, Hideo
Nishioka, Midori
Nomaguchi, Takashi
Noumura, Tetsuo
Nozaki, Masumi
Numata, Hideharu
Nunomura, Noboru
Nunomura, Wataru

Obara, Yoshitaka
Oda, Shujitsu
Ogasawara, Tsuyoshi
Ogawa, Mizuho
Oguro, Chitaru
Okada, Toyohi
Okutani, Takashi
Ooishi, Shigeko
Ooka, Hiroshi
Ooka, Sadako
Oota, Yoshihiko
Osanai, Minoru

Oshima, Noriko

Saito, Takehiko
Sakai, Hidetsugu
Sakaizumi, Mitsuru
Sakurai, Sho
Sasayama, Yuichi
Satoh, Noriyuki
Seki, Tatsunori
Shibuya, Tatsuaki
Shikama, Keiji
Shima, Akihiko
Shimada, Hiraku
Shimozawa, Atsumi
Shirai, Hiroko
Shiraishi, Satoshi
Shozawa, Akiko
Suzuki, Kunio
Suzuki, Minoru
Suzuki, Norio
Suzuki, Noriyo
Suzuki, Tomohiko
Suzuki, Yoshihide

Tachi, Chikashi
Tachi, Sumie
Takabatake, Ikuo
Takahashi, Hiroya
Takahashi, Nobuaki
Takahashi, Sohji
Takasugi, Susumu
Takata, Kenzo
Takeda, Naokuni
Takei, Yoshio
Takeuchi, Nobuo
Tanaka, Shigeyasu
Tateda, Hideki
Tobari, Yoshiko
Tochinai, Shin
Tomino, Shiro
Tomita, Hideo
Tsuchiya, Kimiyuki
Tsuneki, Kazuhiko
Tsutsui, Kazuyoshi

Uchiyama, Minoru
Ueda, Hiroshi
Uehara, Tsuyoshi

1207

1208

Uemura, Haruko
Ueno, Shun-Ichi
Urano, Akihisa
Uwa, Hiroshi

Wada, Masaru
Wakabayashi, Katsumi
Watabe, Hide-aki
Watanabe, Takao K.
Watanabe, Yoko
Watanabe, Yoshio
Watanabe, Yuichi G.

Yamagami, Kenjiro
Yamada, Mayumi
Yamaguchi, Takeo
Yamaguchi, Toshio
Yamaguchi, Tsuneo
Yamana, Kiyotaka
Yanagimachi, Ryuzo
Yoneda, Mitsuki
Yonezawa, Satoshi
Yoshihara, Masayoshi
Yui, Ryogo

1209

ERRATUM

In the article of Volume 7, Number 4, pp. 739-746 entitled “Inner structures of the cerebral vesicle in
the ascidian larva, Styela plicata: a SEM study” by Hisashi Ohtsuki, the title should be read “Inner
structures of the cerebral vesicle in the ascidian larva, Styela plicata: an SEM study”.

INTERNATIONAL UNION OF BIOLOGICAL SCIENCES ;

83RD INTERNATIONAL CONGRESS OF
COMPARATIVE PHYSIOLOGY
AND
BIOCHEMISTRY

TOKYO August 25-30 199]
New Hachioji Campus of Tokyo Metropolitan University

36 SYMPOSIA AND OVER 24 POSTER SESSIONS

PLENARY LECTURES: Knut Schmidt-Nielsen Duke University
Physiological Insight: What Nature Can Teach Us.
Masakazu Konishi California Institute of Technology
Brain Mechanisms of Sound Localization in the Owl.
Yasutomi Nishizuka Kobe University
The Family of Protein Kinase C in Cell Signalling.
Takeo Deguchi Tokyo Metropolitan Institute of Neurosciences
Molecular prnee of bun ae Neurons in Vertebrates.

Hosted Dy the ae Son for Comparative Physiology and Beene tny

Sponsored by: Science Council of Japan; Ministry of Education, Science and Culture; Tokyo
Metropolitan Government; Zoological Society of Japan; Physiological Society of
Japan; Japanese Biochemical Society; Japan Society for Comparative
Endocrinology; Japanese Society of Developmental Biologists — and 19 other
Societies and Organizations.

