wr ^ 1 VI W U U I 9

v^iwrw-irviu^ oivii i nouiilrtli UNO I I I U I Ivli
; CO — 00

UJ /jttffar _ „ ^^nx^N. Ui /

- >4

\£V cr
<

^5

vasn

o

z — _i z

LIBRARIES SMITHSONIAN INSTITUTION NQIXfUIXSNI NVINOSHXIWS S3I8V88I1

O) zz (S) ^ m z x^iiis!iX m

rUTION NOliniUSNI NVIN0SH1IWS S3 I ava 9 n“u B RAR I ES^SMITHSONIAN ~ INSTITUTION 05 NOIXni

*- - w z M 21 A.., C/>

VB8n_ LIBRARIES SMITHSONIAN _ INSTITUTION NOIlfUllSNI NVINOSHXIWS^SB IBVHBn^LIBRAI

— . _ 00 ^ _ _ _ <*> ~ oo —

^ . CO /'^'5SS>>v ^ xXvV^^M'X ^ cu sty&M

^ S tf2P I £ IC^I § ICii

Pn. o.c

O O -*w:- _ >^^Uf£V><' Q -»T 'N^UU3^>/r (0

UTION z N0IXnXIXSNI^NVIN0SHXIWS*S3 I HVHBn^UB R AR I ES^ SMITHSONIANJNSTITUTION ^NOIXflX
- /£2S^\ o A - X^i^ts rn /<r2225x 1 O

-jy
>

f

s^N

18811 U8RARI ES SMITHSONIAN INSTITUTION^NOIXflilXSNl'NVINOSHXIWS^Sa I MVdan ~LI B R Al

JO
CO

3 'W I '*m | *$r i>~"” S W' *■'

UTION ^N0lifUIXSNI_NVIN0SHXIWs“s3 I ava 8 ll\l B RAR I ES^SMITHSONIAN^INSTITUTION ^NOIXfUl
- /£2SS$S\ JV ^ W X^gvSx “ m "

O X^OStj^X __ o

?aan libraries Smithsonian institution NoixnxixsNi~JNviNOSHXiiNSZS3iavaan~JLiBRAfi

Z h- Z r- -, r- 1

^ - - — JJJ 2! m ^ £ X^xixs^ m

UTION NOIXfUIXSNI NVINOSHXIINS S3 I 8 V8 8 11 LI B RAR I E S ^SMITHSONIAN ~ INSTITUTION ^ NOIXfUl

I 1 I *$$> 1 ^ 1 i

fa8n_UBRARIEs"sMITHSONIAN^INSTITUTION"NOIXnilXSNl2NVINOSHXIWs‘/,S3iaVaanf UBRAf

5 2 '"W 5 | 1 ^

ution z NoixnxixsNi^NviNOSHiiws^SB i a va a ii^li b r ar i es^smithsonianjnstitution2 Noixnxi

.— d> 2 /tSSIfes, m XH55X Q jrM’y.. Z xf?S?oX O /SS™?!

- « V. • . UI a r\nn ! to DfVm HiUlN IAIN

_ _ . _ CO

7k UJ

^5

Mi o i i i u m w in ncMjLiu.iiam iNviiNusn±i ifib
co

O _ O X^osv^iX “ O

-J z -J Z _i 2 _

SMITHSONIAN INSTITUTION NOlinilXSNI NVINOSHXIINS S3IHVU8I1 LIBRARIES SMITHSONIAN IN

^ Z r- ^ r- z r-

m ^ XOiii^x m ^ z m z ''40F/ m

NVINOSHXIWS S3 I HVaail LIBRARI ES SMITHSONIAN "INSTITUTION ^NOIlfllllSNl'” NVINOSHIIINS^S"

— ,vv. co z co

z ^ > x^tus^ ^ ^ > '^par

>MITHS0NIAN_INSTITUTI0N^N0Unj.U.SNI_NVIN0SHilWS^S3 I a VB 9 n^LI B R AR I ES^SMITHSONIAN^ IN

«*I ^

^ 5

O ^ ^ _ vgu ws^y o

iviNOSHiiiNS S3i8vyan libraries Smithsonian institution ^NoixnxixsNi^NviNOSHXiws^ss

V o - - d _ _ _ S [I 2 _ f” v 2

— i-n Xfolvox O . ~ o X^ttF^x ~ \v O

_ ^ m ^

MITHSONIAN INSTITUTION NOIJ.rUU.SNI “nVINOSHJ.IWS^SB iavyan_UBRARI ES ^SMITHSONIAN- IN

w 2 CO . ' Z > CO 2

2E ‘ ^ \w > ^

(/) **'x 2 </} " 2 CO A' 2 ^

IVIN0SH1IWS S3IHVaan_UBRARIES SMITHSONIAN INSTITUTION NOIXDXIXSNI NVINOSHXIWS S

UJ 2! ^ Z _ ^ " <”

CO

O X^vos^^X ~ O

-1 2 j| 2 _ _ _,

IMITHSONIAN INSTITUTION NOIi.fUU.SNI NVINOSHilWS S3iaVa8H LIBRARIES SMITHSONIAN IN
r"' z r- r- z r*

.#H|

siovy

IVIN0SH1IWS S3 I avaailLIBRARI ES ^SMITHSONIAN INSTITUTION ^ NOUfUUSNI NVIN0SHillNs”s3
«S»*Sv 2 < .^T s ^ ^ i 4X ? 5 w

~ f£ ,^v< 2 .< Xgsogx 2

MITHSONIAN_ INSTITUTION^NOlifliliSN^NVINOSHilWS^Sa iavaan_LIBRARIES SMITHSONIAN _ IN

, W g <&s. w M /5?S^X ? , w

=! ®

O

c ~ Vo'

^ _ m V/^_ ;^y 5 XorTocjx “ o

VINOSHimS S3 lyvyan LIBRARI ES^SMITHSONIAN^INSTITUTION ZN0IXnXIXSNl“lNVIN0SHXIWSZS3

\ O - - ^ [I ^ z «" _ 2 _ r- v z

^ >XTaTIT/7^. ^%n i7?v. O — - - - —v —

9&H I
H. 1

No. 4

August 1988

ZOOLOGICAL

SCIENCE

An International Journal

PHYSIOLOGY

CELL and MOLECULAR BIOLOGY

GENETICS

IMMUNOLOGY

BIOCHEMISTRY

DEVELOPMENTAL BIOLOGY

REPRODUCTIVE BIOLOGY

ENDOCRINOLOGY

BEHAVIOR BIOLOGY

ENVIRONMENTAL BIOLOGY

ECOLOGY and TAXONOMY

published by Zoological Society of Japan

ISSN 0289-0003

distributed by

Business Center for Academic Societies Japan
VSP, Zeist, The Netherlands

ZOOLOGICAL SCIENCE

The Official Journal of the Zoological Society of Japan

Editor-in-Chief:

Hideshi Kobayashi (Tokyo)
Managing Editor:

Seiichiro Kawashima (Hiroshima)
Assistant Editors:

Takeo Machida (Hiroshima)
Sumio Takahashi (Hiroshima)
Kazuyoshi Tsutsui (Hiroshima)

The Zoological Society of Japan:
Toshin-building, Hongo 2-27-2, Bunkyo-ku,
Tokyo 113, Japan. Tel. (03) 814-5675
Officers:

President: Nobuo Egami (Tokyo)
Secretary: Hideo Namiki (Tokyo)
Treasurer: Tadakazu Ohoka (Tokyo)
Librarian: Shun-Ichi Ueno (Tokyo)

Editorial Board:

Howard A. Bern (Berkeley)
Horst Grunz (Essen)

Susumu Ishii (Tokyo)

Roger Milkman (Iowa City)
Tokindo S. Okada (Okazaki)
Hiroshi Watanabe (Shimoda)

Walter Bock (New York)
Robert B. Hill (Kingston)
Yukiaki Kuroda (Mishima)
Hiromichi Morita (Fukuoka)
Andreas Oksche (Giessen)
Mayumi Yamada (Sapporo)

Aubrey Gorbman (Seattle)
Yukio Hiramoto (Chiba)
Koscak Maruyama (Chiba)
Kazuo Moriwaki (Mishima)
Hidemi Sato (Nagoya)

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 more than 1000 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. Seiichiro Kawashima, Managing Editor, Zoological Science, Zoological Institute, Faculty of
Science, Hiroshima University, 1-1-89 Higashisenda-machi, Naka-ku, Hiroshima 730, 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 1988, The Zoological Society of Japan

Publication of Zoological Science has been supported in part by a Grant-in- Aid for 1
. Scientific Publication from the Ministry of Education, Science and Culture, Japan. J

ZOOLOGICAL SCIENCE 5: 727-731 (1988)

© 1988 Zoological Society of Japan

REVIEW

Taxon-specific Crystallins

G. G. Gause

Institute of Developmental Biology, USSR Academy
of Sciences, Moscow, USSR

ABSTRACT — This is a short review of taxon-specific crystallins identified in eye lenses of various
animal taxons during the last few years. Possible implications of this discovery for understanding
crystallin evolution and lens structure are discussed.

INTRODUCTION

Crystallins are defined as major water-soluble
structural proteins of the vertebrate eye lens.
Research in the field of lens crystallins during the
last two decades was dominated by concepts of (1)
their strict tissue-specificity i.e. occurrence only in
the lens and (2) their universality i.e. lack (with
one exception of delta-crystallins) of any major
qualitative differences between crystallin sets pres¬
ent in lenses of various vertebrate taxons.

Experimental data, which became available dur¬
ing the last few years necessitate partial revision of
these concepts. The data demonstrated the exist¬
ence of taxon-specific crystallins and pointed out
relatedness between these and various cell en¬
zymes.

In this review I shall try to summarize the
available evidence about taxon-specific crystallins
and discuss how this evidence affects our views on
evolution of crystallins and structure of the lens.

“STANDARD” TAXON-UNSPECIFIC
CRYSTALLINS

Historically, lens crystallins were studied mainly
in mammals and birds [1-3]. Mammalian crystal¬
lins belong to two major protein families: the
family of alpha-crystallins and the superfamily of

Received April 12, 1988

beta/gamma crystallins. Alpha- and beta/gamma
families of crystallins (which will be referred to as
standard crystallins) do not show any sequence
similarity with one another. Within the beta /
gamma superfamily the sequence similarity be¬
tween beta- and gamma-crystallins amounts to
about 30%.

In birds and reptiles the beta/gamma superfami¬
ly is incomplete: gamma-crystallins are absent and
another protein family - delta-crystallins - is pres¬
ent alongside alpha- and beta-crystallin families.
The delta-crystallins were the first discovered case
of taxon-specific crystallins.

TAXON-SPECIFIC CRYSTALLINS

Characterization of lens crystallins in the past
was performed by gel-electrophoresis protein se¬
quencing and immunochemistry. More recently
these approaches were complemented by methods
of recombinant DNA and DNA-sequencing.
These studies have been reviewed [4, 5]. As data
accumulated, more species were analyzed in terms
of their lens crystallin composition and more pro¬
teins meeting the operational definition of crystal¬
lins have been found and characterized. Some¬
times they were found only in one or a few genera
of a given systematic group. One of such taxon-
specific crystallins - rho-crystallin - has been
found and characterized in our laboratory in the
frog Rana temporaria [6].

728

G. G. Gause

Delta-cry stallins

There is abundant literature about delta-
crystallins, their genes and expression [5, 7]. For a
long time delta-crystallin was “just another crystal-
lin”, but occurring only in birds and reptiles. An
important event was the discovery of structural
similarity between delta-crystallin and arginino-
succinate lyase [8]. The homology amounted to
about 60% in the comparison between chicken
delta-crystallin and human arginino-succinate
lyase. There are no sequence data about avian
arginino-succinate lyase. It remains to be seen
whether chicken arginino-succinate lyase is coded
by the gene distinct from that coding for delta-
crystallin or there is one gene coding for both
products.

The case of delta-crystallin was the first example
of the recruitment (the term of Wistow and Piati-
gorsky) of a gene coding for an enzyme for the role
of a gene of the crystallin ensemble.

Structural studies of the delta-crystallin gene
and experiments on its expression in various sys¬
tems resulted in the identification of its promoter
[9], which showed no particular expression spec¬
ificity in the eye lens. When recombinant DNA
constructs containing delta-crystallin gene se¬
quences were introduced into cell cultures, the
constructs were expressed effectively in various
cell types, but the expression specificity in the lens
was conferred by an enhancer located within the
third intron of the delta-crystallin gene . This
enhancer was crucial for the lens-specific ex¬
pression.

This finding is very important for the under¬
standing of the evolutionary emergence of the
delta-crystallin gene. It can be proposed that the
transposition of the genetic element containing
lens-specific enhancer into the gene of arginino-
succinate lyase resulted in the appearance of its
ability to lens-specific expression and in this way
arginino-succinate lyase became lens-specific pro¬
tein. It would be interesting to perform a search
for characteristic sequences of mobile transposable
elements in the vicinity of the lens-specific enhan¬
cer. As for possible functional consequences of
such hyperexpression of the arginino-succinate
lyase gene in the lens, nothing definite can be said

at present.

Rho-crystallin of the frog Rana temporaria

Probably this protein was noticed for the first
time in lens extracts of Rana pipiens and described
as an unidentified crystallin. It was characterized
by sequencing of a recombinant cDNA clone
selected by hybridization-translation from the li¬
brary of cDNA corresponding to total lens poly(A)
RNA of Rana temporaria [10]. Rho-crystallin
(originally called by us epsilon-crystallin) is a pro¬
tein with molecular mass of about 35 kDa account¬
ing for about 5-10% of the total frog lens protein
as shown by gel-electrophoresis comparisons and
blot hybridizations between cloned rho-crystallin
cDNA and total lens poly(A) RNA. We do not
have structural information about 25% of the
rho-crystallin sequence near the N-terminus, since
this part was not present in the clone that we have
characterized. We also do not know, whether the
whole of the 35 kDa band in lens extracts of Rana
temporaria is accounted for by rho-crystallin or this
band also contains unrelated proteins with similar
molecular mass. mRNA complementary to rho-
crystallin-specific cDNA is detected both in de¬
veloping embryos of the frog and in RNA prepara¬
tions from adult animals, thus providing indica¬
tions that this gene is expressed effectively at
different stages of ontogeny [11].

Recently two chromosomal clones containing
coding sequences of the rho-crystallin gene (or
pseudogene) have been isolated by us from the
chromosomal library of Rana temporaria. Further
characterization of this clone is now in progress.

At the time when the sequence of rho-crystallin
was deposited with gene and protein sequence
databases, no homologues of this protein was
found. Later it was proposed that by structural
criteria rho-crystallin belongs to the superfamily of
aldose/aldehyde reductases [12]. The homology
between rho-crystallin and human aldose reduc¬
tase amounted to a greater degree. There is
evidence that rho-crystallin can not have the en¬
zymatic activity of aldose reductase, since the
residues of the enzyme active center important for
the activity are not conserved in its molecule [12].

Recently Watanabe et al. [13] reported that
rho-crystallin of the frog (they use the name epsi-

Taxon-specific Crystallins

729

lon-crystallin, which we changed for rho-crystallin)
has a 77% similarity (identical position + conserva¬
tive substitutions) with prostaglandin F synthase
from bovine lung tissue. It thus appears that
rho-crystallin may indeed be the functional pros¬
taglandin F synthase.

On the basis of these data, it is highly probable
that the gene coding for prostaglandin F synthase
or a related reductase was altered in evolution in
such a way that it acquired capacity for high
expression in the eye lens and the product of this
gene became a member of the crystallin ensemble.
We do not know at present, whether this was
associated with gene duplication and whether the
enzymatic activity was retained in the case of
rho-crystallin or it was lost. Since prostaglandin F
synthase is an enzyme with broad tissue specificity,
I assume that its promoter also has a broad spec¬
ificity and the preferential lens-specific expression
of this gene was achieved in evolution by a trans¬
position of the lens-specific enhancer into its vicin¬
ity. In other words the events in the hypothetical
evolutionary history of the rho-crystallin gene may
well turn out to be similar to the case of delta-
crystallin.

Epsilon-cry stallin

Lenses of some aquatic birds as well as of
crocodiles were found to contain epsilon-crystallin
a major protein accounting for about 20% of the
total lens soluble protein and having molecular
weitht of about 40 kDa [14].

Recently it has been demonstrated that the duck
epsilon-crystallin was highly homologous to lactate
dehydrogenase isozyme B4 (94% similarity over
the length equal to 52% of the total LDH mole¬
cule). Probably epsilon-crystallin and LDH-B are
coded by one gene [15]. The LDH activity in the
duck lens extracts is 180 times higher than in chick
lens extracts, where no epsilon-crystallin is present
and the LDH activity of the purified epsilon-
crystallin amounts to about 2/3 of the purified
LDH-B4, the difference being ascribed to epsilon-
crystallin aging in the lens. An interesting con¬
sequence of the recruitment of LDH-B for the
role of crystallin is that it is now subjected to
selection pressure both as an enzyme and as a lens
protein.

In this case again the plausible explanation of
the high expression of LDH-B4 in the lens would
be the migration into the vicinity of the LDH-B
gene of a lens-specific enhancer, which stimulated
its lens-specific expression.

Tau-cry stallin

Tau-crystallin is a major protein component in
lenses of some reptiles, birds and lampreys [16] .
Its subunit molecular mass equals 48 kDa. Se¬
quence comparisons have demonstrated that tau-
crystallin has strong similarity with sequences of
human and yeast enolase [8]. This is another case
of recruitment of a gene coding for an enzyme to a
role of the crystallin gene.

SOME HYPOTHETICAL CONCLUSIONS

Transposon containing the lens-specific enhancer?

The above examples appear to point out a
general trend operating in lens evolution, i.e. that
taxon-specific crystallins originated from cell
housekeeping enzymes by a mechanism which
involved the exaggerated expression of the corre¬
sponding genes in lens cells. This exaggerated
expression was most probably due to the integra¬
tion of a genetic element or transposon containing
lens-specific enhancer into corresponding ancestral
gene. As a result protein coded by such gene
might become lens-specific. Whether this recruit¬
ment of housekeeping enzyme proteins for the role
of crytallin followed some rules other than the
integration of the lens-specific enhancer into forti-
tuous chromosomal sites remains to be seen. The
lens-specific enhancer was identified in the delta-
crystallin gene [9].

There is a question of the evolutionary origin of
the lens-specific enhancer such as one found in the
delta-crystallin gene: where did it come from. At
the moment this question remains without answer.
One possibility is that the lens-specific enhancer(s)
originates from the genes of alpha- or beta/gamma
crystallins. A tissue specific enhancer was de¬
scribed in the gene coding for alpha-crystallin [17].
The alternative sources remain obscure.

730

G. G. Gause

Function and adaptive value

There is interesting problem of requirements if
any that would allow a housekeeping enzyme to
survive as lens crystallin. It appears that the
recruitment for the role of crystallin is not related
with the enzyme activity of the protein that was
recruited. Indeed the survival of the enzyme
activity was observed only in the case of epsilon-
crystallin i.e. the LDH, whereas in cases of other
taxon-specific crystallins no enzyme activity was
reliably demonstrated. Furthermore, the substrate
levels in a metabolically inert lens are low and do
not require any high amounts of the corresponding
enzymes. I propose that the emergence of taxon-
specific crystallins is due predominantly to the
intragenomic process of transposition, creating
within the chromosome constructs, in which the
lens-specific expression of some genes may be
activated presumably by a lens-specific enhancer.
This process is rare, different genes may be acti¬
vated to lens specific transcription in various tax¬
ons and sometimes these evolutionary created
constructs may survive in evolutionary timescale.

This brings us to a question of the adaptive value
if any of the taxon-specific crystallins. The only
plausible hypothesis advanced for the case of epsi-
lon-crystallin [15] was that the possible adaptive
value of its high abundance in the lens may be
associated with the ability of LDH-B to filter-off
short range UV-light due to NADH complexed
with it and having absorption peak at 340 nm. The
resulting light-filter protects retina from the UV-
induced damage. In addition, LDH-B is thermo¬
dynamically stable and “suitable as lens protein in
terms of thermodynamic stability” [15]. The
“light-filter” hypothesis may possibly be invoked
also for rho-crystallin if it retained NADH binding
capacity since the active reductase (prostaglandin
F synthase) also uses NADH as a cofactor, where¬
as this idea does not appear to work for enolase.

Taxon-specific crystallins and lens structure

The very existence of taxon-specific crystallins
demonstrates that the optically transparent “stand¬
ard” crystallin ensemble (i.e. alpha- and beta /
gamma-crystallin families packed in lens fibers) is
tolerant to the incorporation of alien protein com¬

ponents i.e. taxon-specific crystallins. This is in
agreement with the idea that crystallin arrange¬
ment in the lens depends on short-range interac¬
tions of protein molecules [18], and these interac¬
tions do not necessarily imply exact and invariant
fitting of different crystallins to one another. The
evolutionary diversity of crystallin ensembles as
exemplified by the existence of taxon-specific crys¬
tallins depends on this “principle of tolerance” in
the organization of lens crystallin ensemble.

Are taxon-specific organ-specific proteins limited to
the lens?

It may happen that taxon-specific crystallins are
just one example of a more general phenomenon -
existence of taxon-specific variation in organ-
specific proteins. Perhaps taxon-specific crystallins
were discovered simply because the protein spec¬
trum of the eye lens is relatively simple and the
detection of novel protein species is relatively
straightforward. We can not rule out that the
process similar to the evolutionary generation of
new “lens-specific” proteins will be found for other
organs of animals as well. “Taxon-specific” organ-
specific proteins can remain undetected simply
because of the heterogeneity of the protein spec¬
trum in most organs and consequent difficulty of
identifying new proteins. Hypothetical taxon-
specific organ-specific proteins, if they do indeed
exist, would be potentially responsible for the
generation of biochemical diversity in a given
organ in evolution and would give potentially
promising material for adaptations.

CONCLUSION

Discovery of taxon-specific crystallins has
shaken two major dogmas in the crystallin re¬
search. Now we see that (1) at least some crystal¬
lins (e.g. epsilon-crystallin of the duck) are identi¬
cal to housekeeping enzymes and therefore are not
lens-specific; and (2) the crystallin ensembles of
vertebrates show far greater qualitative diversity in
evolution than thought before. The latter cir¬
cumstance is a reflection of the tolerance of the
“standard” i.e. alpha/beta/gamma ensemble to the
incorporation of new molecular species, which
thus do not destroy such important feature of the

Taxon-specific Crystallins

731

ensemble as its transparence. The discovery of
taxon-specific crystallins points out to one (so far
hypothetical) mechanism for the generation of
lens-specific proteins in evolution. In our opinion
the taxon-specific crystallins are evolutionary
young crystallins. It remains to be seen if “stand¬
ard” alpha- and beta/gamma-crystallins also orig¬
inated in this way.

REFERENCES

1 Harding, J. J. and Dilley, K. J. (1976) Exp. Eye
Res., 22.

2 Bloemendal, H. (1981) In “Molecular and Cellular
Biology of the Eye Lens”. Ed. by H. Bloemendal,
John Wiley and Sons, New York, pp. 1-47.

3 Bloemendal, H. (1982) CRC Crit. Rev. Biochem.,
12.

4 Gause, G. G. and Tomarev, S. I. (1987) In “Soviet
Scientific Rev. Physiol, and General Biology”. Gor¬
don and Breach.

5 Piatigorsky, J. (1987) Invest. Ophthalmol. Visual
Sci., 28: 9-28.

6 Gause, G. G., Tomarev, S. I., Zinovieva, R. D.,
Arutyunyan, K. G. and Dolgilevich, S. M. (1986)
In “The Lens: Transparency and Cataract”. Ed. by
G. Duncan, EURAGE, pp. 171-179.

7 Yasuda, K. and Okada, T. S. (1986) Oxford Sur¬
veys on Eukaryotic Genes, 3: 183-209.

8 Wistow, G. and Piatigorsky, J. (1987) Science, 235:
1554-1556.

9 Borras,T., Nickerson, J. M., Chepelinsky, A. B.
and Piatigorsky, J. (1985) EMBO J., 4: 445-452.

10 Hayashi,S., Goto, K., Okada, T. S. and Kondoh,
H. (1987) Genes Dev., 1: 818-828.

11 Zinovieva, R. D., Tomarev, S. I., Gorgolyuk, N.
A. and Gause, G. G. (1987) Doklady AN. SSSR.,
294: 722-725.

12 Carper, D., Nishimura, C., Shinohara, T., Dietz-
chold, B., Wistow, G., Craft, C., Kador, P. and
Kinoshita, J. H. (1987) FEBS Lett., 220: 209-213.

13 Watanabe, K., Fuji, Y., Nakayama, K., Ohkubo,
H., Kuramitsu, S., Kagamiyama, H., Nakanishi, S.
and Hayashi, O. (1987) Proc. Natl. Acad. Sci.,
USA, 84.

14 Stapel, S. O., Zweers, A., Dodemont, H. J., Kan,
J. U. and de Jong, W. W. (1985) Eur. J. Biochem.,
147: 129-136.

15 Wistow, G. J., Mulders, J. W. M. and de Jong, W.
W. (1987) Nature, 326: 622-624.

16 Williams, L. A., Ding, L., Horwitz, J. and Piati¬
gorsky, J. (1985) Exp. Eye Res., 40: 741-749.

17 Okazaki, K., Yasuda, K., Kondoh, H. and Okada,
T. S. (1985) EMBO J., 4: 2583-2595.

18 Delaye, M. and Tardieu, A. (1983) Nature, 302:
415-417.

ZOOLOGICAL SCIENCE 5: 733-742 (1988)

© 1988 Zoological Society of Japan

REVIEW

The Mechanism and Physiological Function of Electrical
Changes during Fertilization of Sea Urchin Gametes

Hideyo Kuroda, Shuichi Obata* 1, Kazuhisa Takemoto,
Motohisa Ishiguro and Hidemi Sato

Sugashima Marine Biological Laboratory, Faculty of Science,

Nagoya University, Sugashima-cho , Toba-shi,

Mie-ken 517, Japan

INTRODUCTION

Following the pioneering studies by Tyler et al.
(1956) on starfish eggs [1] and by Hiramoto (1959)
on sea urchin eggs [2], a transient change in
membrane potential associated with fertilization
was observed in eggs of several different species
(reviews in [3-8]). This change in membrane
potential has been termed the fertilization poten¬
tials or activation potentials. In this review, we
will use the term the “fertilization potential”.

Does this membrane potential change serve
some physiological function? Jaffe [9] proposed
the hypothesis of electrical poly-spermy block
which state that the fertilization potential reduced
the probability of additional sperm penetration
into the egg. While this hypothesis has been
supported by observation on the fertilization of
other animals (reviewed in [4, 6, 10, 11]), it has
also suffered criticism (see [12, 13]). Much of the
criticism is due to the differing opinions concerning
which components comprise a fertilization poten¬
tial under physiological conditions.

Another hypothesis for the function of fertiliza¬
tion potential has been proposed by E. L. Cham¬
bers and his associates [14, 15]. They suggest that
membrane depolarization is required for the incor¬

Received April 27, 1988

1 Present address: Department of Anatomy, Nagoya
University School of Medicine, Tsurumai-cho, Showa-
ku, Nagoya 466, Japan.

poration of a sperm into an egg. At least two
components of fertilization potential are important
in this depolarization process.

Although the fertilization potential is known to
consist of several components (reviewed in [7]),
much experimental data from studies of the ionic
and triggering mechanisms of each components
has been obtained. In this review, we will focus on
discussing three components of the fertilization
potential; step depolarization, component A, and
component B. We will discuss ways in which these
three components might be related, and also what
physiological functions they may serve during the
fertilization potential.

I. FERTILIZATION POTENTIALS OF
SEA URCHIN EGGS

The microelectrode recordings of fertilization
potentials in Pseudocentrotus depressus ( A , B )
Hemicentrotus pulcherrimus (C) and Clypeaster
japonicus ( D ) eggs are show in Figure 1. As
illustrated in A, C and D , the resting potentials
prior to fertilization were around —70 mV, and
there were no significant differences among spe¬
cies. The fertilization potential of Litechinus
variegatus eggs (an American species) was also
shown to be similar [16]. The early phase of the
fertilization potential consisted of two main tran¬
sient components, component A and B. The first
component (denoted by A in Fig. 1 or by segment
(a) in [16]) exhibited a steep rising phase with an

734

H. Kuroda, S. Obata et al.

Fig. 1. Typical fertilization potentials in Pseudocentrotus depressus (A, B), Hemicentrotus pulcherrimus (C) and
Clypeaster japonicus ( D ) eggs. Except in B , the early phase of fertilization potential consisted of two components,
A and B. FM marks the beginning of fertilization membrane elevation, and sp represents the addition of sperm
suspension to an experimental chamber.

overshoot. The second component (B in Fig. 1 or
segment (b) in [16]) was formed during the falling
phase of the first component. In C. japonicus eggs,
the second component was not always obvious,
and appeared as a “shoulder” on chart recordings
(Fig. ID). Figure IB shows the fertilization poten¬
tial of an egg which had a small negative resting
potential (between —10 and —20 mV). In this
case, the first component was never elicited.

At present we wish to emphasize the critical
importance of the relationship between the magni¬
tude of the resting potential and fertilization
potential which follows, as shown in Figure 1 A and
IB.

II. THE TRUE RESTING POTENTIAL
OF UNFERTILIZED EGGS

Using glass microelectrodes, the resting poten¬
tial of unfertilized sea urchin eggs has been ex¬
amined by a number of researchers. At first, only
small negative potential of about — 10 mV was

recorded [2, 17-20]. However, in subsequent
studies, several groups successfully recorded large
negative potentials of about —70 mV [21, 22]. The
potential which immediately followed insertion of
the microelectrode was usually between —5 and
— 15 mV. In a number of cases, the amplitude of
the potential increased with time and eventually
reached a potential of approximately —60 to —80
mV. Since the increased potential is associated
with an increase in input resistance [16, 23], the
small negative potential has generally been con¬
sidered to be an artifact caused by electrical leak¬
age around the microelectrode surface (reviewed
in [4-7, 16, 21, 23]).

Other methods used to estimate membrane
potential have also supported the view that unfer¬
tilized eggs have a large negative membrane poten¬
tial. The fluxes of Na+, K+, CP and Ca2+ [23] or
Na+ and K+ [16] through the membrane of unfer¬
tilized sea urchin eggs were measured using
radioactive tracers. The membrane potential was
calculated from the flux data by means of the

Fertilization Potential of Sea Urchin Gametes

735

constant field equation. The calculated values
were —70 mV [23] and —78 mV [16], respectively.
In the same two studies, membrane potentials
were also measured by microelectrodes. The meas¬
ured values were —75 ±3 mV [23] and —76 + 0.8
mV [16], and were in close agreement with the
results obtained using radioactive tracers. By
another nondamaging but indirect method, action
current associated with fertilization was recorded
extracellularly [24], By simultaneous intracellular
recording, the action current was shown to corres¬
pond to the rising phase of the regenerative action
potential. The occurrence of such an action cur¬
rent at the time of fertilization indicates that the
resting potential of an unimpaled egg must be
more negative than —50 mV, which is the
threshold level necessary to induce an action
potential (a detailed discussion of action potential
threshold will follow).

Dale and De Santis [25], however, suggest that
the small negative resting potential is not an arti¬
fact, but reflects the true resting potential. Using
Paracentrotus lividus and Spaerechinus granulans,
they showed that the eggs from large orange-red
ovaries, which were considered to be mature, had
resting potentials between —10 and —30 mV (will
be denoted by “ — 10 mV”, hereafter). On the
other hand, eggs from small dark-brown ovaries,
which were defined as suboptimal, had a large
negative resting potential between —60 and —90
mV (“ — 70 mV”). Furthermore, these suboptimal
eggs had low fertilization rates. Both mature and
suboptimal eggs had a high specific membrane
resistance. Thus, the small negative potential did
not appear to be an artifact caused by an electrical
leak at the microelectrode surface. They con¬
cluded that (1) both the potential levels of “ — 10
mV” and “—70 mV” were measured correctly,
and (2) the difference between the values reflected
the different degrees of egg maturation.

The results of our experiments using four dif¬
ferent species of Japanese sea urchins:
Pseudocentrotus depressus, Hemicentrotus pulcher-
rimus, Anthocidaris crassispina and Clypeaster
japonicus do not support the conclusion of Dale
and De Santis. The resting potential of unfertil¬
ized eggs of all four species was “—70 mV”. To
test egg maturity, we mixed eggs and sperms from

batches used in electrophysiological measure¬
ments, and determined the rate of fertilization and
the synchrony of the first cleavage. If more than
5% of these eggs failed to complete the first
cleavage, we disregarded our recordings. Our
results suggest that in these four species the true
resting potential of eggs in good condition is “ — 70
mV”. However, this is not to say that all record¬
ings of “ — 10 mV” are an artifact due to electrical
leakage. Figure 2 shows pen-recordings measured
during microelectrode impalement into unfertil¬
ized eggs of Anthocidaris crassispina. In most
cases the potential dropped quickly and attained a
resting potential of “—70 mV” (Fig. 24), although
some eggs maintained a potential of “ — 10 mV”
for 10 min or longer. However, when eggs were
depleted external sodium by perfusing them with
Na-free artificial sea water, the egg membrane
hyperpolarized from “ — 10 mV” to “ — 70 mV”
(Fig. 2 B). This result suggests that Na+ per¬
meability participates in maintaining the small
resting potential of “ — 10 mV”. Feasible ion chan-

25

0

>

E

>

-50

A

■75 L

Fig. 2. Pen-recordings during microelectrode impale¬
ment into unfertilized eggs of Anthocidaris crassispi¬
na. A microelectrode was bringed into contact with
an egg surface at arrowhead, and was applied an
oscillating current to promote microelectrode im¬
palement at arrow. A, After a noisy period of 1.3
min, the potential dropped quickly and attained a
stable level of -75 mV. B, Although a small nega¬
tive potential was maintained for 5.6 min, the poten¬
tial dropped suddenly by the perfusion of sodium-
free artificial sea water (Na-free ASW).

736

H. Kuroda, S. Obata et al.

nels which Na+ could permeate are: (a) an electri¬
cally inexcitable and relatively non-specific chan¬
nel, (b) an electrically excitable and long-lasting
channel, and (c) an intracellular Ca2+ -activated
channel [26].

The existence of the channel like (a) would
strongly support the hypothesis of a small negative
resting potential proposed by Dale and De Santis
[25]. However, it is rather difficult to determine if
such a channel exists because a slight leak around
the microelectrode surface is also permeable to
any kinds of ions. Non-damaging methods, e. g.
patch clamp experiments should solve this prob¬
lem, although none have yet been successful.

We will provide details concerning voltage-gated
channels in Section IV. Briefly at least two kinds
of voltage-gated Ca2+ channels are thought to
exist, although no voltage-gated Na+ channel has
been identified in the membrane of unfertilized sea
urchin eggs. However, it is possible that a voltage¬
gated Na+ channel does exist, or that Na+ passes
through a voltage-gated Ca2+ channel. Such a
channel could be activated by a transient mem¬
brane depolarization ocurring as the result of
microelectrode impalement and may thus cause
the observed small negative potential. Upon mi¬
croelectrode impalement, this small negative
potential of “ — 10 mV” occasionally continues for
10 min or longer before the membrane is repola¬
rized to stable potential of “—70 mV”. However,
most voltage-gated channels close soon after their
activation, and thus, do not elicit such a long-
lasting depolarization. If a voltage-gated, Na+-
permeable channel is involved in maintaining the
resting potential of “ — 10 mV”, it must differ from
the channels which have so far been identified.

A Na+ channel which is activated by an increase
in intracellular Ca2+ concentration ([Ca2+]j) has
not been identified in sea urchin eggs, but has been
observed in starfish oocytes [27]. The resting level
of [Ca2+]j in unfertilized egg was estimated to be
around 10~7M [28-30]. The Ca2+ concentration
of standard artificial sea water is 10-2 M. Since
[Ca2+]j is so lower than the external concentration,
we suggest that egg impalement by a microelec¬
trode is the probable cause of the slight and
transient increase in [Ca2+]i. This increase in
[Ca2+]j may activate Ca2+ -activated Na+ channels

and induce the small negative membrane potential
which is often observed after a microelectrode is
inserted into an egg. The small negative potential
may persist until [Ca2+]j decreases to resting level
by the Ca2+ -pumping systems of organelles and/or
surface membranes.

In conclusion, we suggest that the small negative
resting potential does not reflect the true resting
potential, but rather that it is an artifact induced by
either (1) the electrical leakage around the mi¬
croelectrode surface, or (2) the opening of Ca2+f
-activated Na+ channels.

III. STEP DEPOLARIZATION AS THE INI¬
TIAL ELECTRICAL CHANGE

In the first section of this review, we described
the existence of two depolarizing components
(component A and component B) of the fertiliza¬
tion potential which are thought to serve important
physiological functions. However, component A is
not thought to be initial electrical change which
occurs upon fertilization. Figure 3 A is a typical
recording of the fertilization potential of a P.
depressus egg which has a resting potential of
about —80 mV. Components A and B are clearly
identified although another component which pre¬
cedes component A and appears to be essential for
initiating component A is not observed. High
speed recording (Fig. 3 B) of the same electrical
changes in Figure 3 A showed the existence of
gradual depolarizations (indicated by an arrow)
from —80 mV to —50 mV. At —50 mV, the
threshold level of voltage-activated channels, the
component A is initiated and membrane potential
rises steeply to — 10 mV. Eggs with small resting
potentials did not elicit component A, although
they clearly showed gradual depolarization fol¬
lowed by a phase of long plateau (Fig. 3D). This
depolarization appeared as a step-like change in
low-speed recordings (Fig. 3C). We will designate
such an electrical change as “step depolarization”
hereafter. Step depolarization was originally
observed in “ — 10 mV” eggs. Uehara and Kato
[20] observed the step depolarization during the
fertilization potential of Hemicentrotus pulcherri-
mus eggs, and regarded it as the “fertilization-
wave” which had previously been proposed by

Fertilization Potential of Sea Urchin Gametes

737

Fig. 3. Step depolarizations in Pseudocentrotus depressus eggs. A, The fertilization potential of an egg which had a
large negative resting potential of -80 mV. B, High-speed recording of the early phase of the same fertilization
potential as A. A step depolarization (arrow) from -80 to -50 mV was distinguished from the sharp rising limb of
component A from -50 mV to -10 mV. C, Since component A was not elicited in an egg having a small negative
potential, step depolarization was clearly shown even in a low-speed recording (arrow). D. High-speed recording
of the same fertilization potential as C.

Sugiyama [31]. Ito and Yoshioka [32] showed that
step depolarization occurred concomitantly with
an abrupt decrease in input resistance, and coin¬
cided temporally with the collision between an egg
and a sperm at fertilization. Such a decrease in
input resistance and, furthermore, an increase in
voltage noise during step depolarization were also
observed by Dale et al. [33]. These two groups
have suggested that sperm-egg interaction opened
some channel in the egg membrane and elicited
step depolarization. Dale et al. [33] indirectly
measured the elementary conductance of this
channel and estimated it to be more than 33 pS, a
value similar to conductance value for drug-
induced nonspecific channels. The reversal poten¬
tial of the step depolarization was near 0 mV ([33],
our unpublished data). The amplitude depends
upon the external concentrations of both Na+ and
K+ (our unpublished data). These results suggest
that the channel involved during the step depolar¬

ization has low specificity for these ion species.

Now the question arises as to what is the phys¬
iological significance of the initial electrical change
or step depolarization. Electrical stimuli can elicit
regenerative action potentials in unfertilized egg
membranes, and the threshold level of the action
potential is about —50 mV [16, 23, 24], The action
potential observed on chart recording is similar to,
and has the same threshold level as, component A
of fertilization potentials [16, 23]. Therefore,
component A is most likely the regenerative action
potential which is activated by step depolarization.
Conversely speaking, step depolarization is likely
the trigger which induces component A.

The next question of major interest concerns the
mechanisms generating step depolarizations. At
present, we cannot exclude the possibility for a
mechanism which does not depend on the gating of
an ion channel (e.g. the enhancement of an elec-
trogenic pump). However, noise analysis of the

738

H. Kuroda, S. Obata et al.

membrane potential [33] indicates a possibility for
the participation of an ion channel during step
depolarization. Possible mechanisms which may
generate step depolarization are: (a) the activation
of an egg membrane channel by chemical stimuli,
(b) the activation of an egg membrane channel by
mechanical stimuli, or (c) the activation of an egg
or a sperm membrane channel by the fusion of
sperm and egg membranes. These three mecha¬
nisms are discussed below:

(a) : A chemical substance which is either bound
to the acrosome process or secreted from the
acrosomal granule in the sperm head might acti¬
vate egg membrane channels and elicit step de¬
polarization. However, such a substance has not
been isolated from sea urchin sperm.

(b) : Mechano-receptor channels gated by the
expansion or distortion of cell membranes have
been reported for many cell types from bacteria to
mammalian species (see [35]). Such channels may
exist in egg membranes and may be activated by
the impact generated during sperm-egg collision.
In general, the amplitude of a receptor potential
depends on the intensity of the stimuli. In the case
of step depolarization, the amplitudes are constant
over several measurements. This result suggests
that the impact of sperm-egg collision is constant.
However, it is our understanding that collision
occurs with different velocities and angles of im¬
pact, which would suggest that the impact gener¬
ated during sperm-egg collision would be variable.
Presently, we cannot rule out the possibility that
step depolarization is elicited by mechanical stimu¬
li, although we find little support for this hypoth¬
esis.

(c) : As noted previously, step depolarization is
accompanied by an increase in membrane conduct¬
ance. It is not known whether conductance in¬
crease results from (1) the insertion of new chan¬
nels into the egg membrane or (2) from opening
pre-existing membrane channels which have been
closed. During sperm-egg fusion it is likely that
sperm channels would be incorporated into the
fertilized egg membrane. The fusion must cause
an increase in the total membrane area and inevit¬
ably an increase in capacitance. McCulloh and
Chambers [36] measured the change in conduct¬
ance and capacitance simultaneously during ferti¬

lization in order to distinguish between the two
mechanisms, (a) and (c). They observed corre¬
sponding changes in the conductance and capaci¬
tance, and concluded that an initial increase in
conductance occurred when sperm ion channels
were incorporated into the egg’s membrane during
sperm-egg membrane fusion. However, in a col¬
laborative paper with Longo et al. [37] in the same
year, they presented contradictory results. They
combined electrophysiological and electron mi¬
croscopic methods to correlate the electrical
changes in an egg membrane with the morpholo¬
gical changes resulting during sperm-egg interac¬
tion. Continuity between the plasma membranes
of the sperm and egg was first detected 5 sec after
the onset of the conductance increase. These
results suggest that sperm-egg fusion did not cause,
but rather followed the conductance increase, and
thus their results do not support the conclusion of
McCulloh and Chambers [36]. However, Longo et
al. did point the possibility that membrane changes
which occur during the first 5 sec following the
onset of conductance, are likely to be in a tran¬
sitional or unstable state which is not maintained,
or is reversed when fixative is applied [37]. Thus,
the fact that they did not observe sperm-egg mem¬
brane continuity in samples fixed prior to 5 sec
following conductance onset may reflect the labile
state of early membrane fusion events.

Hinkley et al. [38] devised an elegant technique
for detecting sperm-egg continuity. They pre-
loaded eggs with a DNA-specific fluorochrome and
then fertilized these eggs with unloaded sperms.
When eggs were fixed immediately (at 1 to 4 sec)
after the onset of the conductance change, none of
the sperm which had bound to the egg surface
became fluorescent. However, when the eggs were
fixed at 8-12 sec after the conductance change, a
single sperm attached to the egg surface was
viewed as a brightly fluorescent spot. They con¬
clude that cytoplasmic continuity between the
sperm and the egg was established within 4-8 sec
after the onset of the conductance increase. These
experiments strongly support the hypothesis that
sperm-egg fusion occurs after the onset of the
conductance change. However, Hinkley et al.
acknowledged the possibility that fusion may
actually occur at an earlier period.

Fertilization Potential of Sea Urchin Gametes

739

These experiments of Longo et al. [37] and
Hinkley et al. [38] do not support the hypothesis
that sperm-egg fusion triggers a conductance
change in egg membranes. In contrast, the results
obtained by McCulloh and Chambers [36] do
support this hypothesis, although the details of this
experiment were not described. At present, the
mechanism which elicits step depolarization is un¬
clear.

IV. COMPONENT A OF THE FERTILIZA¬
TION POTENTIAL

If we accept the hypothesis proposed by Jaffe [9]
and by Lynn and Chambers [14], the function of
component A seems to be clear. A sperm could
not easily incorporate into an egg having a resting
potential of “—70 mV” [14]. The step depolariza¬
tion associated with “sperm-egg interaction” acti¬
vates component A at around —50 mV. During
the rising limb of component A, the egg membrane
must pass through a range of membrane potentials
(between +17 and —25 mV in Lytechinus variega-
tus eggs) necessary to facilitate sperm incorpora¬
tion into an egg [14] and, subsequently, a range of
electrical polyspermy block to additional sperm
penetration (> + 18 mV).

What kinds of ion channels are involved in the
generation of component A? In low Na+ sea
water, the action potential which was elicited by an
electrical stimulus or component A of the fertiliza¬
tion potential, diminished not only in amplitude,
but also in duration. A 100-fold reduction in
[Ca2+]0 resulted in the suppression of component
A [16]. Therefore, the action potential must
depend upon both external Ca2+ and Na+. A
typical action potential generated either in normal
sea water ([16] and also see Fig. 2 A) or in high
Ca2+ sea water [39] had a threshold level of about
—50 mV. Once the membrane potential overcame
this threshold, it rose with a steep slope to a
characteristic notch at about 0 mV, and continued
to rise with a greatly diminished slope to a peak of
about +20 mV. These results suggest the exsist-
ence of two kinds of ionic mechanisms, one of
which has a threshold level of about —50 mV and
another with a threshold of around 0 mV. Vol¬
tage-clamp experiments [39] support the idea that

action potentials depend upon two inward cur¬
rents. One of these currents has a threshold level
of —50 mV, and is carried by Ca2 + . However, the
ionic mechanism of the other inward current which
has a threshold of 0 mV has not been well char¬
acterized in sea urchin eggs.

Membrane excitability of starfish oocytes, has
been thoroughly investigated (reviewed in [4]).
Immature starfish oocytes generate action poten¬
tials which are similar to those of sea urchin eggs.
Voltage-clamp studies [40] showed that there were
two kinds of channels refered to as channel I and
II. Channel I was activated between —55 and —50
mV and channel II was activated between —7 and
—6 mV. Many of the electrophysiological prop¬
erties of channel II were similar to those of the Ca
channel characterized in adult tissues. On the
other hand, the current through channel I was
dependent upon both external Ca2+ and Na+.
However, Na+ did not appear to pass through
channel I but did appear to modulate the duration
of the current which passed through channl I. An
interpretation of channel I was recently proposed
by Lansman [27]. In this study, the inward current
activated between —55 and —50 mV consisted of
two components; a rapidly decaying current which
flowed through Ca channels and a more slowly
decaying component which was carried by Na+.
This Na+ current was probably activated by cyto¬
plasmic Ca2+ which flowed into the egg through
channel II and another Ca channel which carries
the rapidly decaying current.

The question as to whether similar channels
concern the action potential or component A in
sea urchin eggs remains unsolved. The threshold
levels of channel I and channel II in starfish
oocytes correspond to the two different threshold
levels observed in sea urchin eggs. Thus, we
suggest that the action potential and the first
component of the fertilization potential in starfish
oocytes corresponds to the action potential and
component A in sea urchin eggs. Furthermore
three kinds of channels appear to be involved in
generating component A: a Ca channel activated
between —55 and —50 mV, a Ca channel acti¬
vated between —7 and —6 mV, and a Na channel
which is activated by intracellular Ca2+.

740

H. Kuroda, S. Obata et al.

V. COMPONENT B OF THE FERTILIZATION
POTENTIAL

As we have shown, component B of the fertiliza¬
tion potential of P. depressus eggs is formed during
the falling phase of component A [26]. The
membrane potential reached a peak value of
around +20 mV and then gradually hyperpolar-
ized. The duration of component A was around 5
sec. The period from the onset of component A to
the beginning of fertilization membrane elevation,
which was thought to be another and more perfect
mechanism of polyspermy block than electrical
polyspermy block, was 20 to 25 sec. Component B
appears to maintain a positive membrane poten¬
tial, which is essential to electrical polyspermy
block, for 15-20 sec between the end of compo¬
nent A and the beginning of fertilization mem¬
brane elevation. Thus, both component A and
component B must play important role during
electrical polyspermy block.

The generation of component B was observed to
be dependent upon the diffusion of Na+ through
the egg membrane [16, 17], However, the mem¬
brane potential at the peak of component B was
less positive than the Nernst potential for Na+
[26]. Consequently, the membrane potential at
this phase cannot be explained solely by the diffu¬
sion of Na+. The peak value was sensitive to the
external concentrations of both Na+ and K+.
These data were fitted to a theoretical line
obtained from the Goldman-Hodgkin-Katz equa¬
tion, using a ratio of PNa: Pk7 Pci=11- 1.0: 0.0
[26]. The results of this analysis suggest that the
permeability of the egg membrane to both Na+
and K+ is responsible for component B. From
voltage-clamp experiments using starfish eggs, fer¬
tilization was observed to induce an increase in
non-selective cation permeability [41].

The resting membrane potential of unfertilized
eggs is largely dependent upon K+ permeability.
What triggers the increase of permeability to both
Na+ and K+ during fertilization? The [Ca2+]j of
sea urchin eggs was observed to increase transient¬
ly at an early stage of fertilization [28-30]. It has
been reported that this increase in [Ca2+]j induces
the cortical reaction leading to the onset of ferti¬
lization membrane elevation (reviewed in [3, 7]),

which was observed early during component B of
the fertilization potential [26]. The calcium
ionophores A23187 or X-537A (Lasalocid) in¬
duced membrane depolarization as observed by
chart recordings, in a manner similar to the ferti¬
lization potential [16, 42, 43], A23187 elicited a
transient depolarization in P. depressus eggs, and
the peak value of this depolarization also de¬
pended on membrane permeability to both Na+
and K+ (our unpublished data). This suggests that
the [Ca2+]j increase is closely associated with the
generation of component B.

In cultured mammalian cells, a Ca2+ -activated
channel which does not discriminate between Na+
and K+ has been observed by single-channel cur¬
rent recording [44-47]. Such a Ca2+ activated
non-selective channel is probably activated at ferti¬
lization. On the other hand, a Ca2+ -activated Na+
conductance [48-50], and a Ca2+ -activated K+
conductance (reviewed in [50], and see [51-53]),
have also been observed. The possibility that a
Ca2+ -activated Na channel and a Ca2+ -activated
K channel open independently but concomitantly
can not be ruled out at present.

The transient increase in [Ca2+]i which occurs
during the early stages of fertilization has been
reported to be dependent upon Ca2+ release from
Ca2+-sequestering endoplasmic reticulum. Fur¬
thermore, Ca2+ release is thought to be triggered
by inositol 1, 4, 5-triphosphate (IP3) (reviewed in
[8, 13, 54]). However, the mechanism by which a
fertilizing sperm induces the hydrolysis of
polyphosphatidyl-inositol 4, 5-bisphosphate and
produces IP3 remains unclear.

Finally a step depolarization does not seem to be
the cause of component B as suggested by results
obtained from studies using either the voltage-
clamp method or the current-clamp method. In
voltage-clamp experiments in which the membrane
potential was clamped at negative potentials of less
than —20 mV, a “brief transient current” was
observed. This current corresponded to a step
depolarization observed during current-clamp ex¬
periment, and was generated without a “prolonged
current” which normally follows and corresponds
to component B [15]. Therefore, the transient
increase in [Ca2 + ]j would not be the effect of the
“sperm-egg interaction” which elicits a step de-

Fertilization Potential of Sea Urchin Gametes

741

polarization.

VI. CONCLUSION

We have shown that the fertilization potential of
sea urchin eggs consists of at least three membrane
events: step depolarization, component A, and
component B. We have reviewed studies concern¬
ing possible ionic channels involved during these
electrical events. We will now briefly summarize
the possible ways in which these membrane events
are related. Sperm-egg interaction apparently
causes two independent events during the initial
stage of fertilization. Both events have not been
well characterized. One event involves the gating
of channels which have low specificity to monova¬
lent cations, and elicits the step depolarization
from about —70 mV to —50 mV. The local cur¬
rent of step depolarization probably activates Ca
channel I, resulting in egg membrane depolariza¬
tion (from —50 mV to about 0 mV). The current
which passes through channel I is thought to
activate Ca channel II. Ca2+ which flows into an
egg through the channels I and II probably acti¬
vates Na channels and elicits membrane depolar¬
ization, which lasts for several seconds. These
three ionic mechanisms combined, are probably
involved in eliciting component A.

Another event which is thought to occur at the
early stage of fertilization is a facilitation of PI-
turnover toward the production of IP3 which, in
turn, increases [Ca2+]j. This Ca2+ increase is
probably responsible for opening channels which
are equally permeable to Na+ and K+, and thus,
results in the initiation of component B. The
physiological functions of components A and B are
not totally understood. However, component A
appears to facilitate the incorporation of a fertiliz¬
ing sperm into an egg and components A and B
both appear to be involved in electrical block to
additional sperm.

ACKNOWLEDGMENTS

We would like to thank Miss Marilyn L. Fitzgeral for
editing this manuscript. This work was supported in part
by Grant 62304062 from the Ministry of Education,
Science and Culture, Japan.

REFERENCES

1 Tyler, A., Monroy, A., Kao, C. Y. and Grundfest,
H. (1956) Biol. Bull., Ill: 153-177.

2 Hiramoto, Y. (1959) Exp. Cell Res., 16: 421-424.

3 Epel, D. (1978) Curr. Top. Dev. Biol., 12: 186-246.

4 Hagiwara, S. and Jaffe, L. A. (1979) Ann. Rev.
Biophys. Bioeng., 8: 385-416.

5 Whitaker, M. J. and Steinhardt, R. A. (1982) Q.
Rev. Biophys., 15: 593-666.

6 Shen, S. S. (1983) In “Mechanism and Control of
Animal Fertilization”. Ed. by J. F. Hartmann,
Academic Press, New York, pp. 213-267.

7 Whitaker, M. J. and Steinhardt, R. A. (1985) In
“Biology of Fertilization, Vol. 3”. Ed. by C. B.
Metz and A. Monroy, Academic Press, New York,
pp. 167-221.

8 Trimmer, J. S. and Vacquier, V. D. (1986) Ann.
Rev. Cell Biol., 2: 1-26.

9 Jaffe, L. A. (1976) Nature (London), 261: 68-71.

10 Jaffe, L. A. and Gould, M. (1985) In “Biology of
Fertilization, Vol. 3”. Ed. by C. B. Metz and A.
Monroy, Academic Press, New York, pp. 223-250.

11 Jaffe, L. A. and Cross, N. L. (1986) Ann. Rev.
Physiol., 48: 191-200.

12 Dale, B. and Monroy, A. (1981) Gamete Res., 4:
151-169.

13 Dale, B. (1987) Nature (London), 325: 762-763.

14 Lynn, J. W. and Chambers, E. L. (1984) Dev.
Biol., 102: 98-109.

15 McCulloh, D. H., Lynn, J. W. and Chambers, E. L.
(1987) Dev. Biol., 124: 177-190.

16 Chambers, E. L. and De Armendi, J. (1979) Exp.
Cell Res., 122: 203-218.

17 Steinhardt, R. A., Lundi, L. and Mazia, D. (1971)
Proc. Natl. Acad. Sci. USA, 68: 2426-2430.

18 Steinhardt, R. A., Shen, S. and Mazia, D. (1972)
Exp. Cell Res., 72: 195-203.

19 Tupper, J. T. (1973) Dev. Biol., 32: 140-154.

20 Uehara, T. and Katoh, K. (1972) Dev. Growth
Differ., 14: 175-184.

21 Ito, S. and Yoshioka, K. (1972) Exp. Cell Res., 72:
547-551.

22 Higashi, A. and Kaneko, H. (1971) Annot. Zool.
Japon., 44: 65-75.

23 Jaffe, L. A. and Robinson, K. (1978) Dev. Biol.,
62: 215-228.

24 Whitaker, M. J. and Steinhardt, R. A. (1983) Dev.
Biol., 95: 244-248.

25 Dale, B. and De Santis, A. (1981) Dev. Biol.. 85:
474-484.

26 Obata, S. and Kuroda, H. (1987) Dev. Biol., 122:
432-438.

27 Lansman, J. B. (1987) J. Physiol., 390: 397-413.

742

H. Kuroda, S. Obata et al.

28 Steinhardt, R. A., Zucker, R. and Schatten, G.
(1977) Dev. Biol., 58: 185-196.

29 Eisen, A., Kiehart, D. P., Wieland, S. J. and
Reynolds, G. T. (1984) J. Cell Biol., 99: 1647-1654.

30 Poenie,M., Alderton,J., Tsein, R. Y. and
Steinhardt, R. A. (1985) Nature (London), 315:
147-149.

31 Sugiyama, M. (1953) Biol. Bull., 104: 210-215.

32 Ito, S. and Yoshioka, K. (1973) Exp. Cell Res., 78:
191-200.

33 Dale,B., DeFelice, L. J. and Taglietti, V. (1978)
Nature (London), 275: 217-219.

34 Nuccitelli, R. and Grey, R. D. (1984) Dev. Biol,
103: 1-17.

35 Kullberg, R. (1987) Trends Neurosci., 10: 387-388.

36 McCulloh, D. H. and Chambers, E. L. (1986) J.
Gen. Physiol., 80: 38a.

37 Longo, F. J., Lynn, J. W., McCulloh, D. H. and
Chambers, E. L. (1986) Dev. Biol., 118: 155-166.

38 Hinkley, R. E., Wright, B. D. and Lynn, J. W.
(1986) Dev. Biol., 118: 148-154.

39 Okamoto, H., Takahashi, K. and Yamashita, N.
(1977) J. Physiol., 267: 465-495.

40 Hagiwara, S., Ozawa, S. and Sand, O. (1975) J.
Gen. Physiol., 65: 617-644.

41 Lansman, J. B. (1983) J. Physiol., 345: 353-372.

42 Hulser, D. and Schatten, G. (1982) Gamete Res.,

5: 363-377.

Steinhardt, R. A. and Epel, D. (1974) Proc. Natl.
Acad. Sci. USA, 71: 1915-1919.

Coluquhoun, D., Neher, E., Reuter, H. and
Stevens, C. F. (1981) Nature (London), 294: 752-
754.

Yellen, G. (1982) Nature (London), 296: 357-359.
Maruyama, Y. and Petersen, O. H. (1982) Nature
(London), 299: 159-161.

Maty, A., Tan, Y. P. and Trautmann, A. (1984) J.
Physiol., 357: 293-325.

Takahashi, K. and Yoshii, M. (1978) J. Physiol.,
279: 519-549.

Jaffe, L. A., Kado, R. T. and Kline, D. (1986) J.
Physiol., 381: 263-278.

Petersen, O. H. and Maruyama, Y. (1984) Nature
(London), 307: 693-696.

Miyazaki, S. and Igusa, Y. (1982) Proc. Natl. Acad.
Sci. USA, 79: 931-935.

Igusa, Y. and Miyazaki, S. (1983) J. Physiol., 340:
611-632.

Igusa, Y. and Miyazaki, S. (1986) J. Physiol., 377:
193-205.

Whitaker, M. J. and Steinhardt, R. A. (1985) In
“Biology of Fertilization, Vol. 3”. Ed. by C. B.
Metz and A. Monroy, Academic Press, New York,
pp. 453-455.

43

44

45

46

47

48

49

50

51

52

53

54

ZOOLOGICAL SCIENCE 5: 743-757 (1988)

© 1988 Zoological Society of Japan

Recovery from Light-induced Sensitivity Loss in the Eye of the
Crustacean My sis relicta in Relation to Temperature:
a Study of ERG-determined V/log I Relationships
and Morphology at 4°C and 14°C

Magnus Lindstrom, Heimo L. Nilsson1 and V. Benno Meyer-Rochow2

Tvarminne Zoological Station, University of Helsinki, SF-10900 Hanko, Finland,

1 Department of Zoology, University of Lund, Helgonavagen 3, S-223
59 Lund, Sweden, and 2Department of Biological Sciences, University
of Waikato, Private Bag, Hamilton, New Zealand

ABSTRACT — Specimens of Mysis relicta from Lake Paajarvi (Finland) were divided into two groups,
one kept at 4°C, the other at 14°C. Immediately following a 1 hr exposure to white light of 4000 lux,
both 4°C and 14°C animals displayed a similar loss in response amplitude, but while sensitivity recovery
in the 14°C group at first progressed more rapidly, it was overtaken by animals from the 4°C population
from about the second day post-exposure as witnessed by electrophysiological recordings that were
carried out routinely for a period of at least 100 hr post-exposure. Supported by electron micrographs of
rhabdom microvilli representing the different experimental groups, it is concluded that the initial
biochemical reconstitution of excitable visual pigment occurs faster at higher temperatures, but that
parallel microvillar instabilities and breakdown also proceed more rapidly at higher temperatures,
leading to ion leakages, sharper response rise times, and flatter than usual V/log I curves. V/log I curves
of eyes with reduced sensitivities also allowed us to conclude that the loss in sensitivity is due, at least in
part, to a net loss of rhodopsin. The fact, however, that indisputable electrophysiological responses to
light can be recorded at all in eyes with seriously distorted visual membranes, suggests that effective
rhodopsin may not be exclusively located in intact rhabdomeric microvilli. Complete recovery of visual
sensitivity to levels of the control animals was not observed either in 14°C or 4°C material, although the
latter was approaching normality approximately 100 hr post-exposure.

INTRODUCTION

Both constant light as well as constant darkness
affect the performance and organization of animal
photoreceptors [1]. In the rat, for example, con¬
stant light leads to a degeneration of visual mem¬
branes and a continuous fall in rhodopsin levels [2]
which would adequately explain the gradual loss of
absolute sensitivity observed in eyes of rats kept
under these conditions [2,3]. The visual process in
the vertebrate eye commences with a change of the
chromophore due to photon absorption and leads,
via several steps, to a destruction of the photopig¬

Accepted October 28, 1987
Received September 2, 1987

2 To whom requests of reprints should be addressd.

ment. A ‘dark’ phase is required in which protein
moiety and chromophore are reassembled; only
then renewed photo-excitation is possible [4].

In arthropods the biochemistry of the visual
process differs and at least in insects does not
normally require a dark phase for photopigment
regeneration [5]. In several species of crustaceans,
however, it was noticed that exposure to even
moderate levels of brightness could suppress visual
sensitivity for several days, if not longer [6-9].
Anatomical studies revealed that light-induced
photoreceptor damage or breakdown, once trig¬
gered, continued to proceed in the dark long after
the initial exposure to light [9-14] and at least in
Jasus edwardsii [8] and Boreomysis megalops [15]
had a direct impact on the behaviour.

The fact that electrophysiologically-recorded

744

M. Lindstrom, H. L. Nilsson and V. B. Meyer- Rochow

sensitivity levels apparently recovered before
structural integrity was re-established, led Meyer-
Rochow and Tiang [8] to suggest a simple model
(Fig. 1). This stated that while degradation and
removal of visual membrane were exceeding mem¬
brane synthesis and re-assembly and, thus, were
the reason for the degenerated appearance of the
microvilli in the electron micrographs one or more
days post-exposure, the biochemical processes of
photopigment regeneration could still operate fast
enough to actually show up as improved sensitivity
in ERG-recordings one or two days post-exposure.
For how long this process would continue and to
what extent it might be affected by ambient
temperatures, however, were open questions,
which to date had not been addressed by the
appropriate research.

Fig. 1. From histological examinations of visual mem¬
branes and electrophysiological tests of visual sensi¬
tivity in the rock lobster Jasus edwardsii [33] the
following hypothesis was formulated: The eye, prior
to illumination, is of a certain adaptational state
( = height on the ordinate). Irreversible loss of
vision ( = damage) results when either visual pig¬
ment curve (solid line) or membrane curve (dotted)
transect the abscissa. Interestingly, the black solid
line mirrors electrophysiologically-recorded sensitiv¬
ity which is higher one day post-exposure than
immediately post-exposure while rhabdom derange¬
ments are considerably more obvious one day post¬
exposure [27] than immediately post-exposure.

On the one hand, Meyer-Rochow and Tiang [16]
in their study of eyes of amphipod crustaceans
from Antarctica have indicated that temperature
alone, in the absence of light, can influence the
ultrastructural organization of visual membranes.
Barnes and Goldsmith [17] and Larrivee and Gold¬

smith [18], on the other hand, have found that in
vitro photopigment regeneration in the lobster
Homarus and the crayfish Procambarus was
temperature-dependent (just like Stieve [19] on
the basis of electrophysiological recordings from
the eye of Eupagurus bernhardus had predicted).
We decided to elucidate the combined effect of
temperature and exposure to light by monitoring
post-exposure ERG-determined absolute sensitivi¬
ties and effects on rhabdom morphology in Mysis
relicta.

Mysis relicta, the northern opossum shrimp, was
chosen because (a) it was readily available, (b) the
structural organization of the eyes of various spe¬
cies of mysid shrimps had previously been de¬
scribed by light- and electron microscopy [20-24],
(c) it was known from ERG-recordings by Lind-
strom and Nilsson [25] that at least specimens from
the dark Lake Paajarvi had photoreceptors which
reacted very sensitively to light, and (d) under
natural conditions animals of this species can ex¬
perience temperatures from near freezing point of
water to at least 15°C [26].

MATERIALS AND METHODS

Electrophysiology

Experimental animals were collected in late May
from 8°C (surface temperature) warm Lake Paa¬
jarvi at night from a depth of 50-60 m and divided
into a 4°C and a 14°C group. Each group was
maintained in thermostat-controlled aerated water
under total darkness, all necessary manipulations,
preparations etc. being carried out in infrared light
of >730 nm with the aid of a “Find-R-Scope”
infrared viewer. While control animals of both
temperature regimes remained in the dark
throughout the entire series of the experiments, at
least two days following capture, two subsets,
representing the 4°C and 14°C groups, were ex¬
posed in a large Petri dish (30 cm diameter) for 1
hr at midday to artificial white light of 4000 lux
coming from a ring-type (24 cm diameter) desk
lamp. Four such exposures, two with 4°C and two
with 14°C animals, were carried out during which
no fatalities amongst experimental animals were
encountered. Following the exposure to the light,

Temperature-dependent Visual Sensitivity Recovery

745

the animals were immediately returned to their
dark environments of their respective groups. Ex¬
posed animals were, of course, kept separately
from the non-exposed controls. Tests of post¬
exposure visual sensitivity commenced 1 hr after
the animals had been returned to the dark to
recover and continued, at the beginning, roughly
every 5 hr. From the second day onwards bigger
intervals were allowed. Care was taken to test 4°C
and 14°C animals alternatingly.

Experimental procedures closely followed those
reported previously [7]. During preparation, using
infra-red image converters mounted on a Wild-5
stereomicroscope, each animal was illuminated by
light that had passed through 2 Kodak Wratten 87
gelatin filters and sometimes a heat filter as well,
which were inserted in the ray path of white light
coming from a microscope lamp. The incident
light, perpendicular to the eye surface, was cen¬
tered around the hole through which the recording
electrode was lowered 40-50 pm into the eye. The
light spot made by the stimulating flash was large
enough to cover the entire eye. Always the same
region of the eye was aimed for when inserting the
electrode. Following preparations, which, on
average, did not take longer than 5-10 min, the
test animals were given 30 min in total darkness to
recuperate from the operation. The system for
stimulation consisted of an extended source
(Osram 6V, 15W microscope lamp, powered by a
constant voltage device) and all recordings were
made in the AC-setting.

All experimental response/stimulus data pairs
were entered into an Exzel TP2 computer in such a
way that response amplitudes were expressed in
/N and stimulus intensities in nd-units. For easier
reference to any particular neutral density one log
unit of reduced stimulus intensity was arbitrarily
divided into 120 equal nd-units, 0.5 of a log unit,
thus, becoming 60, 2 log units being 240 etc. V/log
I curves were printed by a Panasonic KX-P1091P
printer in a double logarithmic plot. The program
used was Lotus 123. Similar double logarithmic
plots were used to demonstrate, response/time and
stimulus intensity/time relationships for fixed neu¬
tral density and response amplitude, respectively.
Finally, based on the widely used formula

n = log(-V- V )/ log f

we determined the slopes (nj) of our V/log I curves
as

n =

log

Vi(Vm— V2)
V2(Vm-V1)

logCIj) — log(I2)

where Vi = response to stimulus intensity lu V2 =
response to stimulus intensity I2, and Vm = max¬
imum response at log 1 = 0.

Electron microscopy

For electron microscopy a number of eyes from
different experimental groups (Table 1) were fixed
for 3 hr in Karnovsky’s well known fixative ad¬
justed to a pH of 7.2 and an osmolarity of 400
mOsmol. Sodium-cacodylate buffer was used
throughout. An initial dissection of the eyes was
carried out in the fixative in darkness (IR-light and
IR-viewers) and after about 5 min a final dissection
into small retinal pieces was carried out in faint
white microscope light (ca. 4 min). The fixation
was then continued in darkness.

Table 1 . Number of eyes available for histolo¬
gy in the different experimental groups

Dark/Light exposure

DA

LA

LA— DA

Ambient water

temperature

+ 4°C

4

4

4

+ 14°C

4

0

4

(DA = dark-adapted, LA = white light exposure
4000 Lux for 1 hr, LA— *D A = light-exposed animals
post-dark-adapted for another 5 days before fixation
of the eyes.

Postfixation was carried out for 2 hr in 2%
Os04. Dehydration was achieved in a graded
series of alcohol and followed by blockstaining in
0.5% uranylacetate/1% phosphotungstic acid and
embedding in Epon. Sections for electron micros¬
copy were stained in uranylacetate and lead for 30
min (LKB grid strainer). Sections were examined
in a Zeiss EM10.

746

M. Lindstrom, H. L. Nilsson and V. B. Meyer-Rqchow

RESULTS

The physiological data obtained indicate that in
Mysis relicta in agreement with the crayfish Pro -
cambarus clarkii [27], but in contrast to the situa¬
tion in insects and cephalopods, where light is
required for the reconversion of photopigment
into its excitable form [5], the process of photopig¬
ment regeneration will take place in the dark. Not
unexpectedly V/log I relationships of both 4°C and
14°C control animals exhibited the same general
shape in all individuals tested and over the entire
experimental period (cf. Figs. 2-4 and 5-7). In
both cases threshold levels of 10 juV response
amplitudes correspond to stimulus intensities of
— 6 log I units or below and maximum amplitudes
of around 10,000 juV were recorded in response to
flashes of the brightest light available (log 1=0).
Absolute sensitivities were severely reduced in
animals of both temperature regimes following a 1
hr exposure to light of 4000 lux though the effect
was particularly pronounced in the 4°C group.
Interestingly, the V/log I curves of eyes with re¬
duced sensitivity in relation to the controls are not
simply shifted down the log-response axis, but also
to the right on the log I axis — a phenomenon
already noticed by Cummins and Goldsmith [6]
and interpreted as a loss of rhodopsin.

Fig. 2. Data of 4°C animals plotted on log-log axes,
showing gradual increases in sensitivity of 1 hr-22 hr
post-exposure animals in relation to 1 day control
animal.

There was a tendency of all animals tested,
irrespective of whether they came from the 4°C
population or 14°C population to exhibit some
degree of post-exposure sensitivity increase, but
the way this sensitivity increase developed and
progressed with time, definitely depended on the
temperature the test animals were kept under
during the recuperation period. When cold (4°C)

Fig. 3. Data of 4°C animals plotted on log-log axes,
showing intensity/ response curves of 37 far-72 hr
post-exposure animals in relation to 2 day control
animal.

Fig. 4. Data of 4°C animals plotted on log-log axes,
showing /intensity response curves of 100 hr- 195 hr
post-exposure animals in relation to 5 day and 14
day control animal.

Temperature-dependent Visual Sensitivity Recovery

747

light-exposed animals were dark-adapted again at
4°C, their eyes did not fully recover to the sensitiv¬
ity of the controls within 9 days, but appeared
nonetheless capable of further sensitivity improve¬
ment, given time (Figs. 2-4, 9 and 11). On the
other hand, light-exposed animals of the 14°C
group kept under that same temperature to recov¬
er, reached corresponding sensitivity levels already
after one day, but from then on their V/log I

□ C Id +1H 0 6H A 12 H X17H

Fig. 5. Data of 14°C animals plotted on log-log axes,
showing intensity /response curves of 1 hr- 17 hr
post-exposure animals in relation to 1 day control
animal.

□ C 3d + 23 H 0 24H A32H X38H

Fig. 6. Data of 14°C animals plotted on log-log axes,
showing intensity /response curves of 23 hr-38 hr
post-exposure animals in relation to 3 day control
animal.

curves became flatter and frequently peak re¬
sponse amplitudes actually fell (Figs. 5-7, 10 and
12). Falling response amplitudes in the 14°C group
were regularly linked to faster rise and fall speeds
of the response, causing the responses to become
sharper and more spike-like in appearance. For
morphological correlates, see below and Figures
13, 14, 15, 16, and 17.

Peak responses in the recovering 4°C group

Fig. 7. Data of 14°C animals plotted on log-log axes,
showing intensity /response curves of 42 hr-58 hr
post-exposure animals in relation to 5 day control
animal.

n

0.8

0.6

0.4

0.2

< o<°

o 4 t

o ox

o °

* O

♦ 4°C controls

° Kc

• 14°C controls

0.4 0.8 1.2 1.6 2.0 2.4 2.8

log (hrs)

Fig. 8. Slopes (n) of V/log I curves, usually calculated
between log I = — 1 and —3, of 4°C and 14°C ex¬
perimental and control animals.

748

M. Lindstrom, H. L. Nilsson and V. B. Meyer-Rochow

Fig. 9. Response values of 4°C animals to fixed stimulus
intensities in relation to post-exposure time plotted
on log-log axes. ND=60, ND = 120, and ND = 180
correspond to log I values of —0.5, —1, and —1.5,
respectively.

log I nd

Fig. 11. Responses of 20 fN and 60 fN and their cor¬
responding stimulus intensities in 4°C animals in
relation to post-exposure time plotted on log-log
axes. Dotted in are the controls.

Fig. 10. Response values of 14°C animals to fixed stim¬
ulus intensities in relation to post-exposure time
plotted on log-log axes. ND=60, ND = 120, and
ND = 180 correspond to log I values of —0.5, — 1,
and —1.5, respectively.

Fig. 12. Responses of 10 fN and 60 fN and their cor¬
responding stimulus intensities in 14°C animals in
relation to time plotted on log-log axes. Dotted in
are the controls.

Fig. 13. Electron micrograph of cross section through the rhabdom layer of a fully dark-adapted eye of My sis relicta
kept at 14°C. With regard to rhabdom organisation, microvillus dimensions, and degree of regularity in the visual
membranes the eyes of dark-adapted 4°C (Fig. 14) animals are no different. All rhabdoms are voluminous and
the screening pigment granules are restricted to a narrow cytoplasmic rim surrounding the former. Abbrevia¬
tions; Rh = rhabdom; SP= screening pigment. The scale is 5 /an.

Temperature-dependent Visual Sensitivity Recovery

749

750

M. Lindstrom, H. L. Nilsson and V. B. Meyer-Rochow

reached maximally 1200 pV, but in the recovering
14°C group 480 juV were not exceeded. While
maximum amplitudes in the 4°C group gradually
increased with time spent in the dark, a tendency
for the peak amplitudes to remain depressed or
actually fall while at the same time thresholds kept
“creeping” towards lower levels was regularly
observed in the 14°C animals. The average slopes
of all the V/log I curves, obtained post-exposure
along the -3 to -1 log I stretch were 0.48 + 0.15 (n
= 11) for the 14°C material and 0.56 + 0.12 (n = 14)
for the 4°C animals. The slopes of the V/log I
curves of the 14°C and 4°C controls in the corre¬
sponding region measured 0.58 + 0.08 (n = 3) and
0.56 + 0.06 (n = 4), respectively (Fig. 8).

The initial somewhat faster sensitivity recovery
can also be demonstrated when response ampli¬
tudes are plotted against time for a given neutral
density filter, i.e. fixed stimulus light intensity.
The data, once again, reveal that while the upward
trend in the 4°C animals continues (Fig. 9), the
14°C animals actually exhibit decreasing response
amplitudes with time (Fig. 10).

Another interesting and instructive way of com¬
paring the results from the 4°C and the 14°C
population of animals is to take the 50 juV response
stretch spanning from the near-threshold response
of 10 /jW to 60 juV and following the corresponding
stimulus intensities required to produce these two
responses in relation to post-exposure time (Figs.
11 and 12). While the 50 juV stretch in all the
controls of both 4°C and 14°C animals typically
corresponds to little more than half of a log unit of
stimulus intensity change (usually within the vicin¬
ity of log 1= —6 and logI=— 5) the situation in
exposed animals is quite different. In both 4°C and
14°C animals the 50 pW stretch immediately fol¬
lowing exposure to the adapting light covers

approximately 2 log units of stimulus intensities
ranging from the brightest light available at log I
= 0 to about 1/100 of this intensity.

With time, thresholds become lower and intensi¬
ties eliciting the 10 pV and 60 juV responses move
closer together: discrimination of intensity shades
improves. But while the response amplitudes in
the 4°C material generally progressed at least until
100 hr post-exposure (Fig. 11), those of the 14°C
animals showed a different trend (Fig. 12) In the
14°C animals the 10-60 p\ bars corresponding to
effective stimulus intensities became increasingly
longer until after 58 hr post-exposure the 50 pV
stretch covered 4 1/2 log units of stimulus intensi¬
ties (Fig. 12). Remarkably enough, thresholds,
however, kept on getting lower so that increasingly
dimmer flashes of light elicited responses. Exactly
the same trend showed up when instead of 10-60
juV 10 and 100 pV were chosen to monitor corre¬
sponding stimulus intensities along the post¬
exposure time axis. The controls, meanwhile,
irrespective of the 4°C or 14°C temperature, re¬
tained their high sensitivity and short 10-60 //V (or
10-100 juV) bars.

Electron microscopy

The morphological changes which are reported
in this section are restricted to the main rhabdoms
of My sis relicta. This species also contains a distal
rhabdom (as well as a so-called epirhabdom) simi¬
lar to the one reported in Praunus flexuosus [20] .

Animals kept in darkness at both +4°C and at
+ 14°C exhibited well preserved visual receptor
cells with the photoreceptive membranes, the mi¬
crovilli, being regularly arranged (Figs. 13 and 14).

Eyes from animals kept at +4°C and exposed to
white light (4000 lux) for 1 hr and then fixed im¬
mediately after exposure to light showed normal

Fig. 14. Electron micrograph of the rhabdom of a 4°C fully dark-adapted specimen not previously exposed to
experimental light. The scale is 0.5 /urn.

Fig. 15. Electron micrograph of the rhabdom of a 4°C specimen that was exposed to white light of 4000 lux for 1 hr
and fixed immediately afterwards. The microvilli are well-preserved and the rhabdom shows no signs of
disruption. The scale is 0.5 /an.

Fig. 16. Electron micrograph of microvilli from the eye of a 4°C animal that was kept in the dark for 5 days following
1 hr exposure to white light of 4000 lux. Note the mixture of swollen microvilli, about 2-3 times the normal size,
and normal-sized ones. The scale is 0.5 /an.

Fig. 17. Electron micrograph of rhabdom microvilli as in Fig. 16, but of the eye of an animal kept at 14°C. Note the
overall severe derangement and loosening of microvilli and membrane whorls (arrow) indicative of damage or
turnover. The scale is 0.5 /an.

752

M. Lindstrom, H. L. Nilsson and V. B. Meyer-Rochow

rhabdoms and well preserved photoreceptor mem¬
branes (Fig. 15). This was in strong contrast to
those animals in which a dark period of 5 days
followed the light exposure. The rhabdoms of
these eyes showed swollen microvilli although the
tight packing of the photoreceptor membranes was
still maintained (Fig. 16).

In terms of photoreceptor membrane integrity
the worst conditions were seen when + 14°C ani¬
mals were treated as above and examined 5 days
post-exposure. A marked derangement of the
photoreceptor membranes was noted in distinct
areas of the rhabdom. No longer was the close
packing of the microvilli to be found in these areas
(Fig. 17) and a rough estimation showed that a-
bout 30-50% of the cross-sectional area of the
electron micrographs was occupied by distorted
visual membranes.

DISCUSSION

Initially, prior to the exposure of the adapting
light, sensitivities were of equal magnitude in the
two 4°C and 14°C populations and the controls.
The ultrastructural organisation of the rhabdoms
of the two temperature groups, too, was identical.
Then, following the 1 hr exposure to the adapting
light, both 4°C and 14°C animals, still in unison,
displayed a similar loss in response amplitude.
Rhabdoms and visual membranes still remained
unaffected. The dichotomy of the two tempera¬
ture groups began with post-exposure sensitivity
recovery in the dark but while 4°C and 14°C groups
recovered differently, controls, kept in the dark
throughout the entire test series, showed no signif¬
icant differences of their V/log I relationships.
This proves that temperature does affect the pat¬
tern of recovery of the absolute sensitivity follow¬
ing exposure to light. To understand how the
changes could have been brought about it seems
essential to briefly examine what could have led to
the dramatic sensitivity loss in the first place.

The electrophysiologically-recorded eye re¬
sponse, which was used to monitor sensitivity,
disregarding purely technical difficulties, depends
on a number of phenomena within the eye. Firstly,
there are the amounts of visually excitable photo¬
pigments and non-excitable photoproducts, which

if not removed may filter and attenuate the light
flux down the rhabdoms [28]. No doubt, photopig¬
ment molecules that have absorbed a photon be¬
come unexcitable and will have to be subtracted
from the pool of photo-excitable substance until
reconstituted. Secondly, there is the amount of
visual membrane available for the photopigment
molecules to be built into. With more membrane
area available for photopigment to associate with,
there can be a greater pool of photoexcitable
substance to start with. Then there is, thirdly, the
quality of the membrane, which determines its
permeability characteristics. In the highly sensitive
dark-adapted eye, an active sodium pump in the
membrane is responsible for low intracellular Na+
levels, but upon photoexcitation Na+ ions rush
inwards to produce the well-known receptor
potentials. A leaky membrane or an intracellular
release of Ca2+ can lead to reduced potentials
similar to light-adaptation [29] and a malfunction¬
ing Na+-pump would prevent a speedy return to
resting potential levels. Fourthly, and finally, the
position, abundance, and chemistry of the screen¬
ing pigments, the dimensions, shapes and contents
of receptor and associated cells all can influence
the light flux reaching the visual membranes.

In many species of crustaceans, including
mysids, a certain amount of screening pigment
migration is the rule, but in the compound eyes of
the mysid Neomysis integer pigment movements
are comparatively small, with distal pigments not
passing beyond the cone and proximal pigment not
leaving the rhabdom region [20]. Complete light-
adaptation in Neomysis integer takes about 20 min
[20]. Since even in the frigid water of the Antarctic
environment, pigment migration speeds were no
different from temperate regions [16, 30] we do
not believe our 4°C and 14°C material would have
reacted significantly differently and assume that in
the eye of Mysis relicta screening pigments would
have been in the extreme light-adapted position
following the 1 hr exposure to the adapting light.

It has been reported that unphysiologically high
temperatures, probably via intracellular C02-
poisoning or a lack of oxygen can cause the
screening pigments to remain locked in the light-
adapted position [31]. For three reasons, however,
we do not believe that this would have happened in

Temperature-dependent Visual Sensitivity Recovery

753

our Mysis relicta for (a) the temperatures were not
unphysiologically high [26], (b) the water was
aerated and control animals displayed normal,
high sensitivity, and (c) the appearance of the eyes
of post-exposure 14°C animals under the dissecting
microscope was no different from that of 4°C
animals.

Much more important than the position of
screening pigment seem the processes involved in
membrane turnover and photopigment recovery.
There is evidence that higher temperatures in
crustaceans can speed up the biochemical process
of photopigment restoration [17-19] so that we
should expect 14°C animals to recover more quick¬
ly. At the same time, however, an animal not
sufficiently acclimatised to the higher temperature
regime exhibits an elevated metabolism (higher
Q10-values) and respiration quotients which indi¬
cate a marked degree of conversion of carbohy¬
drates to lipid (Taylor and Meyer-Rochow, unpub¬
lished). It is known that unsaturation of lipids in
membranes increases with chilling [32, 33] and
presumably the opposite occurs with an increase in
temperature. It is unlikely that lipid movement is
achieved by an actual flow of intact membrane
elements since net growth seldom occurs under
stress, but more probably lipids move via an
exchange process such as that known to be medi¬
ated by phospholipid exchange proteins [34].
Furthermore, Thompson [34] states that because
eukaryotic cells are highly compartmentalized,
most lipid modifications are brought about in one
region of the cell, usually the endoplasmic reticu¬
lum. Thus, if under thermal stress sudden in¬
creases in lipid synthesis for an increased volume
of endoplasmic reticulum were required (and an
increase in ER has, indeed, been described by
Meyer-Rochow and Eguchi [35] for retinula cells
of crayfish kept at 30°C for 3 weeks), the implica¬
tion is obvious and would possibly lead to high
rates of carbohydrate conversion to lipid to replace
thermally-unstable membranes.

The metabolic problems are more complicated
still, for up to 80% of the visual membrane may
not actually represent lipid, but a photopigment
protein like rhodopsin [36]. A search, by the way,
for de-hydro-retinal, a photopigment identified by
HPLC in the eye of cold-adapted freshwater

crayfish [37], occurring additionally to the photo¬
pigment rhodopsin, was negative in Mysis relicta
(Suzuki, personal communication). Digitonin ex¬
tracts of two rhodopsins from the eye of Procam-
barus clarkii by Larrivee and Goldsmith [18] readi¬
ly converted to stable metarhodopsin photoprod¬
ucts at 10°C, but when warmed to 22°C they
bleached within minutes to retinal and opsin. As
long as the photoproducts together with the rho¬
dopsins remain components of the membrane they
add to the stability of the latter, but with the
removal of the stabilizing protein following
bleaching at the warmer temperature, serious
membrane instabilities could result, prompting re¬
pairs and possibly causing leakages of the mem¬
brane with a subsequent Na+-influx and/or Ca2+-
release, both of which would tend to reduce elec-
trophysiologically-recorded response amplitudes
[29, 38]. Our ultrastructural data on 4°C and 14°C
light-exposed eyes, examined 5 days post¬
exposure, add considerable weight to the scenario
suggested above.

When it is remembered that responses were
recorded always from the dimmest to the brightest
lights up and never the other way round and this is
seen in connection with the performance of mem¬
brane Na+-pumps which apparently become im¬
paired at higher temperatures [19], it becomes
obvious that the capacity of the 14°C eyes to
sustain the gradually improving sensitivity (as evi¬
denced by the observed increases in thresholds and
derangement of photoreceptor membrane) is in¬
ferior to that of the 4°C group with regard to
brighter stimuli and possibly dimmer, but longer
stimuli as well.

It is interesting to note that a 2-fold decrease in
response amplitudes has also been reported by
Oakley et al. [39] from the eye of the toad Bufo
when temperature was raised from 13.5°C to
27.5°C. Most likely the performance of membrane
pumps was responsible in both cases, though a
small overall decrease in haemolymphatic Na+
reported to occur with an increase in temperature
[40] could have been involved, at least in Mysis
relicta, as well. That reduced slope of V/log I
curves probably resulted from light-adaptation
taking place during the initial phase of the dark-
adapted receptor’s response to a bright flash, was

754

M. Lindstrom, H. L. Nilsson and V. B. Meyer-Rochow

the conclusion reached by Horridge and Tsuka-
hara [4]. According to Matic and Laughlin [42]
this depression of peak amplitudes and the flatten¬
ing of the V/log I curve, could be explained by
voltage sensitive K+ conductance and a reduction
in the ratio of channels to photons occurring
during the response. However, irrespective of
whether Na+, Ca2+ or K+ concentrations were
involved, since almost all of the discussed effects
are membrane-related phenomena it seems safe to
conclude that as long as the dissolution and break¬
down of visual membrane exceeds membrane
synthesis, and the normal membrane turnover is
out-of-balance, consistent V/log I curves of regular
shapes and slopes are not likely to be encountered.

It is generally assumed that newly synthesised
membrane, containing excitable photopigment, is
transported to the rhabdomere via the cytoplasmic
route. However, the fact that clear electrophys-
iological responses to light can be recorded in
invertebrate as well as vertebrate eyes with
seriously distorted visual membranes shown in this
investigation and others [2, 8], suggests that effec¬
tive rhodopsin may not be exclusively located in
the membranes of the rhabdomeric microvilli or
outer segment discs, respectively. Schwemer [4]
suggested that the cell membrane adjacent to the
rhabdomere may be the site of new membrane
insertion plus photopigment. If this were so it
would explain why Drosophila with severely dam¬
aged rhabdomeres responded electrophysiological-
ly to flashes of light (Eguchi, personal communica¬
tion) and why dermal chromatophores in fish con¬
taining rhodopsin [43] but no microvilli or mem¬
brane discs could react directly to light [44].

The membrane adjacent to the rhabdomere is
essentially a smooth wall with few folds to increase
the surface area and, thus, the holding capacity for
photopigment molecules. Overstimulation would
be a constant danger unless the number of ion
channels to photons was reduced, effectively caus¬
ing a decrease in gain. If the increase in sensitivity
in the 14°C group, because of thermal destruction
of the microvillar membranes of the rhabdom,
depended primarily on the amount of photopig¬
ment within newly synthesized membrane, an ex¬
haustion of photoexcitable molecules would be
expected to occur much more rapidly in the 14°C

than in the 4°C group, which suffered less thermal
damage to its rhabdom. This might also explain
the sharper rise and fall times of the responses
observed under higher temperature in light-
exposed animals ([19], Lindstrom and Meyer-

Light Dark Time

- 4

Light Dark Time

Fig. 18a and b. Based on the results of this paper the
scenario suggested in Fig. 1 is amended to show its
temperature dependence. Prior to illumination the
eye is of a certain adaptational state (DA) = height
on the ordinate, which could differ between warm
and cold animals, but did not in our material. The
temperature difference between 14°C (Fig. 18a) and
4°C (Fig. 18b) is deemed too small to significantly
alter the biochemical reaction speed with which
photopigment and light interact. For that reason the
excitatory visual pigment curve (solid line) follows
an identical shape in (a) and (b). Recovery of
photopigment in the dark is faster for the 14°C
material, but so is the membrane breakdown (dotted
line) which in the 14°C material (a) reaches a level
that shows up as physical damage under the micro¬
scope. In the cold water animals (b) photopigment
recovery is slower and though membrane disrup¬
tions characteristic of light adaptation (LA) also
develop, they did not transgress into the damage
zone; — at least not over the period of our observa¬
tions. Whether the 14°C eye with membrane dam¬
age, given time, can recover or loses all capacity for
vision as in Bathynomus [10] and whether in (b)
membrane damage levels would be reached at a
later date presently remain unknown.

Temperature-dependent Visual Sensitivity Recovery

755

Rochow, in preparation).

In conclusion we can say that the observations
reported in this paper have clearly shown that
temperatures influence the pattern of in vivo post¬
exposure sensitivity recovery and the ultrastruc-
tural organisation of the rhabdom (Fig. 18a, b).
They also indicate that in the 14°C population
threshold sensitivites improve faster compared
with the 4°C group, but that flashes of increasing
brightness lead to faster and longer lasting ampli¬
tude reductions in the 14°C post-exposure animals.
Altered membrane properties, like voltage-
sensitive K+ conductance or impaired function of
Na+-pumps, removal of bleached photopigments
and resulting membrane instability, as evidenced
by the ultrastructure of the My sis rhabdom, and
leakiness in the 14°C material are thought to be the
main reason for the documented difference in
post-exposure ERG-behaviours.

Experiments which we have not yet been able to
do, but which ought to be carried out to further
clarify how temperature and light affect photore¬
ceptor function are (a) to study sensitivity recovery
of animals pre-acclimated to 4°C and 14°C under
reverse temperature conditions, i.e. at 14°C and
4°C, respectively; and (b) to expose animals pre-
acclimated to 4°C and 14°C to light under reverse
temperature conditions. Electrophysiological in¬
vestigations ought then to be paralleled by ultra-
structural examination and an analysis of the de¬
gree of lipid saturation of cell membranes in the
eye; a supplementary freeze-fracture study of the
density of membrane particles would be desirable.
Data on quantities of extracted total photopigment
within the receptors of the various experimental
groups should also be compared. Only then can
we hope to unravel and understand some of the
subtleties of the intricately interwoven processes
involved in determining an eye’s sensitivity.

ACKNOWLEDGMENTS

The authors are indebted to the Finnish Academy of
Sciences for the support rendered towards this project
and Mr. Svante Dagerholm for his help in the field during
our nightly fishing. V. B. Meyer- Rochow, further,
wishes to thank the Alexander von Humboldt Founda¬
tion (Bonn, W. Germany) and Mr. Ivan Manning (Bio¬
logical Sciences, Waikato University) for his skillful

operation of the computer used in analysing the data and
Dr. Mark Schroder (Department of Mathematics,
Waikato University) for his advice on the mathematical
treatment. H. L. Nilsson (electron micrographs) is
grateful to the Swedish Medical Research Council (Plan¬
ning Group for Peripheral Visual Physiology, Grant 145—
3000) and wishes to thank Prof. R. Elofsson (Grant
NFR-2760-113) for his interest and encouragement.
The manuscript was completed in the stimulating sur¬
roundings of Prof. E. Eguchi’s laboratory at Yokohama
City University, Department of Biology.

REFERENCES

1 Eguchi, E. (1986) Eyes and darkness — evolution¬
ary and adaptational aspects. Zool. Sci., 3: 931-943.

2 Rapp, L. M. and Williams, T. P. (1977) Rhodopsin
content and electroretinographic sensitivity in light-
damaged rat retina. Nature, 267: 835-836.

3 Kaitz, M. and Auerbach, E. (1979) Retinal degen¬
eration in RCS rats raised under ambient light
levels. Vision Res., 19: 79-81.

4 Schwemer, J. (1985) Turnover of photoreceptor
membrane and visual pigment in invertebrates. In
“The Molecular Mechanism of Photoreception”.
Ed. by H. Stieve, Springer-Verlag, Berlin, pp. 303-
326.

5 Hamdorf, K. (1979) The physiology of invertebrate
visual pigments. In “Handbook of Sensory Physiolo¬
gy, Vol. VII/6A. Comparative Physiology and
Evolution of Vision in Invertebrates”. Ed. by H.
Autrum, Springer-Verlag, Berlin, pp. 145-224.

6 Cummins, D. R. and Goldsmith, T. H. (1986) Re¬
sponses of crayfish photoreceptor cells following
intense light adaptation. J. Comp. Physiol., A158:
35-42.

7 Lindstrom, M. and Nilsson, H. L. (1983) Spectral
and visual sensitivities of Cirolana borealis Lill-
jeborg, a deep-water isopod (Crustacea:
Flabellifera). J. Exp. Mar. Biol. Ecol., 69: 243-256.

8 Meyer-Rochow, V. B. and Tiang, K. M. (1984) The
eye of Jasus edwardsii (Crustacea, Decapoda, Pali-
nuridae): electrophysiology, histology, and be¬
haviour. Zoologica, 45 (134): 1-85.

9 Nilsson, H. L. and Lindstrom, M. (1983) Retinal
damage and sensitivity loss of a light-sensitive
crustacean compound eye ( Cirolana borealis ) elec¬
tron microscopy and electrophysiology. J. Exp.
Biol., 107: 277-292.

10 Chamberlain, S. C., Meyer-Rochow, V. B. and
Dossert, W. P. (1986) Morphology of the eye of the
giant deep-sea isopod Bathynomus giganteus. J.
Morphol., 189: 145-156.

11 Loew, E. R. (1976) Light, and photoreceptor de¬
generation in the Norway lobster, Nephrops norve-
gicus (L). Proc. R. Soc. London, 193B: 31-44.

756

M. Lindstrom, H. L. Nilsson and V. B. Meyer-Rochow

12 Meyer-Rochow, V. B. (1981) The eye of
Orchomene sp. of O. rossi, an amphipod living
under the Ross Ice Shelf (Antarctica). Proc. R. Soc.
London, 212B: 93-111.

13 Meyer-Rochow, V. B. and Eguchi, E. (1983) Ultra¬
structure du rhabdome de Procambarus clarkii im-
mediatement et 20 heures apres d’un contenu photo-
nique egal. Biol. Cell, 48: 185-190.

14 Shelton, P. M. J., Gaten, E. and Chapman, C. J.
(1985) Light and retinal damage in Nephrops norve-
gicus (L.) (Crustacea). Proc. R. Soc. London, 226B:
217-236.

15 Attramadal, J. G., Fossa, J. H. and Nilsson, H. L.
(1985) Changes in behaviour and eye-morphology
of Boreomysis megalops Sars (Crustacea: Mysi-
dacea) following exposure to short periods of arti¬
ficial and natural daylight. J. Exp. Mar. Biol. Ecol.,
85: 135-148.

16 Meyer-Rochow, V. B. andTiang, K. M. (1979) The
effects of light and temperature on the structural
organisation of the eye of the Antarctic amphipod
Orchomene plebs (Crustacea). Proc. R. Soc. Lon¬
don, 206B: 353-368.

17 Barnes, S. N. and Goldsmith, T. H. (1977) Dark
adaptation, sensitivity, and rhodopsin level in the
eye of the lobster, Homarus. J. Comp. Physiol., 120:
143-139.

18 Larrivee, D. and Goldsmith, T. H. (1982) Spectral
dimorphism of crayfish visual pigment in solution.
Vision Res., 22: 727-737.

19 Stieve, H. (1963) Das Belichtungspotential der Re¬
tina des Einsiedlerkrebses in Abhangigkeit von der
Temperatur. Z. Vergl. Physiol., 46: 249-275.

20 Hallberg, E. (1977) The fine structure of the com¬
pound eyes of mysids (Crustacea: Mysidaceae). Cell
Tissue Res., 184: 45-65.

21 Hallberg, E., Anderson, M. and Nilsson, E.-E.
(1980) Responses of the screening pigments in the
compound eye of Neomysis integer (Crustacea:
Mysidacea). J. Exp. Zool., 212: 397-402.

22 Mayrat, A. (1956) Oeil, centres optiques et glandes
endocrines de Praunus flexuosus. Arch. Zool. Exp.
Gen., 93: 320-366.

23 Meyer-Rochow, V. B. and Juberthie-Jupeau, L.
(1987) An electronmicroscope study of the eye of
the cave mysid Heteromysoides cotti from the island
of Lanzarote (Canary Islands). Stygologia, 3: 24-34.

24 Nilsson, D. -E., Hallberg, E. and Elofsson, R.
(1986) The ontogenetic development of refracting
superposition eyes in crustaceans: transformation of
optical design. Tissue and Cell, 18: 508-519.

25 Lindstrom, M. and Nilsson, H. L. (1984) Synforma-
gan hos pungrakan Mysis relicta i olika ljusmiljoer.
Mem. Soc. Fauna Flora Fennica, 60: 35-38.

26 Stalberg, G. (1933) Beitrag zur Kenntnis der Biolo-
gie von Mysis relicta des Vattern. Arkiv. Zoologi,

26A(15): 1-30.

27 Cronin, T. W. and Goldsmith, T. H. (1984) Dark
regeneration of rhodopsin in crayfish photorecep¬
tors. J. Gen. Physiol., 84: 63-81.

28 Goldsmith, T. H. (1978) The spectral absorption of
crayfish rhabdoms: pigment, photoproduct and pH-
sensitivity. Vision Res., 18: 463-473.

29 Brown, J. E. (1986) Calcium and light adaptation in
invertebrate photoreceptors. In “The Molecular
Mechanism of Photoreception”. Ed. by H. Stieve,
Springer-Verlag, Berlin, pp. 231-240.

30 Obika, M. and Meyer-Rochow, V. B. (1986) Ultra¬
structure of microtubules in dermal melanophores
and spinal nerve of the Antarctic Pagothenia bor-
chgrevinki. Cell Tissue Res., 244: 339-343.

31 Meyer-Rochow, V. B. andTiang, K. M. (1982) Com¬
parison between temperature-induced changes and
effects caused by dark/light adaptation in the eyes of
two species of Antarctic crustaceans. Cell Tissue
Res., 221: 625-632.

32 Hazel, J. R. and Prosser, C. L. (1974) Molecular
mechanisms of temperature compensation in poiki-
lotherms. Physiol. Rev., 54: 620-677.

33 Meyer-Rochow, V. B. and Pyle, C. L. (1980) Fatty
acid analysis of lens and retina of two Antarctic
fishes and of the head and body of the Antarctic
amphipod Orchomene plebs. Comp. Biochem.
Physiol., 65B: 395-398.

34 Thompson, G. A., Jr. (1983) Mechanism of
homeoviscous adaptation in membranes. In “Cellu¬
lar Acclimatisation to Environmental Change (Soc.
for Exp. Biol. Seminar Series No. 17)”. Ed. by A.
R. Cossins and P. Sheterline, Cambridge Univ.
Press, Cambridge, pp. 33-53.

35 Meyer-Rochow, V. B. and Eguchi, E. (1984) The
effects of temperature and light on particles associ¬
ated with crayfish visual membrane: a freeze-
fracture analysis and electrophysiological study. J.
Neurocytol., 13: 935-959.

36 Hamacher, K. and Stieve, H. (1984) Spectral prop¬
erties of the rhodopsin-system of the crayfish Asta-
cus leptodactylus. Photochem. Photobiol., 39: 379-
390.

37 Suzuki, T., Makino-Tasaka, M. and Eguchi, E.
(1984) De-hydroretinal (Vitamin A2-aldehyde) in
crayfish eye. Vision Res., 24: 783-787.

38 Chinn, K. and Lisman, J. (1984) Calcium mediates
the light-induced decrease in maintained K+ current
in Limulus ventral photoreceptors J. Gen. Physiol.,
84: 447-462.

39 Oakley, B., Flaming, D. and Brown, K. T. (1979)
Effects of the rod receptor potential upon retinal
extracellular potassium concentration. J. Gen. Phys¬
iol., 74: 713-737.

40 Burton, R. F. (1986) Tonic regulation in Crustacea:
the influence of temperature on apparent set points.

Temperature-dependent Visual Sensitivity Recovery

757

Comp. Biochem. Physiol., 84A: 135-139.

41 Horridge, G. A. and Tsukahara, Y. (1978) The dis¬
tribution of bumps in the tail of the locust photore¬
ceptor after potential. J. Exp. Biol., 73: 1-14.

42 Matic, T. and Laughlin, S. B. (1981) Changes in the
intensity-response function of an insect’s photore¬
ceptors due to light adaptation. J. Comp. Physiol.,

145: 169-177.

43 Lythgoe, J. N., Shand,J. and Foster, R. G. (1984)
Visual pigment in fish iridocytes. Nature., 308: 83-
84.

44 Iga, T. and Matsuno, A. (1986) Motile iridophores
of a freshwater goby, Odontobutis obscura. Cell
Tissue Res., 244: 165-171.

ZOOLOGICAL SCIENCE 5: 759-766 (1988)

© 1988 Zoological Society of Japan

Mode of Melanosome Migration in Teleostean Melanophores

Tomio Naitoh, Katsumi Takeuchi and Ikuo Takabatake

Department of Biology, Shimane University, 1060 Nishikawatsu,

Matsue, Shimane 690, Japan

ABSTRACT — Using melanophores in scales of the fish Chaenogobius sp. 2, we observed the migratory
behavior of melanosomes that delayed the start of centripetal movement from the tip of cell processes
during melanosome aggregation. Melanosomes with delayed starts migrated smoothly and individually,
or in a very small group in succession. When melanosome dispersion was induced before all
melanosomes had reached the central mass, centripetal migration of the peripheral melanosomes and
centrifugal migration of the central melanosome mass was simultaneously observed within single cell
processes. This indicates that the melanophores possess two distinct intracellular activities concerning
the melanosome migration: one which forces melanosomes in a periphero-central direction and the
other activity which causes dispersion. Velocity for the centripetal migration of individual melanosomes
showed three phases as the granules travelled along cell processes; a first phase of rapid acceleration, a
second phase with slightly increasing speed and a deceleration phase. Differences in the structure of the
cytoplasm along cell processes may account for the differences in velocity during the three phases of
migration. There are no differences in the velocity profiles between melanophores obtained from the
dark stripes of skin and those from light ones. This suggests that the differential responses of
melanophores that cause the patterned skin, do not depend on their intracellular characteristics, but on
the differential extrinsic stimuli.

INTRODUCTION

The migration of pigment granules within fish
chromatophores has been of interest to many
researchers. As shown in reviews [1-4], a variety
of physiological, pharmacological, biochemical
and electronmicroscopy studies have produced
some important concepts on this phenomenon.

Many methods have been used in the physio¬
logical and pharmacological study of chroma¬
tophores [1,2]. Most of them, however, deal with
recording the movement of pigment granule mass.
When this method is used, it is hard to analyse
local activities occurring within chromatophores.
In order to fully understand pigment granule
migration, behavior of pigment granules in each
site of cell processes should be observed. One of
the best ways to study this behavior seems to be
recording the translocation of individual granules
along a process. Although many papers have been
published on the chromatophores, only a few of

Accepted November 13, 1987
Received March 31, 1987

these papers [5-9] have traced the movement of
individual granules.

In the scale melanophores of Chaenogobius
(Gobiidae) species, it is striking that a consider¬
able number of melanosomes are commonly left at
the tip of cell processes during the aggregation
response of the cell, and that those granules later
move to the cell center, in succession, to join the
central melanosome mass. In an attempt to under¬
stand the granule migration mechanism in chroma¬
tophores, we observed the behavior of the melano¬
somes which had delayed their migration in these
melanophores.

MATERIALS AND METHODS

Both sexes of Sumiukigori Chaenogobius sp. 2
[10] were used. They were 7-9 cm in length and
were taken from the streams at Kaga and Chikumi
in Shimane Peninsula, Japan. This fish has the
ability of assuming a patterned background with
dark and light stripes, which depends on the
differential activities of melanophores, in addition
to matching the background shade.

760

T. Naitoh, K. Takeuchi and I. Takabatake

A scale, in which the melanophores are embed¬
ded on the outer side, was plucked from the
postero-lateral region of the body trunk near the
second dorsal fin. It was then set in a trough with
the external surface down and single processes of
the melanophore were observed with a 100 X
oil-immersion objective on a binocular light micro¬
scope (Olympus BH-2). Recording of melano-
some migration was done by two methods; time
lapse microphotography at an interval of 1.4 to 2.6
sec with a motor-driven camera (Olympus OM-2N
and a motor driving unit) and direct measurement
of melanosome location through a 10 X ocular lens
equipped with a micrometer. Super-imposed trac¬
ing of photographs was used to estimate melano¬
some velocity. Melanosome aggregation was in¬
duced by three methods: 1) the administration of
an isotonic KC1 solution or a KC1 solution diluted
2:1 with physiological saline solution; 2) 10
norepinephrine chloride (Sankyo) dissolved in a
physiological saline solution; or 3) a 20 to 50 V-60
Hz alternating current for 60 sec through a glass
capillary electrode that was 25 to 50 pm in tip
diameter [11]. Redispersion of the melanosomes
was caused by: 1) exchanging the KC1 solution
with a physiological saline solution; 2) by nor¬
epinephrine solution for 100 atropine sulfate
(Sigma) dissolved in a physiological saline solu¬
tion; or 3) by removal of the electrical current,
respectively.

The physiological saline solution was composed
of NaCl 128 mM, KC1 2.6 mM, CaCl 2 1.8 mM and
buffered to pH 7.2 with Tris-HCl. The room
temperature ranged between 24.0 and 24.5 °C.

RESULTS

General characteristics of melanophores

Figure 1 shows a melanophore that is located in
a Chaenogobius scale. Melanophores situated in
the dark stripes of the skin were more branched
and darker than those in the light stripes. The
central area of dispersed melanophores was usual¬
ly less granulated. The diameter of a cell in the
dispersed state was 130 to 220 pm. In this species
as in many other teleostean melanophores, ap¬
plication of KC1, norepinephrine or an alternating

30pm

Fig. 1. Scale melanophore of Chaenogobius sp. 2. A
melanophore from a dark stripe of skin is shown in
A and B; a melanophore from a light stripe, in C and
D. A, C: Melanosome dispersed melanophore in
physiological saline solution. B, D: Melanosome
aggregation in isotonic KC1 solution for 10 min.

current caused melanosomes to aggregate in the
cell center, in a clump 15-55 pm in diameter.
Melanosome mass was sharply outlined in the
cytoplasm when mass aggregation was completed.
Removal of alternating current or replacement of
KC1 solution by the physiological saline solution
resulted in melanosome redispersal. After nor¬
epinephrine treatment, melanosome redispersal
was difficult to be produced with the physiological
saline solution alone. However, application of
atropine soon elicited the complete redispersal of
the melanosomes.

Migration of melanosomes separated from the mass

Melanophore response to KC1 or nor¬
epinephrine took up to five minutes to initiate.
With alternating current the response took several
seconds to start. From then on, it took from five to
fifteen minutes to attain the maximal melanosome
aggregation, with the time depending on the indi-

Melanosome Migration

761

vidual differences of melanophores, their locations
in the scale, and the concentration of aggregation
agents or current intensity.

It was notable that a group of melanosomes was
left at the extremity of the processes during the

10pm

Fig. 2. Translocation of the melanosomes separated
from the central mass. A: Melanosome-dispersed
state in physiological saline solution. B-L: Move¬
ment of three melanosomes in a group in isotonic
KC1 solution. Interval between each figure except A
and B: 2.8 sec. Melanosomes indicated by arrow
began their migration from the terminal location (B)
and are moving to the central mass (C-L). The
melanophore used in this preparation is taken from
a light stripe of skin.

aggregation response of melanophores. However,
over a period of time all of these melanosomes
moved individually or in a small group to join the
central mass (Fig. 2). In the melanophores treated
with norepinephrine, it usually took more than 1.5
hr for all of the melanosomes left to move to the
cell center. Melanosome migration was linear, as
though they ran along a narrow track (Fig. 3).
Even when melanosome dispersal was allowed by
removal of KC1 or electric current, or by adminis¬
tration of atropine, distal melanosomes which
were separated from the mass, continued to move
in centripetal direction, until their movements
were overwhelmed by the central mass dispersal
(Figs. 4 and 5). The velocity of the distal melano¬
somes decreased or the melanosomes made U-
turns before they joined the central mass. This
indicates that centrifugal migration effects the cen¬
tripetal movement even before the centrifugal
melanosomes reach the inwardly migrating ones.

It took fifteen to twenty minutes to recover
maximal granule dispersal after removal of either
KC1 or the alternating current. Treatment with
atropine completed the redispersal slightly faster.
During the course of redispersal, melanosomes
moved back and forth over a short distance. In
addition, several pulsations were occasionally
observed, though no pulsations were recorded in
the atropine treated cells.

Velocity of the centripetal migration of individual
melanosomes

The speed of individual granules, which moved
separately after the mass aggregation was com¬
pleted, was frequently faster than that of the mass
aggregation. Furthermore, velocities of individual

Fig. 3. Superimposed traces of melanosome location in
the pictures B to L from Fig. 2. Locations indicated
by the capital letters correspond with those of the
pictures in Fig. 2, respectively.

762

T. Naitoh, K. Takeuchi and I. Takabatake

20pm

Fig. 4. Observation of both centripetal and centrifugal
migration of melanosomes in a sigle cell process. A:
Melanosome-dispersed state. B-G: Series of photo¬
graphs taken at intervals of 2.8 sec in physiological
saline solution after removal of an alternating cur¬
rent of 50 V-60 Hz. Melanosomes, indicated by the

granules were slightly different depending on the
processes. However, the granules which appeared
to move in the same “lane” of a process had similar
velocities. Average moving speed of melanosomes
is shown in Figure 6. It indicates three phases of
velocity; the first phase, in which aggregation of
melanosomes rapidly accelerated; the second
phase, ranging between about 1.5 and 2.5 //m/sec,
where the speed increased only slightly; and the
third phase in which the melanosomes slowed
down. Basic features of the moving speed did not
seem to vary with the different kinds of stimuli.

DISCUSSION

Previous workers [5-7, 12] have noted, that
when melanophores are stimulated to aggregate,
some melanosomes are usually left at the tip of the
processes. The same is true, and rather striking so,
in the melanophores of the present species. To
investigate the dynamics of the granule migration
in chromatophores, melanosomes which separate
from the central mass and then later move indi¬
vidually in succession, seem particularly suitable
for study.

By following the movement of melanosomes at
the central mass and the movement of the melano¬
somes left at the tip of a melanophore process,
Kamada and Kinosita [6] observed both types of
migration, centrifugal and centripetal, occurring at
the same time in one process. In the present study,
the same phenomenon was observed. These iden¬
tical results suggest that intracellular activities are
quite different between the distal region of the
cell, where melanosomes are moving centripetally,
and the proximal region, where melanosomes are
moving centrifugally. From this, we conclude that
melanophores possess two distinct intracellular
activities concerning melanosome migration; one
activity which forces melanosomes to move in a
central direction and the other activity which
makes them disperse. The melanosome driving
force is controlled locally. A balance between the

arrows on the left, are migrating centripetally, while
the central mass, indicated by the arrows on the
right, is dispersing centrifugally. The melanophore
used in this preparation is taken from a light stripe of
skin.

Melanosome Migration

763

Norepinephrine Atropine

Fig. 5. Translocation of melanosomes traced with an ocular micrometer. A, B and C were treated with 10 /jM.
norepinephrine, isotonic KC1 and an alternating current of 50 V, respectively. The stippled portion indicates the
area where mass melanosomes are distributed. 1 is the trace of the outer edge of the melanosome mass. The
remaining are the traces of a single melanosome or a very small number of melanosomes in a group. Both
movements of the individual melanosomes and of the central mass are traced in a given cell process. Note that the
melanosomes left at the extremity of the process move centripetally even when the central mass begins to
disperse. The melanophore used in these preparations are taken from a light stripe of skin.

areas occupied by the two different cell activities
seems to determine a given state of melanosome
distribution, such as complete dispersion, half dis¬
persion or half aggregation, or complete aggrega¬
tion. The origin of the pulsations is considered to
be neural in some melanophores [13]. However,
the phenomenon seems to be ultimately ascribable
to the fluctuation in the balance of the two antago¬
nistic intracellular activities.

Pigment migrating speeds in chromatophores
have been recorded by some investigators [7, 9,
14-19]. In Pterophyllum melanophores, melano¬
somes treated with KC1 to induce centripetal
movement travel at 0.56-3.69 pm/sec [14] or 2
/un/sec [16]; and with epinephrine at 1.5-2 /un/sec
[19]. Green [7] obtained speeds of 2-6 /un/sec in
Fundulus melanophores by epinephrine adminis¬
tration. In Holocentrus erythrophores, a much
higher velocity was reported, 10 //m/sec with KC1
treatment [17] and 16 jum/sec with epinephrine

[15]. The results we obtained are similar to the
velocities in the former two species. Because
pigment granules moved linearly and smoothly to
the cell center, it is not hard to estimate their
velocity of centripetal migration by means of mi¬
crocinematography or time lapse microphotogra¬
phy. However, except for Kamada and Kinosita
[6] who mentioned a slow start for melanosomes
and their slowing down after joining the mass,
there have been no reports which examined the
velocity throughout the whole aggregating event.
Many workers have noted that centripetal migra¬
tion of pigment granules occurs in a uniform
fashion [4, 6, 7, 17, 20, 21]. But, in the present
study three velocity phases were revealed. The
results obtained suggest that the intracellular back¬
ground of the first and second phases is stable,
while that of the third phase is changeable. This is
indicated by the fact that the sites where the first
two phases are observed are determined according

Velocity (pm/sec) Velocity (pm/sec) Velocity (pm/sec)

764

T. Naitoh, K. Takeuchi and I. Takabatake

Fig. 6. Velocity of the centripetal migration of indi¬
vidual melanosomes. A: norepinephrine treatment,
B: isotonic KC1 treatment, C: alternating current
stimulation. •: melanophores obtained from a dark
stripe of skin, <3: melanophores from a light stripe.
Ordinates: Velocity of melanosome aggregation.
Abscissae: Length of the cell process as the percen¬
tage of the distance between the site of the separated
melanosomes in the tip of the cell process (100) and
the site of the edge of aggregated melanosome mass
in the cell center (0). The cell center is located at the
right hand side. Each point is the mean of 10 to 19
measurements and vertical bars represent standard
deviations. Rapid increase of velocity at the begin¬
ning (first phase) is followed by a gradual increase
(second phase), and then deceleration (third phase).

to the site of the processes, but the third phase is
determined, not by the fixed site of the processes,
but by the distance from the edge of the central
mass that moves along the processes. Following
the work of Bikle et al. [22], microtubules have
been considered important in translocation of pig¬
ment granules within chromatophores [2-4].
Melanosome speed during centripetal migration
may depend on intracellular structure; specifically
the delayed start of granules at the extremity of the
cell processes, and the velocity during the first two
phases may be due to difference of microtubular
density in the respective sites along melanophore
processes. Junqueira et al. [18] suggested that
chromatophores with granules that migrate at high
velocity may have more microtubules in the cyto¬
plasm than those with slowly migrating granules.
Centripetal movement of the distal melanosomes
was affected by the centrifugal movement of the
central melanosome mass before they joined the
central mass, resulting in a lower velocity or even
in a U-turn for the distal melanosomes. This
observation seems to support the conclusion of
Byers and Porter [20] and Schliwa [23] that the
reconstruction of a microtrabecular lattice is a
prerequirement for the arrival of pigment granules
in the chromatophore dispersal phase. The third
phase appears to be caused by an effect of granule
dispersal at the perimeter of the central melano¬
some mass.

The fish, Chaenogobius sp. 2, on the patterned
background, such as a white-and-black checker¬
board, shows patterned skin with dark and light
stripes. This is caused by the differential response
of melanophores in the skin; melanophore disper¬
sion in the dark stripes while aggregation in the
light stripes. However, individual melanosomes
migrate similarly in melanophores located in both
the dark stripes of skin and in the light ones. This
suggests that the intracellular activities of mela¬
nophores in both regions are basically the same.
Moreover, we can add that the driving force for
each granule is the same between the mela¬
nophores obtained from the different stripes of the
skin. Therefore, the differential responses of
melanophores on the body surface, which cause
the patterned skin on the patterned background,
seems to depend not on difference in the in-

Melanosome Migration

765

tracellular characteristics of melanophores, but on
possible difference in information from the central
nervous system.

Melanophore response can be produced with
various types of stimulation acting on different
targets; e.g., alternating current stimulates the
nerve fibers [24, 25], KC1 affects nerve terminals
[26, 27], and norepinephrine stimulates mela¬
nophores directly [28]. Irrespective of the kind of
stimulation, individual melanosomes moved in a
similar way, with a similar velocity. Besides, it
seems unlikely that alternating current and KC1
induce the liberation of the same quantity of
transmitter as that of norepinephrine administered
externally. Therefore, it is reasonable to think that
the force for translocation of the individual
melanosomes does not depend on the intensity of
external stimuli.

The direction of movement of individual
melanosomes seems to depend on the balance
between aggregative and dispersal activities. The
velocity of individual melanosomes, in turn, seems
to depend on the shift in the balance between the
two activities.

ACKNOWLEDGMENTS

We would like to thank Prof. R. Wassersug and Miss
K. Murphy of Dalhousie University for constructive
comments on the manuscript. Lise Boylan and Christine
Anjowski graciously helped with word processing.

REFERENCES

1 Fujii, R. (1969) Chromatophores and pigments. In
“Fish Physiology”. Vol. 3. Ed. by W. S. Hoar and
D. J. Randall. Academic Press, New York, pp. 307-
353.

2 Schliwa, M. (1982) Chromatophores: their use in
understanding microtubule-dependent intracellular
transport. In “Methods in Cell Biology. The Cyto-
skeleton. Biological Systems and in vitro Methods”.
Vol. 25B. Ed. by D. Prescott and L. Wilson,
Academic Press, New York, pp. 285-312.

3 Schliwa, M. and Euteneuer, U. (1983) Comparative
ultrastructure and physiology of chromatophores,
with emphasis on changes associated with intracellu¬
lar transport. Am. Zool., 23: 479-494.

4 Obika, M. (1986) Intracellular transport of pigment
granules in fish chromatophores. Zool. Sci., 3: 1-11.

5 Matthews, S. A. (1931) Observations on pigment

migration within the fish melanophore. J. Exp.
Zool., 58: 471-486.

Kamada, T. and Kinosita, H. (1944) Movement of
granules in fish melanophores. Proc. Imp. Acad.
Tokyo, 20: 484-492.

Green, L. (1968) Mechanism of movements of
granules in melanocytes of Fundulus heteroclitus .
Proc. Natl. Acad. Sci. USA, 59: 1179-1186.

Porter, K. R. and McNiven, M. A. (1982) The cyto-
plast: a unit structure in chromatophores. Cell, 29:
23-32.

Matsumoto, J., Akiyama, T. and Nakahari, M.
(1984) Surface morphology of fish pigment cells: its
roles in analyses of their motility and functional
differentiation. In “Scanning Electron Microscopy
III”. Ed. by A. M. F. O’HarC, SEM Inc., Chicago,
pp. 1279-1288.

Masuda, H., Amaoka, K., Araga, C., Ueno, T. and
Yoshino, T. (1984) The Fishes of the Japanese
Archipelago. Tokaidaigaku Shuppankai, Tokyo, p.
265 and pi. 252.

Naitoh, T. (1976) Melanophore pulsations induced
by continuous AC stimulation of the scale of the
bitterling and the crucian carp. Biol. J. Okayama
Univ., 17: 15-21.

Schliwa, M. and Bereiter-Hahn, J. (1973) Pigment
movements in fish melanophores: morphological
and physiological studies. III. The effects of colchi¬
cine and vinblastine. Z. Zellforsch., 147: 127-148.
Fujii, R. and Oshima, N. (1986) Control of chroma-
tophore movements in teleost fishes. Zool. Sci., 3:
13-47.

Egner, O. (1971) Zur Physiologie der Melano-
somenverlagerung in den Melanophoren von Pter-
ophyllum scalar e Cuv. u. Val. Cytobiologie, 4: 262-
292.

Junqueira, L. C. U. and Porter, K. R. (1971)
Studies on the cell physiology of the erythrophores
of the teleost Holocentrus. Rev. Brasil. Biol., 31:
299-304.

Schliwa, M. and Bereiter-Hahn, J. (1974) Morpho-
logische und physiologische Grundlagen der Pig-
mentbewegung in Fisch-Melanophoren; I. Methodik
der Bewegungsanalyse. Microsc. Acta, 75: 235-242.
Porter, K. R. (1973) Microtubules in intracellular
locomotion. In “Locomotion of Tissue Cells”. Ciba
Foundation Symp., 14: 149-169.

Junqueira, L. C., Raker, E. and Porter, K. R.
(1974) Studies on pigment migration in the mela¬
nophores of the teleost Fundulus heteroclitus (L).
Arch. Histol. Jpn., 36: 339-366.

Schliwa, M. and Euteneuer, U. (1978) Quantitative
analysis of the microtubule system in isolated fish
melanophores. J. Supramol. Struct., 8: 177-190.
Byers, H. R. and Porter, K. R. (1977) Transforma¬
tions in the structure of the cytoplasmic ground

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

766

T. Naitoh, K. Takeuchi and I. Takabatake

substance in erythrophores during pigment aggrega¬
tion and dispersion. J. Cell Biol., 75: 541-558.

21 Rozdzial, M. M. and Haimo, L. T. (1986) Reacti¬
vated melanophore motility: differential regulation
and nucleotide requirements of bidirectional pig¬
ment granule transport. J. Cell Biol., 103: 2755-
2764.

22 Bikle, D., Tilney, L. G. and Porter, K. R. (1966)
Microtubules and pigment migration in the mela-
nophores of Fundulus heteroclitus L. Protoplasma,
61: 322-345.

23 Schliwa, M. (1979) Stereo high voltage electron
microscopy of melanophores. Exp. Cell Res., 118:
323-340.

24 Fujii, R. and Novales, R. R. (1969) The nervous
mechanism controlling pigment aggregation in Fun¬
dulus melanophores. Comp. Biochem. Physiol., 29:
109-124.

25 Naitoh, T. and Iwata, K. S. (1976) Physiological
studies on the distribution of the melanophore-
contracting nerves in the scale of the bitterling,
Acheilognathus lanceolata. Biol. J. Okayama Univ.,
17: 1-13.

26 Fujii, R. (1959) Mechanism of ionic action in the
melanophore system of fish. I. Melanophore-
concentrating action of potassium and some other
ions. Annot. Zool. Japon., 32: 47-58.

27 Iwata, K. S., Watanabe, M. and Nagao, K. (1959)
The mode of action of pigment concentrating agents
on the melanophores in an isolated fish scale. Biol.
J. Okayama Univ., 5: 195-206.

28 Fujii, R. (1961) Demonstration of the adrenergic
nature of transmission at the junction between mela-
nophore-concentrating nerve and melanophore in
bony fish. J. Fac. Sci. Univ. Tokyo, Sec. IV, 9: 171—
-196.

ZOOLOGICAL SCIENCE 5: 767-780 (1988)

© 1988 Zoological Society of Japan

Musculature and Innervation of the Internal Reproductive Organs
in the Male Cricket, with Special Reference to the Projection of
Unpaired Median Neurons of the Terminal Abdominal Ganglion

Kouji Yasuyama, Tetsuya Kimura1 and Tsuneo Yamaguchi1, 2

Department of Biology, Kawasaki Medical School, Kurashiki 701-01,
and 1 Department of Biology, Faculty of Science,

Okayama University, Okayama 700, Japan

ABSTRACT — In the male cricket, nickel back-filling of the nerve branches separating from the seventh
nerve root (R7) of terminal abdominal ganglion (TAG) reveals the presence of two types of neurons
running down to the internal reproductive organs (accessory glands, epididymis and ejaculatory duct).
The first type is the neurons (LC neurons) which have the somata forming a cluster in the postero-lateral
region of TAG, and 13 somata are present in the cluster. The second type is the dorsal unpaired median
neurons (DUMR7 neurons), of which the somata form three clusters along the midline of the posterior
half of TAG and the bifurcating axons extend symmetrically to the left and right R7s. The total number
of somata in these clusters is 100 or more. The DUMR7 neurons probably terminate on the muscle
fibers of the reproductive organs. This is based on the following evidences. First, the axon terminals,
which contain both small vesicles (SVs; 40-60 nm) and large vesicles (LVs; 100-150 nm), are found in
the musculature of the reproductive organs. Second, horseradish peroxidase (HRP) back-filled
DUMR7 neurons exhibit that LVs are present in their somata as well as in their axons running to the
reproductive organs. Third, the ligation of either R7 results in the remarkable accumulation of LVs in
the cytoplasm of axon proximal to the ligature.

INTRODUCTION

The basic process of spermatophore formation
in insects requires for both producing various
spermatophore-forming materials in the accessory
glands (AGs) and releasing them from AGs
according to a temporal program [1]. It is most
probable that in the time of release, the contrac¬
tion of muscles of AGs triggered by neural signals
causes the extrusion of materials from the lumens
of the gland tubules. In fact, Odhiambo [2] has
shown that in male desert locust, thin single muscle
layers surround individual tubules of AGs and the
axons originating from the terminal abdominal
ganglion (TAG) make synaptic contacts with the
muscle layers. Further, distribution of the axons
among the muscle layers of AGs has been de-

Accepted December 2, 1987
Received October 21, 1987
2 To whom reprint requests should be addressed.

scribed in other insects (in Leptinotarsa [3]; in
Periplaneta [4]). Recently, Yamaguchi et al. [5]
have reported that in male cricket the dorsal
unpaired median neurons (DUMR7 neurons) of
TAG extend their axons bilaterally through
seventh nerve roots of TAG toward AGs. Howev¬
er, we have rather as yet very little information
about not only the musculature and innervation of
AGs relating to the neural control of sperma¬
tophore formation, but also those of other internal
reproductive organs.

In this paper we describe the gross morphology
of the musculature and innervation of internal
reproductive organs in the male cricket, together
with their fine structures, and discuss the function¬
al role of the morphologically identified neurons,
especially of the dorsal unpaired median neurons,
in the spermatophore formation on the basis of
morphological characteristics.

Preliminary reports of some of this work have
been already presented elsewhere [6, 7].

768

K. Yasuyama, T. Kimura and T. Yamaguchi

MATERIALS AND METHODS

Adult male crickets ( Gryllus bimaculatus )
reared in the laboratory, were used for the experi¬
ments. For nickel back-filling the proximal cut end
of one of seventh nerve roots (R7s) emerging
bilaterally from TAG was plunged into a thin
capillary filled with 1 M NiCl2 for 12-14 hr at 4°C.
Then, nickel was precipitated within the neurons
by addition of rubeanic acid to the saline [8] in
which the preparation was immersed. After fixa¬
tion with 10% formalin, the preparation was dehy¬
drated and cleared for whole-mount viewing.
Usually, the preparation stained with nickel was
intensified following principally the method of
Bacon and Altman [9].

For horseradish peroxidase (HRP) back-filling
the proximal cut end of R7 was plunged into a thin
capillary filled with 20% HRP for 24 hr at 10°C,
and then TAG with the reproductive organs was
fixed in 3% glutaraldehyde in 0.1 M cacodylate
buffer (pH 7.2) for 2 hr, followed by washing in
the buffer. The enzyme reaction was performed
with three drops of 3% H202 in 1% 3, 3’-
diaminobenzidine tetrahydrochloride (DAB) solu¬
tion for 15 min. After washing in the buffer, the
preparation was fixed with 2% 0s04 in the buffer
for 2 hr, then dehydrated through an ethanol series
and embedded in epon resin. Sections, lOjum
thick, were photographed, then reembedded and
cut for electron microscopy. Thin sections were
stained with uranyl acetate and lead citrate.

For the scanning electron microscopical study
the abdomen of which the reproductive organs
were exposed, was detached from the body and
immersed in an ice-cold solution of 2.5% glutaral¬
dehyde in 0.1 M phosphate buffer (pH 7.3) for 12
hr. It was washed and trimmed in the same buffer
to uncover the nerve branches running from TAG
to the reproductive organs. This preparation was
post-fixed with 1% 0s04 buffered in the same
buffer for 3 hr, and dehydrated through an ethanol
series, then dried in a C02 critical-point dryer.
After coating with platinum palladium, the prepa¬
ration was examined in a Hitachi S 570.

For the transmission electron microscopical
study the reproductive organs were isolated care¬
fully with TAG from the body. This preparation

was fixed with an ice-cold 2.5% solution of glutar¬
aldehyde in 0.1 M phosphate buffer (pH 7.3) for 6
hr without cutting into small pieces. The prepara¬
tion was washed with the same buffer and post-
fixed in buffered 1% Os04 for 2 hr. After de¬
hydration in an ethanol series, the preparation was
embedded in epoxy resin. Semiserial thick (10
fjm) sections were cut and mounted on glass slides.
After examination with light microscope to recon¬
struct the nerves innervating the reproductive
organs, appropriate sections were reembedded
and thin sections were cut for electron microscopy.
In some cases, the preparation was cut into small
pieces after fixation with buffered same fixative for
30 min, and fixed in renewed fixative for 2 hr.
These pieces were post-fixed and embedded as
described above. All thin sections were stained
with uranyl acetate and lead citrate and examined
in a Hitachi H 500.

RESULTS

Gross anatomy

AGs of male cricket comprise more than 600
slender tubules, and each gland tubule opens into
the lobed anterior part of ejaculatory duct (ED).
A pair of epididymides (Eps), which are the com¬
pactly coiled parts of vasa deferentia, also open
into there. These reproductive organs are located
in the mass just over TAG (Fig. 1 A). Each tubule
of AGs is surrounded by a thin muscle layer which
consists of striated muscle fibers arranged in a
circular pattern around the longer axis of tubule
(Fig. 2A). This muscular layer is only one myofi¬
bril thick for most of the length of the tubule, but it
becomes multifibrillar toward the junction with the
anterior end of ED (Fig. 2B). Eps also are wound
around by multilayered muscle fibers. The ante¬
rior part of ED is enveloped by a thick muscular
layer, but the remaining parts by an irregular
network of muscle fibers.

AGs, Eps and the anterior part of ED are
innervated by paired nerve branches (Br3s) emerg¬
ing from the left and right R7s of TAG (Fig. IB).
Each R7 arises closely to the eighth nerve root
(R8, cereal nerve) from the postero-lateral margin
of TAG, and soon after running under the R8

Reproductive Organs in Male Cricket

769

Fig. 1. Scanning electron micrographs of the male reproductive organs and their innervation. (A) Dorsal view of the
terminal abdominal ganglion (TAG) and reproductive organs; X31. In this preparation, the accessory glands
(AGs) were drawn back to expose TAG. Arrow heads indicate the left and right nerve branches innervating the
reproductive organs. (B) A higher magnification view of the left nerve branches shown in A; x82. Note that the
seventh nerve root (R7) emerging from TAG is divided into four nerve branches (Brl-Br4). (C) Dorsal view of
Br3s innervating AGs; X33. In this preparation, the Br3s were exposed by removing the gland tubules just
covering them and their neighborhood. Note that the left and right Br3s are joined to form a loop (arrow head).
(D) A higher magnification view of the loop in C; X 100. Two nerve twigs (large arrow heads) with fine twiglets
(small arrow heads) emerge from the loop. Ep, epididymis; R8, eighth nerve root; SS, spermatophore sac; Tr,
trachea; VD, vas deferens.

toward the posterior, it bifurcates: one branch
(Brl) innervates the cereal muscles, and the other
(Br2) divides further into two branches (Br3 and
Br4). The Br3 runs into the mass of the reproduc¬
tive organs lying over TAG, and is joined with the

contralateral Br3, so that both Br3s form a loop
close to where paired Eps open to the anterior end
of ED (Fig. 1C). Two thick nerve twigs spreading
many thin twiglets arise from the region corre¬
sponding to the junction of both Br3s in the loop.

770

K. Yasuyama, T. Kimura and T. Yamaguchi

Fig. 2. Polarized-light microphotographs of the muscle fibers. (A) The striated muscle fibers winding around an
accessory gland tubule; X 320. This microphotograph was taken after the secretions contained in the tubule were
squeezed out by the contraction of tubule which was induced by application of high K+ saline (in mM/1: NaCl, 70;
KC1, 90; CaCl2, 5; buffered with 10 mM/1 Tris to pH 7.2). (B) A cross section of the lobed anterior part of
ejaculatory duct on which the orifices of AGs are located; Xl60.

and they intrude into the muscle layers of the
reproductive organs (Fig. ID). The Br4 reaches
the muscular wall of spermatophore sac.

Nickel back-filling through unilateral Br3 re¬

vealed one lateral and three (first-third) median
clusters of somata in the posterior half of TAG
(Fig. 3). The lateral cluster lies ipsilateral to the
back-filled nerve branch around the site where R7

Reproductive Organs in Male Cricket

771

Fig. 3. Light microphotographs of TAG back-filled with nickel through the right Br3. (A) Dorsal view (montage);
X90. (B) Caudal view; X90. Each arrow head indicates the secondary neurites bifurcating from the primary
neurites of DUMR7 neurons and extending toward the left and right R7s. Black arrows represent the lateral
cluster of somata of LC neurons lying near the site where the right R7 emerges from the ganglion. A white arrow
shows the soma of SR neuron which is located separately from the clusters.

leaves TAG, and consists of 13 somata with the
diameters ranging from 6 to 35 px n. The axons of
the neurons having their somata in this cluster
reach the mass of reproductive organs. This type
of neurons is called “LC neurons”. The first and
second median clusters are located near the dorsal
midline of TAG and the third median cluster
occupies rather a large area near the caudal mid¬
line (Fig. 3B). The positions of these median
clusters correspond to the embryonic segments 8-
11, respectively [10], The features of soma posi¬

tion and profile of neurites show that the neurons
having their somata in the median clusters are
identical with DUMR7 neurons as described by
Yamaguchi et al. [5]. The closer the median cluster
is to the posterior end of TAG, the larger the total
number of somata is in the median cluster. This
number, however, varies in different preparations:
4 to 7 in the first median cluster, 26 to 43 in the
second median cluster and 69 to 89 in the third
median cluster. The diameters of somata range
from 17 to 42 pm in the first median cluster, from 8

772

K. Yasuyama, T. Kimura and T. Yamaguchi

to 46 /j.m in the second median cluster and from 4
to 46 /jm in the third median cluster.

On the other hand, nickel back-filling through
unilateral Br4 showed the presence of a neuron
having its soma (45 fim. in diameter) which is
located in seclusion near the third median cluster
(Fig. 3B). The pathway of this axon is rather
complex: after leaving TAG the axon runs through
the Br3 and loop, passes the contralateral Br3 and
Br4, and then, terminates at one of the retractor
muscles of the spermatophore sac. This neuron is
called “SR neuron”. It is probable that the retrac¬
tor muscle is homologous to the muscle 117 iden¬
tified by DuPorte [11]. That is, it originates from
the anterior end of the ventral valve at its junction
with the ductus, runs obliquely round the sac and is
inserted into the lateral side of the dorsal valve at
the posterior end where the chitinous plates are
given off.

Fine structure of HRP -labeled DUMR7 neurons

HRP applied to a severed Br3 was transported
to the contralateral axons of DUMR7 neurons
beyond their somata within TAG. Figure 4 shows
a HRP-filled soma with its primary neurite in the
second cluster and the labeled axons running
through the contralateral Br3 and projecting via
the site, where both Br3s are joined by their
appearances, into AGs.

In the somata of labeled DUMR7 neurons there
were many mitochondria, ribosomes and Golgi
bodies. Large vesicles (LVs, 100-150 nm in dia¬
meter) were present associated with Golgi body
(Fig. 5A). Similar vesicles could be found even in
the primary and secondary neurites within TAG
(Fig. 5B, C), and also in the peripheral axons
projecting into the mass of reproductive organs
(Fig. 5D). These LVs seem to be produced in the

Fig. 4. Light microphotographs of DUMR7 neurons labeled with HRP through the right Br3. (A) A transverse
section through TAG showing the soma and its primary neurite (arrow) of a labeled DUMR7 neuron; X200. (B)
A transverse section through TAG; X200. Arrows indicate the secondary neurites. (C) A transverse section of
the left Br3 carrying labeled axons (arrows); X545. This section was made at the loop where the left and right
Br3s were joined together. (D) A longitudinal section of a nerve twig arising from the loop; X310. Each arrow
indicates the labeled axon running through the nerve twig.

Reproductive Organs in Male Cricket

773

Fig. 5. Electron microphotographs of HRP-labeled DUMR7. (A) The cytoplasm of labeled DUMR7 neuron soma;
X 12,500. Arrows indicate vesicles which appear to be associated with Golgi bodies (G). (B) A labeled primary
neurite (asterisk) within TAG; X 12,500. The arrow shows a vesicle. (C) A labeled secondary neurite (asterisk)
within TAG; X 12,500. Each arrow indicates a vesicle. (D) A labeled axon (asterisk) passing through the nerve
twig shown in Fig. 4D; X 12,500. This axon also has a vesicle (arrow).

somata and transported from there to the
periphery for local storage and release. To confirm
this presumption, we made short-term ligature
experiments of Br3.

One of paired Br3s was tied by double knots
using a silk thread in the neighborhood of the loop
of Br3s. The other Br3 was not tied as a control.
The preparation was fixed after 6 hr of ligation,
and then it was prepared for electron microscopy.
In the control Br3, the distribution of LVs in the
axons remained unchanged (Fig. 6A), whereas in

the ligated Br3 numerous LVs accumulated in the
cytoplasm of axons proximal to the ligature (Fig.
6B). A cross section of Br3 which was made from
an intact animal showed about 260 transverse
profiles of axons with the diameters ranging from
0.2 to 3.2 jum. Accumulation of LVs in ligated
axons of Br3 was seen in more than 20 axons with
relatively large diameters. It is, therefore, prob¬
able that DUMR7 neurons produce LVs in their
somata and transport them bilaterally to the
periphery.

774

K. Yasuyama, T. Kimura and T. Yamaguchi

Fig. 6. Transverse sections of the nerve branches, Br3s. (A) The non-ligated (control) Br3 contralateral to the
ligated Br3 shown in B; X7,000. (B) The ligated Br3; X 14,000. Note that the accumulation of vesicles is seen in
the axons (ax) with relatively large diameter.

Reproductive Organs in Male Cricket

775

Axon terminals on the muscle fibers of reproductive
organs

The axons containing LVs are widely distributed
along the muscle fibers associated with AGs (Figs.
8, 9), Eps (Fig. 11) and anterior part of ED (Fig.
10), and make neuromuscular junctions with them.
Some axons which have LVs are ensheathed by
Schwann cells and sent from Br3 to the muscular
layer (Fig 7A). Small spherical vesicles (SVs, 40-
60 nm) besides LVs appear in the cytoplasm of
axons near the neuromuscular junctions (Fig. 7B).

At the axon terminal, Schwann cells are absent
partially, and the axon makes synaptic contact with
the muscle fiber (Fig. 8). The neuromuscular
junction is distinguished both by dense postsynap-
tic membrane and by a cluster of SVs at the
presynaptic membrane. The aggregation of LVs is
never found in the axon terminals. This type of
neuromuscular junction is observed in every mus¬
cle layers of AGs, Eps (Fig. 11) and anterior part
of ED (Fig. 10).

The region of the muscle fiber making synaptic
junction with the axon is in the form of extension

Fig. 7. Axons which are distributed among the accessory gland tubules. (A) A longitudinal section of an axon
passing through a thick nerve twig which emerges from the loop of Br3s; X 15,000. Arrows indicate vesicles
containing in the cytoplasm. (B) An oblique section of a thin nerve twiglet near the muscle fibers (mf) of
accessory gland tubules; X 16,000. Note that both small and large vesicles are visible in the cytoplasm of axons
(ax).

776

K. Yasuyama, T. Kimura and T. Yamaguchi

Fig. 8. (A, B). An axon terminal making the synaptic contact (arrows) with muscle fiber (mf) which envelops the

accessory gland tubule; X 27,000. Both A and B show the same axon terminal (ax) in different sections. Note
that the axon terminal contains many large vesicles as well as small vesicles; the latter ones cluster along the
presynaptic membrane, gc, epithelial cell of accessory gland tubule.

from the main body of muscle fiber. The extension
often contains granular sarcoplasm and scarce con¬
tractile filaments (Fig. 8). These are prominent
feature at the multifibrillar layers enveloping the
proximal part of accessory gland tubules where
they open to the anterior end of ED (Fig. 9). In
this region, the extension of muscle fiber forms a
complex series of folds about the axon terminal.

Unfortunately we could not succeed in trans¬
porting HRP as far as the axon terminals under our
experimental procedure. These axon terminals,
however, may be identical with those of DUMR7
neurons, because LVs were present in the axons of
HRP labeled DUMR7 neurons (Fig. 5D) as well as
in their somata (Fig. 5A). The axons which do not
bear LVs are also distributed to the musculature of
the massed reproductive organs. The identified
LC neurons may contribute to these axons. The
investigation of this possibility must be the subject

of future research.

DISCUSSION

Nickel back-filling of Br3 revealed the presence
of 13 bilaterally paired LC neurons and 100 or
more dorsal unpaired median neurons (DUMR7
neurons). All these neurons project into the
muscle fibers of the internal reproductive organs,
AGs, Eps and ED (Figs. 8-11). In addition, the
nickel back-filling of Br4 showed the presence of
SR neuron innervating the contralateral retractor
muscle of spermatophore sac.

Odhiambo [2] described that the individual AG
of the male desert locust has highly developed
muscular layer with innervation, and also that
innervating axons have synaptic vesicles at their
junctional regions. In his observations, however,
no large secretory vesicles were found in any of the

Reproductive Organs in Male Cricket

777

Fig. 9. An axon terminal within the proximal part of the accessory gland tubule, where multilayered muscle fibers
are located; X 29,000. Arrows indicate the dense postsynaptic membranes. Note that large and small vesicles are
present within the axon terminal (ax), and the extension of muscle fiber (mf) forms a complex series of folds about
the axon terminal.

axons attached to the locust AGs. In the present
experiment it was found that the axons containing
LVs widely distributed along the muscle fibers of
the reproductive organs (Fig. 7), and they make
neuromuscular junctions with the muscle fibers,
where LVs coexist with SVs clustering especially at
the presynaptic membranes (Fig. 8). Probably the
terminal of these axons seems to be identical with
those of the axons of DUMR7 neurons, since
vesicles similar to the LVs in appearance are
present in the peripheral axons of HRP-labeled
DUMR7 neurons as well as in their somata (Fig.
5A, D) and the ligation of a Br3 carrying the axons
of DUMR7 neurons reveals remarkable accumula¬
tion of the same type of vesicles as the LVs in the
cytoplasm of axons proximal to the ligature (Fig.
6B). Further, these facts suggest that DUMR7
neurons produce LVs in their somata and trans¬
port them bilaterally to the periphery.

On the other hand, the electron microscopical
observation showed also that there exist the axons
containing only SVs on the muscle fibers of repro¬
ductive organs. These axons may be derived from
LC neurons. It is probable that the muscle fibers
of reproductive organs of male cricket are inner¬
vated by at least two types of neurons, i.e.,
DUMR7 and LC neurons originating from TAG.

Dorsal unpaired medial neurons (DUM
neurons) exhibiting the characteristic bilateral
morphology have been described in several insect
species [e.g., 12-21], since first description of them
in locust by Plotnikova [22]. Among these DUM
neurons, the DUMETi innervating to the extensor
tibiae muscle of locust is thought to release trans¬
mitter adjacent to the muscle fibers without form¬
ing an anatomically specific neuromuscular junc¬
tion, although its preterminal fine branches con¬
tain numerous large dense-core vesicles [23, 24]

778

K. Yasuyama, T. Kimura and T. Yamaguchi

Fig. 10. An axon terminal within the anterior part of the ejaculatory duct where both the accessory gland tubules and
epididymis open; X 29,000. Large and small vesicles are also present in this axon terminal (ax). Numerous small
vesicles cluster together at the presynaptic membrane (arrow), mf, muscle fiber.

and it contains the biogenic amine, octopamine to
modulate the neuromuscular transmission [25-27].
In contrast, the DUM neurons in TAG of firefly
have the primary function of direct initiation and
regulation of an effector response, bioluminescent
response of the lantern through the release of
octopamine which serves a more direct role as a
neurotransmitter than that postulated for its mod¬
ulatory and hormonal functions in other arthropod
systems [19, 28, 29].

In this experiment, it was evident that there
coexist two types of vesicles, LVs and SVs in the
axons which are presumed to be those of DUMR7
neurons: at the axon terminals forming the direct
synaptic junctions on the muscle fibers, SVs are
distributed throughout the profile and aggregate at
the presynaptic membrane, but LVs do not aggre¬
gate near the presynaptic membrane. This fact
suggests the possibility of dual functions of
DUMR7 neurons serving the control of move¬
ments of reproductive organs which are involved in
the spermatophore formation. Recently, we have
found that either of glutamate (10~5 M) and the
neuropeptide proctolin (10-loM) induces a dis¬
tinct tonic contraction of AGs, and further that a
certain amount of proctolin is detectable in the

surrounding medium of AGs while the muscles in
AGs contract tonically in response to high K+
saline [30; Kimura, Yasuyama and Yamaguchi, in
preparation]. It is quite conclusive, therefore, that
DUMR7 neurons are at least proctolinergic,
although a function of proctolin as neurotransmit¬
ter or neuromodulator in the DUMR7 neurons-
AGs system remains to be determined. In this
connection, Agricola et al. [31] reported that TAG
of cockroach contains numerous proctolinergic
fibers whose terminals contain large dense gra¬
nules (140 to 150 nm) and numerous clear vesicles
(40 to 50 nm), and proctolin immunostaining is
visible at the core of large dense granules. This
fact suggests the possibility that LVs may contain
proctolin.

On the other hand, when the electrical stimulus
is antidromically applied to a Br3, AGs exhibit a
monophasic mechanical response (contraction),
and when it is orthodromically applied to that,
AGs show biphasic mechanical response: the first
phase corresponds to the contraction and the
second to a decrease in basal tonus (Kimura,
Yasuyama and Yamaguchi, in preparation).
Therefore it is likely that DUMR7 neurons extend¬
ing their axons through both Br3s are excitatory or

Reproductive Organs in Male Cricket

779

Fig. 11. A transverse section of the epididymis; X 29,000. The epithelial cells of epididymis (ec) are enveloped by
multilayered muscle fibers (mf). Note that an axon terminal (ax) containing both large and small vesicles makes
synaptic contact (arrow) with the muscle fiber.

modulatory neurons, while LC neurons running
only through a Br3 to AGs are inhibitory neurons.
Further experiments on the functional role of
individual neurons innervating the internal repro¬
ductive organs in the spermatophore formation
and courtship behavior are being carried out.

ACKNOWLEDGMENTS

This work was supported in part by a Grant in Aid
from Ministry of Education, Science and Culture of
Japan to T. Y. for scientific research.

REFERENCES

1 Loher, W. (1974) Circadian control of sperma¬
tophore formation in the cricket Teleogryllus corn-
modus Walker. J. Insect Physiol., 20: 1155-1172.

2 Odhiambo, T. R. (1970) The architecture of the
accessory reproductive glands of the male desert
locust. 3. Components of the muscular wall. Tissue
and Cell, 2: 233-248.

3 De Loof, A. and Lagasse, A. (1972) The ultrastruc¬
ture of the male accessory reproductive glands of the
Corolado beetle. Z. Zellforsch., 130: 545-552.

780

K. Yasuyama, T. Kimura and T. Yamaguchi

4 Adiyodi, R. G. and Adiyodi, K. G. (1974) Ultra¬
structure of the utriculi majores in the mushroom¬
shaped male accessory gland of Periplaneta america¬
na (L.). Z. Zellforsch., 147: 433-440.

5 Yamaguchi, T., Kushiro, N. and Waki, T. (1985)
Sexual dimorphism of the terminal abdominal gan¬
glion of the cricket. Naturwiss., 72: 153-154.

6 Kimura, T., Yamaguchi, T. and Yasuyama, K.
(1984) Movement of male accessory glands and its
control by dorsal unpaired median neurons. Zool.
Sci., 1: 883.

7 Yasuyama, K., Kimura, T. and Yamaguchi, T.
(1985) Fine structure of the dorsal unpaired median
neurons innervating the muscle fibers of the acces¬
sory glands, epididymis, and ejaculatory duct of
male cricket ( Gryllus bimaculatus) . Doubutsu Seiri,
2: 173.

8 Levine, R. B. and Murphey, R. K. (1980) Pre- and
postsynaptic inhibition of identified giant inter-
neuons in the cricket (Acheta domesticus). J. Comp.
Physiol., 135: 269-282.

9 Bacon, J. P. and Altman, J. S. (1977) A silver in¬
tensification method for cobalt-filled neurones in
wholemount preparations. Brain Res., 138: 359-
363.

10 Panov, A. (1966) Correlations in the ontogenetic
development of the central nervous system in the
house cricket Gryllus domesticus L. and the mole
cricket Gryllotalpa gryllotalpa L. (Orthoptera, Gryl-
loidea). Entomol. Rev., 45: 179-185.

11 DuPorte, M. (1920) The muscle system of Gryllus
assimilis Fabr. ( = Pennsylvanicus Barum.). Ann.
Ent. Soc. Am., 3: 16-52.

12 Crossman, A. R., Kerkut, G. A., Pitman, R. M.
and Walker, R. J. (1971) Electrically excitable
nerve cell bodies in the central ganglia of two insect
species Periplaneta americana and Schistocerca gre-
garia. Comp. Biochem. Physiol., 40A: 594-597.

13 Crossman, A. R., Kerkut, G. A. and Walker, R. J.
(1971) Axon pathways of electrically excitable cell
bodies in the insect nervous system. J. Physiol., 218:
55-56.

14 Bentley, D. R. (1973) Postembryonic development
of insect motor systems. In “Developmental Neuro¬
biology of Arthropods”. Ed. by D. Young, Cam¬
bridge Univ. Press, Cambridge, pp. 147-177.

15 Hoyle, G., Dagan, D., Moberly, B. and Colqu-
houn, W. (1974) Dorsal unparied median insect
neurons make neurosecretory endings on skeletal
muscle. J. Exp. Zool., 187: 159-165.

16 Clark, R. (1976) Structural and functional changes
in an identified cricket neurons after separation from
the soma. I: Structural changes. J. Comp. Neurol.,
170: 253-256.

17 Casaday, G. B. and Camhi, J. M. (1976) Meta¬

morphosis of flight motor neurons in the moth
Manduca sexta. J. Comp. Physiol., 112: 143-158.
Davis, N. T. (1977) Motor neurons of the indirect
flight muscles of Dysdercus fulvoniger. Ann. En¬
tomol. Soc. Am., 70: 377-386.

Christensen, T. A. and Carlson, A. D. (1981) Sym¬
metrically organized dorsal unpaired median
(DUM) neurons and flash control in the male firefly,
Photuris versicolor. J. Exp. Biol., 93: 133-147.
Lange, A. B. and Orchard, I. (1984) Dorsal un¬
paired median neurons, and ventral bilaterally
paired neurons, project to a visceral muscle in an
insect. J. Neurobiol., 15: 441-453.

Watson, A. H. D. (1984) The dorsal unpaired me¬
dian neurons of the locust metathoracic ganglion:
neuronal structure and diversity, and synapse dis¬
tribution. J. Neurocytol., 13: 303-327.

Plotnikova, S. I. (1969) Effector neurons with
several axons in the ventral nerve cord of the Asian
grasshopper Locusta migratoria. J. Evol. Biochem.
Physiol., 5: 276-278.

Hoyle, G. (1978) The dorsal, unpaired, median
neurons of the locust metathoracic ganglion. J.
Neurobiol., 9: 43-57.

Hoyle, G., Colquhoun, W. and Williams, M. (1980)
Fine structure of an octopaminergic neuron and its
terminals. J. Neurobiol., 11: 103-126.

Evans, P. D. and O’Shea, M. (1977) The identifica¬
tion of an octopaminergic neurone which modulates
neuromuscular transmission in the locust. Nature,
270: 257-259.

Evans, P. D. and O’Shea, M. (1978) The identifica¬
tion of an octopaminergic neurone and the modula¬
tion of a myogenic rhythm in the locust. J. Exp.
Biol., 73: 235-260.

Evans, P. D. and Siegler, M. V. S. (1982) Octopa-
mine mediated relaxation of maintained and catch
tension in locust skeletal muscle. J. Physiol., 324:
93-112.

Christensen, T. A. and Carlson, A. D. (1982) The
neurophysiology of larval firefly luminescence:
direct activation through four bifurcating (DUM)
neurons. J. Comp. Physiol., 148: 503-514.
Christensen, T. A., Sherman, T. G., McCaman, R.
E. and Carlson, A. D. (1983) Presence of octopa-
mine in firefly photomotor neurons. Neuroscience,
9: 183-189.

Kimura, T. and Yamaguchi, T. (1985) Presence of
proctolin in the male accessory reproductive glands
of cricket and its function. Zool. Sci., 2: 860.
Agricola, H., Eckert, M., Ude,J., Birkenbeil, H.
and Penzlin, H. (1985) The distribution of a procto-
lin-like immunoreactive material in the terminal
ganglion of the cockroach, Periplaneta americana L.
Cell Tissue Res., 239: 203-209.

18

19

20

21

22

23

24

25

26

27

28

29

30

31

ZOOLOGICAL SCIENCE 5: 781-790 (1988)

© 1988 Zoological Society of Japan

Functional Innervation of the Intrinsic Thumb Muscles
of the Fruit Bat, P ter opus medius

Khin Maung Saing* 1

Department of Zoology, University of Rangoon, Rangoon, Burma

ABSTRACT — The thumb of the fruit bat was found to be well developed. Three muscles, i.e., flexor
pollicis brevis, adductor pollicis, abductor pollicis brevis were found and their attachments noted. The
muscles were supplied by metacarpal nerve I derived from a single medio-ulnar combined nerve trunk.
The diffuse type of motor endplates constant in size and shape, as well as complex and intermediate
types of neuromuscular spindles, were present in all three muscles. Encapsulated Golgi tendon organs
and Pacinian corpuscles were found profusely in all the muscles. Three muscle fibre types, i.e., red,
white and intermediate based on lipid content were observed in the flexor pollicis brevis. Correla-
tionship of function to the structures in the fruit bat’s short thumb muscles is discussed.

INTRODUCTION

Fingers of the bat are well developed for sup¬
porting the wing membrane, but the thumb has
degenerated into a small process with a claw on the
wing. Such a vestigial thumb is also seen in many
digitigrade mammals as an example of the degen¬
eration in less used organs [1].

In the insectivorous bats (suborder: Microchi-
roptera), the thumbs are most vestigial [2], but in
the fruit bats (suborder: Megachiroptera), they are
comparatively large and useful, because they are
used for clutching the tree branches when the fruit
bats move from branch to branch. Therefore, the
thumbs of the fruit bats seem to retain some
primitive or ancestral pattern. The presence of
various motor and sensory endings has been re¬
ported in striated muscles of the mammals in
general [3-5]. Muscle fibre types different in
color, morphology, lipid content as well as other
properties have been reported in mammals [6-10]
and also in the flight muscles of some bats [11-13].
The relationship between structure and function
concerning the types of motor endings, sensory
endings and muscle fibres has also been widely
studied [14-16]. However, the features and func¬

Accepted November 11, 1987
Received September 3, 1987

1 Present address: Institute of Biological Sciences,
University of Tsukuba, Tsukuba, Ibaraki 305, Japan.

tions of the fruit bat thumbs have never been
studied from such a view-point.

The present work was, therefore, planned in
order to elucidate muscular features and innerva¬
tion patterns in the thumb of the Burman fruit bat,
Pteropus medius, as a primitive model of the bat
thumbs, and to build a foundation for the future
works on the degeneration of structures and func¬
tions in a series of homologous organs with vesti¬
gial counterparts.

MATERIALS AND METHODS

The adult fruit bats ( Pteropus medius) of both
sexes were killed with chloroform and perfused
with neutral formalin (10%) for 3 days. The right
and left forelimbs were then dissected, and the
gross attachments and innervations of the three
short thumb muscles i.e. the flexor pollicis brevis
(FLX), abductor pollicis brevis (ABD) and the
adductor pollicis (ADD) were studied under a
binocular light microscope. Muscle nomenclature
was adopted from Vaughan [2]. For neurohistolo¬
gy, the muscles were dissected out from freshly
killed animals, and were immediately fixed and
stained according to the modified silver nitrate
technique of Barker and Ip [17] for producing
muscle and nerve preparations [18]. Paraffin cross-
sections (10 pm) of all three muscles were made
and stained for neuromuscular spindles with

782

Khin Maung Saing

Weigert’s iron haematoxylin and van Gieson
method [19]. For muscle fibre types, the muscles
were freshly dissected out and fresh frozen cross-
sections (10 p m) were prepared in a cryostat at
— 20°C and stained for lipids with Sudan black B
[20]. All the cross-sections and teased prepara¬
tions were examined under a light microscope and
photomicrographs were taken at 400 X or 100 X
magnifications.

RESULTS

Gross pattern of attachments of the three short
thumb muscles

Three intrinsic thumb muscles, that is, flexor
pollicis brevis (FLX), abductor pollicis brevis

(ABD) and adductor pollicis (ADD) were present
in the hand (Fig. 1). The FLX originated on
medial surface of the fused proximal first and
second carpal bones, and inserted medially on base
of the proximal phalanx of the thumb. The ABD
originated on the scaphoulnar carpal bone and
inserted into lateral half of base of the proximal
phalanx. The ADD arose from the medio-anterior
margin of the ulnar carpal bone and the base of the
second metacarpal bone, and the insertion was on
ulnar-medial surface of the proximal phalanx.

Gross pattern of innervations of the three short
muscles

The medium and the ulnar nerves, derived from
the C8 to T1 spinal nerves in the brachial plexus,
became fused into a single nerve trunk (emu) and

V m4

1 cm

Fig. 1. Ventral view of the hand of the fruit bat showing the three intrinsic thumb muscles and their innervation.
FLX: flexor pollicis brevis, ADD: adductor pollicis, ABD: abductor pollicis brevis, int: interossei muscles, emu:
combined medio-ulnar nerve trunk, mj-nis: metacarpal nerves, nerve branchlets of mj supplying the

FLX, ADD and ABD, ins: insertions and ori: origins of the intrinsic thumb muscles, I-V: proximal phalanges of
the digits one to five, rb: radius bone, wmb: wing membrane.

Innervation of Fruit Bat Short Thumb Muscles

783

ran distally on the medial side of the radial bone
(rb). In the carpal region the combined nerve
(emu) sent off five metacarpal nerves (Fig. 1, mr
m5). The palmar branch of the first metacarpal
nerve (mi) gave off three branchlets (mn1? mn2
and mn3) supplying the ABD, FLX and ADD,
respectively (Fig. 1).

Motor endplates (MEP)

Diffuse type of motor endplates generally con¬
stant in size and terminal arborizations were
observed in all three muscles. They were situated
in the mid-belly regions across the muscles in
motor endplate zones. Generally, an intramuscu¬
lar nerve trunk traversed the muscle fibres at a
right angle giving off many single nerve axons
along the way. Typically, each axon approached
each muscle fibre either obliquely or at a right
angle or in parallel, and ended as a single MEP on
each muscle fibre (Fig. 2A). Double “MEP”s on a
single muscle fibre supplied by different axons
were also noted (Fig. 2 A, two free arrows). Occa¬
sionally, a single axon divided into two branches,
each branch terminating as a MEP containing thick
teloglial cytoplasm on different muscle fibre (Fig.
2B). Rarely, two axons (ep2) supplied a single
MEP on a single muscle fibre (Fig. 2C).

Neuromuscular spindles (MSP)

Complex and intermediate types of MSPs were
profusely found in all three muscles. The complex
type (Fig. 3 A and B) had a long and thick capsule
(c) encasing in average 7 to 10 intrafusal muscle
fibres (although difficult to see in the microphoto¬
graph, the intrafusal muscle fibres were exactly
observed and counted carefully under a light
microscope). An intra-muscular nerve plexus sup¬
plied the MSP. An afferent nerve fibre with large
diameter ended as a flower-spray type of primary
sensory ending (p) among all the intrafusal muscle
fibres in the equatorial zone. Two secondary
sensory endings (s) blunt in shape were supplied
separately by two afferent nerve fibres with large
diameter and were situated in the juxtaequatorial
zone on either side of the primary ending. Two
fine fusimotor nerve fibres formed a diffuse net¬
work (dn) in one mid-polar region with the net¬
work ending as a single small diffuse type of MEP

(ep) in the polar region. Two other fusimotor
nerve fibres entered the other mid-polar region,
forming a diffuse network (dn) and a trail motor
ending (te) in the mid-polar and polar regions,
respectively.

The intermediate type of MSP (Fig. 4A and B)
had a long but thin capsule (c) encasing in average
four intrafusal muscle fibres. The spindle had an
annulo-spiral type of primary sensory ending (p) in
the equatorial zone. A secondary sensory ending
(s) with simple, forked blunt terminations was
present adjacent to the primary ending. Two
afferent nerve fibres with large diameter accompa¬
nied by six fusimotor fine nerve fibres entered the
spindle capsule at the equatorial region. The
primary and the secondary sensory endings were
supplied by each of the large diameter afferents.
The fine fusimotor nerves formed diffused net¬
works (dn) in the mid-polar regions and ended as
small diffused type of motor endplates (ep) in the
polar regions.

One cross-section of a MSP in the ABD muscle
had a thick and circular capsule (c) with the
intracapsular space (s) divided into three compart¬
ments by collagenous septa (Fig. 5, arrowed). The
compartments were occupied by intrafusal muscle
fibres consisted of four large nuclear bag type (mi)
and eighteen small nuclear chain type (m2). No
attempt had been made to differentiate the nuclear
bag type into nuclear bag! and nuclear bag2 cate¬
gories.

Golgi tendon organs (GO)

Numerous single encapsulated Golgi tendon
organs were situated at musculo-tendinous junc¬
tions (Fig. 6). A typical GO had a thick cylindrical
capsule (c) encasing four to five collagenous fibres
which were attached at the tips to extrafusal mus¬
cle fibres (m). The other end of the collagenous
fibres continued into the tendinous part (t). A
large myelinated afferent nerve fibre (nt) entered
the GO capsule at the mid-region and divided into
numerous branchlets ending as flower-sprays
among the collagenous fibres.

Pacinian corpuscles (PC)

Pacinian corpuscles were located near nerve
trunks in the fascia of the muscles (Fig. 7). Each

784

Khin Maung Saing

Innervation of Fruit Bat Short Thumb Muscles

785

Fig. 3. (A) Complex type of neuromuscular spindle in abductor pollicis brevis. Barker and Ip’s silver stain, p:

primary sensory ending, s: secondary sensory ending, dn: diffuse chain network of motor fibres, ep and te:
fusimotor endplate and trail ending, respectively at the polar regions, c: spindle capsule, n: nerve fibres
innervating the spindle. X 100. (B) Line drawing of (A). Intrafusal muscle fibres are shown only at the poles for
clarity.

Fig. 2. Diffuse type of motor endplates in the intrinsic thumb muscles of the fruit bat. Barker and Ip’s silver stain,
mep: motor endplate, m: muscle fibre, a: nerve axon, nt: nerve trunk, ep2: two axons supplying a single motor
endplate. (A) Flexor pollicis brevis; free arrows indicate double motor endplates on a single muscle fibre supplied
by different axons. X400. (B) Adductor pollicis; two motor endplates on different muscle fibres supplied by a
divided single axon. X400. (C) Abductor pollicis brevis; two axons supplied a single motor endplate on a single
muscle fibre. X400.

786

Khin Maung Saing

Fig. 4. (A) Intermediate type of neuromuscular spindle in adductor pollicis. Barker and Ip’s silver stain, nt: nerve

trunk. All other abbreviations are as in Fig. 3. XlOO. (B) Line drawing of (A). Intrafusal muscle fibres are
shown only at the poles for clarity.

Fig. 5. Cross-section of a neuromuscular spindle in flexor pollicis brevis. Weigert’s iron Haematoxylin and van
Gieson stain, m: extrafusal muscle fibres, c: capsule of the spindle, s: intra-capsular space, mi and m2: nuclear bag
and nuclear chain types, respectively of intra-fusal muscle fibres. Free arrows; intra-capsular septa. X400.

Fig. 6. An encapsulated Golgi tendon organ in abductor pollicis brevis. Barker and Ip’s silver stain, c: capsule, nt:
nerve supplying the tendon organ, m: muscle fibres. XlOO.

Innervation of Fruit Bat Short Thumb Muscles

787

788

Khin Maung Saing

capsule (cp) was an elongated oval shape with a
slightly tapered and rounded tip. A small diameter
afferent nerve fibre (n) entered the capsule at the
base and ran along in the central axis toward the
tip. The nerve fibre was covered by a dense layer
of endoneurium which in turn was covered by
squamous epithelium cells and connective tissue.

Fig. 7. Pacinian corpuscles in adductor pollicis. Barker
and Ip’s silver method, cp: capsule of the Pacinian
corpuscle, n: nerve axon innervating the corpuscle.
X100.

Fig. 8. Fresh-frozen cross-section of muscle fibres of
flexor pollicis brevis. Sudan black B stain, r: red
muscle fibre type, w: white muscle fibre type, i:
intermediate muscle fibre type. X400.

Striated muscle fibre types based on lipid content

In all three muscles, each muscle fasciculus was
composed entirely of either red (r), white (w) or
intermediate (i) type of muscle fibres according to
Sudan black B stain densities for lipids (Fig. 8).
The muscle fibre with the most sudanophilic prop¬
erties was assumed to be the red fibre type posses¬

sing high lipid and mitochondria contents. The
white and the intermediate types, low and in¬
termediate in Sudan black B densities, respective¬
ly, were also observed.

DISCUSSION

Based on the present works, it can be said that
the fruit bat intrinsic thumb muscles are profusely
supplied with motor endplates, neuromuscular
spindles, Golgi tendon organs and Pacinian cor¬
puscles as well as three types of muscle fibres (i.e.
red, white and intermediate) based on their lipid
contents reflecting the high functional state of the
fruit bat’s thumb. This is in agreement with the
presence of numerous motor and sensory endings
in the functional limb muscles of other mammals
[5, 21].

The presence of the three well developed intrin¬
sic thumb muscles in the fruit bat in this work is in
agreement with the findings of Vaughan [2] in the
insectivorous bat Eumops perotis. The three
thumb muscles were also described in the mor-
moopid and the leaf-nosed bats, but the muscles
were highly variable with some muscles very small
or missing in certain insectivorous species such as
Mormoops megalophylla and M. blainvillii [22].
The number of the muscles is the same in both the
Eumops and the fruit bat used in this work with
similar insertion points. However, the origins of
the muscles are placed more separately in the fruit
bat which would allow easier control of the thumb.
The fruit bat also uses the thumb in offense and
defense purposes and for locomotion in trees [23].
Moreover, the ulnar bone in the fruit bat’s forearm
is drastically reduced in length with only the radius
bone remaining well developed. This might cause
the fusion of the ulnar and the medium nerves into
a single trunk supplying the hand including the
thumb muscles. Reductive variation in the ulnar
bone is also the general trend among the insecti¬
vorous bats of the families Mormoopidae and
Phyllostomatidae [22].

Muscles associated with skilled activities have
extrafusal motor end-plates constant in size and
appearance, but muscles with continuous function
with short rests such as the diaphragm have irregu¬
lar motor endplates [14]. Tuffery [24] also found

Innervation of Fruit Bat Short Thumb Muscles

789

the same motor endplate structural variation in the
slow muscles of cat. The motor endplates of the
fruit bat short thumb muscles are all diffuse type,
with generally constant size and shape suggesting
skilled activity of the thumb. Some of the single
muscle fibres in the fruit bat must be long and fast
in contraction because they have two motor end-
plates supplied by two different axons (Fig. 2A).
This agrees with the statement that muscle fibres
with double motor endplates might indicate high
contraction speed and extra length in those fibres
[25]. A single axon does not innervate more than
two muscle fibres (Fig. 2B). Only one primary
sensory ending is present in all the spindles ex¬
amined in the fruit bat, with a single and double
secondary sensory endings present in the in¬
termediate and the complex types of spindles,
respectively. All the spindles are well developed
with long capsules, with trail and diffuse motor
endings in the polar regions. This reflects that the
fruit bat spindles would be involved mostly in tonic
response to muscle stretch and contraction as
reported in the snake muscles [15]. Numerous
encapsulated Golgi tendon organs in the fruit bat
thumb muscles indicate that the thumb is well
controlled in function. The tendon organ is a
stretch receptor measuring extrafusal motor end-
plate activity in active muscle contraction, and at
the same time measures passive stretch at the
tendon. The ‘triad’ consisting the neuromuscular
spindles, the tendon organ and the Pacinian cor¬
puscles was said to measure local muscle strength
and tension coordinately [26].

Red muscle fibres were reported to be rich in
lipids and generally involved in slow tonic contrac¬
tion and postural maintenance of the body, where¬
as white fibres were poor in lipids and involved in
first phasic locomotory actions [26]. All three
types of muscle fibres (red, white and intermedi¬
ate) based on sudanophilic properties in the fruit
bat’s thumb suggest that the thumb has skilled
activity capable of finely graded contractions, in
addition to slow and fast movements. The stain
density of Sudan black B was said to correspond to
triglycerides, lipids and mitochondria contents as
well as to succinate dehydrogenase activity in the
striated muscles of the mammals [6, 9, 12]. It was
reported that in the insectivorous little brown bat

{My otis lucifugus), all the muscle fibres in the
pectoralis flight muscle stained intensely by Sudan
black B for neutral fat while the animal was
hibernating in winter, but the fat content was
variable in summer when the bats were actively
flying, reflecting the nonfunctional and functional
states of the same muscle [11].

The similarity of the fruit bat thumb muscles’
functional state to actively functioning muscles of
other mammals is reflected in the presence of two
types of intrafusal muscle fibres (i.e. nuclear bag
and nuclear chain) in the spindles of the fruit bat
thumb muscles. This is in agreement with the
finding of three intrafusal muscle fibre types
(namely, nuclear bagi, nuclear bag2 and nuclear
chain) reported in other mammalian spindles [27],
although no attempt has been made to differenti¬
ate the nuclear bag intrafusal muscle fibres into
nuclear bagi and bag2 types in the present work.

ACKNOWLEDGMENTS

The author wishes to express his gratitude to Prof. K.
Yanagisawa for encouragement and Drs. T. Makioka
and Y. Tanaka for their useful discussions in the prepara¬
tion of this manuscript. The author is also indebted to
Mr. Hidefumi Orii for help in the preparation of figures
and microphotographs. This work was done as part of
M. Sc. thesis submitted at the University of Rangoon,
Burma.

REFERENCES

1 Romer, A. S. (1965) The Vertebrate Body, 3rd ed..
Shorter Version, W. B. Saunders Co., Philadelphia
and London, pp. 198-222.

2 Vaughan, T. A. (1970) The muscular system. In
“Biology of Bats. Vol. 1”. Ed. by W. A. Wimsatt,
Academic Press, New York and London, pp. 140-
194.

3 Zbigniew, L., Kowski, O. L. and Monocha, S. L.
(1973) Muscle spindle. In “The Structure and Func¬
tion of Muscle. Vol. II. Structure, Part 2”. Ed. by G.
H. Bourne, Academic Press, New York and Lon¬
don, pp. 366-474.

4 Couteaux, R. (1973) Motor end plate structure. In
“The Structure and Function of Muscle. Vol. II.
Structure, Part 2”. Ed. by G. H. Bourne, Academic
Press, New York and London, pp. 483-527.

5 Scott, J. A. A. and Young, H. (1987) The number
and distribution of muscle spindles and tendon
organs in the peroneal muscles of the cat. J. Anat.,

790

Khin Maung Saing

151: 143-155.

6 Padykula, H. A. and Gauthier, G. F. (1963)

Cytochemical studies of adenosine triphosphatases
in skeletal muscle fibres. J. Cell Biol., 18: 87-107.

7 Gauthier, G. F. and Padykula, H. A. (1966)

Cytochemical studies of fibre types in skeletal mus¬
cle. A comparative study of the mammalian di¬
aphragm. J. Cell Biol., 28: 333-354.

8 Davies, A. S. and Gunn, H. M. (1972) Histo-
chemical fibre types in the mammlian diaphragm. J.
Anat., 112: 41-60.

9 Burke, R. E., Levine, D. N., Tsairis, P. and Zjac
III, F. E. (1973) Physiological types and histo-
chemical profiles in motor units of the cat gastro¬
cnemius. J. Physiol., 234: 723-748.

10 Tulsi, R. S. (1975) Observations on the structure of
the dorsal muscle in the bottle-nose dolphin ( Tur -
siops truncatus ). J. Anat., 119: 39-48.

11 Armstrong, R. B., Ianuzzo, C. D. and Kunz, T. H.
(1977) Histochemical and biochemical properties of
flight muscle fibres in the little brown bat, Myotis
lucifugus. J. Comp. Physiol., 119: 141-154.

12 Hermanson, J. W. and Foehring, R. C. (1982) A-
natomy and histochemistry of flight muscles in free¬
tailed bats. Am. Zool., 22: A873.

13 Hermanson, J. W. and Foehring, R. C. (1985) His¬
tochemistry of the flight muscle in a phyllostomid
bat. Am. Zool., 25: A21.

14 Cole, W. V. (1957) Morphological characteristics of
the motor endplate of the rat. Anat. Rec., 98: 393-
400.

15 Fukami, Y. and Hunt, C. C. (1970) Tonic and phas¬
ic muscle spindles in snake. J. Neurophysiol., 33:
28-45.

16 Dias, P. L. R. (1979) Effects of tenotomy on the
motor endplates of fast and slow twitch muscles of
the rabbit. J. Anat., 129: 399-404.

17 Barker, D. and Ip, M. C. (1963) A silver method
for demonstrating the innervation of mammalian
muscle in teased preparations. J. Physiol., 69: 73-
74.

18 Barker, D., Scott, J. J. A. and Stacey, M. J. (1985)
Sensory reinnervation of cat peroneus brevis muscle
spindles after nerve crush. Brain Res., 333: 131—
138.

19 Lillie, R. D. (1958) Histopathologic Technic.
Blackiston, New York, pp. 193-196.

20 Ruthmann, A. (1970) Methods in Cell Research.
Cornell Univ. Press, Ithaca, New York, p. 176.

21 Dias, P. L. R. (1974) Surface area of motor end-
plate in fast and slow twitch muscles of the rabbits.
J. Anat., 117: 453-462.

22 Vaughan, T. A. and Bateman, G. C. (1970) Func¬
tional morphology of the forelimb of mormoopid
bats. J. Mammol., 51: 217-235.

23 Jepsen, G. L. (1970) Bat origins and evolution. In
“The Biology of Bats. Vol. 1”. Ed. by W. A.
Wimsatt, Academic Press, New York and London,

pp. 1-62.

24 Tuffery, A. R. (1971) Growth and degeneration of
motor endplates in normal cat hindlimb muscles. J.
Anat., 110: 221-247.

25 Grinyer, I. and George, J. C. (1965) An electron-
microscopic study of the pigeon breast muscle. Can.
J. Zool., 47: 517-523.

26 Cooper, S. (1960) Muscle spindles and other muscle
receptors. In “The Structure and Function of Mus¬
cle. Vol. 1, Structure”. Ed. by G. H. Bourne,
Academic Press, New York, pp. 381-420.

27 Banks, R. W., Harker, D. W. and Stacey, M. J.
(1977) A study of mammalian intrafusal muscle
fibres using a combined histochemical and ultra-
structual technique. J. Anat., 123: 783-796.

ZOOLOGICAL SCIENCE 5: 791-799 (1988)

© 1988 Zoological Society of Japan

Light and Electron Microscopic Observations of Holospora
obtusa : A Macronucleus-specific Bacterium
of the Ciliate Paramecium caudatum

Masahiro Fujishima and Kazumi Hoshide1

Biological Institute, Faculty of Science, and 1 Biological Institute,

Faculty of Education, Yamaguchi University, Yamaguchi 753, Japan

ABSTRACT — Holospora obtusa is a macronucleus-specific bacterium of the ciliate Paramecium
caudatum. Infectious form of this bacterium penetrates the macronuclear envelope via the food
vacuoles and grows exclusively in the nucleus, but never penetrates the micronuclear envelope.

In this work, structures of the infectious form of H. obtusa were examined by light and electron
microscopy. Phase-contrast microscopic observations showed that the infectious form of H. obtusa
consisted of a refractile part with a non-refractile tip and a dark-looking part. Since the bacterium
always penetrates the macronuclear envelope with the non-refractile tip but never with its other end,
this special tip seems to have functions to distinguish the nuclear envelopes of two kinds of nuclei,
macro- and micronucleus and penetrate only the macronuclear envelope. Fluorescence microscopy
showed that a DNA specific fluorochrome, 4', 6-diamidino-2-phenylindole (DAPI) bound only to the
refractile part. This indicates that the refractile part is a nucleoid of this bacterium. It should be noted
that two dotted small regions in the D API-positive part showed especially strong fluorescence.
Transmission electron microscopy showed that the non-refractile tip, the refractile part and the
dark-looking part were composed by an electron-translucent material, a homogeneous electron-dense
material and a typical bacterial cytoplasm, respectively. Scanning electron microscopy showed that the
bacterium has a smooth surface and has no remarkable difference at the both ends, but the one end
which corresponds to the non-refractile tip was always depressed when the bacteria were dried without
using a critical-point dryer, suggesting that the non-refractile tip has an inner-structural property to be
easily depressed by a surface tension.

INTRODUCTION

The gram-negative bacterium Holospora obtusa
is a macronucleus-specific symbiont of Para¬
mecium caudatum. The bacterium shows two
morphologically distinct forms during its life cycle:
an infectious long form (10-15 pm in length), a
reproductive short form (1.0- 1.5 pm in length)
and intermediate forms of various lengths between
the former two forms. The infectious long form is
observed predominantly in the nuclei of starved
host cells and the reproductive short form in those
of vegetatively growing cells. The intermediate
forms are observed in the cells at a transition phase
from log phase to stationary phase in growth, and
are also observed in those at an early infection

Accepted November 9, 1987
Received September 17, 1987

process. Only the infectious long form can pene¬
trate the host macronuclear envelope within 10
min, via the food vacuoles, when homogenate of
bacteria-bearing paramecia or purified bacteria are
added to bacteria-free paramecia [1-5]. On the
other hand, the reproductive short form and the
intermediate forms are always digested in the food
vacuoles so that they cannot infect the host nu¬
cleus.

This bacterium never penetrates the micronu¬
clear envelope. Because one end of the bacterium
can be distinguished from the other as being
non-refractile under the phase-contrast micro¬
scope, it has been observed that the infectious long
form of H. obtusa always penetrates the macronu¬
clear envelope with the non-refractile tip and
never with its other [4]. This observation strongly
suggests that an affinity between this special tip of
the bacterium and the host macronuclear envelope

792

M. Fujishima and K. Hoshide

is responsible for the nuclear specificity of the
infectivity. In other words, this special tip seems to
have functions for recognition of some difference
between the nuclear envelopes of macro- and
micronucleus and for penetration of the macronu-
clear envelope. Furthermore, it has been found
that although maintenance of H. obtusa in the host
macronucleus is achieved only in certain strains of
P. caudatum, the bacterium can infect the macro¬
nuclei not only of P. caudatum, but also of P.
aurelia species complex and P. multimicronu-
cleatum. The bacterium cannot infect the macro¬
nuclei of P. jenningsi, P. bursaria, P. trichium, P.
duboscqui, P. woodruffi, and P. calkinsi [4, 5].
This indicates that the infection and the mainte¬
nance of H. obtusa are separately controlled phe¬
nomena, and that the special tip of H. obtusa has
functions not only for recognition of the macro-
and the micronuclear membranes but also for that
of some difference of the macronuclear envelopes
of Paramecium species. The control mechanisms,
how the bacterium distinguishes the nuclear en¬
velopes of the two kinds of nuclei and penetrates
only the macronuclear envelope with its special
tip, are not yet clear.

The present study was intended to observe de¬
tails of the inner and the surface structures of the
infectious long form of H. obtusa, especially on its
special tip, by means of light and electron micros¬
copy.

MATERIALS AND METHODS

Strains and culture conditions

The cells used in this study were Paramecium
caudatum syngen 3, mating type V, strain 27aG3.
The original Holospora- bearing strain C101
(syngen unknown) was collected in Munster,
FRG, by Dr. H.-D. Gortz. Later, the strain 27aG3
was infected by H. obtusa and used for obtaining
the symbionts in this study. The culture medium

used was 1.25% (w/v) fresh lettuce juice in Dryl’s
solution [6] inoculated with a non-pathogenic
strain of Klebsiella pneumoniae for 1 day before
use [7]. In ordinary cultures several hundred cells
were inoculated into 2 ml of culture medium and
then 4 ml, 10 ml and 10 ml of fresh medium were
added on successive days. Cultures were kept at
25°C.

Light microscopy

Ordinary observation of the symbionts in host
macronuclei was made with a Nomarski-
differential-interference-contrast microscopy
(Olympus, BHS-N) at a magnification of X 1000 or
X400. Phase-contrast and fluorescent microscopy
was made as follows. Infectious forms of H. obtusa
isolated by squashing starved host cells were air-
dried on a glass slide, fixed in an ethanol/acetic
acid (3:1) mixture for 10 min and air-dried again.
Then, a DNA-binding fluorochrome, 4', 6-
diamidino-2-phenylindole (DAPI, 1 ^g/ml) dis¬
solved in deionized water was dropped onto the
preparations and covered with cover glasses.
Observations were made with phase-contrast and
fluorescent optics (Olympus, BH2-RFL) at a mag¬
nification of X400.

Electron microscopy

For transmission electron microscopy (TEM),
H. obtusa- bearing paramecia in late log phase
were fixed in 5% (v/v) glutaraldehyde in 0.2 M
cacodylate buffer (pH 7.2) for 1 hr at room
temperature, and washed with the same buffer.
After post-fixation in cold 1% (v/v) Os04 in the
cacodylate buffer for 1 hr, the cells were dehy¬
drated in a cold ethanol series and embedded in
Araldite. Thin sections obtained with an LKB
ultratome were stained with saturated aqueous
uranyl acetate and lead citrate, and observed with
a JEOL-100C transmission electron microscope.

For scanning electron microscopy (SEM), 250
ml of stationary phase cultures of H. obtusa-

Fig. 1. Photomicrographs of a living cell of P. caudatum with infectious long forms of H. obtusa in the macronucleus
(A), and infectious long forms (B and C). A: Nomarski-differential-interference-contrast. Many rod-shaped
bacteria can be seen in the macronucleus. B: Phase-contrast. Note that each infectious long form consists of a
dark-looking part (single arrowhead) and a refractile part (double arrowhead) with a non-refractile tip (triple
arrowhead). C: DAPI fluorescence of the bacteria in B. Note that the fluorescence appears only at the refractile
part of each bacterium, and strong fluorescence can be seen in two dots. Bars, 20 // m (A) and 10 /jrn (B and C).

Morphological Characters of Holospora

793

Fig. 1

794

M. Fujishima and K. Hoshide

bearing paramecia were harvested by centrifuging
at 380 X g for 3 min, homogenized by hand in a
Teflon homogenizer at 0-4°C, and fixed in 10
volumes of 5% (v/v) glutaraldehyde in 0.2 M
cacodylate buffer (pH 7.2) for 1 hr at 4°C. Then,
the homogenate was washed twice by centrifuging
at 880 X g for 20 min with the same buffer, fixed in
cold 1% (w/v) Os04 in the cacodylate buffer for 1
hr, dehydrated in a cold ethanol series and incu¬
bated in 100% (v/v) isoamyl acetate. The sample
was dried in a critical point dryer, coated with
gold, and observed with a Hitachi- Akashi Mini-
SEM scanning electron microscope. An aliquot of
the sample was dried without using the critical
point dryer.

RESULTS

Nomarski-differential-interference-contrast, phase-
contrast and fluorescence microscopy

Figure 1A shows a well-starved living cell of P.
caudatum which bears many infectious long forms
of H. obtusa in the macronucleus. Figure IB
shows the infectious long form of H. obtusa under
a phase-contrast microscope. It is evident that
nearly half of the bacterial length is consisted of
dark-looking part, and the remaining consisted of
a large refractile part with a small non-refractile
tip. Therefore, on the basis of the morphological
character, both ends of the bacterium can easily be
distinguished. In the same bacteria, a locality of a
fluorescence of DNA binding fluorochrome,
DAPI, was observed with a fluorescence micro¬
scope (Fig. 1C). It was found that DAPI fluores¬
cence appeared only at the refractile part of Figure
IB, but not at the non-refractile tip and dark¬
looking part. This indicates that DNA of the
infectious form of H. obtusa localizes only at the
refractile part, and that the bacterium enters the
macronucleus with the DNA present side fast. It is
of particular interest to note that each infectious
long form has two dotted small regions in the

DAPI positive part in which DAPI fluorescence is
especially strong.

Transmission electron microscopy

Cells of P. caudatum in late log phase of growth
contain various developmental stages of PI. obtusa
in their macronucleus (Fig. 2A, 2B): the infectious
long form (10-15 pm in length), the reproductive
short form (1.0- 1.5 pva in length) and the in¬
termediate forms of various lengths between the
former two types.

The infectious long form can be seen as being
constituted with the following three distinct parts
(Fig. 2A). One end of the bacterium has a sperm
acrosome-like structure, in which electron-
translucent material is composed. This special
structure can not be observed at the other end.
Remaining part of the bacterium can be divided
into two parts: a homogeneous electron dense
region adjacent to the acrosome-like structure and
a bacterial cytoplasmic region. When the mor¬
phology of the bacterium in Figures IB and 2 A
were compared, it is apparent that the acrosome-
like structure, the homogeneous electron dense
region and the bacterial cytoplasmic region in
Figure 2A correspond to the non-refractile tip, the
refractile part and dark-looking part in Figure IB,
respectively. Therefore, this electron-translucent
material in the acrosome-like tip and/or its surface
membrane seem to possess functions for nuclear
specificity and species specificity of the infectivity.
Furthermore, this electron dense region in Figure
2A can be considered as a fine structure of the
nucleoid of this bacterium. In the fluorescence
microscopy, two dotted DAPI positive regions
were found (Fig. 1C). Therefore, we searched the
corresponding structures by TEM, but could not
find them.

The infectious long forms are always situated
individually in an empty hole in the macronucleus,
and have a smooth surface though a shallow de¬
pression can be seen at the border between the
acrosome-like region and the electron dense re-

Fig. 2. Transmission electron photomicrographs of H. obtusa in the host macronucleus. A: Longitudinal sections of
infectious long forms. The infectious form consists of bacterial cytoplasmic region (single arrowhead), electron
dense region (double arrowhead) and electron translucent tip (triple arrowhead). B: Cross sections of infectious
long forms and reproductive short forms. The section of the infectious form (single arrowhead) is a circular
shape, while that of the reproductive form (double arrowhead) is an irregular shape. Bars, 1 pm.

Morphological Characters of Holospora

795

1 hi 1

|§pi

4r ^

%>».

796

M. Fujishima and K. Hoshide

gion (Fig. 2A, 2B). On the other hand, the
reproductive short form and the intermediate
forms of various lengths do not show such in¬
tracellular differentiations as observed in the infec¬
tious long form. They have a bumpy surface and
seem to be closely connected with the macronu-
clear chromatin.

Scanning electron microscopy

If the specific infection of H. obtusa to the
macronucleus depends on an affinity between the
surface membrane of the acrosome-like tip of the
bacterium and the macronuclear envelope, it can
be expected that the surface structure of the spe¬
cial tip may be somewhat different from those of
other parts of the bacterium. To confirm this
possibility, infectious long forms of H. obtusa were
observed with SEM.

Figure 3A through 3C show the infectious long

forms of H. obtusa which were dried with or
without using a critical-point dryer. As shown in
Figure 3 A and 3B, however, the bacterium has a
smooth surface and no local difference could be
detected on the surface structure. But, as shown in
Figure 3C, one of the both ends of the bacterium
was always depressed when the bacterium was
dried without using a critical-point dryer. This
depressed part seems to correspond to the non-
refractile tip in phase-contrast microscopy, or to
the sperm acrosome-like tip in TEM, because the
size of depressed part in SEM is almost the same
with that of the special tip and no similar-sized
structure was observed at the other end of the
bacterium as shown in Figures IB and 2A. It is
evident that this special tip has an inner-structural
property to be easily depressed by a surface ten¬
sion. On the other hand, a shallow depression
which was observed by TEM at the border be-

Fig. 3. Scanning electron micrographs of the infectious long form of H. obtusa. A and B: Both ends of the bacterium
dried with a critical point dryer. C: Bacterium dried without using the critical point dryer. Note that only one end
of the bacterium (arrowhead) is depressed by surface tension. Bars, 1 /j.m.

Morphological Characters of Holospora

797

tween the acrosome-like region and electron dense
region was not observed in Figure 3 A and 3B.
Therefore, the depression observed by TEM
seems to be an artifact.

DISCUSSION

In the ciliate Paramecium six species of Holo¬
spora are known to occur as endonuclear sym¬
bionts [8-11]. All of them exhibit not only a
species specificity but also a nuclear specificity in
their habitats: H. obtusa in the macronucleus of P.
caudatum, H. elegans and H. undulata in the
micronucleus of P. caudatum, H. caryophila in the
macronucleus of P. biaurelia, H. curviuscula in the
macronucleus of P. burs aria, and H. accuminata in
the micronucleus of P. burs aria. All of them are
gram-negative bacteria and cannot be cultivated
outside of the host cells.

In the present work, we observed morphological
features of the infectious long form of H. obtusa by
means of a phase contrast microscope, a fluores¬
cence microscope, a transmission electron micro¬
scope. Schematic representations of each observa¬
tion are summarized in Figure 4. We found that
DAPI fluorescence appeared only at the refractile
part, and that in the D API-positive refractile part
especially strong DAPI fluorescence appeared as
two dotted regions. Locality of the DAPI fluores¬
cence had been reported in an infectious long form
of H. elegans, too [12]. However, unlike the

present results, in H. elegans the fluorescence had
appeared only at the dark-looking part (a cytoplas¬
mic part) and no dotted strong fluorescence had
been observed there. It seems unlikely that such
difference is due to species-difference, because
both species have almost the same inner structures
in TEM. Therefore, reobservations of the locality
of DAPI fluorescence are needed in H. elegans.

Notwithstanding that the infectious long form of
H. obtusa always penetrates the macronuclear
envelope with the non-refractile tip. we could not
find any local difference on its surface structure.
As the failure to observe the surface membrane
differetiation at the special tip of the bacterium,
following possibilities are conceivable. Namely, if
the difference was really present, it might be so
minute to be detected by the SEM used in this
study, or such differentiation may occur at or after
the bacterium is incorporated into the host food
vacuoles in an early infection proccess. Recently,
Fujishima and Nagahara [3] succeeded in purifying
the infectious long forms from homogenates of
bacteria-bearing cells by a Percoll density gradient
centrifugation. They also succeeded in purifying
the bacterium from cells in an early infection
process, and found that before penetrating the
host macronuclear envelope the infectious form
changed its buoyant density from 1.16 g/ml to 1.13
g/ml and 1.11 g/ml as the time passed and its
nucleiod dispersed. Furthermore, they found that
the bacteria, of which buoyant densities were 1.13

Phase-contrast Fluorescence Transmission Scannning

microscopy microscopy electron electron

microscopy microscopy

non-ref racti

le

DAP I -negat i ve

composed with easily
a homogen ious depress b;
electron trans- a surface
lucent material tension

refractile

DAP I -positive composed with

(two dotted a homogen ious

points show electron dense

especialy strong material
f 1 uorescence )

non-refractile DAPI-negati ve composed with

a typical
bacterial
cytoplasm

entirely

smooth

surface

Fig. 4. Schematic representation of the morphological characteristics of the infectious
long form of H. obtusa.

798

M. Fujishima and K. Hoshide

g/ml and 1.11 g/ml, were easily digested in the food
vacuoles when the bacteria were added again to
paramecia. Therefore, the bacterium which just
penetrated the macronuclear envelope is not the
same one with the original infectious long form.
Since the infectious long form is hardly digested in
the host food vacuoles, the above observations
suggest that the nature of the bacterial surface
changed immediately after invasion into the host
cells. This evokes a possibility that the original
infectious long form has not the infectivity to the
macronuclear envelope but acquires it when the
bacterium changes its buoyant density to 1.13 g/ml
or to 1.11 g/ml after invasion into the host cell. If
this is true, the surface structure of the special tip
of the bacterium may also change when the infec¬
tious long form reduces its buoyant density. This
possibility remained to be confirmed in future.