For information please write or FAX to: Congress Secretariat
3rd International Congress of
Comparative Physiology and Biochemistry
c/o Zoological Institute, Faculty of Science,
University of Tokyo
Hongo, Tokyo 113, Japan

FAX: (Japan 81) 3 816 1965 (After Jan. 1, 1991: 3 3816 1965)

D eve | O m e nt Published Bimonthly by the Japanese Society of
Pp Developmental Biologists
Distributed by Business Center for Academic

Growth & Differentiation Societies Japan, Academic Press, Inc.

Papers in Vol. 32, No. 6. (December 1990)

63. REVIEW: P. W. Ingham and Y. Nakano: Cell patterning and segment polarity
genes in Drosophila

64. N. Zisckind and R. P. Elinson: Gravity and microtubules in dorsoventral polariza-
tion of the Xenopus egg

65. H. Grunz: Homoiogenetic neural inducing activity of the presumptive neural plate
of Xenopus laevis

66. A. Mizoguchi, T. Oka, H. Kataoka, H. Nagasawa, A. Suzuki and H. Ishizaki:
Immunohistochemical localization of prothoracicotropic hormone-producing neuro-
secretory cells in the brain of Bombyx mori

67. K. Abe, N. Takagi, M. Naiki and M. Sasaki: A new monoclonal antibody that
recognizes 180 kDa polypeptide expressed on early mouse embryos and mouse
embryonal carcinoma cells

68. H. Satoh and A. Shinagawa: Mechanism of dorso-ventral axis specification in
nuclear transplanted eggs of Xenopus laevis

69. M. Murakami, |. luchi and K. Yamagami: Yolk phosphoprotein metabolism during
early development of the fish, Oryzias latipes

70. K. Shimizu-Nishikawa, H. Katow and R. Matsuda: Micromere differentiation in the’

sea urchin embryo: Immunochemical characterization of primary mesenchyme
cell-specific antigen and its biological roles

71. D. Zitfhian, F. Sehnal, A. Mizoguchi, H. Ishizaki, H. Nagasawa and A. Suzuki:
Developmental changes in the bombyxin- and_ insulin-like immunoreactive
neurosecretory system in the waxmoth, Galleria mellonella

72. K. Mitsunaga, S. Shinohara and |. Yasumasu: Does protein phosphorylation by

protein kinase C support pseudopodial cable growth in cultured micromere-derived
cells of the sea urchin, Hemicentrotus pulcherrimus?

Author Index to Volume 32

Development, Growth and Differentiation (ISSN 0012-1592) is published bimonthly by The Japanese Society of
Developmental Biologists, Department of Developmental Biology, Mitsubishi Kasei Institute of life Science,
Minami-ootani 11, Machida, Tokyo 194, Japan. 1989: Volume 31. Annual subscription for Vol. 32, 1990: U. S.$
148,00, U.S. and Canada: U.S. $ 163,00, all other countries except Japan. A\ll prices include postage, handling
and air speed delivery except Japan. Second class postage paid at Jamaica, N.Y. 11431, U.S.A.

Outside Japan: Send subscription orders and notices of change of address to Academic Press, Inc., Journal
Subscription Fulfillment Department, 1 East First Street, Duluth, MN 55802, U.S. A. Send notices of change of
address at least 6-8 weeks in advance. Please include both old and new addresses. U.S.A. POSTMASTER:
Send changes of address to Development, Growth and Differentiation, Academic Press. Inc., Journal Subscription
Fulfillment Department, 1 East First Street, Duluth, MN 55802, U.S. A.

In Japan: Send nonmember subscription orders and notices of change of address to Business Center for
Academic Societies Japan, 16-3, Hongo 6-chome, Bunkyo-ku, Tokyo 113, Japan. Send inquiries about
membership to Business Center for Academic Societies Japan,4—16, Yayoi 2-chome, Bunkyo-ku, Tokyo 113, Japan.

Air freight and mailing in the U.S. A. by Publications Expediting, Inc., 200 Meacham Avenue, Elmont, NY
11003, U.S.A.

Sophisticated Balance between
Safety and Centrifugation
Capability without Compromise.

Centrifuge in
Integrated with A
Refrigerator

Extra-Quiet
Operation

Ease of Loading/
Unloading
The Rotors

Quick Start/
Quick Stop

High Quality

Triple Safety

PORE AS ose

US NARA qe ey

Design
Corrosion
Resistance
HIGH SPEED
REFRIGERATED
MICRO CENTRIFUGE
move. MR-150
TOMY CORPORATION SOLE AGENT

201 SOLEIL NARIMASU BLDG. 31-8. NARIMASU 1-CHOME,
ITABASHI-KU,. TOKYO 175 JAPAN

TEL:(03)976-341 1

CABLE: TOMYSHO TOKYO FAX:(GIII GII)(03)930-7010

TOMY SEIKO CO.,LTD. MANUFACTURER

2-2-12. ASAHICHO NERIMA-KU. TOKYO 176 JAPAN

TEL:(03)976-3111

TLX:02723111 TOMYCO J

NARISHIGE

THE ULTIMATE NAME IN MICROMANIPULATION

OUR NEW MODELS WR-88 and MO-102M
MAKE PRECISION MICROMANIPULATION SO EASY!