Besides the differences in morphology, DNA
content and transcriptional activity, several differ¬
ences have been found between the macro- and the
micronucleus of ciliates: ribosomal RNA gene
amplification [13-15] and lack of some micronu-
clear DNA sequences in the macronucleus [16—
20], and the presence of a macronucleus-specific
histone [21-23] and a micronucleus-specific his¬
tone [24]. However, difference between the nu¬
clear membranes of two kinds of nuclei, or some¬
thing associated with them, have not yet been
detected. In this respect, to know the infection
mechanism of the Holospora species to a specific
nucleus, either the macro- or the micronucleus,
can be expected to provide new evidence on
nuclear membrane differentiation between the two
kinds or nuclei of common genetic origin.

ACKNOWLEDGMENTS

This work was supported by grants to M. F. from the
Ministry of Education, Science and Culture of Japan
(Nos. 61740422 and 62740428), the Itoh Science Founda¬
tion, the Inamori Foundation, and the Naitoh Founda¬
tion.

REFERENCES

1 Ossipov, D. V., Skoblo, I. I. and Rautian, M. S.
(1975) Iota-particles, macronuclear symbiotic bac¬
teria of ciliate Paramecium caudatum clone Ml 15.

Acta Protozool., 14: 263-280.

2 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.

3 Fujishima, M. and Nagahara, K. (1984) Isolation of
endonuclear symbiont Holospora obtusa from mass
cultures of Paramecium caudatum. VII Int. Congr.
Protozool. (Nairobi). Abst., p. 301.

4 Fujishima, M. and Fujita, M. (1985) Infection and
maintenance of Holospora obtusa, a macronucleus
specific bacterium of the ciliate Paramecium cauda¬
tum. J. Cell Sci., 76: 179-187.

5 Fujishima, M. (1986) Further study of the infectivity
of Holospora obtusa, a macronucleus-specific bac¬
terium of ciliate Paramecium caudatum. Acta Pro¬
tozool., 25: 345-350.

6 Dryl, S. (1959) Antigenic transformation in Para¬
mecium aurelia after homologous antiserum treat¬
ment during autogamy and conjugation. J. Pro¬
tozool., 6 (suppl.): 25.

7 Hiwatashi, K. (1968) Determination and inheri¬
tance of mating type in Paramecium caudatum.
Genetics, Prinston, 58: 378-386.

8 Hafkine, W. M. (1890) Maladies infectieeuses des
paramecies. Annls. Inst. Pasteur, Paris, 4: 148-162.

9 Preer, L. B. (1969) Alpha, an infectious macronu¬
clear symbiont of Paramecium aurelia. J. Pro¬
tozool., 16: 570-578.

10 Ossipov, D. V., Borchsenius, O. N. and Podlipaev,
S. V. (1980) Some peculiarities in organization of
nuclear apparatus in a ciliate, Paramecium bursaria,
infected with symbiotic bacteria, Holospora acumi¬
nata. Acta Protozool., 19: 315-326.

11 Borchsenius, O. N., Skoblo, I. I. and Ossipov, D.
V. (1983) Holospora curviuscula - A new species of
macronuclear symbiotic bacteria of Paramecium
bursaria. Tsitologiya, 25: 91-97.

12 Gortz, H.-D. and Dieckman, J. (1980) Life cycle
and infectivity of Holospora elegans Hafkine, Protis-
tologica, 16: 591-603.

13 Gall, J. G. (1974) Free ribosomal RNA genes in the
macronucleus of Tetrahymena. Proc. Natl. Acad.
Sci. U.S.A., 74: 3078-3081.

14 Yao, M.-C., Blackburn, E. and Gall, J. G. (1979)
Amplification of the rRNA genes in Tetrahymena.
Cold Spring Harbor Symp. Quant. Biol., 43: 1293-
1296.

15 Yao, M.-C. (1981) Ribosomal RNA gene amplifica¬
tion in Tetrahymena may be associated with chromo¬
somal breakage and DNA elimination. Cell, 24:
765-774.

16 Ammermann, D., Steinbruck, G., Von Berger, L.
and Henning, W. (1974) The development of the
macronucleus in the ciliated protozoan Stylonychia
mytilus. Chromosoma, 45: 401-429.

Morphological Characters of Holospora

799

17 Lauth, M. R., Spear, B. B., Heumann, J. and Pre¬
scott, D. M. (1976) DNA sequence diminution dur¬
ing macronuclear development of Oxytricha. Cell, 7:
67-74.

18 McTavish, C. and Sommerville, J. (1980) Macro-
nuclear DNA organization in Paramecium pri-
maurelia. Chromosoma, 78: 147-164.

19 Yao, M.-C. (1982) Elimination of specific DNA
sequences from the somatic nucleus of the ciliate
Tetrahymena. J. Cell Biol., 92: 783-789.

20 Karrer, K. M. (1983) Germ line specific DNA se¬
quences are present on all five micronuclear
chromosomes in Tetrahymena thermophila. Mol.
Cell. Biol., 3: 1909-1919.

21 Allis, C. D., Glover, C. V. C., Bowen, J. K. and
Gorovsky, M. A. (1980) Histone variants specific to

the transcriptionally active, amitotically dividing
macronucleus of the unicellular eukaryote, Tetrahy¬
mena thermophila. Cell, 20: 609-617.

22 Allis, C. D., Ziegler, T. S., Gorovsky, M. A. and
Olmsted, J. B. (1982) A conserved histone variant
enriched in nucleoli of mammalian cells. Cell, 31:
131-136.

23 Wenkert, D. and Allis, C. D. (1984) Timing of the
appearance of macronuclear-specific histone variant
hvl and gene expression in developing new macro¬
nuclei of Tetrahymena thermophila. J. Cell Biol., 98:
2107-2117.

24 Allis, C. D., Glover, C. V. C. and Gorovsky, M. A.
(1979) Micronuclei of Tetrahymena contain two
types of histone H3. Proc. Natl. Acad. Sci. U.S.A.,
76: 4857-4861.

ZOOLOGICAL SCIENCE 5: 801-808 (1988)

© 1988 Zoological Society of Japan

Light-induced Michaelis Constant Increase is Rapid and
Inherent in cGMP Phosphodiesterase
in Frog Rod Outer Segments

Satoru Kawamura and Motohiko Murakami

Department of Physiology , Keio University School of Medicine, Shinano-machi 35,
Shinjuku-ku, Tokyo 160, Japan

ABSTRACT — On cGMP phosphodiesterase (PDE) activation, the Michaelis constant (Km) has been
shown to increase in an intact preparation of rod outer segments from the value of O.lmM in the dark to
1 mM in the light. In the present study, firstly we obtained the evidence that this light-induced Km
increase is not an artifact caused by reduced substrate accessibility. Then we tried to specify the
molecular species responsible for this Km increase. An exogenous PDE activator (transducin-GMPPNP
complex) induced a high Km value in the dark in intact membranes but not in freeze-thawed membranes
that had lost the capability of the light-induced Km increase. The result indicated that intact structure of
PDE is required to induce the Km increase. The light-induced Km increase occurred within 200 msec
after a light stimulation. When in situ cGMP concentration is taken into account, the rapid Km increase
might be associated with rapid turn-off of cGMP hydrolysis in situ.

INTRODUCTION

In rod outer segments, light illumination induces
cGMP phosphodiesterase (PDE) activation [1]
that probably leads to reduction in internal free
cGMP concentration [2, 3] and thereby the reduc¬
tion in the light-sensitive conductance (see for
reviews [4, 5]).

In an intact preparation1 } of rod outer segments,
the Michaelis constant (Km) of PDE in the dark is
about 0.1 mM and it increases to about 1 mM after
light illumination [6]. However, in purified,
freeze-thawed or sonicated disk membranes, the
Km value stays almost unchanged on light illu¬
mination (0. 1-0.3 mM, [1, 7, 8]). In our previous
report [8], we characterized the chemical nature of
the light-induced increase in the Km value (light-
induced Km increase). It was found that the
light-induced Km increase is labile to experimental
manipulations such as freeze-thawing, sonication
and pipetting.

Since stacked disk membranes are usually
observed in intact preparations (see Materials and
Methods), it has been suggested that the high Km
value is due to reduced substrate accessibility in

Accepted November 17, 1987
Received October 23, 1987

the space between stacked membranes and that
freeze-thawing increases the accessibility (see dis¬
cussion in [9]). Though this possibility was ques¬
tioned previously [8, 10], firstly in the present
study, it was further examined. Secondly, we
extended our previous study and tried to under¬
stand why the light-induced Km increase is not
observed in freeze-thawed membranes. Finally we
examined the rapid time course of the light-
induced Km increase.

MATERIALS AND METHODS

Preparation of intact disk membranes and freeze-
thawed membranes

Retinas were obtained from dark-adapted bull¬
frogs ( Rana catesbeiana). All the following ma-

Formerly [8], we termed “crude” preparation. Howev¬
er, this preparation is a minimally disrupted one and
therefore must retain intactness of the membrane. For
this reason we termed “intact” preparation throughout
this article.

Abbreviations:

PDE, cGMP phosphodiesterase; GMPPNP, guanylylimi-
dodiphosphate; HEPES, 4-(2-hydroxyethly)-l-
piperazineethanesulfonic acid.

802

S. Kawamura and M. Murakami

nipulations were carried out in the dark with the
aid of an infrared image converter (NVR 2015,
NEC, Tokyo, Japan). Detailed procedures for
isolation of rod outer segment appeared in our
previous report [8]. However briefly, rod outer
segments were obtained by sucking the retinas into
a pipette tip in a pseudo-intracellular solution (110
mM KC1, 10 mM NaCl, 2mM MgCl2, 0.1 mM
CaCl2, 2.78 mM EGTA, 10 mM HEPES, 1 mM
dithiothreitol, pH 7.8). The plasma membranes of
the rod outer segments thus obtained (intact rod
outer segments) are permeable to cGMP [6].
Though freshly isolated rod outer segments show a
high Km value in the light, the maximal PDE
activity gradually increases for 2 hr until it reaches
to a steady level. For quantitative analysis, the
maximal PDE activity should be constant during
the course of an experiment. For this reason,
intact rod outer segments were kept at room
temperature for at least 3 hr or at 4°C overnight
before use. During this period, rod outer segments
decomposed to disk membranes (intact disk mem¬
branes [8]). However, all of the membranes were
not dispersed but stacked membranes were usually
observed as a minor constituent.

The reason why we used intact preparation was
firstly that it contains all of the necessary compo¬
nents to produce a high Km value and secondly
that in frogs, the light-induced Km increase is
easily lost during purification process of disk mem¬
branes [8].

Freeze-thawed disk membranes that show low
Km values both in the dark and in the light were
obtained by storing the intact rod outer segments
at — 20°C for one or a few days. In most cases,
single freeze-thawing reduced the Km value in the
light to less than 0.4 mM. However, occasionally,
the reduction was partial (—0.6 mM). Though the
reason for this was not known, repetition of freeze¬
thawing of this preparation reduced the Km value
to less than 0.4 mM.

Measurement of PDE activity

PDE activity was estimated from pH change
caused by hydrolysis of cGMP as reported pre¬
viously [11, 12] after some modification of the
method of Yee and Liebman [7]. A 200-/4 portion
of a suspension of disk membranes was made 0.5

mM in both GTP and ATP. For the pH measure¬
ment except for the experiment in Figure 3, MI-
410 glass electrode (Microelectrodes Inc., London¬
derry, USA) was connected to a pH meter (PHM
82, Radiometer, Copenhagen, Denmark). In Fig¬
ure 3, where pH change was measured with fast
time resolution, a metal-type pH electrode (Beet-
rode MEPH, WPI Instruments Inc., New Haven,
USA) was connected to a pre-amplifier (MEZ
8201, Nihon Kohden, Tokyo, Japan). According
to the manufacturer’s catalogue, a 90% response is
obtained in 100 msec with this electrode2). Though
the drift of this electrode was larger than that of
the glass electrode, it did not disturb the rapid pH
measurement that was completed within 2 sec after
a light flash. In either pH measurement, calibra¬
tion was made by addition of known amounts of
HC1 to the pseudo-intracellular solution. PDE
activity was estimated from a tangent of the pH
recording and was expessed as moles of cGMP
hydrolyzed per mole of rhodopsin present per min
(cGMP/rh-min). Since the dark PDE activity was
negligibly low in our preparation [8], it was not
subtracted from the activity in the light.

The Km value in the light was determined from
a single continuous recording of the pH change
under steady illumination [8] except for the experi¬
ment in Figure. 3. Since the PDE activity is high
immediately after the onset of light illumination
but becomes constant after 15 sec [8], disk mem¬
branes were illuminated for 1 min prior to cGMP
addition (final concentration, 2.7 mM) and pH
change was continuously recorded. In the experi¬
ment in Figure 3, we examined the rapid time
course of the light-induced Km increase. For this
purpose, we added known amounts of cGMP to
intact disk membrane preparations in the dark and

2) In WPI’s 1985 catalogue, the response time of this
electrode was specified as 100 msec while in their 1987
catalogue, it increased to 1 sec. In response to our
inquiry, they indicated that 1 sec response time is
obtained by moving the electrode between two differ¬
ent pH buffer solutions. In this case, the response time
depends on how quickly the first solution is replaced
by the second at the electrode-solution interface.
When the pH changes homogeneously throughout the
solution like in our experiment, they indicated, the
response time is faster and on the range of 100 msec.

Km Increase of cGMP Phosphodiesterase

803

measured the PDE activities at various cGMP
concentrations at each fixed time after a light flash.
Because of low PDE activity in the dark in the
intact preparation [8], the measurement of dark
Km value was relatively difficult with the pH assay
method. However, the dark Km value of PDE in
our intact preparation has been reported to be
about 0.1 mM [6, 8]. Rhodopsin concentration
was 8-12 p. M. When cAMP was used as the
substrate, rhodopsin concentration was raised to
25-40 fjM, since higher rhodopsin concentration
was necessary to obtain reliable pH recordings.
Light intensity was 3.7 XlO7 rhodopsin molecules
bleached per outer segment per sec.

Preparation of transducin' guanyly limidodiphos-
phate (transducin' GMPPNP) complex

PDE was activated by transducin- GMPPNP
complex in Figure 2. The complex was obtained
after the method of Kuhn [13] using mainly freeze-
thawed disk membranes. Protein concentration
was determined with the method of Lowry et al.

[14].

Chemicals

cGMP, cAMP, ATP, GTP, polyethylene glycol
were obtained from Sigma (St. Louis, USA) and
the other chemicals from Nakarai (Kyoto, Japan).

RESULTS

Evidence that the high Km value in the intact
preparation is not an artifact

The high Km value of PDE is observed in intact
rod outer segments or intact disk membranes [6,
8]. In the following, in order to show that it is not
an artifact caused by reduced substrate accessibil¬
ity, firstly we tried to constitute a model system in
which most probably the reduction in the substrate
accessibility increased the Km value even in
freeze-thawed membranes. Then we compared
the mechanism of this Km increase with that of the
light-induced Km increase.

Polyethylene glycol is known to induce mem¬
brane aggregation [15] and its addition to freeze-
thawed disk membranes actually caused mem¬
brane aggregation (data not shown). In the pre¬

sence of the reagent, kinetic parameters of PDE
were determined. The addition of polyethylene
glycol increased the Km value in the light (filled
circles in Fig. la) and decreased the maximal PDE
activity (open circles) in a dose-dependent man¬
ner. The effect of the reagent was reversible, since
when the polyethylene glycol concentration was
reduced by washing, the Km value and the maxi¬
mal PDE activity returned to the control value.

We believe that in the aggragate membranes
prepared as shown above, substrate accessibility
was reduced for the following reasons. (1) It is
reasonable to assume that membrane aggregation
inhibits free access of the substrate to the enzyme.
In fact, polyethylene glycol-induced Km increase
did not show substrate specificity (see below). (2)
Though addition of polyethylene glycol increased
the viscosity of the solution, the increased viscosity
was not the determinant of the high Km value in
Figure la. This conclusion was drawn from a
control experiment in which the viscosity of the
solution was raised with sucrose. In the presence
of 19% (w/w) polyethylene glycol, the Km was 2.5
mM (Fig. la) and the relative viscosity to water in
this solution was 5.5. Even at a higher viscosity of
9 in a sucrose solution (44%; w/w), the Km was
less than 0.5 mM. (3) The Km increase by
polyethylene glycol is not due to the direct effect of
this reagent on PDE, since solubilized and trypsin
activated PDE did not increase the Km value even
at 12% (w/w) polyethylene glycol concentration
(data not shown).

Then we tried to compare the mechanism of the
aggregation-induced Km change with that of the
light-induced Km increase. By adding appropriate
amounts of polyethylene glycol to the freeze-
thawed membrane preparation, we tried to obtain
a preparation in which the Km value for cGMP in
freeze-thawed membrane increased close to the
value observed in intact disk membranes. The Km
values in untreated freeze-thawed membranes,
aggregated membranes in the presence of 10.6%
(w/w) polyethylene glycol and intact disk mem¬
branes are shown on the upper panel in Figure lb
and labeled as f thawed, -\-PEG and intact , re¬
spectively. In the same preparations used above,
we measured the Km values for cAMP. Addition
of polyethylene glycol to freeze-thawed mem-

Michaelis constant (•) (mM)

804

S. Kawamura and M. Murakami

Cb)

Michaelis constant (mM]

0 0.5 1

i - 1 - r

f .thawed

-1

cGMP

+ PEG

— 1

intact

D

5

1 0

1 -

f.thawed

i

— 1

!

cAMP

+ PEG

- 1

intact

Fig. 1. Effects of polyethylene glycol on the phospho¬
diesterase activity (PDE activity) and the Michaelis
constant.

(a) PDE activities were measured in freeze-thawed
membranes from a continuous pH recording in the
presence of known concentrations of polyethylene
glycol. The Michaelis constant and the maximal
PDE activity were determined from the double¬
reciprocal plot. The obtained maximal PDE activity
(open circles) and Michaelis constant (filled circles)
were plotted against polyethylene glycol concentra¬
tion (w/w). Each symbol represents single deter¬
mination.

(b) Michaelis constants in freeze-thawed membranes in
the absence and presence of 10.6% (final cone.)
polyethylene glycol (f. thawed and + PEG, respec¬
tively). The value in intact disk membranes is also
shown (intact). In cGMP experiment (upper panel),
the Km values shown are 0.38 + 0.02 mM (f. thawed ;
n=3, a mean±SE), 0.76 + 0.04 mM ( + PEG; n = 3)
and 0.86 + 0.13 mM ( intact ; n = 8). In cAMP experi¬
ment (lower panel), the values are 3.8 ±0.3 mM (/.
thawed ; n=3), 8.2+ 1.4 mM ( + P£G; n = 3) and 3.5
±0.7 mM ( intact ; n = 17).

branes increased the Km value for cAMP as well
(- \-PEG in lower panel in Fig. lb). However,
freeze-thawing did not reduce the Km value for
cAMP (compare the Km value of “/. thawed ” with
that of “ intact ” in lower panel). The result indi¬
cated that the Km change induced by membrane
aggregation and thus, reduced substrate accessibil¬
ity is non-specific to substrate species while light-
induced Km increase is specific to cGMP. There¬
fore, the two types of the Km changes are based on
different mechanisms and the reduction in the
substrate accessibility is not the mechanism that
accounts for the light-induced Km increase.

One may think that to solve the above issue,
there might be a direct approach such as a recon¬
stitution experiment using purified components.
However, at least in frogs, the light-induced Km
increase is easily lost during purification process [8]
and at present, it is not possible.

Km increase induced by transducin' guanylylimido-
diphosphate complex (transducin' GMPPNP com¬
plex) in the dark

In order to study why the light-induced Km
increase is not observed in freeze-thawed mem¬
branes, PDE was activated directly with trans-
ducin-GMPPNP complex in the dark.

In agreement with the report of Fung et al. [16],
transducin- GMPPNP complex activated PDE in
intact disk membranes without light illumination.
Half-maximal activation was observed with addi¬
tion of 5 p\ of the extract (0.4 pg protein). Under
the condition of full activation (7.7 pg protein),
PDE activities in the dark were determined in
intact disk membranes and in freeze-thawed mem¬
branes. The double-reciprocal plot showed that
the Km was 1.1 mM in intact membranes (filled
circles and solid line in Fig. 2a) and 0.23 mM in
freeze-thawed membranes (filled circles and solid
line in Fig. 2b). As controls, the PDE activities in
the two types of the membranes were measured in
the absence of the extract but in the light. The
measured Km value was 0.90 mM in intact mem¬
branes (open circles and dotted line in Fig. 2a) and
0.28 mM in freeze-thawed membranes (open cir¬
cles and dotted line in Fig. 2b). It is evident from
Figure 2 that transducin-GMPPNP complex in¬
creased the Km in intact membranes (Fig. 2a) but

Km Increase of cGMP Phosphodiesterase

805

1/cGMP CmM'1 )

Fig. 2. Effect of transducin’ guanylylimidodiphosphate
(transducin- GMPPNP) complex on the Km values in
intact and freeze-thawed membranes.

(a) Phosphodiesterase activity in intact disk mem¬
branes was measured in the presence of trans¬
ducin’ GMPPNP complex (TGMppNP; 7.7 pg protein)
either in the dark or light. Each data point shown
represents the mean±SE in triplicate determina¬
tions. Assuming linear relations, best fit lines were
determined in the double-reciprocal plot in the dark
(filled circles and solid line) and light (dashed line).
To avoid crowdedness, the data points in the light
are not shown, but the position of the data points
and their deviations are quite similar to those in the
dark. As for a control, relation between PDE
activity and cGMP concentration was also measured
in intact disk membranes in the light in the absence
of the complex (open circles and dotted line).

(b) Phosphodiesterase activities were measured in
freeze-thawed membranes in the presence of trans¬
ducin’ GMPPNP complex in the dark (Km = 0.23
mM; filled circles and solid line) and in the absence
of the complex in the light (Km = 0.28mM; open
circles and dotted line).

not in freeze-thawed ones (Fig. 2b). The result
suggests that PDE molecules are in different con¬
formation in the two types of the membranes. In
other words, PDE intact conformation is required
to exert the light-induced Km increase.

In the presence of transducin-GMPPNP, the Km
value in intact membranes was also determined in
the light. The value was almost the same (1.1 mM,
dashed line in Fig. 2a) as that obtained in the dark.

Rapid time course of light-induced Km increase

As has been hypothesized [8], the light Km
value might increase in a time-dependent manner
after light stimulation. Because of low time resolu¬
tion of the pH electrode used in our previous work
(~2sec), rapid kinetic measurement was not
possible. In the present study, however, using a
metal-type pH electrode, rapid time course of the
light-induced Km increase was measured as early
as 200 msec after a light stimulation.

A suspension of intact disk membranes was
divided into 8 portions. In the dark, ATP, GTP
and known amounts of cGMP were added to one
portion and a light flash of 6 msec duration was
given. With various starting cGMP concentrations
in 8 portions, pH changes were recorded after light
flashes. Some of the recordings are shown as an
inset in Figure 3a.

For a reliable measurement, we used an intense
light (20% bleach/flash) to accelerate PDE activa¬
tion kinetics [7] and raised rhodopsin concentra¬
tion to 30 pM. PDE activities were determined at
various cGMP concentrations at various time in¬
tervals after the flash. Two examples of the
double-reciprocal plot at 200 msec and 1 sec after
the flash are shown in Figure 3a. The Km values
were 0.76 mM at 200 msec (open circles and solid
line in Fig. 3a) and 0.74 mM at 1 sec (filled circles
and dashed line in Fig. 3a). Though the Km value
before light flash was not measured in the present
experiment (see Materials and Methods), it is
about 0.1 mM in our intact preparation [8].

The Km values were plotted against time after
the flash together with maximal PDE activities
(Fig. 3b). PDE activity (open circles in Fig. 3b)
increased gradually and reached its peak at about 1
sec after the flash, while the Km value (filled
circles in Fig. 3b) increased within the time resolu¬
tion of the present measurement and stayed almost
constant during this period. Though the data point
slightly scattered at 150 msec (data not shown), the
Km value was also found to be high. Since the Km
increase is more rapid than the activity increase

806

S. Kawamura and M. Murakami

1/cGMP (mM-1)

CbD

400 > §

O ®

CO ^

LLI O

g |

200 .1 |
co O

'o

0 1 2
Time after flash (sec)

Fig. 3. Rapid time course of the Km increase.

(a) The double-reciprocal plot of phosphodiesterase
(PDE) activity versus cGMP concentration at 200
msec (open circles and solid line) and 1 sec (filled
circles and dashed line) after a light flash. Inset
recordings: pH recordings at varying cGMP concen¬
trations (0, 0.44 and 2.24 mM from bottom to top).
The records were low-pass-filtered at 15 Hz with an
8-pole active filter. A light flash was given at the
time indicated by an arrow.

(b) Time courses in the changes of the Michaelis con¬
stant and the maximal PDE activity after a light
stimulation. The Michaelis constant (filled circles)
and the maximal PDE activity (open circles)
obtained from the double reciprocal plot were plot¬
ted against time after the light flash.

(Fig. 3), one might suspect that the Km value
increases before PDE is activated. However, it is
not the case: since the activated enzyme contri¬
butes mainly in the calculation of the Km value,
the calculated Km values reflect the Km value of
the activated PDE. Therefore, the result showed
that the Km value becomes high within 200 msec
after PDE activation. Though the Km decreased
at 2 sec after the flash in Figure 3, it was not
observed in other experiments. The deviation was
within error of our Km estimation.

DISCUSSION

The present work revealed the following results.
(1) Freeze-thawing specifically reduces the Km
value for cGMP but addition of polyethylene gly¬
col increases the Km values non-specifically (Fig.
1). (2) When PDE is activated in the dark by
transducin- GMPPNP complex, the Km value is
high in intact disk membranes but low in freeze-
thawed membranes (Fig. 2). (3) The light-induced
Km increase occurs within 200 msec after a light
flash (Fig. 3).

Mechanism of light-induced Km increase

There has been a suggestion that reduced sub¬
strate accessibility is the cause of the high Km
value for cGMP in intact disk membranes in the
light. In the present work, we tried to reduce the
substrate accessibility to the enzyme by forming
membrane aggregation (Fig. 1). The Km values
for both cGMP and cAMP increased, but freeze¬
thawing reduced the Km value only for cGMP.
This experiment indicated that the reduced sub¬
strate accessibility is not the determinant of the
high Km value in intact disk membranes. Our
previous work showed that even after rod outer
segments were ruptured by lysis, PDE still shows a
high Km value [8]. All these observations are
consistent with the idea that mM level of the Km
value for cGMP in intact disk membranes is not
due to reduced substrate accessibility. Using re¬
constituted cattle disk membranes which showed
high Km value, Sitaramayya et al. [10] found that
the Km value was constant irrespective of the
extent of the local rate of cGMP consumption. If
the result obtained in their reconstituted system
could apply to that in the intact membranes, their
result also supports our conclusion. Therefore, it
seems reasonable to conclude that the Km of in
situ PDE in the light is close to 1 mM.

As shown in Figure 2, intact conformation of
PDE seemed to be required to cause the light-
induced Km increase. Since we obtained trans¬
ducin' GMPPNP complex from freeze-thawed
membranes (see Materials and Methods) and it
had the ability to induce high Km value in intact
membranes, obviously transducin is not responsi¬
ble for the light-induced Km increase. Even when

Km Increase of cGMP Phosphodiesterase

807

the freeze-thawed membranes were fused with
intact membranes, PDE in freeze-thawed mem¬
branes did not restore the ability of the light-
induced Km increase (Kawamura and Murakami,
unpublished observation). From these results, it
seems that the PDE conformational change in
freeze-thawed membranes is the primary cause for
the loss of the light-induced Km increase. Howev¬
er, since we cannot totally exclude the possibility
of the presence of the Km-increase factor [6, 8]
from these results, further work is necessary to
elucidate the molecular mechanism of the light-
induced Km increase. If the light-induced Km
increase is attained by PDE molecule itself, the
chemical nature of the light-induced Km increase
studied in our previous report [8] is that of PDE
molecule itself.

One might suspect that the high Km value is an
artifact caused by the pH assay method. However,
this is not the case, since conventional radioisoto¬
pic assay also shows the light-induced Km increase
[6],

Possible physiological significance of Km increase

Using lsO labeling technique, Ames et al. [17]
reported that the temporal on/off resolution of the
activated PDE is fast and in a range of less than 1
sec. However, in our electrophysiological PDE
activity measurement done by intracellular injec¬
tion of cGMP, the PDE activity slowly turned off
after a light flash in a range of tens of sec [18, 19].
This apparent discrepancy may be explained by the
following control mechanism on the PDE activity
through Km regulation.

Though the Km value in the light is close to 1
mM as concluded above, this does not necessarily
imply that the value is constant during the entire
period of PDE activation. Based on the fact that
the PDE has an ability to show lower Km value in
various types of preparations even in the light, we
previously hypothesized that the light Km value
increases in a time-dependent manner after light
stimulation [8]. In the present work, it is still
possible that under in situ condition, during a short
period after PDE activation, the Km stays at a low
value of 0.1-0.3mM as in purified disk mem¬
branes, and then rapidly increases to 1 mM within
200 msec (Fig. 3). Since internal cGMP concentra¬

tion is estimated to be in the order of pM [20, 21],
the Michaelis-Menten equation predicts that this
Km increase results in a decrease in in situ PDE
activity to 1/3-1/10 of the initial value within 200
msec after a light flash. If this self- regulatory Km
increase mechanism operates, it explains not only
the in situ rapid turn-off of PDE but also the slow
PDE activity recovery measured electrophysiologi-
cally: intracellular cGMP injection increases in situ
substrate concentration to reduce the effect of the
Km increase, and therefore, PDE hydrolyzes the
injected cGMP. None of the known reactions
participating in the PDE inactivation is rapid
enough to match the recovery time course of
photoreceptor potential after a light stimulus.
Therefore, even though at present, we do not have
the evidence for the time-dependent Km increase
in the light, the above possibility seems to deserve
further investigation.

Though internal free cGMP concentration might
be lower than the Km value of PDE, it may be
associated with the physiological significance of
PDE. If the dark Km value is lower than internal
free cGMP concentration, considerable amounts
of cGMP bind to PDE in the dark. As far as the
physiological role of PDE is concerned, the hy¬
drolysis of such bound cGMP does not contribute
to rapid hydrolysis of free cGMP. Therefore, the
amounts of bound cGMP are minimized in a way
that the Km of PDE in the dark is above the free
cGMP concentration. As stated above, we believe
that 0.1 mM is the Km of the active form of in situ
PDE. This value is also higher than internal free
cGMP concentration. According to Fersht [22],
the high Km value in the light could be the result of
molecular evolution: an enzyme evolves so that it
binds to the transition state of the substrate firmly,
which reduces the activation energy of the enzyme
reaction and induces high catalytic activity. The
firm binding of PDE to the transition state of the
substrate leads to weak binding to free cGMP and
is characterized by a high Km value.

ACKNOWLEDGMENT

This work was supported by Grants-in-Aid from the
Ministry of Education, Science and Culture of Japan to
S. K. (No. 60780286) and to M. M. (No. 60304098).

808

S. Kawamura and M. Murakami

REFERENCES

1 Miki, N., Keirns, J. J., Marks, F. R., Freeman, J.
and Bitensky, M. W. (1973) Regulation of cyclic
nucleotide concentrations in photoreceptors: an
ATP-dependent stimulation of cyclic nucleotide
phosphodiesterase by light. Proc. Natl. Acad. Sci.
USA, 70: 3820-3824.

2 Woodruff, M. L. and Bownds, M. D. (1979) Ampli¬
tude, kinetics and reversibility of a light induced
decrease in 3', 5'-cyclic monophosphate in frog
photoreceptor membranes. J. Gen. Physiol., 73:
629-653.

3 Cote, R. H., Biernbaum, M. S., Nicol, G. D. and
Bownds, M. D. (1984) Light-induced decreases in
cGMP concentration precede changes in membrane
permeability in frog rod photoreceptors. J. Biol.
Chem., 259: 9635-9641.

4 Stryer, L. (1986) Cyclic GMP cascade of vision.
Ann. Rev. Neurosci., 9: 87-119.

5 Kaupp, U. B. and Koch, K.-W. (1986) Mechanism
of photoreception in vertebrate vision. Trends
Biochem. Sci., 11: 43-47.

6 Robinson, P. R., Kawamura, S., Abramson, B. and
Bownds, M. D. (1980) Control of the cyclic GMP
phosphodiesterase of frog photoreceptor mem¬
branes. J. Gen. Physiol., 76: 631-645.

7 Yee, R. and Liebman, P. A. (1978) Light-activated
phosphodiesterase of the rod outer segment: kinetic
parameters of activation and deactivation. J. Biol.
Chem., 253: 8902-8909.

8 Kawamura, S. and Murakami, M. (1986) Charac¬
terization of the light-induced increase in the
Michaelis constant of the cGMP phosphodiesterase
in frog rod outer segments. Biochim. Biophys. Acta,
870: 256-266.

9 Applebury, M. L., Cavaggioni, A., Chabre, M.,
Findlay, J. B. C., Goldberg, N. D., Kaupp, U. B.,
Kuhn, H., Liebman, P. A., Prinz, H., Saibil, H. R.,
Schleicher, A. and Stryer, L. (1986) Triggering and
amplification. In “The Molecular Mechanism of
Photoreception”. Ed. by H. Stieve, Springer-
Verlag, Berlin, pp. 401-429.

10 Sitaramayya, A., Harkness,J., Parkes, J. H., Gon-
zalez-Oliva, C. and Liebman, P. A. (1986) Kinetic
studies suggest that light-activated cyclic GMP phos¬
phodiesterase is a complex with G-protein subunits.

Biochemistry, 25: 651-656.

Kawamura, S. and Bownds, M. D. (1981) Light
adaptation of the cyclic GMP phosphodiesterase of
frog photoreceptor membranes mediated by ATP
and calcium ions. J. Gen. Physiol., 77: 571-591.
Kawamura, S. (1983) Involvement of ATP in activa¬
tion and inactivation sequence of phosphodiesterase
in frog rod outer segments. Biochim. Biophys. Acta,
732: 276-281.

Kiihn, H. (1980) Light- and GTP-regulated interac¬
tion of GTPase and other proteins with bovine
photoreceptor membranes. Nature, 283: 587-589.
Lowry, O. H., Rosebrough, N. J., Farr, A. L. and
Randall, R. J. (1951) Protein measurement with the
folin phenol reagent. J. Biol. Chem., 193: 265-275.
Schramm, M. (1979) Transfer of glucagon receptor
from liver membranes to a foreign adenylate cyclase
by a membrane fusion procedure. Proc. Natl. Acad.
Sci. USA, 76: 1174-1178.

Fung, B. K.-K., Hurley, J. B. and Stryer, L. (1981)
Flow of information in the light-triggered cyclic
nucleotide cascade of vision. Proc. Natl. Acad. Sci.
USA, 78: 152-156.

Ames III, A., Walseth, T., Heyman, R., Barad, M.,
Graeff, R., and Goldberg, N. (1985) Light-induced
increases in cGMP metabolic flux correspond with
electrical response of photoreceptors. J. Biol.
Chem., 261: 13034-13042.

Kawamura, S. and Murakami, M. (1983) Intracellu¬
lar injection of cyclic-GMP increases sodium con¬
ductance in gecko photoreceptors. Jpn. J. Physiol.,
33: 789-800.

Kawamura, S. and Murakami, M. (1986b) In situ
cGMP phosphodiesterase and photoreceptor poten¬
tial in gecko retina. J. Gen. Physiol., 87: 737-759.
Yau, K.-W. and Nakatani, K. (1985) Light-
suppressible, cyclic GMP-sensitive conductance in
the plasma membrane of a truncated rod outer
segment. Nature, 317: 252-255.

Fesenko, E. E., Kolesnikov, S. S. and Lyubarsky,
A. L. (1986) Direct action of cGMP on the conduct¬
ance of retinal rod plasma membrane. Biochim.
Biophys. Acta, 856: 661-671.

Fersht, A. (1985) Enzyme structure and mecha¬
nism. W. H. Freeman and Co., New York, 2nd ed.,
pp. 311-331.

11

12

13

14

15

16

17

18

19

20

21

22

ZOOLOGICAL SCIENCE 5: 809-813 (1988)

© 1988 Zoological Society of Japan

Isolation and Identification of Crucian ( Carassius auratus L.)
Hemoglobin and its Subunits* 1

Fang Hung and Yan Shaoyi2
Institute of Developmental Biology, Academia Sinica, Beijing, China

ABSTRACT — Hemoglobin was isolated from the red blood cells of a common fresh-water fish, crucian
( Carassius auratus L.) by using DEAE-cellulose chromatography. Two types of globin chains of the Hb
were separated by using reverse phase HPLC and urea-Triton acid polyacrylamide gel electrophoresis.
The N-terminal amino acids of a- and /9-chains were both identified as Val-Glu-Try. Our results showed
no obvious polymorphism of Hb in crucian hemolysate.

INTRODUCTION

Fish hemoglobins (Hbs) are not commonly stud¬
ied as compared with Hbs of other vertebrates. So
far, only studies on the primary structure of the
Hbs of carp ( Cyprinus carpio) [1], catostomid
{Catostomus clarkii ) [2], gold-fish ( Carassinus au¬
ratus) [3], and trout ( Salmo irideus ) [4], and a brief
study on the heterogeneity of Hbs in some Ama¬
zon River fish were reported [5], In addition, the
ontogenetic changes of a few kinds of trout Hbs
were also reviewed recently [6], During the past
few years, the Hbs from the blood of seven kinds
of teleosts of the family Cyprinidae, i.e., crucian
( Carassius auratus L.), carp ( Cyprinus carpio ),
blunt-snout bream ( Megalobrama amblycephala ),
grass carp ( Ctenopharyngldon idellus ), black carp
(Mylopharyngldon piceus ), silver carp ( Hypoph -
thalmichthys molitrix) and variegated carp ( Ari -
stichys nobilis ) have been investigated as one of
the biochemical markers for identifying their spe¬
cies-dependent characteristics in our laboratory by
using starch gel electrophoresis [7]. The starch gel
electrophoretograms of these fish Hbs showed that
crucian Hb has only one major band while all the
other Hbs have two to four major bands. This

Accepted December 18, 1987
Received December 22, 1986

1 This work was supported by research grants from the
Chinese Academy of Sciences, and RF 84031 from the
Rockefeller Foundation, U.S.A.

2 To whom requests of reprints should be addressed.

clearly demonstrates that polymorphism is com¬
mon to these fish Hbs except crucian. This unique
characteristic of crucian Hb stimulated us to make
further investigations on its structural properties.

This paper reports the isolation of crucian Hb
and its subunits, and the identification of their
structural characteristics.

MATERIALS AND METHODS

Crucian ( Carassius auratus L.) of the Family
Cyprinidae, Genus Carassius Nilsson, is one of the
most familiar, edible fresh-water fish in China.
Crucians of about one year age weighing about 500
g were bought from Beijing Wan Quanzhong
Fishery in different seasons. DEAE-cellulose was
purchased from Whatman (DE-52). Methanol
and acetonitrile (HPLC grade) were products of
Zhejiang Huangyan Chemical Factory.

Isolation and purification of crucian Hb

The crucian hemolysates were prepared by the
method of Jeppssen et al. [8]. Crucian Hb was
isolated from the hemolysates by using DEAE-
cellulose column chromatography. A brief de¬
scription of the chromatography is given in the
legend of Figure 1.

Separation of crucian globin subunits

Heme was removed by the acid-acetone proce¬
dure described by Rossi et al. [9]. The constituent
chains of crucian Hb were separated by using

810

F. Hung and Y. Shaoyi

Fig. 1. Chromatography of crucian Hb on DEAE-
cellulose column. Crucian hemolysates were applied
to a DEAE-cellulose column (1 X 15 cm) which had
been equilibrated with 50 mM Tris-HCl buffer, pH
8.6. The materials were eluted with a linear gradient
obtained from 300 ml each of the Tris buffer and 0.2
M NaCl solution at contant pH of 8.6. Every 3 ml of
the effluent was collected in each tube at a flow rate
of 36 ml/hr.

Waters HPLC system according to Shelton et al.
[10] with a minor modification. The chain separa¬
tion was carried out by reverse phase chromatogra¬
phy on a Waters /d3ondapak Qg column (10 pm, 4
X 300 mm); specific conditions for elution are
given in the legend of Figure 2.

Characterization of some structural properties of
crucian globin subunits

The urea-Triton acid polyacrylamide disc gel
electrophoresis method of Alter et al. [11] was
used for identification of crucian globin subunits.

Analysis of the N-terminal sequences of globin
subunits was carried out by the double-coupling

method of Chang et al. [12]. The produced
DABITH-terminal amino acid residues were de¬
tected on a polyamide filter.

RESULTS AND DISCUSSION

A typical elution profile of crucian hemolysate
on a DEAE-cellulose column is shown in Figure 1.
This profile demonstrates that crucian Hb consists
essentially of a single type which was detected at
540 nm. The peak appeared between fractions No.
22-25 in accordance with the sites for 0.022-0.025
M Na+ concentration. The yield was 86%.

From the above results, since only a single type
of Hb was isolated, it looks clear that there is no
obvious polymorphism in crucian Hb, while the
polymorphism of Hbs have been commonly re¬
ported in various kind of fish. These results are
also consistent with those reported by Braunitzer
et al. [13]. They analysed the amino acid se¬
quences of a- and /9-chains of a major component
of goldfish Hb, and no indication of polymorphism
of goldfish Hb could be found in their report.
Since crucian is the wild type of the species of
Carassius auratus L. while the goldfish is the
domestic variety of the same species, it is reason¬
able to think that the basic structure of Hb will
reveal the possibility in keeping its conservation
nature during their long history in evolution. Any¬
way, this non-poly morphic characteristic of cru¬
cian and goldfish Hbs is a rare case among the fish
Hbs.

Fourie et al. [17] reported the phenomenon that
the types of carp Hb varied in different seasons.
He suggested that it might be due to the globin
gene expression which was regulated and control¬
led by the change of environmental factors. In our
experiment, no distinct seasonal changes in crucian
Hb were observed.

Two globin subunit chains of crucian hemoglo¬
bin were separated by reverse phase HPLC as
shown in Figure 2. Each peak was identified by
urea-Triton-PAGE to be /?- and a- type globin
chain in the elution order (Fig. 3). According to
the migration rates of two globin chains on the
urea-Triton gel, the first peak was identified as
/?-type globin chain and the second a-type. This
result is similar to the results obtained by other

Characterization of Crucian Hb Subunits

811

Fig. 2. HPLC of crucian globin subunits on C18 column.

Buffer and solvents: Phosphate buffer (P): 6.66 g/1 of KH2P04, pH 2.5; Acetonitrile (A);
Methanol (M). Solvent A consisted of 42.5% (A): 5% (M): 52.5% (P). Solvent B consisted of
50% (A): 5% (M): 45% (P). The globin sample was dissolved in solvent A and the insoluble
materials were removed by centrifugation prior to application to the column. The column was
washed with solvent A during the first 23 min then a linear gradient of Solvent B from 0% to
100% was employed for the elution over the subsequent 20 min. The column was washed
with 100% solvent B to the end continuously. The flow rate was 1.5 ml/min. Proteins were
detected at 280 nm.

a b c

Fig. 3. Electrophoretograms of crucian globin subunits
on urea-Triton acid PAGE.

a: Main fraction in Fig. 1, b: The first peak in Fig.
2 (/?-chain). c: The second peak in Fig. 2 (ar-chain).
12% polyacrylamide gels containing 6 M urea and
2% Triton X-100 in 5% acetic acid were used. The
gels were laid with 10 /ug of proteins for each tube.
Electrophoresis was run in 5% acetic acid following
the procedure of Alter et al [11].

researchers [13-15] although it seems different
from Shelton’s results [10] in which the first peak
of human Hb was identified as a-chain and the
second one as /?-chain. The reason for this differ¬
ence is still unknown.

It has been reported that unlike mammalian
Hbs, fish Hb tetramers in some species seem to be
composed of two to four different subunit chains
[16]. However, our results show that there are
only two common subunit types, a-type and /?-type
in the crucian Hb.

The first three N-terminal amino acid sequences
of the a- and /?-chains of crucian Hb were both
identified as Val-Glu-Try. Data of sequences for
several N-terminal amino acids of fish globins are
listed in Table 1. It indicated that both the N-
terminal amino acids of a- and /?-c hains in carp,
catomostomid, trout and goldfish have the feature,
Ac-Ser-Leu-Ser for a-chain and Val-Glu-Try for
/?-chain. In contrast, the N-terminal amino acids of
crucian a-chain are different from those of other
fish Hbs. The unusual N-terminal sequence of
crucian /?-chain is of particular interest because it
differs significantly from that of goldfish a-chain
[3]. The goldfish was separated from the crucian
through an artificial selection in China about six

812

F. Hung and Y. Shaoyi

Table 1. Comparison of N-terminal amino acid sequences of fish Hbs

a-chain

/?-chain

Ref.

Crucian

( Carassius auratus L.)

Val-Glu-Try

Val-Glu-Try

This report

Carp

(' Cyprinus carpio )

Ac-Ser-Leu-Ser

Val-Glu-Try

[1], [18]

Catostomid
{Catostomus clarkii )

Ac-Ser-Leu-Ser

Val-Glu-Try

[2]

Trout

{Salmo irideus )

Ac-Ser-Leu-Ser

Val-Glu-Try

[4]

Goldfish

{Carassius auratus )

Ac-Ser-Leu-Ser

Val-Glu-Try

[3]

hundred years ago. Is it possible that such a
domestic selective process caused the difference of
the N-terminal sequence between the ochains of
crucian and goldfish Hbs? In order to clarify this
question, further investigation is necessary. Espe¬
cially it should be focused on the identification and
comparison of the complete amino acid sequences
for both fish Hbs.

ACKNOWLEDGMENTS

The authors wish to thank Professor Huang Jifang,
Institute of Developmental Biology, Academia Sinica for
his reading the English manuscript and Mr. Xu
Xiochang, Institute of Biophysics, Academia Sinica for
his assistance in analysing the N-terminal amino acid of
crucian globins.

REFERENCES

1 Gillen, R. G. and Riggs, A. (1972) Structure and
function of the hemoglobins of the carp, Cyprinus
carpio. J. Biol. Chem., 247: 6036-6046.

2 Powers, D. A. and Edmundson, A. B. (1972) Mul¬
tiple hemoglobins of catostomid fish II. The amino
acid sequence of the major a chain from Catostomus
clarkii hemoglobins. J. Biol. Chem., 247: 6694-
6707.

3 Braunitzer, G. and Rodewald, K. (1980) Die Se-
quenz der a- und /?-ketten des Hamoglobins des
Goldfisches ( Carassius auratus). Hoppe-Seyler’s Z.
Physiol. Chem., 361: 587-590.

4 Bossa, F., Barra, R., Petruzzelli, R., Martini, F.
and Brunori, M. (1978) Primary structure of hemo¬
globin from trout ( Salmo irideus ): amino acid se¬
quence of a chain of Hb trout I. Biochim. Biophys.
Acta, 536: 298-305.

5 Unni, E. H. Fyhn, Fyhn, H.J., Davis, B.J., Pow¬
ers, D. A., Fink, W. L. and Garlick, R. I. (1979)

Hemoglobin heterogeneity in Amazonian fishes.
Comp. Biochem. Physiol., 62 A: 39-66.

6 Iuchi, I. (1985) Cellular and molecular bases of the
larval-adult shift of hemoglobins in fish. Zool. Sci.,
2: 11-23.

7 Jingzhi, Y., Guon, W. and Shaoyi, Y. (1986)
Analysis of starch gel electrophoresis patterns of
hemoglobin and red blood cell LDH isozymes of
four kinds of fresh water teleost ( Mylopharyngodon
piceus, Ctenopharyngodon idellus , Hypophthalmich-
thys molitrix , Aristichthys nobilis). Hereditas (Beij¬
ing), 8(2): 25-27.

8 Jeppssen, J. O. (1977) Application Note, 307. LKB
Inc.

9 Rossi, F. A. and Antonini, E. (1958) Studies on the
structure of humoglobin. 1. Physicochemical prop¬
erties of human globin. Biochem. Biophys. Acta,
30: 608-615.

10 Shelton, J. B. and Shelton, J. R. (1979) Preliminary
experiments in the separation of globin chains by
HPLC. Hemoglobin, 3: 353-358.

11 Alter, B. P., Goff, S. C., Eqremov, G. D., Gravely,
M. E. and Huisman, T. H. J. (1980) Globin chain
electrophoresis: A new approach to the determina¬
tion of the Gy/A/ ratio in fetal haemoglobin and to
studies of globin synthesis. Br. J. Haematol., 45:
527-534.

12 Chang, J. Y. and Creaser, E. H. (1976) A novel
manual method for protein-sequence analysis.
Biochem. J., 157: 77-85.

13 Congote, L. F. and Kendall, A. G. (1982) Rapid
analysis of labeled globin chains without acetone
precipitation or dialysis by high-pressure liquid chro¬
matography and ion-exchange chromatography.
Anal. Biochem., 123: 124-132.

14 Shelton, J. B., Shelton, J. R. and Schroeder, W. A.
(1984) High performance liquid chromatographic
separation of globin chains on a large-pore C4
column. J. Liquid Chromatogr., 7: 1969-1977.

15 Huisman, T. H. J. and Webber, B. (1981) In “Ad¬
vances in Hemoglobin Analysis”. Ed. by S. M.

Characterization of Crucian Hb Subunits

813

Hannsh and G. J. Brewer. Alan R. Liss Inc., New
York, p. 23.

16 Fyhn, U. W. H. and Sullivan, B. (1974) Hemoglo¬
bin polymorphism in fishes, I. Complex phenotypic
patterns in the toadfish, Opsanus tau. Biochem.
Genetics, 11: 1974.

17 Fourie, F. L. and VanVuren, J. H. J. (1976) A sea¬
sonal study on the hemoglobins of carp ( Cyprinus

carpio ) and yellowfish ( Barbus holubi) in South
Africa. Comp. Biochem. Physiol, 55B: 523.

18 Hilse, K. and Braunitzer, G. (1966) Zur Phylogenie
des Hamoglobinmolekiils fiber den Polymorphismus
und die N-terminalen Aminosauren des Karpfenha-
moglobins. Hoppe-Seyler’s Z. Physiol. Chem., 344:
166-168.

-

ZOOLOGICAL SCIENCE 5: 815-822 (1988)

© 1988 Zoological Society of Japan

Meiotic Studies of Interracial Hybrids from the Wild
Population of the Large Japanese Field Mouse,
Apodemus speciosus speciosus

Masako Saitoh* 1 and Yoshitaka Obara2

Department of Biology, Faculty of Science, Hirosaki University,
Hirosaki 036, Japan

ABSTRACT — The meiosis and gametogenesis were examined in interracial hybrids (Rb heterozygotes,
2n=47) between two karyotypically different races A (2n = 48) and B (2n=46) of Apodemus speciosus
speciosus. Histologically gametogenesis appeared to be normal in either sex. The anomalous
segregation in anaphase I disjunction was relatively higher in the male Rb heterozygotes (27.9%) than in
a race A male (6.2%), though the formation of a chain trivalent between Rb metacentric and twin
acrocentric chromosomes was observed in the hybrids. The post-reductional segregation of the XY
bivalent was confirmed. These observations suggested that the parental forms (races A and B) have
been semiisolated by the hybrid inferiority with a narrow hybrid zone and that the meiotic misdivision in
the Rb heterozygotes may, at least in part, function as a post-mating isolation mechanism for
maintaining the parapatric distribution of these two chromosomal races.

INTRODUCTION

The large Japanese field mouse, Apodemus spe¬
ciosus speciosus, consists of two chromosomal
races (race A and race B) which have been formed
through a Robertsonian (Rb) rearrangement [1,
2]. They distribute parapatrically with a narrow
contact zone named “Toyama-Hamamatsu line”:
race A (2n=48) ranges to the northward of this
line and race B (2n=46) to the southward [3-6].
The interracial hybrids (Rb heterozygotes) can
easily be detected within the contact zone [4-6].

Chromosomal races produce relatively unfit
heterozygotes which therefore leave fewer prog¬
eny, and ultimately the pure genotypes may be
maintained by their hybrid inferiority which is
expected to be equivalent to the vagility [7]. Thus,
in general interracial hybrids can be highly ex¬
pected to undergo meiotic drive, which causes the
production of unbalanced gametes. Similar situa¬

Accepted October 26, 1987
Received September 21, 1987

1 Present address: Chromosome Laboratory, Shiseikai
Dai-ni-Hospital, 5-19-1 Kamisoshigaya, Setagaya-ku,
Tokyo 157, Japan.

2 To whom reprints should be requested.

tion would also be expected in A. s. speciosus : the
fitness of the interracial hybrids (2n = 47) may be
reduced to a certain degree due to unusual seg¬
regation during meiosis.

Meiotic anaphase I malsegregation attributable
to the nondisjunction of the Rb trivalents has been
observed in the mice heterozygous for Rb
metacentrics [8-13], Such meiotic drive system
would be one of the most likely factors for induc¬
ing partial sterility of the interracial hybrids [14,
15]. Therefore, it may be indispensable for inquir¬
ing into the nature of the raciation mechanism of
A. s. speciosus to examine in detail the malseg¬
regation rate in meiosis of the interracial hybrids,
or Rb heterozygotes. Up to now, no practical
reproduction analysis of the interracial hybrids of
A. s. speciosus has been made except for the cross
breeding experiments of Tsuchiya [16], by which
he regarded the interracial hybrids to possess “very
good” fertility.

The aim of this report is to investigate the
fertility in the wild specimens of the interracial
hybrids of A. 5. speciosus from the following
viewpoints: (1) spermatogenesis and oogenesis, (2)
meiotic pairing and (3) meiotic anaphase I malseg¬
regation.

816

M. Saitoh and Y. Obara

In some Apodemus species including A. s. spe-
ciosus, the post-reductional segregation of the sex
chromosomes, which is a quite unique reduction
system in mammals, has been suggested by several
workers on the basis of the examination of the
testis sections [17-23]. In the present study the
reduction system of A. s. speciosus will also be
analyzed by the air-drying and C-banding
methods.

MATERIALS AND METHODS

Wild-caught specimens of the three karyological
forms of Apodemus speciosus speciosus were used
in this study; race A-type mice were from Aomori
and Nagano Prefs., race B-type ones from Tottori
and Nagano Prefs. and their hybrids from Ina,
Nagano Pref. These karyological forms were de¬
termined by their bone marrow karyotypes.

Histologic examination : Testes and ovaries
(race A-5 $ , 3-?-; hybrid-3 7-£; race B-2$,
2 -£) were kept until use in PFA-3 fixative. After
paraffin-embedding, the gonads were serially sec¬
tioned at 8 /an, and stained with Mayer’s haema-
toxylin and eosin.

Meiotic preparation : Fourteen male specimens
(race A-6 $ , hybrid-6 $ , race B-2 $ ) were pro¬
vided for meiotic study. For meiotic preparation,
the protocol of Evans et al. [24] was adopted. The
flame-dried preparations were conventionally
stained with Giemsa for chromosome counting and
the air-dried ones were C-banded by the BSG

Fig. 1. Gonadal transection of the interracial hybrids of
and (b) oogenesis in the ovary. Scale bar, 10 /an.

method [25].

Chiasma counts : Fifty-eight metaphases of the
first meiotic division (M I) from three male Rb
heterozygotes were photographed under micro¬
scope, and the interstitial chiasmata of the Rb
trivalents were counted with the printed M I
figures.

M II counts : The nondisjunction rate in the M
I was evaluated by scoring the frequency of the
heteroploid (hypo- and hyperploid) secondary
spermatocytes of a specimen of race A and two
hybrids. Seven hundred and twenty-two well-
spread metaphases of the second meiotic division
(M II) were photographed and the printed figures
of the M II metaphases were carefully traced as
line drawings for reliable and correct identification
of the chromosomes.

RESULTS

Histology of the testis and ovary

Figure 1 shows the spermatogenesis (a) and
oogenesis (b) of the interracial hybrids, respective¬
ly. Within the seminiferous tubules various stages
of the spermatogenic cells such as spermatogonia,
primary and secondary spermatocytes, spermatids
and spermatozoa were in rows in proper order
from the basal lamina to the lumen. All the testes
of the hybrids examined were apparently normal in
the cellular morphology as well as in the constitu¬
tion of the germ-lines. Moreover, just as with the

A. s. speciosus: (a) germ-lines in the seminiferous tubules

Meiosis of Apodemus s. speciosus

817

Table 1. Counting of CH III caused by the interstitial chiasmata in three male Rb
heterozygotes at late diakinesis/M I

a

b

c

d

e

f

Total

H-l

6

2

1

1

3

13

H-2

3

1

—

■■

—

4

H-3

17

10

6

7

—

1

41

Total(%)

26(44.8)

13(22.4)

7(12.1)

8(13.8)

3(5.2)

1(1.7)

58

a, b, c, d, e and f correspond to those of Fig. 3.

parental forms, some bunches of the mature
sperms connected to the Sertoli’s cells. In seven
female specimens of the interracial hybrids, the
transition of small primary follicles to secondary
ones was almost the same as that of the parental
forms, race A and race B (Fig. lb). No anomalous
feature could be detected, at least at the histologi¬
cal level, in the gametogenic cells of the interracial
hybrids, irrespective of age and sex.

Meiotic pairing

Chromosomal compatibility in the interracial
hybrids can be evaluated through meiotic normal¬
ity. In the present Rb heterozygotes, a typical
chain trivalent (CH III) between the Rb
metacentric (SMI; refer to the previous work [6])
and the twin-acrocentrics (Nos. 10 and 17) was
detected without exception in addition to the twen¬
ty-one autosomal and one XY bivalents which
were quite identical to those of the parental forms
(Fig. 2). Such a perfect meiotic pairing in the Rb

heterozygotes which can be well expected from a
high degree of G-band homology between the two
parental forms [6] may allow the subsequent nor¬
mal gametic differentiation. However, the chias-
ma configuration in the CH Ills varied from cell to
cell, as presented in Figure 3 a to f. The occurr¬
ence frequency of interstitial chiasma was 68.9%
(a + b + f) in the side of the short arm-No. 17 and
62.1% (a-f-c + e) in that of the long arm-No. 10
(Table 1). What the chiasma-like loose-
associations (Fig. 3 e and f) stand for, that is,
whether they resulted from asynapsis or desynapsis
was still uncertain.

Meiotic anaphase I malsegregation

The M II spreads of the male Rb heterozygotes
consisted of seven classes of chromosome constitu¬
tion; n = 23 with SMI, n = 24 without SMI, n = 24
with SMI, n = 23 without SMI, n = 25 with SMI, n
=22 without SMI and the other types. According
to the classification of Ford and Evans [10], the

Fig. 2. (a) Diakinesis/M I of an interracial hybrid showing a normal configuration including an Rb CH III (arrow)

and a sex bivalent (arrow head), (b) A diagram of a typical Rb CH III.

818

M. Saitoh and Y. Obara

a b c d e f

Fig. 3. Demonstration of the Rb CH Ills carrying chi-
asmata in Nos. 10 and 17/either arm of SMI (a), one
chiasma either in the short arm/No. 17 (b) or the
long arm/No. 10 (c), no chiasma (d) and loose-as¬
sociations of No. 17 (e) or No. 10 (f).

first two can be attributed to “normal disjunction”,
the third and fourth to “partial nondisjunction”
and the fifth and sixth to “total nondisjunction”.
The former six are schematically depicted in Fi¬
gure 4, A, B and C, in which the unbalanced

haploid conditions are owing to the absence or the
surplus of the chromosome arms taking part in the
formation of the trivalent, and the last one in¬
cludes the aneuploidy due to the nondisjunctions
in both of the trivalent and bivalents (Table 2, D).
Taking these conditions into consideration, the M
II spreads were examined in detail, paying atten¬
tion to the Rb metacentric and the chromosome
number (Fig. 4). The SMI of the Rb heterozy¬
gotes could be easily distinguished from the four
small metacentrics intrinsic to this species by their
relative length and arm ratio at the second metaph¬
ase (Fig. 5). X2 test showed that the data from
the two heterozygotes (H-l and FI-2) of Table 2
were homogeneous with each other. Therefore,
they were summed up. After all the anomalous
segregation occurred with a relatively high fre¬
quency, 27.9% (hyperploid: 11.2%; hypoploid:
16.7%) (Table 2). The frequency of spontaneous

Table 2. Counting of M II figures in two male Rb heterozygotes

A

B

C

D

Total cells

23

24

24

23

25

22

scored

H-l

93

124

27

36

1

4

17

302

H-2

96

126

20

35

3

6

21

307

Total

189

250

47

71

4

10

38

609

A,B and C correspond to those of Fig. 4. D includes the nondisjunctions in both of the trivalent
and bivalents. B + C+D

% nondisjunction(anomalous segregation) =— — — — — — — XlOO

@Normal D

©Partial ND

©Total ND

23

24

25

N

RlJtln

24

23

22

Ufln

(fi): in

Fig. 4. Diagrams of the segregation patterns in the Rb CH Ills: (A) normal disjunction, (B) partial
nondisjunction and (C) total nondisjunction. Upper row: SMl-carrying side, Lower row: SMI-missing
side. The Arabic numerals indicate the chromosome numbers in M II.

Meiosis of Apodemus s. speciosus

819

ft

M

1

I

*

1

SMI

SM2

M3

SM4

M 5

L

®

1 7 A

^ * H • t

i y

iCUCUUTftn,,...

1 9 A

Fig. 5. M II karyotypes from a male heterozygote (n=23, with SMI) (a) and a male homozygote of race A (n=24,
without SMI) (b). Seventeen (a) and nineteen (b) acrocentric bivalents are all symmetric in morphology, being
like a boomerang in some bivalents and straight in others. The sex chromosome is apparently asymmetric in
either case, plainly indicating the post-reductional segregation in the sex bivalent of this species.

malsegregation, if any, can be estimated to a
certain degree by analyzing that of the homozygo-
tic specimens (race A or race B). One hundred
and thirteen M II spreads of a homozygote of the
race A were examined in detail: 93.8% of the
spreads had an intrinsic haploid set of chromo¬
somes (n = 24) and 4.4% was hyperploid (n = 25)
and 1.8% hypoploid (n=23). Thus, a homozygote
of A. s. speciosus may possess ordinarily some 6%
of gametic aneuploidy as spontaneous malsegrega¬
tion.

Reduction mechanism of the sex chromosomes dur¬
ing meiosis

The conventionally stained X-Y bivalents were
demonstrated together with their schematic draw¬
ings in Figure 6 a and b. The centromeric regions
looked like small dense bodies associated side-by¬

side, and the proximal regions connecting to the
centromeres were slightly off with a certain space
between the chromatids. These configurations
were certainly regarded as the chiasmata caused by
the crossing-over at the proximal regions of the X
and Y chromosomes. The proximal regions
arranged side-by-side were darkly stained when
C-banded, as indicated by the correspondence of
the C-band pattern between the somatic and meio-
tic cells (Fig. 6c). Following crossing-over be¬
tween the chromatids of the X and Y chromo¬
somes, the centromeres of each chromosome are
splitted and the resulting sister centromeres may
shift to the opposite direction together with the
sister centromere of the partner to make their
configuration stable. In this way, the X-Y pair of
chromatids may keep on the telosynapsis at the
proximal ends (centromere regions). If this is the

820

M. Saitoh and Y. Obara

Y • fe

* % 1 4 m mm ^ f y

SM 2 M3 SM4 M5 ^ X

T- I

JjP

i { +

: C 1 (i i

* * w ^ \

i ^ „ _ / . ■**

• w

*

x 5 M

¥

C

1 2 e , '

d 1 2 '* I * - ' ^

- 8 - 19—

Fig. 6. Crossing-over and meiotic segregation of the X and Y of a male A. s. speciosus: (a) three examples of the X-Y
bivalents in late diakinesis/M I showing chiasma formation and (b) their handwritings (solid line for the X and
dotted line for the Y). (c) Coincidence of C-banding pattern between the sex bivalent (right) and the X and Y
chromosomes from a bone marrow cell (left), (d) A C-banded M II spread (right) and its karyotype (left). Note
the end-to-end pairing at the centromeric C-band regions of the sex bivalent. Darkly-stained stable C-bands are
shown by arrow heads.

case, then an asymmetric chromosome, i.e., a
chromatid pair of the X and Y, should be detected
in the M II figures by applying C-staining. Figure
6d just substantiates this notion, as clearly shown
by an arrow, and hence the post-reduction system
in the sex chromosomes of A. s. speciosus was
positively proved by C-staining for the air-dried
preparations. Since the arms of the autosomal
bivalents were all opened in morphology, carrying
small centromeric C-bands in the middle of them,
and the distal ends of the satellite chromosome
(No. 7) and the long arms of M5 were darkly
stained in particular (Fig. 6d), the application of
C-banding to the meiotic cells may not affect, at
least to a visible extent, the nature and distribution
of C-bands in comparison with those of the somatic
cells previously demonstrated [6].

DISCUSSION

It is well-accepted that Rb rearrangements may
have played a certain role in raciation and/or
speciation in a variety of animals, and in some
cases certainly played a major role in these evolu¬
tionary events of animals. In order to inquire into
their mechanisms, the meiotic process of the in¬
terracial and/or interspecific Rb heterozygotes has
been studied in several mammalian species, focus¬
ing a special attention on the isolation factors such
as the chromosomal nondisjunction, production of
aneuploid gametes and hybrid sterility [10, 11, 14,
26-28].

As revealed in this study, the male wild-caught
interracial hybrids of A. s. speciosus carry some
28% of aneuploid spermatozoa, whereas a male
race A-type homozygote only 6%. An all-

Meiosis of Apodemus s. speciosus

821

acrocentric laboratory mouse strain, NMRI, had
7. 0-9. 5% of aneuploid M II [8, 9, 13]. There
seems to be no significant difference in the fre¬
quency of the aneuploid M II between the
homozygotic specimens of A. speciosus and Mus
musculus. It was estimated, taking these findings
into consideration, that the meiotic anaphase I
malsegregation responsible for the nondisjunction
of the Rb CH III may amount to approximately
22%. In M. musculus, the malsegregation in the
male heterozygotes carrying one Rb metacentric
varied from 2% to 35% in frequency, depending
on the combination of the acrocentrics taking part
in the Rb rearrangement [9, 10, 14].

The aneuploid spermatids from the Rb heter¬
ozygotes of M. musculus are scarcely eliminated
during spermatogenesis, as revealed by measuring
the DNA content of their spermatozoa [15, 29]. In
other words these aneuploid sperms are able to get
to the ova through female genital tracts, i.e.,
uterus and Fallopian tube after copulation. This
view was ascertained by the back-cross experi¬
ments in the interracial hybrids of the house mouse
[9-11] and the common shrew [26, 27]: the pre-
and post-implantation losses in the back-cross hy¬
brids were of the same order as the frequency of
the aneuploid spermatocytes resulting from the
unusual segregation, because hyper- and hypo-
ploidy inevitably produced trisomy and mono¬
somy, respectively, both of which become extinct
practically during the early stage of the embryonic
development. These findings do imply the absence
of the selection pressure against the unbalanced
gametes in heterozygotes untill fertilization.

After all, it may also be true for the wild
interracial hybrids of A. s. speciosus that the
meiotic disturbance such as the malorientation of
trivalents and nondisjunction leads to the selective
elimination of unbalanced zygotes through the
developmental breakdown and zygotic mortality at
the early stages of development, and as a natural
consequence reduces the litter size of hybrids, up
to almost three-fourth of that in the homozygotes.
Under such condition the hybrid inferiority and
the intermigration of the parental forms (race A
and race B) might have been well balanced in a
narrow zone of hybridization, and therefore, so-
called meiotic drive (i.e., chromosomal nondis¬

junction) might have played a primary role in
maintaining stability of their parapatric distribu¬
tion.

Since the reduced fertility of the interracial
heterozygotes certainly restricts the gene flow
across the hybrid zone, the genetic discrepancy will
grow more and more with the progression of a vast
number of generations. Moreover, since the cross¬
ing-over is slightly suppressed in general in Rb CH
Ills [9, 30], the genetic variation between races will
increase in a synergistic manner. Once such a
condition of semiisolation or isolation is estab¬
lished, further diversification of genetic variation
would be possible for such newly-established
forms. Therefore, the two races of A. s. speciosus
distributing parapatrically in the northern and
southern districts with a narrow zone of hybridiza¬
tion may be most likely just in the speciation
process through the post-mating isolation arising
from such cytologic mechanisms.

The post-reductional segregation has been sug¬
gested, from the inspection of the sectioned mate¬
rials, in all Apodemus species so far examined (A.
sylvaticus, A. agrarius, A. flavicollis, A. semotus
and A. speciosus), except for A. argenteus which in
contrast shows the pre-reductional segregation
[17-23]. In the present study, the post-reduction
of the XY bivalent of A. s. speciosus was con¬
firmed by the C-banding method. So far as A.
speciosus is concerned, the post-reduction of the
XY bivalent was considered to be induced by the
crossing-over between their proximal chromatids
consisting of C-heterochromatin which probably
contain repetitive units similar to each other.
Thus, the chiasma formation in the sex bivalent
may be, at least in this field mouse, a probable
factor inducing the post-reduction of the sex
chromosomes.

ACKNOWLEDGMENT

We would like to express our appreciation to Professor
Kazuo Saitoh, Faculty of Science, Hirosaki University,
for his helpful advice and suggestion during the course of
this study and reading the manuscript with expert crit¬
icism.

822

M. Saitoh and Y. Obara

REFERENCES

1 Shimba, H. and Kobayashi, T. (1969) A Robertso¬
nian type polymorphism of the chromosomes in the
field mouse, Apodemus speciosus. Jpn. J. Genet.,
44: 117-122.

2 Tsuchiya, K. and Yosida, T. H. (1971) Distribution
of two chromosomal types of Japanese wood mouse,
Apodemus speciosus. Ann. Rep. Natl. Inst. Genet.
Japan, 21: 45-50.

3 Tsuchiya, K., Moriwaki, K. and Yosida, T. H.
(1973) Cytogenetical survey in wild population of
Japanese wood mouse, Apodemus speciosus and its
breeding. Exp. Anim., 22: 221-229.

4 Tsuchiya, K. (1974) Cytological and biochemical
studies of Apodemus speciosus group in Japan. J.
Mammal. Soc. Japan, 6: 67-87 (In Japanese, with
English summary).

5 Harada, M., Hamada, S., Koyasu, K. and Miyao,
T. (1984) Studies on a contact zone between two
chromosomal races of Apodemus speciosus. J. Mam¬
mal. Soc. Japan, 10: 101-102 (Abstract).

6 Saitoh. M. and Obara, Y. (1986) Chromosome
banding patterns in five intraspecific taxa of the large
Japanese field mouse, Apodemus speciosus. Zool.
Sci., 3: 785-792.

7 Hewitt, G. M. and Barton, N. H. (1980) The struc¬
ture and maintenance of hybrid zones as exemplified
by Podisma pedestris. Symp. Roy. Entomol. Soc.
Lond., 10: 149-169.

8 Tettenborn, U. and Gropp, A. (1970) Meiotic non¬
disjunction in mice and mouse hybrids. Cytogene¬
tics. 9: 272-283.

9 Cattanach, B. M. and Moseley, M. (1973) Nondis¬
junction and reduced fertility caused by the tabacco
mouse metacentric chromosomes. Cytogenet. Cell
Genet., 12: 264-287.

10 Ford, C. E. and Evans, E. P. (1973) Robertsonian
translocations in mice: segregational irregularities in
male heterozygotes and zygotic unbalance. Chromo¬
somes Today, 4: 387-397.

11 Gropp, A., Kolbus, U. and Giers, D. (1975) Sys¬
tematic approach to the study of trisomy in the
mouse. II. Cytogenet. Cell Genet., 14: 42-62.

12 Capanna, E., Gropp, A., Winking, H., Noack, G.

and Civitelli, M.-V. (1976) Robertsonian

metacentrics in the mouse. Chromosoma (Berl.),
58: 341-353.

13 Ferri, E. and Capanna, E. (1979) Segregation dis¬
orders in multiple heterozygous Robertsonian mice.
Acc. Lincei Rend. Mat. Fis. Nat., 66: 598-603.

14 Gropp, A. and Winking, H. (1981) Robertsonian
translocations: cytology, meiosis, segregation pat¬
terns and biological consequences of heterozygosity.
Symp. Zool. Soc. Lond., 47: 141-181.

15 Redi, C. A., Garagna, S., Pellicciari, C., Manfredi

Romanini, M. G., Capanna, E., Winking, H. and
Gropp, A. (1984) Spermatozoa of chromosomally
heterozygous mice and their fate in male and female
genital tracts. Gamete Res., 9: 273-286.

Tsuchiya, K. (1979) Notes on breeding of wood
mouse groups for laboratory animal. Rep. Hok¬
kaido Inst. Public Health, 29: 102-106 (In

Japanese).

Oguma, K. (1934) A new type of the mammalian
sex-chromosome found in a field mouse, Apodemus
speciosus. Cytologia, 5: 460-471.

Tateishi, S. (1934) A preliminary report on some
peculiar shaped chromosomes in three species of
Apodemus. Trans. Nat. Hist. Soc. Formosa, 24: 15-
17 (In Japanese).

Tateishi, S. (1935) On the sex chromosomes of
eleven species of the Muridae, etc. Proc. Jpn. Ass.
Adv. Sci., 10: 1007-1012 (In Japanese).

Matthey, R. (1936) La formule chromosomiale et
les heterochromosomes chez les Apodemus euro-
peens. Z. Zellf. mikr. Anat., 25: 501-515 (In
French).

Raynaud, M. (1936) Les heterochromosomes du
mulot. C. R. Acad. Sci., Paris, 202: 1942-1944 (In
French).

Roller, P. C. (1941) The genetical and mechanical
properties of the sex chromosomes. VII. Apodemus
sylvaticus and A. hebridensis. J. Genet., 41: 375-
389.

Makino, S. (1951) Studies on the murine chromo¬
somes. V. A study of the chromosomes in
Apodemus, especially with reference to the sex
chromosomes in meiosis. J. Morphol., 88: 93-126.
Evans, E. P., Breckon, G. and Ford, C. E. (1964)
An air-drying method for meiotic preparations from
mammalian testes. Cytogenetics, 3: 289-294.
Sumner, A. T. (1972) A simple technique for de¬
monstrating centromeric heterochromatin. Exp.
Cell Res., 75: 304-306.

Searle,J. B. (1984) Nondisjunction frequencies in
Robertsonian heterozygotes from natural popula¬
tions of the common shrew, Sorex araneus L.
Cytogenet. Cell Genet., 38: 265-271.

Searle, J. B. (1986) Meiotic studies of Robertsonian
heterozygotes from natural populations of the com¬
mon shrew, Sorex araneus L. Cytogenet. Cell
Genet., 41: 154-162.

White, M. J. D. (1978) Modes of Speciation.
W. H. Freeman & Co., San Francisco, pp. 169-226.
Redi, C. A. and Capanna, E. (1978) DNA-content
variation in mouse spermatozoa arising from irregu¬
lar meiotic segregation. Boll. Zool., 45: 315-322.
Cattanach, B. M. (1978) Crossover suppression in
mice heterozygous for tobacco mouse metacentrics.
Cytogenet. Cell Genet., 20: 264-281.

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

ZOOLOGICAL SCIENCE 5: 823-832 (1988)

© 1988 Zoological Society of Japan

Sexual Dimorphism in the Genital Tubercle of the Duck:
Studies on the Normal Development and Histogenesis

Hideho Uchiyama and Takeo Mizuno* 1

Zoological Institute, Faculty of Science, University of Tokyo,
7-3-1 Hongo, Bunkyo-ku, Tokyo 113, Japan

ABSTRACT — The development of the genital tubercle of the duck embryo was studied by light and
scanning electron microscopy. The genital tubercle started to develop from the anterior region of the
cloaca at day 7 in both sexes, and was initially consisted of the epidermis, the mesenchyme and the
urethral plate. A blind duct became separated from the urethral plate between day 10 and day 12 in a
proximal to distal direction. Marked sexual dimorphism started at day 10. Thereafter, the male genital
tubercle grew spirally and proceeded to the histogenesis of a pair of cavernous bodies, an elastic
ligament, a blind duct, and so on. Whereas female genital tubercle shrank continuously and did not
proceed to any more histogenesis. The shrinkage of the female genital tubercle was assumed to be due
to a sharp decrease of the mitotic activity and the occasional cell death in the mesenchyme of the genital
tubercle.

INTRODUCTION

Among birds, male duck, goose, swan, and
ostrich have an exceptionally well-developed
copulatory organ, Phallus protrudens [1]. The
anatomy of this organ in the adult duck has been
well described [2-6]. However, the process of the
development of this organ has not been described
enough. Pomayer [7] outlined the development of
the genital tubercle at its early stages in the male
but not the female embryos. Wolff [8] described
the genital tubercle in both sexes along the course
of development, but mentioned little on the histo¬
logical aspect.

The genital tubercle of the duck had been a good
meterial for studying the avian sexual dimorphism.
The male genital tubercle regresses to become the
female type by the application of estrogen while
the female one overgrows to become the “pseudo”
male type by the application of androgen [8] . The
genital tubercle develops into the male type
irrespective of the sex of the embryo when the

Accepted October 28, 1987
Received October 13, 1987

1 Present address: Department of Biology, Faculty of
Pharmaceutical Sciences, Teikyo University, Sagami-
ko-machi, Kanagawa 199-01, Japan.

embryo is gonadectomized [9] or when cultured in
vitro early in its development [10]. Hence, the
male type has been considered a “neutral” type
and the female type, an “induced” type by
estrogens.

As a first step toward analyzing the action of sex
hormones in the genital tubercle of the duck
embryo, we studied the ontogeny of the sexual
dimorphism histologically and by scanning elec¬
tron microscopy. We obtained the result that the
regression of the female genital tubercle surely
accompanies the gradual decrease of its volume,
and that the decrease is probably due to a decrease
of the mitotic activity and to the occasional cell
death in the mesenchyme.

MATERIALS AND METHODS

Animals

Fertilized eggs of Pekin and Aokubi duck were
incubated in a humidified incubator at 37.5°C. In
these conditions, most of the egg hatched on day
27, and a few on day 28. Embryos were staged
according to Kaltofen [11], The sex of the embryos
was determined by the macroscopical observation
of the gonad. This method was only available after
day 11.

824

H. UCHIYAMA AND T. MlZUNO

Histology

Cloacal tissues were excised and fixed in Bouin’s
fluid. Embryos of older than 16 days were anesthe¬
tized in ovo for a few minutes with chloroform or
diethylether before dissection. Fixed specimens
were transferred to 70% ethanol, dehydrated in a
graded series of buthanol: ethanol mixtures, stored
in absolute buthanol, embedded in paraffin, and
sectioned at 5 pm. Sections were stained with
hematoxylin and eosin, or Heidenhain’s Azan
method.

Scanning electron microscopy
Excised cloacal tissues were fixed in 2.5% glu-

6 5 4

taraldehyde in 0.1 M cacodylate buffer (pH 7.4)
for 2 hr at 4°C, postfixed in 1% osmium tetroxide
in the cacodylate buffer for 2 hr at 4°C, dehydrated
with graded ethanol series, transferred into
isoamyl acetate, dried in a critical point drier with
liquid C02, mounted onto alminum stabs, coated
with gold in an ion spatter, and examined with a
Hitachi S430 scanning electron microscope.

Measurement of the volume with serial sections

The outlines of serial frontal sections (the outer
line of the epidermis) of the genital tubercle of 5
pm thick were drawn on sheets of paper at every 5
sections by a camera lucida at the magnification of
XlOO to X200. These lines were traced with a

Fig. 1. Definition of the boder of the genital tubercle among other cloacal tissues. A scheme of a mid-sagittal section
and several frontal sections of the genital tubercle at 10 days of incubation are shown. Anterior is to the top. The
urethral plate (Up) elongates more anteriorly in deeper regions. Therefore, the anterior limit of the urethral
plate (B) crosses the plane made by the anterior limits of the right and left proctodeal wall (A and C) at some
depth in the cloaca (depth-5 in the figure). The genital tubercle was defined as the region more distal to the
depth-5, and more posterior to the plane made by the anterior limits of the proctodeal wall (dotted portions).
Although this criterion seems to be meaningless from the anatomical point of view, it is a practical way to
discriminate the undifferentiated and borderless genital tubercle from surrounding tissues. Ur, urodeum; Pr,
proctodeum; Cr, cloacal rim.

Development of Duck Genital Tubercle

825

cursor on a degitizer (K-510 mk2, Kanto Denshi,
Japan) connected with a microcomputer (PC-
9801, NEC, Japan). As the degitizer emitted a
stream of X, Y coodinates at every 0.5 sec, the
figure enclosed by a smooth line was converted
into a polygon with 30 to 100 apexes, whose inner
area was computed. The volume of the section was
calculated by multiplying the area by the thickness
(5 pm). The volume of a genital tubercle was
approximated with the summation of the volume
of each unit (5 times the volume of each section).
The definition of the border between the genital
tubercle and the proctodeal wall is shown in Figure
1.

Colcemid treatment

With a 25 p\ microsyringe (1TO, Japan), 1.2
pg/g body weight of colcemid (Nakarai, Japan; 66
pglm\ in Tyrode’s solution) was dropped onto the
chorio-allantoic membrane just above the embryo
through a window (about lxl cm) opened on a
shell. The body weight of our embryos was
presumed from the data by Romanoff [12]. After
the treatment, the window was sealed with a
cellophane tape, and the egg was incubated for
another 4 hr. The genital tubercle was excised,
fixed in Bouin’s, and embedded in paraffin. Serial
sections were cut frontally at 5 pm. The mitotic
activity in the genital tubercle was calculated as
shown in the legend of Figure 7.

Histochemical staining for acid phosphatase

Excised genital tubercles were fixed in 2.5%
glutaraldehyde in 0.1 M cacodylate buffer (pH 7.4)
for 2 hr at 4°C, washed in two or three exchanges
of Apathy’s gum syrup for several days at 4°C,
embedded in Tissue Tek II (Miles laboratories,
USA), frozen in liquid isopentane cooled with
liquid nitrogen, and cut at 10 pm. in a cryostat
(Cryocut II, American Optical, USA) at — 20°C.
Sections were floated in the ice-cold cacodylate
buffer and stained with the azo-indoxyl method

[13].

Nomenclature

Anatomical nomenclature was mainly according
to King [6].

RESULTS

Observations by the scanning electron microscopy

a) The genital tubercle during sexually uni-
morphic phase

At day 6 of incubation, a swelling is recognized
between the tail and the belly (Fig. 2a). This
swelling becomes the genital tubercle at day 7 (Fig.
2b). The genital tubercle first grows posteriorly
(Fig. 2c), then gradually turns anteriorly around its
base (Fig. 2d). The seminal groove (the anlage of
Sulcus phalli, the ejaculatory groove) appears
along the midline of the posterior surface at day 8.
Although the sex of the embryo was not deter¬
mined before day 11, more than 10 samples of the
genital tubercle showed similar appearances at
each stage, suggesting that the external morpholo¬
gy of the genital tubercle before day 11 is sexually
unimorphic.

b) Development of the male genital tubercle
after day 11

The male genital tubercle grows spirally from
day 11 onward (Figs. 2e, 3a, and 3c). The rotation
occurs only in the proximal part. The distal part
becomes glans-like in appearance. Primitive style
of Rugae phalli, the circular folds, appears from
about day 20.

c) Development of the female genital tubercle
after day 11

The female genital tubercle is often slightly
twisted at day 11. However, the growth and spiral
morphogenesis never proceed further. Conse¬
quently, the female genital tubercle keeps a similar
appearance from day 12 to day 18 (Figs. 2f, and
3b). From day 18, many irregular furrows appear
on the epidermis of the proximal part (Fig. 3d).
The genital tubercle of a newly hatched duckling is
recognized as like a short rod derived from the
distal part. The proximal part seems to be almost
“absorbed” into the proctodeal wall (Fig. 3e). The
female genital tubercle keeps this short rod-like
structure at least 2 months after hatching.

Observations by the light microscopy
a) The genital tubercle during sexually unimor¬
phic phase

At day 5, a small swelling is recognized just

826

H. UCHIYAMA AND T. MlZUNO

Fig. 2. Development of the genital tubercle as observed by scanning electron microscopy, I. Anterior is to the top.
(a) The cloacal region at day 6. A swelling (marked by an arrowhead) appears between the tail and the belly. X
73. (b) The cloacal region at day 7. Gt: genital tubercle; Cr: cloacal rim. x60. (c) A genital tubercle at day 8. X
70. (d) Lateral view of a genital tubercle at day 10. Seminal groove is seen on the posterior surface. X70. (e)
Caudal-lateral view of a male genital tubercle at day 12. The proximal part of the genital tubercle has begun to
grow spirally. In this specimen, the rotation angle is almost 45°. X 92. (f) Caudal view of a female genital tubercle
at day 12. Note the small distal portion compared with the male one (e). X92.

Development of Duck Genital Tubercle

827

Fig. 3. Development of the genital tubercle as observed by scanning electron microscopy, II. Anterior is to the top.
(a) Lateral view of a male genital tubercle at day 18. X56. (b) Caudal view of a female genital tubercle at day 16.
X80. (c) Caudal view of a male genital tubercle of a newly hatched duckling. Rugae phalli (R), the undulations
on the epidermis, are seen. S: Sulcus phalli. X34. (d) Caudal view of a female genital tubercle at day 20.
Irregular furrows (arrowheads) are seen on the epidermis. X 120. (e) Caudal view of a female genital tubercle of
a newly hatched duckling (an arrow). The distal part is retained as a rod-like protrusion, whereas the proximal
part is almost absorbed into the cloacal wall. X41.

anterior to the cloacal membrane. This swelling
gradually grows posteriorly with the cloacal mem¬
brane, forming the genital tubercle by day 7.
Accordingly, the urodeum forms a pocket-like
protrusion into the genital tubercle (Fig. 4a). This

pocket-like structure gradually becomes flatter
along the mid-sagittal plane of the body by day 9,
so that the “pocket” becomes a plate-like structure
to which we will refer as the “urethral plate”
hereafter in this paper. Near the urethral plate,

828

H. UCHIYAMA AND T. MlZUNO

Development of Duck Genital Tubercle

829

mesenchymal cells are dense and parallel to the
plate. From day 10, a blind duct (’der Blind-
schlauch’ in Pomayer [7]) becomes separated from
the urethral plate in a proximo-distal sequence
(Fig. 4b). The basement membrane becomes
obscure at the middle of the urethral plate, and the
plate becomes separated. Simultaneously mesen¬
chymal cells fill the gap between the duct and the
plate. Again, more than 10 samples at each stage
showed similar histology, suggesting that the geni¬
tal tubercle is sexually unimorphic histologically
before day 10.

b) Development of the male genital tubercle
after day 10

The separation of the urethral plate is completed
by day 12. The resulting blind duct becomes
equipped with a little but distinct lumen from day
11. Mesenchymal cells become concentrically
arranged around the blind duct. From day 16, the
epidermis covering the genital tubercle multiplies
its layers earlier than the epidermis in other re¬
gions of the body does. In the inner layer of the
urethral plate, narrow extracellular spaces grad¬
ually appear and fuse one another to open to the
outside along its entire length, resulting in the
formation of Sulcus phalli. Simultaneously, in the
mesenchyme of the proximal part, Corpus fibro-
lymphaticum dextrum and sinistrum, a pair of
cavernous bodies, begin to be formed. From day
18, Ligamentum elasticum phalli can be recognized
as a collagen-rich dermal cell aggregates in the
Corpus fibrolymphaticum dextrum. Additionally,
Suspensorium phalli, also a collagen-rich tissue
connected to and parallel with the blind duct, can
also be recognized (Fig. 4c). In contrast, the
mesenchyme of the distal part remains undiffer¬
entiated. From day 20, the epidermis becomes
thick at each Rugae phalli. The epithelium of the
blind duct is two layered at day 25. This epithe¬
lium multiplies its layers after hatching, but does

not differentiate into glandular cells before a week
after hatching.

c) Development of the female genital tubercle
after day 10

The separation of the urethral plate is completed
by day 12. However, the resulting blind duct lacks
a distinct lumen and ordered close cell contacts
which are characteristic to the epithelial tissues.
Therefore, the blind duct is a mass of cells rather
than a duct (Fig. 4e). Along with the separation of
the urethral plate, disintegration of the blind duct
proceeds also in a proximal to distal sequence from
day 10, so that the duct becomes progressively
shorter. After day 10, cells with completely
basophilic nuclei and therefore assumed to be
dying, are observed among viable cells. Among
these seemingly-dying cells are those with com¬
pletely basophilic nucleus of almost normal size,
with several middle-sized fragments of heterochro¬
matin, and with a tiny fragment of heterochroma¬
tin and barely recognizable cytoplasm (Fig. 4d).
Acid phosphatase-positive cells are also prominent
in the female genital tubercle (Fig. 5b), and its
distribution and density are quite consistent to
those of dying cells. Since phagocytes are only
infrequently observed, this consistency suggests
that the dying cells have acid phosphatase activity.
Such cells are especially prominent in the mesen¬
chyme around the blind duct, but are also dis¬
tributed everywhere in the mesenchyme of the
genital tubercle. The proportion of the dying¬
looking cells in the mesenchyme at day 12 is
roughly 3.5% of total cells. After day 16, the
epidermis multiplies its layers as in males. Howev¬
er, the histogenesis in the female genital tubercle
does not proceed further (Fig. 4d), and Rugae
phalli is never formed.

Fig. 4. Histology of the developing genital tubercle. Anterior is to the top. (a) The cloacal region at day 7. The
urodeum forms a pocket-like protrusion (U) in the genital tubercle. Cm: cloacal membrane. Hematoxylin and
eosin. X 110. (b) A frontal section of a genital tubercle at day 10. The urethral plate is separated into a blind duct
(Bd) and a “definitive” urethral plate (Up). Hematoxylin and eosin. X270. (c) A frontal section of a male genital
tubercle at day 25. 1: Corpus fibrolymphaticum dextrum-, 2: Corpus fibrolymphaticum sinistrum-, 3: Suspensorium
phalli', 4: Ligamentum elasticum-, 5: Sulcus phalli; 6: the blind duct. Azan. X86. (d) Dying cells (arrowheads) in a
female genital tubercle at day 12. D: the blind duct. Hematoxylin and eosin. X610. (e) A frontal section of a
female genital tubercle at day 16. Azan. X90.

830

H. UCHIYAMA AND T. MlZUNO

Fig. 5. Histochemical staining for acid phosphatase in the genital tubercle at day 12. Enzyme active sites are shown
as black large spots, (a) is male and (b), female. Acid phosphatase-positive cells are very prominent near the
blind duct (D) of the female genital tubercle.

Volume changes in the genital tubercle during de¬
velopment

We tried to determine whether the female geni¬
tal tubercle only arrests to grow or it shrinks along
with the development. The genital tubercle during
its early stages was too small and too ambiguously
demarcated from the proctodeal wall when seen
from outside for us to excise it and measure its wet
weight easily. Alternatively, we measured the
volume of the genital tubercle by the calculation

Fig. 6. Volume changes in the genital tubercle. The
volume of the genital tubercle was measured by the
summation of the volume of serial sections. The
volume of each genital tubercle is represented with a
closed circle (female), a closed square (male) or an
open circle (sex not determined). The sexual
dimorphism begins from about day 10 and becomes
marked after day 12. After day 12, the volume of
female genital tubercle decreases steadily.

based on serial sections as described in Materials
and Methods. The volume of the male genital
tubercle increased slowly from day 10 to day 16,
then fast until day 20. Whereas the volume of
female one increased more slowly than in the male
from day 10 until day 12, then decreased con¬
tinuously until day 20 (Fig. 6).

incubation time (day)

Fig. 7. The mitotic activity in the genital tubercle dur¬
ing development. Genital tubercles of colcemid-
treated embryos were serial-sectioned frontally.
The percentage of metaphase cells in the mesen¬
chymal cells was scored at three regions; 1/3, 2/3,
and 3/3 of the length of the genital tubercle from the
apex. At each region, over 1300 nuclei were
counted. The mitotic activity in the genital tubercle
was defined as the average of the percentages of
metaphase cells at three regions. Mean + SD of
three to five specimens. After day 12, the mitotic
activity in the female genital tubercle was strikingly
low.

Development of Duck Genital Tubercle

831

Mitotic activity in the genital tubercle

The mitotic activity (defined in the legend of
Fig. 7) in the genital tubercle of each sex was
compared after day 8. Since cell death as the
histological characteristics for the degeneration of
the female genital tubercle was more prominent in
the mesenchyme than in the epidermis, we
counted the percentage of metaphase-arrested
cells in the mesenchyme of the genital tubercle of
the embryo treated with colcemid in ovo for 4 hr.
Early genital tubercle had relatively high mitotic
activity. After day 12, the mitotic activity was
above 2.3% in the male, whereas less than 0.8% in
the female.

DISCUSSION

There are two phases in the development of the
genital tubercle: the sexually unimorphic phase
(day 7-10), and the sexually dimorphic phase
(after day 10). In the sexually unimorphic phase,
the genital tubercle is made up simply with the
epidermis, the mesenchyme, and the urethral plate
(or the pocket-like structure from the urodeum).
This organization is strikingly similar to that of
embryonic human penis during its early phase [15,
16]. In the sexually dimorphic phase, the male
genital tubercle grows spirally and forms a pair of
fibrolymphatic bodies, an elastic ligament, a blind
duct and other tissues. The fast increase of the
volume after day 16 is assumed to be due in part to
the formation of cavernous tissues. In the female
embryo, the genital tubercle arrests growth, mor¬
phogenesis, and differentiation, followed by the
total shrinkage. We assumed that this shrinkage is
primarily accounted in terms of the cell kinetics
within the genital tubercle. After day 12 in the
female genital tubercular mesenchyme, less than
0.8% of the cells proliferated within 4 hr, whereas
the proportion of dying cells was about 3.5%.
Though we have no data on how long do dying
cells remain visible by the light microscopy, if we
assume the period to be 7 hr according to the
majority of cases of chromatopycnotic cells [14],
about 7% of the total cells would be lost in a day.
This can explain semiquantitatively the slow
shrinkage of the female genital tubercle. As an

alternative possibility, it is still not excluded that
the cells in the proximal part of the female genital
tubercle migrate out of the genital tubercle.

We suggested that the dying cells in the female
genital tubercle have acid phosphatase activity.
The parallelism between the cell death and the
elevation of acid phosphatase activity has also
been shown in the case of the limb bud of wingless
chick embryo [17], in the interdigital cells of the rat
foot [18], and in the Mullerian ducts of the male
chick embryo [19]. It is possible that the cell in the
female genital tubercle suicides by the activation of
acid hydrolases of its own.

Cell death can be classified into two groups;
necrosis and apoptosis [20]. It is argued that
necrosis usually affects tracts of contiguous cells,
and is associated with a gross departure from
physiological conditions, whereas apoptosis char¬
acteristically affects scattered individual cells and
is usually seen in embryonic life or steady-state
kinetics of tissues. Not using transmission electron
microscopy, we could not characterize the cell
death in the genital tubercle by their criteria.
However, nuclei condensed rather than swelled,
and these cells were sparsely dispersed among
viable cells, showing that the cell death in the
genital tubercle resembles to apoptosis.

Wolff and Wolffs experiments indicated that
gonads are not necessary for the development of
the male-type genital tubercle [9, 10], and that the
factor(s) to induce the regression of the genital
tubercle is mimicked by estrogens [8]. Toward
determining whether the cell death and the arrest
of mitosis are evoked directly by estrogens, we are
investigating the effect of estrogen and anti¬
estrogen on the histology of the genital tubercle,
and the ontogeny of sex-steroid receptors.

REFERENCES

1 Gerhardt, U. (1933) Kloake und Begattungsorgane.
In “Handbuch der vergleichenden Anatomie der
Wirbeltiere”. Ed. by L. Bolk, E. Goppert, E. Kal-
lius and W. Lubosch, Vol. 6. pp. 267-350.

2 Liebe, W. (1914) Das mannliche Begattungsorgan
der Hausente. Jena. Z. Naturwiss., 51: 627-696.

3 Guzsal, E. (1974) Erection apparatus of the copula-
tory organ of ganders and drakes. Acta Vet. Acad.
Sci. Hung., 24: 361-373.

832

H. UCHIYAMA AND T. MlZUNO

4 Komarek, V. V. (1969) Die mannliche Kloake un-
serer Entenvogel. Anat. Anz., 124: 434-442.

5 Komarek, V. V. and Marvan, F. (1969) Beitrag zur
mikroskopischen Anatomie des Kopulationsorganes
der Entenvogel. Anat. Anz., 124: 467-476.

6 King, A. S. (1979) Systema Urogenitale. In “Nomi¬
na Anatomica Avium”. Ed. by J. Baumel, A. King,
A. Lucas, J. Breazile and H. Evans, Academic
Press.

7 Pomayer, C. (1902) III. Die Vogel. In “Morpholo-
gische Studien fiber Kloake und Phallus der
Amnioten”. Ed by A. Fleischman, Morphol. Jahrb.,
30: 614-652.

8 Wolff, Em. (1950) La differenciation sexuelle nor-
male et le conditionnement hormonal des caracteres
sexuels somatiques precoces, tubercule genital et
syrinx chez l’embryon de Canard. Bull. Biol. France
Belgique, 84: 121-193.

9 Wolff, Et. and Wolff, Em. (1951) The effects of
castration on bird embryos. J. Exp. Zool., 116: 59-
98.

10 Wolff, Et. and Wolff, Em. (1951) Sur la differencia¬
tion in vitro du tubercule genital de l’embryon de
Canard. C. R. Soc. Biol., 146: 492-493.

11 Kaltofen, R. S. (1971) Embryonic Development in
the Eggs of the Pekin Duck. Center for Agricultural
Publishing and Documentation, Wageningen.

12 Romanoff, A. (1960) The Avian Embryo. Macmil¬

lan, New York.

13 Lojda, Z., Gossrau, R. and Schiebler, T. H. (1979)
Enzyme Histochemistry: A Laboratory Manual.
Springer Verlag, Berlin.

14 Glucksmann, A. (1951) Cell deaths in normal verte¬
brate ontogeny. Biol. Rev., 26: 59-86.

15 Glenister, T. W. (1954) The origin and fate of the
urethral plate in man. J. Anat., 88: 413-425.

16 Glenister, T. W. (1958) A correlation of the normal
and abnormal development of the penile urethra
and of the infraumbilical abdominal wall. Br. J.
Urol., 30: 117-126.

17 Hinchliffe, J. R. and Ede, D. A. (1973) Cell death
and the development of limb form and skeletal
pattern in normal and wingless (ws) chick embryos.
J. Embryol. Exp. Morphol., 30: 753-772.

18 Ballard, K. J. and Holt, S. J. (1968) Cytological and
cytochemical studies on cell death and digestion in
the foetal rat foot. The role of macrophages and
hydrolytic enzymes. J. Cell Sci., 3: 245-262,

19 Scheib-Pfleger, D. and Wattiaux, R. (1962) Etude
des hydrolases acides des canaux de Muller de
l’embryon de Poulet. I. Activites totales et solubles
des canaux d’embryons de 8 a 10 jours d’incubation.
Dev. Biol., 5: 205-217.

20 Wyllie, A. H., Kerr, J. F. R. and Currie, A. R.
(1980) Cell death: The significance of apoptosis. Int.
Rev. Cytol., 68: 251-306.

ZOOLOGICAL SCIENCE 5: 833-845 (1988)

© 1988 Zoological Society of Japai

Atrial Natriuretic Peptide (ANP)-immunoreactivity
and Ultrastructures of Cardiocytes in Fish

Tohru Hirohama, Haruko Uemura, Sumio Nakamura
and Tomoji Aoto* 1

Biological Laboratory, Kanagawa Dental College, 82 Inaoka-cho,
Yokosuka-shi 238, Japan

ABSTRACT — The hearts of three freshwater teleosts ( Oryzias latipes, Cyprinus carpio and Tribolodon
hakonensis ), three marine teleosts ( Chrysiptera cyanea, Sebastes inermis and Hexagrammos otakii) and
one elasmobranch ( Triakis scyllia) were investigated immunohistochemically and ultrastructurally. The
immunohistochemical method using an antiserum against a-human atrial natriuretic peptide (hANP)
indicated stronger response in the atrium than ventricle of Cyprinus, Tribolodon and Triakis. No
hANP-immunoreactive material in the cardiocytes of Oryzias and of any of the marine teleosts in this
study could be detected.

Ultrastructurally, the atrial and ventricular cardiocytes of all the seven species of fish studied were
found to have electron-dense secretory granules varying widely in diameter. Generally, granule size in
the fish was much less than that so far reported in mammalian species. However, the ventricular
cardiocytes had far less granules than the atrial cardiocytes. The negative immunohistochemical
response of cardiocytes in Oryzias and marine teleosts may be not possibly due to the antibody used in
this study but rather to the very small amount of atrial natriuretic peptide (ANP)-like substance
remaining in the cardiocytes, since the number of electron-dense granules was very few in these species.
These findings indicate that the ANP-like substance is apparently produced by both atrial and
ventricular cardiocytes in fish.

INTRODUCTION

The discovery of electron-dense granules in a-
trial cardiocytes is attributed to Kisch [1], the first
to notice in the guinea pig “many very small
osmiophilic bodies of apparently spherical shape”
and about 200 nm in diameter between myofibrils.
Since then, a group of peptides that influence
blood pressure, renal function and salt balance has
been isolated from several mammalian cardiac
atria and designated as the atrial natriuretic pep¬
tide (ANP) (for review, see [2]). Subsequent
investigations have indicated specific granules in
atrial cardiocytes to be sites for the storage of ANP
[3,4]. “ANP-granules” have also been found even
in the heart of non-mammalian vertebrates [5-8].
However, studies in this regard on fish are quite

Accepted December 25, 1987
Received October 13, 1987

1 To whom request of reprints should be addressed.

limited at present. In the present research a study
was conducted on the immunohistochemistry and
ultrastructure of cardiocytes in various teleosts and
one species of elasmobranchs.

MATERIALS AND METHODS

Three species of freshwater teleosts, Japanese
medaka Oryzias latipes (2. 3-2. 8 cm in total
length), common carp Cyprinus carpio (18-20 cm)
and river-dwelling Japanese dace Tribolodon
hakonensis (15-18 cm), three marine teleosts,
cobalt damselfish Chrysiptera cyanea (3. 2-4. 5 cm),
rockfish Sebastes inermis (15-20 cm) and greenling
Hexagrammos otakii (26-38 cm), and one elas¬
mobranch, banded dogfish Triakis scyllia (65-95
cm), of both sexes were used. These fishes were
obtained from commercial sources except the
rockfish and banded dogfish, which were caught in
the sea near Yokosuka and Misaki, respectively, in
Kanagawa prefecture. Also, for comparison.

834

T. Hirohama, H. Uemura et al.

mouse ( Mus musculus) atrial cardiocytes were
studied ultrastructurally.

An anesthetic was used only in dogfish which
were immersed in 0.02% MS222 (Sandoz) solution
prior to dissection. For the immunohistochemical
study, the heart was dissected out from decere¬
brated fish, fixed in Bouin’s solution for 24 hr at
4°C and 6 jun i thick paraffin sections were made.
Deparaffinized and hydrated sections were incu¬
bated in a solution containing 10% H2O2 in metha¬
nol at 4°C for 15 min and then treated according to
the peroxidase-antiperoxidase (PAP) method [9]
using an antiserum (Peptide Institute, Inc., Osaka)
against a-human ANP (hANP) diluted 1: 660 in 0.1
M phosphate-buffered saline adjusted to pH 7.3.
The specificity of the immunohistochemical stain¬
ing was confirmed by replacing the antiserum with
that preabsorbed by excess hANP (10 ^g/ml of
hANP antiserum diluted 1:660) at 4°C for 48 hr.
This procedure resulted in the complete loss of
immunoreactivity in cardiocytes in all cases. For
the electron microscopical study, the tissues dis¬
sected out from the animals were immediately
fixed in 2% glutaraldehyde adjusted to pH 7.4 with
cacodylate buffer for 2 hr at 4°C, washed in the
same buffer for 24 hr and then fixed in 1% osmium
tetroxide for 1 hr. The tissues were embedded in
Spurr resin [10] and their sections examined with a
JEM 100-B electron microscope following double
staining with uranyl acetate and lead citrate. A-
trial cardiocyte ultrastructures were observed and
only granules with a clear core surrounded by a
distinct limiting membrane in 15-30 cells in each
species had their diameter measured. In order to
know whether these are more than one type car¬
diocytes containing granules of particular size, ten
cardiocytes of a Hexagrammos atrium, containing
relatively large numbers of granules, were chosen
and in each of these ten cardiocytes the granule
diameter was measured. The homogeneity of
these cardiocytes with respect to distribution of
granule size was analyzed by the Kruskal-Wallis
method. Ventricular cardiocytes were examined
for the presence of secretory granules in all species
studied.

RESULTS

In the mouse, atrial cardiocytes contained
numerous electron-dense granules of 120-430 nm
(the median, 290 nm) in diameter, near the nu¬
cleus (Fig. 1). Well-developed myofibrillar bun¬
dles, Golgi complex and numerous mitochondria
were also present (Fig. 2). But in the atrial car¬
diocytes of the teleosts, myofibrillar bundles, elec¬
tron-dense granules and Golgi complex were poor¬
ly developed in most cases. The ventricular car¬
diocytes had far fewer granules than the atrial
cardiocytes in the teleosts. The cardiocyte charac¬
teristics of each species are described below.

In Oryzias, no ANP immunoreactive material
(ir-ANP) was detected in either atrial or ventricu¬
lar cardiocytes. However, only a few electron-
dense granules were noted in some atrial car¬
diocytes. The granules, 50-290 nm in diameter
(the median, 160 nm) (Fig. 1), were scattered
throughout the cytoplasm, without being concen¬
trated at any particular area in the cell (Fig. 3). A
very few ventricular cardiocytes were found to
each contain a small number of electron-dense
granules.

In the Cyprinus atrium, nearly all cardiocytes
retained a moderate amount of ir-ANP about the
nucleus (Fig. 4a), but did not in the ventricle
except for a few cells in the inner layer of the
ventricular wall and the proximal part of the
intraventricular, trabeculae carneae-iormmg septa
which showed weak ANP immunoreactivity (Fig.
4b). Electron microscopy showed the cardiocytes
to contain a moderate number of granules situated
near the nucleus and along the cell periphery (Fig.
4c). Diameter ranged from 80-340 nm (the me¬
dian, 180 nm) (Fig. 1). In a few ventricular car¬
diocytes, a very small number of granules was
situated about the nucleus.

In the Tribolodon atrium, ir-ANP was rather
widely distributed in most cardiocytes (Fig. 5a). In
the ventricle, most cardiocytes in the inner layer of
the ventricular wall and intraventricular septa re¬
tained moderate or scarce ir-ANP around the
nucleus (Fig. 5b). In other areas, only a few
cardiocytes were noted to contain ir-ANP and only
in a small amount. Ultrastructurally, the atrial
cardiocytes of this species contained thick myofi-

ANP-like Substance in Fish

835

DIAMETER ( "«" )

Fig. 1. Size distribution of round electron-dense secretory granules in atrial cardiocytes. Median values are
indicated by broken lines, n, number of fish used.

brillar bundles (about 1 pm) and a relatively large
number of electron-dense granules situated more
or less about the nucleus (Fig. 6a). In some cases,
however, the granules were dispersed rather wide¬
ly. The granules ranged from 50-270 nm (the
median, 140 nm) in diameter and appeared to have
three peaks of occurrence (Fig. 1). Compared to

other freshwater teleosts, Tribolodon possessed a
relatively large number of ventricular cardiocytes
containing granules (Fig. 6b).

In Chrysiptera, neither the atrium nor ventricle
contained ir-ANP. Atrial cardiocytes contained
rather thick myofibrillar bundles (1.0-1. 2 /an),
and poorly developed Golgi complex. Not many

836

T. Hirohama, H. Uemura et al.

ANP-like Substance in Fish

837

t

w

Fig. 4. Cardiocytes of Cyprinus. (a) ANP-immunoreactive material is accumulated around the nucleus of the atrial
cardiocyte. X640. (b) Small amount of similar material (arrows) is seen in ventricular cardiocytes. X640.

w, ventricular wall; s, proximal part of the intraventricular septa, (c) An electron micrograph of atrial
cardiocytes, showing a moderate number of granules (Sg). Other abbreviations are the same as those in Fig. 2.

Fig. 2. Longitudinal section of mouse atrial cardiocytes, showing abundant electron-dense secretory granules (Sg) of
wide range of size. G, Golgi complex; Mi, mitochondria; My, myofibril; N, nucleus.

Fig. 3. An Oryzias atrial cardiocyte. Only a few secretory granules (Sg) are seen. Other abbreviations are the same
as those in Fig. 2.

838

T. Hirohama, H. Uemura et al.

v i */

Fig. 5. Cardiocytes of Tribolodon. (a) Abundant ANP-immunoreactive material is rather
widely distributed in the atrial cardiocytes. X640. (b) Ventricular cardiocytes in the
inner layer of the ventricular wall (arrow heads) and intraventricular septa (arrows)
retaining ANP-immunoreactive material. X640.

granules were present in the cell, and most were
situated along the periphery, only a few being
visible about the nucleus (Fig. 7). Diameter was
from 60-260 nm (the median, 130 nm) (Fig. 1).
Only a very few electron-dense granules were
present in ventricular cardiocytes.

In Sebastes, ir-ANP was not detected in either
the atrium or ventricle. The atrial cardiocytes
were relatively wide and their myofibrillar bun¬
dles, rather thin (0.6-0. 8 fjm). Granules, not
abundant, were distributed along the periphery of

the cells (Fig. 8) and ranged from 60-280 nm in
diameter (the median, 160 nm) (Fig. 1). In a few
ventricular cardiocytes, a few granules were dis¬
persed throughout the cells.

No ANP-immunoreactivity could be detected in
Hexagrammos heart tissue. Ultrastructurally,
however, a few granules, 80-250 nm (the median,
150 nm) in diameter (Fig. 1), were dispersed along
the periphery of the atrial cardiocytes. The cells
were filled with many myofibrillar bundles and
numerous, more or less congregated mitochon-

Fig. 6. A part of a Tribolodon atrial cardiocyte, showing many secretory granules (Sg) near the nucleus (N). Other
abbreviations are the same as those in Fig. 2.

ANP-like Substance in Fish

839

. ' v • . , ,r:

840

T. Hirohama, H. Uemura et al.

ANP-like Substance in Fish

841

dria. The Golgi complex was not well-developed
(Fig. 9a). A few ventricular cardiocytes contained
a small number of electron-dense secretory gran¬
ules (Fig. 9b).

In ten atrial cardiocytes of Hexagrammos, gran¬
ule diameter’’ was measured and the results are
listed in Table 1. The Kruskal-Wallis test showed
granule size distribution homogeneity in a group of
cardiocytes A, E, F, G, H and I, and in another
group, C, D, and J. But if combined, these nine
cardiocytes cease to show such homogeneity (P<
0.01). Thus, the two groups were tentatively
designated as types I and II, respectively. In the
cardiocyte B granule size distribution differed so
much from that of the others that, on combining it
with type I, the homogeneity ceased to exist (P<
0.02). Consequently, this cell was classified as type
III, thus making a total of three types of car¬
diocytes discernible on the basis of diameter dis¬
tribution. The median diameters of the granules in
these types were 150, 135 and 200 nm, respec¬
tively.

In the Triakis atrium, most cardiocytes in the
outer layer of the atrial wall showed weak ANP-
immunoreactivity and immunoreactive products
were dispersed throughout the entire cell (Fig.
10a). In other areas, only a few cardiocytes
retained a small amount of ir-ANP near the nu¬
cleus. In the ventricle, some cardiocytes showed
weak immunoreactivity near the nucleus (Fig.
10b). These cells were situated primarily in the
inner layer of the ventricular wall and proximal
part of the septa. Electron microscopy indicated
myofibrillar bundles in atrial cardiocytes to be
thick (1.5 pm) and coarse. A few secretory gran¬
ules, considerably larger in size than those in
teleosts studied, were dispersed throughout the
cell. Both electron-dense granules, 110-400 nm
(the median, 230 nm) in diameter (Fig. 1), and
small electron-lucent vesicles, about 90 nm in di¬
ameter, could be observed about the nucleus.
Some vesicles, about 126 nm in diameter, had a
fuzzy coating. The Golgi complex was only poorly

developed (Fig. 10c). Very few ventricular car¬
diocytes possessed granules whose number was
quite small.

DISCUSSION

Electron microscopy demonstrated secretory
granules, which are probably carriers of ANP, in
the cardiocytes of both atrium and ventricle in
fishes. However, in the mouse (Fig. 2) and in
other mammals [11], granules are found mainly in
the atrium. It would be interesting to explore the
phylogeny of the distribution of secretory granules
in the heart tissue. Cardiocytes of fish generally
have Golgi complex developed to a lesser extent
than in mammals. The electron-dense granules are
also smaller (Fig. 1) and fewer in number in fish
than in mouse. It thus appears that secretory
activity in fish cardiocytes is not very high.
Although all the species examined contained elec¬
tron-dense granules in the cardiocytes of both
atrium and ventricle, immunohistochemically both
the atrial and ventricular cardiocytes of only Cypri-
nus and Tribolodon were reactive to hANP anti¬
serum and both cardiocytes in the remaining four
species were non-reactive. The negative im¬
munoreactivity in these four species may be due to
insufficient ANP in the cell. In fact, cardiocytes in
these species were found ultrastructurally to con¬
tain few secretory granules. However, meager
ANP might be detected even in these species,
provided antisera against their intrinsic ANP, in¬
stead of hANP antiserum, were used. Anyway, it
is quite likely that both atrial and ventricular
cardiocytes in fish possess peptides with some
amino acid sequence in common with the hANP.
A weak ANP-immunoreaction in ventricles has
also been observed in three freshwater teleost
species [12].

Among the teleosts studied, Tribolodon had the
highest number of granules and Golgi complex,
and in their cardiocytes, ANP-immunoreactivity
was most evident, thus indicating the possibility of

Fig. 7. Chrysiptera atrial cardiocytes. A few secretory granules (Sg) are seen. Other abbreviations are the same as
those in Fig. 2.

Fig. 8. Sebastes atrial cardiocytes, containing many well-developed mitochondria (Mi) and a few secretory granules
(Sg). Other abbreviations are the same as those in Fig. 2.

842

T. Hirohama, H. Uemura et al.

ANP-like Substance in Fish

843

Table 1. The size distribution of round electron-dense granules in ten cardiocytes (A-J) in a
Hexagrammos atrium

Granule

diameter

(nm)

Number of granules

A

B

C

D

E

F

G

H

I

J

80

1

1

90

100

2

2

1

1

1

110

1

1

1

1

120

3

1

5

1

3

1

6

130

1

3

2

2

3

3

2

140

1

1

2

2

2

1

1

5

150

5

2

4

4

1

2

3

5

4

3

160

1

2

1

2

2

2

1

1

170

2

4

3

3

1

1

180

1

1

3

1

4

2

1

190

2

2

200

3

1

210

4

3

1

1

220

1

1

2

1

1

230

1

1

1

1

1

240

250

1

1

1

1

Total

number

18

19

9

18

17

17

20

16

13

21

their strong secretory activity. Tribolodon h. is a
euryhaline fish and therefore, its high adaptability
to a wide range of salinity may be closely related
with the active secretion of ANP-like material of
this fish.

The cardiocytes of marine species and Oryzias
contained only a few granules. This feature
perhaps can be explained by either less active
synthesis or rapid release following synthesis in
these species.

In the spiny dogfish, Squalus acanthias, heart
extracts have been shown to contain an ANP-like
substance capable of in vitro stimulation of rectal
gland chloride secretion [5]. In Triakis, a similar
substance from cardiocytes is assumed to play
some role in osmoregulation. The presence of only

a small number of granules and weak ANP-
immunoreactivity in this elasmobranch may thus
be deciphered as the reflection of the active release
rather than the poor synthesis of this substance.

In nearly all species of fish examined, granules
ranging widely in size could be discerned in the
atrial cardiocytes, showing seemingly three or
more peaks of distribution (Fig. 1). Whether the
difference in size of the granules represents dif¬
ferent steps of ANP maturation or whether the
cardiocytes secrete, in addition to ANP, some
substance(s) other than ANP is not known. In this
respect, at least in Hexagrammos three types of
atrial cardiocytes could be distinguished on the
basis of granule diameter distribution (Table 1).
The biological significance of occurrence of these

Fig. 9. Hexagrammos cardiocytes. (a) Atrial cardiocytes, containing numerous mitochondria (Mi) and myofibrillar
bundles (My). A few secretory granules (Sg) are dispersed along the periphery of the cells. Also seen on the
periphery are many electron-lucent vesicles (arrows), some of which show reverse-pinocytotic feature, (b)
Ventricular cardiocytes. Very few secretory granules (Sg) are seen near the nucleus (N). Other abbreviations are
the same as those in Fig. 2.

844

T. Hirohama, H. Uemura et al.

Fig. 10. Triakis cardiocytes. (a) ANP-immunoreactive cardiocytes (arrows) in the outer layer of the atrial wall.
ANP-immunoreactive products are dispersed throughout the entire cell. X640. (b) A weak ANP-

immunoreactivity is mostly localized in the juxtanuclear area (arrows). X640. (c) An atrial cardiocyte, having
several secretory granules (Sg) and electron-lucent vesicles (arrows). Other abbreviations are the same as those
in Fig. 2.

three types of cardiocytes is a problem for future
research.

ACKNOWLEDGMENTS

The authors are grateful to Dr. Hideshi Kobayashi for

his keen interest in the present research. Thanks are due
to Dr. S. Yoshitake for her competent assistance. We
are also thankful to the staffs of the Misaki Marine
Biological Station, University of Tokyo for collecting the
banded dogfish and permitting the use of facilities. This
study was supported in part by a Grant-in-Aid for
Scientific Research from the Japan Ministry of Educa-

ANP-like Substance in Fish

845

tion, Science and Culture.

REFERENCES

1 Kisch, B. (1956) Electron microscopy of the atrium
of the heart. I. Guinea pig. Exp. Med. Surg., 14:
99-112.

2 Atlas, S. A. (1986) Atrial natriuretic factor: A new
hormone of cardiac origin. Recent Prog. Horm.
Res.. 42: 207-249.

3 de Bold, A. J. (1982) Tissue fractionation studies on
the relationship between an atrial natriuretic factor
and specific atrial granules. Can. J. Physiol. Pharma¬
col., 60: 324-330.

4 Cantin, M., Gutkowska, J., Thibault, G., Milne, R.
W., Ledoux, S., MinLi,S., Chapeau, C., Garcia,
R., Hamet, P. and Genest, J. (1984) Immunocy-
tochemical localization of atrial natriuretic factor in
the heart and salivary glands. Histochemistry, 80:
113-127.

5 Solomon, R., Taylor, M., Dorsey, D., Silva, P. and
Epstein, F. H. (1985) Atriopeptin stimulation of
rectal gland function in Squalus acanthias. Am. J.
Physiol., 249: R348-R354.

6 Yoshitake, K. (1985) Comparative anatomic and
electron microscopic study on specific granules of
heart muscle cells. Okayama Igakkai Zasshi, 97:

713-748.

7 Netchitailo, P., Feuilloley, M., Pelletier, G., Can-
tin, M., de Lean, A., Leboulenger, F. and Vaudry,
H. (1986) Localization and characterization of atrial
natriuretic factor (ANF)-like peptide in the frog
atrium. Peptides, 7: 573-579.

8 Reinecke,M., Betzler, D. and Forssmann, W. G.
(1987) Immunocytochemistry of cardiac
polypeptide hormones (Cardiodilatin/atrial natriure¬
tic polypeptide) in brain and hearts of Myxine
glutinosa (Cyclostomata). Histochemistry, 86: 233-
239.

9 Sternberger, L. A., Hardy, P. H. Jr., Cuculis, J. J.
and Meyer, H. G. (1970) The unlabeled antibody
enzyme method of immunohistochemistry. J. His-
tochem. Cytochem., 18: 315-333.

10 Spurr, A. R. (1969) A low- viscosity epoxy resin
embedding medium for electron microscopy. J.
Ultrastruct. Res., 26: 31-43.

11 Tomisawa, M. (1969) Atrial specific granules in
various mammals. Arch. Histol. Japon., 30: 449-
465.

12 Chapeau, C., Gutkowska, J., Schiller, P. W.,
Milne, R. W., Thibault, G., Garcia, R., Genest, J.
and Cantin, M. (1985) Localization of immunoreac-
tive synthetic atrial natriuretic factor (ANF) in the
heart of various animal species. J. Histochem.
Cytochem., 33: 541-550.

ZOOLOGICAL SCIENCE 5: 847-853 (1988)

© 1988 Zoological Society of Japai

Absence of Blood Vessels in the Brain of Six Species
of Primitive Salamanders

Kazuhiko Tsuneki* 1 and Masami Ouji
Department of Biology, Shimane University, Matsue, Shimane 690, Japan

ABSTRACT — The brains of six species of salamanders belonging to the Hynobiidae ( Onychodactylus
japonicus, Pachypalaminus boulengeri, Hynobius retardatus, H. nigrescens, H. okiensis, and H.
stejnegeri ) were studied histologically. These salamanders are sluggish animals living under relatively
cool habitats. In all species, the brain parenchyma was not vascularized, except for the olfactory bulb
and neural lobe. Their brain parenchymas are relatively thin and neuronal perikarya are distributed
mainly in the periventricular region facing cerebrospinal fluid which is nourished by blood vessels of the
large choroid plexus.

INTRODUCTION

Recently, we reported that the bulk of the brain
parenchyma of three species of salamanders be¬
longing to the genus Hynobius (H. nebulosus, H.
naevius, and H. kimurai) lacks blood vessels [1].
This was the first report that showed the presence
of vertebrates lacking parenchymal vascularization
in the brain. However, the hynobiid salamanders
that we have studied were all collected from a
restricted part of Japan, the Shimane Prefecture.
It was unknown whether this peculiar nature in the
brain also occurs in the other species of the genus
Hynobius and the species belonging to the Hyno¬
biidae but not to Hynobius. In this paper, we
report whether the species of the Hynobiidae such
as Onychodactylus japonicus, Pachypalaminus
boulengeri, Hynobius retardatus (the northernmost
Hynobius in Japan), and Hynobius stejnegeri (the
southernmost Hynobius in Japan) also lack paren¬
chymal vascularization in the bulk of the brain.

MATERIALS AND METHODS

The following six species belonging to the Hyno¬
biidae were studied: Onychodactylus japonicus
(8.9 to 15.7 cm in total length = TL), Pachypalami-

Accepted November 13, 1987
Received September 5, 1987

1 Present address: College of General Education, O-
saka University, Toyonaka, Osaka 560, Japan.

nus boulengeri (7.3 to 15.7 cm in TL), Hynobius
retardatus (8.5 to 9.2 cm in TL), Hynobius nigres¬
cens (13.8 cm in TL), Hynobius okiensis (14.9 cm
in TL), and Hynobius stejnegeri (13.5 cm in TL).
After decapitation, the brains were fixed in
Bouin’s solution, were embedded in paraffin, and
were serially sectioned transversely at 7 pm. The
sections were stained with Masson-Goldner’s
method with or without prestaining with paral¬
dehyde fuchsin. For comparison, the brains of the
California newt, Taricha torosa (14.0 cm in TL),
belonging to the Salamandridae, and the spring
salamander, Gyrinophilus porphyriticus (15.5 cm
in TL), belonging to the Plethodontidae, were
processed for light microscope observation as the
brains of the hynobiid salamanders. When
needed, we also studied histological preparations
of the brains of various vertebrates of our own
collection.

In Onychodactylus japonicus and Pachypalami¬
nus boulengeri, the vascular system was perfused
with India ink. Thus, under anesthesia with MS
222, the heart was exposed and was perfused at
first with a 0.6% NaCl solution and then with India
ink. The perfused brain with surrounding tissues
was fixed in Bouin’s solution, embedded in par¬
affin, serially sectioned transversely at 10 pm, and
observed without staining.

In Onychodactylus japonicus, the vascular sys¬
tem also was demonstrated with a corrosion cast
method. Under anesthesia, a 0.6% NaCl solution

848

K. Tsuneki and M. Ouji

1 * j cam

Avascular Brain in Onychodactylus

849

and methacrylate resin (Mercox (Dainippon Ink
Chemical) diluted with methyl methacrylate) were
successively perfused intracardially. After
polymerization of the resin, soft tissues were dis¬
solved with 20% NaOH. The vascular cast was
observed under a binocular microscope.

RESULTS

Light microscopical histology

Most cells of urodele amphibians, including
erythrocytes and endothelial cells, are very large.
Blood vessels can be easily identified in light
microscopical preparations.

In all studied species belonging to the Hyno-

Fig. 5. The left olfactory bulb of Hynobius nigrescens. Blood vessels are not restricted in the glomerular layer
(double arrows), but also seen in the plexiform layer (arrows), lv, lateral ventricle.

Fig. 6. The right olfactory bulb of Onychodactylus japonicus. Blood vessels are restricted within the glomerular
layer (arrows). The slit on the left represents the rostral tip of the lateral ventricle.

Both figures represent transverse sections magnified to Xl05.

Fig. 1. Left cerebral hemisphere of Hynobius retardatus. The telencephalic choroid plexus (cp) is seen in the lateral
ventricle, dp, dorsal pallium; mp, medial pallium (primordium hippocampi).

Fig. 2. Preoptic area of Hynobius stejnegeri. At this level, the third ventricle appears to be goblet-shaped. The
choroid plexus is seen in the top of the goblet. The bottom of the goblet represents the preoptic recess.

Fig. 3. Diencephalic-mesencephalic region of Pachypalaminus boulengeri. The diencephalic choroid plexus is seen
in the wide ventricle. Arrow heads indicate the paraventricular organ, p, pineal; sco, subcommissural organ.
Fig. 4. The hypothalamus of Onychodactylus japonicus. Sporadic cells in the white matter are glia cells. Arrow
heads indicate the paraventricular organ.

All figures represent transverse sections magnified to X105.

850

K. Tsuneki and M. Ouji

biidae, the brain parenchyma is not vascularized
(Figs. 1-4) except for two small areas; the olfac¬
tory bulb (Figs. 5 and 6) and the neural lobe (pars
nervosa). In Pachypalaminus boulengeri and four
species belonging to Hynobius, a small number of
blood vessels occur not only in the glomerular
layer of the olfactory bulb but also in the plexiform
layer of the olfactory bulb (Fig. 5), as in three
hynobiid species studied in a previous paper [1].
(In the previous paper [1], the moderately vascula¬
rized plexiform layer of the olfactory bulb was
misidentified as the rostral tip of the cerebral
hemisphere.) In Onychodactylus japonicus, the
blood vessels in the olfactory bulb are restricted
within the glomerular layer and never extend to
the plexiform layer (Fig. 6).

The brain parenchyma of Taricha torosa and

Gyrinophilus porphyriticus is vascularized not only
in the olfactory bulb and the neural lobe, but also
in most regions of the brain (Figs. 7 and 8).

In all species studied, blood vessels are distrib¬
uted in the meninx, the paraphysis, and the cho¬
roid plexus. The large telencephalic and
diencephalic choroid plexuses are observed in the
ventricle (Figs. 1-4, 7-10). However, these
epithelial structures invaginate from the brain sur¬
face into the ventricle during the course of de¬
velopment. The blood vessels of the choroid
plexus are originally meningeal blood vessels.
Another epithelial structure, the paraphysis, is an
evaginated circumventricular organ and the blood
vessels of the paraphysis are of course meningeal
blood vessels.

In all species studied, the brain parenchyma is

Fig. 7. The left cerebral hemisphere of Taricha torosa. Note blood vessels (arrows) in the parenchyma, dp, dorsal
pallium; mp, medial pallium.

Fig. 8. The right cerebral hemisphere of Gyrinophilus porphyriticus. Note blood vessels (arrows) in the
parenchyma, dp, dorsal pallium; mp, medial pallium.

Both figures represent transverse sections magnified to X105.

Avascular Brain in Onychodactylus

851

aggregation of perikarya) and the perimeningeal
white matter (an aggregation of nerve fibers) (Figs.
1-4, 7 and 8). Glia cells also are found sporadical¬
ly in the white matter.

The thickness from the meningeal surface to the
ventricular surface of ependymal cells of the me¬
dial pallium is 350 to 500 pm in the hynobiid
salamanders including Onychodactylus japonicus
and Pachypalaminus boulengeri. In Gyrinophilus
porphyriticus and Taricha torosa, the thickness of
the medial pallium is about 390 and 400 pm,
respectively. All these values were obtained from
the materials embedded in paraffin. Real values in
the living state should be slightly larger than these
values. In any case, the thickness of the brain
parenchyma of hynobiid salamanders is thin,

although not particularly so compared to the non-
hynobiid salamanders of similar size range.

India ink perfusion

Blood vessels filled with India ink are not found
in the brain parenchyma except for the olfactory
bulb and the neural lobe in Onychodactylus japoni¬
cus and Pachypalaminus boulengeri (Figs. 9 and
10).

Blood vessels filled with India ink are seen in the
ventricle of Figures 9 and 10. They represent
blood vessels of the telencephalic choroid plexus.
Dark structures seen in the meninx are in majority
meningeal blood vessels filled with India ink.
Melanophores are dispersed in the meninx, espe¬
cially in the meninx covering the midbrain, but the

Fig. 9. The left cerebral hemisphere of Pachypalaminus boulengeri perfused with India ink. India ink-filled blood
vessels are restricted in the choroid plexus (left) and the meninx (right) which separates the bilateral cerebral
hemispheres.

Fig. 10. The right cerebral hemisphere of Onychodactylus japonicus perfused with India ink. Blood vessels are seen
only in the choroid plexus (center) and the meninx (top and lower right).

Both figures represent transverse sections magnified to X105.

852

K. Tsuneki and M. Ouji

dark structures in the meninx of Figures 9 and 10
are roundish and it is unlikely that they represent
melanophores.

Corrosion cast

Because of the shallow focus depth of a binocu¬
lar microscope, we could not produce photographs
of good quality. Even under a binocular micro¬
scope, however, we could observe sufficient details
of the vascular arrangement.

Blood vessels in the meninx appear to be a
net-like basket. In the telencephalic region, the
basket contains the core in the center, which is the
cast of entangled blood vessels of the telencephalic
choroid plexus. There are no apparent twigs that
leave the superficial basket net or the central core.
A net in the region corresponding to the glomeru¬
lar layer of the olfactory bulb is somewhat ruffled.

DISCUSSION

In addition to three species studied in a previous
paper [1], this study has revealed that Onychodac-
tylus japonicus, Pachypalaminus boulengeri, and
four species belonging to Hynobius also lack
parenchymal vascularization in the bulk of the
brain. These species were brought from various
regions of Japan, from Hokkaido to Kyushu, and
include both torrent-breeders and stagnant pond-
breeders. Therefore, it is highly probable that all
Japanese salamanders belonging to the Hyno-
biidae lack blood vessels in the bulk of the brain
parenchyma, although several species ( Salaman -
drella keyserlingi and six species belonging to
Hynobius) are not yet studied. Unfortunately,
there appears to be no report on the brain struc¬
ture of hynobiid salamanders distributed in the
East Asia outside Japan.

The reason why the olfactory bulb is vascula¬
rized even in the hynobiid salamanders is un¬
known. In the couse of study of the brain develop¬
ment in the giant salamander, Andrias japonicus,
we noticed that the olfactory bulb is vascularized
first during development [2]. If the development
would be stopped at this stage, the adult condition
of the hynobiid brain might be obtained. Howev¬
er, this developmental consideration does not ex¬
plain the occurrence of the highly vascularized

brain in neotaneous salamanders such as Siren
intermedia [1].

Exceptionally in the Hynobiidae, Onychodacty-
lus japonicus is a lungless salamander [3], In this
species, gas exchange mechanisms between the
external and internal environments and between
blood vessels and brain parenchyma are thus most
peculiar. This salamander lives in the deep forest
and breeds in small torrents hardly accessible by
men [4]. Under a condition of relatively high
oxygen content in the relatively cool habitat, this
salamander might not necessitate the lung and the
vascularized brain parenchyma. Onychodactylus
japonicus is also unusual in totally lacking blood
vessels even in the plexiform layer of the olfactory
bulb. Thus, Onychodactylus japonicus possesses
the least vascularized brain among the salamander
species so far studied.

The main lungless salamander family is the
Plethodontidae. Gyrinophilus porphyriticus is a
relatively large species and possesses a vascula¬
rized brain parenchyma as Taricha torosa belong¬
ing to the Salamandridae. It may be interesting to
examine brain vascularization in very small species
of the Plethodontidae such as Thorius spp.

The distance which materials such as oxygen,
carbon dioxide, nutrients, metabolic wastes, and
hormones must traverse is relatively thin in hyno¬
biid salamanders. This feature may at least partly
compensate the absence of parenchymal vascular¬
ization in the brain of these salamanders.

The preferential distribution of perikarya in the
periventricular region also may be critical. The
materials essential for neuronal activity must dif¬
fuse either from the meningeal blood vessels or
from the choroid plexus blood vessels. The occur¬
rence of the periventricular grey matter suggests
material exchange between the periventricular
grey matter and the choroid plexus blood vessels
via cerebrospinal fluid. Even in non-hynobiid
amphibians, it is known that a substantial portion
of blood entering the brain flows to the choroid
plexus [5], which is the main site of cerebrospinal
fluid production.

The brain parenchyma of hynobiid salamanders
lacks “the first circulation” by blood vessels. The
brain parenchyma in general lacks “the second
circulation” by lymphatic vessels. Therefore, so-

Avascular Brain in Onychodactylus

853

called “the third circulation” by cerebrospinal fluid
[6] may be important for metabolism in the brain
parenchyma of hynobiid salamanders. However,
the preferential occurrence of the periventricular
grey matter is not restricted to the hynobiid sala¬
manders, but is found in most urodeles. After a
survey of the brain sections of various vertebrates
of our collection, it has become apparent that in
many groups of lower vertebrates the perikarya
tend to appear around the ventricle more or less,
but highly preferential aggregation of perikarya in
the periventricular region is found in urodeles and
lungfishes. Therefore, spinal cord-like distribution
of perikarya and nerve fibers in the brain may
characterize these animals, but this feature itself
does not fully explain the occurrence of an avascu¬
lar brain parenchyma and utmost importance of
cerebrospinal fluid in the hynobiid salamanders.
There might be various reasons that caused the
absence of blood vessels in the brain parenchyma
of hynobiid salamanders.

ACKNOWLEDGMENTS

Most salamanders studied in this paper were obtained
by courtesy of the following people: S. Ishii (Tokyo), M.

Kakegawa (Tokyo), K. Kubokawa (Tokyo), S. Kuzumi

(Nagoya), H. Nambu (Toyama), and S. Ogasawara

(Kouchi). This study was partly supported by a Grant-in-

Aid for Scientific Research from the Ministry of Educa¬
tion, Science and Culture of Japan.

REFERENCES

1 Tsuneki, K., Ouji, M., Akiyoshi, H. and Ichihara, K.
(1985) Absence of blood vessels in the brain paren¬
chyma of hynobiid salamanders. Experientia, 41:
1400-1402.

2 Tsuneki, K., Kuwabara, K., Kobara, J. and Ouji, M.
(1983) Development of the adenohypophysis and
circumventricular organs in the giant salamander,
Andrias japonicus. Zool. Mag. (Tokyo), 92: 149-163
(in Japanese with English summary).

3 Kudo, T. and Kitani, N. (1934) Ontogenetical and
phylogenetical considerations on the lung of Onycho¬
dactylus japonicus. Acta Anat. Nippon., 7: 425-433
(in Japanese).

4 Akita, Y. (1982) Notes on the egg-laying site of
Onychodactylus japonicus on Mt. Hodatsu. Jpn. J.
Herpetol., 9: 111-117 (in Japanese with English
summary).

5 Heisey, S. R. (1968) Brain and choroid plexus blood
volumes in vertebrates. Comp. Biochem. Physiol.,
26: 489-498.

6 Milhorat, T. H. (1975) The third circulation revis¬
ited. J. Neurosurg., 42: 628-645.

.

■

-

■ •

'

ZOOLOGICAL SCIENCE 5: 855-861 (1988)

© 1988 Zoological Society of Japan

Factors Regulating Urination Patterns in Male and
Female Mice ( Mus musculus)

Masashi Daumae and Takeji Kimura

Department of Biology, College of Arts and Sciences, University of Tokyo,
Meguro-ku, Tokyo 153, Japan

ABSTRACT — Urine marking pattern of male and female mice was analyzed under several hormonal
treatments. Males, ovariectomized and androgenized females and neonatally androgenized females
deposited larger number of urine spots than normal females. Castration of males decreased the number
of spots in a 1.5-hour test but not in a 16-hour test, suggesting that castration may induce changes in
motivational state involved in the initiation of marking in novel environment; however, it does not
change the mode of urination. It is probable that the behavior pattern of deposition of large number of
urine spots is dependent on neonatal androgen as well as androgen in later life. Analysis of the
distribution of different sizes of urine spots in the chosen size categories revealed the negative
correlation between spot size and length of external genitalia. Neonatally androgenized females which
have shorter genitalia than males deposited spots of larger sizes than males. It is suggested that sexual
dimorphism in urination pattern of mice is basically determined by neonatal androgen, and maintained
by later androgen which develops external genitalia to the form suitable for deposition of tiny spots on
the one hand, and keeps motivation for frequently urinating behavior pattern on the other.

INTRODUCTION

Scent marking behavior is generally observed in
many mammalian species. House mice ( Mus mus¬
culus) mark their home range with their own urine,
usually without any particular postures. This be¬
havior is observed more frequently in males than
in females. The use of ultra-violet light techniques
to visualize the urine spots has facilitated the study
of mouse urination pattern [1] and has provided
various information about it. Male and female
mice increase their rate of urine deposition when
they are exposed to conspecifics of the opposite sex
[2, 3]. Male mice also increase the rate in the
presence of other male mice [1]. These facts
suggest that mice use urine for chemical com¬
munication. However, previous works dealt only
with differences in number of spots and spatial
distribution patterns. If urine marking is impor¬
tant for a mouse to inform other conspecifics about
its status including sex and reproductive ability,
more quantitative analyses must be performed in

Accepted November 19, 1987
Received August 12, 1987

order to understand the meaning of the marking.
Here, we report the difference in distribution of
urine spots of different sizes between male and
female mice under different endocrinological
treatments. We think this kind of analysis will
provide more precise information concerning dif¬
ferences in urination pattern among mice of var¬
ious conditions including sex differences.

The urine marking pattern of male mice is
known to be dependent on androgen in adulthood
[4], and it has been shown that neonatal androgen
treatment enhances the marking activity in female
mice [5]. In the present investigation, analysis of
the marking pattern was focused on the effect of
androgen given in neonatal period as well as in
adulthood, and the distributions of different sizes
of urine spots were compared.

MATERIALS AND METHODS

Animals

The sexually naive male and female mice of
ICR/JCL strain, aged about 70 days, were used.
They were reared in a room maintained at 23°C

856

M. Daumae and T. Kimura

with light:dark cycle of 14: 10 (lights on at 5 a.m.).
All animals were weaned at 4 weeks of age and
then males and females were separated. Mice of
the same sex were housed in groups of 5-6 indi¬
viduals until use.

The urination patterns of castrated males were
analyzed 2 weeks after the operation at 70-80 days
of age and compared with those before the castra¬
tion. To see the effect of androgen in females, we
first recorded the urination pattern in intact
females, and then ovariectomized them at 70-80
days of age and recorded their urination pattern 2
weeks after the operation. Starting on the next
day, subcutaneous injection of 50 pg testosterone
propionate (TP)/0.05 ml sesame oil was given ev¬
ery day for 18 days. Urination patterns of these
females were recorded at 3 days interval through¬
out hormone injection period.

Neonatally androgenized females were provided
by daily subcutaneous injection of 20 pg TP/0.02
ml sesame oil for 3 days beginning at 3 days old.
Oil injected females served as controls. In order to
eliminate the influence of ovarian hormones to
urine marking pattern in these females, the ex¬
perimental animals were ovariectomized after re¬
cording of marking pattern at approximately 100
days of age and tested again 2 weeks after the
operation.

Testing procedures

Urination patterns were collected in the arena
measuring 50x30x30 cm, enclosed with walls of
vinyl chloride plates, the floor being a sheet of
polyethylene-backed absorbent paper (Benchkote,
Whatman) with absorbent surface upper side.

Subjects were tested in the arena overnight for
16 hr (1900-1100), though in some normal and
castrated males 1.5-hr test (1900-2030) was also
performed.

To examine the relationship between the urina¬
tion pattern and the size of the external genital
organ, the length of the penis in males and clitoris
in females was measured for 6 animals of each
experimental group. Experimental animals were
killed and the entire genital organ was fixed in
Bouin’s solution, and the length of penis and
clitoris, from the tip to the base, was measured
with a caliper rule along the caudal midline.

Data analysis

After the test in the arena, the subject was
removed, and the absorbent paper sheets were
dried in air. Urine spots were visualized under
ultra-violet light (365 nm) to trace the outlines of
urine spots with a pencil on the paper. The area of
each spot was measured and recorded using a
graphic digitizer (Oscon SQ-3000) and a micro¬
computer (NEC PC-9801). We used the three
parameters to assess the urination pattern.

(1) Spot number: the total number of urine spots
recorded for each paper, irrespective of size.

(2) Spot size: The area of each urine spot. We put
the data into the histogram according to the 4
gradations of spot area to examine the difference
of urination pattern.

(3) Total area: In order to see the effect of the
urine amount to the urination pattern, we re¬
corded total area of urine spots deposited on the
paper.

Statistical analysis of data was made with Wil-
coxon’s matched-pair, signed ranks test, Mann-
Whitney’s U-test, Friedman’s test or Kruskal-
Wallis test. Lengths of penis and clitoris were
analyzed with t-test.

RESULT

1. Effect of androgen on the urination pattern of
females

As shown in Figure la, intact female mice depo¬
sited significantly fewer urine spots than males (N
= 6, U=0, P<0.05, Mann-Whitney’s U-test).
Ovariectomy did not induce significant change in
the urination pattern of females (N=6,
T=6.5, P>0.05, Wilcoxon’s test), but the number
of spots gradually increased following daily injec¬
tions of TP. Significant increase was observed
after the 15th day as compared to data collected
when they were intact (T=0, P<0.05) or those
obtained after ovariectomy (T=0, P<0.05) (Fig.
lb).

2. Effect of castration in males

The male urine spot number did not decrease
significantly by castration in 16-hr tests (N = 6, T=

Urination Patterns in Mice

857

(a) (b)

Fig. 1. Mean number of urine spots ( + SE) deposited by male and female mice, (a) Mean number of urine spots in
intact males (N = 16) and females (N = 12). (b) Changes in females after ovariectomy and daily injections of TP.
During TP treatment, urination pattern was recorded at 3 days interval (N = 6). OVX=ovariectomized, 3TP =
ovariectomized and TP injected for 3 days.

1 000'
900.
800 ■

c/3 6 00

E-t

o

c/3 500

400

300

1 00

1 6 hours
N = 6

1

INTACT CASTRATED
MALE MALE

200

200

1 50 --

1 . 5 hours
N=7

X

INTACT CASTRATED
MALE MALE

Fig. 2. Mean number of urine spots deposited by male
mice before and after castration.

Left; Overnight test for 16 hr. The difference is
not significant (N = 6, P>0.05, Wilcoxon’s test).
Right; Short time test for 1.5 hr. The difference is
significant (N=7, P<0.05, Wilcoxon’s test).

5, P>0.05, Wilcoxon’s test) (Fig. 2). This result
seems to be inconsistent with the previous work [4]
which demonstrated that castration induced a de¬
crease in urine marking frequency during a 1-hr
test. In order to confirm whether the long term
test conceals the effect of castration, we tested

castrated males for 1.5 hr and found significant
decrease in the number of urine spots (N=7, T=
0, P<0.05, Wilcoxon’s test) (Fig. 2).

3. Effect of neonatal treatment of females with
androgen

The number of urine spots of neonatally
androgenized females was significantly larger than
that of intact females (U = 2, P<0.05, Mann-
Whitney’s U-test) and oil injected controls (U=0,
P<0.01, Mann-Whitney’s U-test), but smaller
than that of intact males (U = 4, P<0.05, Mann-
Whitney’s U-test) (Fig. 3). The spot number of
neonatally androgenized female mice increased by
ovariectomy performed at 100 days of age (Fig. 3).

4. Size distribution analysis

The distribution of size of the urine spots was
estimated in various groups of mice. All data were
collected by 16-hr test. Marked difference was
found between male and female mice. Urine spots
of normal male mice, as well as those of castrated
males, are concentrated in smaller area group (less
than 10 mm2) (normal males, £2 = 43.95. P<0.001;
castrated males, S = 156.5, P<0.001, Friedman’s
test). On the other hand, in females no significant
difference was found among different size groups

858

M. Daumae and T. Kimura

+OVX

Fig. 3. Mean number of urine spots deposited by
neonatally androgenized female mice.

NO = neonatally oil-injected (N = 6).

NA = neonatally androgenized (N = 6).
OVX=ovariectomized (N = 6).

(X2 = 3.2, P>0.05, Friedman’s test). In ovariecto-
mized and TP treated females, urine spots were
concentrated in smaller area group (S = 170, P<
0.001, Friedman’s test), and being similar to male
pattern. Neonatally androgenized females showed
a unique pattern. Their urine spot size was con¬
centrated in middle size (10-100 mm2) (S = 104, P
<0.01, Friedman’s test), and a similar pattern was
found in those androgenized females after ovariec¬
tomy (S=170, P<0.001, Friedman’s test) (Fig. 4).

5. Total area

The mean values of total area of urine spot of 5
groups tested for 16 hr did not show any significant
difference ( X2=1.82 , P>0.05, Kruskal-Wallis
test) (Fig. 5).

6. Correlation between urination pattern and the
length of external genital organs

The lengths of external genital organ of 5 ex¬
perimental groups are shown in Table 1. Castra¬
tion did not induce significant change in penis

length in males. Androgen treatment in ovariecto-
mized females caused a very significant increase in
the length of the clitoris, while the length of clitoris
in neonatally androgenized females were not signif¬
icantly different as compared to intact females.
Differences among each group suggest that the
lengthening of external genitalia is necessary for
the deposition of small spots. There was a signif¬
icant correlation between the number of small
urine spots and the length of the penis or clitoris
(r=0.961, P<0.05).

DISCUSSION

Female mice deposited significantly smaller
number of urine spots than males. This result
corresponds to other previous works [6]. Female
mice increased their urine spot number showing
male-like urination pattern following androgen
treatment, we have confirmed that this effect of
androgen did not last after cessation of injections
(unpublished data). Therefore, androgen seems to
increase the urine spot number in females revers¬
ibly. Castration of males did not decrease urine
spot number when 16-hr tests were adopted, while
in 1.5-hr test, castrated males clearly showed fewer
spots than intact ones. Mean numbers of urine
spots deposited per hour by intact males were
110.3 + 27.9 (±SE) in 1.5-hr tests and 35.4 + 9.4 in
16-hr test, while the corresponding numbers in
castrated males were 22.1 + 7.4 and 22.7 + 5.3.
This suggests that castration induces change in
motivational state which is involved in the initia¬
tion of marking in novel environment, but long¬
term test masks the difference since the mode of
urination is unchanged by castration. Therefore, it
is highly probable that, in males, frequent urina¬
tion with small amount is established by irreversi¬
ble effect of androgen in neonatal period.

Neonatally androgenized female mice deposited
more number of urine spots than control female
mice injected with sesame oil, though the number
was less than that of male mice. This suggests the
importance of early androgen secretion in estab¬
lishment of male type of urination pattern. In
these females, however, though urine spots were
scattered all over the arena as observed in male
mice, the distribution of size of spots are markedly

Urination Patterns in Mice

859

/ / / / / / / /

10 100 1000
SPOT SIZE (mm2 )

10 1001 000
SPOT SIZE (mm2 )

INTACT FEMALE
N= 1 2

r*u i — i r~i i — 3

0 10 100 1000

/ / / /

10 100 1000

SPOT SIZE (mm2 )

OVX FEMALE

1 8 DAYS TP

NA FEMALE

NA+OVX FEMALE

10 100 1000
SPOT SIZE (mm2 )

10 1001 000
SPOT SIZE (mm2)

10 100 1000
SPOT SIZE (mm2 )

Fig. 4. The
area; 0-

urine spot distribution according to the size of each spot. Each urine spot was classified into 4 groups by
10 mm2, 10-100 mm2, 100-1000 mm2, larger than 1000 mm2.

different from that of males. In males, most of
spots were sized less than 10 mm2, while in neona-
tally androgenized females, spots were concen¬
trated in middle size range (10-100 mm2). Maru-
niak et al. [7] noted that deermice and house mice

have relatively long prepuces and distribute their
urine in the form of many small discrete marks,
while hamsters and gerbils have relatively short
prepuces and deposit their urine only infrequently
and characteristically in large pools. Length of

860

M. Daumae and T. Kimura

N=6 IN EVERY GROUP

MALE FEMALE

Fig. 5. Mean total area of urine spots in 5 groups.
O VX = ovariectomized female, TP = ovariectomized
and androgenized female with 18 days TP treatment,
NA = neonatally androgenized female.

Table 1. Mean length of penis and clitoris

Group

Mouse

Length of penis or
clitoris(mean ± S . E . )

A

Male (N=6)

3.98 + 0.12

B

Castrated (N = 6)

4.07 + 0.12

C

Female(TP) (N = 6)

3.39 + 0.13

D

Female (N = 6)

2.28 + 0.13

E

Female(NA)(N = 6)

1.82 + 0.07

t-Test, A vs. B, P>0.05; A vs. C, P<0.01; C vs.

D, PC0.001; D vs. E, P<0.05.

Female(TP) = Ovariectomized females injected

TP for 18 days.

Female(NA) = Neonatally androgenized female.

clitoris of neonatally androgenized females was
significantly shorter than normal females (P<
0.05), being much shorter than penile length of
males (P< 0.001) as shown in Table 1. Taking this
into consideration together with the spot size dif¬
ference, it is suggested that, although the frequent
urination is induced by neonatal androgen regard¬
less of sex, spot size is determined by the morpho¬
logical characteristics of the external genitalia.
The fact that lengthening of the clitoris in females
after androgen treatment in adulthood well corre¬
lates with the small urine spot size also supports
this idea.

As well as Maruniak et al. [7], Powell and Wolff
[6] also considered the genital morphology as an

important factor affecting the sex difference in
spot number. However, there must be other
factors that affect the trait of urine marking.
Difference between urination pattern of males and
that of neonatally androgenized females can be
explained by the difference in genital morphology,
but the difference between intact females and
neonatally androgenized females cannot be at¬
tributed to morphological difference. Clearly,
there must be marked difference in the manner of
urination. Although thorough observation of
urination behavior was not performed because of
lack of any significant urinating posture in mice,
the spatial distribution pattern of spots suggests
that males urinate as they were moving around
while females seem to urinate when they are not
walking. This difference is likely to be related to
the higher mechanism which controls the scent
marking behavior as a whole. Neonatal effect of
androgen seems to affect the mechanism possibly
exists in the brain and promotes its differentiation
into male-type behavior pattern. Androgen in
adulthood seems to act in males to maintain the
motivation level and in females to develop both
motivation and genital morphology.

Ovariectomy in neonatally androgenized
females enhanced the frequency of urination,
while in normal females, ovariectomy did not
cause any significant change. This is contradictory
with our previous experiment where the object
marking in both normal and neonatally androge¬
nized females was inhibited by ovariectomy [5]. It
is probable that marking behavior in novel en¬
vironment is differently enhanced as compared to
presentation of novel object in the familiar en¬
vironment. Further research under various ex¬
perimental situations must be done to elucidate the
relationship between estrogen and urine marking
behavior.

As shown in Figure 5, the total area of spots was
not different among 5 groups, suggesting that
whole amount of urine excreted in overnight time
length is almost the same regardless of the sex and
hormonal conditions. Desjardins et al. [1] also
demonstrated that the amount of urine was not
different between dominant and subordinate
males, though they showed markedly different
urination pattern.

Urination Patterns in Mice

861

Urine marking in mice is generally considered to
be important in the social and reproductive be¬
havior [e.g. 8-10]. Assuming that mice, especially
males, use their urine to advertise their existence
and obtain any benefit, it may be an advantage to
distribute the same amount of urine by small drops
all over the home range. In male mice, such basic
urination pattern is assumued to be established by
neonatal androgen, and their urogenital morphol¬
ogy enables to deposit tiny drops of urine.
Androgen in adulthood probably maintains the
motivation level which is necessary for frequent
urination. These facts seem to show that both
behavioral and morphological adaptations are im¬
portant for urine marking in male mice. It is
suggested that male mice actively use urine odor
for communication as compared to females. In
females, we have reported that vaginal rather than
urine odor affects both approach and sexual be¬
havior of males [5]. Further quantitative re¬
searches concerning the effect of scattering small
amount of urine on social interaction will explain
the significance of urine marking of mice.

ACKNOWLEDGMENTS

This study was supported in part by grants-in-aid for
special research project on the mechanisms of animal
behavior and on biological aspects of optimal strategy
and social structure, and also by a grant No. 61540542
from the Japan Ministry of Education, Science, and
Culture.

REFERENCES

1 Desjardins, C., Maruniak, J. A. and Bronson, F. H.
(1973) Social rank in house mice; differentiation
revealed by ultra-violet visualization of urinary
marking patterns. Science, 172: 939-941.

2 Maruniak, J. A., Owen, K., Bronson, F. H. and
Desjardins, C. (1974) Urinary marking in male
house mice: responses to novel environmental and
social stimuli. Physiol. Behav., 12: 1035-1039.

3 Maruniak, J. A., Owen, K., Bronson, F. H. and
Desjardins, C. (1975) Urinary marking in female
house mice: effects of ovarian steroids, sex experi¬
ence and type of stimulus. Behav. Biol., 13: 211—
217.

4 Maruniak, J. A., Desjardins, C. and Bronson, F. H.
(1977) Dominant-subordinate relationships in cas¬
trated male mice bearing testosterone implants.
Am. J. Physiol., 233: E495-E499.

5 Kimura, T. and Hagiwara, Y. (1985) Regulation of
urine marking in male and female mice: effects of
sex steroids. Horm. Behav., 19: 64-70.

6 Powell, A. J. and Wolff, P. R. (1982) Sex differ¬
ences in mouse urination patterns. Anim. Behav.,
30: 1207-1211.

7 Maruniak, J. A., Desjardins, C. and Bronson, F. H.
(1975) Adaptations for urinary marking in rodents:
prepuce length and morphology. J. Reprod. Fertil.,
44: 567-570.

8 Eisenberg, J. F. and Kleiman, D. G. (1972) Olfac¬
tory communication in mammals. Ann. Rev. Ecol.
Syst., 3: 1-32.

9 Bronson, F. H., (1976) Urine marking in mice:
causes and effects. In “Mammalian Olfaction, Re¬
productive Processes and Behaviour”. Ed. by R. L.
Doty, Academic Press, New York, pp. 119-141.

10 Wolff, P. R. and Powell, A. J. (1984) Urine pattern
in mice: An analysis of male/female counter¬
marking. Anim. Behav., 32: 1185-1191.

'

ZOOLOGICAL SCIENCE 5: 863-868 (1988)

© 1988 Zoological Society of Japai

Coprophagy as an Innate Behavior in the Mouse

Koichi Y. Ebino, Koichiro Yoshinaga, Toru R. Saito* 1
and Kazuaki W. Takahashi2

Toxicology Division, Institute of Environmental Toxicology, Uchimoriya-cho 4321,
Mitsukaido-shi, Ibaraki 302-02, and 2Department of Laboratory
Animal Science, Nippon Veterinary & Zootechnical College, Kyonan-cho
1-7-1, Musashino-shi, Tokyo 180, Japan

ABSTRACT — In order to verify whether or not coprophagy is an innate behavior in the mouse, two
kinds of experiments were conducted. In the first experiment, a fecal suspension was administered to
mice to see if the frequency of coprophagy could be reduced by such treatment. Neither the frequency
nor the diurnal pattern of coprophagy was affected by the treatment, which demonstrates that the
nutrients in the administered feces do not cause negative feedback on the frequency of coprophagy. In
the second experiment, sucklings, while still unable to open their eyes, were treated so that their ears
would not open. They were forcedly weaned and reared in isolation to determine if coprophagy would
appear spontaneously. The onset and posture of coprophagy in the forcedly-weaned sucklings were
essentially identical to those of the controls. Moreover, there were no differences in the frequency and
in the diurnal pattern of coprophagy between the forcedly-weaned sucklings and the controls at 28 days
old. From these results, we concluded that coprophagy is an innate behavior in the mouse.

INTRODUCTION

The authors have previously studied coprophagy
in mice from the viewpoints of behavior and
nutrition [1-7]. The mouse obtains B vitamins
through coprophagy, especially vitamin Bi2, and
these vitamins are synthesized by intestinal mi¬
croorganisms. Thus, coprophagy seems to play an
important role in the nutrition of the mouse [3].
However, germ-free mice exhibit coprophagy
identical to that observed in conventionally housed
mice [4]. In this case, coprophagy in the mouse
occurred even when it was no longer nutritionally
beneficial to the mouse. It appears, then, that
coprophagy is essentially an innate behavior in the
mouse which is performed habitually.

The present study was designed to confirm that
coprophagy is inherent in the mouse. The follow¬
ing two kinds of experiments were carried out. In

Accepted November 19, 1987
Received July 17, 1987

1 Present address: Department of Pharmacology,

Kyorin University School of Medicine, Mitaka, Tokyo
181, Japan.

the first experiment, mice were administered feces
in a suspension to see if the frequency of cop¬
rophagy could be reduced. In the second experi¬
ment, sucklings forcedly weaned before their eyes
and ears were open were reared in isolation and
observed to determine whether or not coprophagy
would appear spontaneously.

MATERIALS AND METHODS

Experiment I: Effect of oral administration of a
fecal suspension on the frequency of coprophagy

Male mice of the ICR strain, purchased at 6
weeks old (CLEA Japan, Tokyo), were acclima¬
tized to the experimental environment until they
were 10 weeks old. Animals were kept in a
semibarrier-sustained room with an ambient
temperature of 24 + l°C and a relative humidity of
55 + 5%. The room was ventilated 15 times per
hour with all fresh air and illuminated for 14 hr a
day (lights on at 05:00 and off at 19:00). Eight
animals were housed in each of several stainless
steel cages (310W X 440D X 230 H mm) with raised

864

K. Y. Ebino, K. Yoshinaga et al.

wire-mesh floors and supplied with a pellet diet
MF (Oriental Yeast Co., Ltd., Tokyo) and water
ad libitum.

The suspension of feces was prepared as follows.
Feces excreted by 50 male mice from 07:00 to
1 1 : 00 were collected and water was added at the
rate of 1.6 ml per 9 pellets. This mixture was
kneaded together by mortar and pestle to make a
homogeneous suspension. The suspension was
further subjected to a homogenizer (Kinematica
GmbH, Switzerland) for a few minutes. The
suspension was administered by a stomach tube to
8 mice at the rate of 1.0 ml per animal twice in the
afternoon (around 12:00 and 15:00) when cop-
rophagous activity was ordinarily lower.

The number of feces administered daily, 9 pel¬
lets per mouse, was almost equal to or more than
the number of pellets ingested by coprophagy, i.e.,
8.6 pellets per day in male mice of 10 weeks old
[5],

water a fecal suspension water

nmn|uiitmmm|nutn

ft 1} ft ft ft ft

.oration of w - 2 w - 1

treatment

j : administration of water or a fecal suspension
: recording of animals

Fig. 1. Schedule of experiment I.

The experiment was carried out as shown in
Figure 1. Twelve mice were given water for one
week; then 8 mice were selected by reference of
weight or condition, and were given the fecal
suspension for 2 weeks. Finally, water was given
again for one week. When water was adminis¬
tered, the volume and the frequency of administra¬
tion were the same as those of the suspension.
With this schedule, coprophagy by individual mice
was recorded by videotape recorder at set points
(24 hr each). An infrared ray projector was used
during the dark hours of the photoperiod. Cop¬
rophagy was observed later by reproducing the
tapes. When more than half of a fecal pellet was
eaten, it was recorded as one act of coprophagy.
During the treatment period, mice were weighed
twice a week.

Experiment II: Coprophagy in sucklings forcedly
weaned before their eyes and ears were open, and
reared in isolation thereafter

Six pregnant mice were housed individually in
aluminum cages with raised wire-mesh floors
(210W X 330D X 140 H mm) in the same room used
for Experiment I, and were allowed to give birth.
On day 4 of lactation (the day of parturition was
counted as day 0 of lactation), the number of
sucklings was adjusted to 6 (3 animals of each sex).
On day 12, the auricles of the sucklings were
treated with a cyanacrylate binding agent to pre¬
vent development of the ears and hearing. On day
13 or 14 of lactation, 1 to 3 males were forcedly
weaned from each litter after it was confirmed that
their eyes and ears had not opened. Twelve male
sucklings were then housed individually in polycar¬
bonate cages with raised wire-mesh floors (92W X
205D X 127 H mm) and were reared in isolation
from other mice. Animals were provided with a
pellet diet MF and water ad libitum.

Coprophagous activity by these isolated mice
was recorded from day 16 to day 20, and on day 28.
For controls, coprophagy in another litter or 4
males and 4 females, housed in a polycarbonate
cage (210W X330D X 180 H mm) with the mater¬
nal animal, was recorded from day 14 to day 19.
On day 28, only the activities of 4 males were
recorded.

Statistical analysis

Student’s Ftest was applied to evaluate any
significant differences.

RESULTS

Experiment I

No specific abnormalities were found in body
weight or general condition of mice treated with
the fecal suspension, demonstrating that the treat¬
ment did not have any harmful effects on them
(Table 1). One out of the eight mice died from a
technical error during the administration of the
suspension.

Observation of coprophagy revealed that nei¬
ther the frequency nor the diurnal pattern of

Coprophagy in the Mouse

865

Table 1. Body weight changes during fecal suspension treatment

Water

Fecal suspension

Water

0

2

5

9

13

16

20

23

27 days

Body

37.0

37.7

36.8

36.6

36.1

36.1

36.1

36.3

36.9

weight(g)

±1.6

±1.9

±2.1

±1.3

±1.6

±1.3

±1.6

±1.4

±1.3

[8]

[8]

[8]

[8]

[8]

[7]

[7]

[7]

[7]

Values represent means ± standard deviation.
[ ]: No. of animals.

Table 2. Frequency of coprophagy during fecal suspension treatment

Water

Fecal suspension

Water

7

9

15

21

23

28 days

Frequency of

9.9

11.3

9.3

13.7

12.6

11.0

coprophagy per

±4.1

±4.0

±3.7

±5.6

±6.6

±3.5

mouse per day

[8]

[8]

[8]

[7]

[7]

[7]

Values represent means ± standard deviation.
[ ]: No. of animals.

o 0.5 --

on.

- 0500 1900

Li.

Time of day

on the 7th day of water administration on the 7th day of a fecal suspension administration

Fig. 2. Comparison of diurnal pattern of coprophagy between controls and treated mice. Each bar represents mean
frequency of coprophagy by 8 mice.

coprophagy was affected by treatment with the
fecal suspension (Table 2, Fig. 2).

Experiment II

The twelve forcedly-weaned sucklings opened
their eyes and ears at 14 or 15 days old; they took
food and water by themselves and excreted feces

autonomously. After being weaned, the animals
exhibited a loss of body weight for a few days but
gained weight thereafter and no abnormalities
were found in their general condition and be¬
havior. Coprophagy was first observed in six of the
animals at 16 days old, five more animals at 17 days
and the last at 18 days old. Each animal exhibited

866

K. Y. Ebino, K. Yoshinaga et al.

Table 3. Frequency of coprophagy in naturally-weaned and forcedly-weaned pups

The day of

Frequency of

Group

onset of

coprophagy at

coprophagy

28 days old

Naturally-weaned pups

17 to 18 days old

12.0+1.8

[4]

Forcedly-weaned pups

16 to 18 days old

13.4 + 3.5
[12]

Values represent means + standard deviation.
[ ]: No. of animals.

Naturally- weaned pups

Forcedly-weaned pups

Fig. 3. Comparison of diurnal pattern of coprophagy between naturally-weaned and forcedly-weaned pups at 28
days of age.

coprophagy as soon as it could reach its anus with
its mouth. The first act of coprophagy occurred
quite naturally, as of the animal had been attracted
by its own feces.

In the control animals, coprophagy was first
observed at 17 or 18 days old, 4 or 5 days after the
opening of the eyes and the ears. Coprophagy first
occurred a few days after sucklings began to ex¬
crete feces autonomously. As seen in the forcedly-
weaned animals, sucklings at this age were barely
able to reach their anuses with their mouths.

There were no differences between the ex¬
perimental and control mice in regard to the age of
initiation of coprophagy or in the frequency and
diurnal pattern of coprophagy at 28 days of age
(Table 3, Fig. 3).

DISCUSSION

From the results of Experiments I and II, cop¬
rophagy appears, in fact, to be an innate behavior
in the mouse.

Experiment I demonstrated that nutrients in
feces could not subdue the appetite for coprophagy
in the mouse. Moreover, it has become evident
that coprophagy itself has a steady and proper
rhythm.

Then, how does the appetite for coprophgy
occur, and how is it subdued? From the viewpoint
of behavior, coprophagy in the mouse is divided
into four steps, in principle. First, judging from its
posture, the mouse seems to be aware of feces
coming near to the anus. The mouse excretes the
feces by a defecation reflex, and takes its feces

Coprophagy in the Mouse

867

directly by the mouth. Finally, the mouse masti¬
cates the feces while holding it with the forelimbs
[2], Therefore, each round of coprophagy seems
to begin with the entry of feces into the rectum,
since the feces stretch the rectal wall and this
stimulation is transmitted to the spinal cord and
brain through the afferent pelvic nerve [8]. This
must be the first signal to the mouse informing it of
the presence of feces to be ingested. In other
words, the feces itself existing in the rectum may
be the sign stimulus for each round of coprophagy.
However, since mice do not eat all their feces,
there must be something characteristic about the
feces to release coprophagy. Possibilities include
the size of the feces, the number of feces con¬
nected in the colon and rectum, or the descending
speed of the feces from the colon to the rectum, for
mice have been observed to reingest 2 or 3 fecal
pellets consecutively [2]. Furthermore, in rabbits,
reingested soft feces have been reported to be
excreted rapidly [9] as clusters of small, moist,
discrete pellets with a distinctive sheen [10]. As for
size, a cluster of soft feces is much larger than each
hard fecal pellet. We suspect that the mechanism
informing the host as to which feces is to be
ingested is probably identical among species which
practice coprophagy.

On the other hand, the appetite is subdued when
the end has been completed in general. Why then,
were the mice not satisfied with the administered
feces? Perhaps the appetite for coprophagy is
subdued when each component of coprophagy has
been successfully performed, and consequently,
the feces have been reingested. It is certain that
each component of coprophagy is conducted via a
reflex circuit through the spinal cord and/or the
brain, which is initiated by the presence, smell,
vision, touch or taste of the feces. The mouse does
not seem to be satisfied if the end result of
coprophagy is achieved unless it experiences the
feces with its own sensory organs. Being a kind of
food when reingested, the feces must be a great
attraction, particularly to the sense of smell or
gustation.

Then why were the nutrients in the administered
feces unable to affect the frequency of cop¬
rophagy? We have already observed the suppres¬
sive effect of a vitamin B12 fortified diet on the

frequency of coprophagy in the mouse [3], The
fortification was magnified 365 fold against the
content in the basal diet. In the present study, the
amount of vitamin B12 in the administered feces,
the most probable nutrient associated with cop¬
rophagy in the mouse, was estimated to be 4 fold
greater, at maximum, against the daily intake of
vitamin B12 from the normal diet. Therefore, the
amount of nutrient(s) in the administered feces
might not have been enough to affect the frequen¬
cy. Although further exeriments may be needed to
answer the above question convincingly, we now
feel that the frequency of coprophagy is regulated
largely by the quality of the diet.

The second experiment was conducted at Tin¬
bergen’s recommendation [11]. He said, in his
book entitled The Study of Instinct, “The only way
to find out what behaviour is innate and what is
acquired during individual life is to raise indi¬
viduals in isolation, to observe the development of
their behaviour, and to study the influence of
different environments upon it”. Since the forced¬
ly-weaned sucklings were taken before their eyes
and ears were open and were reared in isolation
and sheltered from other mice, they did not have a
chance to imitate their dams or other animals.
Even under such conditions, the onset of cop¬
rophagy, the posture, the frequency and the diur¬
nal pattern of coprophagy at 28 days of age in the
experimental group were as same as those of the
controls. Furthermore, there was no chance for
“trial and error” or “learning” of coprophagy
judging from the quite natural and immediate
onset of coprophagy when accompanied by
appropriate physical development. Since any
effects of isolation on coprophagy by the forcedly-
weaned sucklings were absent, we have concluded
that coprophagy is an innate behavior in the
mouse.

In summary, coprophagy is an important be¬
havior nutritionally beneficial to the mouse, and
appears to be controlled as follows. Being an
innate behavior, each round of coprophagy is
performed reflexively as estimated above and re¬
peated to a certain degree during the day based on
the quality of the daily diet. Then, a relatively
steady and proper diurnal pattern of coprophagy is
formed in association with the 24 hr rhythm of the

868

K. Y. Ebino, K. Yoshinaga et al.

meal pattern, as has been reported in rabbits [12].
In addition, the frequency of coprophagy vacillates
with changes in the nutritional requirements of the
host, i.e., those caused by physiological conditions
such as growth [5], pregnancy or lactation [6].

ACKNOWLEDGMENTS

The authors are grateful to Dr. Yasuhiko Shirasu,
Director of Toxicology Division, Institute of Environ¬
mental Toxicology, for his advice on this study. The
authors are also grateful to Messrs. Hisao Kawakatsu,
Yoshio Takizawa, Hiro-o Suzuki and Yoshihiro Horie,
also of Toxicology Division for their assistance.

REFERENCES

1 Takahashi, K. W., Saito, T. R., Suzuki, W., Ka-
tsuyama, M., Sakuma, S., Tauchi, K., Kawanishi,
H. and Imamichi, T. (1983) Studies on coprophagy
in the mouse. Zool. Mag., 92: 397-401.

2 Takahashi, K. W., Ebino, K. Y., Saito, T. R. and
Imamichi, T. (1985) Strain difference in copropha-
gous behavior in laboratory mice ( Mus musculus).
Zool. Sci., 2: 249-255.

3 Ebino, K. Y., Suwa, T., Kuwabara, Y., Saito, T. R.
and Takahashi, K. W. (1986) Analyses of consti¬
tuents of feces and the effect of a vitamin B12
fortified diet on coprophagy in the mouse. Exp.

Anim., 35: 381-386 (In Japanese with English ab¬
stract).

4 Ebino, K. Y., Amao, H., Suwa, T., Kuwabara, Y.,
Saito, T. R. and Takahashi, K. W. (1987) Cop¬
rophagy in the germfree mouse. Exp. Anim., 36:
33-37 (In Japanese with English abstract).

5 Ebino, K. Y., Suwa, T., Kuwabara, Y., Saito, T. R.
and Takahashi, K. W. (1987) Lifelong coprophagy
in male mice. Exp. Anim., 36: 273-276 (In Japanese
with English abstract).

6 Ebino, K. Y., Suwa, T., Kuwabara, Y., Saito, T. R.
and Takahashi, K. W. (1988) Coprophagy in female
mice during pregnancy and lactation. Exp. Anim.,
37: 101-104.

7 Ebino, K. Y., Yoshinaga, K., Saito, T. R. and
Takahashi, K. W. (1988) A simple method for pre¬
vention of coprophagy in the mouse. Lab. Anim.,
22: 1-4.

8 Jensen, D. (1976) Gastrointestinal motility and its
regulation. In “The Principles of Physiology”.
Prentice-Hall, New York, pp. 835-845.

9 Eden, A. (1940) Coprophagy in the rabbit: Origin
of ‘Night’ faeces. Nature, 145: 628-629.

10 Thacker, E. J. and Brandt, C. S. (1955) Cop¬
rophagy in the rabbit. J. Nutr., 55: 375-386.

11 Tinbergen, N. (1969) The Study of Instinct. The
Clarendon Press, Oxford, re-issued., pp. 51-54.

12 Hornicke, H. (1981) Utilization of caecal digesta by
caecotrophy (soft faeces ingestion) in the rabbit.
Livestock Production Science, 8: 361-366.

ZOOLOGICAL SCIENCE 5: 869-874 (1988)

© 1988 Zoological Society of Japan

Ecological Distribution and Habitat-linked Density
of Colonies of Stenogastrine Wasps
in Tropical S. E. Asia* 1

Ryoh-ichi Ohgushi, Soichi Yamane2 and Shoichi F. Sakagami3

Department of Biology, Faculty of Science, Kanazawa University, Kanazawa 920,

2 Biological Laboratory, Faculty of Education, Ibaraki University, Mito 320, and 3 Institute
of LowTemperature Science, Hokkaido University, Sapporo 060, Japan

ABSTRACT — The species diversity and the colony density of stenogastrine wasps were compared
among 3 types of vegetation in tropical S. E. Asia. A climax vegetation of submontane rain forest in
Malay Peninsula, showed the largest number of both species and colonies. Native palm orchards with
scattered remains of immature rain forest in Sumatera Barat, had fewer species and colonies than in the
climax rain forest. Finally, a farmland at the foots of hills in Sumatera Barat, was poorest in both species
diversity and colony density. Liostenogaster and Eustenogaster species were mostly confined to humid
forests, while Parischnogaster , especially species of the P. mellyi- group, extended their distribution to
opener habitats. The open farmland was occupied only by poor populations of P. mellyi and
Parischnogaster aff. jacobsoni.

INTRODUCTION

Stenogastrine wasps are in distribution restricted
to tropical and subtropical S. E. Asia and New
Guinea. By its primitively social life [1], this group
seems to be an important “missing ring” of social
evolution in wasps. But our knowledge about their
behavior and ecology is still insufficient because of
their small and inconspicuous form and secretive
life style.

In recent years a considerable number of papers
on the stenogastrine biology have been published.
Most of them, however, either describe the nest
architecture [2-5] and behavior patterns [5, 6], or
deal with social relations among females [6-8].
Aspects which relate to the ecological distribution,
habitat preference, colony density, etc., have not
yet been thoroughly studied.

In paralell with taxonomic and biological sur¬
veys of the stenogastrine wasps in Malaysia and
Indonesia since 1964, we have made ecological
censuses of their colonies in relation to environ¬

Accepted October 23, 1987

Received June 12, 1987

1 Contribution No. 24 of Sumatra Nature Study (En¬

tomology)

mental conditions. In this paper, we try to com¬
pare the species diversity and the colony density of
these wasps among three types of habitats, which
may represent the humid virgin forests, immature
forests intervened with palm orchards and open
cultivated farmland, respectively.

AREAS AND METHODS FOR STUDY

Field studies were carried out at three localities,
one in Peninsular Malaysia and two in the Province
of Sumatera Barat, Indonesia. These areas repre¬
sent three vegetations, as mentioned above. The
location of areas and the type of environments are
summarized as follows.

Bukit Frazer (Fig. 1): Situated on the SW-slope
of 600 m in altitude of a mountaneous district at
about 30 km northeast of Kuala Lumpur. Covered
with a dense forest nearly comparable to the
subalpine primary rain forest defined by Richards
[9] and penetrated by a paved road, which was
partly shaded with the forest crown. Environmen¬
tal conditions seemed to be comparable with the
rain forest edge. Its ground was covered with
bushes, vines and ferns. Soil was humid and run
through by small streams.

870

R. Ohgushi, S. Yamane and S. F. Sakagami

Fig. 1. Submontane rain forest at Bukit Frazer study
area.

Teluk Kabung (Fig. 2): Situated on hills at about
12 km south of Padang, along an unpaved road
connecting Padang and Painan. The road ran up
and down the hills of 80—150 m in altitude, which
closely approached the coast of the Indian Ocean.
The hills were covered mostly with planted palms
mixed with some native trees and scattered re¬
mains of immature rain forest. Since the both sides
of the road were relatively open, not covered with
tall trees, the road was exposed to the direct sun.
Tall native trees remained only sporadically.
Several small streams crossed the road and the
forest remained near streams or ravins. The
ground of the forest and plantations were not
much humid. Stenogastrine nests were mainly
found in humid and shaded places near the water.

Lubuk Mintrun (Fig. 3): Situated in an orchard
of the Provincial Horticultural Experiment Station
at about 6 km northeast of Padang. It was in a
cultivated area of lowland plain covered with fruit
trees of citrus, rambutan, and palms at hillside,
and a rice field at the basin. Laterite soil was
exposed to sunbeams and dried up. Trees in the

Fig. 2. Secondary forest along a road at Teluk Kabung
study area.

orchards and plantations were planted sparsely
and most of fruit trees were young. A small paddy
field, 10 houses, 7 of which were inhabited by
farmers, and a small cottage were incorporated in
the area.

In each area, the number of colonies was
counted by a line census method.

Bukit Frazer: Line census of 54 person hours

Fig. 3. Citrus and rambutan orchard in a farmland at
Lubuk Mintrun study area.

Ecological Distribution of Stenogastrine Wasps

871

along the both sides (ca. 6 m wide for each) of the
road ca. 3 km long, Teluk Kabung: Line census as
above, but 30 person hours, Lubuk Mintrun: Cen¬
sus of 60 person hours in citrus and rambutan areas
covering a total of 10 ha.

Nests were discovered on twigs of tree, leaves of
fern, tree rootlets, inside the hollows made at
roadcut, underside of overhanging cliffs or rocks,
edges of roofs of cottages, etc. For each nest, the
species name, or if unknown, the nest type [4] was
recorded. Some nests were identified only to the
genus. Henceforth, the colonies mean active nests
which were occupied by one or more adult wasps.

Although the census at Lubuk Mintrun was not
the line census in a strict sense, the counts were
chiefly made along rows of planted trees, and the
values could be converted into those of line-base,
this area nearly equaling 4.5 km of a census line.

RESULTS

Table 1 shows the numbers of colonies found in
the three areas and those per 100 m of census line.

The values may roughly reflect the relative density
of colonies in these areas, being highest in Bukit
Frazer (B. F.) and lowest in Lubuk Mintrun (L.
M.). The number of species was also abundant in
B. F. with 9 nest types (or 9 species), while only 2
types (species) in L. M. Teluk Kabung (T. K.)
occupied a position intermediate between B. F.
and L. M.

In the genus Parischnogaster , 5 nest types were
found in B. F., 4 in T. K. and 2 in L. M. Type Ps4
(a type of the P. striatula-g roup, cf. [4]) was found
only in B. F., but most abundant there. On the
contrary, in T. K. P. striatula (type Psl) and P.
mellyi (type Pml) were most abundant. P. mellyi
was the commonest species in the three areas. This
species seems to have adopted to relatively open
and less humid environments such as secondary
forests and abandoned farms. In L. M., only a few
colonies of P. mellyi and Pjl (a type of the P.
jacobsoni- group) were found. Two nest types
(probably belonging to different species) of Lioste-
nogaster were very common at B. F. Two nest
types of Eustenogaster were found in B. F. and T.

Table 1 . Number of colonies (active nests) of social wasps in three areas surveyed

Species or nest types

Bukit Frazer

Teluk Kabung

Lubuk Mintrun

Stenogastrinae
genus Parischnogaster

P. striatula (type Psl)
type Ps4

92 ( 3.1 )

161 ( 5.4 )

127 ( 4.2 )

P. mellyi (type Pml)

3 ( 0.1 )

114 ( 3.8 )

26 ( 0.6 )

type Pjl

20 ( 0.7 )

3( 0.1 )

5( 0.1 )

type Pj2

1 ( 0.03)

2 ( 0.07)

genus Liostenogaster

L. flavolineata
type L3

301 (10.0 )

57 ( 1.9 )

genus Eustenogaster

type El
type E2

7( 0.2 )

3( 0.1 )

2 ( 0.07)

Total

other social wasp

645 (21.5 )

248 ( 8.3 )
not sensused

31 ( 1.0 )

Polistinae

8( 0.3 )

21 ( 0.7 )

type Ps = the P. striatula-group\ type Pm = the P. mellyi- group;

type Pj = the P. jacobsoni- group; type L = Liostenogaster-, and type E = Eustenogaster. Details see

[4]-

( ) = number of colonies per 100 m of census line.

872

R. Ohgushi, S. Yamane and S. F. Sakagami

Table 2. Nest sites and substrates of stenogastrine wasps observed in three areas surveyed

Species or
nest types

Rootlet at
thin

(<0.3 mm)

roadcut

thick

(0.4 mm<)

Leaf or stem
of fern in
underbush

Vine or fiber
on tree or
eave of hut

Innersurface
of concrete
tunnel

Parischnogaster

P. striatula

213

6

type Ps4

P. mellyi

3

140

161

type Pjl
type Pj2

3

29

Liostenogaster

L. flavolineata

301

type L3
Eustenogaster

57

type El andE2

12

For the nest types, see Table 1 and [4],

In these areas, stenogastrine nests in tree hollows or underground cavities were not found.

K., with a larger number of colonies in the former.
The number of polistine colonies (genera Ropali-
dia and Polistes ) found in the surveyed area (ex¬
cluding T. K.) were scarce as given in Table 1.

To evaluate the diversity of the stenogastrine
faunas in the three areas, Simpson’s index of
diversity [13] was calculated. The values were
0.302 for B. F., 0.471 for T. K. and 0.720 for L.
M., respectively. Showing that the fauna is most
complex and diverse in the rain forest and opposite
in the farmland.

As seen in Table 2, there are clear differences
among the species or nest types in preference for
nest sites and substrates. Two species of Liosteno-
gaster and one Parischnogaster (Ps4) formed dense
nest aggregations inside concrete cement pipes
used for watercourse, which offered a dark and
humid environment. In contrast to them, some
Parischnogaster species, especially P. mellyi and
Pjl built their nests in open, dry environments
such as farmland, garden and edge of immature
rain forests.

DISCUSSION

It is well known that the land faunas of mammals
and insects in Malay Peninsula and Sumatra re¬
semble each other because of their connection
during most time of glacial ages [10, 11], Because

the areas surveyed are geographically not much
distant and belonging to the same climatic zone,
we can regard them as belonging to the same bio-
and ecogeographical zone when we discuss the
habitat preference.

For social wasps and bees, the habitat offers not
only the feeding and mating sites, but also the sites
for settling their colonies. Thus, environmental
conditions are important to determine both the
diversity of species involved and the colony den¬
sity.

Up to the present, some field entomologists
have pointed out that the stenogastrine wasps
mostly built their nests in shaded and humid places
such as underbushes, underside of overhanging
roadcuts, stream bank and roof of artificial build¬
ings [2, 3, 5]. Based on the present observations
and previous studies [4, 5], we classify their nesting
sites as follows, according to the micro-
topographical factors and the conditions of sub¬
strates.

a) Undersurfaces of overhanging roadcut or
stream banks.

b) Undersurfaces of overhanging rocks, con¬
crete bridges and inner surfaces of concrete
cement tunnels for watercourse.

c) Inside of tree hollows or underground cavi¬
ties.

d) Twigs, stems, leaves and vines of under-

Ecological Distribution of Stenogastrine Wasps

873

bushes.

e) Undersurfaces of roofs, the eaves and floor
of huts and houses.

The nest site preference by various wasps and
bees may relate to their behavioral (including
social organization) and ecological prerequisites.
The stenogastrine wasps, whose distribution is
limited to southeast Asia and New Guinea [12],
might have evolved in these areas where rain forest
has been fairly prosperous. From Table 1 and a
calculation of Simpson’s index, we can recognize
that the highest diversity of species (or nest types)
and density of colonies are found in the mature
rain forests. Colonies were also abundant in
immature rain forest areas, but they were found
there mostly at shaded and humid microhabitats.
This suggests that areas of the tropical rain forests
have been their main homeland and opener and
drier habitats have been secondarily colonized by
them.

Observations conducted at various environ¬
ments of tropical Asia [5] and results shown in
Table 1 suggest that subfamilies of the family
Vespidae have selected different types of environ¬
ments for nidification. Generally speaking, the
Stenogastrinae have chosen dark and humid places
in the climax and subclimax rain forests, while the
Polistinae did lighter and drier places such as
grassland and the early stages of succession.

Some species of the P. mellyi- group and P.
jacobsoni- group nest in open habitat outside rain
forests such as roadcuts in open places, orchards
and houses in villages. Hansell [14] supposed that
the suburban habitats of P. mellyi seem not to be
native for the stenogastrines. The P. striatula-
group, another species group of Parischnogaster,
however, seems to remain at darker and more
humid habitats close to rain forests, such as the
roadcuts under the forest canopy, underground
cavities, etc.

Most colonies of Liostenogaster and Eustenogas-
ter were found in dark and humid places as under¬
bushes near the ground of rain forests. These
environments seem to offer main habitats to them.
Among these groups, L. flavolineata and L3, as
well as Ps4, were found to nest inside artificial
constructs, e.g. concrete cement tunnels. Their
original nest sites were probably natural caves and

underground cavities. These environments are
presently not restricted to the rain forests, but also
occur in opener and drier places. So, if these
environments exist, the wasps can inhabit even the
secondary forests [15]. This shows that conditions
of microhabitat also affect the distribution of these
wasps.

ACKNOWLEDGMENTS

We would like to express our hearty thanks to Prof.
Tatuo Kira, the Chairman of the Osaka City University
Expedition to Southeast Asia, Prof. Shunzo Kawamura,
the Chairman of the Sumatra Nature Study Project and
the staffs of Andalas University, Indonesia, for their help
during the course of our study. We also indebted to Dr.
Kimio Yoshikawa, who made the field survey in Bukit
Frazer with us. These studies were suppported by
Grants-in-Aid for Overseas Scientific Survey (1981
No. 56041027, 1983 No.58041032) from the Ministry of
Education, Science and Culture, Japan.

REFERENCES

1 West-Eberhard, M. J. (1978) Phylogeny and evolu¬
tion of social behaviour in wasps. J. Kansus En-
tomol. Soc., 51: 832-856.

2 Williams, F. X. (1919) Philippine wasp studies. Part
II. Description of new species and life history
studies. Rep. Exp. Stn. Hawaii Sugar Plant. Assoc.,
Entomol. Ser., 14: 19-196.

3 Pagden, T. H. (1958) Some Malayan social wasps.
Malay. Nat. J., 12: 131-148.

4 Ohgushi, R., Sakagami, F. S., Yamane, S. and
Abbas, N. D. (1983) Nest architecture and related
notes of stenogastrine wasps in the Province of
Sumatera Barat, Indonesia (Hymenoptera, Vespi¬
dae). Sci. Rep. Kanazawa Univ., 28: 27-58.

5 Ohgushi, R. (1986) A review on the biology and
ecology of stenogastrine wasps. Hymenoptelist’s
Communication, No. 25/26: 1-76 (in Japanese).

6 Pardi, L. and Turillazzi, S. (1981) Behavior and
social organization of Parischnogaster nigricans ser-
rei (Du Buysson) (Hymenoptera, Vespidae). Moni-
tore Zool. Ital., 15: 322-323.

7 Yoshikawa, K., Ohgushi, R. and Sakagami, F. S.
(1969) Preliminary report on entomology of the
Osaka City University 5th Scientific Expedition to
Southeast Asia 1966, with description of two genera
of stenogastrine wasps by J. Van Der Vecht. Nat.
Life S. E. Asia, 6: 158-182.

8 Yamane, S., Sakagami, F. S. and Ohgushi, R.
(1983) Multiple behavioral options in a primitively
social wasp, Parischnogaster mellyi. Insect. Soc., 30:

874

R. Ohgushi, S. Yamane and S. F. Sakagami

412-415.

9 Richards, P. W. (1952) The Tropical Rain Forest.
Cambridge University Press, London.

10 Unbgrove, J. H. F. (1949) Structural History of the
East Indies. Cambridge University Press, London.

11 Kuenen, H. (1950) Marine Geology. John Willy &
Sons, Inc., New York & London.

12 Vecht, V. D. (1975) A review of the genus Steno-
gaster Guerin (Hymenoptera: Vespidae). J. Aust.
Entomol. Soc., 14: 283-308.

13 Simpson, E. H. (1949) Measurement of diversity.

Nature, 16: 688.

14 Hansell, M. H. (1983) Social behaviour and colony
size in the wasp Parischnogaster mellyi (Saussure),
Stenogastrinae (Hymenoptera, Vespidae). Proc.
Natl. Akad. Wet. C., 86: 167-178.

15 Ohgushi, R. and Yamane, S. (1983) Supplementary
notes on the nest architecture and biology of some
Parischnogaster species in Sumatera Barat (Hyme¬
noptera, Vespidae). Sci. Rep. Kanazawa Univ., 28:
69-78.

ZOOLOGICAL SCIENCE 5: 875-881 (1988)

© 1988 Zoological Society of Japan

Pliciloricus hadalis (Pliciloricidae), a New Loriciferan Species Collected
from the Izu-Ogasawara Trench, Western Pacific

Reinhardt M0bjerg Kristensen and Yoshihisa Shirayama1

Institute of Cell Biology and Anatomy, University of Copenhagen, Universitetsparken 15,
DK-2100 Copenhagen 0, Denmark, and 1 Ocean Research Institute, University
of Tokyo, 1-15-1 Minamidai, Nakano-ku, Tokyo 164, Japan

ABSTRACT — A new species of the Loricifera, Pliciloricus hadalis nov, sp. (Pliciloricidae), is described.
This species is distinguished from the other Pliciloricus species by its jar-shaped lorica, the mucous coat
on the larval lorica and the 30 sculptured plates in the adult thorax. This is the first discovery of
Loricifera not only from the hadal bathymetric zone but also from the fine sediment (red clay).
Moreover, this is the first record of Loricifera from the Western Pacific area. These facts suggest that
Loricifera has a high ability of adaptation and is distributed ubiquitously in the marine environment.

INTRODUCTION

After the first recognition of meiofauna [1],
zoologists have paid keen attention to this small¬
sized benthic organisms. Their efforts have led to
numerous discoveries of new taxa. One of the
most distinctive taxonomic work is the description
of new phylum, Loricifera, by Kristensen [2]. This
group of animals was characterized by the follow¬
ing features. 1) The body is segmented into five
regions: mouth cone, head, neck, thorax, and
abdomen. 2) A variable number of sensory (and
possibly locomotory) appendages (scalids) are
attached to the head and the neck regions. 3) The
abdomen is covered with a lorica (girdle), which
the name, Loricifera, has been derived from.

The first species of the Loricifera, Nanaloricus
mysticus Kristensen, 1983, was found in clean shell
gravel at 25 m deep. Since the first description,
nine species, representing three genera, two fami¬
lies and one order, have been described, all from
coarse subtidal sediments of the Atlantic Ocean [2,
3],

This article reports the first discovery of lori¬
ciferan from the Izu-Ogasawara Trench at a depth
of 8260 m. The present discovery of new lori¬
ciferan from red clay of the hadal zone [4] of the

Accepted November 27, 1987
Received September 17, 1987

Western Pacific Ocean in addition to several unde¬
scribed species from other areas, suggests a world¬
wide distribution of Loricifera [5] inhabiting a wide
range of depths and sediment types within the
marine environment.

MATERIALS AND METHODS

The present specimens were found from a box
core sample collected at ST. 9 of a cruise, KH-80-
1, of R/V Hakuho-Maru, Ocean Research Insti¬
tute, University of Tokyo (ORIUT). This station
was established close to the axis of the Izu-
Ogasawara Trench at a depth of 8260 m [6]. This
hadal-depth station was deeper than the calcite
compensation depth, and the sediment consisted
of pelagic “red clay” interbedded with silt and sand
laminae of turbidite origin.

The sediment sample was fixed and preserved
using 5% seawater formalin with Rose Bengal on
board ship. In the laboratory on land, meiofaunal
organisms were sorted out from the sediment
under a dissecting binocular microscope.

Three loriciferan individuals were found from
the sample. These specimens were transferred to
glycerol and observed under a compound micro¬
scope. Fortunately these specimens consisted of a
female, a male and a large larva, and they were
revealed to be a new species of the genus Plicilori¬
cus recently described from the upper-bathyal

876

R. M. Kristensen and Y. Shirayama

zone (289-439 m) off the coast of North and South
Carolina, U.S.A. [3]. A full description of the
present species follows.

DESCRIPTION

Order Nanaloricida Kristensen, 1983
Family Pliciloricidae Higgins and Kristensen, 1986
Genus Pliciloricus Higgins and Kristensen, 1986

Pliciloricus hadalis nov. sp.

(Figs. 1-4)

Diagnosis'. Adults with a small mouth cone without
mouth tube. The first row of scalids consisting of
the eight clavoscalids with pod-like tips. The
fourth row consisting of 15 fleshy and claw-tipped
spinoscalids, alternating with 15 unmodified very
long spinoscalids. The neck with the last row of
scalids consisting of seven double and eight single
trichoscalids of the same length. The thorax with
sculpture close to the edge of the lorica. The
midventral plicae of the lorica differ only slightly
from other 20 plicae. The single pair of posterior
flosculi located very caudally on the ventral site.
The lorica jar-shaped and globular posteriorly.

The free-living larva (Higgins-larva) with a
mouth cone without ventral oral setae. The buccal
canal separated into an anterior part, the mouth
cavity and a complex hexagonal cuticular appa¬
ratus, which joins the pharyngeal bulb. Spinosca¬
lids long (^55 ^m). Collar well defined. Lorica
covered with mucoid detritus. Toes straight, spi-
nose and very long (153 ^m), and pointing ante¬
riorly. Three pairs of setae at the posterior end of
lorica. All setae without hair.

Etymology : hadalis from the Greek word Hades
(hell) used here as the Pliciloricus from the hadal
bathymetric zone of the deep sea.