SOME FEATURES of THE WR-88 WATER ROBOT MICROMANIPULATOR (3-DIMENSIONAL)
* Drift-free, the new WR-88 has a DRIFT movement of less than 2 microns.
* The new WR-88 has a SMOOTH MICRODRIVE MECHANISM.
* An Aqua Purificate remote control ensures totally vibration-free operation.

NARISHIGE SCIENTIFIC INSTRUMENT LAB.

9-28 KASUYA 4-CHOME SETAGAYA-KU, TOKYO 157, JAPAN
PHONE (INT-L) 81-308-8233, FAX (INT-L) 81-3-308-2005
CABLE : NARISHIGE LABO, TELEX, NARISHIGE J27781

NARISHIGE |

JAPAN

my

at

‘i

i"

ZOOLOGICAL SCIENCE

VOLUME 7 NUMBER 6 DECEMBER 1990
CONTENTS
REVIEWS
Fukumoto, M.: Morphological aspects of ascidian fertilization ............. 02. eee eee cece eee ees 989
Hill, R. B. and K. Kuwasawa: Neuromuscular transmission in molluscan hearts .................. 999

PROCEEDINGS OF THE 61ST ANNUAL MEETING OF THE ZOOLOGICAL SOCIETY OF JAPAN

Abstracts of papers presented at the 61st Annual Meeting of the Zoological Society of Japan’ .... 1013

Abstracts of papers read by the Zoological Society Prize winners

Asashima, M.: Embryonic induction and cell differentiation during the amphibian

development. «os... 60h eee clei g gee Senn oGagd eee ded digs on Oe ee 1015
Mabuchi, 1.: .Mechanism of cytokinesis im-animal cells ©... 2-2-4. .452----e eee 1017
Physiology ....... Pee bes goncdcesuhe tothe ave meee eumeteds ogee eae ae 1019
GEMELICS {0 eee ee hed ee einen ee eee wie snes Hak Ghee See eed Deeb ee 1053
Biochemistry: 2.0.5. dices deed (ae eae saws sacs ae eee ieee tes are ioe LOSe
Developmental biology 2.6. b ee fee ee ee cae bd nba i a sen ee 1082
Endocrinology: —o.2 2022s code dec gee Feces becca be weenie awe ons ae ATO O) se eee 1129
Cell biology and Morphology ©... ...0....0. 0.2 ne cc bee ene ce oo oe 1154
Behavior biology and Ecology. <2... ... 2. essence ives vas oink See 1174
Taxonomy and Systematics - 9. 020.2205. ebb koe pete cee neals os bale 0 eo coe 1181
AMMOUNCEMENUS, 92.6). i. ee es eS de ae west Sele odd ee ae baw sib es oe edi Oe 1186
ACKnOWleGdements = 3085 ase ws aceon ce oo ea Se cies sales ele Gate 1186, 1206
NuthOr INdeX? 2. Ses wl awe eRe oe Lacie ine SR LAW Wow dene + eon ee eee 1187
GRAGUIM oe 5 acs oes oa Gat od ae 8 Ae RS im ae Sea cu tite ke Wink Sone anne ere dee oO eee 1209
Contents of ZOOLOGICAL SCIENCE, Vol. 7, Nos. 1-6: —.....2...2.2.22-00.. 2 ee ee i
INDEXED IN: | Issued on December 15
Current Contents/LS and AB & ES, Printed by Daigaku Letterpress Co., Ltd.,
Science Citation Index, Hiroshima, Japan
ISI Online Database,

CABS Database, INFOBIB

eee ea a <

RAS

cP ER 2 EE ge) = -ée gel > © G7 = Ee
2 ¢ A Ee % ks) es rea e ro) a = Vy ije a 7: : Ze
m CtiNoseD o Ms Washi m ASS = <j Waser m iF ie m 22)
ISHLIWS S3 \uYvud ee BRARI ES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S43 1d
‘w - (ap) ,
= = = < = wy, = sy
= S A Z IN a ly = =i
ae \N “fy,
8 NAG 5 Wi YZ3 z 9
-. = AW tr WN YH t= oS G7
=. = RN SS = = AS Tg Ie = a
> = ~. > = YN > = a
2 ” = ” Loess z= “” =

ISONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS saiuvugi LIBRARIES SMITHSONIAN INSTI

= a) — ” = ae
Uy, & 2 av : 5 u 3
yp Y oie | xg 2 x - ff tf ee Hi, mt
fe? & oc 4 s& NW = S ow S
ee fan} oO ~ a -— foo =
2 ie 2 a = z ie il
S3IYVNaIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI ee fi -
z 5 a as 7 2 is i
= 20 > Bs a xy =
= > = oa > ra > roe
= Pe) & A — Pe] —
i = uv) a a i oD .
Pe o = o = o Mie
ISONIAN (a NOILALILSNI ara eva ati BRARIES SMITHSONIAN INST
} = = Fee SO = x R
= “4 ze = z NS
: i Gif 2 : a
S Zz F z= = N
eS Pi s
a = ‘ w
Sagiuvugiy LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS

phate | $3 r
PK

a
SMITHSONIAN

NOILOLILSNI

NOILNLILSNI
LIBRARIES

INSTITUTION NOILNLILSNI SJIYVUSIT. LIBRARIES INST

MR We
‘S

SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3te

SS \ ¢ In, 47
S3'1¥yVYditl LIBRARIES SMITHSONIAN

INSTITUTION NOILNIILSNI
SJIUVYSIT LIBRARIES

INSTITUTION
INSTITUTION
SaiyVvysl)

ISHLINS S3IYVUaIT LIBRARIES

we

INSTITUTION NOILNLILSNI NVINOSHLIWS

we

SS'1YVYUdIT LIBRARIES SMITHSONIAN INST

NVINOSHLINS S31YVUGIT LIBRARIES SMITHSONIAN

NVINOSHLINS
SMITHSONIAN

Na
Sy VAS
NS
SMITHSONIAN
NVINOSHLIWS

NVINOSHLIAS

Wii
Dj
TS

LIBRARIES SMITHSONIAN

NOILNLILSNI

NOILNLILSNI
NOILNLILSNI

NOILNLILSNI

LIBRARTES

SHLINS S3J!IYVYSIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31

w
uj
ce
<<
a
om.
5
y z iz z - Zz fs z
AS = 0.8) 2 “Oo 2 ti =
Re ce 2 = me) & w =
» Ee ss a = = 2 iE
ce a a Os Z & Ae:
SONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS S43 iuvud riot BRARI ES OMe
2 3g = < = ie eS Zz
SS ae’ 4 ‘ a =| = ®S
ZN fF S a S = a
2 N&R 2 z By 2 z 2N
E Z Ee ey E Z E .

ee) Wg 5 7 oe Ss s] N
ie SS E a = = Se
we mn : Z ae
Ww ee w ion =
SRARIES SMITHSONIAN INSTITUTION NOILMLILSNI_ NVINOSHIIWS, SAlyvydiy LI BRARIES SMIT
‘ za * sue .
= Za = ee = < =
= = oN 3 G%,= = S Sj
= ra N\ = GY O =r rs) Se
8 2 N38 GG? S we
= ay AS Se Aha Z ae z
Se ae : 2 3 :
_NWINOSHLINS S3!UYVNGI1_ LIBRARIES INSTITUTION _NVI
i - a
a ud a ws = us mice
ae ac a e a Fa aha
: AY ce 5 oe = c a4
° aN = S a 3 Z Si
at
BRARIES SMITHSONIAN INSTITUTION NOILMLILSNI NVINOSHLINS S3!luVvudit LIBRARIES SMIT
5 = = SE z S
= wy 2 w .) = o
EGG E on F = ary 4
= — wn <= op) =
LNLILSNI_ NVINOSHLINS $3 lYVUGIT LIBRARIES INSTITUTION NOILMLILSNI_ NVIN
ax ce 7)
‘S eat jf hg ; = = Zz = z
; 2 “iy F 2 : 2 =
= = as A ~*~ = 2 ae S =
SMITHSONIAN INSTITUTION NOILOLILSNI NVINOSHLINS S3INVUEIT SMIT
‘ z a Be Z f
li un =e ap) AS RS wn aes
es: = ao Ae LEO WY ox pete (a a ee
a = ze : NAS < za < ivf
ac 4 & a4 aN WN me 4 Ee
= S = S ee S eae
if ; z ae Zz eee Zz 3
LIMLILSNINVINOSHIINS (S31YVYGIT LIBRARIES SMITHSONIAN, INSTITUTION | NOILOLILSNI_NVIN
o aes - O. Hae st ie svi 5 fs =
as] S = ow = w = w X
ps) EWN = Pe) = ual 5 ie
a SA Bs zs : ve E cao
a KG - a Ee a = a fess
= SOS a “Le rm no m
o ee oD z o z a

BRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31YVYUSIT LIBRARIES SMIT

w
g = g Z i yee =
: 2 Ng : 2 : : :
fend. OF) SX eu OD OD n ” ”
O +S Ww? WE G 32 O = O
a - AS Zz. om Z a =
wos = YW >" = > = w =
> = >. =
NHLNLILSNI NVINOSHLINS S3S!YVYYSIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_ NVIN
=e iW Z n S @ Eee Pk
n me ” a oO a wn :
4 & a = Fae = =
= e = aad bar fa + |
O es O an o eS O
za =J =z =) Fas ad =
BRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IY¥VYSII DERG Zap
= z
= Ow je) 6 S)
= > e) = - i TF. 5
% kK Ts wy
E ° = a = = Z
Ww) a n” Ww) w 4
= a = - =

S SateVesl 1] LIBRARIES NOILNLILSNI NVI

YVINOSHLINS S3ZIlYVudtit

VINOSHLINS S3ZI1YVYUEIT

JINOSHLIWS
MITHSONIAN

MITHSONIAN
MITHSONIAN

SMITHSONIAN IN

Ab ents Pbu ARS? ; “4 i i
+ i 2 r + ‘ Wei: bol “1 yet
hae es] ni Matis'§ ae A : eta H 4 ; " ' is
en Sh fh ‘ fs ao i ; A ; : PE ‘
rec hoaat Mi ADR Beh a ii : : t } et
Abie nt Minn eg ay ‘ OMEN af ; ' aga i ‘
7 sss TmARe vib ee , nets t . et j ‘ failed dgein aa Pay % :

OND ee CoAT HD hap

een Nani a Sa a iy ‘ Trae a ne ie in 3 908 88 012 261 1 2735

AD in ea) is bee hres. Ha
Tih SF Vo par ah.
a> tnty tee fie
DODD esds | ikon pay
Noh Be ey Royer
any Ava t Sen Cee DS
i wher ome) iat
VA ety bP eles ey

Ay dente ‘ne ' ‘ cathy, i STE AW Sets
Spe sintaty itp hehe) 5 S Z a

ae
TAP
Nae A eos
PB re ated eb on
7
So andl se ge
4855 Wa AIN Phe Bodie yh Ae Be
Ave pide

weg e G3
fenreig

AP ewlavenet
LORE LA vive)
Sere are
Se eT Case eens

SNe diets ae sbeee
ae oe

SA Ra TP A ee
Rind dm Orato

Tks edarbres

eb aR2SB Ss Ba
Sed. witiea ree

-TEEXIIBTSAD >

reach ep sbe “es

Suehageens,
Say &
APR At Soe ney

ate
PRU 6M

vy “4 Dy

PEE Un
STA
Pe ker be,

; Op aed,
nrg Peta oe
may crs Aaa EY a ,
Seo as i Hebe ew BYNES Veh we
nes ides

BN be IONE Mii cE AB
ae pet Fe DNE Se hone aaah Par a

APU Migedigo darn hades fe Y AL os)

yt euys
tee Meath) Ys sae a eee Dab as
ee By 19 PEs, Vargpe ses fr yee
ee yma a a» Ks
AYE Me WEGNER Dy OH,
MPI, Alt we ve)

por yad* Path Megas ay)! Ob § PNM ne SE, YO
HR Deaen Perey, oat » 8 £ 8 cart Pac |
PRL Baye ay Sep par F

PQ 4 WINE Ae Te

ok Debi Layee) 2 vt

being 4:

KY BOYS

NPSL eee

a2 Abn da ty hy

Tae TWD,

my

ry y Bike % hen, Bd,
He oyiee is Yet Pi ENE ey:

7% ener Wy PAS YT Og, w Sy? i,
il gr GA Ped, ey ye o 4 ee

AOA HU BEF ae rg!
ke 2 DB) AVE,
HN Dh EO ain at ak es Ee