Type material: In the Zoological Museum of
Copenhagen (ZMC), Denmark. The holotypic
adult female was collected on March 3,1980, at ST.
9 (28°28.3’N 143°19.6’E; 8260 m) of the cruise
KH-80-1 of R/V Hakuho-Maru, ORIUT [6]. The
animal was found in the layer of 1-2 cm from the

surface of the sediment consisting of red clay with
silt and sand laminae. The allotypic adult male and
the paratypic larva were found from the same
station (ST. 9) as the holotype, but were located in
the surface 0-1 cm layer of the sediment.

Description: The holotypic female (Figs, la and 2)
was originally 219 pm long, but by an accident, the
animal was drying out and partially destroyed.
The mouth cone, completely extended, is 23 pm
long and consists of three regions. The basal
portion has eight retractor muscles, the second
region has eight oral ridges, and the third region
consists of tubular extensions and the flexible
buccal canal, which can withdraw into the second
region of the mouth cone.

The head (introvert) has nine rows of appen¬
dages of typical Pliciloricus type. The first row
consists of eight uniform clavoscalids, identical in
both female and male. All eight clavoscalids have
three distinct segments in addition to a spinose tip.
The terminal segment is pod-shaped. The second
row of scalids consists of five long leg-like spino¬
scalids and two partially fused smaller scalids, the
double-organ [3], located in the midventral line.
The long leg-like spinoscalids are segmented,
while the double-organ is unsegmented and has a
thin cuticle. All seven spinoscalids have large stiff
hairs along the anterior margin. The third row of
scalids consists of 15 small leg-shaped appendages,
each with enlarged double bases and a tubercular
joint with two lateral cusps. The fourth row of
appendages consists of 15 long spine-shaped sca¬
lids alternating with 15 short and fleshy, claw-
tipped scalids. The fifth to seventh rows of appen¬
dages are uniform, and each row consists of 30
spinoscalids. Each of the 30 spinoscalids of the
eighth row possesses a basal plate that bears
serrate filiform spines. The ninth row consists of
30 short, robust and scale-like scalids.

The neck is marked by two constrictions, the
limits of both the head and the thorax are distinct.
This region has seven double and eight single
trichoscalids. Both of them have two basal plates,
but only the upper basal plate of the double
trichoscalid has a long spine. The double (100 pm
long) and the single (95 pm long) trichoscalids are
serrated with a central canal in which lies a single

New Loricifera from the Deep Sea

877

Fig. 1. Microscopic photographs of Pliciloricus hadalis sp. nov. a) Holotypic female. Ventral view. Note that the
lorica is reflecting the light, b) Paratypic Higgins-larva. Lateral view. Scale bar=50/mi.

878

R. M. Kristensen and Y. Shirayama

me

in

ne

th

- cs

- sr2

- do

- sr3

-sr4

-cl

“Sr9
“ Sr5

-sr6

“ Sr7

“ tP
-ed

sr8

lo

tr2

tn

Pi

fl

Fig. 2. Holotypic female of Pliciloricus hadalis sp. nov. collected in the Izu-Ogasawara Trench at a depth of 8260 m.
Ventral view; cl = claw-tipped spinoscalids (sr4); cs = clavoscalids (srx); do = double organ; ed = edge of lorica; fl =
flosculus; in=introvert (head); lo = lorica; lr = double ridge of lorica; me = mouth cone; ne = neck; pl = plica; s^-
sr9=scalid rows 1-9; th = thorax; tp = thoracical plate; trx = primary trichoscalid; tr2= secondary trichoscalid.

cilium. Each of the double trichoscalids has a
single base from which extend primary and sec¬
ondary trichoscalids of nearly the same size.

The thorax has no appendages. The anterior
part is unsculptured, with a very thin cuticle. The

posterior part of the thorax is sculptured. The
sculptured part consists of 15 small and 15 large
plates. The large plate has a cuticular ridge. The
thorax is well separated from the lorica. The lorica
is a typical Pliciloricus type consisting with 22

New Loricifera from the Deep Sea

879

double folds (plicae). The cuticle lacks the honey¬
comb structure found in many loriciferans. The
anterior margin of the lorica has two arch-like

structures associated with each of 22 plicae. The
midventral plica differs only slightly from the other
plicae. A single pair of posterior flosculi is present

OS

id

ph

cs

sr2

Sr3

Sr4

Sr5

Sr6

sr7

to

ISl

1*2

se !
se2

se3

Fig. 3. Paratypic larva of Pliciloricus hadalis sp. nov. Ventral view. cs=clavoscalid (srx); co = collar; ia = internal
armature of larval buccal canal; in = introvert; lo = lorica; lsx = anterolateral seta; ls2 = anteroventral seta; mc =
mouth cone; os = oral stylet; ph= pharyngeal bulb; sei = posterodorsal seta; se2 = posterolateral seta; se3 =
posteroterminal seta; sr1-sr7=scalid rows 1-7; th = thorax; to = toe.

880

R. M. Kristensen and Y. Shirayama

in the ventral part of two other plicae. The shape
of the lorica is atypical for the genus Pliciloricus
that the lorica consists of a single piece and is
shaped like a jar. The internal anatomy was
destroyed in the holotypic female, but a single
large egg has originally filled the abdomen. The
gonopores are located caudo-dorsally close to the
dorsal anus.

The allotypic male has retracted maximally and
therefore is only 145 pm long. The mouth cone,
the head, the neck and the unsculptured part of the
thorax are located inside of the lorica. All appen¬
dages are oriented anteriorly and extruded from
the anterior margin of the lorica in the opposite
way to the fully extended female (first the tricho-
scalids, second the spinoscalids and third the clavo-
scalids). The internal anatomy is preserved in an
excellent condition. Two large dorsal testes, each
with a caudal seminal vesicle, contained filiform
sperms. A pair of protonephridia is located close
to the genital system. Both the anus and the two
lateral gonopores are located caudo-dorsally.

The single paratypic larva is a typical Higgins-
larva [2] of the Pliciloricus type (Figs, lb and 3). It
is 262 pm long with spinose toes (153 pm long).
The mouth cone is tripartite, and the two terminal
portions were slightly retracted telescopically. The
head, the neck, the thorax and the abdomen
extended completely. The mouth opening is sur¬
rounded by six cuticular valves. The internal
mouth armature is hexaradially symmetrical with
six stylet-like structures. The buccal canal consists
of a heavily sclerotized mouth in front of the long
oval pharyngeal bulb. The pharyngeal bulb is
glandular and only a few myoepithelial cells are
seen. The first row of appendages consists of the
eight uniform clavoscalids, each of which consists
of two laterally compressed basal segments and
two spinose terminal segments. The spinoscalids
of the second to the fifth rows consist of two
segmented, long and robust scalids. The sixth row
of head appendages consists of eight double scales,
called protoscalids. The seventh row of appen¬
dages consists of seven protoscalids, forming a
W-shaped structure, alternating with eight single
papillae each located on a basal plate. The closing
apparatus, or the collar [3], is formed by the neck
when the head is retracted. It is regularly folded,

Fig. 4. Paratypic larva of Pliciloricus hadalis sp. nov.

Lateral view, an = anus; de = detritus.

but lacks collar pores. The thorax has five rows of
15 to 20 plates, forming an accordion-like system.

The larval lorica has a thick unsculptured cuticle
with 20 ridges. The fine structure of the lorica is
hidden by a mucous layer, on which detrital mate¬
rial attached (Fig. 4). The abdomen is very wide
(87 pm) and balloon-shaped. The very long toes
extend from the lateroventral region of the caudal
end and is oriented anteriorly. About 40% of the
proximal length of the toes is hollow, while the
remaining terminal 60% narrows to a solid spine.
The two pairs of anterior lorical setae are branched
near the tip. The three pairs of posterior lorical
setae are simple. The caudal pair of lorical setae
(inserting near the dorsal anus) are hollow and
perhaps glandular. Flosculi are absent.

New Loricifera from the Deep Sea

881

DISCUSSION

Only five species, Pliciloricus orphanus, P. pro¬
fundus, P. enigmaticus, P. gracilis and P. dubius,
have been described previously for the genus. The
present species, P. hadalis , is closely related to P.
orphanus and P. profundus, on the basis of its
larval characteristics, and to a lesser extent to P.
enigmaticus and P. gracilis. The larva of P. dubius
is unknown. The adult of P. hadalis closely
resembles P. profundus in the presence of single
pair of posterior flosculi, the type of mouth cone
and clavoscalids and the claw-shaped spinoscalids.
Unique to P. hadalis is the shape of the lorica of
both adults and larva, the mucous coat on the
larval lorica and the 30 sculptured plates in the
adult thorax.

The habitat of Pliciloricus hadalis is different
completely from the other congeneric species. All
described species so far have been found exclusive¬
ly from sublittoral coarse sediments off the south¬
eastern coast of the U.S.A., e.g. coarse quartz
sand of 400-439 m, coarse oolytic sand of 294 m
and medium black phosphorite sand of 289 m deep
[3]. On the other hand, P. hadalis was found from
very fine red clay at 8260 m in the Western Pacific
Ocean. This finding reveals that the genus Plicilor¬
icus thus far is the most eurybathic meiofaunal
genus. Moreover, in contrast to the interstitial
nature of the other species, P. hadalis most prob¬
ably burrows within the sediment, because it was
collected not only from the surface but also from
the subsurface layer of the sediment. Such wide
variety of habitats as well as modes of life suggest
that Pliciloricus has the ability to adapt to various
environmental conditions without marked mod¬
ification in morphology. In addition to P. hadalis,
we have three additional undescribed deep-sea
loriciferan species and even an undescribed family

from the Shatzky Rise, central Pacific at depths of
2430 and 3160 m, and still more species are being
discovered from various areas of the world ocean.
These data suggest a ubiquitous distribution of
Loricifera not only in the coastal waters but also in
the deep sea. Considering that loriciferans are
now known to comprise many species and genera,
and are distributed widely in various types of
marine sediments, it is astonishing that this group
of animals was not described until 1983. The
reason probably lies in the labor necessary for the
intensive process of meiofaunal research, the mag¬
nitude of required patience, and luck.

ACKNOWLEDGMENT

We thank Captain Tadama and other officers and crew
of R/V Hakuho-Maru for their help in sampling, and our
colleague, Dr. R. P. Higgins, National Museum of Natu¬
ral History, Smithsonian Institution, U.S.A. for his
critical reading of the manuscript.

REFERENCES

1 Mare, M. F. (1942) A study of a marine benthic
community with special reference to the microorgan¬
isms. J. Mar. Biol. Ass. U. K., 25: 517-554.

2 Kristensen, R. M. (1983) Loricifera, a new phylum
with aschelminthes characters from the meiobenthos.
Z. Zool. Syst. Evol.-forsch., 21: 163-180.

3 Higgins, R. P. and Kristensen, R. M. (1986) New
Loricifera from Southeastern United States coastal
waters. Smithson. Contrib. Zool., No. 438: 1-70.

4 Bruun, A. F. (1957) Deep sea and abyssal depths. In
“Treatise on Marine Ecology and Paleoecology, Vol.
1, Ecology”. Ed. by J. W. Hedgpeth, The Geological
Society of America, Memoir 67, pp. 641-672.

5 Kristensen, R. M. (1986) Loricifera. In “Stygofauna
Mundi”. Ed. by L. Botosaneanu, E. J. Brill, Leiden,
pp. 119-121.

6 Shirayama, Y. (1984) The abundance of deep-sea
meiobenthos in the Western Pacific in relation to
environmental factors. Oceanol. Acta, 7: 113-121.

ZOOLOGICAL SCIENCE 5: 883-892 (1988)

© 1988 Zoological Society of Japan

Mites of the Genus Myobia (Trombidiformes, Myobiidae) Parasitic
on Apodemus mice in Korea and Japan, with Reference
to their Immature Stages* 1

Kimito Uchikawa, Keisuke Nakata2 and Fritz S. Lukoschus3

Department of Parasitology, Shinshu University School of Medicine, Matsumoto 390,
2Hokkaido Forest Experiment Station, Bibai, Hokkaido 079-01, Japan, and
3 Department of Aquatic Ecology, Catholic University, Nijmegen, The Netherlands

ABSTRACT — Myobia nodae, Myobia apodemi, Myobia agraria and Myobia kobayashii were taken
from Apodemus speciosus sspp., Apodemus argenteus, Apodemus agrarius (Korea) and Apodemus
peninsulae sspp. (Korea and Hokkaido), respectively, suggesting that each mite is so strictly monoxenic
or specific to a host species that it may serve as an indicator in the classification of Apodemus in Korea as
well as in Japan. Immature stages of the 4 Myobia species were depicted and measured. It was
demonstrated that these mites were identified almost correctly in the proto- and deuto-nymphal stages
and precisely in the tritonymphal stage, making mites of the genus Myobia more reliable indicators in
the host classification.

INTRODUCTION

The genus Myobia von Heyden, 1826, is associ¬
ated exclusively with the rodent family Muridae.
Myobia musculi (Schrank) infesting Mus musculus
was described as early as in 1781, and its morphol¬
ogy has been observed in detail since that time. In
Japan, a total of 5 species within the genus Myobia
has been recorded as monoxenic mites occurring
on mice of the genera Mus, Micromys and
Apodemus [1], suggesting that these mites may
serve as labels in classifying host mice. Myobia
kobayashii Uchikawa et Mizushima, 1975, is the
fifth species found on Apodemus giliacus ( =A .
peninsulae ) subsequent to the finding of this mouse
in Hokkaido by Kobayashi and Hayata in 1971 [2].
If myobiids of Apodemus mice had thoroughly
investigated much earlier, the occurrence of A.
giliacus there could have suggested from acaro-
logical data.

Mammalogists’ interest currently been con¬
cerned with systematics of Apodemus mice dis¬

Accepted November 9, 1987
Received October 13, 1987

1 We began this study together, but Dr. FritzS. Lukos¬

chus passed away before reading the manuscript of the

present paper.

tributed in Korean Peninsula and mainland China
[3,4], To cope with this trend, a double-purposed
study of mites of the genus Myobia parasitic on
Apodemus in Korea and Japan was carried out.
First, A. agrarius and A. peninsulae from Korea
and A. speciosus tusimaensis from Tsushima Is¬
land, Japan, located between Japan and Korea
were examined for myobiids specific to them.
Second, immature stages of all Myobia species
parasitic on Apodemus mice from Japan and
Korea were compared with one another to check
the usefulness of them as indicators in host clas¬
sification. Incidentally, those stages usually occur
more abundantly on hosts than adults, and are not
exactly studied yet morphologically.

MATERIALS AND METHODS

Skin specimens of A. peninsulae , Hang-ge-ryon,
Kanwangdo, Korea, 1981-VII-9, alcoholic speci¬
mens of A. agrarius, Monehi, Korea, 1951-VIII-
27, and A. speciosus tusimaensis, Hisada, Tsushi¬
ma Island, Japan, 1957- VI-13, all of which are
deposited in the collection of the National Science
Museum, Tokyo, were examined for myobiids
under the dissecting microscope at a magnification
XlO, combing hairs with the forceps. Mites were

884

K. Uchikawa, K. Nakata and F. S. Lukoschus

also taken directly as above from alcoholic speci¬
mens of A. speciosus speciosus from Matsumoto,
Honshu, Japan, 1985-V-21, A. speciosus ainu
from Sapporo, Hokkaido, Japan, 1983-IX-28, A.
argenteus from Kamikochi, Honshu, 1983-X-18,
and A. peninsulae giliacus from Oiwake-cho,
Yufutsu-gun, Hokkaido, 1985-X-8.

Mites were mounted on slides in the modified
Hoyer’s (= Andre’s) solution according to the
routine procedures, and examined microscopi¬
cally.

RESULTS

Host-parasite relationships

The host Apodemus mice examined and Myobia
mites taken from them are shown in Table 1. A.
agrarius from Korea was proved to be the host of
Myobia agraria Gorissen et Lukoschus, which was
originally found on A. agrarius from Poland and
Germany [5]. A. peninsulae from Korea harbored
Myobia kobayashii. A. speciosus tusimaensis
yielded Myobia nodae Matuzaki. Both A. spe¬
ciosus speciosus from Matsumoto and A. speciosus
ainu from Sapporo were infested with M. nodae.
A. argenteus from Kamikochi and A. peninsulae
giliacus from Hokkaido harbored Myobia apodemi
Uchikawa and M. kobayashii , respectively.

The host-relation of the Myobia species parasitic
on mice of the genus Apodemus in Korea and
Japan is confirmed to be strictly monoxenic as far
as the species level of hosts is concerned.

Immature stages

Larval and nymphal stages of the 4 Myobia
species are depicted in Figures 1-25; the leg
chaetotaxy is shown in Table 2; and measurements
in pm are given in Table 3.

Leg I is 3-segmented in the larval and 3 nymphal
stages as well as in the adult. Other legs consist of
the 4 segments, trochanter, femuro-genu, tibia and
tarsus throughout all immature stages, whereas
legs II-IV are 5-segmented in the adult.

The larva is hexapodal, with leg III being much
shorter than leg II and lacking terminal claw. Two
pairs of propodosomal setae, vi, ve, sc i and sc e,
are distinctly inferior in size to 6 pairs of hyster-

osomal setae, dx_3 and lU3 (Figs. 1, 10, 15 and 21,
Table 2). Only a single pair of simple setae are
present on coxal regions I (Fig. 7).

The protonymph is also hexapodal, with leg III
bearing a terminal claw. A pair of setae each are
added dorsally ( d4 ) and ventrally (modified ones)
to the idiosomal chaetotaxy of the larva (Figs. 2
and 8). Leg chaetotaxy is the same as that of the
larva (Table 2).

The deutonymph is octopodal, but legs IV are
still primordial. Idiosomal setations on the dorsum
and venter are quite the same as those of the
protonymph. Setae on the idiosomal dorsum ex¬
hibit great variations in shape, length and position
individually and bi-laterally (Figs. 3 and 4, 12 and
13, 17 and 18, 22 and 23; Table 3). Those varia¬
tions are too diverse and irregular to ascribe them
to the sexual dimorphism. Femuro-genua II and
III bear 3 and 1 seta, respectively, instead of 2 and
0 on the corresponding segments of the pro¬
tonymph (Table 2).

The tritonymph is octopodal, with leg IV being
fully developed. Although setation on the idio¬
somal dorsum is constant throughout all nymphal
stages, some setae are more developed in the
tritonymph than in the preceding nymphal stages.
Three pairs of setae are present on coxal regions I
(Fig. 9) instead of 2 pairs of setae in the proto- and
deutonymphs (Table 2). Trochanters II and III as
well as tarsi II and III bear one more seta than
corresponding segments of the deutonymph (Table
2). In this stage, there are larger-sized individuals
with the elongate opisthosoma and smaller-sized
ones with the rather contracted opisthosoma (Figs.
5 and 6, 19 and 20, 24 and 25; Table 3). This
difference in idiosomal size and form may be
ascribable to the sexual dimorphism. Idiosomal
size is generally larger in advanced stages than in
preceding stages (Table 3), yet it is not possible to
separate 2 successive stages precisely from each
other according to difference in size alone.

In the immature stages of the 4 Myobia species,
no specific characters were found in all legs and
chaetotaxy of legs and idiosomal venter. Only
chaetotaxy on the idiosomal dorsum seems to be
characteristic of each species as shown in Figures
1-26. It is evident that M. nodae and M. apodemi
are closer in general appearance that is largely

Myobia Infesting Apodemus in Korea and Japan

885

formed by chaetotaxis of the idiosomal dorsum to
each other than to M. agraria and M. kobayashii,
which closely resemble to each other. According¬
ly, the differentiation of the paired species is of
practical importance to identify the 4 Myobia
species in all immature stages. Instead of describ¬
ing immature stages of each species in detail, a key
to the 4 species and developmental stages is made
as follows:

1. Hexapodal. Femuro-genu II bearing 2 setae;

femuro-genu III lacking seta . 2.

Octopodal. Femuro-genua II and III bearing
3 and 1 seta, respectively . 7.

2. Third leg without terminal claw. Hyster-
osomal setae consisting of 6 pairs; d4 lacking.
No seta ventrally on gnathosoma. A pair of
setae ventrally on coxal regions I. ... Larva
. 3.

Third leg with a terminal claw. Hysterosomal
setae consisting of 7 pairs; d4 present. A pair
of seta ventrally on gnathosoma. Two pairs
of setae ventrally on coxal regions I. ... Pro¬

tonymph . 4.

3. Larva

Third leg primordial . apodemi *.

Third leg well developed. ...agraria, koba¬
yashii, nodae.

4. Protonymph

Distance between l3 subequal to that between

h . 5.

Distance between /3 distinctly larger than that
between l2 . 6.

5. Seta d4 longer than 7 pm . agraria*.

Seta d4 shorter than 5 pm . kobayashii* .

6. No setae dorsally on idiosoma remarkably

long . nodae*.

Seta /3 distinctly longer than other setae.

. apodemi, nodae.

7. Fourth leg primordial or not fully developed.
Two pairs of setae ventrally on coxal regions
I. Trochanters II— III bearing no seta; tarsus

III with 5 setae. ... Deutonymph . 8.

Fourth leg well developed. Three pairs of
setae ventrally on coxal regions I. Trochan¬
ters II-IV bearing 1 seta; tarsus III with 6
setae. ... Tritonymph . 11.

8. Deutonymph

Distance between /3 subequal to that between

l2, /3 situated anterad to d3 . 9.

Distance between /3 distinctly larger than that
between /2; /3 situated almost on basal level of
d3 . 10.

9. Seta vi shorter than 15 pm. Posterior subme¬
dian setae d2, l2, /3, d3 and d4, inferior in size
to other 6 pairs of setae (Figs. 17 and 18).

. agraria.

Seta vi longer than 15 pm. Propodosal setae,
d\ and /i well developed; d2, l2 and /3 superior
in size to d3 and d4 (Fig. 22), with a few

exceptions (Fig. 23) . kobayashii.

10. Seta vi shorter than 13 pm. Forth leg primor-
dium small and thumb-like (Figs. 3 and 4).
Setae sc e and /3 long but l\ short on some
individuals (Fig. 3), while all setae rather

short on others (Fig. 4) . nodae*.

Seta vi longer than 18 pm. Forth leg primor-
dium large, triangular, without segmentation
(Figs. 12 and 13). Setae sc e, lx and /3 long on
a larger portion of individuals (Fig. 12).

. apodemi* .

11. Tritonymph

Seta /3 situated antero-extriad of d3. No dor¬
sal setae remakably long . 12.

Seta /3 situated laterad of d3. Setae sc e, l\
and /3 remarkably longer than other setae.
. 13.

12. All setae dorsally on idiosoma shorter than
30 //m(Table 3); submedian setae vi, dx, d2,
l2, /3, d3 and d4 inferior in size to other setae

(Figs. 19 and 20) . agraria*.

Some setae dorsally on idiosoma longer than
40 pm (Table 3); setae d3 and d4 distinctly
inferior in size to other setae (Figs. 24 and
25) . kobayashii* .

13. Seta vi shorter than 17 pm (Table 3).

. nodae*.

Seta vi longer than 20 pm (Table 3).
. apodemi* .

The asterisked species in the key are dif¬
ferentiated precisely from other species in
each developmental stage. It is thus difficult
to identify all species correctly in the larval
stage. Only M. apodemi is separable from
the other 3 species.

In the protonymphal stage, variation in

886

K. Uchikawa, K. Nakata and F. S. Lukoschus

Figs. 1-9. Idiosoma and legs I of Myobia nodae Matuzaki. 1: Dorsum of larva. 2: Dorsum of protonymph. 3:
Dorsum of deutonymph, A type. 4: Dorsum of deutonymph, B type. 5: Dorsum of male tritonymph. 6: Dorsum
of female tritonymph. 7: Venter of larva. 8: Venter of proto- and deutonymphs. 9: Venter of tritonymph. Bar:
50 fim.

Myobia Infesting Apodemus in Korea and Japan

887

Figs. 10-14. Dorsal view of idiosoma of Myobia apodemi Uchikawa. 10: Larva. 11: Protonymph. 12:

Deutonymph, A type. 13: Deutonymph, B type. 14: Male tritonymph. Bar: 50 pm.

Figs. 15-20. Dorsal view of idiosoma of Myobia agraria Gorissen and Lukoschus. 15: Larva. 16: Protonymph. 17:

Deutonymph, A type. 18: Deutonymph, B type. 19: Male tritonymph. 20: Female tritonymph. Bar: 50 pm.
Figs. 21-26. Dorsal view of idiosoma of Myobia kobayashii Uchikawa and Mizushima. 21: Larva. 22: Protonymph.
23: Deutonymph, A type. 24: Deutonymph, B type. 25: Male tritonymph. 26: Female tritonymph. Bar: 50 pm.

888

K. Uchikawa, K. Nakata and F. S. Lukoschus

Myobia Infesting Apodemus in Korea and Japan 889

Table 2. Leg chaetotaxy of immature and adult stages of mites of the genus Myobia

Legs and segments

L

PN

DN

TN

A

trochanter

_

_

_

_

3

I

femuro-genu

6

6

6

6

5

tibio-tarsus

7

7

7

7

15

trochanter

_

_

_

1

3

femuro-genu

2

2

3

3

5 femur

II

7 genu

tibia

4

4

4

4

6

tarsus

6

6

6

6

6

trochanter

—

—

_

1

3

femuro-genu

—

—

1

1

3 femur

III

6 genu

tibia

3

3

3

3

6

tarsus

5

5

5

6

6

trochanter

1

3

femuro-genu

1

3 femur

IV

tibia

3

5 genu

6

tarsus

5

6

L, PN, DN and TN as in Table 1. A: Adult.

length of seta /3 of M. nodae makes it im¬
possible to separate this species from M.
apodemi completely, while M. agraria and
M. kobayashii are separable from each other.
A total of 3 individuals of M. nodae that
beared long /3 was found in the populations
from Matsumoto and Sapporo (Table 1).
Although Matuzaki [6] depicted long /3 ( = 6b
in Matuzaki) of the protonymph ( = 2nd larva
in Matuzaki) in the original description of M.
nodae , she measured 3 examples of /3 as 35.0,
13.8 and 16.3 pm, indicating the presence of
both long and short /3. It is thus reasonable
to presume that both types of protonymphal
M. nodae, with or without long /3, occur
concomitantly in a given population, and that
the protonymphs of this species may easily be
identified on the basis of individuals with
short /3.

The deutonymphal stage is characteristic of
its vast variation in arrangement and length
of setae on the idiosomal dorsum. Although
setal arrangement is almost specific to each
mite species, the key character to separate 2
species alike to each other is the length of
seta vi. M. agraria and M. kobayashii are not
completely separable from each other be¬
cause of the overlap of ranges of the setal
lengths at both ends, yet several individuals

in a population probably indicate which spe¬
cies they belong to.

The tritonymphal stage does not exhibit
remarkable variation in the chaetotaxy of the
idiosomal dorsum. All the 4 species are
identified precisely in this stage as shown in
the above key.

DISCUSSION

Mice of the genus Apodemus distributed in
Korea and Japan have yielded a total of 4 species
of myobiids of the genus Myobia. Since A. penin-
sulae and A. giliacus have been thought to be
conspecific [3, 4] as was dealt with above, a
Myobia mite is confirmed to be monoxenic or
specific to a Apodemus mouse in Korea and Japan.
Some mammalogists are inclined to separate many
subspecies within a Apodemus species [3, 7]. And
recent karyological analyses have shown the
occurrence of polymorphism in the karyotype of a
given Apodemus mouse [4]. Mites of the genus
Myobia parasitic on different subspecies of
Apodemus species belong, however, to a single
species (Table 1). Moreover, all the available data
indicate that M. nodae is the sole myobiid parasitic
on A. speciosus sspp. from various parts of Japan
from Hokkaido to Tokara Islands [8], suggesting
that the karyological difference within a host spe-

890

K. Uchikawa, K. Nakata and F. S. Lukoschus

Table 3. Measurements in /mi for immature stages of 4 species of the genus Myobia parasitic
on Apodemus mice in Korea and Japan

nodae

apodemi

agraria

kobayashii

Larva(n)

(5)

(20)

(9)

(10)

BL

175-245

143-195

175-235

165-220

BW

120-170

93-133

100-145

100-145

vi

8- 10

5- 8

7- 10

5- 8

12- 16

11- 16

11- 13

10- 13

Protonymph(n)

(10)

(10)

(10)

(10)

BL

203-260

170-230

208-280

193-258

BW

153-190

130-163

128-160

130-165

vi

10- 13

8- 14

7- 10

7- 11

di

7- 13

5- 11

5- 8

5- 8

d4

8- 13

7- 10

7- 8

3- 5*

h

5- 10

6- 17

3- 5

5- 6

h

12- 17
(33-35)f

27- 43*

5- 8

6- 10

Deutonymph

A-type(n)

(10)

(10)

(10)

(10)

BL

220-270

173-240

230-270

220-290

BW

170-210

153-185

153-190

160-190

vi

10- 13

18- 35*

7- 15

15- 23*

sc e

55- 80

70- 90

9- 18

18- 30*

dj

13- 18

7- 23

8- 15

15- 26*

d2

8- 17

5- 20

6- 10

7- 17

h

8- 16

65- 85*

10- 15

18- 38*

h

18- 55

45- 55

7- 9

10- 15*

B-type(n)

(10)

(1)

(4)

(8)

BL

237-320

220

240-290

200-305

BW

175-235

185

170-190

165-200

vi

12- 13

23*

8- 10

15- 20*

sc e

7- 10

16

12- 20

5- 11*

di

4- 11

8

8- 10

7- 15

d2

5- 10

10

7- 8

6- 8

h

4- 10

9

8- 11

4- 15

h

12- 20

23

7- 7

8- 10

Tritonymph
(Small) (n)

(10)

(10)

(10)

(10)

BL

222-355

220-300

220-280

270-340

BW

165-255

165-213

150-180

180-240

vi

12- 17

20- 25*

10- 14

20- 26*

sc e

73-103

88-108

20- 28

25- 40

d ,

6- 21

14- 20

12- 17

22- 30*

d2

16- 20

17- 22

10- 16

18- 28*

h

78- 95

85-103

16- 23

30- 55*

l2

16- 22

17- 23

10- 15

20- 28*

h

53- 68

55- 70

8- 12

16- 28*

Myobia Infesting Apodemus in Korea and Japan

891

Table 3. (Continued)

nodae

apodemi

agraria

kobayashii

(Large) (n)

(10)

(8)

(10)

(10)

BL

410-465

325-390

310-460

400-440

BW

280-305

220-235

200-240

230-280

VI

14- 17

22- 30*

10- 16

22- 26*

sc e

80-105

95-110

18- 27

28- 38*

dj

15- 20

13- 23

11- 17

23- 28*

d2

15- 20

13- 23

10- 15

22- 26*

h

78-100

82-100

13- 25

26- 50*

h

15- 20

16- 27

10- 18

18- 28*

h

25- 70

58- 70

7- 12

19- 27*

BL: Body( = gnathosoma + idiosoma) length, BW: Body width.

* A key character for separating related species from each other,
f Setal length for a total of 3 individuals with long l3 .

cies is so far indifferent to the speciation of the
Myobia mite. Thus mites of the genus Myobia may
serve as indicators or labels in sorting out host
mice at a species ranking. As was demonstrated in
the present study, the mites are identified almost
precisely in the proto- and deuto-nymphal stages
and precisely in the tritonymphal stage. The
nymphal stages are usually found even on dead
hosts preserved for a long time, still attaching
firmly on host skin with their oral organs. Accord¬
ingly transfer of live mites to strange hosts and
post-morten transfer of mites hardly oocur in those
immature stages. Moreover, the tritonymph seems
to be the most dominant stage in a give population
of mites of the genus Myobia (Table 1). These
facts make mites of the genus Myobia more reli¬
able indicators in the host classification.

Intrageneric relationships among component
myobiid mites are sometimes deducible from the
structure of male genital shield, including genital
setae [9-11]. In the genus Myobia , the male
genital shield and genital setae are, however, not
so variable as to deduce such information. The
present study indicates that M. nodae and M.
apodemi are closer in nymphal stages to each other
than to M. agraria and M. kobayashii, which are
alike to each other. Our comparative study of
immature M. musculi as well as the detailed study
by Grant [12] demonstrates that M. musculi resem¬
bles more closely M. agraria and M. kobayashii
than M. nodae and M. apodemi , although M.

musculi is easily separable from the other 4 species
in the immature stages, including the larval stage.
It is necessary to analyse phylogenetic significance
of similarity or dissimilarity in general appearance
throughout whole developmental stages of all
mites of the gunus Myobia.

ACKNOWLEDGMENTS

We are indebted to Dr. Mizuko Yoshiyuki, Depart¬
ment of Zoology, National Science Museum, Tokyo, for
allowing KU to examine Apodemus specimens from
Korea and Tsushima Island deposited in the museum
collection. This study was supported financially in part
by a Grant-in-Aid for Scientific Research to KU
(No. 57480392) from the Ministry of Education, Science
and Culture, Japan.

REFERENCES

1 Uchikawa, K. (1977) Notes on the myobiid mites
parasitic on Insectivora and Rodentia in Japan (In
Japanese). In “Contribution to Acarology in
Japan”. Ed. by M. Sasa and J. Aoki, Zukan-no-
Hokuryukan, Tokyo, pp. 415-432.

2 Kobayashi, T. and Hayata, I. (1971) Revision of the
genus Apodemus in Hokkaido. Annot. Zool.
Japon., 44: 236-240.

3 Xia, W. (1985) A study on Chinese Apodemus and
its relation to Japanese species. In “Contemporary
Mammalogy in China and Japan”. Ed. by T. Kawa-
michi, Mamm. Soc. Jpn., pp. 76-79.

4 Kobayashi, T. (1985) Taxonomic problems in the
genus Apodemus and its allies. In “Contemporary
Mammalogy in China and Japan”. Ed. by T. Kawa-

892

K. Uchikawa, K. Nakata and F. S. Lukoschus

michi, Mamm. Soc. Jpn., pp. 80-82.

5 Gorissen, M. M. and Lukoschus, F. S. (1982)
Myobia ( Myobia ) agraria sp. n. (Acarina: Prostig¬
mata: Myobiidae) from Apodemus agrarius (Roden-
tia: Muridae: Murinae) with a key to the known
species. Ann. Zool. (Warszawa), 36: 567-575.

6 Matuzaki, S. (1965) A new mite of genus Myobia
(Acarina: Myobiidae) from small mammals in
Japan. Bull. Kochi Women’s Univ., 13 (Ser. Nat.
Sci.): 1-10.

7 Imaizumi, Y. (1960) Coloured Illustrations of the
Mammals of Japan. Hoikusha, Osaka, pp. 141-147.

8 Uchikawa, K. and Suzuki, H. (1984) Medico-
zoological studies in Tokara Archipelago, Kagoshi¬
ma Prefecture, Japan — Records of acari and flea
found on an Apodemus mouse on Nakanoshima

Island comprising the archipelago. Trop. Med., 26:
17-25.

9 Uchikawa, K. and Harada, M. (1981) Evaluation of
bat-infesting Myobiidae (Acarina, Trombidiformes)
as indicators in taxonomy and phylogeny of host bats
(Chiroptera). Zool. Mag., 90: 351-361.

10 Uchikawa, K. (1985) Calcarmyobia from the Ethio¬
pian region (Acarina, Myobiidae). Bull. Br. Mus.
nat. Hist. (Zool.), 48: 45-55.

11 Uchikawa, K. (1986) Mites of the genus Binuncus
Radford (Trombidiformes, Myobiidae) and in¬
formation on host taxonomy deduced from them. J.
Parasit., 72: 257-270.

12 Grant, C. D. (1942) Observations on Myobia mus-
culi (Schrank) (Arachnida: Acarina: Cheyletidae).
Microent., 7: 64-76.

ZOOLOGICAL SCIENCE 5: 893-895 (1988)

© 1988 Zoological Society of Japan

[COMMUNICATION]

Phosphocalcic Response to Vitamin D3 Treatment in
Freshwater Snake, Natrix piscator

Ajai K. Srivastav and Ladli Rani

Department of Zoology, University of Gorakhpur,
Gorakhpur-237 009, India

ABSTRACT — Vitamin D3 (20 IU/100 g bw) was admin¬
istered to the freshwater snake, Natrix piscator for 15
days. This treatment evokes hypercalcemia from day 3 to
day 5. This response declines at day 10 and day 15.
Vitamin D3 also induces hyperphosphatemia from day 5
till the end of the experiment (day 15).

INTRODUCTION

Little information is available on the role of
Vitamin D and its metabolites in the lower
vertebrates. There exist a few reports regarding
the effects of vitamin D3 and its metabolites in
teleosts [1-5] and amphibians [6-8]. However, in
reptiles there is a single report [9] of vitamin
D3-induced hypercalcemia. There seems to be no
study concerning the effect of vitamin D3 on the
serum inorganic phosphate concentration of rep¬
tiles. Hence, the present study was designed to
investigate such an effect in freshwater snake,
Natrix piscator.

MATERIALS AND METHODS

Adult specimens (both sexes) of N. piscator (bw
90-130 g) were collected locally and acclimatized
to the laboratory conditions (ambient temperature
and light ranged between 22-25°C and 10: 20-
10: 50 hr, respectively) for one week prior to use.
An initial sampling of blood (from 6 specimens)
was taken from the non-treated (normal) speci¬
mens before the start of the experiment. Then the

Accepted November 24, 1987
Received September 21, 1987

remaining animals were divided into two groups
(A and B) of 30 specimens (20 males and 10
females) each.

Snakes from the both groups were given daily
intraperitoneal injections of the following treat¬
ments for 15 days: Group A: 0.1 ml/100 g bw of
vehicle (95% ethanol) and Group B: 20 I.U. of
vitamin D3/100 g bw (dissolved in 95% ethanol).
Vitamin D3 was provided by Prof. A. W. Norman
(Univ. of Calif.). In all cases, the injection volume
was 0.1 ml/100 g bw. The injections were per¬
formed at 10: 00 each day.

Six specimens (4 males and 2 females) from each
group were anesthetized with ether 4 hr after the
last injection on the 1st, 3rd, 5th, 10th and 15th
day of the experiment. Blood samples were
collected by caudal amputation. After clotting of
the blood the sera were separated by centrifuga¬
tion (3500 rpm in a refrigerated centrifuge Janetzki
K 24) and analysed for serum calcium and serum
inorganic phosphate according to Trinder’s [10]
and Fiske and Subbarow’s [11] methods, respec¬
tively.

The animals were not fed during experiment.
Differences in the serum calcium and inorganic
phosphate levels among different groups were
analysed by Student’s r-test.

RESULTS

In non-treated specimens the values of serum
calcium and serum phosphate are 10.42 + 0.28 and
4.38 + 0.18 mg/100 ml, respectively.

Serum calcium level of N. piscator remains

894

A. K. Srivastav and L. Rani

Table 1. Effect of vitamin D3 on serum calcium and phosphate in freshwater snake, Natrix
piscator

Days

Serum calcium
(mg/100 ml)

Serum phosphate
(mg/100 ml)

Group A

Group B

Group A

Group B

1

10.50 + 0.18

10.95 + 0.32

4.31 + 0.11

4.47 + 0.20

3

10.63 + 0.27

11.66 + 0.293

4.37 + 0.13

4.86 + 0.18

5

10.58 + 0.26

12.38±0.29d

4.27 + 0.16

4.92 + 0. 12b

10

10.45 + 0.28

12.08±0.25c

4.42 + 0.12

5.05±0.14b

15

10.63 + 0.16

11.12 + 0.28

4.32 + 0.20

5.18±0.18b

The values are mean ± S.E. of 6 determinations.

a, b, c, and d, indicate significant responses: P<0.05, P<0.01, P< 0.005 and P< 0.001,
respectively.

unaffected after day 1 following vitamin D3 treat¬
ment. After day 3, the value increases significant¬
ly. This increase continues up to day 5. Thereaft¬
er, a decline is noticed in the calcium level on day
10 and day 15 (Table 1 and Fig. 1). However, the

DAYS

Fig. 1. Serum calcium and phosphate in freshwater
snakes ( N . Piscator ) at various time intervals after
vitamin D3 treatment. The values are mean ± S.E.
a, b, c, and d indicate significant responses, P<0.05,
<0.01, <0.005 and <0.001, respectivety.

calcium value at day 10 is still significantly higher
than the control level.

Serum inorganic phosphate levels of vitamin
D3-treated specimens were not significantly devi¬
ated from the controls on day 3. From day 5
onwards, increased levels of phosphate are noticed
(Table 1 and Fig. 1).

DISCUSSION

Our data demonstrate that vitamin D3 is effec¬
tive in inducing hypercalcemia and hyperphos¬
phatemia in N. piscator. These results concur with
earlier studies conducted with different species of
vertebrates [1-9]. As the animals were not fed in
the present study, the calcium needed for the
hypercalcemic response to vitamin D3 possibly was
derived from the demineralization of bones. The
probability of getting calcium from water cannot
be ruled out.

Parathyroid hormone has been reported to
facilitate the transformation of the vitamin D3
metabolite 25-hydroxycholecalciferol to the active
form 1, 25-dihydroxycholecalciferol [12, 13]. It has
also been reported that increased calcium ions can
inhibit this transformation by the direct effect on
the parathyroid and calcitonin secretion rates [14].
The observed decrease in calcemic value in Natrix
piscator on day 15 may be explained on the fact
that the elevated blood calcium may have suppres¬
sed the secretion of parathyroid hormone and thus
affecting the formation of 1 , 25-

dihydroxycholecalciferol. The decline of calcium

Phosphocalcic Response to Vitamin D3

895

level on day 15 can also be attributed to the
possible activation of the secretion of ultimobran-
chial gland, which has been reported to contain a
hypocalcemic factor - calcitonin [15, 16], in re¬
sponse to prolonged hypercalcemic challenge.
Suzuki et al. [17] have reported that the small
granular cells of ultimobranchial gland in sea
snake, Laticauda semifasciata contain highly de¬
veloped Golgi complexes which are surrounded by
immature granules in the process of formation.
They have explained that the activity of the
glandular cells is high because the snakes live in
sea where the calcium load is high.

The hyperphosphatemic action of vitamin D3
suggests that nondietary phosphorus, possibly
from the bone or soft tissues, can be mobilized.

Prior to the present report no data have been
available concerning vitamin D3 effect on serum
inorganic phosphate level from reptiles.

ACKNOWLEDGMENTS

This study was supported by a research grant F 23-
334/84-SR II from University Grants Commission, India
to A. K. Srivastav.

REFERENCES

1 MacIntyre, I., Colston, K. W., Evans, I. M., Lopez,
E., Macauley, S. J., Peignoux-Deville, J., Spanose,
E. and Szelke, M. (1976) Clin. Endocrinol., Suppl.,
5: 85.

2 Lopez, E., Peignoux-Deville, J., Lallier, F., Col¬
ston, K. W. and MacIntyre, I. (1977) Calcif. Tissues

Res., Suppl., 22: 19-23.

3 Srivastav, A. K. (1983) J. Fish Biol., 23: 301-303.

4 Fenwick, J. C., Smith, K., Smith, J. and Flik, G.
(1984) Gen. Comp. Endocrinol., 55: 398-404.

5 Swarup, K., Norman, A. W., Srivastav, A. K. and
Srivastav, S. P. (1984) Comp. Biochem. Physiol.,
78B: 553-555.

6 Robertson, D. R. (1975) Endocrinology, 96: 934-
940.

7 Bentley, P. J. (1983) Comp. Biochem. Physiol.,
76B: 717-719.

8 Srivastav, A. K., Rani, L. and Swarup, K. (1987)
Can. J. Zool. , 65: 2111-2112.

9 Srivastav, A. K., Srivastav, S. P., Srivastav, S. K.
and Swarup, K. (1986) J. Physiol., Paris, 81: 17-18.

10 Trinder, P. (1960) Analyst, 85: 889-894.

11 Fiske, C. H. and Subbarow,Y. (1925) J. Biol.
Chem., 66: 375-400.

12 Rasmussen, H., Wong, M., Bikle, D. and Good¬
man, D. B. P. (1972) J. Clin. Invest., 51: 2502-
2504.

13 Fraser, D. R. and Kodicek, E. (1973) Nature, New
Biol., 241: 163-166.

14 Boyle, I. T., Gray, R. W. and Deluca, H. F. (1971)
Proc. Natl. Acad. Sci. U. S. A., 68: 2131-2134.

15 Uchiyama, M., Yoshihara, M., Murakami, T. and
Oguro, C. (1978) Gen. Comp. Endocrinol., 36: 59-
62.

16 Uchiyama, M., Yoshihara, M., Murakami, T. and
Oguro, C. (1981) Gen. Comp. Endocrinol., 43:
259-261.

17 Suzuki, K., Yoshizawa, H., Yoshihara, M.,
Sasayama, Y. and Oguro, C. (1984) In “Endocrine
Control of Bone and Calcium Metabolism”. Ed. by
D. V. Cohn, J. T. Potts, Jr. and T. Fujita, Elsevier
Science Publishers B. V., Amsterdam, pp. 203-205.

-r

ZOOLOGICAL SCIENCE 5: 897-900 (1988)

© 1988 Zoological Society of Japan

[COMMUNICATION]

Establishment of a Multiple Recessive Tester Stock
in the Fish Oryzias latipes

Atsuko Shimada, Akihiro Shima and Nobuo Egami1

Laboratory of Radiation Biology, Zoological Institute, Faculty of
Science, University of Tokyo, Bunkyo-ku, Tokyo 113, and
lThe National Institute for Environmental Studies,

Ibaraki 305, Japan

ABSTRACT — A multiple recessive tester stock homozy¬
gous for 5 loci has been established in the fish Oryzias
latipes. The loci chosen as markers were b. Da, gu, pi,
and r. No linkage relationship between Da and gu, and
Da and pi was suggested so far as being judged from the
analyses of segregation ratio in the F2 generation.
However, no clear-cut conclusion could be obtained with
regard to gu-pl pair. We believe that the first
establishment of this stock obviously increased the
usefulness of the Medaka as a laboratory animal particu¬
larly suitable for studying germ cell mutations by the
specific-locus method.

INTRODUCTION

The specific-locus method has been used mainly
in mice to detect visible recessive mutations in
germ cells induced by radiations and/or chemicals
[1]. In this method, the tester mice homozygous
for 7 visible recessive loci were mated to wild-type
partners subjected to various treatments. There¬
fore, it is possible and hence the major advantage
of this method that recessive mutations at all the
loci concerned are detected in the first filial
generation after treatment of parents.

We have been using the fish Medaka Oryzias
latipes as a laboratory animal for studying radia¬
tion biology, aging and cancer. Very recently, we
showed the validity of the Medaka for detection of
germ cell mutations at the b locus in males exposed

Accepted December 12, 1987
Received November 19, 1987

to acute gamma-ray irradiation [2], To make
multiple visible recessive genes homozygous in a
tester stock will increase the usefulness of the
Medaka for detection of mutagens particularly in
the aquatic environment. Hence, in order to
establish the tester fish in the Medaka which is
valid for the specific-locus method, attempts were
made to obtain multiple visible recessive tester
fish. In this report, the procedures for producing
the tester as well as some problems encountered
during the production process are presented.

MATERIALS AND METHODS

Mutant genes used as markers

In the Medaka, over 50 visible mutant genes
have already been known [3-6]. Out of these, we
tentatively chose 5 mutant genes ( b: colorless
melanophores, Da: double anal fins, gu: reduced
deposition of guanine in iridocytes, pi: no pectoral
fins throughout life, r: colorless xanthophores),
based on the following criteria: (1) The recessive
homozygotes are easy to culture, and (2) easy to
distinguish from their wild-type counterparts. (3)
The mutant phenotypes are expressed early during
developmental stages and stable throughout the
life of the fish. All the mutant Medaka were kindly
given by Dr. H. Tomita of Laboratory of Freshwa¬
ter Fish Stocks, Nagoya University.

898

Step No.
(1)

A. Shim ad a, A. Shim a and N. Egami

(1) b/b +/+ r/r x +/+ Do/Da +/+

I

(2) b/+ Da/+ r/+

I

(3) b/b +/+ pl/pl r/r x b/b Da/Da +/+ r/+ (5) b/b +/+ r/r x b/b gu/gu +/+

(4)

(7)

(8)

(9)

(10)

b/b Da/+ pl/+ r/r

I

b/b Da/Da +/+ pl/pl r/r

(6) b/b gu/+ r/+

I

b/b +/+ gu/gu +/+ r/r

b/b Da/+ gu/+ pl/+ r/r

I

b/b Da/+ gu/gu pl/pl r/r

I

b/b Da/Da gu/gu pl/pl r/r

Fig. 1. Mating procedure for producing a multiple recessive tester stock of the Medaka. Numbers in parentheses
represent the steps for crossing of the males and females. In the steps 1, 3, 5, and 7, females were shown to the
left and males to the right, respectively.

Procedure for producing the multiple recessive
tester stock

Figure 1 shows the mating procedure for pro¬
ducing the multiple recessive homozygotes. The
fish of the genotypes (b/b r/r), (Da/ Da), (b/b
gu/gu), and (b/b pl/pl r/r) were the starting
materials. Adult fish were cultured in the glass
vessell 24 cm in diameter. Each fertilized egg was
plated one by one into a well of plastic microtiter
plates (Microtestplatten 96-fach Type-U, Kontron
Precision Products, Switzerland) [2]. This use of
microtiter plates made it possible to scrutinize for
any deviation in each embryo from the wild
phenotypes, and also to follow the developmental
fates of each embryo [2]. Both fish and eggs were
kept in a culture room at 26-29°C under the
illumination conditions of 14 hr light-10 hr dark.
In each mating step, the Fi progeny were obtained
from one pair of the parent (Steps 1, 3, 5, 7 in Fig.
1), whereas the F2 progeny were from two pairs of
the F! progeny (Steps 2, 4, 6, 8, 9 in Fig. 1).

Linkage analysis of genes used as markers

The segregation ratios in the F2 progeny just
after hatching (Step 8 in Fig. 1) were statistically
examined using the chi-square test method.

RESULTS AND DISCUSSION

Phenotypic expressions in the multiple recessive
tester stock

In the multiple recessive tester Medaka which
are homozygous for the 5 loci, the body color is
white (b/b r/r), deposition of guanine in iridocytes
around eyes and abdomen are reduced (gu/gu),
dorsal fins resemble anal fins (Da/ Da), and pecto¬
ral fins do not exist (pl/pl). The phenotypic
characters of the mutant genes and the develop¬
mental stages at which phenotypes were expressed
were not affected by the presence of multiple
recessive genes in homozygosity. Hence, the 5
mutant genes can be considered to be expressed
independently or to have no pleiotropic effect on
phenotypic expression of other’s. It is noteworthy
that the vertebral column of the adult females
homozygous for b. Da, pi, r, or b. Da, gu, pi, r
(Fig. 1, females in the Steps 7, 10) were frequently
curved dorso-ventrally. Moreover, viability of the
fish homozygous for b. Da, gu, pi, and r was very
low; out of 100 hatched embryos, about 10 on
average grew up to adult fish.

Segregation analysis for some loci

The data on numbers of F2 offspring obtained

Multiple Recessive Tester Medaka

899

during the production of the tester were used for
segregation analysis of the 3 locus-pairs, Da-gu,
Da-pl, and gu-pl, respectively (step 8 in Fig. 1).
The parents were a ( b/b Dal Da +/+ pH pi r/r )
female and a ( bib +/+ gu/gu +/+ rlr) male (Step
7 in Fig. 1). The F: progeny were homozygous for
b and r, but heterozygous for Da, gu, and pi.

Table 1A shows the segregations within one
locus. In the Da locus, the segregation ratio was
not significantly different from those to be ex¬
pected (X2—3.28, 0.05<P<0.10). In the gu and pi
loci, the segregation distortion was observed (gu:

X2 = 8.67, P<0.01; pl:X2=U. 54, P<0.01). This
may be due to zygote lethality in ( bib gu/gu r/r)
and (bib pl/pl r/r) embryos, although we do not
have any direct evidence.

For linkage analysis, we examined the segrega¬
tions within the two pairs of alleles, Da-gu , Da-pl ,
and gu-pl (Table IB). In every pair, X2 value was
very large (Da-gu: X2= 14.02, P <0.01; Da-pl: X2=
14.89, P<0.01; gu-pl: X2=25J4, P<0.01), in¬
dicating highly significant differences between
observed and expected numbers. In the Da-gu and
Da-pl pairs, however, we could not totally reject

Table 1. Segregation analysis for the Da, gu, and pi loci using F2 progeny just after hatching3

A. Segregations in the single locusb

Genotype

Observed (expected) number

X2

Dal Da

384 (416)

3.28

0.05<P<0.10

Dal+ and +/+

1280 (1248)

d.f. = lc

gu/gu

364 (416)

8.67

P<0.01

gu / + and +/+

1300 (1248)

d.f. = 1

pl/pl

pl/+ and +/+

356 (416)

1308 (1248)

11.54

P<0.01
d.f. = 1

B. Segregations in the two locid

Genotype

Observed (expected) number

y2

y2e

y2f

gu/gu gu/+ and +/+

X

X

X

Dal Da

Da/+ and +/+

94 (104) 290 (312)

270 (312) 1010 (936)

14.02

P<0.01

11.95

2.07

P>0.15
d.f. = 1

pl/pl pll+ and +/+

Da/Da

Da/+ and +/+

83 (104) 301 (312)

273 (312) 1007 (936)

14.89

PC0.01

14.82

0.07

P>0.7
d.f. = 1

pl/pl pl/+ and +/+

gu/gu

gu/ -T and +/+

94 (104) 270 (312)

262 (312) 1038 (936)

25.74

PC0.01

20.21

5.53

0.01<P<0.02
d.f. = 1

a Parents were a (bib Da/Da +/+ pl/pl rlr) female, and a (bib +/+ gu/gu +/+ r/r) male (Fig. 1, Step 7).

Fx progeny were (bib Da/+ gw/+ pl/+ r/r) fish (Step 8 in Fig. 1).
b The remaining two loci were ignored.
c Degree of freedom.
d The remaining one locus was ignored.
e Sum of the X2 value for the two single locus.

f Difference between sum of the X2 values for the two single locus and X2 value for the locus-pair.

900

A. Shimada, A. Shima and N. Egami

the hypothesis that these loci were independent
one another because the distortion within a locus-
pair could be explained by distortion in the
corresponding loci. For instance, sum of the X2
value for segregation in the Da and gu loci was
11.95 ( = 3.28 + 8.67 in Table 1 A), while the X2
value in the Da-gu pair was 14.02. The subtraction
of the former X2 value (11.95) from the latter
(14.02) will give the X2 value for the independence
of the two loci concerned. The X2 value thus
obtained was 2.07, giving P>0.15 at any degree of
freedom. This result suggested that the segrega¬
tion distortion in the Da-gu pair was not due to the
linkage relationship between the loci but rather
due to the segregation distortion in the loci
concerned. With regard to the Da-pl pair, the
same results were obtained; difference between
the sum of the X2 value for the Da and pi loci and
the X2 value in the Da-pl pair was 0.07 ( = 14.89
— 14.82, P>0.7 at any degree of freedom. Table
IB). Therefore, we may conclude that no linkage
relationship was suggested between the Da and gu ,
and Da and pi loci.

In the gu-pl pair, however, the distortion could
not be explained even by the distortions in the loci
concerned; difference between the sum of the X2
value in the gu-pl pair was 5.53 (0.01 <P< 0.02).
The biological meaning of this result is not known,
and hence no clear-cut conclusion was obtained for
relationship between the gu and pi loci.

Attempts for segregation analysis by a backcross
method are now under way in order to obtain some
clue to understand the relationship between these
loci. In this connection, it is noteworthy that
Naruse et al. [7] recently found no linkage rela¬
tionship between the following locus pairs by the
use of diploid gynogenesis; pi and b, gu and b, and
Da and b.

ACKNOWLEDGMENT

We wish to thank Dr. H. Tomita of Laboratory of
Freshwater Fish Stocks, Nagoya University for his kind
supply of the fish. This work was supported in part by a
grant-in-aid and a subsidy from Ministry of Education,
Science, and Culture, Japan, to which we are very
grateful.

REFERENCES

1 Russell, W. L. and Kelly, E. M. (1982) Proc. Natl.
Acad. Sci., USA., 79: 542-544.

2 Shima, A. and Shimada, A. (1988) Mutation Res.,
198: 93-98.

3 Yamamoto, T. (1975) Medaka (Killifish) — Biology
and Stains, Keigaku Publishing Company, Tokyo,
pp. 1-365.

4 Tomita, H. (1982) Medaka, 1: 7-9.

5 Tomita, H. (1984) Medaka, 2: 1-5.

6 Tomita, H. (1985) Medaka, 3: 5-16.

7 Naruse, K., Shimada, A. and Shima, A. (1988) Zool.
Sci., 5: 489-492.

ZOOLOGICAL SCIENCE 5: 901-903 (1988)

© 1988 Zoological Society of Japan

[COMMUNICATION]

Karyotype of a Scincid Lizard, Carlia fusca ,
from Guam, the Mariana Islands

Hidetoshi Ota, Tsutomu Hikida, Masafumi Matsui1
and Masami Hasegawa2

Department of Zoology, Faculty of Science, and 1 Biological Laboratory,
Yoshida College, Kyoto University, Sakyo, Kyoto 606, and 2Office
for the Establishment of Prefectural Museum of Chiba,
Katsuragi 2-10-1, Chiba 280, Japan

ABSTRACT — Karyotype of a scincid lizard, Carlia
fusca, is described. This is the first report for the
karyotype of the genus. The species has 2n — 30 chromo¬
somes forming two size groups. All the four pairs
belonging to the larger group are metacentric elements.
Pairs 5 and 6 also appear metacentric chromosomes,
whereas pairs 7 submetacentric, pairs 8 and 9 sub-
telocentric, and pairs 10 to 15 acrocentric elements.
Among the other scincids, Leiolopisma species most
closely resemble C. fusca from the karyological aspect.

INTRODUCTION

The genus Carlia belongs to the subfamily
Lygosominae, the most advanced subgroup of the
family Scincidae, and accomodates some 40 spe¬
cies distributed in Australia, Papua New Guinea,
eastern part of Indonesia, and the Mariana Islands
[1-6]. Recently, several taxonomists have inten¬
sively investigated and described external and
skeletal morphology of some members of this
genus [4-9]. However, no information has ever
been provided regarding the karyotype of Carlia
species.

In the present paper, we will describe the
karyotype of C. fusca, a species distributed in
Australia, Papua New Guinea, and the Marianas
[1, 6], for the first time as a part of a revisional
work on the cytotaxonomy of the family Scincidae,

Accepted December 11, 1987
Received November 16, 1987

one of the most poorly studied families among the
suborder Lacertilia [10].

MATERIALS AND METHODS

Two males and six females, collected from
Guam, the Mariana Islands in December 1986 by
MH, were examined. Air-dried mitotic cells,
prepared after Ota et al. [11], were stained in 20%
Giemsa solution, and were photographed for
detailed observations. The karyotype was deter¬
mined on the basis of more than 20 well-spread
cells for each animal.

The voucher specimens were deposited in the
Osaka Museum of Natural History, Australian
Museum, and the herpetological collection of the
Department of Zoology, Kyoto University.

RESULTS AND DISCUSSION

The chromosome number was determined as 2n
= 30 (Fig. 1). No sex chromosome heteromor¬
phism was recognized. Chromosome pairs 1 to 4
were distinctly larger than the remainder. All
chromosomes belonging to the larger group were
metacentric. Of the smaller group, pairs 5 and 6
were metacentric, pairs 7 submetacentric, pairs 8
and 9 subtelocentric, and pairs 10 to 15 acrocentric
elements.

Among the other scincids hitherto karyotyped,
13 lygosomine species, Mabuya m. mabouya.

902 H. Ota, T. Hikida et al.

IX

1

¥8

2

8K

3

»«

4 5

••

6

*«

7

mm

8

* a

9

m m

10

mm

11

m m

12

*1?

m m>

14

• a

15

Fig. 1. Karyotype of a famale Carlia fusca. The scale represents 10 pm.

Scincella laterale, S. himalayanum, Riopa sunde-
valli, Ctenotus lesueurii, Sphenomorphus tympa¬
num, Ctenotus spaldingi, Cyclodina oliveri, Leiolo-
pisma weekesae, L. entrecasteauxii, L. guichenoti,
L. trilineatum and L. telfairi, share diploid
chromosome number (30) with Carlia fusca [12-
17]. From the former six species, Carlia fusca
differs in possessing subtelocentric chromosomes
for pair 8, since all those have metacentric pair 8.
Similarly, the pair 7 of Ctenotus spaldingi is
metacentric, whereas that of Carlia fusca is dis¬
tinctly submetacentric. The karyotype of Cyclodi¬
na oliveri is distinct from that of Carlia fusca in
exhibiting male sex chromosome heteromorphism.
Thus, the karyotype of C. fusca most closely
resembles those of the Leiolopisma species, and
this seems to coincide with the results of morpho¬
logical analysis by Greer [3-5], who assumed a
close phylogenetic relationship between Carlia and
Leiolopisma.

Several authors have surmised the family Scinci-
dae to be karyologically relatively conservative
[10, 18]. The present results seem to support this
assumption, since the karyotype of Carlia fusca
exhibited several characteristics common to the
known scincid karyotypes - moderate diploid
number, subdivision into two size-groups, and the
numerical dominance of biarmed elements in the
larger group [10, 18]. Even so, however, very few
karyological data have, as yet, been available

considering the large number and high morpho¬
logical and ecological diversity of species belong¬
ing to the family. The lygosominae is the largest
scincid subfamily and encompasses 40 to 60 genera
and more than 600 species [2]. However, only
about 25 species of eight genera have hitherto been
karyologically examined in this subfamily.
Moreover, some of these species were karyotyped
only by a testis-sectioning technique, which may
produce incorrect data [18]. Thus, further studies
on extensive taxa are required to correctly outline
the karyological divergence of the family Scin-
cidae.

ACKNOWLEDGMENTS

We thank H. G. Cogger for confirming our identifica¬
tion of the present materials, and H. Fukada, S. Ishi-
hara, H. Nakamura and Y. Shibata for providing litera¬
ture cited here. This work was supported in part by a
Grant-in-Aid for Special Project Research on Biological
Aspects of Optimal Strategy and Social Structure from
the Japan Ministry of Education, Science and Culture.

REFERENCES

1 Dryden, G. L. and Taylor, E. H. (1969) Univ. Kan¬
sas Sci. Bull., 48: 269-279.

2 Greer, A. E. (1970) Bull. Mus. Comp. Zool., 139:
151-184.

3 Greer, A. E. (1974) Aust. J. Zool., Suppl. Ser., 31:
1-67.

Karyotype of Carlia fusca

903

4 Greer, A. E. (1976) Herpetologica, 32: 371-377.

5 Greer, A. E. (1979) Rec. Aust. Mus., 32: 339-371.

6 Cogger, H. G. (1979) Reptiles and Amphibians of
Australia (Revised ed.). Ralph Curtis Book,
Florida.

7 Mitchell, F. J. (1953) Rec. S. Aust. Mus., 11: 75-
90.

8 Storr, G. M. (1974) Rec. W. Aust. Mus., 3: 151-
165.

9 Greer, A. E. (1975) Herpetologica, 31: 70-75.

10 Bickham, J. W. (1984) In “Chromosomes in Evolu¬
tion of Eukaryotic Groups. Vol. II”. Ed. by A. K.
Sharma and A. Sharma, CRC Press, Florida, pp.
13-40.

11 Ota, H., Matsui, M., Hikida, T. and Tanaka, S.
(1987) Experientia, 43: 924-925.

12 Becak, M. L., Becak, W. and Denaro, L. (1972)
Caryologia, 25: 313-326.

13 King, M. (1973) Aust. J. Zool., 21: 21-32.

14 Wright, J. W. (1973) Chromosoma (Berl.), 43: 101—
108.

15 Hardy, G. S. (1979) N. Z. J. Zool., 6: 609-612.

16 DeSmet, W. H. O. (1981) Acta Zool. Pathol.
Antverp., 76: 35-72.

17 Duda, P. L. and Gupta, A. (1981) Chrom. Inf.
Serv., 30: 23-25.

18 Gorman, G. C. (1973) In “Cytotaxonomy and
Vertebrate Evolution”. Ed. by A. B. Chiarelli and
E. Capanna, Academic Press, New York, pp. 349-
424.

j

j

ZOOLOGICAL SCIENCE 5: 905-909 (1988)

© 1988 Zoological Society of Japan

[COMMUNICATION]

S-100 Protein-like Xmmunoreactive Cells in the Brain-Midgut
Endocrine System of the Insect Periplaneta arnericana

Yasuhisa Endo and Toyoshi Endo1

Department of Applied Biology, Kyoto Institute of Technology, Matsugasaki,
Sakyo-ku, Kyoto 606, and 1 Department of Medicine, Yamanashi
Medical College, Tamaho, Yamanashi 409-38, Japan

ABSTRACT— Using an antiserum to S-100 protein
which is known as a unique protein to glial cells in the
central nervous system of mammals, the brain-midgut
endocrine system of the insect Periplaneta arnericana was
investigated by immunohistochemistry. Unexpectedly,
immunoreactivity of S-100 protein was detected in the
medial neurosecretory cells of the brain, and in the
endocrine cells of the midgut epithelium. Immunoreac-
tive neuronal processes from the medial neurosecretory
cells extend to the corpora cardiaca, a neurohemal organ
of insects. These results suggest that an S-100 protein¬
like substance, i.e. a substance reactive to the anti-S-100
protein serum, may be secretory and bioactive in insects.

INTRODUCTION

It is well known that, in mammals, brain and
gut, neurons and endocrine cells, share a number
of bioactive peptides. Recently Nishiitsutsuji-Uwo
and Endo [1] have revealed that the insects,
phylogenetically far from mammals, have a homo¬
logous brain-midgut endocrine system. The ultra¬
structure of gut endocrine cells in insects was very
similar to that of mammals [2-4]. The immunohis-
tochemical works indicated that the immunoreactiv-
ities of pancreatic polypeptide, somatostatin and
enteroglucagon were detected in different types of
endocrine cells in the cockroach midgut [5, 6].
Furthermore, pancreatic polypeptide- and soma-
tostatin-like immunoreactivities were frequently
found in many neurons in the central and visceral

Accepted October 31, 1987
Received October 5, 1987

nervous system of the cockroaches [7, 8].

S-100 protein is a Ca2+ -binding protein and first
isolated as a soluble protein of the central nervous
system of the cow, rabbit and pig [9]. The
immunohistochemical works indicated that this
protein is localized in glial cells of the central and
peripheral nervous system [10, 11]. However, as
later immunological studies indicated, this protein
is not unique to glial cells, but contained in a wider
range of cell types; for example, fat cells [12],
chondrocytes [13], reticular cells of lymphoid
organs [14], stellate cells of anterior pituitary [15]
and some islet cells of pancreas [16]. In inverte¬
brates, there have been little informations on S-
100 protein. In the present study, we have
investigated immunohistochemically of the pre¬
sence of S-100 protein in the brain-midgut endoc¬
rine system of the cockroaches.

MATERIALS AND METHODS

Adult American cockroaches (Periplaneta amer-
icana) of both sexes were used. The brain and
midgut were removed in 0.01 M phosphate buf-
fered-saline (0.8% NaCl, pH 7.4) and then fixed in
Bouin’s fluid for 2-3 hr. The tissues were dehy¬
drated in ethanol series, cleared in xylene and
embedded in paraffin. Serial sections were cut at
4-6 pm thickness and mounted on a slide glass
coated with chrome allum-gelatin.

Dewaxed sections were immunostained by the
peroxidase-antiperoxidase (PAP) method [17]
using a rabbit antiserum to bovine S-100 protein

906

Y. Endo and T. Endo

[18]. The specificity of the antiserum was de¬
scribed previously [18]. The dilutions of anti-S-
100 protein serum, goat anti-rabbit IgG serum
(Miles) and rabbit PAP complex (Miles) were
1: 1000, 1: 100 and 1: 100, respectively. Incubation
of first layered antiserum was carried out at 4°C for
12-16 hr and that of other antisera was done at
room temperature for 0.5-1 hr. After coloration
with 3, 3' diaminobenzidine and H202, the sec¬
tions were wealky stained with hematoxylin.

The staining specificity was checked as follows;
1) omission of first layered antiserum, and 2)
replacement of first layered antiserum to the
antiserum preabsorbed with excess antigen
(bovine S-lOOb protein, 100 ^g/ml of the diluted
antiserum, at 4°C for 24 hr). In both tests, the
specific staining was ascertained.

RESULTS

We have expected that, if S-100 protein is
present in the brain of the cockroaches, this
protein should be contained in the glial cells as in
the case of mammals. However, the cells reactive
for anti-S-100 protein serum were neurons. Fur¬
thermore, the reactive neurons were only a small
population and located in a specific region, pars
intercerebralis of protocerebrum (Fig. 1). This
region is known as the site of medial neurosecre¬
tory cell group. The size of the immunoreactive
neurons was not uniform; Large (30-40 /urn in
diameter) and medium (20-30 jam) sized neurons
were mixed. Immunoreactivity of S-100 protein
was found in their cytoplasm but not in the nuclei
(Fig. 2). Their neuronal processes gathered into a
bundle per each hemisphere and extended to the
corpora cardiaca, neurohemal organ of insects
(Fig. 3). No strongly immunoreactive cells were
found in other regions of the protocerebrum
including lateral neurosecretory cell group, the
deutocerebrum and the tritocerebrum, although
weak immunoreactive neurons were found in the
ventral portion of protocerebrum.

In the midgut of cockroaches, small number of
epithelial cells were reactive for anti-S-100 protein
serum (Fig. 4). These cells were cone-shaped and
scattered in the basal portion of epithelium.
Immunoreactivity of S-100 protein was found in

the cytoplasm, especially in the perinucleic region.
The shape and location of these cells indicated that
these cells apparently belonged to basal-
granulated cells. Frequency of S-100 protein
immunoreactive cells was not so high in the total
population of basal-granulated cells.

DISCUSSION

The present study revealed that the presence of
S-100 protein-like substance in the brain-midgut
endocrine system of the cockroaches. The fact that
this substance was contained in the neurosecretory
cells and gut endocrine cells was unexpected but
not surprising. Even in mammals, as mentioned in
the introduction, a wider range of cell types
contain this protein [12-16]. The present study
suggested that S-100 protein (-like substance) may
be universal in vertebrates and invertebrates,
although this protein was not contained in glial
elements of insects.

The nature of S-100 protein-like substance in
insects is totally unknown. However, the specific
localization of this substance in the brain and
midgut suggested that -this substance may be
secretory, bioactive substance in insects. Alterna¬
tively, there is a possibility that this substance may
be a carrier protein containing the antigenic
determinants common to S-100 protein.

In mammals, two interesting data were reported
on S-100 protein-producing cells except glial cells:
According to Ishikawa et al. [19], cultured stellate
cells of rat anterior pituitary secrete S-100 protein
which stimulates release of prolactin from prolac¬
tin-secreting clonal cells in vitro. According to
Girod et al. [16], some islet cells of monkey
pancreas are immunoreactive for anti-S-100 pro¬
tein serum, but not for anti-insulin, glucagon,
somatostatin and pancreatic polypeptide sera.
These reports suggested the possibility that S-100
protein is a chemical messenger acting intercellu-
larly. But the ultrastructural localization of this
protein was not reported in both cases, although
Ishikawa et al. [19] described that the stellate cells
were agranular cells. Immunoelectron microscopic
study should be necessary to elucidate whether or
not this substance is localized in the membrane
bound-secretory granules of the neurosecretory

S-100 Protein Immunoreative Cells in Insects

907

Fig. 1. Frontal section view of brain of cockroach. S-100 protein immunoreactive cells are seen in the pars
intercerebralis (pi) of the protocerebrum. X80.

Fig. 2. Sagittal section view of pars intercerebralis of brain. Immunoreactivity of S-100 protein is found in the
cytoplasm of large (arrowhead) and medium (arrows) sized neurons. X800.

908

Y. Endo and T. Endo

4

Fig. 3. Corpus cardiacum (cc) and corpus allatum (ca). Nerve fibers containing the immunoreactivity of S-100
protein (arrows) locate densely in the peripheral part of the corpus cardiacum, a neurohemal organ of insects. No
immunoreactive fibers are seen in the corpus allatum. X700.

Fig. 4. Cross sectional view of midgut epithelium. Immunoreactivity of S-100 protein is detected in the cytoplasm of
basal-granulated cell (arrowhead). Arrows indicate non-immunoreactive basal-granulated cells, n: nidus of
generative cells. X600.

S-100 Protein Immunoreative Cells in Insects

909

cells and gut endocrine cells of the cockroaches.

ACKNOWLEDGMENTS

We thank Dr. N. Agui (The National Institute of

Health, Tokyo) for a generous supply of cockroaches.

REFERENCES

1 Nishiitsutsuji-Uwo, J. and Endo, Y. (1984) In “In¬
sect Neurochemistry and Neurophysiology”. Ed. by
A. B. Borkovec and T. J. Kelly, Plenum Publ., New
York, pp. 451-453.

2 Nishiitsutsuji-Uwo, J. and Endo, Y. (1981)
Biomed. Res., 2: 30-44.

3 Endo, Y. and Nishiitsutsuji-Uwo, J. (1981)
Biomed. Res., 2: 270-280.

4 Endo, Y. and Nishiitsutsuji-Uwo, J. (1982) Cell
Tissue Res., 222: 515-522.

5 Iwanaga,T., Fujita, T., Nishiitsutsuji-Uwo, J. and
Endo, Y. (1981) Biomed. Res., 2: 202-207.

6 Endo,Y., Nishiitsutsuji-Uwo, J., Iwanaga, T. and
Fujita, T. (1982) Biomed. Res., 3: 454-456.

7 Endo, Y. , Iwanaga, T., Fujita, T. and Nishiitsutsuji-
Uwo, J. (1982) Cell Tissue Res., 227: 1-9.

8 Fujita, T., Yui, R., Iwanaga, T., Nishiitsutsuji-

Uwo,!., Endo, Y. and Yanaihara, N. (1981) Pep¬
tides, 2, suppl. 2: 123-131.

Moore, B. W. (1965) Biochem. Biophys. Res. Com-
mun„ 19: 739-744.

Matus, A. and Mughal, S. (1975) Nature, 258: 746-
748.

Ferri, G. L., Probert, L., Cocchia, D., Michetti, F.,
Marangos, P. J. and Polak, J. M. (1982) Nature,
297: 409-410.

Michetti, F., Dell’Anna, E., Tiberio, G. and Coc¬
chia, D. (1983) Brain Res., 262: 352-356.
Stefansson, K., Wollmann, R. L., Moore, B. W.
and Arnason, B. G. W. (1982) Nature, 295: 63-64.
Iwanaga, T., Fujita, T., Masuda, T. and Takahashi,
Y. (1982) Arch. Histol. Jpn., 45: 393-397.
Nakajima, T., Yamaguchi, H. and Takahashi, K.
(1980) Brain Res., 191: 523-531.

Girod, C., Durand, N. and Raccurt, M. (1987) Cell
Tissue Res., 247: 11-16.

Sternberger, L. A. (1979) Immunocytochemistry.
Wiley, New York, 2nd ed., 354 pp.

Hidaka, H., Endo, T., Kawamoto, S., Yamada, E.,
Umekawa, H., Tanabe, K. and Hara, K. (1983) J.
Biol. Chem., 258: 2705-2709.

Ishikawa, H., Nogami, H. and Shirasawa, N. (1983)
Nature, 303: 711-713.

9

10

11

12

13

14

15

16

17

18

19

ZOOLOGICAL SCIENCE 5: 911-914 (1988)

© 1988 Zoological Society of Japan

[COMMUNICATION]

Changes of Acetylcholinesterase Activity in Rat Supraoptic
Nucleus Cell Bodies during Water Deprivation

Yuta Kobayashi and Seiichiro Kawashima1

Department of Pharmacology, Shimane Medical University, Izumo,
Shimane 693, and 1 Zoological Institute, Faculty of Science,

Hiroshima University, Hiroshima 730, Japan

ABSTRACT — Changes of acetylcholinesterase (AChE)
activity in the cell bodies of the supraoptic nucleus
(SON) of the Sprague-Dawley/Tw male rats during water
deprivation were studied using microspectrophotometry.
The AChE activity was measured as the optical
density at 490 nm in a 6 /mi spot on the histochem-
ical reaction product in the cell bodies. A significant
reduction in the mean maximum AChE activity was
observed after 2 days of water deprivation. Following a
moderate recovery at 4 days, the enzyme activity
significantly elevated at 8 days as compared with the
initial control level. The number of AChE positive cells
in a section per rat increased with the dehydration
period, attaining more than 2-fold at 8 days compared
with that of the controls. The reduction in AChE activity
at 2 and 4 days may indicate the reduction of the total
enzyme activity per cell. However it may be caused by
the decrease in density due to the hypertrophy of the cell
bodies. Notwithstanding the hypertrophy, AChE activity
was the greatest after 8 days of water deprivation. The
present results suggest that the changes in AChE activity
in the SON cell bodies are closely related with the
response of these neurons to osmotic stress.

INTRODUCTION

Chronic water deprivation or salt loading effec¬
tively reduces the vasopressin and oxytocin con¬
tent in the neural lobe of rats [1], showing that
vasopressin- and oxytocin-producing neurons are
responsive to osmotic stress. Previous investiga¬
tors concluded that cholinergic mechanism local¬
ized in the supraoptic nucleus (SON) was involved
in mediating vasopressin release ([2], for review

Accepted November 20, 1987
Received October 28, 1987

see [3]). In effects, acetylcholinesterase (AChE)
activity was histochemically demonstrated in the
SON by Uemura [4] and Shute and Lewis [5], and
was micro-radiometrically quantified using a map¬
ping technique by Packman et al. [6]. Although
histological changes of the hypothalamo-neuro-
hypophysial system in rats to osmotic stress has
been well documented (for review, see [7]),
detailed morphometrical analysis on the changes
of AChE activity in the SON has not yet been
carried out. Therefore, microspectrophotometry
was applied to clarify this point in the present
study.

MATERIAL AND METHODS

Animals Twenty male rats of the inbred Spra¬
gue-Dawley/Tw strain were kept in separate cages
and fed laboratory chow (CA-1, CLEA Japan
Inc.) and tap water ad libitum until experiments.
At 3 months of age animals were divided into four
groups of five rats each; control, 2 days, 4 days and
8 days of water deprivation.

Histochemistry For the demonstration of
AChE activity, Karnovsky’s method [8] was em¬
ployed with a slight modification. After decapita¬
tion, the hypothalamic region was quickly dis¬
sected out in each rat and fixed in 4% formalin (pH
7.6 with 0.1 M phosphate buffer) for 16 hr at 4°C.
After washing for 24 hr in 0.88 M sucrose, the
tissue was frozen in isopentane cooled with dry
ice-acetone and frontal sections were cut at 20 pm
thickness on a cryostat and mounted on a slide.

912

Y. Kobayashi and S. Kawashima

The sections on four slides from four different
group rats were incubated as a single batch for 30
min at 37°C in a freshly prepared reaction mixture
consisting of (1) 60 mg of acetylthiocholine iodide
dissolved in 78 ml of 0.1 M sodium hydrogen
maleate buffer at pH 6.0, (2) 6 ml of 0.1 M sodium
citrate, (3) 12 ml of 30 mM copper sulfate, (4) 12
ml of 1 mM tetra-isopropylpyrophosphoramine
(isoOMPA), (5) 12 ml of 5 mM potassium ferricy-
anide and (6) 18 g of sucrose. IsoOMPA was used
as a blocker of nonspecific cholinesterase.

Microspectrophotometry Reaction product was
reddish brown precipitates and showed the max¬
imum absorption wave length at 490 nm. In each
rat one section at the middle part of the left SON
was used for measurement. The optical density
(OD) at 490 nm in spot of 6 pm in diameter was
measured in each AChE positive cell body distrib¬
uted in the SON using a microspectrophotometer
(Olympus, MMSP) and the maximum OD de¬
tected by scanning whole area of each cell body
was recorded. All the cell bodies bearing the
cytoplasmic space larger than the spot size were
measured.

For the statistical analysis, Student’s t-test and
X 2-test were employed.

RESULTS

In the hypothalamus, relatively large amounts of
AChE reaction product were found in many
magnocellular cells of the SON (Fig. 1). The
number of AChE positive cells measured in one
section per rat of left SON increased significantly
with dehydration period: control, 110 + 9; 2-day
dehydration group, 182 ±17 (P<Y).05 compared
with the control); 4-day dehydration groug, 188 ±
26 (P<0.05); 8-day dehydration group, 267 ±25
(P< 0.005). In all water deprived rats the hyper¬
trophy of AChE positive cells was evident.

The distribution of the maximum OD of the
SON cell bodies in five sections in each group was
shown in Figure 2. After 2 days of dehydration,
cells showing weak AChE activities (OD less than
0.06) increased in frequency and the median of the
distribution was the smallest among the four
groups. After 4 days of dehydration, the frequen¬
cy of cells showing strong AChE activity (OD

Fig. 1. Acetylcholinesterase positive cell bodies in the
rat supraoptic nucleus without water deprivation.
The spot with an arrow indicates the size of a
scanning spot of 6 pm in diameter. OC; optic
chiasma. The bar indicates 50 urn.

greater than 0.11) became more than that of the
2-day dehydration. After 8 days of dehydration,
cells showing strong AChE activity increased much
more and the median was the largest among the
four groups.

A significant decrease in the mean maximum
OD was found in AChE positive cell bodies in rats
2 days after water deprivation (Fig. 3). It recov¬
ered slowly. However, it was still significantly
lower after 4 days of dehydration than the initial
control level. In contrast, it significantly exceeded
the control level after 8 days of dehydration.

DISCUSSION

Previous investigators demonstrated AChE
activity in the SON [4, 5]. Almost all the
magnocellular neurons containing immunoreactive
vasopressin or oxytocin [9] seemed to show AChE
activity.

Present results showing a significant change in
AChE activity in the cell bodies of the SON after
dehydration are in harmony with the previous
notion that cholinergic system in the SON is
involved in the releasing mechanism of vasopressin
([2], for review, see [3]).

The changes in AChE activity are known to
occur in physiological conditions. For example,
rapid decrease and restoration of AChE activity
have been reported in the anterior part of the

Number of measured cell bodies

AChE in SON cells during Dehydration

913

Fig. 2. Effects of dehydration on the distribution of the
maximum acetylcholinesterase (AChE) activity in
neurosecretory cells of the supraoptic nucleus of
male rats of the Sprague-Dawley/Tw strain. Results
obtained from five slides of five rats were compiled.
AChE activity is expressed in terms of the optical
density (OD) at 490 nm in a scanning spot of 6 jum in
diameter. Arrows with numbers indicate the me¬
dian of the distribution. The median was significant¬
ly different between any two groups ( X 2-test, P<
0.001). Broken lines indicate arbitrary limits of the
cells showing weak AChE activity (OD less than
0.06) and those showing strong activity (OD greater
than 0.11).

hypothalamus of the rat within 15 min of suckling
stimulus [10]. In a separate study, a reduction of
AChE content in the SON with aging was
observed [11]. After 2 or 4 days of water
deprivation, the mean and median of the max¬
imum OD were decreased. The decrease may be
due to the reduction of enzyme density with the

Fig. 3. Changes of the maximum acetylcholinesterase
(AChE) activity in neurosecretory cells of the su¬
praoptic nucleus after water deprivation. The num¬
bers in parentheses indicate the number of cells with
AChE activity. Each point depicts the mean±S.E.
of the maximum optical density (OD) at 490 nm of
each cell body in a 6-jum scanning spot. Significance
of difference from the control value: dehydration for
2 days, P<0.001; 4 days, P<0.001; 8 days, P<0.001
(Student’s t-test).

hypertrophy of AChE positive cell bodies and/or
the increase in the number of cells with weak
AChE activity. Increase in somatic size of the
magnocellular neurons of the SON in response to
osmotic stress has already been reported by
Hatton and Walters [12], They reported that the
increase was about two fold after 7 days of water
deprivation. In addition, the attachment of
neurons to neurons with less interposing glial cells
became conspicuous [7]. Such a hypertrophy of
neuronal cell bodies may explain, at least in part,
the increase in the number of cell bodies measured
in the present study. Furthermore, the possibility
of the change of some AChE negative cells to
positive ones after water deprivation cannot be
denied. In this case, it is possible that total AChE
content may increase by dehydration, which was
apparent in rats at 8 days after dehydration.

At 4 days after dehydration, cell bodies showing
strong AChE activity began to increase and at 8
days after dehydration, the maximum AChE
activity increased significantly concurrently with
the increase in the numbers of AChE positive cells
and the hypertrophy of cell bodies. Thus, increase
of AChE in the SON cell bodies after 8 days of

914

Y. Kobayashi and S. Kawashima

dehydration was evident. The present results
suggest that the cholinergic system in the SON is
concerned with the osmotic stress-induced vaso¬
pressin release in the rat.

REFERENCES

1 Jones, C. W. and Pickering, B. T. (1969) J. Phys¬
iol., 203: 449-458.

2 Sladek, C. D. and Joynt, R. J. (1979) Endocrinolo¬
gy, 104: 659-663.

3 Kawashima, S., Kawamoto, K. and Kobayashi, Y.
(1986) Zool. Sci., 3: 227-244.

4 Uemura, H. (1965) Annot. Zool. Japon., 38: 79-96.

5 Shute, C. C. D. and Lewis, P. D. (1966) Br. Med.
Bull., 22: 221-226.

6 Packman, P. M., Godfrey, D. A., Williams, A. D.
and Matschinsky, F. M. (1977) J. Histochem.
Cytochem., 25: 1237-1246.

7 Hatton, G. I. (1986) Fed. Proc., 45: 2328-2333.

8 Karnovsky, M. J. (1964) J. Cell Biol., 23: 217-232.

9 Rhodes, C. H. Morrell, J. I. and Pfaff, D. W.
(1981) J. Comp. Neurol., 198: 45-64.

10 Oba, T., Ota, K. and Yokoyama, A. (1971) Endo¬
crinol. Japon., 18: 23-26.

11 Kobayashi, Y. and Kawashima, S. (1988) J. Sci.
Hiroshima Univ. Ser. B, Div. 1, 33: (in press)

12 Hatton, G. I. and Walters, J. K. (1973) Brain Res.,
59: 137-154.

ZOOLOGICAL SCIENCE 5: 915-918 (1988)

© 1988 Zoological Society of Japan

[COMMUNICATION]

Invasion of Drosophila albomicans to the Mainland of Japan

Nobuhiko Asada

Department of Biology, Faculty of Science, Okayama University
of Science, Okayama 700, Japan

ABSTRACT — Drosophila albomicans Duda, a species
belonging to the D. nasuta subgroup of the immigrans
species group, has been known to distribute widely in
tropical and subtropical regions in Asia, but never
reported in the mainland of Japan. In 1984, the species
was first captured in the Inland Sea region of Japan and is
now widening its distribution area in western Japan.

INTRODUCTION

Drosophila albomicans Duda 1924 belongs to
the D. nasuta subgroup of the immigrans species
group, widely distributing in the tropical and
subtropical regions of Asia (Thailand, Pescadores
Islands, and Formosa). In Japan, this species has
been reported in the Ryukyu Islands (Iriomote-
jima, Ishigaki-jima, Okinawa, Tokunoshima, and
Amami-oshima), and was first collected in 1966 by
Okada and Kitagawa [1]. Since then, Amami-
oshima, 28°N and 129°E, has been regarded as the
northernmost distribution area [2], and extensive
surveys of Drosophila populations around the
mainland of Japan have failed to obtain D.
albomicans at all [3-5].

Since 1984, unexpectedly, D. albomicans has
been frequently collected at many localities on the
coast of the Inland Sea, 34-35°N and 133-135°E,
in western Japan [6-7]. In this article, I report on
the invasion of D. albomicans to the mainland of
Japan.

METHODS

Sixteen collection sites shown in Table 1 and

Accepted November 11, 1987
Received October 12, 1987

Figure 1 were chosen where traps were set, in
human habitation, orchard, winery, brewery, and
grassland. At each site, five to ten traps containing
fermented banana baits were placed 10-20 meters
apart from late May to late October in each year of
1984-1987. Drosophila flies in the traps were
collected three times a day for two days. Net
sweeping was also carried out in some places.

RESULTS AND DISCUSSION

Table 1 shows the summary of the collection
data of Drosophila albomicans. The places where
D. albomicans were collected is also shown in
Figure 1. It is clear that D. albomicans has been
collected in the Inland Sea region only, especially
the eastern half of the Sea; e.g., Okayama (Site 1,
Okayama Pref., orchard, 6 females and 5 males in
1986; 2 males in 1987), Akasaka (Site 2, Okayama
Pref., winery, 4 females and 4 males in 1986; none
in 1987), Seto (Site 3, Okayama Pref., brewery, 6
females and 3 males in 1986; none in 1987),
Ko-no-shima (Site 5, Kasaoka, Okayama Pref.,
orchard, 1 female in 1985; 19 females and 18 males
in 1986), Nishi-oshima (Site 6, Kasaoka, Okayama
Pref., orchard, more than 26 females in 1986),
Takamatsu (Site 10, Kagawa Pref., orchard, 14
females and 16 males in 1986), Tokushima (Site
11, Tokushima Pref., park, 8 females and 6 males
in 1986), Akashi (Site 13, Hyogo Pref., park, 9
females and 12 males in 1986), Ieshima (Site 14,
Hyogo Pref., human habitation, some individuals
in 1984; 11 females and 7 males in 1986), and
Sumoto (Site 15, Hyogo Pref., orchard, 41 females
and 86 males in 1986). On the other hand, no

916

N. AS ADA

Table 1 . Summary of the collection data of Drosophila albomicans

Locality

Number of specimens

Female Male Total

Year and month of collections and collectors

Chugoku district

1. Okayama (Okayama)

6

5

11

1986

Aug.

Asada

0

2

2

1987

Aug.

Asada and Kaneko

2. Akasaka (Okayama)

4

4

8

1986

Oct.

Asada and Shiino

0

0

0

1987

Aug.

Asada and Kaneko

3. Seto (Okayama)

6

3

9

1986

Oct.

Asada and Shiino

0

0

0

1987

Aug.

Asada and Kaneko

4. Mabi (Okayama)

0

0

0

1986

May and Jun. Asada et al.

5. Ko-no-shima (Okayama)

1

0

1

1985

Oct.

Asada

19

18

37

1986

May and Jun. Asada et al.

6. Nishi-oshima (Okayama)

>26

?

>26

1986

May and Jun. Asada et al.

7. Kawakami (Okayama)

0

0

0

1985

Jul.

Asada et al.

0

0

0

1986

Jul.

Asada

8. Taisha (Shimane)

0

0

0

1986

Aug.

Fukumitsu

9. Tsuma (Shimane)

0

0

0

1985

May

Asada

Shikoku d.

10. Takamatsu (Kagawa)

14

16

30

1986

Oct.

Shiino

11. Tokushima (Tokushima)

8

6

14

1986

Oct.

Shiino

Kinki d.

12. Kobe (Hyogo)

0

0

0

1986

Oct.

Shiino

13. Akashi (Hyogo)

9

12

21

1986

Oct.

Shiino

14. Ieshima (Hyogo)

?

?

>1

1984

Oct.

Mikasa

11

7

18

1986

Oct.

Asada and Shiino

15. Sumoto (Hyogo)

41

86

127

1986

Oct.

Shiino

16. Kimiidera (Wakayama)

0

0

0

1986

Oct.

Shiino

Total

>145

159

>305

specimens were caught in some localities mostly
inland from the Inland Sea; e.g., Mabi (Site 4,
Okayama Pref., bank, 1986), Kawakami (Site 7,
Okayama Pref., grassland, 1985 and 1986), Taisha,
(Site 8, Shimane Pref., brewery, 1986), Tsuma
(Site 9, Shimane Pref., moor, 1985), Kobe (Site
12, Hyogo Pref., human habitation, 1986) and
Kimiidera (Site 16, Wakayama Pref., orchard,
1986). As mentioned above, D. albomicans has
been widening its distribution area along the coast
of the Inland Sea. The situation suggests a
possibility of the invasion of D. albomicnas in both
the Kyushu district and the western part of the
Kinki district in the near future. The sudden
appearance of D. albomicans in the mainland of
Japan, especially in the southern parts of Japan, is

quite interesting from the standpoint of population
biology; when and from where D. albomicans flies
invaded into these areas, how they can succeed to
widen their distribution areas from south to
northward, and how is their adaptive strategy to
the low-temperature northern zone, and so on.
The collection data suggested a high colonizing
ability of D. albomicans within a few years from
1984 although the first landing place is unknown.

Ieshima Islands, where the first mainland
population of D. albomicans was found, are small
islands located near Kobe and Osaka, both famous
for their international trading ports. Populations
in the Ieshima Islands hold an important key for
understanding the process of invasion of D.
albomicans into the mainland of Japan. Because

Drosophila albomicans in Japan

917

128

130

132

134

136 ° E

Amami-oshima

d

Fig. 1. Map showing the collection sites in western Japan given in Table 1. White circles
represent where D. albomicans were caught and the black ones represent areas where none
were captured.

of the extremely small size of the Ieshima Islands,
it is very unlikely that D. albomicans has inhabited
only that area for an extended time. Recent
invasion of the species may be a more reasonable
understanding. As to the route to the invasion into
Japan, two possibilities are suggested. First,
populations invaded some regions of southern
Kyushu from the southern regions such as Amami-
oshima, then widened its distribution area step-by-
step through Kyushu, western Chugoku, and the
eastern Chugoku districts to Okayama Prefecture,
or the Ieshima Islands. Second, spontaneous
invasion occurred accompanied with traffic trans¬
portation into the Osaka region directly, then the
distributed area enlarged to the neighborhood

regions. The latter possibility seems to be more
likely than the former, although field surveys in
Kyushu, western Chugoku and the Kinki districts
are necessary to arrive at any conclusions.

Similar to D. albomicans , D. simulans Sturte-
vant is also known as a species recently colonized
in the mainland of Japan [8]. D. simulans had
never been captured in Japan except in the Bonin
Islands, 27°N and 142°E, until one male was
collected in Yakushima in 1972. Since then, its
distributing area has expanded centering around
both Kyushu and the Kanto-Tokai districts [9-10].
Morphological and electrophoretic studies sug¬
gested that the newly colonized mainland popula¬
tions of D. simulans did not derive from the Bonin

918

N. AS ADA

Islands populations [11-12]. In order to estimate
the ancestral population(s) and to reveal the
process(es) of colonization of D. albomicans, in
the mainland of Japan, not only further field
surveys, but also molecular analyses are needed in
the same manner as D. simulans.

ACKNOWLEDGMENTS

The author wishes to express his many thanks for
invaluable criticisms to Profs. O. Kitagawa, Tokyo Met¬
ropolitan University, S. Ohba, Okayama University of
Science, and T. K. Watanabe, National Institute of
Genetics. He also wishes to express his hearty thanks to
Messrs. K. Mikasa, Josai Dental University and T.
Shiino, Tokyo Metropolitan University, who kindly
provided their unpublished data, and to Dr. D. M.
Kimble, Okayama University of Science for critical
constructive reading of the manuscipt. To the late Prof.
H. Ikeda, Ehime University, I dedicate this issue with my
most heartfelt gratitude.

REFERENCES

1 Wilson, F., Wheeler, M., Harget, M. and Kam-

bysellis, M. (1969) Univ. Texas Publ., No. 6918:
207-253.

Kitagawa, O., Wakahama, K. I., Fuyama, Y., Shi-
mada, Y., Takanashi, E., Hatsumi, M., Uwabo, M.
and Mita, Y. (1982) Jpn. J. Genet., 57: 113-141.
Watanabe, T. K. and Kawanishi, M. (1978) Zool.
Mag., 87: 109-116.

Watada, M., Inoue, Y. and Watanabe, T. K. (1986)
Zool. Sci., 3: 873-883.

Asada, N. and Kaneko, A. (1986) Bull. Okayama
Univ. of Science, 21A: 235-244.

Asada, N. and Kaneko, A. (1987) Bull. Okayama
Univ. of Science, 22A: 313-324.

Asada, N., Fukumitsu, T. and Kaneko, A. (1988)
Bull. Okayama Univ. of Sci., 23A: 75-90.
Watanabe, T. K. and Kawanishi, M. (1976) Proc.
Jpn. Acad., 52: 191-194.

Kikkawa, H. and Peng, R. T. (1938) Jpn. J. Zool.,
7: 507-552.

Okada, T. (1956) Systematic Study of Drosophili-
dae and Allied Families of Japan. Gihodo, Tokyo.
Watada, M., Tobari, Y. N. and Ohba, S. (1978)
Jpn. J. Genet., 53: 458.

Watada, M. and Ohba, S. (1981) Jpn. J. Genet., 56:
648-649.

2

3

4

5

6

7

8

9

10

11

12

development

Published by

the Japanese Society of Developmental Biologists

The journal is devoted to the publication of original papers dealing with any aspects of developmental
phenomena in all kinds of organisms, including plants and micro-organisms. Papers in any of the following fields
will be considered: developmental genetics, growth, morphogenesis, cellular kinetics, fertilization, cell division,
dormancy, germination, metamorphosis, regeneration and pathogenesis, at the biochemical, biophysical and
analytically morphological levels ; reports on techniques applicable to the above fields. At times reviews on subjects
selected by the editors will be published. Brief complete papers will be accepted, but not preliminary reports.

Members of the Society receive the Journal free of charge. Subscription by institutions is also welcome.

Papers in Vol. 30, No. 1. (Feburary, 1988)

31. REVIEW: C. C. Lambert and R. A. Koch: Sperm binding and penetration during ascidian
fertilization

32. REVIEW: O. Yamashita and L. S. Indrasith: Metabolic fates of yolk proteins during
embryogenesis in arthropods

33. R. Masho: Fates of animal-dorsal blastomeres of eight-cell stage Xenopus embryos vary
according to the specific patterns of the third cleavage plane

34. T. Kominami: Changes in the pattern of intercellular junctions during early embryogenesis of
the starfish, Asterias amurensis

35. T. Nishikata, I. Mita-Miyazawa and N. Satoh: Differentiation expression in blastomeres of
cleavage-arrested embryos of the ascidian, Halocynthia roretzi

36. G. Nardi and M. Cipollaro: l-Methyl-4-thiohistidine and glutathione in the developing
embryo of the sea urchin, Paracentrotus lividus

37. A. S. Suzuki and K. Harada: Prospective neural areas and their morphogenetic movements
during neural plate formation in the Xenopus embryo. II. Disposition of transplanted
ectoderm pieces of X. borealis animal cap in prospective neural areas of albino X. laevis
gastrulae

Abstracts of the papers presented at the 21st Annual Meeting of the Japanese Society of
Developmental Biologists, 1988

Development, Growth and Differentiation (ISSN 0012-1592) is published bimonthly by The Japanese Society of
Developmental Biologists, Department of Biology, School of Education, Waseda University, Tokyo 160, Japan.
1988: Volume 30. Annual subscription U. S. $ 110.00 including air speed delivery except Japan. Application to
mail at second class postage rate is pending at Jamaica, NY 11431, U. S. A.

Outside Japan: Send subscription orders and notices of change of address to Academic Press, Inc., Journal
Subscription Fulfillment Department, 6277 Sea Harbor Drive, Orlando, FL 32887, 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, 6277 Sea Harbor Drive, Orlando, FL 32887, 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-chorne, 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.

TOM Y CORPORATION SOLE AGENT

1002 SOLEIL NARIMASU BLDG.. 31-8. NARIMASU 1-CHOME.

ITABASH1-KU. TOKYO 175 JAPAN

TEL:(03)976-341 1 TLX: 027231 1 1 TOMYCO J

CABLE: TOM YSHO TOKYO FAX: (GUI G II ) (03)930-70 1 0

TOMV SEIKO CO., LTO. MANUFACTURER

2-2-12. ASAHICHO NERIMA-KU. TOKYO 176 JAPAN
TEL:(03)976-3 1 1 1

TOIV1Y

MR-1SO

HIGH SPEED
REFRIGERATED
MICRO CENTRIFUGE

MOOEL MR-1 50

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

NARISHIGE

THE ULTIMATE NAME IN MICROMANIPULATION

OUR NEW MODELS MO-102 and MO-103
MAKE PRECISION MICROMANIPULATION SO EASY!

(Photo: by courtesy of Olympus Optical CO., LTD.)

SOME FEATURES of MO-102 and MO-103:

* The manipulator head is so small that it can be mounted directly on the
microscope stage. There is no need for a bulky stand.

* Hydraulic remote control ensures totally vibration-free operation.

* 3-D movements achieved with a single joystick.

Micromanipulators Microelectrode pullers Stereotaxic instruments

NARISHIGE' NARISHIGE SCIENTIFIC INSTRUMENT
LABORATORY CO., LTD.

4-9-28, Kasuya, Setagaya-ku, Tokyo 157 JAPAN
Telephone: 03-308-8233 Telex: NARISHG J27781

(Contents continued from back cover )

Endo, Y. and T. Endo: S-100 protein-like im-
munoreactive cells in the brain-midgut en¬
docrine system of the insect Periplaneta

americana (COMMUNICATION) . 905

Kobayashi, Y. and S. Kawashima: Changes
of acetylcholinesterase activity in rat sup¬
raoptic nucleus cell bodies during water de¬
privation (COMMUNICATION) . 911

Morphology

Tsuneki, K. and M. Ouji: Absence of blood
vessels in the brain of six species of primitive
salamanders . 847

Behavior Biology

Daumae, M. and T. Kimura: Factors regulat¬
ing urination patterns in male and female

mice ( Mus musculus ) . 855

Ebino, K. Y., K. Yoshinaga, T. R. Saito and
K. W. Takahashi: Coprophagy as an innate
behavior in the mouse . 863

Ecology

Ohgushi, R., S. Yamane and S. F. Sakagami:
Ecological distribution and habitat-linked
density of colonies of stenogastrine wasps in

tropical S. E. Asia . 869

Asada,N.: Invation of Drosophila albomi-
cans to the mainland of Japan (COM¬
MUNICATION) . 915

Taxonomy

Kristensen, R. M. and Y. Shirayama: Plici-
loricus hadalis (Pliciloricidae), a new lori-
ciferan species collected from the Izu-

Ogasawara Trench, Western Pacific . 875

Uchikawa, K., K. Nakata and F. S. Lukos-
chus: Mites of the genus Myobia (Trombi-
diformes, Myobiidae) parasitic on
Apodemus mice in Korea and Japan, with
reference to their immature stages . 883

ZOOLOG

VOLUME 5 NUMBER 4

ICAL SCIENCE

AUGUST 1988

CONTENTS

REVIEWS

Gause, G. G.: Taxon-specific crystallins ..727
Kuroda, H., S. Obata, K. Takemoto, M. Ishi-
guro and H. Sato: The mechanism and
physiological function of electrical changes
during fertilization of sea urchin gametes

ORIGINAL PAPERS

Physiology

Lindstrom, M., H. Nilson and V. B. Meyer-
Rochow: Recovery from light-induced sen¬
sitivity loss in the eye of the crustacean Mysis
relicta in relation to tempertature: a study of
ERG-determined V/log I relationships and

morphology at 4°C and 14°C . 743

Naitoh, T., K. Takeuchi and I. Takabatake:
Mode of melanosome migration in tele-
ostean melanophores
Yasuyama, K., T. Kimura and T. Yamaguchi:
Musculature and innervation of the internal
reproductive organs in the male cricket, with
special reference to the projection of un¬
paired median neurons of the terminal abdo¬
minal ganglion . 767

Khin Maung Saing: Functional innervation of
the intrinsic thumb muscles of the fruit bat

Pteropus medius . 781

Srivastav, A. K. and L. Rani: Phosphocalicic
response to vitamin D3 treatment in freshwa¬
ter snake, Natrix piscator (COMMUNICA¬
TION) . 893

Cell Biology

Fujishima, M. and K. Hoshide: Light and
electron microscopic observations of Holo-
spora obtusa : A macronucleus-specific bac¬

terium of the ciliate Paramecium caudatum
. 791

Biochemistry
Kawamura, S. and M. Murakami: Light-
induced Michaelis constant increase is rapid
and inherent in cGMP phosphodiesterase in

frog rod outer segments . . 801

Hung, F. and Y. Shaoyi: Isolation and iden¬
tification of crucian ( Carassius auratus L.)
hemoglobin and its subunits . . 809

Genetics

Saitoh, M. and Y. Obara: Meiotic studies of
interracial hybrids from the wild population
of the large Japanese field mouse, Apodemus

speciosus speciosus . 815

Shimada, A., A. Shima and N. Egami:
Establishment of multiple recessive tester
stock in the fish Oryzias latipes (COM¬
MUNICATION) . . 897

Ota, H., T. Hikida, M. Matsui and M. Hasega-
wa: Karyotype of a scincid lizard, Carlia
fusca , from Guam, the Mariana Islands
(COMMUNICATION) . 901

Developmental Biology

Uchiyama, H. and T. Mizuno: Sexual
dimorphism in the genital tubercle of the
duck: Studies on the normal development
and histogenesis . 823

Endocrinology

Hirohama,T., H. Uemura, S. Nakamura and
T. Aoto: Atrial natriuretic peptide (ANP)-
immunoreactivity and ultrastructures of car-
diocytes in fish . 833

( Contents continued on inside back cover)

INDEXED IN: Issued on August 15

Current Contents/LS and AB & ES, Printed by Daigaku Printing Co., Ltd.,

Science Citation Index, Hiroshima, Japan

I SI Online Database,

CABS Database, INF OB IB

Vol. 5 No. 5

October 1988

ZOOLOGICAL

SCIENCE

An International Journal

PHYSIOLOGY

CELL and MOLECULAR BIOLOGY

GENETICS

IMMUNOLOGY

BIOCHEMISTRY

DEVELOPMENTAL BIOLOGY

REPRODUCTIVE BIOLOGY

ENDOCRINOLOGY

BEHAVIOR BIOLOGY

ENVIRONMENTAL BIOLOGY

ECOLOGY and TAXONOMY

published by Zoological Society of Japan

.... . . Business Center for Academic Societies Japan

distributed by ysp ^ Jhe Nether|an(Js

SSN 0289-0003

ZOOLOGICAL SCIENCE

The Official Journal of the Zoological Society of Japan

Editor-in-Chief:

Hideshi Kobayashi (Tokyo)
Managing Editor:

Seiichiro Kawashima (Tokyo)
Assistant Editors:

Takeo Machida (Urawa)

Sumio Takahashi (Hiroshima)
Kazuyoshi Tsutsui (Hiroshima)

The Zoological Society of Japan:
Toshin-building, Hongo 2-27-2, Bunkyo-ku,
Tokyo 113, Japan. Tel. (03) 814-5675
Officers:

President: Nobuo Egami (Tokyo)
Secretary: Hideo Namiki (Tokyo)
Treasurer: Tadakazu Ohoka (Tokyo)
Librarian: Shun-Ichi Ueno (Tokyo)

Editorial Board:

Howard A. Bern (Berkeley)
Horst Grunz (Essen)

Susumu Ishii (Tokyo)

Roger Milkman (Iowa City)
Tokindo S. Okada (Okazaki)
Hiroshi Watanabe (Shimoda)

Walter Bock (New York)
Robert B. Hill (Kingston)
Yukiaki Kuroda (Mishima)
Hiromichi Morita (Fukuoka)
Andreas Oksche (Giessen)
Mayumi Yamada (Sapporo)

Aubrey Gorbman (Seattle)
Yukio Hiramoto (Chiba)
Koscak Maruyama (Chiba)
Kazuo Moriwaki (Mishima)
Hidemi Sato (Nagoya)

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 1000 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:

Prof. C. Oguro, Managing Editor (for 1989-), 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 1988, The Zoological Society of Japan

T Publication of Zoological Science has been supported in part by a Grant-in-Aid for 1
1 Scientific Publication from the Ministry of Education, Science and Culture, Japan. J

ZOOLOGICAL SCIENCE 5: 919-924 (1988)

© 1988 Zoological Society of Japan

REVIEW

The Pathway of Sperm-Egg Interaction
in Ascidians: Biology and Chemistry

Rosaria De Santis1 and Maria Rosaria Pinto1,2

1 Department of Cell and Developmental Biology, Stazione Zoologica di Napoli,
Villa Comunale, 80121 Napoli, and 2lnstitute of Protein
Biochemistry and Enzymology, CNR, 80072
Arco Felice, Napoli, Italy

INTRODUCTION

Marine organisms have been the elective mate¬
rial for the understanding of many challenging
problems in biology. This is particularly the case
of the sea urchin, which since the second half of
19th century, has played a key role in the field of
developmental biology, equaled only by amphib¬
ians.

At the beginning of the century another model
was proposed by E. D. Conklin. He described in
detail the organization and cell-lineage of the
ascidian egg, indicating that ascidians provide one
of the best examples for the study of cell line
segregation [1], Conklin’s comprehensive work
raised a series of questions on different and also
temporally distant events, paving the way for the
studies that were to follow: along this line, T. H.
Morgan resumed studies of the early stages of
sperm-egg interaction and, in particular, that of
the genetic regulation of self-sterility in Ciona
intestinalis [2] . In fact ascidians are hermaphrodit¬
ic and, to some extent, self-sterile animals, hence
they offer a unique opportunity to study the prob¬
lem of self-nonself recognition. Morgan suggested
that in Ciona intestinalis the self-sterility barrier is
under the control of more than one allele and that
it resides in the egg investments [2]. These
observations have been confirmed and extended:
in fact it has been found that this barrier resides in

Received May 11, 1988

the vitelline coat (VC) which modulates the self¬
sterility [3-5]. Since Morgan’s observations, a
considerable amount of data from several labora¬
tories has shown that in ascidians, species specific
binding of spermatozoa occurs on the egg ex¬
tracellular coat, the VC, and that binding is a key
step in fertilization, preliminary to all the events
leading to gamete fusion [5-8]. Recently efforts
have been made to isolate the molecules of the
vitelline coat responsible for the sperm binding:
these molecules have been also characterized both
biochemically and functionally [9-11].

Few studies have been conducted on the coun¬
terpart of the VC on the spermatozoon. Most of
the data available deal with the structure and the
physiology of this cell in relation to the binding
process [12-19]. Recently studies have been
undertaken aimed at identifying the molecules
involved in the interaction and at clarifying the
mechanism underlying these events [7, 9, 10, 12,
20-23],

In this paper we review studies on the early
stages of sperm-egg interaction in the ascidians
with special attention to the biology and chemistry
of the binding. Similarities with sea urchin and
mammal fertilization are also touched upon.

THE MAIN EVENTS IN SPERM-EGG
INTERACTION

A general and simplified scheme of the events
leading to the penetration of spermatozoa into the

920

R. De Santis and M. R. Pinto

perivitelline space can be drawn from observations
carried out in Ciona intestinalis : the spermatozoon
passes between the follicle cells and binds firmly to
the VC. It is activated, it undergoes the acrosome
reaction and then penetrates through the VC.

Even though follicle cells are the first barrier the
spermatozoon encounters in approaching the egg,
they do not seem to be necessary for a successful
interaction [6], A chemotactic role for the follicle
cells has been proposed by Miller, who examined
several ascidian species [23, 24]. This observation
has been confirmed in Phallusia and in Asidia
nigra , in which removal of the follicle cells results
in a slight reduction of the fertilization rate [7, 25].
So far, only in the Japanese ascidian Halocynthia
roretzi the follicle cells seem to be strictly neces¬
sary for fertilization [3].

The spermatozoon reaches the VC and within a
few minutes binds firmly and species-specifically to
the outer surface of the coat by the plasma mem¬
brane of its tip. Binding occurs at the whole
surface of the VC not covered by the follicle cells,
apparently without any polarity [5-8, 18]. In
ascidians binding is neither preceded by a loose
and nonspecific “attachment”, as described in the
mouse, nor followed by a cortical reaction, as in
mammals and echinoderms.

Upon binding, the spermatozoon undergoes a
series of changes, that can be referred to as sperm
“reaction” or “activation” [16, 26]: these consist of
the swelling and translocation of the cylindrical
mitochondrion towards the tail and of concomitant
physiological changes. The reaction culminates in
the exocytosis of the vesicle/s present at the sperm
tip, which is interpreted as an acrosome reaction
[6, 7, 19]. The acrosome reaction leads to the
penetration of the spermatozoon through the VC.
Trypsin-like and chymotrypsin-like proteases con¬
tained in the vesicles seem to be the lytic agents
that allow the spermatozoon to penetrate through
the VC [7, 27-30].

The spermatozoon passes between the test cells
apparently unimpeded by these cells. Since no tool
has been found whereby the activity of these cells
can be discriminated from that of the VC, only
indirect experimental evidence is available.
Hence, even though test cells seem not to have a
primary role, their participation in the interaction

cannot be ruled out [3].

Once the spermatozoon has crossed this last
barrier, it makes contact with the egg plasma
membrane, on which fusion occurs at the vegetal
pole [4, 31].

SPERM-EGG BINDING

Most of the data about sperm-egg interaction in
ascidians concern the binding of the spermatozoa
to the VC, a preliminary and essential step for
species-specific recognition between the gametes.

Binding is established between the tip of sperm
head and fibrils emerging from the outer surface of
the VC [6, 7]. Screening of the surface of the VC
with lectins and the effect of correspondent com¬
petitive sugars on binding and fertilization re¬
vealed that in Ciona intestinalis, egg fucosyl sites
present on the fibrils have a key role in sperm
binding [32, 33]. In Phallusia mammillata and in
Ascidia nigra, N-acetylglucosamine residues pres¬
ent on the VC have the same function in sperm
binding [7, 8, 34]. The biochemical and functional
characterizations of the molecules involved in this
interaction have been conducted mainly on Ciona
intestinalis. A glycoprotein complex with sperm
receptor activity has been isolated from the VC of
the mature egg. SDS-PAGE analysis of this com¬
plex followed by incubation of the gel with 3H-
Lotus tetragonolobus agglutinin (LTA) revealed
the presence of five fucosyl-containing glycopro¬
teins (FP). FP exhibited the properties expected
for the receptor: i.e. they inhibited binding of the
spermatozoa to the VC and fertilization. Further¬
more, spermatozoa incubated with FP were acti¬
vated and triggered into the acrosome reaction [9].
FP are synthesized by the oocyte during the vitel¬
logenic stages of oogenesis [35]. This demon¬
strates that the glycoprotein complex with sperm
receptor activity, on which the fine tuning and in
fact the success of fertilization depend, is synthe¬
sized under the control of the oocyte genome.

These findings led to a further biochemical and
functional analysis of FP purified by affinity chro¬
matography from the ovary [10], The separation
by gel filtration of the fucosyl glycoproteins into
the units identified by SDS-PAGE, always resulted
in the elution of FP with the void volume, thus

Sperm-Egg Interaction in Ascidians

921

indicating that the sperm receptor on the VC is a
highly complex glycocon jugate with an extraordi¬
narily high apparent molecular weight (>107 D)
and a protein-carbohydrate ratio of 2: 1. FP inhibit
the binding of the spermatozoa to the VC and
induce sperm activation and the acrosome reaction
[10]. Exhaustive proteolytic digestion of FP yields
high molecular weight glycopeptides (>4xl05),
which contain N-acetylgalactosamine, fucose,
galactose and rhamnose. These glycopeptides,
when compared with the intact FP, have much
lower capacity to inhibit sperm binding, but they
do not induce sperm activation and, in conse¬
quence, the acrosome reaction [10]. These
findings raise the question of the involvement of
the polypeptide component in the sperm-egg bind¬
ing process, either directly or as a carrier of the
carbohydrate chains. In the latter case, the
polypeptide chain could provide a backbone for
the specific assembly or conformation of the car¬
bohydrate component, thus allowing the most
effective molecular match with the counterpart on
the spermatozoon [10].

Hence, FP share many of the characteristics of
both the sperm receptors of the egg of Strongy-
locentrotus purpuratus [36] and of the zona pelluci-
da of the mouse [37]. In the first case, in fact, the
receptor for the spermatozoa is a high molecular
weight proteoglycan-like molecule (greater than
107) that is responsible for the species specific
binding of the spermatozoa to the egg. As to the
biological activity, FP are more similar to ZP3, the
zona pellucida glycoprotein of the mouse egg,
which serves as both receptor for spermatozoa and
inducer of the sperm acrosome reaction. Also in
this case the sperm receptor activity is dependent
only on its carbohydrate component, whereas the
whole complex is responsible for the acrosome
reaction-inducing activity [37].

While the sperm receptors on the egg have been
isolated and extensively characterized in several
systems, their counterparts on the spermatozoon
and the mechanisms underlying the sperm-egg
interaction have yet to be elucidated. In the sea
urchin, upon contact with the egg jelly through a
sperm plasma membrane glycoprotein [38], the
spermatozoon undergoes the acrosome reaction,
thus exposing “bindin”, a protein associated with

the inner acrosomal membrane [39]. Bindin in¬
teracts species-specifically with the glycoprotein
sperm receptor on the VC by a lectin-
polysaccharide type of interaction [40], In mouse,
it has been suggested that different kinds of pro¬
teins associated with the sperm plasma membrane
participate in the sperm-zona interaction [37]. A
galactosyltransferase localized at the sperm surface
is at least one of the components involved in the
sperm binding to the egg zona pellucida through an
enzyme-substrate mechanism [41, 42].

Also in Ciona intestinalis there is evidence that
different proteins present on the surface of the
spermatozoa are involved in the sperm-egg in¬
teraction. Concanavalin A-binding sites are pres¬
ent at the tip of the sperm head: these glycopro¬
teins have been isolated by affinity chromatogra¬
phy from the tip of the sperm head and they
proved to be active in the inhibition of binding
[20].

Furthermore, it has been suggested that an
a-L-fucosidase, isolated from Ciona spermatozoa
is involved in the very early stages of binding [21,
43]. This enzyme would interact with the fucosyl-
glycoproteins of the VC by forming an enzyme-
substrate complex. Whether this interaction pro¬
ceeds with the hydrolysis of fucose, has yet to be
elucidated. In Phallusia mammillata where ter¬
minal N-acetylglucosamine is functional in sperm
binding [7, 34], it has been found that a /J-D-N-
acetylglucosaminidase retains the highest activity
among sperm glycosidases [21, 43], These findings
support the hypothesis that in different systems
glycoproteins of the egg envelope and carbohy¬
drate-binding proteins of the spermatozoon medi¬
ate the sperm-egg recognition and binding. These
proteins can be roughly classified into two groups:
enzymes, such as the galactosyltransferase of
mouse and the a-L-fucosidase of ascidians, and
lectin-like proteins, such as the “bindin” of the sea
urchin spermatozoon.

Indeed, the sperm-egg interaction proceeds
through a far more complicated series of con¬
nected and ordered reactions, that involve differ¬
ent molecules and/or different functional sites on
the same molecule [37, 44, 45]. In fact, in mouse
there is evidence that, besides galactosyltrans¬
ferase, also a trypsin-like proteinase may be in-

922

R. De Santis and M. R. Pinto

volved in binding [44, 46]. Furthermore, a fucose-
binding protein isolated from boar spermatozoa
has been demonstrated to be identical with the
sperm proteinase acrosin. This molecule combines
specific proteolytic activity with zona- and car¬
bohydrate-affinity properties, thus suggesting that
the lectin-like activity of the acrosin directs the
proteolytic activity to its structural target [45].

SPERM ACTIVATION, ACROSOME
REACTION AND PENETRATION

Upon binding to the VC ascidian spermatozoon
undergoes a number of morphological and phys¬
iological changes called “sperm reaction” [16, 26].
Spermatozoa lack a midpiece and they have a
single cylindrical mitochondrion alongside the nu¬
cleus in the head. When the spermatozoon binds
to the VC by the tip of the head, the mitochon¬
drion swells, becomes spherical and starts sliding
towards the tail. Occasionally, the mitochondrion
may even be cast off. This reaction has been also
observed in spermatozoa bound to isolated VC
[16] or can be induced by the calcium ionophores
[14, 22]. Since the presence of actin and myosin
has been revealed by indirect immunofluorescence
in the spermatozoa of several species of ascidians,
it has been proposed that actin-myosin sliding is
responsible for mitochondrion translocation [18].
In the sperm traversing the VC and the peri-
vitelline space, the membrane of the spermato¬
zoon is anchored to the VC by the outer surface
overlying the mitochondrion and to the mitochon¬
drion by its inner surface. As the mitochondrion
moves along the tail, the sperm is driven across the
perivitelline space. When the mitochondrion
reaches the tip of the tail, it is released.

Sperm reaction is accompanied by a release of
H+; this causes intracellular alkalinization, which
in turn increases permeability to calcium [22, 47].
Hence, the onset of binding might trigger the
activation of calcium channels that promotes the
uptake of Ca2+ into spermatozoa and the acro-
some reaction.

Changes at the tip of the spermatozoa referred
to as an acrosome reaction have been described
only in Ciona intestinalis and in Phallusia mammil-
lata [6, 7, 19]. The reaction proceeds through the

exocytosis of one or some acrosomal vesicles of the
spermatozoa just about to penetrate the VC. It
can be inferred that the acrosomal vesicles contain
lysins necessary for the penetration of the sperma¬
tozoa through the VC [19]. In fact, a preliminary
screening to detect protease activity involved in
sperm penetration produced evidence indicating
that in Pleurogona both chymotrypsin- and tryp-
sin-like activities are necessary, while in Enterogo-
na only the chymotrypsin-like activity is required
for the penetration through the VC [30, 43, 48].
More recently, two types of trypsin-like proteases
have been purified from spermatozoa of Halo-
cynthia roretzi [28]. Whereas one of the enzymes
showed properties closely related to those of
mammalian acrosin, the other seemed to be a
novel type of sperm acrosin-like enzyme with very
strict substrate specificity. These enzymes, named
ascidian acrosin and spermosin respectively, ex¬
erted lytic activity on the coat only together with
chymotrypsin-like enzyme(s). It was suggested
that ascidian spermosin, acrosin and chymotryp¬
sin-like enzyme are involved in a “lysin system”

[28] : the timing of action of the three sperm
proteases was examined by adding specific pro¬
tease inhibitors at various times after insemination

[29] . It seems likely that spermosin and the
chymotrypsin-like enzyme function at an early
stage of the process of penetration through the egg
investment, while acrosin functions at a later stage
[29].

Concerning the number of spermatozoa that
cross the VC, it is generally agreed that rare
acrosome-reacted spermatozoa can be observed on
the VC and that “few spermatozoa” can be found
within the perivitelline space [4-7]. The sugges¬
tion that the VC acts as a barrier for sperm
penetration and the apparent discrepancy between
the number of spermatozoa bound and those
penetrating the VC, indicated that in order for the
attachment to be followed by the acrosome reac¬
tion there must be a precise “molecular match”
between receptors on the egg and the counterpart
on the spermatozoon [9].

In ascidians, morphological differences of the
VC, as a consequence of the sperm penetration,
have never been observed. Only in the Japanese
ascidian Halocynthia roretzi is there an expansion

Sperm-Egg Interaction in Ascidians

923

of the perivitelline space following fertilization,
and there is evidence for the participation of a
trypsin-like enzyme in this change [49]. By analo¬
gy with the role of the trypsin-like enzyme of the
sea urchin, this enzyme may function in a limited
proteolysis of the VC, which becomes more resist¬
ant to proteolysis than the VC of the unfertilized
eggs.

A block to polyspermy that requires Na+ has
been found in Ascidia nigra [8] ; this block involves
also the release of a protein, either an enzyme or a
soluble lectin, that modifies VC N-acetylglucos-
amine residues required for sperm binding. This in
turn causes a rapid decline in the number of sperm
bound.

Information about the pathway of the spermato¬
zoon after its passage through the VC is still
limited. Most of the data on the steps following
penetration concern the site of fusion of the sper¬
matozoon with the egg plasma membrane [4, 31].
Spermatozoa come in contact with the egg plasma
membrane and can be found randomly distributed
along the animal-vegetal axis of the egg, except at
the animal pole [4], At the time ooplasmic seg¬
regation is nearly completed, spermatozoa are
localized only at the vegetal pole, where, after
fusion, the male pronucleus is located [1, 4].
Sperm-egg fusion occurs between the plasma
membrane of the postacrosomal region of the
sperm head and the egg plasma membrane [7].

CONCLUDING REMARKS

The process whereby gametes recognize and
adhere to each other is one of the most challenging
problems in cell biology. The study of the mecha¬
nisms of this interaction showed that the informa¬
tion for establishing species-specific recognition
and binding lies in the expression and organization
of molecules at the surface of both gametes. In this
respect, gamete interaction, although retaining
peculiar features, can be regarded as any other
cell-cell interaction system [41-52].

In ascidians, as in the vast majority of animals,
sperm receptors are glycoproteins assembled in an
extracellular envelope, which is a vanguard for the
interaction with the spermatozoa. In these organ¬
isms the involvement of the carbohydrate and the

protein components of the sperm receptors in the
binding has to be further elucidated. Concerning
the molecular mechanism which regulates binding,
a preliminary and essential step of the interaction,
there is evidence that glycosidases on the sperma¬
tozoa interact with the glycoproteins of the VC by
way of an enzyme-substrate mechanism. This
mechanism has also been proposed in mouse ferti¬
lization and in other cell-cell interaction systems.
However, this seems to be an oversimplification of
the binding process, in the pathway of which other
as yet unidentified molecules may take part. As
emerges from this review, knowledge of the com¬
plex cascade of events that starts with gamete
recognition and leads to the fusion of the pronu¬
clei, has grown in the last decade: however, a
fundamental support to the understanding of this
process will certainly come from new technologies
of molecular biology.

In this context, also one of the most stimulating,
and still unsolved, problems in ascidian fertiliza¬
tion, namely the control of self-sterility, will cer¬
tainly benefit from these technologies. In fact,
studies with the colonial ascidian Botryllus have
suggested that the fine tuning of the gametic
self-incompatibility is controlled by genes linked or
identical to those that regulate the colony fusibil¬
ity. It seems likely that gametic and somatic
self-recognition genes in protochordates represent
ancient functions of primitive major histocompati¬
bility complex genes [53]. A complete understand¬
ing of this problem still awaits much more informa¬
tion. What is evident is that in this respect
ascidians have proved to be a good example of how
the choice of a model can answer more general
questions.

REFERENCES

1 Conklin, E. G. (1905) J. Acad. Natl. Sci. Phila., 13:
1-119.

2 Morgan, T. H. (1944) J. Exp. Zool., 95: 37-59.

3 Fuke, M. T. (1983) Wilhelm Roux’s Arch. Dev.
Biol., 192: 347-352.

4 Kawamura, K., Fujita, H. and Nakauchi, M. (1987)
Dev. Growth Differ., 29: 627-642.

5 Rosati, F. and De Santis, R. (1978) Exp. Cell Res.,
112: 111-119.

6 DeSantis, R., Jamunno, G. and Rosati, F. (1980)

924

R. De Santis and M. R. Pinto

Dev. Biol., 74: 490-499.

7 Honegger, T. G. (1986) Dev. Biol., 118: 118-128.

8 Lambert, C. C. (1986) Dev. Biol., 116: 168-173.

9 De Santis, R., Pinto, M. R., Cotelli, F., Rosati, F.,
Monroy, A. and D’Alessio, G. (1983) Exp. Cell
Res., 148: 508-513.

10 De Santis, R. and Pinto, M. R. (1987) Dev. Growth
Differ., 29: 617-625.

11 Pinto, M. R., DeSantis, R., D’Alessio, G. and
Rosati, F. (1981) Exp. Cell Res., 132: 289-295.

12 Casazza, G., De Santis, R. and Pinto, M. R. (1984)
Exp. Cell Res., 155: 261-266.

13 Cotelli, F., DeSantis, R., Rosati, F. and Monroy,
A. (1980) Dev. Growth Differ., 22: 561-569.

14 De Santis, R., Hoshi, M., Cotelli, F. and Pinto, M.
R. (1983) In “The Sperm Cell”. Ed. by J. Andre,
Martinus Nijhoff Publ., The Hague, pp. 274-276.

15 Fukumoto, M. (1984) Ultrastruct. Res., 87: 252-
262.

16 Lambert, C. C. (1982) Am. Zool., 22: 841-849.

17 Lambert, C. C. and Lambert, G. (1983) Gamete
Res., 8: 295-307.

18 Lambert, C. C. and Lambert, G. (1984) Dev. Biol.,
106: 307-314.

19 Rosati, F., Pinto, M. R. and Casazza, G. (1985)
Gamete Res., 11: 379-389.

20 Casazza, G., De Santis, R. and Pinto, M. R. (1988)
Dev. Growth Differ., 30: 147-158.

21 Hoshi, M., DeSantis, R., Pinto, M. R., Cotelli, F.
and Rosati, F. (1985) Zool. Sci., 2: 65-69.

22 Lambert, C. C. and Lambert, G. (1981) Dev. Biol.,
88: 312-317.

23 Miller, R. L. (1975) Nature, 254: 244-245.

24 Miller, R. L. (1982) Am. Zool., 22: 827-840.

25 Lambert, C. C. and Lambert, G. (1981) J. Exp.
Zool., 217: 291-295.

26 Lambert, C. C. and Epel, D. (1979) Dev. Biol., 69:
296-304.

27 Hoshi, M., Numakunai, T. and Sawada, H. (1981).
Dev. Biol., 86, 117-121.

28 Sawada, H., Yokosawa, H. and Ishii, S. (1984) J.
Biol. Chem., 259: 2900-2904.

29 Sawada, H., Yokosawa, H., Numakunai, T. and
Ishii, S. (1986) Experientia, 42: 74-75.

30 Yokosawa, H., Numakunai, T., Murao, S. and
Ishii, S. (1987) Experientia, 43: 925-927.

31 Sawada, T. and Osanai, K. (1981) Wilhelm Roux’s
Arch. Dev. Biol., 74: 490-499.

Rosati, F., De Santis, R. and Monroy, A. (1978)
Exp. Cell Res., 116: 419-427.

Rosati, F. and De Santis, R. (1980) Nature, 283:
762-764.

Honegger, T. G. (1982) Exp. Cell Res., 138: 446-
451.

Rosati, F., Cotelli, F., DeSantis, R., Monroy, A.
and Pinto, M. R. (1982) Proc. Natl. Acad. Sci.
USA, 79: 1908-1911.

Rossignol, D. P., Earles, B. J., Decker, J. L. and
Lennarz, W. J. (1984) Dev. Biol., 104: 308-321.
Wassarman, P. M. (1987) Science, 235: 553-560.
Podell, S. B. and Vacquier, V. D. (1985) J. Biol.
Chem., 260: 2715-2718.

Vacquier, V. D. and Moy, G. W. (1977) Proc. Natl.
Acad. Sci. USA, 74: 2456-2460.

Glabe, C. G., Grabel, L. B., Vacquier, V. D. and
Rosen, S. D. (1982) J. Cell Biol., 94: 123-128.
Lopez, L. C., Bayna, E. M., Litoff, D., Sharper, N.
L., Sharper, L. H. and Shur, B. D. (1985) J. Cell
Biol., 101: 1501-1510.

Shur, B. D. and Hall, N. G. (1982) J. Cell Biol., 95:
574-579.

Hoshi, M. (1984) In “Advances in Invertebrate
Reproduction”. Ed. by W. Engels, W. H. Clark,
Jr., A. Fischer, P. J. W. Olive and D. F. Went,
Elsevier Science Publ. B. V., Vol. 3, pp. 27-40.
Florman, H., Saling, P. M. and Storey, B. T. (1982)
J. Androl., 3: 373-381.

Topfer-Petersen, E. and Henschen, A. (1987)
FEBS Letters, 226: 38-42.

Saling, P. M. (1981) Proc. Natl. Acad. Sci. USA,
78: 6231-6235.

Lambert, C. C. (1982) Dev. Biol., 91: 257-262.
Sawada, H., Miura, M., Yokosawa, H. and Ishii, S.
(1984) Biochem. Biophys. Res. Commun., 121:
598-604.

Sawada, H., Kawahigashi, M., Yokosawa, H. and
Ishii, S. (1985) J. Biol. Chem., 260: 15694-15698.
Adair, W. S., Monk, B. C., Cohen, R., Hwang, C.
and Goodenough, U. W. (1982) J. Biol. Chem.,
257: 4593-4602.

Humphreys, S., Humphreys, J. and Sano, J. (1977)
J. Supramol. Struct., 7: 339-351.

Springer, W. R. and Barondes, S. H. (1982) Proc.
Natl. Acad. Sci. USA, 79: 6561-6565.

Scofield, V. L. (1985) In “MBL Lectures in Biolo¬
gy”, Ed. by H. O. Halvorson and A. Monroy, Alan
R. Liss Inc., New York, Vol. 7, pp. 213-220.

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

ZOOLOGICAL SCIENCE 5: 925-937 (1988)

© 1988 Zoological Society of Japan

REVIEW

Neuroendocrine Control of Anuran Anterior Preoptic
Neurons and Initiation of Mating Behavior

Akihisa Urano* 1

Department of Regulation Biology , Faculty of Science,
Saitama University, Urawa, Saitama 338, Japan

INTRODUCTION

The preoptic area plays an important role in the
evocation of sexual behavior in many vertebrate
species [1-6]. In the anuran brain, the anterior
part of the preoptic nucleus (APON) is considered
to be the center for triggering male mate calling
behavior. After careful re-examination of the
lesion study by Aronson and Noble [7] on the brain
of the leopard frog Rana pipiens , Schmidt [2, 8]
proposed that the APON initiates an organizing
activity of mating call patterns in the brain stem
“call center” which is composed of the pre-
trigeminal nucleus and the laryngeal motor
neurons. The same system may function in the
brain of the Japanese toad, Bufo japonicus [9, 10].
The APON is further concerned with female
orientation to conspecific mating calls in R. pipiens

pi].

Anuran mating behavior is controlled not only
by neural input signals but also by various hormo¬
nal signals. Testosterone seems to be not a sole,
but a crucial hormonal factor in the initiation of
male mate calling behavior [12-14], In gravid
female Rana pipiens , vasotocin, an anuran
neurohypophysial hormone, suppressed female re¬
lease calling behavior, and in turn elevated sexual
receptivity [15]. Luteinizing hormone-releasing
hormone (LHRH) also increased sexual receptiv¬
ity in female Xenopus laevis [16].

Received May 27, 1988

1 Present address: Ocean Research Institute, University
of Tokyo, Minamidai, Nakano-ku, Tokyo 164, Japan.

The amphibian brain shows a fundamental organ¬
izational pattern common to the structure of the
vertebrate brain [17]. Clarification of neuroendo¬
crine control of mating behavior in amphibians is
thus important for understanding phylogenetically
fundamental control mehanisms of sexual be¬
havior. In this paper, I first describe the morpho¬
logical characteristics of the preoptic nucleus
(PON), and then review recent studies concerning
the neural and hormonal control mechanisms for
the initiation of seasonal reproductive behavior in
anuran amphibians.

CYTOARCHITECTURE OF THE PON

Subnuclear Organization

The PON is a neuronal cell mass which sur¬
rounds the preoptic recess, and is considered to be
an important neuroendocrine center. It is easily
divisible into anterior and posterior halves by a
relatively cell-poor zone along the sulcus preopti¬
cus (Fig. 1), and each half is composed of several
subnuclei, some of which include many neuro¬
secretory neurons. Recent physiological studies
indicate that these subnuclei are functionally dis¬
tinctive [18].

In the Japanese toad, the PON can be divided
into seven subnuclei: the anterior part of the PON
(APON), the dorsal and the ventral periventricu¬
lar parts, the dorsal and the ventral magnocellular
parts ( dmc and vmc), the suprachiasmatic part,
and the posterior part of the PON (PPON) [18].
The APON is composed of a rather compact cell

926

A. Urano

Fig. 1. Diagram of the parasagittal midplain of the toad
diencephalon. Note that the preoptic nucleus which
locates antero-dorsal to the optic chiasma (OC) is
divisible into the anterior part (APON) and the
posterior part (PPON). AC, anterior commissure;
EP, epiphysis; LT, lamina terminalis; ME, median
eminence; NID, nucleus infundibularis dorsalis;
NIV, nucleus infundibularis ventralis; OL, optic
lobe; PD, pars distalis of hypophysis; PI, pars inter¬
media of hypophysis; PN, pars nervosa of hypoph¬
ysis; TEG, tegmentum mesencephali; TEL, telen¬
cephalon; IIIv, third ventricle, (from Urano and
Ishihara [110])

mass surrounding the anterior portion of the
preoptic recess in the toad brain, while, in many
ranid species, APON neurons are well organized
in a laminar pattern [19-21]. The dmc and the vmc
consist of magnocellular neurosecretory neurons
which contain immunoreactive (ir) vasotocin [18,
22], mesotocin [22], somatostatin [23] and other
peptides [24]. The toad suprachiasmatic part in¬

cludes small ir-vasotocin neurons and receives
direct retinal innervation (Shimotoso and Urano,
unpublished). Synapses between the optic termi¬
nals and the PON neurons were observed in Rana
temporaria [25]. The PPON, the name of which
follows nomenclatorial usage in the brain atlas of
Rana pipiens [21], is a common preoptic structure
in many anuran species.

Sexual Dimorphism

Differences in patterns of sexual behavior be¬
tween males and females are partly due to differ¬
ences between male and female neuronal circuit¬
ries, i. e., sexual dimorphism of nervous systems.
As in songbirds [26], the Japanese quail [27], and
some mammalian species [28-31], the brain loci
concerning male mate calling behavior showed
sexual dimorphism in Xenopus [32], Rana pipiens
[33] and Bufo japonicus [34]. As for the toad
PON, the nuclear volume of the APON in the
male was significantly larger than that in the
female (male/female, 1.25-1.39). The mean of the
total cell number in the male APON was 1.20
times that of the female, while the cell sizes of
male and female APON neurons were in the same
range [35]. These results suggest that the sexual
difference in the APON volumes was caused by a
greater cell number in the male APON. In addi¬
tion, the amygdala pars medialis (Am), which
forms a morphological and probably functional
complex with the APON [36], showed a similar

Table 1. Volumes (Xl0-3 mm3, mean + S.E.) of the anterior part of the preoptic nucleus (APON)
and the amygdala pars medialis (Am) in 1-year old toadlets, and hibernating and post-breeding
adult toads

Subnuclei

Male

Female

M/F

APON

1-year old toad

103.6±4.5

98.1+3.2

1.06

hibernating

311.5 + 23.0

224.1 + 14.2

1.39*

post-breeding

250.0 + 12.3

199.3 + 9.7

1.25**

Am

1-year old toad

45.9 + 2.1

35.5 + 3.1

1.30*

hibernating

153.8 + 9.5

90.1 + 4.2

1.71**

post-breeding

110.1 + 8.4

58.9 + 3.2

1.87**

*, P<0.02; ** P<0.005 by the t-test.

M/F, the ratio of the nuclear volume in the male to the female.

Mating Behavior and Preoptic Neurons

927

sexual difference. Ontogenetically, the sexual
difference in the Am volumes appeared in yearling
toadlets, preceding the development of the APON
(Table 1).

The sexual dimorphism in the APON seems to
be a fundamental part of vertebrate brain structure
in general, since the preoptic area is sexually
dimorphic in the quail [27], the rat [30] and the
macaque monkey [29].

MATE CALLING TRIGGER CENTER

Location in the APON

Mate calling can be induced by electrical stim¬
ulation of the APON. In freely moving Rana
pipiens, the effective sites for electrical evocation
of calling were located in the rostral extreme of the
PON [37]. Using isolated pipiens brain prepara¬
tions in which neural correlates of mate calling can
be recorded, the area most sensitive to electrical
stimulation was localized in the ventral half of the
APON [38]. The ventrolateral border of the
preoptic gray was much more sensitive than the
ventricular surface. Stimulation of the same locus
was also effective in inducing electrical activity

correlating to mate calling in in situ perfused brains
of Japanese toads (Fig. 2).

Scanning Electron Microscopy of the Ventricular
Surface

In the anuran brain, the ventricular wall of the
dorsal part of the APON is ciliated as are the
surfaces of the ventricular walls in many other
brain loci. However, the wall of the ventral part is
only sparsely ciliated; instead, it is studded with
numerous large bulbous protrusions which are
large cytoplasmic extensions of ependymal cells
and intraventricular dendritic end bulbs of secre¬
tory neurons [36, 39-41], This area includes
almost all the portions of the ventricular wall of the
rostro ventral part of the APON, and corresponds
topographically to the area mentioned above that
is most sensitive to the initiation of calling be¬
havior by electrical stimulation.

Steroid Hormone Accumulating Neurons

The APON has been considered to be a sex
steroid-sensitive center that triggers sexual be¬
havior in anurans. Autoradiographic studies of the
brains in Xenopus laevis [42, 43] and Rana pipiens
[44] showed that many APON neurons can

0.2 mA

— T*'-' I 50 pV

0.6 mA

20 jjV

1 mA 1 5

Fig. 2. Neural correlates of mate calling induced by electric stimulation (2 ms biphasic pulse, 50
Hz for 1 sec) of the APON through a fine bipolar electrode in a perfused toad brain.
Electrical activity correlating mate calling was recorded at the root of the vagus nerve (n.X,
upper trace) and the hyoglossal muscle (m. hg, lower trace). Magnitudes of the neural
correlates increased when the intensity of stimulation was elevated from 0.2 mA to 0.6 and
1.0 mA.

928

A. Urano

accumulate sex steroid hormones. Such neurons
were localized throughout the dorsal-ventral ex¬
tent in the rostral part of the APON, but were
found ventrally in the caudal part. This distribu¬
tional pattern of sex steroid-accumulating APON
neurons corresponds to the localization pattern of
the ventricular bulbous protrusions. In mammals,
the medial preoptic nucleus, which is the homo-
logue of the APON, includes many sex steroid-
accumulating neurons [45, 46].

Intracranial implantation of testosterone into or
near the APON enhanced the incidence of mate
calling which was evoked acoustically by play-back
of tape-recorded conspecific mating calls in Rana
pipiens [14]. In mammals, steroid hormones are
reported to excite electrical activity of central
neurons [47-50]. At present, it is not clear
whether testosterone can modulate electrical activ¬
ity of APON neurons in the amphibian brain due
to the lack of any experimental data.

Projection of the APON to the Brain Stem Call
Center

Mate calling behavior and its neural correlates
could be evoked even after massive ablations or
lesions to brain areas including almost all of the
telencephalon, the dorsal thalamus, the torus
semicircularis, the dorsal part of the isthmic nu¬
cleus and the infundibulum [38, 51]. These results
indicate the presence of pathways at the ventro¬
lateral border of the central gray from the APON
to the isthmo-trigeminal tegmentum.

The retrograde axonal transport study in which
afferents to laryngeal motor neurons were traced
in the brain of Xenopus laevis showed the presence
of direct projection of APON neurons to the
pre-trigeminal nucleus of the dorsal tegmental area
(DTAM). DTAM neurons send efferent fibers to
laryngeal motor neurons [52]. The projection from
the APON to the DTAM is less flourishing in the
female than in the male.

HUMORAL SIGNAL RECEPTIVITY

As is described above, the APON from which
efferent fibers project to the brain stem call center
is believed to be the androgen-sensitive center for
male mate calling behavior. An electrophysiolo-

gical study of this locus in Rana pipiens showed the
presence of neurons responsive to auditory stim¬
ulation by playback of conspecific mating calls, and
injections of pituitary homogenate significantly
increased the percentage of these units excited by
the calls [53], Activity of APON neurons thus can
be modulated by humoral signals. A Golgi-
electron microscopic study revealed that a portion
of the APON neurons have the proper anatomical
features for detecting humoral signals (Fig. 3) [36,
54].

Fig. 3. Diagram showing CSF-contacting and BC-
contacting neurons in the APON. These neurons
have the proper morphological features to receive
both humoral and neuronal synaptic, either chemi¬
cal or electric, input signals, be, blood capillary; EP,
ependymal layer; Med, medial gray of the APON;
Lat, lateral white of the APON; POR, preoptic
recess, (from Urano and Ishihara [110])

Cerebrospinal Fluid (CSF) -Contacting Neurons

In many anuran species, the APON, especially
its ventral half, contains CSF-contacting neurons
[36, 39-41]. These neurons were rapid-Golgi
stained and gold-toned, and then were examined
by scanning electron microscopy after removal of
surrounding tissues with hydrochloric acid and
collagenase. The somata of stripped neurons bear
debris of nerve terminals on their surfaces, sug¬
gesting that the neuronal activity of these cells can
be modulated by various synaptic input signals.

Processes of CSF-contacting neurons, probably
dendritic, protrude into the preoptic recess. The
dendrites projecting into the preoptic recess from
preoptic neurosecretory cells can be equipped to
serve both secretory and sensory functions [55].
The CSF-ventricular system is thought to distri-

Mating Behavior and Preoptic Neurons

929

bute biologically-active hormonal substances with¬
in the brain, since many researchers have found
various hormones in the CSF, such as LHRH and
thyrotropin-releasing hormone (TRH) [56], oxyto¬
cin and vasopressin [57], and melatonin [58]. The
concentrations of these hormones varied according
to various physiological statuses. Further, single
intraventricular injections of LHRH and TRH
increased the amplitudes and frequency of elec-
troencephalographic (EEG) activity recorded
from the brain of a hibernating Japanese toad (Fig.
4) [59]. The effective dose of 1 jug needed for
enhancement of EEG activity through a single
intraventricular injection of LHRH or TRH was
much less than that needed for systemic injections.
It is thus possible that the CSF-contacting neurons

YE 63 2

85/4/3

Ji_n_

T LHRH 1 m9 icv.

JlL

JU LJLIL

JLIL

12 16 20 0 4 8

Dfl 4/8 T Saline 1 pi icv

'!]_n _ IS _ iul

IV

m-

n-

i->-

4/9

8 12 16 20

J u

6 ' ' ' 4 ' ' ' 8

jltt juuuuulI TlT1

PVl

8 12 16 20 0 4 8

Fig. 4. Effects of intraventricularly injected LHRH and
TRH on EEG activity of a hibernating Japanese
toad. According to amplitudes and frequency, EEG
activity was categorized into 4 levels: I, resting; II,
awake; III, active; and IV, very active. Each level
corresponds well to a behavioral state. Note that
both LHRH and TRH induced dual fast and slow
enhancements of EEG activity. Although the EEG
record is not shown, the pattern of LHRH-induced
fast response includes highly synchronized bursting
waves that were not observed upon injection of
control saline. (Fujita, thesis, Saitama University)

whose dendritic processes protrude into the preop¬
tic recess detect changes in ventricular hormonal
status and motivate the neuronal circuitry in pre¬
paration for mating behavior in pre-breeding
anurans.

Blood Capillary (BC) -Contacting Neurons

The presence of BC-contacting neurons is in¬
compatible with the general concept of the rela¬
tions between brain neurons and capillaries.
Blood capillaries in the vertebrate brain are gener¬
ally surrounded by astrocytic endfeet with an in¬
tervening basement membrane, so that brain
neurons, even fish hypothalamic neurosecretory
cells, are separated from the vascular endothelium
[60]. Nonetheless, neurosecretory cells which
directly come into contact with blood capillaries
were shown in the toad preoptic nucleus [61].
Recently, it was found that a considerable number
of peptidergic neurons come into contact with
blood capillaries with only an intervening base¬
ment membrane in the APON of both the bullfrog
and the Japanese toad [36].

BC-contacting neurons send their dendrites
laterad toward the preoptic white matter.
Although arborization is rather poor, the dendrites
usually bifurcate several times and form dendritic
fields. There, many axon terminals form synapses
on the dendritic spines of these neurons. It is
highly probable that APON neurons receive the
input signals of the afferent fibers mainly through
the dendritic synapses in the preoptic white matter
along the border of APON cell mass, since Hal-
pern [62] noted that terminal degeneration by the
telencephalic lesions was located along the lateral
edges of cell masses in the frog hypothalamus. The
single BC-contacting neurons thus detect changes
in titers of blood-born hormones, preferably sex
steroid hormones which have activational effects
on APON neurons, and further receive neuronal
input signals through dendritic synapses to inte¬
grate hormonal and neural signals concerned with
the initiation of sex behavior.

930

A. Urano

AFFERENTS OF THE APON

Retrograde Horseradish Peroxidase (HRP) Study

It is important to know what neuroanatomic
afferent relations the APON has with other parts
of the brain. Such information is requisite for
better understanding of the sensory modalities and
activating or inhibiting pathways that might trigger
or modulate sexual behavior through the PON.
Thus, the afferents of the APON were examined in
Rana pipiens (Urano and Gorbman, unpublished)
and Bufo japonicus [63] using the retrograde HRP
method, which is a particularly useful tool in
studies of neural connections.

Evidence of retrogradely transported enzymatic
activity was observed in perikarya and neuropil in
the following brain regions: the ventro-medial
limbic cortex, the posterior part of the preoptic
nucleus including the magnocellular part, the in¬
fundibular nuclei, the thalamic area, the subtectal
and tegmental regions including the reticular
formation, and the rhombencephalic central gray.
Neurons in these regions appear to send their
axons to the APON mainly via the medial and
lateral forebrain bundles. Localization of some
HRP-labeled perikarya and fibers coincides with
that of immunoreactive perikarya and fibers con¬
taining either LHRH, vasotocin or TRH which
have been considered to project to the APON [64,
65],

Particular HRP-labeled loci in the ventro-medial
limbic cortex included the nucleus medialis septi,
the nucleus lateralis septi, the nucleus accumbens
septi, the amygdala pars medialis and the nucleus
of the diagonal band of Broca. The amygdala-
preoptic tract may exist in all vertebrate classes
from cyclostomes to mammals [66]. The septal
projection to the preoptic area in the leopard frog
and the Japanese toad has an apparently homolo¬
gous relationship to a similar pattern in the lizard
[67] and the rat [68, 69]. Although in anurans, the
physiological significance of amygdaloid and septal
projections to the APON is not clear at present, it
is possible that these projections are concerned
with the control of sexual behavior as has been
claimed in mammals [70, 71].

HRP-labeled structures in the subtectal and

tegmental regions were the nucleus anterodorsalis
tegmenti mesencephali, the torus semicircularis,
the nucleus posteroventralis tegmenti mesen¬
cephali, the nucleus isthmi and the mesencephalic
reticular nuclei. Mesencephalic projections to the
anterior hypothalamus are well known in amphib¬
ian brains [17, 19, 66, 72] as well as in other
vertebrate classes [66, 73, 74], The mammalian
preoptic area is directly continuous with a vast
nonspecific neuronal apparatus of the brain stem
reticular formation [75]. In frog brains, the
mesencephalic reticular system receives afferents
from various parts of the brain, such as the telen¬
cephalon [62, 76], the optic tectum [77], and the
superior olivary nucleus [78]. The presence of
multimodal inputs suggests a nonspecific or gener¬
alized character of function of the anuran reticular
formation as a possible activating or inhibitory
regulatory system which may influence the neural
substrate for mating behavior.

Chemical Neuroanatomy of the APON Afferents

Information on the chemical nature of APON
afferents is important for the examination of con¬
trol mechanisms of APON neuronal activity at the
cellular and molecular levels.

The HRP study mentioned above showed the
presence of HRP-labeled neurons in the mag¬
nocellular part of the PON, and in the nuclei
infundibularis dorsalis and ventralis in the toad
brain. These regions are rich in vasotocinergic and
mesotocinergic neurosecretory neurons [22], and
TRH neurons [65, 79], respectively. Jokura and
Urano [64] verified that varicose ir-vasotocin fibers
are found in the ventrolateral region of the APON
where the APON neurons have their dendritic
fields. Some ir-vasotocin fibers from the vmc
protrude into the APON cell mass, and appeared
to come into contact with somata of APON
neurons. In Japanese toads, ir-TRH neurons were
localized mainly in the nucleus infundibularis ven¬
tralis (NIV) [65]. Ir-TRH fibers arising from the
NIV neurons project to the median eminence to
form the hypothalamo-hypophysial tract. In addi¬
tion, a considerable number of ir-TRH fibers in¬
nervate into the APON. In the APON, varicose
ir-TRH fibers are scattered widely among the
neuronal cell mass and the white matter.

Mating Behavior and Preoptic Neurons

931

Other important loci in the toad brain where
HRP-labeled neurons were found include the nu¬
cleus medialis septi and the nucleus of the diagonal
band of Broca. These loci contained many ir-
LHRH neurons which project to the APON [64],
Most ir-LHRH fibers emanating from the nucleus
medialis septi form a loose fiber bundle with those
arising from the diagonal band of Broca. These
ir-LHRH fibers, which have typical beaded fea¬
tures, project to the ventrolateral border of the
preoptic gray.

In mammalian brains, peptidergic axonic pro¬
cesses form ordinary synapses [80] and en passant
synapses with dendritic profiles [81]. Therefore, it
is highly probable that varicose ir-vasotocin, ir-
TRH and ir-LHRH fibers form ordinary or en
passant synapses in the dendritic fields of APON
neuron in the toad brain.

Functional Significance of the APON Aff events

The retrograde HRP study indicates that there
are multimodal inputs to the APON from various
regions of the brain. The septal nuclei, which send
ir-LHRH fibers to the APON, receive olfactory
inputs through the medial olfactory tract [82, 83],
and the amygdala is innervated by projections
from the accessory olfactory bulb [84]. These
limbic nuclei, from which affe rents to the APON
arise, may relay olfactory signals to the APON
neurons. In addition, the terminal nerve, which
may function in odor processing, sends an ir-
LHRH-ergic projection to the preoptic region in
the tiger salamander and the bullfrog [85].

Visual cues can be conveyed through direct
retinal projection to the suprachiasmatic part of
the PON. This was clarified in the Japanese toad
by use of a cobaltic lysine method (Shimotoso and
Urano, unpublished). The presence of direct
retino-preoptic projetion has also been supposed
in the brain of Rana temporaria [25]. Acoustic
signals which excite APON neurons may reach the
preoptic region through at least two ascending
pathways in the brain stem [78]. One is the
pathway relayed through the nucleus oliva superi¬
or and the nucleus profundus mesencephali; the
other is that relayed through the nucleus oliva
superior and the torus semicircularis. The thala-
mo-preoptic connection is a possible pathway for

transmission of tactile signals. Thus, the APON
neurons may be influenced by various kinds of
sensory inputs, although almost all sensory modali¬
ties are relayed and may be regulated either by sex
steroid hormones or by neurohormones released
from extrahypothalamic terminals of neurosecre¬
tory neurons [86, 87]. Since the electrical activity
of many APON units was excited by iontophoreti-
cally applied LHRH, TRH and vasotocin (Fujita
and Urano, in preparation), the APON neurons
probably integrate various sensory inputs under
the influence of peptidergic neurosecretory
neurons, and then generate neural signals for the
initiation of mate calling behavior.

SEASONAL VARIATIONS

Many anurans, especially those in the temperate
zone, are typical seasonal breeders which spawn in
spring or early summer. The neuroendocrine
systems associated with reproductive behavior also
show seasonal changes in their synthetic and secre¬
tory activities. In bullfrogs, the plasma level of
luteinizing hormone (LH), which can increase
androgen secretion from the testes [88], was ele¬
vated during the breeding season [89]. In the
Xenopus hypothalamus, the contents of LHRH,
which can stimulate pituitary gonadotropin release
in bullfrogs [90] and plasma androgen levels in
Japanese toads [91], varied seasonally in corre¬
spondence to reproductive physiological states
[92]. Ishii and his collaborators measured circan-
nual changes in plasma levels of various pituitary
hormones [93, 94], thyroid hormones [95], adrenal
steroids [96, 97] and sex steroids [98]. Most of
these hormones showed marked increases in their
plasma titers prior to or during the breeding sea¬
son. Further, classical histochemical and mor¬
phometric studies showed seasonal morphological
changes in the hypothalamo-neurohypophysial
neurosecretory system in Rana temporaria [99,
100]. The results of these studies suggest the
possibility that the APON neurons do show some
seasonal changes in their morphological and func¬
tional features, since the activity of APON
neurons was modulated by administrations of tes¬
tosterone [14] and pituitary homogenates [53].

932

A. Urano

Seasonal Changes in the Volumes of PON Sub¬
nuclei

Seasonal variations in the volumes of several
preoptic and amygdala subnuclei were found be¬
tween hibernating and postbreeding Japanese
toads [34]. The APON in the hibernating males
was 125% larger than that in the post-breeding
animals. The seasonal difference in the female
APON was not statistically significant. In both
sexes, the hibernating animals had larger ventral
magnocellular parts of the PON, amygdala
medialis and amygdala lateralis than the post¬
breeding animals did. The seasonal variation in
the volumes of several subnuclei mentioned above
may be due to hypertrophy of the neurons in these
loci, since cell nulear volumes of PON neurons
increase prior to the breeding season [100]. Mor¬
phological changes in the APON and the amygdala
thus precede physiological and behavioral changes
in the breeding season. A similar result was
observed in the song control nucleus in the brain of
the male canary which is larger during breeding
season than in the fall when the animal is sexually
inactive [101], probably because sex steroids in¬
duce dendritic growth in this nucleus [102].

Immunoreactivity of Neuroendocrine Cells

It is highly probable that neuronal input signals
to APON neurons differ seasonally. Therefore,
seasonal variations in LHRH, TRH and vasotocin
that were localized in varicose afferent fibers to the
APON were examined immunohistochemically in
toad forebrains and neurohypophyses [65, 103].
The immunohistochemical technique utilized was
the avidin-biotin-peroxidase complex (ABC)
method, which is superior to the peroxidase-
antiperoxidase method in a quantitative study.
LHRH immunoreactivity (ir) was strong in both
perikarya and fibers in animals captured in spring
and autumn, while in summer animals, LHRH-ir
was weak. Seasonal changes in TRH-ir were
similar to those in LHRH-ir, while significant
seasonal variations were not found in vasotocin-ir.
The circannual changes in LHRH-ir appear to
correspond with seasonal variations in plasma ste¬
roid levels reported by Inoue et al. [98]. This
coincidence implies that LHRH and sex steroids

can have synergistic effects on the control of
APON neurons.

Effects of Castration

As is described above, testosterone may mod¬
ulate neural activity of the APON to initiate
reproductive behavior. Structures of the nervous
system are modified by sex steroids during both
fetal and adult periods in many vertebrate species
[101, 102, 104]. In Japanese toads, the volumes of
the APON and the amygdala in the male are larger
than those in the female. Furthermore, the
volumes of these nuclei and LHRH-ir also changed
seasonally. These changes appear to correlate with
the annual variation of plasma testosterone levels.
Castration experiments, in which the role of testos¬
terone in the control of the phenomena mentioned
above was examined, showed that the effects of
castration differ seasonally [105]. The volume of
the amygdala medialis in autumn toads was signif¬
icantly reduced by castration; however, the reduc¬
tion in spring animals was not statistically signif¬
icant. Meanwhile, castration did not modify
LHRH-ir in the median eminence in either spring
or autumn toads, although dense ir-LHRH fibers
were observed in the mesencephalic tegmental
region in the castrated spring toads but not in the
autumn toads either intact or castrated. These
results suggest that seasonal influences on the
effects of castration were not uniform among the
different brain loci.

COORDINATION OF NEURAL AND
ENDOCRINE ACTIVITY

Temporal coordination of neural and endocrine
events is a cruial requisite for successful reproduc¬
tion. Plausible candidates for coordinating the
brain and endocrine functions are LHRH-ergic
and vasotocinergic neurosecretory systems, both
of which send fine varicose fibers to various ex-
trahypothalamic brain loci other than the median
eminence and the pars nervosa [86, 87]. Both
LHRH-ergic and vasotocinergic fibers innervate
either sensory or motor centers concerned with
reproductive behavior.

LHRH applied by microiontophoresis increased
discharge rates of individual neurons in the septal

Mating Behavior and Preoptic Neurons

933

preoptic area of the rat [106], and in the APON of
the Japanese toad (Fujita and Urano, in prepara¬
tion). An intracellular study showed that LHRH
can mimic slow excitatory postsynaptic potentials
when applied to the sympathetic neurons in the
bullfrog [107]. LHRH at the synaptic level may
play a role in increasing neuronal excitability in the
loci where LHRH fibers innervate. On the other
hand, LHRH applied systemically or intraventri-
cularly can stimulate the pituitary-gonadal axis to
elevate plasma androgen levels in male Japanese
toads (Fig. 5). This evidence suggests that LHRH
simultaneously affects both the neuronal activity of
the APON neurons as an excitatory neurotrans¬
mitter or neuromodulator and the endocrine
events of the pituitary-gonadal axis as a hypothala¬
mic releasing hormone.

The endocrine functions of vasotocin in amphib¬

ians are well documented in many endocrine text¬
books. In addition, vasotocin and its homologues
can excite unit-spike activity of neurons in the rat
supraoptic and paraventricular nuclei [106, 108],
the eel preoptic nucleus [109] and the toad APON
(Fujita and Urano, in preparation). Vasotocin
thus may facilitate the activity of many central
neurons as a neuromodulator or a local hormone.
The latter possibility is supported by the fact that
10_9M vasopressin, comparable to the effective
dose of vasopressin necessary for peripheral
targets, can excite rat paraventricular neurons
[108].

At present, it is difficult to account for the
temporal discrepancy between LHRH-induced
neural events (Fig. 4) and endocrine events (Fig.
5). When LHRH functions as a neurotransmitter
or a neuromodulator, its influence on target

Time of ler Administration (hours) Time after Administration (hours)

Time af ter Administration (hours)

Fig. 5. Changes in the plasma testosterone levels after administrations of intravenous and intraventricular LHRH.
Each curve represents a change in testosterone levels in an individual male toad, a, effects of intravenous saline as
a control; b, single intravenous injection of 10 fig LHRH; c, single intravenous injection of 100 jug LHRH; d,
continuous infusion of LHRH at a dose of 100 //g/hour for 4 hours; and e, intracerebroventricular administration
of 100 ng LHRH. Note that the dose of intracranial LHRH which markedly elevated plasma testosterone was
much less than that of intravenous LHRH. (Fujita, thesis, Saitama University)

934

A. Urano

neurons lasted for within the order of seconds or
minutes. However, endocrine events, e.g., the
secretion of androgen, take a much longer time.
Since the APON neurons are sex steroid-sensitive
and are excited by LHRH, some unknown intrinsic
cellular mechanisms within the APON neurons
and neurons having the same characteristics may
regulate the above temporal discrepancy in order
to complete seasonal breeding successfully.

REFERENCES

1 Demski, L. S. and Knigge, K. M. (1971) The tel¬
encephalon and hypothalamus of the bluegill
( Lepomis macrochirus ): evoked feeding, aggres¬
sive and reproductive behavior with representative
frontal sections. J. Comp. Neurol., 143: 1-16.

2 Schmidt, R. S. (1968) Preoptic activation of frog
mating behavior. Behaviour, 30: 239-257.

3 Wheeler, J. M. and Crews, D. (1978) The role of
the anterior hypothalamus-preoptic area in the
regulation of male reproductive behavior in the
lizard, Anolis carolinensis : lesion studies. Horm.
Behav., 11: 42-60.

4 Barfield, R. J. (1969) Activation of copulatory
behavior by androgen implanted into the preoptic
area of the male fowl. Horm. Behav., 1: 37-52.

5 Morali, G. and Beyer, C. (1979) Neuroendocrine
control of mammalian estrous behavior. In “En¬
docrine Control of Sexual Behavior”. Ed. by C.
Beyer, Raven Press, New York, pp. 33-175.

6 Larssen, K. (1979) Features of the neuroendocrine
regulation of masculine sexual behavior. In “En¬
docrine Control of Sexual Behavior”. Ed. by C.
Beyer, Raven Press, New York, pp. 77-163.

7 Aronson, L. R. and Noble, G. K. (1945) The sex¬
ual behavior of anura. II. Neural mechanisms
controlling mating in the male leopard frog, Rana
pipiens. Bull. Am. Mus. Nat. Hist., 86: 83-140.

8 Schmidt, R. S. (1976) Neural correlates of frog
calling: isolated brainstem. J. Comp. Physiol., 108:
99-113.

9 Urano, A. (1983) Neural correlates of mating
behavior in perfused toad brains. Zool. Mag., 92:
601.

10 Satou, M., Oka, Y. and Ueda, K. (1987) The
nervous system. In “Biology of Toad”. Ed. by A.
Urano and K. Ishihara, Shokabo, Tokyo, pp. 62-
90.

11 Schmidt, R. S. (1969) Preoptic activation of mat¬
ing call orientation in female anurans. Behaviour,
35: 114-127.

12 Schmidt, R. S. (1966) Hormonal mechanisms of
frog mating calling. Copeia, 1966: 637-644.

13 Palka, Y. S. and Gorbman, A. (1973) Pituitary
and testicular influenced sexual behavior in male
frogs, Rana pipiens. Gen. Comp. Endocrinol., 21:
148-151.

14 Wada, M. and Gorbman, A. (1977) Relation of
mode of administration of testosterone to evoca¬
tion of male sex behavior in frogs. Horm. Behav.,
8: 310-319.

15 Diakow, C. and Nemiroff, A. (1981) Vasotocin,
prostaglandin, and female reproductive behavior in
the frog, Rana pipiens. Horm. Behav., 15: 86-93.

16 Kelley, D. B. (1982) Female sex behaviors in the
South African clawed frog, Xenopus laevis : gona¬
dotropin-releasing, gonadotropic, and steroid hor¬
mones. Horm. Behav., 16: 158-174.

17 Herrick, C. J. (1948) The brain of tiger salaman¬
der, Ambystoma tigrinum. Univ. Chicago Press,
Chicago.

18 Takami, S., Jokura, Y. and Urano, A. (1984) Sub-
nuclear organization of the preoptic nucleus in the
toad, Bufo japonicus. Zool. Sci., 1: 759-770.

19 Frontera, J. G. (1952) A study of anuran
diencephalon. J. Comp. Neurol., 96: 1-69.

20 Hoffman, H. H. (1963) The olfactory bulb, acces¬
sory olfactory bulb and hemisphere of some anur¬
ans. J. Comp. Neurol., 120: 317-368.

21 Wada, M., Urano, A. and Gorbman, A. (1980) A
stereotaxic atlas for diencephalic nuclei of the frog,
Rana pipiens. Arch. Histol. Jpn., 43: 157-173.

22 Vandesande, F. and Dierickx, K. (1976) Im-

munocytochemical demonstration of separate vaso-
tocinergic and mesotocinergic neurons in the
amphibian hypothalamic magnocellular
neurosecretory system. Cell Tissue Res., 175: 289-
296.

23 Vandesande, F. and Dierickx, K. (1980) Im-

munocytochemical localization of somatostatin-
containing neurons in the brain of Rana tempo -
raria. Cell Tissue Res., 205: 45-53.

24 Yui, R. (1983) Immunohistochemical studies on
peptide neurons in the hypothalamus of the bull¬
frog Rana catesbeiana. Gen. Comp. Endocrinol.,
49: 195-209.

25 Vullings, H. B. G. and Heussen, A. M. A. (1975)
Electron microscopic observations on the retino-
preoptic pathway of Rana temporaria. Cell Tissue
Res., 161: 177-182.

26 Arnold, A. P. (1980) Sexual differences in the
brain. Am. Sci., 68: 165-173.

27 Panzica, G. C., Viglietti-Panzica, C., Calacagni,
M., Anselmetti, G. C., Schumacher, M. and Balth-
azart, J. (1987) Sexual differentiation and hormon¬
al control of the sexually dimorphic medial preoptic
nucleus in the quail. Brain Res., 416: 59-68.

28 Arimatsu, Y., Seto, W. and Amano, T. (1981)
Sexual dimorphism in alpha-bungarotoxin binding

Mating Behavior and Preoptic Neurons

935

capacity in the mouse amygdala. Brain Res., 213:
432-437.

29 Aynob, D. M. and Greenough, W. T. (1983) Sex
differences in dendritic structure in the preoptic
area of the juvenile macaque monkey brain. Sci¬
ence, 219: 197-198.

30 Gorski, R. A., Gordon, J. H., Shryne, J. E. and
Southam, A. M. (1978) Evidence for a morpholo¬
gical sex difference within the medial preoptic area
of the rat brain. Brain Res., 148: 333-346.

31 Matsumoto, A. and Arai, Y. (1986) Morphologi¬
cal evidence for sexual dimorphism in wiring pat¬
tern in the neuroendocrine brain. In “Pars Distalis
of the Pituitary Gland-Structure, Function and
Regulation”. Ed. by F. Yoshimura and A. Gorb-
man, Elsevier, Amsterdam, pp. 239-245.

32 Hannigan, P. and Kelley, D. B. (1981) Male and
female laryngeal motor neurons in Xenopus laevis.
Soc. Neurosci. Abstr., 7: 269.

33 Schmidt, R. S. (1982) Sexual dimorphism in succi¬
nic dehydrogenase staining of toad pretrigeminal
nucleus. Exp. Brain Res., 45: 447-450.

34 Takami, S. and Urano, A. (1984) The volume of
the toad medial amygdala-anterior preoptic com¬
plex is sexually dimorphic and seasonally variable.
Neurosci. Lett., 44: 253-258.

35 Takami, S., Higashiyama, T. and Urano, A. (1984)
A quantitative analysis of sexual dimorphism in the
anterior part of the preoptic nucleus in the toad.
Neurosci. Lett., S17: S35.

36 Urano, A. (1984) A Golgi-electron microscopic
study of anterior preoptic neurons in the bullfrog
and the toad. Zool. Sci., 1: 89-99.

37 Wada, M. and Gorbman, A. (1977) Mate calling
induced by electrical stimulation in freely moving
leopard frogs, Rana pipiens. Horm. Behav., 9:
141-149.

38 Schmidt, R. S. (1984) Neural correlates of frog
calling: preoptic area trigger of ‘mating calling’. J.
Comp. Physiol., 154: 847-853.

39 Nakai,Y., Ochiai,H., Shioda, S. and Ochi, J.
(1977) Cytological evidence for different types of
cerebrospinal fluid-contacting subependymal cells
in the preoptic and infundibular recesses of the
frog. Cell Tissue Res., 176: 317-334.

40 Fasolo, A. and Franzoni, M. F. (1977) A Golgi
study on the hypothalamus of amphibia. The
neuronal typology. Cell Tissue Res., 178: 341-354.

41 De Waele, G. and Dierickx, K. (1979) Scanning
electron microscopy of the wall of the third ventri¬
cle in the brain of Rana temporaria. Part IV. Cell
Tissue Res., 203: 53-64.

42 Kelley, D. B., Morrell, J. I. and Pfaff, D. W.
(1975) Autoradiographic localization of hormone¬
concentrating cells in the brain of an amphibian,
Xenopus laevis. I. Testosterone. J. Comp. Neurol.,

164: 47-62.

43 Morrell, J. I., Kelley, D. B. and Pfaff, D. W.
(1975) Autoradiographic localization of hormone¬
concentrating cells in the brain of an amphibian,
Xenopus laevis. II. Estradiol. J. Comp. Neurol.,
164: 63-78.

44 Kelley, D. B., Lieberburg, I., McEwen, B. S. and
Pfaff, D. W. (1978) Autoradiographic and biochem¬
ical studies of steroid hormone-concentrating cells
in the brain of Rana pipiens. Brain Res., 140: 287-
305.

45 Sar, M. and Stumpf, W. E. (1975) Distribution of
androgen-concentrating neurons in rat brain. In
“Anatomical Neuroendocrinology”. Ed. by W. E.
Stumpf and L. D. Grant, S. Karger, Basel, pp.
120-133.

46 Pfaff, D. W. (1980) Estrogen and Brain Function.
Springer-Verlag, New York, Heidelberg, Berlin.

47 Yagi, K. (1973) Changes in firing rates of single
preoptic and hypothalamic units following an in¬
travenous administration of estrogen in the cas¬
trated female rat. Brain Res., 53: 343-352.

48 Bueno, J. and Pfaff, D. W. (1976) Single unit
recording in hypothalamus and preoptic area of
estrogen-treated and untreated ovariectomized
female rats. Brain Res., 101: 67-78.

49 Kow, L.-M. and Pfaff, D. W. (1985) Estrogen
effects on neuronal responsiveness to electrical and
neurotransmitter stimulation: an in vitro study on
the ventromedial nucleus of the hypothalamus.
Brain Res., 347: 1-10.

50 Nabekura,J., Oomura, Y., Minami,T., Mizuno,
Y. and Fukuda, A. (1986) Mechanism of the rapid
effect of 17-estradiol on medial amygdala neurons.
Science, 233: 226-228.

51 Schmidt, R. S. (1966) Central mechanisms of frog
calling. Behaviour, 26: 251-285.

52 Wetzel, D. M., Haerter, U. L. and Kelley, D. B.
(1985) A proposed neural pathway for vocalization
in South African clawed frogs, Xenopus laevis. J.
Comp. Physiol. A, 157: 749-761.

53 Urano, A. and Gorbman, A. (1981) Effects of
pituitary hormonal treatment on responsiveness of
anterior preoptic neurons in male leopard frogs,
Rana pipiens. J. Comp. Physiol., 141: 163-171.

54 Urano, A. (1985) The anterior preoptic neurons of
the frog, Rana catesbeiana and toad, Bufo bufo
japonicus: control of mating behavior. In “Current
Trends in Comparative Endocrinology”. Ed. by B.
Lofts and W. N. Holmes, Hong Kong Univ. Press,
Hong Kong. pp. 1125-1126.

55 Smoller, C. G. (1965) Neurosecretory processes
extending into third ventricle: secretory or sen¬
sory? Science, 147: 882-884.

56 Joseph, S. A., Sorrentino, S., Jr. and Sundberg,
D. K. (1975) Releasing hormones, LRF and TRF,

936

A. Urano

in the cerebrospinal fluid of the third ventricle. In
“Brain-Endocrine Interaction II. The Ventricular
System in Neuroendocrine Mechanisms”. Ed. by
K. M. Knigge, D. E. Scott, H. Kobayashi and S.
Ishii, Karger, Basel, pp. 306-312.

57 Robinson, I. C. A. F. and Jones, P. M. (1982)
Neurohypophysial peptides in cerebrospinal fluid:
recent studies. In “Neuroendocrinology of Vaso¬
pressin, Corticoliberin and Opiomelanocortins”.
Ed. by A. J. Baertchi and J. J. Dreifuss, Academic
Press, London, pp. 21-31.

58 Klein, D. C. (1979) Circadian rhythms in the
pineal gland. In “Endocrine Rhythms”. Ed. by D.
T. Krieger, Raven Press, New York, pp. 202-223.

59 Fujita, Y. and Urano, A. (1985) Enhancement of
EEG activity by injection of LHRH and TRH in
hibernating toad, Bufo japonicus. Zool. Sci., 2:
979.

60 Reaves, T. A., Cumming, R. and Hayward, J. N.
(1982) Light- and electron-microscopic character¬
ization of electrophysiologically identified, horse¬
radish peoxidase-injected magnocellular neuroen¬
docrine cells in goldfish preoptic nucleus. Neurosci¬
ence, 7: 1545-1557.

61 Murakami, M. (1964) Electronenmikroskopisch
Untersuchungen am Nucleus Praeopticus der
Krote (Bufo vulgaris formosus). Z. Zellforsch., 63:
208-225.

62 Halpern, M. (1972) Some connections of the tel¬
encephalon of the frog, Rana pipiens. An ex¬
perimental study. Brain Behav. Evol., 6: 42-68.

63 Fujita, Y. and Urano, A. (1986) The afferent
projection to the anterior part of the preoptic
nucleus in Japanese toads, Bufo japonicus. Zool.
Sci., 3: 677-686.

64 Jokura, Y. and Urano, A. (1985) Projections of
luteinizing hormone-releasing hormone and vaso¬
tocin fibers to the anterior part of the preoptic
nucleus in the toad, Bufo japonicus. Gen. Comp.
Endocrinol., 60: 390-397.

65 Fujita, Y. and Urano, A. (1986) Immunohistochem-
ical localization of thyrotropin-releasing hormone
(TRH) in the hypothalamus of the toad, Bufo
japonicus. Proc. Jpn. Soc. Com. Endocrinol., 1:
23.

66 Crosby, E. C. and Showers, M. J. C. (1969) Com¬
parative anatomy of the preoptic and hypothalamic
areas. In “The Hypothalamus”. Ed. by W.
Haymaker, E. Anderson and W. J. H. Nauta,
Charles, C. Thomas, Springfield, pp. 61-135.

67 Hoogland, P. V., Ten Donkelaar, H. J. and Cruce,
J. A. F. (1978) Efferent connections of the septal
area in a lizard. Neurosci. Lett., 7: 61-65.

68 Raisman, G. (1966) The connections of the sep¬
tum. Brain, 89: 317-348.

69 Swanson, L. W. and Cowan, W. M. (1976) Auto¬

radiographic studies of the development and con¬
nections of the septal area in the rat. In “The Septal
Nuclei. Adv. Behav. Biol., Vol. 20 ”. Ed. by J. E.
DeFrance, Plenum, New York, London, pp. 37-
64.

70 Lisk, R. D. (1967) Sexual behavior: hormonal con¬
trol. In “Neuroendocrinology. Vol. II”. Ed. by L.
Martini and W. F. Ganong, Academic Press, New
York, pp. 197-239.

71 Malsbury, C. W. and Pfaff, D. W. (1974) Neural
and hormonal determinants of mating behavior in
adult male rats. A review. In “Limbic and Auto¬
nomic Nervous Systems Research”. Ed. by L. V.
DiCara, Plenum, New York, London, pp. 85-136.

72 Kicliter, E. and Northcutt, R. (1975) Ascending
afferents to the telencephalon of Ranid frogs: an
anterograde degeneration study. J. Comp.
Neurol., 161: 239-254.

73 Ariens Kappers, C. U., Huber, G. C. and Crosby,
E. C. (1960) The Comparative Anatomy of the
Nervous System of Vertebrates, including Man.
Vol. 2. Hafner, New York.

74 Zyo, K., Oki, T. and Ban, T. (1963) Experimental
studies on the medial forebrain bundle, medial
longitudial fasciculus and supraoptic decussations
in the rabbit. Med. J. Osaka Univ., 13: 193-239.

75 Nauta, W. J. H. and Haymaker, W. (1969)
Hypothalamic nuclei and fiber connections. In
“The Hypothalamus”. Ed. by W. Haymaker, E.
Anderson and W. J. H. Nauta, Charles C. Tho¬
mas, Springfield, pp. 136-209.

76 Kokoros, J. J. (1972) Amphibian telencephalic
efferents: an experimental study. Am. Zool., 12:
727.

77 Rubinson, K. (1968) Projection of the tectum
opticum of the frog. Brain Behav. Evol., 1: 529-
561.

78 Rubinson, K. and Skiles, M. (1973) Efferent pro¬
jections from the superior olivary nucleus in Rana
catesbeiana. Anat. Rec., 175: 431-432.

79 Seki,T., Nakai,Y., Shioda, S., Mitsuma, T. and
Kikuyama, S. (1983) Distribution of immunoreac-
tive thyrotropin-releasing hormone in the forebrain
and hypophysis of the bullfrog, Rana catesbeiana.
Cell Tissue Res., 233: 507-516.

80 Buijs, R. M. (1985) Extrahypothalamic pathways
of a neurosecretory system: their role in neuro¬
transmission. In “Neurosecretion and the Biology
of Neuropeptides”. Ed. by H. Kobayashi, H. A.
Bern and A. Urano, Jpn. Sci. Soc. Press, Tokyo,
pp. 279-286.

81 Silverman, A. J. (1983) Luteinizing hormone re¬
leasing hormone (LHRH) synapses in the diagonal
band (DBB) and preoptic area (POA) of the
guinea pig. Soc. Neurosci. Abstr., 9: 1182.

82 Scalia, F., Halpern, M., Knapp, H. and Riss, W.

Mating Behavior and Preoptic Neurons

937

(1968) The efferent connections of the olfactory
bulb in the frog: a study of degenerating unmyelin¬
ated fibers. J. Anat., 103: 245-262.

83 Northcutt, R. G. and Royce, G. J. (1975) Olfac¬
tory bulb projections in the bullfrog, R. catesbeiana
Shaw. J. Morphol., 145: 251-268.

84 Scalia, F. (1972) The projection of the accessory
olfactory bulb in the frog. Brain Res., 36: 409-411.

85 Wirsig, C. R. and Getchell, T. (1986) Amphibian
terminal nerve: distribution revealed by LHRH
and AChE markers. Brain Res., 385: 10-21.

86 Jokura, Y. and Urano, A. (1986) Extrahypothala-
mic projection of luteinizing hormone-releasing
hormone fibers in the brain of the toad, Bufo
japonicus. Gen. Comp. Endocrinol., 62: 80-88.

87 Jokura, Y. and Urano, A. (1987) Extrahypothala-
mic projection of immunoreactive vasotocin fibers
in the brain of the toad, Bufo japonicus. Zool. Sci.,
4: 675-681.

88 Muller, C. H. and Licht, P. (1980) Gonadotropin
specificity of androgen secretion by amphibian
testes. Gen. Comp. Endocrinol., 42: 365-377.

89 Licht, P., McCreery, B. R., Barnes, R. and Pang,
R. (1983) Seasonal and stress related changes in
plasma gonadotropins, sex steroids, and cortico¬
sterone in the bullfrog, Rana catesbeiana. Gen.
Comp. Endocrinol., 50: 124-145.

90 Daniels, E. and Licht, P. (1980) Effects of gona¬
dotropin-releasing hormone on the levels of plasma
gonadotropins (FSH and LH) in the bullfrog, Rana
catesbeiana. Gen. Comp. Endocrinol., 42: 455-
463.

91 Fujita, Y. and Urano, A. (1987) Effects of LHRH
on EEG activity and plasma androgen levels in
Japanese toads. Zool. Sci., 4: 1079.

92 King, J. A. and Millar, R. P. (1979) Hypothalamic
luteinizing hormone-releasing hormone content in
relation to the seasonal reproductive cycle of Xeno-
pus laevis. Gen. Comp. Endocrinol., 39: 309-312.

93 Yoneyama, H., Ishii, S., Yamamoto, K. and
Kikuyama, S. (1984) Plasma prolactin levels of
Bufo japonicus before, during and after breeding in
the pond. Zool. Sci., 1 :969.

94 Itoh, M. and Ishii, S. (1986) Establishment of
radioimmunoassays of toad ( Bufo japonicus) gona¬
dotropins and annual cycle of plasma FSH and LH
levels in a wild toad population. Zool. Sci., 3: 1078.

95 Tasaki, Y., Inoue, M. and Ishii, S. (1986) Annual
cycle of plasma thyroid hormone levels in the toad,
Bufo japonicus. Gen. Comp. Endocrinol., 62: 404-
410.

96 Jolivet-Jaudet, G., Inoue, M., Takada, K. and
Ishii, S. (1984) Circannual changes in corticoster¬
one plasma levels and binding of corticosterone to
plasma in Bufo japonicus formosus. Zool. Sci., 1:
317-325.

97 Jolivet-Jaudet, G., Inoue, M., Takada, K. and
Ishii, S. (1984) Circannual changes in plasma
aldosterone levels in Bufo japonicus formosus.
Gen. Comp. Endocrinol., 53: 163-167.

98 Inoue, M., Takada, K. and Ishii, S. (1982) Season¬
al changes in plasma androgen concentrations and
reproductive behavior in toads. Zool. Mag., 91:
568.

99 Dierickx, K. and Van Meirvenne, N. (1961)
Karyometric studies of the preoptic nucleus of
Rana temporaria. Gen. Comp. Endocrinol., 1: 51—
58.

100 Dierickx, K. and Vandesande, F. (1965) The mag-
nocellular preoptic nuclei and the reproduction in
Rana temporaria. Z. Zellforsch., 68: 190-193.

101 Nottebohm, F. (1981) A brain for all seasons:
Cyclical anatomical changes in song control nuclei
of the canary brain. Science, 214: 1368-1370.

102 DeVoogd, T. and Nottebohm, F. (1981) Gonadal
hormones induce dendritic growth in the adult
avian brain. Science, 214: 202-204.

103 Jokura, Y. and Urano, A. (1985) An Immunohis-
tochemical study of seasonal changes in luteinizing
hormone-releasing hormone and vasotocin in the
forebrain and the neurohypophysis of the toad,
Bufo japonicus. Gen. Comp. Endocrinol., 59: 238-
245.

104 Kurz, E. M., Sengelaub, D. R. and Arnold, A. P.
(1986) Androgens regulate the dendritic length of
mammalian motoneurons in adult-hood. Science,
232: 395-398.

105 Fujita, Y., Jokura, Y., Takami, S. and Urano, A.
(1987) Effects of castration on volumes of the
preoptic nucleus and the amygdala and on im-
munoreactivity of LHRH fibers in the brain of the
toad Bufo japonicus. Gen. Comp. Endocrinol., 68:
278-285.

106 Baker, J. L. (1977) Physiological roles of peptides
in the nervous system. In “Peptides in Neurobiolo¬
gy”. Ed. by H. Gainer, Plenum Press, New York,
pp. 295-343.

107 Jan, L. Y., Jan, Y. N. and Brownfield, M. S.
(1980) Peptidergic transmitters in synaptic boutons
of sympathetic ganglia. Nature, 288: 380-382.

108 Inenaga, K. and Yamashita, H. (1986) Excitation
of neurones in the rat paraventricular nucleus in
vitro by vasopressin and oxytocin. J. Physiol., 370:
165-180.

109 Sugita, R. and Urano, A. (1986) Responses of
magnocellular neurons in in vitro eel preoptic nu¬
cleus (PONmg) to acetylcholine, catecholamines,
vasotocin, isotocin, angiotensin, and Na+. Zool.
Sci., 3: 1081.

110 Urano, A. and Ishihara, K. [eds] (1987) Biology of
Toad. Shokabo, Tokyo.

'

-

■

ZOOLOGICAL SCIENCE 5: 939-945 (1988)

© 1988 Zoological Society of Japai

Activation of Respiration and Initiation of Motility
in Rainbow Trout Spermatozoa

Toshio Inoda* 1, Hideki Ohtake2 3 4 and Masaaki Morisawa3, 4

Ocean Research Institute, University of Tokyo, Nakano-ku, Tokyo 164, and
2Dokkyo University, School of Medicine, Mibu,

Tochigi 321-02, Japan

ABSTRACT — It is well established that sperm motility of rainbow trout is initiated by the decrease in
K+ concentration surrounding sperm which triggers the intracellular c AMP-dependent initiation
process. Present study showed that K+ did not affect sperm respiration but inhibited flagellar
movement and thus suggested that K+ regulates sperm motility through its effect on flagellum. On the
other hand, inhibitors of respiratory chain or uncoupler of oxidative phosphorylation affected sperm
respiration and inhibited sperm motility, suggesting that energy producing system at mitochondria
contributes to sperm motility. Motility was initiated even if 02 was eliminated from dilution medium,
although C02 suppressed both respiration and motility. This result suggested that sperm motility is not
(Delimited but C02 is responsible for the regulation of sperm motility through the activation of
respiration. It is likely that regardless of K + -dependent cAMP system at sperm flagella, there is another
system at mitochondria: enhancement of respiration by the release from C02 suppression at spawning
may relate to the initiation of sperm motility in rainbow trout.

INTRODUCTION

Spermatozoa are immotile in undiluted semen
and initiate motility on dilution into appropriate
medium. As factors to cause the phenomenon,
many things in the seminal plasma have been
proposed (see [1]). Rothschild [2] postulated that
low 02 tension in the seminal plasma is most likely
responsible for the sperm immotility in the repro¬
ductive organ and that increase in 02 tension
surrounding spermatozoa at spawning causes in¬
itiation of motility. Carbon dioxide was also
proposed as another possible factor from the re¬
sults that C02 inhibits both respiration and motil¬
ity in sea urchin sperm [3, 4]. Johnson et al. [4]
also suggested that 02 does not affect sperm

Accepted January 8, 1988
Received October 24, 1987

1 Present address: Department of Biology, Faculty of
Science, Toho University, Funabashi, Chiba 274,

Japan.

3 Present address: Misaki Marine Biological Station,
University of Tokyo, Miura, Kanagawa 238-02, Japan.

4 To whom reprints should be requested.

motility, since motility initiation occurs when O?
was eliminated by blowing N2 gas over a thin layer
of semen. These studies have focused on the
contribution of energy supply system to the motil¬
ity initiation; however it is still unclear which
factor is the physiological initiator of sperm
motility.

Morisawa and collegues recently proposed a
motility initiation system from another point of
view. They showed that motility of spermatozoa in
salmonid fishes is suppressed by K+ and spermato¬
zoa become motile in the K+ deficient medium [5].
However, K+ can not inhibit motility of trout
spermatozoa of which plasma membrane and
mitochondria are removed with the detergent [6],
implicating that the site of K+ action is not
mitochondria but flagella. By regulating flagellar
motility with or without K+, it is possible to
separate the mitochondrial function from the
flagellar function. Consequently, salmonid sperm
seems to offer an especially convenient material
for investigating which factor contributes to
mitochondrial metabolism or flagellar mechanism
in the initiation of sperm motility.

940

T. Inoda, H. Ohtake and M. Morisawa

For clarifying this point, we compared the re¬
spiration and motility in trout sperm in the pres¬
ence or absence of K+ and furthermore examined
the effects of aerobic or anaerobic condition and
C02 on the sperm respiration and motility. The
results suggested that K+ dependent initiation
system is present in flagella, and that increase of
energy supply at mitochondria by decrease of C02
may possibly contribute to the initiation of trout
sperm motility.

MATERIALS AND METHODS

Mature male rainbow trout ( Salmo gairdneri )
was obtained from Oshino Branch of Yamanashi
Prefectural Fisheries Experimental Station. They
were kept in an aquarium with circulating and
aerating water at 10°C. The semen was collected
by inserting a pipette into the sperm duct. Col¬
lected sperm was preserved on ice without dilution
for several hours during the experiments.

For investigating the effects of K + , dilution,
inhibitors of respiratory chain and uncoupler of
oxidative phosphorylation on sperm respiration
and motility (Figs. 1-3), 100 mM NaCl or KC1
solution was kept without bubbling with any gases.
With 3 ml of the above solutions 0. 1 ml semen was
diluted with various conditions in the chamber of
oxymeter and oxygen consumption was measured.
Each plot in Figures 2, 3 and 5 was calculated from
the oxygen consumption in 5 sec after dilution. On
a glass slide without cover 0.1 /A of semen was
suspended in 50 jA of 100 mM NaCl solution and
sperm motility was observed by light microscopy
using dark illumination. NaN3 and KCN were
each dissolved in distilled water. CCCP (carbonyl
cyanide ra-chlorophenyl hydrazone) was dissolved
in 4% ethanol which did not affect sperm motility
and respiration.

For studying the effect of 02 (Fig. 4a), N2 gas
was introduced into 200 ml of 100 mM NaCl solu¬
tion in an Erlenmeyer flask from a N2 gas cylinder.
Amount of dissolved 02 was checked at an
appropriate time interval with an oxymeter. A
closed chamber (Bellco: 0.75 ml) was filled with
the solution containing various concentrations of
02 using a syringe and 1 jA of semen was injected
with a microsyringe, and then the motility of sperm

in the chamber was observed under microscope.

The effect of completely-02-eliminated condi¬
tion (Fig. 4b) was investigated in 100 mM NaCl
solution containing various concentrations of
Na2S204, which was introduced into both a closed
chamber and an oxymeter, and sperm motility and
oxygen content were measured.

Effect of C02 on sperm motility was investi¬
gated (Fig. 5) as follows. C02 gas was bubbled
into 200 ml of 100 mM NaCl solution in the flask
for a few hours. pH value of the solution de¬
creased during C02-bubbling and finally reached
6.0. Media containing various concentrations of
C02 were prepared with mixing the C02 saturated
medium with 100 mM NaCl solution and pH was
adjusted to 6.0 with HC1. Each medium was
introduced into the closed chamber and oxymeter,
and sperm motility and oxygen consumption were
measured. Amount of C02 and pH value in these
media were checked with a carbon analyzer (Mod¬
el 524 C, O. I. Corporation, U. S. A.) and pH
meter respectively before experiment.

Oxygen consumption was measured with an
oxymeter (Yanagimoto Co., Ltd.) for 30 to 60 sec
at a chart speed of 30 or 60cm/min. Solutions
were buffered with 20 mM Hepes-NaOH at pH 8.0
(Figs. 1-4) and 6.0 (Fig. 5). Experiments were
carried out at 10°C (Figs. 1-4) or 20°C (Fig. 5).

Tracks of sperm were recorded by VTR through
a video camera connected with a microscope and
percentage of motile sperm and swimming speed in
Figures 4 and 5 were measured as described pre¬
viously [7]. In Figures 1 and 3, the number of
moving spermatozoa was evaluated in terms of
grade ( — , + , +): grade +, at least over half of
spermatozoa were motile in the field of view of
microscope; grade ±, below half of spermatozoa
were motile; grade — , all spermatozoa were im-
motile.

RESULTS

Effect of potassium

As shown in Figure 1, when spermatozoa were
suspended into 100 mM NaCl solution at a dilution
ratio of 1:30, in which spermatozoa initiated for¬
ward motility, they consumed oxygen at the rate of

Respiration and Trout Sperm Motility

941

1

1

o 4 8

Time after dilution (sec)

Fig. 1. Change in the oxygen consumption of rainbow
trout spermatozoa in NaCl and KC1 solutions. Se¬
men at the volume of 0.1 ml was diluted with 100

mM NaCl solution ( - ) or 100 mM KC1 solution

( — ) buffered with 20 mM Hepes at pH 8.0.
Arrows indicate the time of adding the semen.
Sperm motility was exhibited in parentheses.

66.9 + 3.8 nmol/ml semen/sec from three experi¬
ments in Figure 1 in 5 sec after dilution and then
the rate decreased. Spermatozoa diluted in 100
mM KC1 (1:30 dilution) were completely im-
motile, however, they consumed oxygen at 67.1 +
11.7 nmol/ml semen/sec in 5 sec and then the ox¬
ygen consumption became lower. Namely the rate
of oxygen uptake of the sperm which were quies¬
cent in the presence of K+ was almost the same as
that of the sperm which initiated motility in the
absence of K+.

Effect of dilution

Oxygen consumption of undiluted trout semen
was almost zero (Fig. 2). When semen was diluted
in lOOmM NaCl solution (1: 15 dilution), sperma-

450

Dilution rate

Fig. 2. Effect of dilution on the oxygen consumption of
rainbow trout spermatozoa. The appropriate
volume of semen was added to 3 ml of 100 mM NaCl
solution buffered with 20 mM Hepes, pH 8.0. Ver¬
tical bars represent Means + S.E. in 3 experiments.

tozoa consumed oxygen at 45 + 20 nmol/ml semen/
sec. Oxygen consumption increased with the in¬
crease of a dilution rate and reached almost max¬
imum at a dilution rate of 1: 240 (365 + 26 nmol/ml
semen/sec). The level was maintained until a
dilution rate reached 1:480.

Effects of NaN 3, KCN and CCCP

Oxygen consumption of spermatozoa in 5 sec in
100 mM NaCl solution at a dilution rate of 1:30
was 69.4 + 6.2 nmol/ml semen/sec (Fig. 3a), which
was almost equal to that in Figure 1. When the
dilution medium contained NaN3, oxygen con¬
sumption of sperm decreased with the increase of
concentration of NaN3: In the medium containing
10 mM NaN3, it was 67% of that in the NaN3 free
medium. Sperm motility was almost suppressed
with 5 mM NaN3 and completely suppressed with
10 mM NaN3.

Oxygen consumption and motility of the sperm
decreased as the concentration of KCN increased
(Fig. 3b). The oxygen consumption reached to
67% of that in the KCN free condition in the
presence of 10 mM KCN. Sperm motility became
feeble by the addition of 5 mM KCN and was
completely suppressed by 10 mM KCN.

As shown in Figure 3c, when spermatozoa were
diluted with lOOmM NaCl solution containing

Oxygen consumption Oxygen consumption Oxygen consumption

(nmol 02/ml semen/s) ZI (nmol 02/ml semen/s) " CT = (nmol 02/ml semen/s)

942

T. Inoda, H. Ohtake and M. Morisawa

a

I _ II _ | _ I _ 1 _ l _ l _ l

0 || 1.25 2.5 5 10 2 0 40

CCCP (carbonyl cyanide ra-chlorophenyl hydra-
zone), slight enhancement of oxygen consumption
was observed: In the presence of 40 juM of CCCP,
it was 118 + 54 nmol/ml semen/sec, that was 1.57-
fold in the absence of CCCP. Spermatozoa
showed active motility at the concentrations of less
than 2.5 //M of CCCP and they were completely
immotile at 10 //M of CCCP.

Effects of O 2 and C02

Sodium chloride solution at the concentration of
100 mM contained 345 nmol 02/ml. This value
was designated as 100%. As shown in Figure 4a,
when N2 gas was bubbled into the 100 mM NaCl
solution for 10 min, oxygen concentration rapidly
decreased to 111.2 nmol/ml (32.2%) and reached
23 nmol/ml (6.7%) with 160 min bubbling. Sper¬
matozoa were motile at any oxygen level.

Oxygen concentration in 100 mM NaCl solution
at pH 8.0 was reduced in the presence of Na2S204
(Fig. 4b): Only 115 nmol 02/ml (11.1%) was con¬
tained at 2.5 mM of Na2S204 and oxygen content
was zero at 5 mM of Na2S204. In spite of the
drastic reduction of 02 content, all spermatozoa
exhibited full motility with constant velocity of
156-172. 5//m/sec in 0 to 40 mM Na2S204.

When the semen was diluted with 100 mM NaCl
solution at pH 6.0, spermatozoa moved with the
velocity of 189 + 7.5 //m/sec and their oxygen con¬
sumption was 69 + 16.3 nmol/ml semen/sec (Fig.
5). However, when C02 content in the medium
increased, motility, velocity and oxygen consump¬
tion decreased and became zero at 50 mM of C02.

DISCUSSION

Sperm motility of rainbow trout is suppressed in
the sperm duct by seminal K+, and decrease in K+
concentration surrounding sperm at spawning into
fresh water induces the initiation of sperm motility

Fig. 3. Effects of inhibitors of respiratory chain and an
uncoupler of oxidative phosphorylation on the ox¬
ygen consumption and motility of rainbow trout
spermatozoa. Semen was added to 100 mM NaCl
containing various concentrations of NaN3 (a), KCN
(b) or CCCP (c). Vertical bars represent Means ±
S.E. in 3 experiments. Sperm motility was exhibited
in parentheses.

CCCP (^M)

Respiration and Trout Sperm Motility

943

a

100 r QA— A-A -A -A-A-A-A-A A A A - A -.100

0-0^0'0N

oN

O — O-

I L

•O - O

4)

a

(/)

0 10 30 50 70 100 120 140 160

Bubbling time (min)

HH - 1 - 1 - 1 - 1

0 2.5 5 10 20 40

Na2S204 (mM)

Fig. 4. Effect of 02 on the motility of rainbow trout spermatozoa, (a). Relative percentage of
dissolved oxygen (o) and motility of spermatozoa (a) in 100 mM NaCl when N2 was bubbled
for indicated time, (b). Percentage of dissolved oxygen in 100 mM NaCl containing Na2S204
(O)- Percentage (a) or swimming velocity (a.) of motile spermatozoa. Vertical bars
represent Means + S.E. in 20-25 spermatozoa.

Swimming velocity (jum/s)

944

T. Inoda, H. Ohtake and M. Moris a wa

C02 (mM)

Fig. 5. Effect of C02 on the oxygen consumption and motility of rainbow trout spermatozoa. Percentage (a) or
swimming velocity (a) of motile spermatozoa. Oxygen consumption of spermatozoa (#). Vertical bars represent
Means + S.E. in 3 experiments.

[5]. Although our recent studies have demon¬
strated the detailed mechanism of K+ dependent
initiation process of trout sperm motility [1], the
target site of K+ has been left somewhat unclear.
In this paper, it was shown that the oxygen con¬
sumption of sperm, of which motility was suppres¬
sed by K+, was almost similar to that of motile
spermatozoa in the K+ free medium (Fig. 1). This
suggested that K+ does not suppress mitochon¬
drial respiration but do flagellar movement. Fur¬
thermore, target of K+ may be plasma membrane
of sperm flagella since flagella of which plasma
membrane was removed are able to beat in the
presence of K+ [6].

It has been reported that immotile trout sperma¬
tozoa retain a high concentration of ATP, while a
rapid decrease of ATP level occurs within very
short period when spermatozoa initiate motility
[8]. This phenomenon might be correlated with
the short term oxygen consumption of trout sper¬
matozoa at the initiation of motility which occurs
within a very short period. The short term oxygen
consumption of trout sperm increased with in¬
crease of dilution ratio (Fig. 2). From the result, it
seems to be considered that gradual activation of
mitochondrial function occurs at natural spawning

when spermatozoa are released and gradually di¬
luted in water. In the process, some changes of
volatile factor in the circumstance of sperm may
possibly relate to the initiation of energy supply
and sperm motility. Thus there is some room for
further examining the correlation between sperm
respiration and initiation of motility.

NaN3 and KCN, inhibitors of respiratory chain,
or CCCP, an uncoupler of oxidative phosphoryla¬
tion, suppressed sperm motility (Fig. 3), suggest¬
ing that sperm motility seems to be restricted by
the energy supplying systems. These results con¬
firmed our preliminary data [9].

Many investigators reported that sperm respira¬
tion and motility are affected considerably by 02
and C02 (see [1]). Rothschild [2] reported that sea
urchin spermatozoa in a gas-tight chamber were
immotile when N2 was introduced, however sper¬
matozoa became motile when 02 was introduced
into the chamber. However, the opposite conclu¬
sion was proposed by Johnson et al. [4]. In
rainbow trout, as shown in Figure 4, spermatozoa
could initiate and maintain motility in 02 deficient
medium, even in a completely anaerobic medium
obtained by the addition of Na2S204. This result
suggests that 02 is not a limiting factor for sperm

Respiration and Trout Sperm Motility

945

motility in this species. A change from anaerobic
to aerobic condition, which might occur at natural
spawning, may not affect sperm motility.

C02 is reported as a suppressor of sperm motil¬
ity in many animals [1]; for example, motility and
respiration of sea urchin spermatozoa are revers¬
ibly suppressed by C02 [4]. In rainbow trout, CO?
influenced inhibitorily to the sperm respiration and
motility (Fig. 5). Thus, C02 seems to be an
attractive candidate as the factor for suppressing
the sperm respiration and motility in the semen in
reproductive organ in which C02 is present [10].

In conclusion, although there are some doubts
whether volatile factor(s) physiologically restricts
the initiation of trout sperm motility, it is attractive
to predict that a volatile factor dependent system
at mitochondria may contribute to the initiation of
trout sperm motility, independently of the estab¬
lished K+ dependent initiation mechanism at
flagella.

ACKNOWLEDGMENTS

We thank the director and staff of Oshino Branch of
Yamanashi Prefectural Fisheries Experimental Station
for providing material. This work was supported in part
by grant-in-aid from the Japanese Ministry of Education,
Science and Culture to M. M.

REFERENCES

1 Morisawa, M. (1985) Initiation mechanism of sperm
motility at spawning in teleosts. Zool. Sci., 2: 605-
615.

2 Rothschild, Lord (1948) The physiology of sea-
urchin spermatozoa: lack of movement in semen. J.
Exp. Biol., 25: 344-352.

3 Mohri, H. and Yasumasu, I. (1963) Studies on the
respiration of sea-urchin spermatozoa. V. The effect
of Pco2. J. Exp. Biol., 40: 573-586.

4 Johnson, C. H., Clapper, D. L., Winkler, M. M.,
Lee, H. C. and Epel, D. (1983) A volatile inhibitor
immobilizes sea urchin sperm in semen by depressing
the intracellular pH. Dev. Biol., 98: 493-501.

5 Morisawa, M., Suzuki, K. and Morisawa, S. (1983)
Effects of potassium and osmolality on spermatozoan
motility of salmonid fishes. J. Exp. Biol., 107: 105—
113.

6 Morisawa, M. and Okuno, M. (1982) Cyclic AMP
induces maturation of trout sperm axoneme to initi¬
ate motility. Nature, 295: 703-704.

7 Morisawa, M., Morisawa, S. and De Santis, R.
(1984) Initiation of sperm motility in Ciona intestina-
lis by calcium and cyclic AMP. Zool. Sci., 1: 237-244.

8 Christen, R., Gatti, J. and Billard, R. (1987) Trout
sperm motility — The transient movement of trout
sperm is related to the changes in the concentration of
ATP following the activation of the flagellar move¬
ment. Eur. J. Biochem., 166: 667-671.

9 Morisawa, M., Okuno, M., Suzuki, K., Morisawa, S.
and Ishida, K. (1982) Initiation of sperm motility in
teleosts. J. Submicrosc. Cytol., 15: 61-65.

XU ,

ZOOLOGICAL SCIENCE 5: 947-950 (1988)

© 1988 Zoological Society of Japan

Excitatory and Inhibitory Junction Potentials Recorded
from the Red Muscle of Marine Teleost, Puffer Fish

Tohoru Hidaka and Shiro Yukiyama1

Department of Biology, Faculty of General Education and
1 Faculty of Science, Kumamoto University,

Kumamoto 860, Japan

ABSTRACT — Junction potentials were recorded from the red muscle of two species of the marine
teleost, puffer fish, Takifugu poecilonotus and T. rubripes. Three types of potential change, the
excitatory junction potential (ejp), the inhibitory junction potential (ijp) and the diphasic junction
potential composed of ejp and ijp, were elicited by the nerve stimulation, and the miniature excitatory
junction potential (mejp) and the miniature inhibitory junction potential (mijp) were observed in the
resting muscle. Thus, this muscle received the innervations from both excitatory and inhibitory nerves
and the excitatory innervations were much more abundant than the inhibitory ones. Nicotinic
antagonist of acetylcholine (ACh), d-tubocurarine (d-TC), suppressed ejp, ijp, mejp and mijp and
anticholinesterase, neostigmine, augmented them, while muscarinic antagonist of ACh, atropine, did
not affect them. The results suggested that the excitatory and the inhibitory neuromuscular
transmissions of this muscle were cholinergic and the nature of the receptors was nicotinic. The present
observations obtained in the marine teleost were almost the same as those reported in the freshwater
teleost, silver carp.

INTRODUCTION

In the vertebrate skeletal muscle, it is well
known that there are two muscle types, referred as
fast and slow, phasic and tonic or white and red,
respectively. As reviewed by Hess [1] and Morgan
and Proske [2], the white muscle produces the
action potential in response to the nerve stimula¬
tion, whereas the red muscle does not initiate the
spike but responds with non-propagating junction
potential in various kinds of vertebrate skeletal
muscle. Similar observations were obtained in the
white and red muscle of freshwater teleost [3, 4].
The red muscle of a silver carp was found to elicit
the excitatory junction potential (ejp) by the nerve
stimulation and to generate the miniature excita¬
tory junction potential (mejp) in the resting muscle
[4]. Recently, it was reported in the same nerve-
muscle preparation that not only ejp but also the
inhibitory junction potential (ijp) and the diphasic
junction potential composed of ejp and ijp were

Accepted February 2, 1988
Received January 8, 1988

elicited by the nerve stimulation and that the
miniature inhibitory junction potential (mijp) as
well as mejp could be recorded from the resting
muscle [6]. This observation was the first demon¬
stration that ijp and mijp were recorded in the
vertebrate skeletal muscle.

The experiment reported below was carried out
in order to evaluate if the inhibitory innervation
was present in red muscle of marine teleost and to
compare the innervation pattern between the red
muscles of freshwater and marine teleost fishes.

MATERIALS AND METHODS

Two species of puffer fish, Takifugu poecilono¬
tus and Takifugu rubripes , 12-27 cm in body
length, were used. Because significant differences
in the results were not found between two species,
the results obtained using T. poecilonotus were
presented in this paper. Fishes were purchased
from a fishery and were kept in the natural sea
water saturated with air up to about a month. The
nerve-muscle preparation was dissected from the
red muscle of both sides of pectoral fin which was

948

T. Hidaka and S. Yukiyama

innervated by the spinal motor nerves, Th. 1 and
Th. 2, as described previously [5]. The prepara¬
tion was placed in a chamber of 5 ml in volume and
was perfused at a constant flow rate of 5 ml/min
with artificial sea water (ASW) of the following
composition (mM); NaCl 462, KC1 9.4, CaCl2
10.8, MgCl2 48.2 and NaHC03 6.0 (pH 8.0). The
methods of recording of the electrical responses
and the nerve stimulation were the same as those
in the previous study [6]. The following drugs were
used; d-tubocurarine (d- TC, Sigma), neostigmine
methylsulfate (Sigma) and atropine sulfate
(Nakarai Chemicals). The effects of these drugs
dissolved in ASW were tested by the bath applica¬
tion. The experiment was carried out at room
temperature (18-24°C).

RESULTS

The resting potentials of the red muscle fiber
were —70 + 1.1 mV (mean + SE, n— 25) in Takifu-
gu poecilonotus and —69.5 + 0.8 mV (mean + SE,
n = 25) in T. rubripes. The significant difference in
the resting potential was not noted between the
freshwater and the marine teleosts, the corre¬
sponding value of silver carp being —73.1 mV [5].

Figure 1 shows the typical response to the single
nerve stimulation recorded from the red muscle

|lOmV

100msec

Fig. 1. Three types of response elicited by single nerve
stimulation, a; ejp, b; ijp, c; diphasic junction
potential. Records taken from different muscle
fibers. Dots in this and Fig. 3 indicate nerve stimula¬
tion.

fiber of T. poecilonotus. These potentials were ejp
(a), ijp (b) and diphasic junction potential (c). The
amplitude and the duration of three types of the
response were different from fiber to fiber ex¬
amined. In the diphasic junction potential con¬
sisted of ejp and ijp, ejp consistently proceeded

ijp-

The amplitudes of ejps and ijps were 5.4 ±0.6
mV (mean + SE, n = 20) and 2.6 ±2.0 mV (mean±
SE, n=20) respectively, and the durations of ejp
and ijp were 30.7 + 3.7 msec (mean + SE, n=20)
and 130.0 + 14.2 msec (mean + SE, n = 20) respec¬
tively. The amplitudes of depolarizing phase and
hyperpolarizing phase of diphasic junction poten¬
tials were 6. 7 + 0. 9 mV (mean + SE, n=20) and
2. 2 + 0. 4 mV (mean + SE, n=20) respectively, and
the duration of depolarizing phase and hyperpolar¬
izing phase of diphasic junction potentials were
18.4 + 2.3 msec (mean + SE, n=20) and 115.6 +
14.9 msec (mean + SE, n = 20) respectively. Thus,
the amplitude of ejp exceeded that of ijp and the
duration of ejp was shorter than that of ijp. This
was the case in depolarizing phase and hyperpolar¬
izing phase of diphasic junction potential.

In the resting muscle, the miniature junction
potentials (mjps) were generated spontaneously.
Figure 2 showed the sample records in which mejp
(a), mijp (b) and both mejp and mijp (c) were
observed in three different muscle fibers. An
example generating only mijp was rare and was
observed in only one fiber throughout the present
experiment (b). It was noticed that mejps having
two different, fast and slow, rise times were re¬
corded from the same muscle fibers (a and c).
From the observations presented in Figures 1 and
2, it was suggested that the red muscle of puffer
fish might receive the double innervations from
both the excitatory and the inhibitory nerves and
the excitatory innervation might be distributed
along the muscle fiber multiply.

Table 1 shows the number and the percentage,
of the junction potentials (A) and the miniature
junction potentials (B), which were recorded from
the muscle fibers inserted with the microelectrode
arbitrarily. Out of 425 muscle fibers examined, the
percentages of ejp, ijp and diphasic junction
potential were 56%, 16% and 28% respectively
(A). Out of 120 muscle fibers, the percentages of

Ejp and Ijp in Puffer Fish

949

Fig. 2. Sample records of mejp (a), mijp (b) and both mejp and mijp (c) recorded continuously
from different muscle fibers.

Table 1. Number (percentage) of junction potentials (jp) recorded from 425 muscle fibers (A)
and of miniature junction potentials (mjp) recorded from 120 muscle fibers (B) of T.
poecilonotus

(A) Number (%) of jp

(B) Number (%) of mjp

ejp

238 (56)

mejp

103 (86)

flP

68 (16)

mijp

1 ( 1)

diphasic jp

119 (28)

mejp and mijp

16 (13)

total

425

total

120

the fibers generating mejp, mijp and both mejp
and mijp were 86%, 1% and 13% respectively (B).
These percentages might roughly reflect the dis¬
tribution of the excitatory and the inhibitory in-

10"6M d-TC 10"eM neostigmine

50msec

Fig. 3. Effects of d-TC and neostigmine on diphasic
junction potential, al and bl; control. a2; 15 min
after 10-6 M d-TC. b2; 15 min after 10-6M neo¬
stigmine. a and b; records from different muscle
fibers. Partial recovery 60 min after washing (not
shown).

nervations of this muscle and might suggest that
the excitatory innervations are more abundant
than the inhibitory ones.

To investigate the nature of the neurotransmit¬
ter and the receptor mediating the neuromuscular
transmission of this muscle, the effects of d-TC and
neostigmine on the diphasic junction potential
were examined. As shown in Figure 3, diphasic
junction potential was almost eliminated by 10~6
M d-TC (a2) and was augmented by 10-6 M
neostigmine (b2). Similar effects of d-TC and
neostigmine were observed on ejp, ijp and mijp.
Atropine (10~6M), the muscarinic antagonist of
ACh receptor, had no appreciable effects on all
types of junction potentials. The results indicate
that the transmission is cholinergic at the neuro¬
muscular junction of this muscle and the nature of
the receptor is nicotinic.

DISCUSSION

It was reported in the previous paper that de¬
polarizing ejp, hyperpolarizing ijp and diphasic

950

T. Hidaka and S. Yukiyama

junction potential were recorded from the red
muscle of the freshwater fish, silver carp [6]. This
finding was the first demonstration of the inhibi¬
tory innervation in the vertebrate skeletal muscle.
The present results confirmed this finding and
extended it to the marine teleost, puffer fish. As
presented in Table 1, ijp and mijp were recorded
from extremely fewer muscle fibers than ejp and
mejp in puffer fish. The generation of ijp was
observed in 68 (16%) of 425 muscle fibers in which
junction potentials were recorded, and mijp
appeared in only one of 120 fibers in which minia¬
ture junction potentials were recorded. Thus, the
inhibitory innervations seem to be also far fewer
than the excitatory ones in the red muscle of puffer
fish, as suggested in silver carp [6].

It was demonstrated that both the excitatory and
the inhibitory neuromuscular transmissions were
cholinergic and were mediated by nicotinic recep¬
tors in the freshwater [6] and the marine teleosts.
In such situation, the same neurotransmitter, pre¬
sumably ACh, would be concerned in the genera¬
tion of both ejp and ijp. At present, it is difficult to
determine whether both ejp and ijp are elicited by
the transmitter released from the same nerve ter¬
minal or from the different nerve terminals. Also,
it is not certain whether the same receptor con¬
tributes to the generation of ejp and ijp or the
different receptors elicit ejp and ijp separately.

The possibility that the transmitters released
from the different nerve terminals activate the
different receptor sites and produce ejp and ijp
separately might be considered. The result that

ejp and ijp were converted to diphasic junction
potential by increasing the intensity of nerve stim¬
ulation [6] would support this possibility. Fur¬
thermore, since the length constant of the red
muscle fiber of silver carp was measured to be 2. 1
mm [5], it might be possible that diphasic junction
potential or both mejp and mijp generating at the
receptors few millimeters apart were recorded with
the microelectrode inserted into a certain position
of the muscle fiber. However, in such a situation,
since depolarizing ejp and hyperpolarizing ijp or
mejp and mijp are generated by the different ionic
mechanisms, it is likely that the receptor sites
operating by the different ionic mechanisms are
distributed in patches on a surface of a single
muscle fiber.

REFERENCES

1 Hess, A. (1970) Vertebrate slow muscle fibers. Phys¬
iol. Rev., 50: 40-62.

2 Morgan, D. L. and Proske, U. (1984) Vertebrate
slow muscle: Its structure, pattern of innervation, and
mechanical properties. Physiol. Rev., 64: 103-169.

3 Takeuchi, A. (1959) Neuromuscular transmission of
fish skeletal muscles investigated with intracellular
microelectrode. J. Cell. Comp. Physiol., 54: 211-220.

4 Hidaka, T. and Toida, N. (1969) Neuromuscular
transmission and excitation-contraction coupling in
fish red muscle. Japan. J. Physiol., 19: 130-142.

5 Hidaka, T. and Toida, N. (1969) Biophysical and
mechanical properties of red and white muscle fibres
in fish. J. Physiol., 201: 49-59.

6 Hidaka, T. and Miyahara, T. (1987) Excitatory and
inhibitory neuromuscular transmission in fish red
muscle. Zool. Sci., 4: 819-823.

ZOOLOGICAL SCIENCE 5: 951-957 (1988)

© 1988 Zoological Society of Japai

The Involvement of Microtubules in the Light
Response of Medaka Melanophores

Sumiko Negishi

Department of Biology, Keio University,
Yokohama 223, Japan

ABSTRACT — The light response of isolated medaka melanophores is accompanied by remarkable
changes in cell shape; the peripheral region is extended in light and retracted in the dark. This change in
the peripheral region of the cytoplasm can be prevented by the antimitotic agents, colcemid and
nocodazole, and by cytochalasin B. The dark-induced aggregation of melanosomes is scarcely inhibited
by antimitotic agents or cytochalasin B, while the pigment dispersion induced by irradiation is prevented
almost completely by antimitotic agents and partially by cytochalasin B. When melanophores are
treated with colcemid or nocodazole in the dark, their radially-arrayed microtubules disappear and the
cell outline becomes irregular. However, the distribution of microtubules is preserved in the
light-dispersed cells even after treatment with antimitotic agents. These findings suggest that
microtubules of the normal number and distribution, and not microfilaments, are required for the
centrifugal pigment migration of melanophores responding to light and for the retainment of cell shape.

INTRODUCTION

Color changes of the integument, which are
widely observed in lower vertebrates, occur due to
pigment translocation in the chromatophores [1].
The force-generating mechanism in transporting
pigments has been studied in conjunction with
other non-muscle cell motility. However, the
mechanism causing pigment migration has been
too complex to be elucidated [2]. For example, for
the microtubules, two studies have given contra¬
dictory results. In one, the number of microtu¬
bules in angelfish melanophores with aggregated
pigments decreased to 40% in the cells with dis¬
persed pigments [3], while in another study on
Fundulus melanophores, the number of microtu¬
bules in the dendritic processes remained un¬
changed before and after pigment withdrawal [4].
In general, microtubules are thought to be indis¬
pensable for the fast transport of pigments within
chromatophores in some teleost species [4-6] and
to be required for the development of normal cell
shape and pigment distribution [7].

In Oryzias melanophores, pigment aggregation
at the cell center occurs in response to a-adrenergic
agonists [8], melatonin [9], melanophore-concen-
trating hormone (MCH) [10, 11] and changes in
illumination [12]. The reactions to the former
three agents are completed within 2 min to punc¬
tate state in cultured melanophores [11], whereas
30 min is required to complete the aggregation in
the dark [12], The difference in the velocity
between these reactions raises the question of
whether the same motive force is used for both
reactions, i.e., the microtubules may not be in¬
volved in the slow reaction, the dark-induced
melanosome aggregation.

The present experiment was undertaken to
elucidate the role of microtubules in light-
dependent pigment displacement by using anti¬
mitotic agents, colcemid and nocodazole, and
cytochalasin B, a drug that disrupts cytoplasmic
microfilaments. A specific immunological probe
for microtubules was used to study the changes in
their distribution in melanophores responding to
illumination.

Accepted February 10, 1988
Received January 16, 1988

952

S. Negishi

MATERIALS AND METHODS

Melanophores were isolated from scales of the
wild type medaka, Oryzias latipes (BBRR) [13],
and cultured in L-15 medium supplemented with
5% FCS for 2-6 days at 25°C before use. Melano-
some translocation, which was induced by changes
in the illumination, was observed using an inverted
microscope (Nikon Diaphoto). The intensity of
the incident light was adjusted to 600 lux with a
halogen lamp (12 V, 50 W) equipped with neutral
density filters. The responses of melanophores
were recorded with an image analyzer, Planimex
25 (Nireco). The full-aggregated state attained by
epinephrine was usually corresponding to about
15% of the full dispersion in the culture medium.
For indirect immunofluorescence microscopy,
melanophores cultured on a cover glass were per-
meabilized with Brij 58 and polyethylene glycol for
1 min [11]. After fixation with 3.7% formaldehyde
in phosphate-buffered saline (PBS) for 6 min at
room temperature, the cell preparations were
rinsed with PBS. Next, a 1:20 dilution of anti¬
tubulin antibody (Miles Yeda) was applied to the
specimen for 1 hr at 37°C. After washing with
PBS, the specimen was incubated with FITC
labeled IgG (Miles Yeda) for 1 hr at 37°C followed
by washing with PBS, and enclosed in 90% (vol/
vol) glycerol/PBS on a slide glass. The fluores¬
cence was observed with a Nikon fluorescence
microscope. All drugs, colcemid (Sigma) and
epinephrine (Sigma) except cytochalasin B
(Aldrich) were directly dissolved in the saline
solution (127 mM NaCl/2.7 mM KCl/1.8mM
CaCl2/5 mM Tris-HCl buffer, pH 7. 3/5. 5 mM glu¬
cose). Cytochalasin B was diluted from a stock
solution dissolved in dimethyl sulfoxide.

RESULTS

When melanophores of Oryzias (Fig. la) were
moved to the dark, centripetal translocation of the
melanosomes occurred within 30 min at the veloc¬
ity of 0.6 + 0.02 //m/min, accompanied by retrac¬
tion of the cytoplasm in the marginal area (Fig.
lb). Subsequent illumination brought about ex¬
tension of the cytoplasm in the peripheral region
(Fig. lc), and then the pigment granules moved

centrifugally within 5 min at the rate of 4.9 ±0.5
pm/mm (Fig. Id). These alterations in the cell
outline occurred with all changes in illumination,
though the dark-induced alteration in the cell
shape was more remarkable than the light-induced
one. There was little shape alteration in the cells
when pigment was aggregated with epinephrine
(10/iM) within 2 min at the rate of 20 + 1.2 pm/
min.

Effect of colcemid

A melanophore, when incubated with 5 pM
colcemid in darkness for 60 min, showed centripe¬
tal migration of its pigment and decreased in cell
size by about 35% (Fig. 2b). Subsequent illumina¬
tion for 20 min did not cause dispersion of the
melanosomes (Fig. 3), and the cytoplasm in the
marginal area did not extend but retracted as
shown in Figure 2c.

Effect of nocodazole

Nocodazole, like colcemid, had an inhibitory
effect on the melanophores (Fig. 4). When the
melanophores were exposed to 1 pM nocodazole
for 20 min in the dark, the melanosomes migrated
to the cell body. However, the cell shape changed
remarkably, as shown in Figure 4b. The cell
processes were almost retracted and the cell size
became exceedingly small. Centrifugal migration
of melanosomes scarcely occurred with subsequent
irradiation, while slight extension of the cytoplasm
perceived (Fig. 4c). As the time period of nocoda¬
zole treatment was prolonged to 30 min, mela¬
nophores whose outlines became irregular were
incapable of pigment dispersion (data not shown).

Effect of cytochalasin B

When melanophores were exposed to cytochala¬
sin B (10 pg/m\) in the dark for 60 min, melano¬
somes aggregated more or less incompletely as
shown in Figure 5b, and the retraction in the
peripheral region of cytoplasm occurred, though
its extent was insufficient when compared to the
untreated melanophores. Subsequent irradiation
for 10 min in the drug caused pigment dispersion,
whereas the extent of dispersion was inhibited by
20% (Fig. 3). The marginal area of cytoplasm
scarcely expanded by this irradiation (Fig. 5c).

Role of Microtubules in Melanophores

953

Fig. 1. Melanophore response to illumination, a. Dispersed state in light, b. Aggregated state after 60 min in the
dark, c, d. Re-dispersed state after 2 min and 5 min illumination. (In all figures, the bar=30 /urn)

Fig. 2. Light micrographs showing the effect of colcemid (5 /iM). a. Control, dispersed state, b. Aggregated state
after 60 min colcemid treatment in the dark. c. Aggregated state after 20 min illumination followed the same
treatment as in b. The cell shape is drawn below the micrograph of each cell.

954

S. Negishi

Fig. 3. Effect of colcemid (5 /iM) and cytochalasin B
(10 //g/ml) on light response of melanophores.
Melanophores were treated in the dark with col¬
cemid for 60 min (a), or cytochalasin B for 60 min
(•) or 120 min (o), or in the saline (□), and then
illuminated at 600 lux. Each point represents the
mean + S.E. for 23-30 cells.

When exposure time to cytochalasin B under the
dark condition was prolonged to 120 min, pigment
dispersion in melanophores by the following irra¬
diation was inhibited by 60% (Fig. 3).

Immunofluorescence staining of cytoplasmic micro¬
tubules

As shown in Figure 6, a large number of micro¬
tubules are radially arrayed in an epinephrine-

aggregated melanophore (Fig. 6a). This situation
is similar to those in a dispersed cell (Fig. 6b),
while fewer microtubules are found in the dark-
aggregated cell (Fig. 6c). When a melanophore
was exposed to 5 //M colcemid in the dark for 60
min, radial alignments of microtubules were dis¬
rupted in most areas except the cell body (Fig. 6e).
In a melanophore exposed to 5 pM. colcemid in
light for 60 min, a number of microtubule arrays
were preserved as indicated in Figure 6d. The
disruption of microtubules as described for Figure
6e was similarly observed in melanophores after
treatment with nocodazole (1 ^M) (data not
shown).

DISCUSSION

The present results suggest that diassembly and
assembly cycles of microtubules in the cytoplasm
participate in the light-dependent migration of
pigments in Oryzias melanophores. The microtu¬
bules are thought to be indispensable for the rapid
pigment migration in fish chromatophores [4-6].
However, the role of microtubules in centrifugal
pigment migration remains to be clarified. In the
present study, colcemid-treated melanophores,
where radial alignments of microtubules almost
disappeared (Fig. 6e), responded no longer to
illumination with pigment dispersion as shown in

Fig. 4. Light micrographs showing the effect of nocodazole (1 //M). a. Control, dispersed state, b. Aggregated state
after 20 min nocodazole treatment in the dark. c. Aggregated state after 10 min illumination followed the same
treatment as in b. The cell shape is drawn below the micrograph of each cell.

Role of Microtubules in Melanophores

955

Fig. 5. Light micrographs showing the effect of cytochalasin B (10 /ig/ml). a. Control, dispersed state, b.
Aggregated state after 60 min cytochalasin B treatment in the dark. c. Dispersed state after 10 min illumination
followed the same treatment as in b. The cell shape is drawn below the micrograph of each cell.

Figure 3. This observation suggests that assembly
of the cytoplasmic microtubules plays an important
role in the centrifugal pigment transport induced
by light.

The fact that the rate of centripetal melanosome
migration in the dark was 20 times slower than that
in the epinephrine-treated one, may suggest a
difference in the transport mechanism between
both kinds of pigment aggregation. The number of
radial arrays of microtubules is distinctly less in the
dark-aggregated melanophores than in the
epinephrine-aggregated ones, suggesting that the
dark-induced pigment aggregation is closely re¬
lated with disassembly of cytoplasmic microtu¬
bules. The concept that the slow speed of melano¬
some aggregation induced in the dark is probably
caused by disassembly of cytoplasmic microtubules
does not contradict the findings that disassembly of
microtubules by means of cold treatment or treat¬
ment with antimitotic reagents causes remarkable
retardation of centripetal pigment movement
attained by the application of a-adrenergic agonists
[3-6, 14-16].

In the melanophores, whose outlines were not
changed by the treatment with cytochalasin B,
pigment migration occurred in both directions.

This observation suggests that microfilaments are
not always important as the motile system of
pigment transport in light adaptation of Oryzias
melanophores. The cell shape of melanophores
changed greatly with the colcemid treatment, sug¬
gesting that the microtubules, which run from the
cell center to the periphery, most likely participate
in the maintenance of cell shape. This agrees with
the results with Holocentrus erythrophores [7].
Melanophores, in their light response, retract the
cytoplasmic periphery in the dark and extend it in
light, suggesting that the fluid flux in the cytoplasm
may be involved in the melanosome movement
derived by light on and off as demonstrated in
Xenopus tadpoles [17]. However, cytochalasin
B-treated cells, whose cytoplasm in the periphery
scarcely changed by irradiation, are capable of
transporting melanosomes. These facts suggest
that the fluid flux of melanophores is not necessari¬
ly required as the driving force of pigment trans¬
location in light adaptation. Although the
machanism is still not clear, it seems that light on
and off starts a cycle of assembly and disassembly
of microtubules, which is involved in pigment
displacement of melanophores, and results in the
alteration of cell shape in response to light. Thus,

956

S. Negishi

Fig. 6. Localization of microtubules in melanophores revealed by immunofluorescence with tubulin antibody, a.
Melanophore aggregated by epinephrine, b. The cell under light, c. The dark aggregated cell, d, e.
Colcemid-treated cells in light (d) and darkness (e) for 60 min.

normal number and distribution of microtubules
appears to be required to support cell shape.

REFERENCES

1 Bagnara, J. T. and Hadley, M. E. (1973) Chroma-

tophores and Color Change. Prentice-Hall, Engle¬
wood Cliffs, New Jersey.

2 Obika, M. (1986) Intracellular transport of pigment
granules in fish chromatophores. Zool. Sci., 3: 1-11.

3 Schliwa, M. and Euteneuer, U. (1978) Quantitative
analysis of the microtubule system in isolated fish
melanophores. J. Supramol. Struct., 8: 177-190.

4 Murphy, D. B. and Tilney, L. G. (1974) The role of
microtubules in the movement of pigment granules
in teleost melanophores. J. Cell Biol., 61: 757-779.

5 Porter, K. R. (1973) Microtubules in intracellular

Role of Microtubules in Melanophores

957

locomotion. Ciba Found. Symp., 14: 149-166.

6 Schliwa, M. and Bereiter-Hahn, J. (1973) Pigment
movements in fish melanophores: Morphological
and physiological studies. III. The effects of colchi¬
cine and vinblastine. Z. Zellforsch., 147: 127-148.

7 Ochs, R. L. (1982) The role of microtubules in cell
shape and pigment distribution in spreading
erythrophores. Eur. J. Cell Biol., 28: 226-232.

8 Fujii, R. (1961) Demonstration of the adrenergic
nature of transmission at the junction between mela-
nophore-concentrating nerve and melanophore in
bony fish. J. Fac. Sci. Univ. Tokyo, Sect. IV, 9:
171-196.

9 Obika, M. (1976) An analysis of the mechanism of
pigment migration in fish chromatophores. Pigment
Cell (Basel), 3: 254-264.

10 Fujii, R. and Oshima, N. (1986) Control of chroma-
tophore movements in teleost fishes. Zool. Sci., 3:
13-47.

11 Negishi, S., Fernandez, H. R. and Obika, M. (1985)
The effects of dynein ATPase inhibitors on melano-
some translocation within melanophores of the
medaka, Oryzias latipes. Zool. Sci., 2: 469-475.

12 Negishi, S. (1985) Light response of cultured mela¬
nophores of a teleost adult fish, Oryzias latipes. J.
Exp. Zool., 236: 327-333.

13 Obika, M. (1976) Pigment migration in isolated fish
melanophores. Annot. Zool. Japon., 49: 157-163.

14 Wikswo, M. and Novales, R. R. (1972) Effect of
colchicine on microtubules in the melanophores of
Fundulus heteroclitus. J. Ultrastruct. Res., 41: 189—
201.

15 Byers, H. R. and Porter, K. R. (1977) Transforma¬
tions in the structure of the cytoplasmic ground
substance in erythrophores during pigment aggrega¬
tion and dispersion. I. A study using whole-cell
preparation in stereo high voltage electron micros¬
copy. J. Cell Biol., 75: 541-558.

16 Obika, M., Turner, W. A., Negishi, S., Menter, D.
G., Tchen, T. T. and Taylor, J. D. (1978) The
effect of lumicolchicine, colchicine and vinblastine
on pigment migration in fish chromatophores. J.
Exp. Zool., 205: 95-109.

17 Wise, G. E. (1969) Ultrastructure of amphibian
melanophores after light-dark adaptation and hor¬
monal treatment. J. Ultrastruct. Res., 27: 472-485.

ZOOLOGICAL SCIENCE 5: 959-964 (1988)

© 1988 Zoological Society of Japan

A Simple and Efficient Method for Photometric Estimation
of the State of Pigment Aggregation in Fish Melanophores

Nils Grundstrom1, Hans Sundgren2, Jan-Olof G. Karlsson1 3
and Hans Elwing2

1 Department of Pharmacology, 2Department of Applied Physics and
3 Department of Biology, Linkoping University,

581 85 Linkoping, Sweden

ABSTRACT — The existing photometric methods for the assessment of the state of pigment aggregation
in fish scale melanophores are not well suited for pharmacological or related investigations when an
appreciable number of scales are studied simultaneously. We have therefore developed a method which
make use of a new photometric apparatus, a scale photometer, for the assessment of the state of pigment
aggregation; this apparatus allows a pharmacological methodology that is simple and time-efficient, but
also allows investigations of time dependent processes, initialized by drugs or electrical nerve
stimulation. In this article we describe the method and the novel scale photometer. The method is
evaluated by means of the aggregating responses elicited either by noradrenaline and medetomidine or
by electrical field stimulation. The method is discussed in relation to previously applied methods and
some potential applications are suggested.

INTRODUCTION

The study of pigment cells with translocatable
pigment granules, chromatophores, has for a long
time attracted large interest [1] and a number of
methods for assessing the state of pigment
aggregation in chromatophores have been em¬
ployed.

We have recently attempted a pharmacological
characterization of the pigment aggregating adre¬
noceptors of fish melanophores using isolated
scales [2, 3]. In our opinion the most efficient
method for estimation of the degree of pigment
aggregation in response to various pharmacologic¬
al agents has been an ocular method [2], even if a
photometric method in many aspects would be
preferable. The existing photometric methods,
however, typically include a microscope stage in
the measuring setup and only one preparation can
be studied at a time [4-6], this make these
methods both expensive and impractical for phar¬
macological or related applications.

Accepted February 9, 1988
Received December 2, 1987

We have therefore developed a method which
make use of a new photometric apparatus for the
assessment of the state of pigment aggregation;
this apparatus allows a pharmacological methodol¬
ogy that is simple and time-efficient, but also
allows investigations of time dependent processes,
initialized by drugs or electrical nerve stimulation,
since it is easy to connect the apparatus to a chart
recorder.

In this article we describe our method and the
novel scale photometer. The method is discussed
in relation to previously applied methods and some
potential applications are suggested.

MATERIAL

Isolated fish scales were obtained from the dark
areas of the dermis of the Cuckoo Wrasse ( Labrus
ossifagus L.) as previously described [2]. The
isolated scales were suspended in a saline buffer
solution of the following composition (concentra¬
tions in mM): NaCl 150.0, KC1 5.2, CaCl2-2H?0
2.9, MgS04-7H20 1.8, Na?P04-2H?0 2.4,

NaHCC>3 17.9 and glucose 5.6 [7], The solution
was equilibrated with 5% CO? in 02 and kept at

960

N. Grundstrom, H. Sundgren, et al.

20° and pH 7.3.

The following drugs were used: 1-noradrenaline
bitartrate (Sigma Chemical Company, St. Louis,
Mo., USA), medetomidine hydrochloride (Far-
mos Group Ltd, Turku, Finland) and yohimbine
hydrochloride (Sigma Chemical Co.). The drugs
were dissolved in saline buffer solution.

METHODS

Ocular estimation of aggregation

The scales were placed on glass microscope
slides with the dermal side down and immersed in
50 ju\ of buffer. The scales were viewed in a
microscope (Leitz SM Lux) and the state of
aggregation was evaluated according to a modified
melanophore index [2, 8]. This index estimates the
degree of pigment aggregation by the use of a scale
ranging from 1 to 5 (with half-step resolution); 1
denotes complete aggregation and 5 denotes com¬
plete dispersion of the pigment.

During the course of the experiments the mi-
croscopist was unaware of what kind of agent, if
any, that was applied to a scale.

The scale photometer

In combination with a well trained observer the
ocular method is simple, sensitive and useful. The
result of the method however is highly dependent
on the ability of the observer. In order to minimize
the influence of the observer, increase the resolu¬

tion, sensitivity and reproducibility and to make
the observing more convenient we have developed
a simple apparatus, the scale photometer.

In the state of complete pigment dispersion the
melanosomes are spread out and all pigment parti¬
cles are completely exposed to the incoming light.
In the opposite state, when maximal aggregation
occurs, the melanosomes cluster and pigment
particles are only partially exposed to the incoming
light. It is then expected that the amount of
transmitted light through the fish scale is de¬
pendent on the state of pigment dispersion. Con¬
sequently the basic principle of an instrument
would be to detect transmitted visible light through
the pigmented area of a fish scale.

The measuring cuvettes of the scale photometer
(see Fig. 1A) are arranged five in a row in a sliding
multi sample holder which can be slid into one out
of five measuring positions. The multi sample
holder is removable and can easily be exchanged
with other identical multi sample holders. The
bulk material of the multi sample holder is black
acrylic plastic. The use of black material mini¬
mizes the influence of ambient light. The cuvettes
are milled in the bulk material and covered by
glass on two sides (see Fig. 1A). The top of the
cuvettes are open to allow mounting of the scales
and the addition of solutions. A scale is slid into a
U-shaped groove on one of the sides of the cuvette
(see Fig. 1A) and is then clamped into position by
inserting a plastic strip into the upper part of the
groove. The epidermal side of the scale is oriented

Fig. 1. Schematic representations showing the essential parts of the scale photometer (A) and the principles of
operation (B). The light source is positioned to the left in the figures and consists of a light emitting diode (LED)
and a reference light detector (that is part of an intensity controller); the main light detector is found to the right
in the figures. In between the light source and detector are the cuvette, containing the scale, and two apertures
(12 mm apart). All parts shown in (A) with the exception of glass slides and electronic components are made of
black acrylic plastic.

Simple Evaluation of Pigment Aggregation

961

towards the interior of the cuvette.

As light source a light emitting diode (LED) is
used. This LED (H-2000, Stanley Electric Co.,
Ltd., Japan) has a high luminous intensity of about
2000 millicandela. The peak wavelength is 660 nm
with a spectral half bandwidth of 25 nm. The light
from the LED is detected by a reference photode¬
tector and a detector for the transmitted light
(both detectors are PIN silicon photo diodes,
BPW34, Siemens, W. Germany). The signal from
the reference photodetector is fed to a light in¬
tensity controller (see Fig. IB) and compared to a
preset value. The current through the LED is
controlled by the light intensity controller resulting
in a very stable light intensity. The emitted light
passes through an aperture (diameter 1.5 mm), the
cuvette containing the fish scale, another aperture
(diameter 1.5 mm) and finally the remaining light
is detected by the photodetector. The signal from
the detector is converted to a voltage output in the
current amplifier and the voltage level is directly
proportional to the light transmission through the
scale.

Reducing the light beam through the apertures
makes it possible to choose a very small detected
area of the scale. The investigated area on the
scale is 1.8 mm2. The radiant sensitivity of the PIN
silicon photo diode is about 0.4 A/W at 660 nm and
the active area is 7.3 mm2. The maximum light
power exposed to the scale is less than 10 /AV
which minimizes heating of the scale and buffer
medium. Furthermore the power unit is separated
from the rest of the device to avoid excess heating
of the scales when the multi sample holders are
positioned in the scale photometer.

In the prototype version of the scale photometer
an external digital voltmeter was used and a two-
channel recorder was connected.

Photometric estimation of aggregation

The scales were mounted in the cuvettes and
immersed in 50 jA of the saline buffer. The trans¬
mission range was defined by the intensities mea¬
sured when the light beam was either totally
interrupted (0%) or allowed to pass through a
buffer-filled cuvette without a scale in position
(100%).

Stimulation of the melanophores

The melanophores were stimulated to aggregate
their pigment granules, either by the addition of a
pharmacological agent or by electrical field stim¬
ulation of intrinsic nerves. The electrical field
stimulation was performed by means of two silver
wire electrodes (0.2 mm in diameter) that was
mounted, 4 mm apart, in a cuvette. A Grass S88
stimulator equipped with an isolation unit (Grass
SIU 5) was used to deliver trains of varying
duration at a frequency of 20 Hz (1 msec biphasic
square pulses, 60 V nominally out from the stimu¬
lator).

All data are presented as means + S.E.M.

RESULTS

The melanophores of isolated scales from Lab-
rus ossifagus maintain a state of pigment disper¬
sion in saline buffer solution. The dispersed state
was also maintained when the scales were posi¬
tioned for measurement of transmission in the
scale photometer. It was noted that the interfer¬
ence from stray light was negligible. In Figure 2 A
the size and time course of the transmission re¬
sponse after addition of two pharmacological
agents are shown. Noradrenaline completely
aggregated the melanophores, as confirmed in the
microscope, whereas yohimbine (alpha2-
adrenoceptor selective) effectively antagonized the
aggregation. Repeated experiments, with addition
of noradrenaline, on different scales (n — 25)
showed a typical baseline transmission of 38 + 2%
and a stimulated transmission of 57 + 2% after
maximal aggregation as confirmed in the micro¬
scope.

The function of the scale photometer was also
evaluated in concentration-response tests and
compared with our ocular method for estimation
of pigment aggregation. Accumulated concentra¬
tion-response curves were obtained using either of
the two methods. In both methods the prepara¬
tions were allowed a 5-min resting period in day
light after each addition of drug. Two adrenergic
drugs were used: noradrenaline and medetomi-
dine, a potent alpha2-adrenoceptor selective ago¬
nist [9]. Following the resting period the state of

962

N. Grundstrom, H. Sundgren, et al.

55

50

45

40

35

Fig. 2. Effects of drugs on the state of aggregation of melanophores. (A) shows a representative recording of light
transmission. The aggregating effect of noradrenaline (NA) and its reversal by the alpha2-adrenoceptor selective
antagonist yohimbine (YOH) can be observed. In (B) cumulative concentration-effect curves of aggregation are
shown. Filled symbols indicate data evaluated by ocular estimation (for comparison the melanophore index scale
was transformed to a percentage scale) and open symbols indicate data evaluated by the photometric method.
Two agonistic drugs were used: medetomidine (a, a) and noradrenaline (o, #)• The vertical bars indicate
S.E.M. (n=5).

AGGREGATION (%)

TIME (min)

TIME (s)

0 5 10 15 20 25

TIME (min)

Fig. 3. Typical effects of electrical stimulation on the state of aggregation of melanophores. In (A) the transmission
response to a short stimulation (10 s, curve I) and a more prolonged stimulation (60 sec, curve II) are shown. In
(B) the reproducibility of repeated short (10 sec) stimulations are shown. The dots indicate the incidence of a
stimulation.

aggregation was evaluated using either of the two
methods. The results of these experiments are
presented in Figure 2B as normalized concentra¬
tion-aggregation curves.

The size and time course of the transmission
response following electrical nerve stimulation of
the scales are shown in Figure 3A. The response
is, as expected, much faster than after addition of
drugs (c.f. Fig. 2A). The reproducibility of re¬
peated electrically induced aggregation was also
tested as shown in Figure 3B.

DISCUSSION

In this paper we describe an efficient method for
estimation of the state of aggregation of mela¬
nophores on fish scales; the method includes a
novel apparatus, a scale photometer.

The advantages of using a dedicated apparatus
for the estimation of aggregation are several; e.g.
there is no need for a microscope as part of the
setup and the apparatus can be made very efficient
for the evaluation of the state of aggregation when
several scales are used in the experiments, which is

Simple Evaluation of Pigment Aggregation

963

the case in pharmacological work.

The apparatus was designed to be simple and
reliable. The light source was chosen to be a
powerful light emitting diode, emitting in the red.
The advantage of using red light is that the in¬
terfering effects of other chromatophores than the
melanophores, i.e. xantophores and erythro-
phores, can be virtually eliminated [10].

The recording obtained by the use of the scale
photometer (see Figs. 2A and 3A, B) are compa¬
rable to the recordings obtained by the use of other
methods [e.g. 4, 6]. It can be noted that the
change of transmission following maximal stimula¬
tion of the melanophores by drugs (Fig. 2A) or by
electrical field stimulation (Fig. 3A) is of compara¬
ble size. We have previously shown, by the use of
tetrodotoxin [4], that the electrical field stimula¬
tion applied in this work activates intrinsic nerves
only and not the melanophores directly. The
aggregating response to repeated electrical stim¬
ulation (Fig. 3B) shows a high degree of reproduci¬
bility, which may be of importance in experiments
when previously induced responses are used as
controls to responses elicited after treatment.

The scale photometer is especially useful when
concentration response curves are obtained since
several multi sample holders (with five cuvettes
each) can be used simultaneously. In Figure 2B
the aggregating responses elicited by two
adrenergic agonists are shown as a function of
concentration; the response to each agonist was
evaluated either by the ocular or the photometric
method. It can be concluded from the figure that
the two methods give very similar results, especial¬
ly if the deviations are taken into consideration.
Furthermore, the form of the curve for the two
agonists agree regardless of method. It is evident
from the curves in Figure 2B that the deviations
are substantial, this appears however to be more a
characteristic of the scale melanophores them¬
selves since similar deviations appear regardless of
the method applied.

We have developed a method which apply a new
photometric apparatus for the assessment of the
state of pigment aggregation in melanophores.
This apparatus, a scale photometer, allows a
methodology that is simple and time-efficient and
the method could find potential applications espe¬

cially in the pharmacological field where it could,
e.g., complement existing models for characteriza¬
tion of alpha2-selective drugs. The method could
also be of potential value as an assay during
purification of pertussis toxin [11] or even used as
part of a diagnostic method for whooping cough.

ACKNOWLEDGMENT

The excellent technichal assistance by Mr. P.-E. Ster¬
ner is gratefully acknowledged. We also wish to thank
Prof. J. O. Stromberg at Kristineberg Marine Biological
Station for placing laboratory resources at our disposal.
Medetomidine was a gift from Farmos Group Ltd,
Turku, Finland. This work was supported by grants from
the Swedish Technical Development Board (86-4494),
Trion AB and Centrala Forsdksdjursnamnden (87-18).

REFERENCES

1 Fujii, R. (1969) Chromatophores and pigments. In
“Fish Physiology, vol. 3”. Ed. by W. S. Hoar and D.
J. Randall, Academic Press, New York, pp. 307-
353.

2 Andersson, R. G. G., Karlsson, J. O. and Grund-
strom, N. (1984) Adrenergic nerves and the alpha2-
adrenoceptor system regulating melanosome
aggregation within fish melanophores. Acta Physiol.
Scand., 121: 173-180.

3 Karlsson, J. O. G., Andersson, R. G. G., Elwing,
H. and Grundstrom, N. (1987) Comparative studies
on nerve- and noradrenaline-induced melanosome
aggregation within different species of fish. Comp.
Biochem. Physiol., 88C: 287-291.

4 Fujii, R. and Novales, R. R. (1968) Tetrodotoxin:
Effects on fish and frog melanophores. Science, 160:
1123-1124.

5 Nagahama, H. and Katayama, H. (1982) Ba-
pulsations of innervated and denervated mela¬
nophores of the teleost fish in the presence of
adrenaline. J. Sci Hiroshima Univ., Ser B. Div. 1,
30: 159-171.

6 Oshima, N. and Fujii, R. (1984) A precision photo¬
electric method for recording chromatophore re¬
sponses in vitro. Zool. Sci., 1: 545-552.

7 Nilsson, S. (1971) Adrenergic innervation and drug
responses of the oval sphincter in the swimbladder
of the cod ( Gadus morhua). Acta Physiol. Scand.,
83: 446-453.

8 Hogben, L. T. and Slome, D. (1931) The pigmen¬
tary effector system VI. The dual character of
endocrine coordination in amphibium colour
change. Proc. R. Soc. B.. 108: 10-53.

9 Savola, J.-M., Ruskoaho, H., Puurunen J.,

964

N. Grundstrom, H. Sundgren, et al.

Salonen, J. S. and Karki, N. T. (1986) Evidence for
medetomidine as a selective and potent agonist of
a2-adrenoceptors. J. Auton. Pharmacol., 5: 275-
284.

10 Fujii, R. (1961) Demonstration of the adrenergic
nature of transmission at the junction between mela-
nophore-concentrating nerve and melanophore in

bony fish. I. Fac. Sci., Univ. Tokyo, Sect. IV, 9:
171-196.

11 Karlsson, J. O. G., Grundstrom, N., Wikberg, J. E.
S., Friedman, R. and Andersson, R. G. G. (1985)
The effect of pertussis toxin on alpha-2-
adrenoceptor-mediated pigment migration in fish
melanophores. Life Sci., 37: 1043-1049.

ZOOLOGICAL SCIENCE 5: 965-971 (1988)

© 1988 Zoological Society of Japan

Non-Synaptic Release of Transmitter-Containing Vesicles from
the Enteric Neurons of the Rat Small Intestine

Yasuhisa Endo

Department of Applied Biology, Kyoto Institute of Technology,
Matsugasaki, Sakyo, Kyoto 606, Japan

ABSTRACT — Exocytotic release of transmitter-containing vesicles from the neurons of the rat small
intestine was investigated by electron microscopy using a perfusion method of high K+ Ringer solution
to stimulate release of transmitters. Omega-shaped figures suggestive of exocytosis of large cored
vesicles were frequently found at the non-synaptic sites. Exocytosis occurred not only in the surface of
nerve fascicles but also in the sites facing other neuronal processes and Schwann cell’s processes. These
results suggest that the terminal varicosities may not always release their chemical messengers only in
the sites directly facing the target tissues.

INTRODUCTION

It is generally accepted that neurons have var¬
ious chemical messengers as neurotransmitters and
release them toward the effector cells through
synapses. However, recent electron microscopic
studies demonstrated the occurrence of non-
synaptic release of neurotransmitters from the
neurons in the central nervous system [1] and the
adrenal medulla [2].

Most of the enteric neurons of mammals have
various neuropeptides which are packed in large
cored vesicles (cf. [3]). Several transmission elec¬
tron microscopic studies on the enteric nervous
system have been done, but exocytotic release of
large cored vesicles from the neurons has not been
described (cf. [4]). Recently, Endo and Kobayashi
[5] studied on the three-dimensional structure of
the autonomic groundplexus in the small intestine
of guinea pigs, by scanning electron microscopy
using a HCl-digestion method to eliminate connec¬
tive tissues. They revealed that the terminal
portion of the enteric nervous system was a con¬
tinuous network of unmyelinated nerve fascicles
and the many, varicose neuronal processes were
exposed to the surface of unmyelinated nerve
fascicles. They postulated that these exposed area

Accepted January 6, 1988
Received October 21, 1987

of neuronal processes may be the sites of interac¬
tion between the enteric neurons and their effector
cells.

In the present study, I investigated the exocyto¬
tic release of transmitter-containing vesicles from
the neurons of the rat small intestine by transmis¬
sion electron microscopy. In order to stimulate the
exocytotic release of them, perfusion of high K+
Ringer solution was applied (cf. [6]). Tannic
acid-glutaraldehyde-osmium tetroxide fixation is
known as an useful method to demonstrate the
exocytotic figures of secretory granules, because
this fixation renders the extracellular secretory
substances highly electron dense and does not
affect the ultrastructure of other cellular elements
[7, 8]. This fixation method was used in this study,
combining with a perfusion of high K+ Ringer
solution.

MATERIALS AND METHODS

Adult male Wistar rats (100-200 g body weight)
were anesthetized by an intraperitoneal injection
of sodium pentobarbital (50 mg/kg body weight).
Three animals were perfused with 50 ml of normal
Ringer solution (145 mM NaCl, 5.6 mM KC1, 2.3
mM CaCl2) through the left ventricle, and then
perfused with 200 ml of the fixative containing
2.5% glutaraldehyde and 0.5% tannic acid in 0.1
M phosphate buffer, pH 7.4. Five animals were

966

Y. Endo

perfused with 50 ml of high K+ Ringer solution (30
mM NaCl, 120 mM KC1, 2.3 mM CaCl2) to stimu¬
late the release of various transmitters, and then
perfused with the fixative as mentioned above.
The duodenum was removed and immersed in the
fresh fixative for 2-3 hr. After rinsing in the
phosphate buffer, the tissues were cut into small
pieces and postfixed for 1 hr with 1% 0s04 or
Karnovsky’s reduced 0s04 [9] which was a mix¬
ture consisting of one part 2% 0s04 and one part
3% potassium ferrocyanide. The tissues were
dehydrated in a graded series of ethanol, passed
through propylene oxide and embedded in Epon
812. Thin sections were stained with uranyl ace¬
tate and lead citrate, and examined with a Hitachi
H-700 electron microscope at 75 kV.

RESULTS

The enteric nervous system consisted of ganglia
(myenteric and submucous) and unmyelinated
nerve fascicles. The latter were distributed
throughout the gut wall, i.e. the longitudinal and
circular muscle coats, submucosa and mucosa.
Most of neuronal processes in the fascicles had
small clear vesicles and/or large cored vesicles
which were thought to include various chemical
messengers, i.e. acetylcholine, catecholamines and
neuropeptides (cf. [3, 4]). Conventional synaptic
junctions were sometimes found in the ganglia
(Fig. 1), but rarely seen in other regions.

Omega-shaped figures indicative of exocytosis of
neurotransmitters were rarely seen in the enteric
nervous system of the rats perfused with normal
Ringer solution. In the rats stimulated by perfu¬
sion of high K+ Ringer solution, omega-shaped
figures of transmitter-containing vesicles were fre¬
quently encountered. Therfore, the results men¬
tioned below were obtained in the rats perfused
with high K+ Ringer solution.

In the ganglia, small clear vesicles accumulated
at the synaptic sites where characteristic thickening

of cell membranes was obvious, but large cored
vesicles generally located at the portion distant
from the synaptic sites (Fig. 1). The exocytotic
figures of large cored vesicles were found in the
interspaces between neuronal processes and be¬
tween neuronal process and cell body, where no
membrane specialization indicative of synaptic
contact was found (Fig. 1).

In the muscular coats, a large number of un¬
myelinated nerve fascicles were distributed among
smooth muscle cells, especially in the innermost
layer of the circular muscle coat (i.e. the deep
muscular plexus). Unmyelinated nerve fascicles
consisted of Schwann cells and many neuronal
processes (Fig. 2). Irregularly-shaped processes of
Schwann cells enclosed several neuronal pro¬
cesses. Some of neuronal processes were exposed
directly to the surface of nerve fascicles. General¬
ly, nerve fascicles and smooth muscle cells ran at a
distance, and both of them were wrapped with
each basement membrane (Fig. 2). A few nerve
fascicles were in close contact with smooth muscle
cells, where the basement membrane of both ele¬
ments were absent (Fig. 3). However, the synaptic
specialization was not found at these contacts (Fig.
3). Here, small clear vesicles accumulated and
omega-shaped figures indicative of exocytotic re¬
lease of their contents were sometimes found (Fig.
3, arrow). Omega-shaped figures of small clear
vesicles also occurred in the surface of neuronal
processes relatively distant from smooth muscle
cells (Fig. 4, arrow). Omega-shaped figures of
large cored vesicles occurred in the nerve fascicles
distant from smooth muscle cells, not only in the
surface of nerve fascicles (Fig. 5) but also in the
interspaces between neuronal processes (Fig. 6).

In the lamina propria mucosae and the muscu-
laris mucosae, numerous unmyelinated nerve
fascicles were distributed, but the synaptic contact
between neuronal processes and other cells was
not found. Omega-shaped figures of large cored
vesicles were found not only in the surface of nerve

Fig. 1. A synaptic ending in the myenteric ganglion. Small clear vesicles accumlate at the presynaptic site.
Exocytotic figure of large cored vesicle (arrow) is seen at the site distant from the synapse. M: mitochondria. X
65,000.

Fig. 2. Unmyelinated nerve fascicle and smooth muscle cells (S) in the circular muscle coat. A nerve fascicle consists
of a Schwann cell and many axons. Some axons are exposed to the surface of nerve fascicle. F: fibroblasts, N:
nucleus of Schwann cell. X 16,000.

Non-Synaptic Exocytosis in Neurons

967

968

Y. Endo

Fig. 3. A nerve fascicle in close contact with smooth muscle cell (S) in the circular muscle coat. Exocytotic figure of
small clear vesicle (arrow) is seen. X 40,000.

Fig. 4. Exocytotic figure of small clear vesicle (arrow) at the site distant from smooth muscle cell (S) in the circular
muscle coat. X 55 ,000.

fascicles (Fig. 7) but also in the interspaces be¬
tween neuronal processes and between neuronal
and Schwann cell’s processes (Fig. 8).

In spite of applying the tannic acid fixation, the
electron density of the exocytosed vesicles was
relatively low when compared with the results of
previous studies [1, 2, 7, 8]. This may be due to the
use of Karnovsky’s reduced Os04 [9] which is
known as the fixative preserving membrane struc¬
ture excellently.

DISCUSSION

In the present study, I demonstrated by electron
microscopy that the exocytosis of large cored vesi¬
cles and small clear vesicles occurred in the termi¬
nal varicosities of the enteric neurons. The exocy¬
tosis theory has been commonly accepted as a
release mechanism of secretory granules from the

neurosecretory cells, various endocrine and ex¬
ocrine cells (cf. [6, 10, 11]). However, there have
been only a few studies which demonstrated the
exocytotic release of neurotransmitters in the au¬
thentic neurons. According to Nagasawa [6], the
frequency of encountering the exocytotic figures is
extremely rare in the neurosecretory axon termi¬
nals of the posterior pituitary when compared with
other secretory cells such as the anterior pituitary
or the adrenal medulla. But the stimulation of a
solution containing high KC1 and CaCl2 increases
the frequency of the exocytotic figures. In the
present study, similar phenomenon was found also
in the enteric neurons. Although the detailed
mechanism of KCl-induced exocytosis is not fully
understood, the sites and mode of transmitter
release seem to be similar to those of the normal
condition.

Generally, the exocytotic release of small clear

Non-Synaptic Exocytosis in Neurons

969

Fig. 5. Exocytotic figure of large cored vesicle (arrow) at the surface of nerve fascicle, distant from smooth muscle
cell (S) in the circular muscle coat. X 67,000.

Fig. 6. Exocytotic figure of large cored vesicle (arrow) at the site facing other neuronal process, in the circular
muscle coat. X 61,000.

vesicles in the neuromuscular junctions is known
to occur at the conventional synaptic sites [12]. In
the present study, I found the exocytotic figures of
small clear vesicles not only in the vicinity of
smooth muscle cells but also in the sites distant
from smooth muscle cells. This difference may be
caused by the absence of apparent synaptic contact
between the enteric neurons and smooth muscle
cells.

In the central nervous system of the rabbits,
Nitsch and Rinne [13] observed the exocytotic
figures of large cored vesicles at the synaptic
junctions of the hippocampal mossy fibers.
However, recent electron microscopic studies [1,
2] demenstrated the presence of non-synaptic re¬
lease of large cored vesicles as well as the synaptic
and neurohemal release. In the present study, I
also demonstrated the non-synaptic release of
large cored vesicles in the enteric neurons.

As reported in the previous scanning electron
microscopic study [5], terminal portion of the
enteric nervous system consists of a continuous
network of unmyelinated nerve fascicles and does
not have a specialized end apparatus facing the
effector cells. Therefore, the non-synaptic release
coud be expected from these observations. But it
was unexpected that the exocytosis also occurred
in the sites facing other neuronal processes and
Schwann cell’s processes. Similar phenomenon
was found in the neurosecretory axon terminals of
the rat posterior pituitary, where the exocytotic
figures of large cored vesicles occurred not only in
the sites surrounding blood capillaries but also in
the sites facing other neuronal processes (Y.
Endo, unpublished data). These facts indicate that
the terminal varicosities of neurons may not always
release their chemical messengers only in the sites
directly facing the target tissues.

970

Y. Endo

Fig. 7. Exocytotic figure of large cored vesicle (arrow) at the surface of nerve fascicle, distant from smooth muscle
cell (S) in the muscularis mucosae. X 69,000.

Fig. 8. Exocytotic figure of large cored vesicle (arrows) at the site facing Schwann cell’s process (SC), in the lamina
propria mucosae. X 47,000.

ACKNOWLEDGMENTS

The author thanks Prof. S. Kobayashi (Department of
Anatomy, Yamanashi Medical College) and Prof. Y.
Waku (Department of Applied Biology, Kyoto Institute
of Technology) for their encouragements and valuable
discussion. This work was supported by a Grant-in- Aid
for scientific research from the Ministry of Education,
Science and Culture of Japan (No. 62770012).

REFERENCES

1 Buma, P. and Roubos, E. W. (1986) Ultrastructural
demonstration of nonsynaptic release sites in the
central nervous system of the snail Lymnaea stagna-
lis , the insect Periplaneta americana, and the rat.
Neuroscience, 17: 867-879.

2 Golding, D. W. and Pow, D. V. (1987) ‘Neuro¬
secretion’ by a classic cholinergic innervation appa¬
ratus. A comparative study of adrenal chromaffin
glands in four vertebrate species (teleosts, anurans.

mammals). Cell Tissue Res., 249: 421-425.

3 Furness, J. B. and Costa, M. (1980) Types of nerves
in the enteric nervous system. Neuroscience, 5: 1-
20.

4 Gabella, G. (1979) Innervation of the gastrointes¬
tinal tract. Int. Rev. Cytol., 59: 129-193.

5 Endo, Y. and Kobayashi, S. (1987) A scanning
electron microscope study on the autonomic ground-
plexus in the lamina propria mucosae of the guinea-
pig small intestine. Arch. Histol. Jpn., 50: 243-250.

6 Nagasawa, J. (1977) Exocytosis: the common re¬
lease mechanism of secretory granules in glandular
cells, neurosecretory cells, neurons and para-
neurons. Arch. Histol. Jpn., 40 (Suppl.): 31-47.

7 Roubos, E. W. and van der Wal-Divendal, R. M.
(1980) Ultrastructural analysis of peptide-hormone
release by exocytosis. Cell Tissue Res., 207: 267-
275.

8 Endo, Y. and Nishiitsutsuji-Uwo, J. (1982) Ex¬
ocytotic release of secretory granules from endo¬
crine cells in the midgut of insects. Cell Tissue Res.,

Non-Synaptic Exocytosis in Neurons

971

222: 515-522.

9 Karnovsky, M. J. (1971) Use of ferrocyanide-
reduced osmium tetroxide in electron microscopy.
Proc. 14th Annu. Meet. Am. Soc. Cell Biol., pp.
146.

10 Douglas, W. W. and Nagasawa, J. (1971) Mem¬
brane vesiculation at sites of exocytosis in the
neurohypophysis and adrenal medulla: a device for
membrane conservation. J. Physiol., 213: 94-95.

11 Normann, T. C. (1976) Neurosecretion by exocyto¬

sis. Int. Rev. Cytol., 46: 1-77.

12 Heuser, J. E. and Reese, T. S. (1981) Structural
changes after transmitter release at the frog neuro¬
muscular junction. J. Cell Biol., 88: 564-580.

13 Nitsch, C. and Rinne, U. (1981) Large dense-cored
vesicle exocytosis and membrane recycling in the
mossy fibre synapses of the rabbit hippocampus
during epileptiform seizures. J. Neurocytol., 10:
201-219.

>

ZOOLOGICAL SCIENCE 5: 973-978 (1988)

© 1988 Zoological Society of Japan

Fibrillar System and Possible Control Mechanism for the Cycle
of Contraction and Elongation of Spirostomum ambiguum

Hideki Ishida* 1, Yoshinobu Shigenaka2 and Masako Imada

Laboratories of Cell Biology, Faculty of Integrated Arts and
Sciences, Hiroshima University, Hiroshima 730, Japan

ABSTRACT — A large heterotrichous ciliate, Spirostomum ambiguum, shows the characteristic twisting
contraction when stimulated by certain mechanical or chemical factors. This contraction is attributed to
a special contractile fibrillar system which is termed myoneme, while the subsequent elongation after the
contraction occurred might be induced by another type of fibrillar system, longitudinal microtubular
sheets (LMSs). The myoneme is composed of a great number of fine filaments which are heavily packed
and lie in parallel with each other. The diameter of individual filament varies from 3 to 5 nm in the
elongated state but raises up to 7 to 9 nm in the contracted state. Every nonciliated basal body has a
rootlet-like structure, which is attached to the myoneme not only in the contracted state but also in the
elongated state. If the myoneme and LMSs have antagonistic functions to each other, the force of each
fibrillar system is assumed to be transmitted by these rootlet-like structures. On the other hand, anterior
fiber sheet was found to be attached to the LMS in the extended state but detached from that in the
contracted state of organism. From these observations, the switching or control mechanism was
proposed and discussed for the cycle of contraction and elongation of the organism.

INTRODUCTION

Among a number of heterotrichous ciliates,
there are some typical contractile ones such as
Condylostoma , Spirostomum and Stentor. Howev¬
er, Spirostomum ambiguum is the only one show¬
ing its own characteristic twisting contraction [1-
3]. Although the contraction of the cell body of
these organisms is easily induced by various kinds
of stimuli this is believed to be caused directly by a
special contractile fibrillar system which is termed
myoneme [1-6] or M-band [7]. On the other hand,
the subsequent elongation of the organism after
the contraction occurred has been proposed to be
attributed mainly to another fibrillar system, longi¬
tudinal microtubular sheets (LMS) [2, 3].

Regarding to the twisting contraction mecha¬
nism of Spirostomum , Yogosawa-Ohara et al. [3]
have proposed the following sequence: (1)

Accepted January 28, 1988
Received January 6, 1988

1 Present address: Biological Institute, Shimane Uni¬
versity, Matsue 690, Japan.

2 To whom reprint requests should be forwarded.

Shortening along the long axis occurs due to the
activity (probably sliding) of myonemal filaments,
(2) at the moment when the body length decreases
all of the neighboring LMSs slide with each other
possibly involving the anterior fiber sheet, and (3)
the curving of each LMS is derived from the sliding
between the LMS and the sasa structure (SS) by
means of obliquely arranged projections on the SS
which is a bamboo leaf-like structure attached to
the proximal region of every LMS. Coordination
of these phenomena might result in the twisting
contraction of the organism.

However, the ultrastructural details have not
thoroughly been studied especially with respect to
the switching mechanism of contraction and
elongation of Spirostomum. Therefore, the pre¬
sent study aimed to examine the details of
myoneme, longitudinal microtubular sheets and
their associated structures for elucidating the
structural changes between two states of contrac¬
tion and elongation. As the results, a possible
mechanism controlling the contraction-elongation
cycle has been proposed here.

974

H. Ishida, Y. Shigenaka and M. Imada

MATERIALS AND METHODS

Organisms

Live samples of Spirostomum ambiguum Ehren-
berg were originally collected and kindly supplied
by Dr. T. Suzaki from the Research School of
Biological Sciences, the Australian National Uni¬
versity, Canberra, A. C. T. 2601, Australia. They
were cultured in our laboratory using 0.01% Knop
solution as culture medium containing some boiled
wheat grains and small quantity of 0.5% hay
infusion at about 24°C. Other special food sources
were not supplied to the medium. Subculturing
was carried out at a regular interval of about 3
weeks.

Light Microscopy

Light micrographs were obtained from the living
organism in both states of contraction and elonga¬
tion by using a differential interference microscope
(Olympus, BH). To cause the contraction, mechan¬
ical stimuli were applied to the organisms by
means of tapping the specimen-loaded glass slide
or electrical stimuli (35 V, DC) were given directly
to the organism. In case of electrical stimulation,
the organism was placed at the center between the
two platinum electrodes (5 mm in distance).

Electron Microscopy

Prior to the fixation, living organisms were
rinsed twice to clean them up with fresh culture
medium, 0.01% Knop solution. Preparations in
the elongated state were prepared by treating the

organisms primarily with the relaxation medium
containing 10 mM EGTA, 3 mM MgSOt and 10
mM phosphate buffer (pH 6.8) for 10 min just
before the fixation. The other preparations in the
contracted state were obtained by stimulating
them in the fresh culture medium (0.01% Knop
solution) by means of tapping the tube containing
the organisms.

The pre-fixation was performed in 1% glutaral-
dehyde with 33 mM phosphate buffer (pPI 6.8) for
30 min at 0°C. The fixed samples were rinsed with
33 mM phosphate buffer (pH 6.8). Following that,
they were post-fixed in 1% Os04 with 33 mM
phosphate buffer (pH 6.8) for 30 min at 0°C. After
the double fixation, the samples were dehydrated
with a graded ethanol series and embedded into
Spurr’s low viscosity embedding medium [8].

Ultrathin sections were produced with a Porter-
Blum ultramicrotome (MT-1) equipped with glass
knives. The sections were stained with 3%
aqueous uranyl acetate for 7 min and Reynolds’
lead citrate stain [9] for 3 min at room tempera¬
ture. Observations were carried out under a
transmission electron microscope (JEOL, J EM
100S) at the accelerating voltage of 80 kV.

RESULTS

A large heterotrichous ciliate, Spirostomum
ambiguum, shows the characteristic contraction
when applied by chemical, mechanical or electrical
stimuli [1-5, 10]. Figure 1 shows the two states of
the same free-swimming organism; Figure la is in
the elongated stage and Figure lb in the electrical-

Fibrillar System of Spirostomum

975

ly stimulated and contracted state.

When observed under an electron microscope,
the cortical region of organism is characterized by
three typical types of fibrillar systems and their
associated structures; sub-pellicular microtubules,
longitudinal microtubular sheets and myonemes
(Fig. 2a). The myoneme is located at the tran¬
sitional plane between ectoplasm and endoplasm
and featured by being surrounded by various sizes
of vacuoles. Every myoneme is composed of a
great number of filaments which are heavily pack¬
ed and lie in parallel with each other inside the
myoneme. The diameter of individual filament

varies from 3 to 5 nm in elongated state of the
organism (Fig. 3a), but raises up to 7 to 9 nm in the
contracted state (Fig. 3b).

As a whole, the bundles of myonemal filaments
demonstrate a mesh-like distribution throughout
the cell body just beneath the arrays of longitudi¬
nal microtubular sheets as described by Yogosawa-
Ohara and Shigenaka [2]. In addition to that, the
present study revealed that major bundles of the
myonemal meshes ran in parallel with the body
axis in the elongated state of organism but altered
their axes diagonally in the contracted state. In the
contracted state, on the other hand, the myonemal

Fig. 2. Electron micrographs of vertical sections through the area of cell cortex. Microtubular sheets (mi) and
myoneme (my) lie just under the cell surface (Fig. 2a). An inserted electron micrograph (Fig. 2b) is of the
rootlet-like structure in the contracted state, in which a nonciliated posterior basal body was found to be
connected with the myoneme by the rootlet-like structure (arrowhead). Just under the cell membrane,
sub-pellicular microtubules could be seen (s). (X 24,500).

976

H. Ishida, Y. Shigenaka and M. Imada

Fig. 3. Electron micrographs of longitudinal sections through the myoneme in the extended (Fig. 3a) and the
contracted (Fig. 3b) states. The bundle of myoneme consists of 3 to 5 nm filaments in the extended state and 7 to
9 nm filaments in the contracted state. Myonemal filaments run in parallel with each other. (x40,700).

meshes became more compact since distances of
branch to branch of the myonemal meshes became
shorter.

As already known, the microtubular sheets run
longitudinally and just along the ciliary lines.
Typically in this area, a number of mitochondria
can be seen. The longitudinal microtubular sheets
(LMS) do not alter the structure of their own
components in both states of elongation and con¬
traction of the organism. Each LMS is composed
of about 20 microtubules which are arranged in
parallel with each other and connected by numer¬
ous links to one another, and derived from the
nonciliated one of every basal body pair and run
toward the posterior end of the organism. When
the cell shape changes from extended state to
contracted one, the overlapping LMSs increase in
number. At the same time, the center-to-center
distance of two neighboring and antero-posteriorly
arranged ciliary bases becomes shorter from 3.05

±0.05 /urn (n = 14) in the elongated state to 2.25 +
0.05 fxm (n=25) in the contracted state.

Along every ciliary line, there can be seen an
array of paired basal bodies which are ciliated and
nonciliated or barren. Every nonciliated posterior
basal body was found to be connected with the
myoneme by a rootlet-like structure (Fig. 2b),
which is consisted of a great number of fine
filaments and connected to the myoneme in both
states of contraction and elongation of the organ¬
ism. On the other hand, it is noteworthy that the
anterior fiber sheets derived from the space be¬
tween the pair of basal bodies were found to be
attached to the LMSs in the extended state (Fig.
4a) but to be detached from them in the contracted
state (Fig. 4b). This phenomenon has not been
noticed by Yogosawa-Ohara et al. [3] and other
investigators [1], although it might be closely re¬
lated to controlling the contraction-elongation cy¬
cle of the organism.

Fig. 4. Electron micrographs of the longitudinal microtubular sheets (LMSs). The anterior fiber sheets derived
between the pair of basal bodies (arrow heads) are attached to the LMSs in the extended state (Fig. 4a) but
become detached from them in the contracted state (Fig. 4b). (X 12,000).

Fibrillar System of Spirostomum

977

Just under the cell membrane, so-called sub-
pellicular microtubules could be seen (Fig. 2a).
These microtubules are closely associated to the
cell membrane and run in parallel with the ciliary
line, suggesting that they might have a cytoskeletal
role without being related to the contraction-
elongation cycle.

DISCUSSION

Just like in the present organism, another heter-
otrichous ciliate, Stentor coeruleus , also shows the
rapid contraction [11-13], although the style is not
twisting contraction. This organism also has the
similar fibrillar systems, myoneme and microtubu¬
lar sheets to those in Spirostomum. In this Stentor ,
every myonemal filament is known to be 4 nm in
diameter in the extended state. In the contracted
state, however, the filaments (10 to 12 nm in
diameter) appear instead of them, which are to be
of tubular profiles with a wall thickness of 4 to 5
nm. In the contracted state, the wall of filaments is
made up of four to six subunits.

As to the chemical nature of myoneme, Hobbs
et al. [14] have described in Spirostomum teres that
arrowhead decoration was not observed in cyto¬
plasmic filament bundles although the myosin sub¬
fragment S-l was introduced into the cells for
incubation under conditions suitable for actin dec¬
oration. Furthermore, Yogosawa-Ohara and
Shigenaka [2] have published in Spirostomum
ambiguum that cytochalasin B treatment did not
cause degradation of myonemal filaments even at a
higher concentration (50^g/ml). These observa¬
tions strongly suggest that the myonemal filaments
may not be actin-like, but may be similar to the
spasmonemal filaments of peritrichous ciliates,
Vorticella and Carchesium [15, 16] or the retrac¬
tion fiber filaments of a dinoflagellate, Ceratium

[17]-

On the other hand, microtubular sheets which
are derived from antero-posteriorly arranged cili¬
ary pairs are overlapping and might slide relatively
with each other. When contracted, the overlap¬
ping microtubular sheets increase their number in
cross setions as described by Huang and Pitelka
[12]. When the cell was fixed in isometric contrac¬
tion, the microtubular sheets were in the state of

elongation and the overlapping microtubular
sheets were at minimum in number. Moreover,
the internal structure of the myoneme altered in
the contracted state; the myonemal filaments be¬
came to be of tubules with the diameter of 10 to 12
nm as described above.

This observation has suggested that the
myoneme generates the motive force resulting into
cell shortening [11]. It is thought that the
myoneme and the microtubular sheets might func¬
tion as antagonistic elements to each other. That is
to say, the myoneme generates the motive force
for cell contraction, although the microtubular
sheets slide with each other to cause only cell
elongation.

The present Spirostomum demonstrates the fine
structures which are quite similar to those in
Stentor-, the myonemal filaments change their di¬
ameters and structures themselves and the micro¬
tubular sheets increase in number as seen in a cross
section of organism in the contracted state. There¬
fore, it may be said that the myoneme and micro¬
tubular sheets of Spirostomum might function just
like those of Stentor.

If the myoneme and the microtubular sheets
have a function as an antagonistic system, the force
of each fibrillar system must be transmitted to
induce the movement of antagonistic systems of
them. As the candidate for this, the rootlet-like
structures might be considered to transmit the
force of antagonistic system. On the other hand,
the anterior fiber sheets are attached to the micro¬
tubular ribbons in the elongated state but detached
from them in the contracted state, so they may
have a function as the “trigger” or “switch” for
inducing contraction and/or elongation of the cell
body.

REFERENCES

1 Yagiu, R. and Shigenaka, Y. (1963) Electron mi¬
croscopy on the longitudinal fibrillar bundle and the
contractile fibrillar system in Spirostomum ambi¬
guum. J. Protozool., 10: 364-369.

2 Yogosawa-Ohara, R. and Shigenaka, Y. (1985)
Twisting contraction mechanism of a heterotrichous
ciliate, Spirostomum ambiguum. 1. Role of the
myoneme. Cytobios, 44: 7-17.

3 Yogosawa-Ohara, R., Suzaki, T. and Shigenaka, Y.
(1985) Twisting contraction mechanism of a hetero-

978

H. Ishida, Y. Shigenaka and M. Imada

trichous ciliate, Spirostomum ambiguum. 2. Role of
longitudinal microtubular sheet. Cytobios, 44: 215—
230.

4 Hamnilton, T. C. and Osborn, D. (1976) Measure¬
ments of contraction latencies to mechanical and
electrical stimulation of the protozoan, Spirosto¬
mum ambiguum. J. Cell Physiol., 91: 403-408.

5 Hawkes, R. B. and Holberton, D. V. (1974)
Myonemal contraction of Spirostomum. I. Kinetics
of contraction and relaxation. J. Cell Physiol., 84:
225-236.

6 Yagiu, R. and Shigenaka, Y. (1961) Electron mi¬
croscopical observation of Condylostoma spatiosum
Ozaki & Yagiu, in ultra-thin section. VIII.
Myoneme (the contractile fibrils). Zool. Mag., 70:
110-114.

7 Randall, J. T. and Jackson, S. F. (1958) Fine struc¬
ture and function in Stentor polymorphus. J. Bio-
phys. Biochem. Cytol., 4: 807-830.

8 Spurr, A. R. (1969) A low-viscosity eopxy resin
embedding medium for electron microscopy. J.
Ultrastruct. Res., 26: 31-43.

9 Reynolds, E. S. (1963) The use of lead citrate at
high pH as an electron-opaque stain in electron
microscopy. J. Cell Biol., 17: 208-212.

10 Ettienne, E. E. (1970) Control of contractility in

Spirostomum by dissociated calcium ions. J. Gen.
Physiol., 56: 168-179.

11 Bannister, L. H. and Tatchell, E. C. (1968) Con¬
tractility and the fiber systems of Stentor coeruleus.
J. Cell Sci., 3: 295-308.

12 Huang, B. and Pitelka, D. R. (1973) The contractile
process in the ciliate, Stentor coeruleus. I. The role
of microtubules and filaments. J. Cell Biol., 57: 704-
728.

13 Newman, E. (1972) Contraction in Stentor
coeruleus : A cinematic analysis. Science, 177: 447-
449.

14 Hobbs, V. S., Jenkins, R. A. and Bamburg, J. R.
(1983) Evidence for the lack of actin involvement in
mitosis and in the contractile process in Spirosto¬
mum teres. J. Cell Sci., 60: 169-179.

15 Amos, W. B. (1972) Strucuture and coiling of the
stalk in the peritrich ciliates Vorticella and Carche-
sium. J. Cell Sci., 10: 95-122.

16 Amos, W. B. (1975) Contraction and calcium bind¬
ing in the vorticellid ciliates. In “Molecule and Cell
Movement”. Ed. by S. Inoue and R. E. Stephens,
Raven Press, New York, pp. 411-436.

17 Maruyama, T. (1981) Motion of the longitudinal
flagellum in Ceratium tripos (Dinoflagellida): a re¬
tractile flagellar motion. J. Protozool., 28: 328-336.

ZOOLOGICAL SCIENCE 5: 979-988 (1988)

© 1988 Zoological Society of Japan

Gonadal Maturation Independent of Photic Conditions
in Laboratory-Reared Sea Urchins, Pseudocentrotus
depress us and Hemic entrotus pulcherrimus

Masamichi Yamamoto, Mieko Ishine and Masao Yoshida

Ushimado Marine Laboratory, Okayama University,

Ushimado, Okayama 701-43, Japan

ABSTRACT — The gametogenesis in two species of the laboratory-reared sea urchins, Pseudocentrotus
depressus and Hemicentrotus pulcherrimus , were compared under different artificial photic conditions.
Juvenile sea urchins derived from zygotes of the same batch were reared under constant illumination at
20°C until they were transferred to experimental environments, where they were kept for about a year.
The experimental environments were: continuous light at constant temperature (20°C) (20 LL);
continuous light at ambient temperatures (Amb LL); continuous darkness at ambient temperatures
(Amb DD); changing photoperiod in phase with ambient daily photoperiod at 20°C (In-phase);
changing photoperiod 6 months out of phase with ambient daily photoperiod at 20°C (Out-of-phase);
and ambient light at ambient temperatures (Control). The stages in gametogenesis were determined
according to a histological standard.

In P. depressus, the gonads in the ambient temperature groups (Amb DD, Control) matured
simultaneously; the gonads in the constant temperature groups (20 LL, In-phase, Out-of-phase) also
became ripe at the same time, but earlier than those in ambient temperature groups (Amb LL was lost
by an accident). In H. pulcherrimus , the gonads in ambient temperature groups (Amb LL, Amb DD,
Control) ripened simultaneously; but in the groups kept at 20°C (20 LL, In-phase, Out-of-phase), the
gonads remained very immature. These results indicate that in the two species no specific photic
conditions are required for gametogenesis to proceed. The results in H. pulcherrimus also suggest the
presence of a critical temperature to permit the progression of gametogenesis.

INTRODUCTION

Sea urchins in shallow waters of the Temporal
Zone generally have a specific breeding season.
The annual reproductive cycles have been studied
in detail in many sea urchin species [1-11]. In
these studies, the environmental factors control¬
ling the reproductive phenomena are evaluated
from links between reproductive cycles and en¬
vironmental fluctuations, or from comparisons be¬
tween populations of the same species collected
from different localities — photoperiods, tempera¬
tures, foods and the lunar cycle have been pre¬
sumed to serve as seasonal cues to synchronize the
reproductive cycles. There are, however, a few
data on the gonadal response to experimentally

Accepted February 17, 1988
Received December 26, 1987

manipulated environmental factors. To our know¬
ledge, such studies have been done only in Stron-
gylocentrotus purpuratus collected from the west
coast of North America [12-16]. Recently Pearse
etal. [15, 16] have experimentally demonstrated in
S. purpuratus that growth and gametogenesis are
under photoperiodic control.

We have started a series of long-term experi¬
ments on environmental control of gametogenesis
in Japanese sea urchins. We used animals reared
from fertilized eggs under a constant artificial
condition in the laboratory so that they are as free
as possible from the influence of the natural en¬
vironment. In contrast to the Pacific coast of
North America, where the seasonal change in sea
temperature is moderate, it is marked along the
coast of Japan. It is thus conceivable that Japanese
sea urchins are under different environmental con¬
trol from the -S', purpuratus. In this paper, we

980

M. Yamamoto, M. Ishine and M. Yoshida

present experimental evidence showing that
photoperiod is not at least a main environmental
cue to synchronize gametogenesis in two common
Japanese sea urchins, Pseudocentrotus depressus
and Hemicentrotus pulcherrimus .

MATERIALS AND METHODS

Adult sea urchins

Adults of Pseudocentrotus depressus for obtain¬
ing gametes were collected from the subtidal zone
near the Gaikai Fishery Experimental Station of
Yamaguchi Prefecture in November 1985. The
breeding season of this species there is from Octo¬
ber to December. Adults of Hemicentrotus pul¬
cherrimus for obtaining gametes were collected in
the intertidal zone near the Ushimado Marine
Laboratory, Okayama Prefecture, in March 1986.
The breeding season of this species there is from
January to March.

Rearing of juvenile sea urchins

The experiments were carried out in the Ushi¬
mado Marine Laboratory. Embryos were
obtained by artificial fertilization of gametes
spawned from one male and female on 26 Novem¬
ber 1985 in P. depressus and on 29 March 1986 in
H. pulcherrimus. Larvae and juvenile sea urchins
were reared according to the method by Kakuda
[17-19] after slight modifications. The embryos
were kept in 30 liter tanks (600 embryos/liter) at
20°C and fed with diatom Chaetoceros gracilis that
was cultured in the laboratory. When the echinus
rudiment became prominent in 8-armed pluteus
larvae (21 day and 15 day after fertilization in P.
depressus and H. pulcherrimus , respectively),
metamorphosis was induced by giving plastic
plates on which a film of diatoms had been pre¬
pared by soaking the plates in running sea water
for several days. The plastic plates with meta¬
morphosed juvenile sea urchins were transferred
into a 500 liter tank provided continuously with
small amount of running sea water. The tank was
kept in a water bath at 20°C under continuous
illumination of the ceiling lights. When the test of
the juvenile sea urchins reached about 1 mm in
diameter, the sea urchins began to be fed with sea

lettuce, Ulva pertusa. When the test diameters
became about 7 to 12 mm in P. depressus (18
March 1986) and about 4 to 7 mm in H. pulcherri¬
mus (21 June 1986), juvenile sea urchins were
divided into 7 groups of 70-80 individuals that
were matched in size distribution and transferred
into experimental environments.

Experimental environments

Individuals in each group were placed in a round
plastic cage, 30 cm in diameter, and kept under
one of the following experimental environments:

Group 1 (20 LL); constant temperature (20°C)
and continuous light.

Group 2 (20 DD); constant temperature (20°C)
and continuous darkness.

Group 3 (In-phase); constant temperature (20°C)
and in-phase photoperiod regime (see
below).

Group 4 (Out-of-phase); constant temperature
(20°C) and out-of-phase photoperiod
regime (see below).

Group 5 (Amb LL); ambient temperature and
continuous light.

Group 6 (Amb DD); ambient temperature and
continuous darkness.

Group 7 (Control); ambient temperature and
ambient light.

For maintaining each of Groups 1-6, we used
separate indoor aquaria (120x65x45 cm), each
equipped with a system to recirculate constant
temperature sea water. In the aquaria for constant
temperature groups (Groups 1-4), the recirculat¬
ing sea water kept at 20±1°C was continuously
replaced little by little by new sea water. The
aquaria for ambient temperature groups (Groups 5
and 6) were continuously supplied with running
sea water at ambient temperatures. The illumina¬
tion for continuous light in Groups 1 and 5 was
given by some 40 W fluorescent bulbs (about 2000
lux). The aquaria for continuous darkness and
photoperiodic regimes (Groups 2, 3, 4 and 6) were
made light-tight using opaque plastic boards. The
illumination for photoperiodic regimes (Groups 3
and 4) was given by a 40 W waterproof fluorescent
bulb equipped inside each aquarium and control-

Gonadal Maturation of Sea Urchin

981

led by a time switch outside. In the in-phase
photoperiod regime, the light went on at local
sunrise and off at local sunset. In the out-of-phase
photoperiod regime, the light went on and off at
the time when the light in the in-phase aquarium
went off and on, respectively, thus the photo¬
period being set to be 6 months out of phase with
the ambient daily photoperiod. The time switches
were reset every 7 days according to the changes in
the ambient daily photoperiod. The control group
(Group 7) was kept in an outdoor tank supplied
continuously with running sea water at ambient
temperatures.

The sea urchins were provided with an unlimited
amount of food. They were fed with Ulva pertusa
from March to October and some species of Sar-
gassum (S. horneri, S. serratifolium, S. tortile etc.)
from November to February. Feces and sediments
were removed from the aquaria once a week. The
lid of the aquaria for the continuous dark groups
(Groups 2 and 6) was opened briefly (less than 15
min) every 3 or 4 days at an unfixed time for
feeding and cleaning.

Sampling

The test diameters and wet weights of all ani¬
mals were periodically measured. Some randomly
selected animals from each group were periodical¬
ly dissected for histological analysis. The gonads
were weighed and pieces were fixed in Bouin’s
solution. The paraffin-embedded gonadal tissues
were sectioned and stained in Papanicolous’
haematoxylin and eosin. The experiments were
continued until all individuals were sampled.

RESULTS

Stages in gonadal development

The gonad index (percentage of wet gonad
weight in wet animal weight) is often used as an
indication of the gonadal maturation, but we think
it is unsuitable to the present experiments; not
only exact measuring of wet gonad weights in small
sea urchins is difficult but the gonad indices do not
fully correlate with gonadal maturities because of

Fig. 1. Sections of the ovary in Hemicentrotus pulcherrimus. a, Stage 1; b. Stage 2; c, Stage 3; d. Stage 4.
Scale, 100 /an.

982

M. Yamamoto, M. Ishine and M. Yoshida

rapid growths of sea urchins themselves. In the
present study we found it most suitable to deter¬
mine the stages of gonadal development according
to a morphological standard defined in advance.

Previous authors have divided the process of
gonadal maturation in sea urchins into 4 to 9 stages
[2, 4, 10, 11]. After a survey of many histological
sections, we defined the following 5 stages in
development of the sea urchin gonad.

Stage 0 : No obvious germ cells are found; sexes of
the gonad cannot be identified in the
section.

Stage 1 : Ovary — A few small oocytes are present
along the periphery of the ovarian wall.

No large oocytes were found (Fig. la).
Testis — Small clusters of spermatogenic
cells are present along the periphery of
the testicular wall (Fig. 2a).

Stage 2: Ovary — Many large oocytes with a
prominent germinal vesicle are present in
the ovarian wall. Some oocytes migrate
toward the center of the ovarian lobe
(ovarian cavity). No mature eggs are
found (Fig. lb).

Testis — The wall of the testicular lobe is
lined with columns of spermatocytes.
Small masses of spermatozoa are present
in the center of the testicular lobe (lu¬
men) (Fig. 2b).

Fig. 2. Sections of the testis in Hemicentrotus pulcherrimus. a. Stage 1; b. Stage 2; c, Stage 3; d, Stage 4.
Scale, 100 pm.

Fig. 3. The stages of the gonads of Pseudocentrotus depressus maintained under 5 experimental environments. Each
symbol represents one individual. Sea urchins were reared from zygotes of the same batch fertilized on 26
November 1985 and were kept under constant illumination at 20°C until they were transferred to one of the
following experimental environments on 18 March 1986: continuous light at 20°C (20 LL); changing photoperiod
in phase with ambient daily photoperiod at 20°C (In-phase); changing photoperiod 6 months out of phase with
ambient daily photoperiod at 20°C (Out-of-phase); continuous darkness at ambient temperatures (Amb DD);
ambient light at ambient temperatures (Control). Sea water temperature near the Ushimado Marine Laboratory
is shown at the top.

Gonadal Maturation of Sea Urchin

983

984

M. Yamamoto, M. Ishine and M. Yoshida

Stage 3: Ovary — Numerous mature eggs are pre¬
sent in the ovarian cavity (Fig. lc).

Testis — The lumen is filled with large
numbers of spermatozoa (Fig. 2c).

Stage 4: Ovary — The ovarian cavity contains
only a few relict eggs. Oocytes are few in
the ovarian wall (Fig. Id).

Testis — The lumen is almost empty with
a few relict spermatozoa. Spermatogenic
cells are few in the testicular wall (Fig.
2d).

Gonadal response

Figure 3 shows the stage of the gonad in each
individual of P. depressus examined during the
course of the experiment. When animals were
transferred into the experimental environments in
March, all gonads were rudimentary and their
sexes cannot be identified from the histological
sections (Stage 0). The sexes became evident
(Stage 1) in animals larger than about 15 mm in
test diameter. In January 1987, the tests had
grown to 25-35 mm in diameter and the gonads to
0.6-1.2g in wet weight of the 5 lobes. We had
accidentally lost Groups 2 (20 DD) and 5 (Amb
LL) on the way. The gonads in all remaining
groups had reached maturity (Stage 3) within a
year after fertilization. The testes generally be¬
came ripe earlier than the ovaries. In the two
ambient temperature groups (Amb DD, Control),
the gonads reached Stage 3 almost simultaneously
irrespective of the photic conditions. In the three
constant temperature groups (20 LL, In-phase,
Out-of-phase), the gonads reached Stage 3 earlier
than those in the ambient temperature groups
(Amb DD, Control), but among these three
groups the timing of gonadal maturation is almost
identical irrespective of the photic conditions. The
gonads in the continuously illuminated group (20
LL) entered the post-spawned stage (Stage 4)

earlier than those in the groups under photo-
periodic regimes (In-phase, Out-of-phase).

Figure 4 shows the stage of the gonad in each
individual of H. pulcherrimus examined during the
course of the experiment. We had lost Group 2 (20
DD) on the way by an accident. In February 1987,
the tests had grown to about 20 mm in diameter
and the gonads had grown from a rudimentary
state in July to about 0. 6-1.0 g in wet weight of the
5 lobes. In the three ambient temperature groups
(Amb LL, Amb DD, Control) the ovaries and the
testes had reached the maturity (Stage 3) in some
individuals in December and in almost all indi¬
viduals in January irrespective of the photic condi¬
tions. Again in H. pulcherrimus , the gonads in the
continuously illuminated animals (Amb LL) en¬
tered post-spawned stage (Stage 4) earlier than
those in animals kept under continuous darkness
(Amb DD) or ambient daily photoperiod (Con¬
trol). In the three groups kept at 20°C (20 LL,
In-phase, Out-of-phase), the gonads never de¬
veloped over Stage 1 at least up to January; the
gonads were as large as or rather a little larger than
those in the ambient temperature groups but they
were full of nutritive cells and in many animals
even the sexes cannot be identified (Stage 0).
Some animals were kept under continuous light
(20 LL) until June 1987 but their gonads remained
at Stage 0 or 1.

Body growth

Pearse et al. [15] have reported in Strongy-
locentrotus purpuratus that the peak in body
growth rate could be sifted 6 months in animals
kept under 6 months out-of-phase photoperiodic
regime. We compared the growth rates among
three groups kept under three different light re¬
gimes at a constant water temperature of 20°C (20
LL, In-phase, Out-of-phase). In P. depressus , the
group kept under continuous light (20 LL) grew
significantly faster than groups under in-phase and

Fig. 4. The stages of the gonads of Hemicentrotus pulcherrimus maintained under 6 experimental conditions. Each
symbol represents one individual. Sea urchins were reared from zygotes of the same batch fertilized on 29 March
1986 and were kept under constant illumination at 20°C until they were transferred to one of the following
experimental environments on 21 June 1986: continuous light at 20°C (20 LL); changing photoperiod in phase
with ambient daily photoperiod at 20°C (In-phase); changing photoperiod 6 months out of phase with ambient
daily photoperiod at 20°C (Out-of-phase); continuous light at ambient temperatures (Amb LL); continuous
darkness at ambient temperatures (Amb DD); ambient light at ambient temperatures (Control).

Gonadal Maturation of Sea Urchin

985

Amb L L

4 ?

A

©

A

A

$

A

A

•

A

A

•

mm

A

A

s®

mm

■ $

A 1

1 A |

l ®®

1 1

STAGE 0 A

1 □ o

2 0®

3 ■ •

4 ®

Amb D D

A

A

A A

A A

A | | A | j

■

■«

mm

St

1 |

•

i

i

i

E# | |

mm
mm ,

|

i i i

Control a

A

•

•

A

A

<§>

m

■

A

A

<§>

m

■$

A

A

■<D

A

A

0®

A 1

1 AD1 1

| 0® |

■<s> |

20 In phase

A

A

A

A

A

A

AO

AO

AO

A A
A A
ADO A
ADO AD
| ADO | ADQ|

20 Out of phase

A

A

A

A

A

A

A

A

A

A

AO

A

A A
A A
AO ADO
AO ADO
AO 1 ADO 1

20 LL

A

A

A

A

A

A

t>D>0

ooo

A

AO

AO

AO

A

A O
ADO
ADO

A

A

A

ADO

A

A

A

A

Jul

Aug

Sep

Oct

Nov

Dec

Jan

Feb

Mar

Apr

A

A

A

A A

A A

AD AO

ADO ADO

1986

1987

986

M. Yamamoto, M. Ishine and M. Yoshida

9

Fig. 5. The wet weight of Pseudocentrotus depressus and Hemicentrotus pulcherrimus maintained under three
different light regimes at 20°C: continuous light (20 LL), changing photoperiod in phase with ambient ones
(In-phase) and changing photoperiod 6 months out-of-phase with ambient ones (Out-of-phase). Each point
represents the mean of all individuals maintained, which were about 75 at the beginning of the experiment but
decreased in number by being sampled for histological analysis during the experiments.

6 months out-of-phase photoperiodic regimes but
the latter two groups grew at the same rate
throughout the experimental period (Fig. 5). In
H. pulcherrimus , the growth rates did not differ
among the three groups under the different light
regimes (Fig. 5).

DISCUSSION

It is difficult to evaluate the contribution of
environmental factors to gonadal maturation un¬
less the conditions of animals prior to the experi¬
ment were fully evident. Boolootian [12] has
reported in S. purpuratus that a population main¬
tained in a constant laboratory condition remained
reproductive synchrony with the field ones over
three years. Pearse et al. [20] suggested in the sea
star, Pisaster ochraceus , the presence of an en¬
dogenous circannual rhythm that is set very early

in life and insensitive to experimentally fixed day-
lengths. In the present experiments we used sea
urchins derived from fertilized eggs of the same
batch and reared under a constant environment
(continuous light at 20°C) until they were transfer¬
red to experimental environments.

The experiments by Pearse et al. [15] have
showed that in Strongylocentrotus purpuratus,
both gametogenesis and body growth are regulated
by seasonal changes in photoperiod. The normal
winter to spring spawning period of this species
could be shifted to summer to fall by exposing sea
urchins to changing photoperiod set 6 months out
of phase with ambient daily photoperiod. Game¬
togenesis in S. purpuratus is also responsive to
fixed photoperiod [16]. After one year mainte¬
nance under fixed short days (8L: 16D) or fixed
neutral days (12L: 12D) the gonads became ripe,
but under fixed long days (16L: 8D) the gonads did

Gonadal Maturation of Sea Urchin

987

not have gametes.

The present experiments show that in two com¬
mon Japanese sea urchins, Pseudocentrotus de-
pressus and Hemicentrotus pulcherrimus , game-
togenesis is under different environmental control
from that in S. purpuratus ; no specific photic
conditions are required for gametogenesis to pro¬
ceed. In H. pulcherrimus , the timings of gonadal
maturation were identical among the three groups
kept under different photic conditions (continuous
darkness, continuous light and ambient photo¬
period) as long as the animals were kept at
ambient temperatures. In P. depressus kept at
ambient temperatures, the groups under con¬
tinuous darkness and ambient photoperiod had
mature gonads simultaneously. Moreover, in P.
depressus kept at 20°C, the gonads reached the
maturity at the same time in three groups under
different light regimes (continuous light, in-phase
and out-of-phase photoperiod). These results indi¬
cate that gametogenesis proceeds independently of
environmental photic conditions. In the two spe¬
cies, unlike in S. purpuratus [15], growth was also
unaffected by difference in photoperiodic regimes,
although the continuous light seems to promote
the growth in P. depressus.

Our experiments suggest that temperature is a
factor controlling gametogenesis in P. depressus
and H. pulcherrimus. In all groups of H. pulcherri¬
mus maintained at 20°C, the gonads remained
undifferentiated or immature after the gonads in
the ambient temperature groups became ripe. In
contrast, in all groups of P. depressus kept at 20°C,
the gonads reached maturity within a year after
fertilization. These results suggest the existence of
a species-specific temperature range permitting the
progression of gametogenesis. The effect of
temperatures on sea urchin gametogenesis is now
being analyzed in the succeeding experiments.

Although photic conditions do not seem to
regulate the onset of gametogenesis in the present
species, they seem to relate to the end of game¬
togenesis. In the gonads of both H. pulcherrimus
and P. depressus , entrance to the post-spawned
stage was promoted by a continuous illumination.
S. purpuratus is also a contrast to the present two
species in this respect. In the former, cold temper¬
ature treatments did not affect the onset of game¬

togenesis but high temperature treatments induced
gamete resorption [13].

The difference in regulatory mechanisms of
gametogenesis between the present species and S.
purpuratus may be ascribed to the difference in the
condition of water temperatures in the sea where
they live. Orton [21] have argued that sea temper¬
ature is the main environmental factor controlling
reproductive activity of many shallow-water
marine animals. Seasonal changes in sea tempera¬
ture are marked along the coast of Japan: monthly
means of sea water temperature range 9 to 26°C
near the Ushimado Marine Laboratory. In con¬
trast, there are only weak and poorly defined
seasonal changes in sea temperature on the Pacific
coast of California: Monthly means of sea water
temperature range 12 to 17°C [20] at Santa Cruz,
California where the experiments in S. purpuratus
by Pearse et al. [15] were carried out. In such a
condition, the photoperiod can be a main factor to
control the reproduction of marine invertebrates.
Pearse’s group has reported the presence of photo-
periodic control of the annual reproductive cycles
in two species of sea stars from California [20, 22,
23].

Since the present materials were reared under
the constant illumination at 20°C from the begin¬
ning of their lives, the experimental groups main¬
tained under continuous light at 20°C (20 LL)
appear to be free from any environmental cues to
start the gametogenic cycle. In such a group of P.
depressus , the gonads ripened synchronously at a
different time from those in the groups maintained
at ambient temperatures. There may be an en¬
dogenous temporal program to start the game¬
togenic cycle.

ACKNOWLEDGMENTS

We thank Mr. Nobutaka Kakuda of the Gaikai Fishery
Experimental Station, Yamaguchi Prefecture for kindly
supplying sea urchins. Thanks are also due to Dr. Kozo
Ohtsu and Messrs. Waichiro Godo and Masao Isozaki of
this marine laboratory for their help in rearing and
maintenance of sea urchins. This work was supported by
a Grant-in-Aid for Special Project Research from the
Ministry of Education, Science and Culture (No.
62124037).

988

M. Yamamoto, M. Ishine and M. Yoshida

REFERENCES

1 Boolootian, R. A. (1966) Reproductive physiology.
In “Physiology of Echinodermata”. Ed. by R. A.
Boolootian, Interscience Publishers, New York,
London, Sydney, pp. 465-486.

2 Yoshida, M. (1952) Some observations on the
maturation of the sea urchin, Diadema setosum.
Annot. Zool. Japon., 25: 265-271.

3 Holland, N. D. (1967) Gametogenesis during the
annual reproductive cycle in a cidaroid sea urchin
(Stylocidaris affinis). Biol. Bull., 133: 578-590.

4 Pearse, J. S. (1969) Reproductive periodicities of
Indo-Pacific invertebrates in the Gulf of Suez. II.
The echinoid Echinometra mathaei (de Blainville).
Bull. Mar. Sci., 19: 580-613.

5 Gonor, J. J. (1973) Reproductive cycles in Oregon
populations of the echinoid, Strongylocentrotus pur-
puratus (Stimpson) I. Annual gonad growth and
ovarian gametogenic cycles. J. Exp. Mar. Biol.
Ecol., 12: 45-64.

6 Mori, T., Tsuchiya, T. and Amemiya, S. (1980)
Annual gonadal variation in sea urchins of the order
Echinothurioida and Echinoida. Biol. Bull., 159:
728-736.

7 Iliffe, T. M. and Pearse, J. S. (1982) Annual and
lunar reproductive rhythms of the sea urchin, Diade¬
ma antillarum (Philippe) in Bermuda. Int. J. Inver-
tebr. Reprod., 5: 139-148.

8 Falk-Petersen, I. B. and Lonning, S. (1983) Repro¬
ductive cycles of two closely related sea urchin
species, Strongylocentrotus droebachiensis and S.
pallidus. Sarsia, 68: 157-164.

9 Lessios, H. A. (1985) Annual reproductive
periodicity in eight echinoid species on the Carib¬
bean coast of Panama. In “Echinodermata”. Ed. by
B. F. Keegan and B. D. S. O’Connor, A. A. Balke-
ma, Boston, pp. 303-311.

10 Nichols, D., Bishop, G. M. and Sime, A. A. T.
(1985) Reproductive and nutritional periodicities in
populations of the European sea urchin. Echinus
esculentus (Echinodermata: Echinoidea) from the
English channel. J. Mar. Biol. Assoc. U. K., 65:
203-220.

11 Hori, R., Phang, V. P. E. and Lam, T. J. (1987)
Preliminary study on the pattern of gonadal de¬
velopment of the sea urchin, Diadema setosum, off
the coast of Singapore. Zool. Sci., 4: 665-673.

12 Boolootian, R. A. (1964) Die Bedeutung der abio-
tischer Faktoren fur die Gonadenentwicklung und
Fortpflanzung der mariner Evertebraten. Helgolan-
der wiss. Meeresunters., 10: 118-139.

13 Cochran, R. C. and Engelman, F. (1975) Environ¬
mental regulation of the annual reproductive season
of Strongylocentrotus purpuratus (Stimpson). Biol.
Bull., 148: 393-401.

14 Leachy, P. S., Hough-Evans, B. R., Britten, R. J.
and Davidson, E. (1981) Synchrony of oogenesis in
laboratory-maintained and wild populations of the
purple sea urchin ( Strongylocentrotus purpuratus ).
J. Exp. Zool., 215: 7-22.

15 Pearse, J. S., Pearse, V. B. and Davis, K. K. (1986)
Photoperiodic regulation of gametogenesis and
growth in the sea urchin Strongylocentrotus purpu¬
ratus. J. Exp. Zool., 237: 107-118.

16 Bay-Schmith, E. and Pearse, J. S. (1987) Effect of
fixed daylengths on the photoperiodic regulation of
gametogenesis in the sea urchin Strongylocentrotus
purpuratus. Int. J. Invertebr. Reprod. Dev., 11:
287-294.

17 Kakuda, N. and Nakamura, T. (1975) Studies on
the artificial seedling of the sea urchin. II. On the
food for larvae of Pseudocentrotus depresssus.
Aquiculture, 22: 56-60.

18 Kakuda, N. (1978) Studies on the artificial seedling
of the sea urchin. III. On mass culture of pluteus
larvae. Aquiculture, 25: 121-127.

19 Kakuda, N. (1978) Studies on the artificial seedling
of sea urchin. IV. On culture of juvenile sea urchins.
Aquiculture, 25: 128-133.

20 Pearse, J. S., Eernisse, D. J., Pearse, V. B. and
Beauchamp, K. A_ (1986) Photoperiodic regulation
of gametogenesis in sea stars, with evidence for an
annual calendar independent of fixed daylength.
Am. Zool., 26: 417-431.

21 Orton, J. S. (1920) Sea temperature, breeding and
distribution in marine animals. J. Mar. Biol. Assoc.
U. K., 12: 339-366.

22 Pearse, J. S. and Beauchamp, K. A. (1986) Photo¬
periodic regulation in a brooding sea star from
central California. Int. J. Invertebr. Reprod. Dev.,
9: 289-297.

23 Pearse, J. S. and Eernisse, D. J. (1982) Photo¬
periodic regulation of gametogenesis and gonadal
growth in the sea star Pisaster ochraceus. Mar. Biol.,
67: 121-125.

ZOOLOGICAL SCIENCE 5: 989-998 (1988)

© 1988 Zoological Society of Japai

Normal Embryonic Stages of the Pygmy Cuttlefish,
Idiosepius pygmaeus paradoxus Ortmann

Masamichi Yamamoto

Ushimado Marine Laboratory, Okayama University,
Ushimado, Okayama, 701-43 Japan

ABSTRACT — A normal table of embryonic stages for the pygmy cuttlefish, Idiosepius pygmaeus
paradoxus Ortmann, a suitable material for developmental study of the cephalopod, is presented. A
female cuttlefish kept in still sea water in a small vessel repeatedly lays 30-80 fertilized eggs every 2-7
days for more than a month. Thirty stages are morphologically defined during the embryonic period
from oviposition to hatching. Hatching occurs 16-18 days after oviposition at 20°C. The external shape
and some internal structures of the living embryo at each stage are illustrated, and the chronological
ages and main developmental events are assembled in a tabular form.

INTRODUCTION

The cephalopod is unique in its developmental
pattern: unlike other molluscs, its eggs are
telolecithal, show no spiral cleavage and develop
directly to miniature adult forms. However the
developmental processes are not fully analyzed
due to the difficulty of obtaining eggs, especially at
early stages of development. At present, Loligo
pealii [1,2] and Loligo vulgaris [3] are mainly used
in experimental studies of cephalopod develop¬
ment because their eggs are obtainable compara¬
tively easily. The pygmy cuttlefish, Idiosepius
pygmaeus paradoxus, can be a more useful ex¬
perimental organism for the developmental study
of the cephalopod. They are widely distributed
along the coast of Japan [4]. They are small in size
(about 15 mm in mantle length) and are easily
collectable by a hand net in the eelgrass bed. The
maintenance of adults and obtaining of eggs are
very easy [5, 6]; no special equipment is necessary
to keep them in the laboratory. They lay eggs
repeatedly in small vessels; in this laboratory just
laid eggs are available every day for at least half an
year. The egg is very transparent and its size is
suitable for handling (about 1 mm in longer di¬
ameter).

Accepted December 25, 1987
Received November 10, 1987

The cephalopod generally has a long embryonic
period; a miniature adult form is gradually com¬
pleted in the egg capsule. Natsukari [6] has
described an outline of the development of Idio¬
sepius pygmaeus paradoxus and reported the
embryonic period to be 15-17 days at about 20°C.
The speed of development is so susceptible to
environmental conditions, such as temperature
and egg density, that it is almost impossible to
designate developmental stages correctly only by
chronological means. Thus, the morphological
staging of Idiosepius.pygmaeus paradoxus seems to
be useful for future developmental studies of this
species.

MATERIALS AND METHODS

Adult pygmy cuttlefish, Idiosepius pygmaeus
paradoxus Ortmann were collected by a hand net
or a tow net of a coarse mesh size in eelgrass beds.
Near the Ushimado Marine Laboratory, the cut¬
tlefish can be easily obtained from April to August
but they disappear from autumn to winter with
regression of the eelgrass beds. The male can be
distinguished from the female by the white testis
visible through the mantle at the caudal end.

Female cuttlefish were kept in still sea water in
plastic vessels (3-4 individuals per vessel of 25 cm
in diameter) in a constant temperature room (20 +
1°C). The water was changed every 2 or 3 days.

990

M. Yamamoto

Aeration was not necessary. The cuttlefish were
usually fed with gammaridian amphipods (mainly
Ampithoe valida) collected by shaking sea weeds
(2-3 amphipods per individual per day.) Many
other small crustaceans such as marine and terres¬
trial isopods, fully grown Artemia and marine and
fresh water shrimps were usable as foods as long as
they are alive and the size is appropriate.

Several pieces of microscopic slides, one surface
of which was covered with black adhesive tape,
were leaned against the inner wall of the vessel
with covered face upward. The cuttlefish usually
attach to the under surface of the slide with the
adhering gland present on the dorsal side of the
mantle. Females begin to lay eggs in several days.
The eggs are fertilized when they are laid; females
have usually received spermatophores from males
in the sea. Two maturation divisions begin succes¬
sively after oviposition. The eggs, each wrapped
with a thin vitelline membrane, several layers of
gelatinous coats and a transparent egg-capsule
(Fig. 2 lb), are laid into an one-layered compact
mass on the under surface of the slide. One female
kept laying 30-80 eggs every 2-7 days for more
than one month in the laboratory; the duration
between two ovipositions varied but the rate of egg
production was not only constant in a female but
almost equal among females (mean egg production

rate (eggs/day)±s.d. = 13.7 ±0.4, n=6).

The slide with eggs on one face was transferred
to a petri dish and kept at 20±0.5°C. Develop¬
ment tended to be retarded in the eggs laid in the
central area of an egg mass. In order to keep
synchronous development among embryos, some
eggs were removed from the slide so that no eggs
were closely surrounded by others. Developing
embryos were periodically observed under a
stereoscopic microscope (Nikon, SMZ-10) at the
magnification of 40-60. The external shapes and
some internal structures observable from the out¬
side were sketched with the aid of drawing appa¬
ratus. Preliminary staging was made by comparing
more than 10 series of sketches of separate batch¬
es. The final staging was constructed by correcting
the preliminary one through checking it against the
development of further 8 batches. The chronolo¬
gical ages are determined by averaging the time
data obtained from 10 developmental series of
separate batches.

Explanatory illustrations of organ primordia are
presented in Figure 1 for the sake of convenience.

RESULTS AND DISCUSSION

The developmental period of Idiosepius pyg-
maeus paradoxus from oviposition to hatching was

Fig. 1. Organ primordia in the embryo of Idiosepius pygmaeus paradoxus. The illustrations are only explanatory
and do not represent any particular stages. Left, dorsal view; right, ventral view, a, arm; an; anus; c,
chromatophore; e, eye vesicle; f, fin; ff, funnel fold; g, gill; H . organ of Hoyle; 1, lens; m, mantle; mo, mouth, o,
olfactory organ; ol, optic lobe; r, retina; s, sucker; sc, secondary cornea; sg, shell gland; sp, siphon; st, statocyst;
y, yolk; ys, yolk sac.

Normal Embryonic Stages of Cuttlefish

991

divided into 30 stages on the basis of morphologi¬
cal features of living embryos. The external shapes
and some internal structures of the embryos are
illustrated in Figures 2 to 6. Main developmental

events to characterize each stage and the mean
time to attain the stage after oviposition are pre¬
sented in Table 1.

I think that these 30 stages are discrete but some

Fig. 2. Normal embryonic stages of Idiosepius pygmaeus paradoxus : Stages 1 to 12. The egg envelopes are shown
only in one drawing in the upper left, the lateral view of the egg at stage lb: c, egg capsule; g, gelatinous layer; p,
perivitelline space; v, vitelline membrane. The other drawings are views from the animal pole. Each number
indicates the stage number.

992

M. Yamamoto

1 mm

Fig. 3. Normal embryonic stages of Idiosepius pygmaeus paradoxus : Stages 8-18. The lateral or dorsal views.

Normal Embryonic Stages of Cuttlefish

993

experience and familiality with materials may be
required to determine stages accurately. It must
be taken into consideration that the stages are
artificially constructed by dividing a continuous
change into steps. Naturally, there will often be
discrepancies between actual embryos and illustra¬
tions. Stages of such embryos should be described
such as, for instance, “early stage 20” or “late stage
21.”

Some developmental events described in Table
1 may be less reliable than others as developmental
criteria. It is relatively difficult to distinguish organ
primordia in living embryos from stages 17 to 19,
though they are observable more discretely in fixed
or properly stained embryos. The earliest stage
when the heart beat is recognizable in some
embryos is stage 23 but the heart beat is observable
in all embryos at stage 27. It is also difficult to

19

1 mm

Fig. 4. Normal embryonic stages of Idiosepius pygmaeus paradoxus : Stages 19-21. The left, middle and right
drawings in each stage are the dorsal, ventral and animal pole views, respectively.

994

M. Yamamoto

determine the precise stage for the secondary necessary to use several criteria to identify stages

cornea to cover the eye. The retina shows differ- correctly.

ent colorations under different illuminations. The time data presented in Table 1 is useful only

Hatching occurs precociously at stage 28 if as an approximation. The speed of development

embryos are mechanically agitated. Thus, it is varies among individuals or batches. Cleavage

Normal Embryonic Stages of Cuttlefish

995

Fig. 6. Normal embryonic stages of Idiosepius pygmaeus paradoxus : Stages 27-30. The left and right drawings in
stages 27 and 28 are the dorsal and ventral views, respectively. Only the dorsal views are shown in stages 29 and
30.

996

M. Yamamoto

Table 1. Normal embryonic stages of the cuttlefish Idiosepius pygmaeus paradoxus

Stage

No.

Time after
oviposition
(20±0.5°C)

Main developmental events

1

Precleavage stage. First polar body appears 45-60 min after oviposi¬
tion (la). Blastodisc becomes evident about 2 hr 30 min and second
polar body appears about 3 hr 30 min after oviposition (lb).

2

5.5 hr

2 cell stage. First cleavage plane occurs along the plane of bilateral
symmetry of the embryo.

3

7.0 hr

4 cell stage. Second cleavage plane occurs obliquely to the first one
and the embryonic anterior (future dorsal) and posterior (future
ventral) become evident. The polar bodies are present in the anterior.

4

8.5 hr

8 cell stage. Division is unequal.

5

10.0 hr

16 cell stage. Division is unequal.

6

12.0 hr

32 cell stage. Division is unequal and asynchronous.

7

14.0 hr

About 64 cell stage.

8

16.0 hr

Blastoderm is spreading by division of marginal cells.

9

20.0 hr

Blastoderm is indented in outline with increasing marginal cells.

10

24.0 hr

Blastoderm is decreasing in diameter by centripetal migration of
marginal cells.

11

1 day 05 hr

Blastoderm is smooth in outline and smallest in diameter. A papilla of
yolk is visible in the center of the blastoderm.

12

1 day 14 hr

Yolk papilla becomes flattened. Blastoderm has slightly increased in
diameter.

13

2 day 00 hr

Epiboly begins. About 1/6 of the egg surface is covered by the
blastoderm.

14

2 day 13 hr

About 1/4 of the eggs surface is cellulated.

15

3 day 01 hr

About 1/3 of the egg surface is cellulated.

16

3 day 14 hr

About 1/2 of the egg surface is cellulated.

17

4 day 03 hr

About 2/3 or more of the egg surface is cellulated. Major organ
primordia appear as thickenings in the blastoderm.

18

4 day 16 hr

Most of the egg surface is cellulated. Primordia of the mantle, shell
gland, mouth and arms are discernible but very faintly. Egg changes
in shape from ovoidal (18a) to discoidal (18b) and begins to rotate
very slowly.

19

5 day 05 hr

Cellulation is complete. Primordia of the gills, funnel folds and
statocysts are visible faintly. Invagination of the eyes and shell gland
has begun. Embryos rotate rapidly.

20

5 day 18 hr

Primordia of the mantle, funnel folds, gills and arms become discrete
by elevation from the embryonic surface. Anus primordium is
evident. Sucker primordia appear on some arms. Yolk sac begins
weak pulsation.

21

6 day 07 hr

Embryonic body begins to stand up from the egg surface. Eye vesicles
have just closed. Optic lobe primordia are visible. Sucker primordia
have increased in number. Mantle begins to grow downward.
Ventral funnel folds are bent toward the midline.

(continue)

Normal Embryonic Stages of Cuttlefish

997

Table 1.

(continued)

Stage

No.

Time after
oviposition
(20 + 0.5°C)

Main developmental events

22

7 day 00 hr

Fins are discernible with care. Funnel folds have just formed the
siphon at the ventral proximities. Yolk sac is separated from the
embryonic body by a constriction. Eye vesicle is spherical in shape.
Lens is visible as a small dot. Retina is dish-shaped and very pale
yellow.

23

7 day 18 hr

Mantle completely covers the gills. Fins are evident. Median margins
of the ventral funnel folds are fusing. Retina is cup-shaped and
reddish orange in color. Lens is evident as a refractile rod. Iris fold is
being formed. Rotational movement gradually ends but pulsation of
the yolk sac becomes strong. Heart beat begins in some embryos.

24

8 day 16 hr

Organ of Hoyle is discernible with care. Retina is reddish brown. Iris
is prominent as a colored circle. Mantle completely covers the margin
of the funnel.

25

9 day 16 hr

Chromatophores, very pale yellow in color, appear on the dorsal face
of the mantle, at first, and of the head, later. Organ of Hoyle is
discrete. Secondary cornea has just begun to cover the eye. Retina is
dark brown. Olfactory organs are evident. Pulsation of the yolk sac
weakens. Mantle sometimes contracts.

26

10 day 16 hr

Chromatophores, orange in color, appear on the whole embryonic
surface. Secondary cornea is covering the eye. The mantle, head and
yolk sac are approximately equal in size but the head is somewhat
larger in width than the mantle. Embryos begin to move the fins,
siphon and arms and often change position within the egg capsule.

27

12 day 06 hr

Secondary cornea has just covered the eye. Retina is reddish black.
The mantle is slightly larger in width than the head.

28

13 day 22 hr

The mantle is prominently larger in size than the head. Yolk sac is
decreasing in size. Chromatophores are usually contracted. Embryos
move vigorously in the capsule.

29

15 day 22 hr

Yolk sac is very small or almost invisible but a mass of yolk remains in
the embryonic body. Some embryos hatch.

30

18 day 05 hr

Yolk mass is completely lost. The remaining embryos hatch.

proceeds fairly synchronously within a batch but
after epiboly, time variation between individuals
gradually increases with stages; the fastest reaches
stage 19 about 24 hr earlier than the latest. De¬
velopmental speed is also dependent on tempera¬
ture and density of eggs. The duration from
oviposition to hatching was about 12 days at 23°C
and more than 1 month at 17°C. The embryos
located in the central area of a broad egg mass
always develop more slowly than those in its
periphery and sometimes show abnormalities in
shape, especially at late embryonic stages; the
development may be susceptible to oxygen con¬
centration.

The egg envelopes are omitted from the illustra¬

tions except one drawing in Figure 2. The peri-
vitelline space between the egg and the vitelline
membrane is narrow up to stage 17 but gradually
increases in volume after stage 18. The first slight
increase in volume at stage 18 may give the egg a
room to change the shape and to rotate. Gomi et
al. [7] have suggested in Sepiella that pro¬
teinaceous osmoactive substance produced by the
embryo causes water intake into the perivitelline
space.

There are many reports on the developmental
process of the cephalopod [2, 8] but only a few are
usable as normal tables of embryonic develop¬
ment. Naef [9] has presented the developmental
series for many cephalopods in his extensive

998

M. Yamamoto

monograph but cleavage is neglected in his stag¬
ings. Arnold [1] has divided the embryonic period
of Loligo pealii into 30 stages, which do not always
correspond to the present 30 stages. Yamamoto
[10] has defined 40 stages in the embryonic de¬
velopment of Sepiella japonica. The embryo of
Sepiella japonica is seemingly different from that
of the present material; in the former the orga-
nogenetic period is longer and more stages are
definable there than in the latter.

ACKNOWLEDGMENT

I would like to thank Messrs. W. Godo and M. Isozaki
of this Marine Laboratory for their help in collecting the
cuttlefish. This work is supported by a Grant-in- Aid
from the Japanese Ministry of Education, Science and
Culture (No. 60540463).

REFERENCES

1 Arnold, J. M. (1965) Normal embryonic stages of
the squid, Loligo pealii (Lesueur). Biol. Bull., 128:
24-32.

2 Arnold,!. M. and Williams- Arnold, L. D. (1977)
Cephalopoda: decapods. In “Reproduction of
Marine Invertebrates”. Ed. by A. C. Giese and J. S.

Pearse, Academic Press, New York, pp. 243-290.

3 Marthy, H. J. (1982) The cephalopod egg, a suit¬
able material for cell and tissue interaction studies.
In “Embryonic Development, Part B: Cellular
Aspect”. Ed. by M. M. Burger and R. Weber, Alan
R. Liss, New York, pp. 223-233.

4 Sasaki, M. (1929) A monograph of the dibranchiate
cephalopods of the Japanese and adjacent waters. J.
Coll. Agr. Hokkaido Univ., 20: Suppl., 132-134.

5 Sasaki, M. (1923) On an adhering habit of a pygmy
cuttlefish, Idiosepius pygmaeus Steenstrup. Annot.
Zool. Japon., 10: 209-213.

6 Natsukari, Y. (1970) Egg laying behavior,
embryonic development and hatched larva of the
pygmy cuttlefish, Idiosepius pygmaeus paradoxus
Ortmann. Bull. Fac. Fish. Nagasaki Univ., 30: 15-
29.

7 Gomi, F., Yamamoto, M. and Nakazawa, T. (1986)
Swelling of egg during development of the cuttlefish,
Sepiella japonica. Zool. Sci., 3: 641-645.

8 Hamabe, M. (1983) Cephalopoda. In “Musekitsui
Dobutsu no Hassei, Jou.” Ed. by K. Dan, K. Seki-
guchi, Y. Ando and H. Watanabe, Baifukan,
Tokyo, pp, 343-360.

9 Naef, A. (1928) Die Cephalopoden. Monographic
35, Fauna e Flora del Golfo di Napoli.

10 Yamamoto, M. (1982) Normal stages in the de¬
velopment of the cuttlefish, Sepiella japonica Sasaki.
Zool. Mag., 91: 146-157.

j

ZOOLOGICAL SCIENCE 5: 999-1006 (1988)

© 1988 Zoological Society of Japan

Absence of Androgen Receptors in the Prostatic
Glandular Epithelium Derived from Testicular
Feminization Mutant ( Tfm ) Mice

Takeo Mizuno1,2, Hiroyuki Takeda, Naoya Suematsu,
Noriko Hironaka and Ilse Lasnitzki3

Zoological Institute, Faculty of Science, University of Tokyo,
Tokyo 113, Japan and * 1 2 Strangeway s Research Laboratory,
Wort’s Causeway, Cambridge CB1 4RN, England

ABSTRACT — Urogenital sinus epithelium from male embryos of Tfm androgen receptor defective
mice was combined with urogenital sinus mesenchyme from normal rat embryos and the recombinates
were grown underneath the kidney capsule of male nude mice. The Tfm epithelium formed prostatic
buds which developed into fully formed glands. A few recombinates also formed seminal vesicles and
coagulating glands. A study of androgen receptors by steroid autoradiography showed that labelling
was absent in the epithelium but present in the mesenchyme surrounding the epithelium. In the control
recombinates in which epithelium from normal wild type mice was combined with the rat mesenchyme
both epithelium and mesenchyme were androgen-labelled. A study of the fine structure of the Tfm
epithelium and its enzyme activity showed that both approximate those seen in the control epithelium.
The results suggest that epithelial androgen may not be required for the cytodifferentiation of the
epithelium and that both bud formation and cytodifferentiation are induced by the androgen activated
mesenchyme. The possibility that androgens are necessary for epithelial cell proliferation and the
synthesis of specific prostatic proteins, is discussed.

INTRODUCTION

Prostatic glands develop from the urogenital
sinus as epithelial buds projecting from the sinus
epithelium into the surrounding mesenchyme
under the influence of androgens [1]. Epithelial
mesenchymal recombination experiments between
androgen deficient ( Tfm mutant) epithelium and
wild type mesenchyme [2] suggested that the
mesenchyme is a target for androgen and that
epithelial buds are induced by androgen primed
mesenchyme. This concept has been supported by
steroid autoradiographic analysis which showed
androgen-binding sites in the mesenchyme
throughout all fetal stages [3], In contrast

Accepted January 29, 1988
Received January 7, 1988

1 Present address: Faculty of Pharmaceutical Sciences,
Teikyo University, Sagamiko, Kanagawa, 199-01
Japan.

2 To whom reprints should be requested.

androgen binding sites were absent in the fetal
epithelium but appeared after birth coinciding with
the cytodifferentiation of the postnatal epithelium
[4, 5],

This paper attempts to relate the presence of
androgen to the development and cytodifferentia¬
tion of the mouse prostate gland, and also to
determine whether the urogenital sinus mesen¬
chyme will induce androgen receptors in the
androgen receptor deficient Tfm epithelium. For
this purpose urogenital sinus epithelium from the
Tfm mouse fetuses was combined with urogenital
sinus mesenchyme from normal rat fetuses and the
recombinates grown under the kidney capsule of
normal host mice, for up to 8 weeks. The fine
structure of its epithelium and the enzyme activity
of epithelium and mesenchyme such as alkaline
phosphatase and nonspecific esterase activity were
determined. The uptake of labelled androgen by
the mesenchyme and epithelium was analysed by
steroid autoradiography and the results correlated

1000

T. Mizuno, H. Takeda et al.

with the degree of cytodifferentiation and enzyme
activity.

MATERIALS AND METHODS

Animals

Mice carrying the Tfm- mutation gene and wild-
type mice were kindly provided by Dr. Mary F.
Lyon (MRC, Harwell, Oxon) and bred in inbreed¬
ing in our laboratory. Mutant male (XTfm/Y)
embryos were obtained by mating XTfm/X+
females with wild males (X+/Y). Wistar-Imamichi
rats were purchased from Imamichi Institute for
Animal Reproduction, Oomiya, Japan. The ani¬
mals were mated during the night and copulation
was confirmed by the presence of spermatozoa in
the vaginal smears on the next morning. The
conceptus was considered to be 0.5-day old at
12:00 of this day. Male fetuses of 15.5- and
16.5-days were obtained from mother mice and
rats, respectively. Post-natal Wistar-Imamichi rats
were also used for control.

Identification of Tfm-mutant fetuses

Fragments of pectoral skin of all male mouse
fetuses used for the study were examined histologi¬
cally. Mammary rudiments regressed and no hair
follicles appeared in 15.5-day wild-type (X+/Y)
fetuses, while both rudiments were developed in
mutant (Xr/m/Y) ones.

Chemicals

Cold testosterone (Koch-Light Laboratories,
Colnbrook) and [1, 2, 6, 7-3H] testosterone (83.4
Ci/mmol; Amersham International PLC, Bucks)
dissolved in propylene glycol and this stock solu¬
tion was further diluted with medium 199 (Earle’s
salts, GIBCO Laboratories, Grand Island, NY,
U.S.A.). Collagenase (Worthington Biochemical
Corporation, Freehold, NJ, U.S.A.; Code CLS
149 U/mg) was dissolved in Tyrode’s solution.

Separation of epithelium and mesenchyme

Urogenital sinuses of fetal male mice and rats
were treated with 0.06% collagenase solution
(Worthington Biochemical Corp., Code CLS) for
40 min at 37°C, and the epithelium and mesen¬

chyme were separated carefully by fine two pairs of
forceps. The tissues were then thoroughly washed
in Tyrode’s solution supplemented with 50% fetal
bovine serum.

Recombination of epithelium and mesenchyme and
cultivation of recombinates

Epithelia isolated from Tfm or wild-type fetuses
were recombined with mesenchymes isolated from
normal rat fetuses. Recombinates were cultured in
vitro for one day by a modified Trowell technique
[6, 7] in Earle’s Medium 199 supplemented with
10% fetal bovine serum and 0.01 pglm\ cold testos¬
terone. The recombinates were then grafted be¬
neath the kidney capsule of male athymic ICR
nude mice (Charles River Japan Inc., Atsugi,
Japan), and cultured for 4, 6 and 8 weeks.

Histology and ultrastructural study

The grafts were removed from the kidney cap¬
sule in Tyrode’s sloution supplemented with 10%
fetal bovine serum and were divided into frag¬
ments approximately 1 mm thick. Some fragments
were fixed in Bouin’s fluid and the sections were
stained with haematoxylin-eosin. Some fragments
were fixed in a modified Karnovsky’s fixative [8] at
4°C for 3 hr and post-fixed in 1% osmium tetroxide
in 0.1 M cacodylate buffer (pH 7.5) at 4°C for 2 hr,
and carried through a graded series of alcohols,
and embedded in epoxy resin according to the
method of Luft [9]. Ultra-thin sections were cut on
a Porter-Blum MT2 microtome, and stained with
uranyl acetate and lead citrate. They were ex¬
amined with a JEOL 100 CX electron microscope.

Enzyme histochemistry

Some other fragments of the grafts were fixed in
ice-cold 4% paraformaldehyde in 0.1 M phosphate
buffer (pH 7. 2-7. 4) for 2 hr, washed overnight in
several changes of 5% sucrose at 4°C and frozen in
isopentan ( — 190°C) chilled with liquid nitrogen.
Sections, 5 pm thick, were cut in a cryostat (Cryo-
cut II; American Optical, Buffalo, NY, U.S.A.) at
about — 25°C. Alkaline phosphatase activities
were assessed by tetrazolium reaction (modified
after McGadey, [10]) with 5-bromo-4-chloro-3-
indoxyl-phosphate as the substrate and tetranitro-
blue-tetrazolium as the coupling dye at pH 9.2-

Prostate Formation without Androgen Receptors

1001

9.4. Nonspecific esterase activities were assessed
at pH 5.0 by an azo-coupling method [11] with
1-naphtyl acetate as the substrate and hexazotized
para-rosaniline as the coupling dye.

Labelling procedure and autoradiography

Steroid autoradiography for locating androgen¬
binding sites was carried out according to the
method of Stumpf and Sar [12]. The details of the
labelling procedure and autoradiographic tech¬
niques were described in the previous paper [4],
The hosts were castrated 4-days before labelling to
reduce the level of endogenous androgens. Tissue
recombinates were divided into small fragments
approximately 1-2 mm thick. They were incu¬
bated in Medium 199 supplemented with 10% fetal
bovine serum and 2.5 pC\lm\ [1, 2, 6, 7-
3H]testosterone. The incubation was carried out at
37°C in a gas phase of 50% 02 and 5% C02.

After 15 hr incubation the tissues were washed
with Tyrode’s solution for 3 hr and were rapidly
frozen in liquid nitrogen. Five-//m frozen sections,
cut in a cryostat, were thaw-mounted onto the
emulsion-coated slides and exposed for 2-3 weeks
at 4°C.

To assess the specificity of the binding of
[3H]testosterone, we added a 400-fold excess of
unlabelled testosterone to the medium containing
[3H]testosterone.

Identification of tissues

Sections cut in a cryostat were stained with
Hoechst dye #33258 (Calbiochem-Behring, La
Jolla, CA, U.S.A.). By this staining, we can
distinguish mouse nuclei from rat ones. Mouse
nuclei contain several small discrete intranuclear
fluorescent bodies, absent in rat nuclei.

RESULTS

Morphology of prostate glands induced in recom¬
binates

Histological examination of the recombinates
showed that most of them developed prostate
glands. In recombinates of wild type epithelium
and rat mesenchyme the acini had a wide lumen
lined with columnar secretory epithelium which

exuded much secretory matter. The lamina pro¬
pria surrounding the epithelium was poor in cells
and fibres (Fig. 1). Recombinates with Tfm
epithelium and rat mesenchyme also formed
glands but their tubules were narrower and the
epithelium was lower with a reduction of the
supranuclear cytoplasmic area. In contrast to
glands induced from normal epithelium the con¬
nective tissue surrounding the epithelium was in¬
creased significantly (Fig. 2). With the increase of
the culture period to 6-8 weeks the diameter of the
tubules increased and there was some secretory
matter. Some recombinates also formed seminal
vesicles, and urethral and coagulating glands
(Table 1). Hoechst staining confirmed that the
epithelium in the recombinates consisted of mouse
cells only (Fig. 3).

Ultrastructure of induced glands in the recombi¬
nates

In the recombinates of Tfm epithelium with rat
mesenchyme after 4 weeks growth microvilli de¬
veloped at the apical surface of the epithelial cells,
which also possessed a well developed rough sur¬
faced endoplasmic reticulum and Golgi apparatus.
Many secretory granules were seen in the apical
cytoplasm (Fig. 4). These features were similar to
those of the ventral prostate of the wild-type
mouse at 4 weeks of age (Fig. 5).

Enzyme histochemistry

Alkaline phosphatase and nonspecific esterase
activities were examined histochemically (Table
2). Recombinates of the sinus epithelium of Tfm
mice and the sinus mesenchyme of normal rats
showed similar activities to those of the control
explants; the alkaline phosphatase activity was
localized only in the lamina propria surrounding
the epithelium which showed no activity (Figs. 6
and 7). In contrast, the nonspecific esterase activ¬
ity appeared mainly in the cytoplasm of the epithe¬
lium (Figs. 8 and 9). These features are the same
as those seen in normal prostate glands of the same
age or in glands in the control explants.

Androgen receptors in the explants

We examined, then, whether androgen recep¬
tors were induced in the prostatic epithelium de-

1002

T. Mizuno, H. Takeda et al.

Prostate Formation without Androgen Receptors

1003

Table 1. Prostate gland formation in the recombinates of the urogenital sinus epithelium of Tfm or

wild-type mice

with the urogenital sinus

mesenchyme of normal rats

Recombinates

(epithelium/

mesenchyme)

Culture

period

(weeks)

Prostate

glands

Seminal

vesicles

Coagulating

glands

Urethral

glands

4

15/16

4/16

3/16

9/16

T/m-mouse/

6

5/5

1/5

0/5

4/5

Rat

8

4/4

1/4

0/4

4/4

Wild-mouse/

Rat

4

11/11

6/11

1/11

4/11

Table 2. Alkaline phosphatase and nonspecific esterase activities in the prostatic glands induced in
the explants

Recombinates

(epithelium/

mesenchyme)

Culture

period

(weeks)

Number

of

explants

Tissues

Alkaline

phosphatase

activity

Nonspecific

esterase

activity

4

9

E

—

+ - + + +

T/ra-mouse/

M

++-+++

-

Rat

8

4

E

NE

-

M

NE

-

Wild-mouse/

4

5

E

—

+ - + + +

Rat

M

+ + - + +

-

NE: not examined, — : negative, +: weak, + + : strong, + + + : very strong.
E: epithelium, M: lamina propria.

rived from the sinus epithelium of Tfm mice. The
results are summarized in Table 3. In control
explants, the epithelial nuclei of the glands were
labelled strongly and the nuclei of the lamina
propria moderately (Fig. 10). In contrast, the
nuclei of most of the glandular epithelium derived
from the Tfm mice were negative but the intensity
of labelling in the lamina propria was similar to
that in the control explants (Fig. 11). In competi¬
tion experiments, unlabelled testosterone abol¬
ished nuclear labelling completely. The results
show that in the recombinates with Tfm epithelium
the androgen receptors were localized in the lami¬
na propria of the prostate, that few or no receptors

were induced in the Tfm epithelium, and that
practically no androgen was transferred into the
epithelium from the lamina propria. Nevertheless
prostatic gland formation took place in the recom¬
binates of the androgen-receptor defective epithe¬
lium with the normal mesenchyme.

DISCUSSION

The results presented here show that the sinus
epithelium of Tfm mice forms functional prostatic
glands under the influence of the sinus mesen¬
chyme of wild type rats in the presence of
androgens, that the nuclei of the lamina propria

Figs. 1 and 2. Histology of prostatic glands induced in the recombinates of the sinus epithelium of the wild-type (Fig.
1) or of the Tfm-type (Fig. 2) and the sinus mesenchyme of the wild-type. The culture period was 4 weeks.
Magnification X760. Haematoxylin-eosin staining.

Fig. 3. Hoechst staining of the prostatic glands induced in the recombinates of the mouse epithelium of Tfm- type and
the rat mesenchyme. All the epithelial cells were mouse cells.

Figs. 4 and 5. Epithelial cells of the prostatic glands induced in the recombinates of the Tfm-e pithelium and the rat
mesenchyme (Fig. 4) and those of the ventral prostate of a mouse of 4 weeks old (Fig. 5). Magnification X5,000.

1004

T. Mizuno, H. Takeda et al.

>0 * •*.'

'***' **4 * 1 , ^

. .>• . -

** . r

j>r;- '

\:u *> -

%7v '

* %

^’Wil

v *•>

*-t?r *

*

. .. . . .

■' ••••’• f 3v- /

\* N#/ ‘

. k r. •*.*

• \ »*

* #•

Prostate Formation without Androgen Receptors

1005

Table 3. Incorporation of [3H]testosterone into the nuclei of the epithelium and mesenchyme of the
induced prostatic glands

Recombinates

(epithelium/

mesenchyme)

Culture

period

(weeks)

Number

of

explants

Tissues

Nuclear

androgen

incorporation

T/ra-mouse/

4

5

E

-

Rat

M

+

Wild-mouse/

4

5

E

+ +

Rat

M

+

— : negative, +: positive,

+ + : heavily, E:

epithelium, M:

lamina propria.

surrounding the epithelial buds incorporate
androgens but that the epithelium remains recep¬
tor negative (see also [13]). This is in contrast to
previous findings on androgen receptors in the
developing rat prostate gland [5]. These showed
receptor positive mesenchyme surrounded recep¬
tor negative epithelium during fetal and early
postnatal stages while epithelial receptors
appeared and mesenchyme ones declined from day
10 postnatal coinciding with the beginning of cyto-
differentiation. The results suggested that
androgen-activated mesenchyme induced bud
formation and that epithelial androgen stimulates
cytodifferentiation. The present results do not
bear out this interpretation. The fine structure and
enzymatic activity of the receptor negative Tfm
epithelium approximate that seen in the sinus
epithelium from normal mouse embryos and it is
possible that epithelial androgen is not required
for its cytodifferentiation. However, the result
does not necessarily deny that the Tfm epithelium
does not express some androgen-dependent pro¬
teins which the normal prostatic epithelium ex¬
presses [14].

The exact function of epithelial androgen is still
uncertain. It may be a mitogen. In rat and mouse
prostate glands in organ culture androgens in¬
creased epithelial proliferation in a dose depend¬

ent manner [15], but in cell cultures of isolated rat
epithelium they did not influence the rate of prolif¬
eration [16]. On the other hand, the hormone may
be involved in the production of functional sub¬
stances specific for the fully developed prostate
gland such as prostatic binding protein [17],
prostatic secretion protein [18], prostatic basic
protein, probasin [19] or prostatic acid phospha¬
tase [20].

In addition to the formation of prostate glands
the Tfm epithelium associated with the rat mesen¬
chyme developed a small number of coagulating
glands and seminal vesicles. The coagulating
glands originate from the urogenital sinus and the
seminal vesicles from the proximal part of the
Wolffian ducts and it seems that androgen recep¬
tors in the sinus epithelium or Wolffian ducts are
also dispensable for the development of these
accessory male sex organs. Finally, the results
suggest that the mesenchyme does not induce a
synthesis of androgen receptors in the Tfm epithe¬
lium.

ACKNOWLEDGMENTS

This work was supported by Grants-in-Aid from the
Ministry of Education, Science and Culture of Japan.
We are also greatly indebted to Dr. Mary F. Lyon of the
MRC Radiobiology Unit, for the gift of the mutant Tfm
mice used in the present investigation.

Figs. 6 and 7. Histochemistry of alkaline phosphatase activity of the prostatic glands induced in a recombinate of the
epithelium of wild-type (Fig. 6) or Tfm (Fig. 7) and the rat mesenchyme. Magnification X760.

Figs. 8 and 9. Histochemistry of nonspecific esterase activity of the prostatic glands induced in a recombinate of the
epithelium of wild-type (Fig. 8) and Tfm (Fig. 9) rat mesenchyme. Magnification X760.

Figs. 10 and 11. Autoradiographs of explants incubated with [3H]testosterone. Sections were stained with
haematoxylin. The exposure period was 3 weeks. The glandular epithelium was induced from the sinus
epithelium of a wild-type mouse (Fig. 10) and that of a Tfm mouse (Fig. 11) in the presence of the rat
mesenchyme that incorporated androgen heavily. Magnification X 1,000.

1006

T. Mizuno, H. Takeda et al.

REFERENCES

1 Lasnitzki, I. and Mizuno, T. (1977) Induction of the
rat prostate gland by androgens in organ culture. J.
Endocrinol., 74: 47-55.

2 Lasnitzki, I. and Mizuno, T. (1980) Prostatic induc¬
tion: Interaction of epithelium and mesenchyme
from normal wild-type mice and androgen-
insensitive mice with testicular feminization. J. En¬
docrinol., 85: 423-428.

3 Shannon, J. M. and Cunha, G. R. (1983) Auto¬
radiographic localization of androgen binding in the
developing mouse prostate. Prostate, 4: 367-373.

4 Takeda, H. and Mizuno, T. (1984) Incorporation
des androgenes au moment de la cytodifferenciation
de l’epithelium prostatique chez le Rat. C. R. Soc.
Biol., 178: 572-575.

5 Takeda, H., Mizuno, T. and Lasnitzki, I. (1985)
Autoradiographic studies of androgen-binding sites
in the rat urogenital sinus and postnatal prostate. J.
Endocrinol., 104: 87-92.

6 Lasnitzki, I. (1976) The action of androgens on rat
prostate glands in organ culture. In “Organ Culture
in Biomedical Research”. Ed. by M. Balls and M.
A. Monnickendam, Cambridge Univ. Press, Cam¬
bridge, pp. 241-256.

7 Trowell, O. A. (1959) The culture of mature organs
in a synthetic medium. Exp. Cell Res., 16: 118-148.

8 Karnovsky, M. J. (1965) A formaldehyde-
glutaraldehyde fixative of high osmolality for use in
electron microscopy. J. Cell Biol., 27: 137A.

9 Luft, J. H. (1961) Improvements in epoxy resin
embedding methods. J. Biophys. Biochem. Cytol.,
9; 409-414.

10 McGadey, J. (1979) Tetrazolium reaction. In “En¬
zyme Histochemistry”. Ed. by Z. Lojda, R. Gossrau
and T. H. Schiebler, Springer-Verlag, Berlin, pp.
61-63.

11 Davis, B. J. and Ornstein, L. (1959) Simultaneous
azo-coupling with 1 -naphthyl acetate. In “Enzyme
Histochemistry (1979)”. Ed. by Z. Lojda, R. Goss¬
rau and T. H. Schiebler, Springer-Verlag, Berlin,
pp. 109-110.

12 Stumpf, W. E. and Sar, M. (1975) Autoradio¬
graphic techniques for localizing steroid hormones.
Methods in Enzymol. 36: 135-156.

13 Mizuno, T., Takeda, H. and Suematsu, N. (1986)
Recepteurs d’androgenes de l’epithelium au cours
de l’induction des glandes prostatiques a partir de
Pepithelium du sinus urogenital de la Souris Tfm-
mutant. C. R. Soc. Biol., 180: 593-595.

14 Suematsu, N., Takeda, H. and Mizuno, T. (1988)
Glandular epithelium induced from urinary bladder
epithelium of the adult rat does not show full
prostatic cytodifferentiation. Zool. Sci., 5: 385-395.

15 Lasnitzki, I. (1965) Action and interaction of hor¬
mones and 3-methylcholanthrene on the ventral
prostate gland of the rat in vivo. J. Natl. Cancer
Inst., 35: 339-348.

16 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.

17 Heyns, W. and De Moor, P. (1977) Prostatic bind¬
ing protein. A steroid-binding protein secreted by
rat prostate. Eur. J. Biochem., 78: 221-230.

18 Pousette,A., Bjork, P., Carlstrom, K., Forgren,
B., Hogber, B. and Gustafsson, J.-A. (1981) In¬
fluence of sex hormones on prostatic secretion pro¬
tein, a major protein in rat prostate. Cancer Res.,
41: 688-690.

19 Matsuo, Y., Nishi, N., Muguruma, Y., Yoshitake,
Y., Kurata, N. and Wada, F. (1985) Localization of
prostatic basic protein (“Probasin”) in the rat pros¬
tates by use of monoclonal antibody. Biochem.
Biophys. Res. Commun., 130: 293-300.

20 Perrin, P., Clairet, F. and Fleury-Goyon, M. C.
(1984) Place des phosphatases acides prostatiques
dans le traitement des adenocarcinomes de la pros¬
tate. J. Urol., 1: 19-22.

ZOOLOGICAL SCIENCE 5: 1007-1012 (1988)

© 1988 Zoological Society of Japan

Plasma Levels of Androgens in Growing
Frogs of Rana nigromaculata

Sumiko Tanaka, Hisaaki Iwasawa1 and Katsumi Wakabayashi2

Department of Pathology, Kagawa Medical School, Kagawa 761-07,
1 Faculty of Science, Niigata University, Niigata 950-21, and
~lnstitute of Endocrinology, Gunma University,

Maebashi 371, Japan

ABSTRACT — Using young, subadult and adult Rana nigromaculata of both sexes, plasma testosterone
and dihydrotestosterone (DHT) levels were determined by radioimmunoassay. In young male frogs,
the testosterone concentration was almost constant at ca. 100 pg/ml from just after metamorphosis to
July of the next year, while the DHT concentration showed a gradual upward tendency and reached a
level twice that of newly metamorphosed frogs by May of the next year. In first-year female frogs, the
concentrations of testosterone and DHT were ca. 100 pg/ml, and had doubled by June of the next year.
In second-year male frogs, testosterone and DHT concentrations increased remarkably in autumn,
especially in DHT, and reached 8.7 ng/ml. In second-year females, testosterone also increased from
July on, and reached 1.1 ng/ml in the breeding season (the first half of May) of the following year,
though the DHT level remained low. Androgen concentrations in adult males were remarkably low just
after hibernation compared with those in the preceding autumn (especially in DHT), decreased further
until June, and increased again from July until October as the same manner in second-year males.
Testosterone concentrations in adult females showed similar changes, though always slightly lower than
those of the male frogs of the same age.

INTRODUCTION

The development of radioimmunoassay (RIA)
has enabled one to determine quantitatively very
small amounts of hormones in blood. In adult
frogs of some anuran species, seasonal changes in
the levels of circulating sex hormones have been
reported [1-3]. On the other hand, plasma
androgen levels in young and subadult frogs have
been poorly documented [4]. It is, therefore,
difficult to discuss in detail the correlation between
plasma androgen concentration and the develop¬
ment of genital tracts, or spermatogenic progress.
The present paper describes the changes in the
plasma levels of testosterone and DHT in young,
subadult and adult frogs of Rana nigromaculata,
and discusses the relationship between androgen
levels and the development of sexual organs in this
species [4-8].

Accepted December 26, 1987
Received October 23, 1987

MATERIALS AND METHODS

Animals and serum samples Young, sub¬
adult, and adult frogs of Rana nigromaculata were
collected monthly from May to October, provided
that subadult and adult female frogs were to
September, at Kanazuka, Niigata Prefecture. The
animals collected each time were divided into 6
groups according to sex and age, i.e., first-year
(young), second-year (subadult), and third-year or
more aged adult males and females. The body
length and weight of these frogs were determined,
and animals within the range of standard deviation
were used as materials. Blood samples were
collected in microtubules from the heart of frogs
anesthetized weakly by immersion in a 1% solu¬
tion of MS-222, the blood was centrifuged at
15,000 g for 5 min at 4°C. The separated sera were
stored at — 80°C. Each time serum samples for
RIA were made from 6 points. The number of
frogs used each point was 6-10 young frogs, 2-4
subadults, and one adult.

1008

S. Tanaka, H. Iwasawa and K. Wakabayashi

Separation of testosterone and DHT After
the addition of 1,000 cpm of 3H-testosterone and
3H-DHT, 200 p\ of the serum was extracted twice
with 4 ml of diethylether. After evaporation of the
solvent, the extract was redissolved in 200 p\ of an
isooctane-benzene-methanol mixture (90:5:5).
Testosterone and DHT were then chromato¬
graphed on Sephadex LH-20 (Pharmacia Fine
Chemicals). One half of the testosterone or DHT
fraction was used for assay and the other half was
retained to determine recovery with a liquid scin¬
tillation counter.

RIA RIA was carried out by a double
antibody method according to the method of
Ismail et al. [9] with minor modification. For the
iodination of testosterone-3-carboxymethyloxime-
tyrosinemethylester, the chloramine-T method
[10] was used. The assay system consisted of: 1)
standard or sample dissolved in 200 p\ of 1% BSA
borate buffer (pH 8.0), 2) 100/4 of diluted anti¬
serum (1:8000, Eiken Immunochemical Labora¬
tory, Tokyo) in 1% normal rabbit serum with 0.05
M EDTA borate buffer (pH 8.0), and 3) 100 p\ of
125I-labeled hormone in 1% BSA borate buffer
(pH 8.0). The reaction was carried out at 4°C for 2
days. A goat anti-rabbit gamma-globulin serum
was used as the second antibody. The amount of
labeled testosterone bound in the precipitate was
determined with an automatic gamma counter
(Beckman, System 8000).

Statistical analyses The data were treated
according to the method of one way analysis of
variances, and comparison of means were carried
out by Duncan’s new multiple range test, Student’s
t-test or Cochran-Cox test depending on the uni¬
formity of variances.

RESULTS

Plasma testosterone concentrations in young
male frogs were nearly unchanged from just after
metamorphosis to the spring of the next year,
showing a slight rise in the middle of June (P<0.05
vs. August in first year). DHT levels showed a
tendency to rise from middle July to October in the
first year (P<0.25, one way analysis of variances),

and showed a big variation in May of the next year.
The level was high in early June (P<0.01 vs. July
and August in first year) and then fell down until
late June. This seasonal change was highly signif¬
icant (P< 0.025, analysis of variances).

In young females, on the other hand, plasma
testosterone levels were constant during the first
year, and rose in early June in the second year (P
<0.05 vs. first year September and October).
Plasma DHT levels in females showed a gradual
increase from May to middle June in the second
year, then fell down sharply. These changes were
significant (P<0.05) when examined by the analy¬
sis of variances. The mean level in the middle June
was significantly higher than those in the first year
August and October (P<0.05) and than that in
middle July in the second year (P<0.01) (Fig. 1).

In second-year male frogs, both testosterone
and DHT showed a similar fluctuation pattern.
That is, they clearly increased in August (P<0.1
and P<0.01 vs. July, respectively), the concentra¬
tion of testosterone was ca. 0.5 ng/ml and DHT
was ca. lng/ml, decreased temporarily in Septem¬
ber (P<0.1 and P<0.05 vs. August, respectively),
and then increased markedly in October to 2.5
ng/ml of testosterone and 8.7 ng/ml of DHT. This
rise in October was especially remarkable in DHT;
the level of DHT was three times higher than that
of testosterone.

In adult male frogs, high testosterone levels
were found in May, but noticeable individual
differences were seen. Then testosterone and
DHT concentrations decreased until the middle of
June, and increased again from July (P<0.1 and P
<0.01 middle June vs. September, respectively),
about a monthe earlier than in subadult male
frogs. The concentrations of testosterone and
DHT decreased (P<0.1 and P<0.05 vs. August,
respectively) briefly in September and increased
strikingly in October. As in the case of second-
year male frogs, the concentrations of DHT in the
summer and autumn months were always greater
than those of testosterone.

In second-year females, the DHT concentration
remained at a certain low level from July on. On
the other hand, the concentration of testosterone
gradually increased from July to September (P<
0.001 July vs. September, P<0.01 July vs August),

Plasma Androgens in Growing Frogs

1009

1st year frogs 2nd year frogs

Fig. 1. Concentrations of plasma testosterone and DHT in the frogs from metamorphosis to just before the
appearance of secondary sexual characters. Vertical lines indicate S.E. The relation between the two is *: P<
0.05 and **: P<0.01.

and reached l.lng/ml in adult female frogs just
after hibernation. The level of testosterone was at
its minimum after the breeding season (P<0.05
May vs. middle June), and increased again after
July on (P<0.01 middle June vs. September). In
subadult and adult female frogs, unlike the male
frogs of the same age, no decrease in androgen was
seen in September (Fig. 2).

DISCUSSION

Gonadotropins and plasma androgen levels in male
frogs

As in the reproductive endocrine system in other
vertebrates, a close relationship exists between
gonadotropin secretion and androgen production
in anurans [11-14]. Licht et al. [3] reported that in

male Rana catesbeiana the concentrations of circu¬
lating LH and androgen fluctuated almost in paral¬
lel. From the ultrastrucutural quantitative data on
gonadotropin-producing cells in subadult male
frogs of R. nigromaculata [7], gonadotropin con¬
trol of androgen production is clearly suggested.
That is to say, the changes in the number, size, and
ultrastructure of gonadotropin-producing cells
coincide well with the fluctuation in plasma levels
of androgens, especially in the DHT observed in
the present study.

Plasma androgen level and spermatogenesis

In newly metamorphosed frogs of R. nigromacu¬
lata , the seminiferous tubules are almost filled with
primary spermatogonia. On the other hand, in
young summer and autumn frogs, fairly active
spermatogenesis is seen in some parts of the semi-

1010

S. Tanaka, H. Iwasawa and K. Wakabayashi

2nd year frogs adult frogs

Fig. 2. Concentrations of plasma testosterone and DHT in subadult (2nd-year) and adult frogs. Vertical lines
indicate S.E. The relation between the two is *: P<0.1, **: P<0.05 and ***: P<0.01.

niferous tubules, and most of these spermatogenic
cells disappear during hibernation [6, 8, 15]. Judg¬
ing from the results of the present study, it seems
that this uneffective spermatogenesis is caused by
the deficiency of androgens. Effective sperma¬
togenesis begins in late spring in second-year frogs.
In August, in which a remarkable increase in the
number of spermatids and spermatozoa is seen,
androgen concentrations increased remarkably,
and a further increase was seen in October. The
thumb pads begin to develop in August, and
develop highly in October in this species [4, 5],
Therefore, the progress of spermatogenesis and
the development of thumb pads well agree with the
increase in the concentration of plasma androgens.
The reason for the decrease in androgen seen in
September is unknown.

A wide variation in the concentration of plasma

testosterone was seen in adult male frogs of the
breeding season. This phenomenon may be due to
the use of frogs having different breeding activi¬
ties, and it seems that the testosterone concentra¬
tion rapidly decreases toward the end of the breed¬
ing season. It is interesting that the available
androgen in the breeding period is testosterone
rather than DHT.

The changes in the plasma androgen concentra¬
tion observed in the present study coincided with
the results of Kera and Iwasawa [4], except for the
values in second-year May frogs, and the reason
for this disagreement is not known.

Plasma androgen level and development of genital
tracts

Sexual difference in genital tracts in R. nigromac-
ulata appeared rapidly and remarkably in second-

Plasma Androgens in Growing Frogs

1011

year frogs [8]. This mode of appearance coincides
well with the increasing change in the plasma
concentrations of testosterone and DHT. It is
known that testosterone is capable of conversion
to DHT or estrogen, but DHT is not capable of
further change and is an androgen with the
strongest physiological activities [16, 17]. In
maturing male frogs, the levels of plasma DHT
were always higher than those of testosterone,
whereas in subadult and adult female frogs, only
testosterone fluctuated, and DHT remained at a
constant low level except for a peak which
appeared in the second-year June frogs. It is
known that in adult males of some Rana species
DHT is a major component of serum androgens [3,
18-20]. Iwasawa and Kobayashi [21] reported that
the administration of testosterone and estradiol to
first-year frogs induced considerable development
of both Wolffian and Mullerian ducts. These
results suggest that a good deal of androgen de¬
tected in second-year late spring male frogs and
early summer female frogs has a stimulative effect
on the development of genital tracts.

Role of plasma androgens in female frogs

In the present study, the plasma testosterone
levels in female frogs were not so low as compared
with those in male ones, except for second- and
third-year male frogs in autumn when the concen¬
tration of plasma androgen showed a marked
increase. In R. esculenta [1] and R. catesbeiana [3]
also, the concentrations of testosterone in the
serum of adult female frogs are almost the same,
or even greater than those in adult male ones, and
the levels of testosterone are higher than those of
estradiol.

Delrio et al. [22] showed that testosterone and
DHT were produced in the ovaries of R. esculenta.
Fortune and Tsang [23] reported that the ovaries
of Xenopus laevis produced a large amount of
testosterone after treatment with frog pituitaries.
Habbard and Licht [24] found, in R. pipiens and R.
catesbeiana , that gonadotropin stimulated the
secretion of testosterone and progesterone in the
ovaries, and oocyte maturation occurred simul¬
taneously. Licht et al. [3] reported that the plasma
androgen levels in adult females were highly corre¬
lated to the developmental condition of the Mulle¬

rian ducts. A similar phenomenon was observed in
subadult female frogs of R. nigromaculata [8].
Considering these facts together with the results of
Iwasawa and Kobayashi [21], it is conceivable that
the androgens found in the serum of female R.
nigromaculata are produced by the ovaries, and
that the serum testosterone in the female frogs acts
not only as the precursor estrogens [22], but acts
stimulative ly on the sex organs as an androgen.

We think that the number of adult frogs used in the
present study is not enough to clarify the pattern of
seasonal changes in androgen levels compared with other
anuran species, so we have not referred to the problem in
this paper.

REFERENCES

1 D’Istria, M., Delrio, G., Botte, V. and Chieffi, G.
(1974) Radioimmunoassay of testosterone, 17/?-
oestradiol and oestrone in the male and female
plasma of Rana esculenta during sexual cycle. Ster¬
oids Lipids Res., 5: 42-48.

2 Siboulet, R. (1981) Variations saisonnieres de la
teneur plasmatique en testosterone et dihydrotesto¬
sterone chez le crapaud de Mauritanie ( Bufo mauri-
tanicus ). Gen. Comp. Endocrinol., 43: 71-75.

3 Licht, P., McCreery, B. R., Barnes, R. and Pang,
R. (1983) Seasonal and stress related changes in
plasma gonadotropins, sex steroids, and corticoster¬
one in the bullfrog, Rana catesbeiana. Gen. Comp.
Endocrinol., 50: 124-145.

4 Kera, Y. and Iwasawa, H. (1981) Functional histol¬
ogy of the testies in the process of sexual maturation
in the frog, Rana nigromaculata. Zool. Mag.
(Tokyo), 84: 109-114. (in Japanese with English
summary)

5 Iwasawa, H. and Asai, O. (1959) Histological
observation on the seasonal change of the testis and
the thumb pad in the frog, Rana nigromaculata. J.
Fac. Sci. Niigata Univ., Ser. II, 2: 213-219.

6 Iwasawa, H. and Kobayashi, M. (1976) Develop¬
ment of the testis in the frog Rana nigromaculata,
with special reference to germ cell maturation.
Copeia, 1976: 461-467.

7 Iwasawa, H. and Kera, Y. (1982) Structural change
in gonadotropin-producing cells of male frogs, Rana
nigromaculata, in the process of sexual maturation.
Gen. Comp. Endocrinol., 46: 236-245.

8 Iwasawa, H., Nakazawa, T. and Kobayashi, T.
(1987) Histological observations on the reproduc¬
tive organs of growing Rana nigromaculata frogs.
Sci. Rep. Niigata Univ. Ser. D, 24: 1-13.

9 Ismail, A. A. A., Niswender, G. D. and Midgley,

1012

S. Tanaka, H. Iwasawa and K. Wakabayashi

A. R. (1972) Radioimmunoassay of testosterone
without chromatography. J. Clin. Endocrinol.
Metab., 34: 177-184.

10 Greenwood, F. C. and Hunter, W. M. (1963) The
preparation of 131I-labelled human growth hormone
of high specific radioactivity. Biochem. J., 89: 114—
123.

11 Lofts, B. (1974) Reproduction. In “Physiology of
the Amphibia”. Ed. by B. Lofts, Academic Press,
New York, vol. 2, pp. 107-218.

12 Lofts, B. (1984) Amphibians. In “Marshall’s Phys¬
iology of Reproduction”. Vol. 1, Ed. by G. E. Lam¬
ming, Churchill Livingstone, Edinburgh, pp. 127-
205.

13 Licht, P. (1979) Reproductive endocrinology of
reptiles and amphibians: gonadotropins. Ann. Rev.
Physiol., 41: 337-351.

14 Muller, C. H. and Licht, P. (1980) Gonadotropin
specificity of androgen secretion by amphibian
testes. Gen. Comp. Endocrinol.. 42: 365-377.

15 Kobayashi, T. and Iwasawa, H. (1986) Effects of
testosterone on spermatogenic process and sperm
preservation in late autumn young Rana nigromacu-
lata. Zool. Sci., 3: 387-390.

16 Gower, D. B. and Fotherby, K. (1975) Biosynthesis
of the androgens and oestrogens. In “Biochemistry
of Steroid Hormones”. Ed. by H. L. J. Makin,
Blackwell Sci. Publ., Oxford, pp. 77-104.

17 Gower, D. B. (1984) Biosynthesis of the androgens
and C19 steroids. In “Biochemistry of Steroid Hor¬
mones”. Ed. by H. L. J. Makin. Blackwell Sci.
Publ., Oxford, pp. 170-206.

18 Wada, M., Wingfield, J. C. and Gorbman, A.
(1976) Correlation between blood level of
androgens and sexual behavior in male leopard
frogs, Rana pipiens. Gen. Comp. Endocrinol., 29:
72-77.

19 Muller, C. H. (1976) Steroidogenesis and sperma¬
togenesis in the male bullfrog, Rana catesbeiana :
Regulation by purified bullfrog gonadotropins. Ph.
D. Dissertation, Univ. California, Berkeley.

20 Muller, C. H. (1977) Plasma 5a-dihydrotestos-
terone and testosterone in the bullfrog, Rana cates¬
beiana: Stimulation by bullfrog LH. Gen. Comp.
Endocrinol., 33: 122-132.

21 Iwasawa, H. and Kobayashi, M. (1974) Effect of
testosterone and estradiol on the development of
sexual characters in young Rana nigromaculata.
Biol. Reprod., 11: 398-405.

22 Delrio, G., D’lstria, M., Iela, L. and Chieffi, G.
(1979) The possible significance of testosterone in
the female green frog Rana esculenta during sexual
cycle. Boll. Zool., 46: 1-9.

23 Fortune, J. E. and Tsang, P. C. (1981) Production
of androgen and estradiol-17/? by Xenopus ovaries
treated with gonadotropins in vitro. Gen. Comp.
Endocrinol., 43: 234-242.

24 Hubbard, G. M. and Licht. P. (1986) In vitro ova¬
rian responses to pulsatile and continuous gonado¬
trophin administration on steroid secretion and
oocyte maturation in the frogs, Rana pipiens and
Rana catesbeiana. Gen. Comp. Endocrinol., 61:
417-423.

ZOOLOGICAL SCIENCE 5: 1013-1018 (1988)

© 1988 Zoological Society of Japai

Vascular Supply of Hypophysis in the Turtle,

Geoclemys reevesii

Yoshihiko Oota and Ichiro Koshimizu

Biological Institute, Faculty of Science, Shizuoka University,
Shizuoka 422, Japan

ABSTRACT — The vascular route in the hypothalamo-hypophysial complex was studied histologically in
the turtle, Geoclemys reevesii. Distinct anterior and posterior groups of portal vessels originating from
the anterior and posterior capillary plexus of the median eminence have been demonstrated. The
anterior group of portal vessels breaks up into capillaries in the cephalic lobe of the pars distalis,
whereas the posterior group of portal vessels is continuous with a capillary network in the caudal lobe of
the pars distalis. It is suggested that the presence of two distinct groups of vascular routes is correlated
with the cytological differentiation of the pars distalis.

INTRODUCTION

The vascularization of the hypothalamus and
hypophysis has been described in detail for various
vertebrate species. The reptilian hypophysial por¬
tal system has been investigated previously by
Enemar in the lizard and the snake [1], by Green
in the turtle and several other species [2], by
Wingstrand in the snake [3]. Such studies have
demonstrated the significance of the hypophysial
portal system by which hypophysiotropic neuro¬
hormones may be transported from the median
eminence (ME) into the pars distalis (PD). In the
previous studies on the vascularization of the
hypophysis of the turtle, Pseudemys scripta , the
presence of two distinct anterior and posterior
groups of capillary plexus and portal vessel system
arising from the anterior and posterior primary
capillary plexus of the anterior and posterior divi¬
sions of the ME has been estimated briefly [4].

Reported here are the precise results of an
investigation designed to ascertain whether the
division of portal vessels into anterior and pos¬
terior groups is present in the turtle, Geoclemys
reevesii. Since there is clear differential distribu¬
tion of the PD cells, the correlation between the
vascular supply and the cytological differentiation

Accepted February 8, 1988
Received November 26, 1987

of the PD cells is also examined.

MATERIALS AND METHODS

Adult male and female turtles were obtained
from a dealer. The hypothalamic regions of 10
animals were fixed in Bouin’s solution and pre¬
pared for routine histology. Thin paraffin sections
of 6 pm were stained with Heidenhain’s Azan
technique or Gomori’s paraldehyde fuchsin (AF).
For the study of the vascularization, 8 animals
were injected with filtered India ink-saline solution
(1:1) through the ventricle of the pulsating heart
or into the left carotid artery. After decapitation,
the brains were fixed in Bouin’s solution. Subse¬
quently they were trimmed to small pieces. Thick
paraffin sections of 100 pm were prepared without
any staining procedure.

RESULTS

In the turtle, the ME forms a well-developed
swelling on the ventral side of the infundibulum
and can be divisible into anterior and posterior
regions (Fig. 1). Both regions of the ME are
supplied with distinct components of fibers from
different areas of the hypothalamus, preoptic
hypothalamic, and tuberal regions. The anterior
ME receives a rich supply of fibers containing
AF-stainable neurosecretory material from the

1014

Y. OOTA AND I. KoSHIMIZU

Fig. 1. Mid-sagittal section through the infundibulum and hypophysis. AME, anterior ME; Ca, caudal lobe of PD;
Ce, cephalic lobe of PD; PI, pars intermedia; PME, posterior ME; PN, pars nervosa; PT, pars tuberalis. Dotted
line represents the boundary between Ce and Ca. AF stain. x60.

anterior hypothalamus, whereas the posterior ME
receives only a few fibers of neurosecretory nature.

The ME is covered by a very dense primary
capillary plexus deriving from branches of the
infundibular arteries. Some of the capillaries
penetrate deeply into the palisade layer of the ME.
As shown in Figure 2, there are two distinct ante¬
rior and posterior capillary plexus, corresponding
to the anterior and posterior regions of the ME,
respectively.

At the mid-ventral surface of the ME, these two
groups of capillary plexus converge into two
groups, anterior and posterior, of portal vessels.
The vascular connections between the capillary
plexus of the ME and the portal vessels can be
demonstrated in transverse sections of the India
ink-injected specimens (Fig. 3).

Several large portal vessels pass vertically down¬
ward from the primary capillary plexus of the
anterior ME, and break up into a secondary sinu¬
soidal-capillary net within the anterior portion of
the PD, cephalic lobe. Similarly, the capillaries of
the posterior ME converge to form large portal
vessels leading to the secondary sinusoidal¬
capillary net within the posterior region of the PD,
caudal lobe (Figs. 2 and 3).

The elongated PD consists of well-defined

cephalic and caudal lobes, which are distinct in
their cellular constituents (Fig. 1). At least four
types of secretory cells are characterized by the cell
size and tinctorial affinities for specific dyes. The
regional patterns of the cellular distribution are
demonstrable in the two lobes. The cephalic lobe
is occupied by abundant carminophilic acidophils
and amphophils, whereas the caudal lobe contains
orangeophilic acidophils and basophils (Fig. 4).
The capillaries in the cephalic and caudal lobes are
almost independent of each other and form a
dense network around the secretory cells of ce¬
phalic and caudal lobe, respectively.

DISCUSSION

The vascularization of the turtle hypophysis has
been investigated previously, and the presence of
portal system which supplies the PD with venous
blood from the capillaries of the primary capillary
plexus in the ME has been demonstrated [2, 5-8].
The primary capillary plexus of the ME arising
from the branches of infundibular arteries is well
developed and the blood supply to the hypophysis
is fundamentally the same as in the other reptilian
species [9]. Although there are species differ¬
ences, these previous studies reveal that the portal

Hypophysial Portal System in Turtle

1015

Fig. 2. Mid-sagittal section through the infundibulum and hypophysis of India ink-injected specimen (A). Note the
two groups of primary capillary plexus and the portal vessels (B). a, primary capillary plexus in the anterior ME;
b, anterior group of portal vessels; c, secondary capillary plexus in the cephalic lobe: d, primary capillary plexus in
the posterior ME; e, posterior group of portal vessels; f, secondary capillary plexus in the caudal lobe; Arrows
with A-D represent the transverse planes of the sections in Fig. 3. A, X45; B. X77.

vessels collect the effluent blood from the primary
capillary plexus of the ME and pass vertically
downward to the PD.

The present study indicates that the ME has
distict anterior and posterior regions and is cov¬
ered by a well developed capillary network, corre¬
sponding to the anterior and posterior capillary
plexus. Moreover, there are distinguishable ante¬

rior and posterior groups of portal vessels originat¬
ing from the anterior and posterior capillary plexus
of the ME, respectively. These observations sug¬
gest that there are regional specialization of the
vascular system corresponding to the drainage of
the ME capillary blood containing hypothalamic
neurosecretory factors. The existence of regional
specialization of the vascular system in the ME is

1016

Y. OOTA AND I. KOSHIMIZU

Fig. 3. Transverse sections through the infundibulum and hypophysis of India ink-injected specimen. Primary
capillary plexus penetrating into the anterior ME is indicated in A. Secondary capillary plexus in the cephalic
lobe occupies the bottom of B. Anterior group of portal vessels can be noted in C. Primary capillary plexus in the
posterior ME and secondary capillary plexus in the caudal lobe are indicated in D. A-D, X45.

in good accord with previous findings in the turtle,
Pseudemys scripta [4]. In the turtle, Chrysemys
picta , the primary capillary net forming a dense
plexus in the ME does not show any anatomical
division into anterior and posterior plexus [2].

Holmes and Ball [10] have demonstrated that
the reptilian ME is divisible into anterior and
posterior regions, as in birds. Using electron
microscopy, structural difference between the
anterior and posterior ME has been reported in

Hypophysial Portal System in Turtle

1017

Fig. 4. Part of the cephalic lobe (A) and caudal lobe (B). 1, carminophilic acidophils arranging in cords around
sinusoid; 2, amphophils; 3, orangeophilic acidophils of small size; 4, basophils. Azan stain. A-B, X460.

the turtle [11-13].

The portal vessels break up into secondary capil¬
lary plexus in the PD. In the present studies, the
capillaries deriving from the anterior group of
portal vessels mainly supply blood into the cephal¬
ic lobe of the PD, whereas the capillary plexus
deriving from the posterior group of portal vessels
supply blood into the caudal lobe of the PD. This
arrangement of the vascular supply is also in good
accord with previous reports [4], It should be
noted that the portal vessels and secondary capil¬
lary net in the turtle, Chrysemys picta are not
separated [2].

It has been clearly demonstrated that there is
distinct cytological differentiation between the
cephalic and caudal lobes of the turtle PD. Early
investigations based on tinctorial studies have
shown five chromophilic cell types, consisting of
two types of acidophils, and three basophils in the
turtle PD [8, 9, 14, 15]. The close relationship
between the chromophilic cell types and the spe¬
cific hormone secretions has been studied by using
immunocytochemical techniques [16, 17]. Recent¬
ly, Mikami [18] has identified five types of secre¬
tory cells in the turtle PD ( Geoclemys reevesii ):
cephalic lobe (carminophilic acidophils as prolac-
tin-immunoreactive cells, amphophils as adreno¬

corticotropic hormone-immunoreactive cells), and
caudal lobe (orangeophilic acidophils as growth
hormone-immunoreactive cells, two types of
basophils as gonadotropin- and thyrotropin-
immunoreactive cells, respectively).

In the previous study of the vascularization of
the turtle, Pseudemys scripta, we raised the ques¬
tion that the presence of two distinct groups of
capillary plexus and portal vessel system might be
correlated with the cytological differentiation of
the PD. The present study has clearly showed that
the regional specialization of two distinct groups of
portal vessels is correlated with the cytological
differentiation of the PD in the turtle, Geoclemys
reevesii. It may be also postulated that the anterior
region of the ME controls the cephalic lobe and
the posterior region of the ME controls the caudal
lobe of the PD. Similar structural differentiation
of the vascular system of the hypophysis has been
demonstrated in the bird and the functional rela¬
tionship between ME and PD has been suggested
[19-21].

Few references correlating immunocytochemical
studies can be found for the hypophysiotropic
factors of reptiles. There is strong evidence that
the release of luteinizing hormone from the PD is
controlled by luteinizing hormone releasing hor-

1018

Y. OOTA AND I. KoSHIMIZU

mone (LHRH). The occurrence of LHRH-nerve
terminals has been demonstrated in the ME of
various reptiles including the turtle, Geoclemys
reevesii [22]. In the lizard, neurons containing
immunoreactive somatostatin or /9-endorphin have
been detected in the ME [23, 24], The presence of
the fibers containing the hypophysiotropic factors
in the ME may be involved in the control of the PD
function by portal vascular system.

REFERENCES

1 Enemar, A. (1960) The development of the
hypophysial vascular system in the lizards Lacerta a.
agilis Linnaeus and Anguis fragilis Linnaeus and in
the snake Natrix n. natrix (Linnaeus), with compara¬
tive remarks on the amniota. Acta Zool., 41: 141—
237.

2 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.

3 Wingstrand, K. G. (1951) The Structure and De¬
velopment of the Avian Pituitary. From a Compara¬
tive and Functional Viewpoint. C. W. K. Gleerup,
Lund.

4 Oota, Y. and Kawada, M. (1986) On the vascular
supply of hypophysis in the turtle, Pseudemys scrip-
ta. Proc. Japan Acad., 62B: 69-71.

5 Diepen, R. (1952) Vergleichend-anatomische Un-
tersuchungen fiber das Hypophysen-Hypothalamus-
System bei Amphibien und Reptilien. Verh. anat.
Ges., Jena, 50: 79-89.

6 Tayler, S. J. (1952) Vascularity of the hypophysis of
lower vertebrates. The painted turtle, Chrysemys
picta marginata Agassiz. Can. J. Zool., 30: 134-143.

7 Grignon, G. and Grignon, M. (1962) La vascular¬
ization de l’hypophyse chez la tortue terreste ( Testu -
do mauritanica) . Anat. Anz., 109: 492-506.

8 Saint Giron, H. (1970) The pituitary gland. In
“Biology of the Reptilia Vol. 3”. Ed. by C. Gans,
Academic Press, London and New York, pp. 135—
199.

9 Ball, J. N. (1981) Hypothalamic control of the pars
distaris in fishes, amphibians, and reptiles. Gen.
Comp. Endocrinol., 44: 135-170.

10 Holmes, R. L. and Ball, J. N. (1974) The Pituitary
Gland. A Comparable Account. Cambridge Univ.
Press, London.

11 Tsuneki, K. (1976) Neurosecretory axo-axonic
synapses in the median eminence of the turtle,
Geoclemys reevesii. Neuroendocrinology, 20: 59-67 .

12 Oota, Y. (1980) Ultrastructural aspects of the ante¬
rior region of the median eminence of the turtle.

Geoclemys reevesii. Rep. Fac. Sci., Shizuoka Univ.,
14: 75-88.

13 Oota, Y. (1981) Ultrastructural aspects of the pos¬
terior region of the median eminence of the turtle,
Geoclemys reevesii. Rep. Fac. Sci., Shizuoka Univ.,
15: 29-38.

14 Doerr-Schott, J. (1976) Immunohistochemical de¬
tection, by light and electron microscopy of pituitary
hormones in cold-blooded vertebrates. II. Reptiles.
Gen. Comp. Endocrinol., 28: 513-529.

15 Licht, P. and Pearson, A. K. (1978) Cytophysiology
of the reptilian pituitary gland. Int. Rev. Cytol.,
Suppl., 7: 239-289.

16 Pearson, A. K. and Licht, P. (1982) Morphology
and immunocytochemistry of the turtle pituitary
gland with special reference to the pars tuberalis.
Cell Tissue Res., 222: 81-100.

17 Pearson, A. K., Wurst, G. Z. and Cadle,J. E.
(1983) Ontogeny and immunocytochemical dif¬
ferentiation of the pituitary gland in the sea turtle,
Caretta caretta. Anat. Embryol., 167: 13-37.

18 Mikami, S. (1986) Comparative cytology of the
anterior pituitary of higher vertebrates. In “Pars
Distalis of the Pituitary Gland — Structure, Func¬
tion and Regulation”. Ed. by F. Yoshimura and A.
Gorbman, Elsevier Sci. Publ. B.V., Amsterdam,
pp. 71-79.

19 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.

20 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.

21 Dominic, C. J. and Singh, R. M. (1969) Anterior
and posterior groups of portal vessels in the avian
pituitary. Gen. Comp. Endocrionol., 13: 22-26.

22 Nozaki, M., Tsukahara, T. and Kobayashi, H.
(1984) Neuronal systems producing LHRH in verte¬
brates. In “Endocrine Correlates of Reproduction”.
Ed. by K. Ochiai, Y. Arai, T. Shioda and M. Taka-
hashi, Japan Sci. Soc. Press, Tokyo/Springer-
Verlag, Berlin, pp. 3-27.

23 Fasolo, A. and Gaudino, G. (1982) Immunohis¬
tochemical localization of somatostatin-like im-
munoreactivity in the hypothalamus of the lizard,
Lacerta muralis. Gen. Comp. Endocrinol., 48: 205-
212.

24 Vallarino, M. (1986) /?-Endorphin-like immuno-
reactivity in the brain of the lizard, Lacerta muralis.
Gen. Comp. Endocrinol., 64: 52-59.

ZOOLOGICAL SCIENCE 5: 1019-1032 (1988)

© 1988 Zoological Society of Japan

Fetal and Postnatal Development of Arginine
Vasopressin-Immunoreactive Neurons in the Mouse

Tomoo Yamashita* 1, Keiichi Kawamoto2 and Seiichiro Kawashima3

Zoological Institute, Faculty of Science, Hiroshima University, Hiroshima 730,
and 3 Zoological Institute, Faculty of Science, University of Tokyo, Tokyo 1 13, Japan

ABSTRACT — The ontogeny of arginine vasopressin (AVP)-producing neurons in the hypothalamo-
neurohypophysial system (HNS) was immunocytochemically studied in fetal and postnatal mice.
Presumptive AVP neurons underwent proliferation in the vicinity of the third ventricle and migrated to
settle in the final loci by fetal age of 18 days (FA 18). AVP-immunoreactive neurons were first detected
on FA 14 in the presumptive supraoptic (SON) and retrochiasmatic nuclei (RCN). AVP-
immunoreactive axons and terminals were present in the median eminence and pars nervosa of the
neurohypophysis on FA 15, but not on FA 14. Immunoreactive neurons were recognized in the
paraventricular nucleus (PVN) on FA 15 and their terminals in the external layer of the median
eminence became immunoreactive on FA 18. In the suprachiasmatic nucleus (SCN) AVP-
immunoreactive perikarya appeared on FA 16. The number of AVP neurons in the SON and RCN
markedly increased during fetal life. Postnatal increase in the number of immunoreactive neurons in the
PVN and SCN as well as that of the SON and RCN was apparent. To sum up, the present study shows
that cytodifferentiation of AVP-producing neurons in the HNS takes place during early days of last
trimester of pregnancy and that the HNS completes the general morphological changes before birth.

INTRODUCTION

Arginine vasopressin (AVP) is synthesized
mainly in the magnocellular neurons in the supra¬
optic (SON), retrochiasmatic (RCN) and paraven¬
tricular nuclei (PVN) of the hypothalamus in mam¬
mals [1-5]. AVP is transported through the fiber
layer of the median eminence (ME) to the pars
nervosa (PN) of the pituitary and to the external
layer of the ME [6]. Some other AVP-producing
neurons are present among the parvocellular
neurons in the suprachiasmatic nucleus (SCN), the
axons originating from which projecting to the
brain regions other than the PN and ME [7, 8].
Besides the vasopressinergic pathways from the

Accepted January 5, 1988
Received December 5, 1987
* Fetal ages in the cited paper were rearranged for
convenience of comparison with the present study, so
that the day when vaginal plug was observed was
designated as FA 0.

1 Present Address: Numata High School, Tomo 161-1,
Numata-cho, Asaminami-ku, Hiroshima 731-31.

2 Present Address: Department of Neurosurgery, School
of Medicine, Hiroshima University, Hiroshima 734.

3 Reprint requests to: Dr. S. Kawashima.

PVN to the PN and ME [9-12], extrahypothalamic
projections have also been documented, for exam¬
ple, to the forebrain [8, 12], the brain stem [8, 13]
and the spinal cord [8, 13, 14]. Therefore, AVP
has been proposed to act not only as the antidiure¬
tic hormone but also as a neurotransmitter.

During the last ten years the ontogeny of the
hypothalamo-neurohypophysial system (HNS) has
been investigated in some species of rodents. His¬
tological staining [15], electron microscopy [16-
20], immunocytochemistry [21-23] and radioim¬
munoassay [19, 24, 25] were applied for the study
of development of the HNS. These studies have
demonstrated that AVP synthesis in the mag¬
nocellular neurons begins to occur and rapidly
elevates during the late gestation period and that
the HNS completes its maturation by the end of
the first month of life. Compared to the studies on
other species of mammals, however, developmen¬
tal study on the mouse HNS is scanty.

Therefore, in the present study the development
of AVP-producing neurons of the HNS in fetal and
postnatal mice was examined by means of AVP
immunocytochemistry.

1020

T. Yamashita, K. Kawamoto and S. Kawashima

MATERIALS AND METHODS

Animals

Mice of the C57BL/6NCrj strain maintained in
this laboratory were used in the present study.
They were housed in a temperature-controlled
room at 12-hr light (06: 00-18: 00 hr) and 12-hr
dark cycle with free access to laboratory chow
(CA-1, Japan Clea Inc.) and tap water.

Female mice were placed in a cage with males in
the evening and separated from males in the next
morning. The day was designated as fetal age of 0
day (FA 0) for successful pregnancy. Most preg¬
nant mothers delivered their pups in the morning
on FA 19. The day of birth was designated as
postnatal age of 0 day (PA 0).

Tissue preparation

Fetuses between FA 14 and FA 18 regardless of
sexes were taken out by Caesarean cut at 13: 00 hr
from at least two different litters at each fetal age.
After birth, in order to exclude any possible sex
difference, only male mice were chosen and killed
at 13: 00 hr on PA 2, 14, 30 and 90. Five animals
were used for each fetal and postnatal age groups.

Animals were killed by decapitation, and the
brains were taken out and fixed in Bouin’s fluid for
two days. After trimming all the brains were kept
in 70% ethanol overnight. Dehydration in a
graded series of ethanol and embedding in para-
plast were completed within the following day.
Serial frontal sections were cut at 6 pm in thick¬
ness, and every fifth sections were mounted on
albumin-coated glass slides for immunocytochem-
ical staining. The adjacent sections were stained
with Ehrlich’s haematoxyline-eosin (fetuses) or
thionine (postnatal mice) for general histological
changes during development of the HNS.

Immunocy to chemical procedures

Immunocytochemistry for AVP was performed
by means of the avidin-biotin-peroxidase complex
(ABC) technique [26]. Deparaffinized sections
were reacted with the following sequence of solu¬
tions: (1) rabbit anti-AVP serum (RV-1K, raised
in this laboratory) (1:6,400) for 24 hr at 4°C, (2)
biotinylated goat anti-rabbit IgG serum (Vector

Laboratories, California) (1:200) for 30 min at
room temperature (RT), (3) 0.3% H202 for 30
min at RT, (4) ABC reagents (avidin DH and
biotinylated horseradish peroxidase H) of the
ABC kit (Vector Laboratories, California) (1: 100)
for 30 min at RT and (5) 0.015% 3,3'-

diaminobenzidine in 0.05 M Tris-HCl buffer (pH
7.4) containing 0.01% H202 for 10 min at RT.
Sections were washed three times with 0.01 M
phosphate buffered saline (pH 7.4) at 4°C, 5 min
each, between each step. The characterization of
anti-AVP serum was reported previously [27].

After washing in three changes of distilled wa¬
ter, the preparations were dehydrated with graded
series of ethanol and mounted with balsam.

Morphometry

The right halves of the hypothalamus were used
for morphometry. In immunocytochemically
stained sections, only those cells that showed dis¬
tinct nucleoli and contained brown reaction prod¬
ucts were counted as AVP-producing neurons.
The total number of cells (N) per animal was
calculated from the total number of cells (n) in
every fifth sections by the formula: N = 2x5xn.

RESULTS

1. Development of the HNS

FA 14. General morphology of the hypothala¬
mus was extremely different from that in the adults
(Fig. la). Presumptive magnocellular neurons
appeared to be produced in the diamond-shaped
region around the third ventricle at the junction of
the ventral and medial lobes of the diencephalic
neuroepithelium. Ependymal cells were spindle¬
like in shape, forming two or three layers.

The condensation of the SON, RCN and PVN
was not completed on FA 14. However, in the
presumptive region of the SON, the magnocellular
neurons were a little more concentrated than the
surrounding area. The presumptive PVN con¬
tained the cells migrating from the neuroepithe¬
lium to the final locus in the hypothalamus. The
SCN was not yet demarcated from the adjacent
area. The ependymal cells along the ventral floor
of the third ventricle were still undergoing mitosis.

Development of AVP Neurons in Mice

1021

Fig. 1. Frontal sections of the hypothalamus (right side) during development. Stained with haematoxyline and eosin
(a-c), or thionine (d). a, FA 14; b, FA 16; c, FA 18; d, PA 14. IIIV, third ventricle. Bars: 200 /an.

1022

T. Yamashita, K. Kawamoto and S. Kawashima

The ME was consisted of the ependymal and
fiber layers, and the external layer was not yet
differentiated. Scattered cells of the pars tuberalis
were visible under the ME. The PN was immature
and full of glial cells (primordial pituicytes) sur¬
rounding a central cavity and possesed very few, if
any, axonal termini. The periphery of the PN was
well vascularized.

FA 15. Ependymal cells in the diamond¬
shaped region around the third ventricle continued
to divide, and some of them might differentiate
into the hypothalamic cells, because the migratory
paths of cells from the matrix layer along the third
ventricle to the ventro-lateral region of the
hypothalamus was observed.

At this fetal age the SON could be recognized as
the cell cluster laterally to the optic tract, but many
neurons were still migrating to the SON. The PVN
appeared to contain both migrating and settled
cells. The RCN and SCN were not clearly demar¬
cated.

The fiber layer of the ME was better developed
than that on FA 14. Though the external layer of
the ME was not at all differentiated, the number of
cells of the pars tuberalis was increased on the
ventral surface of the ME. Because of the penetra¬
tion of axons to the outer area of the PN, it was
clearly divided into two areas; the inner pituicyte-
rich area and the outer fibrous area.

FA 16. By this day, most ependymal cells
lining the third ventricle appeared to cease any
further mitotic divisions, and the neurons migrat¬
ing to the ventrolateral regions of the hypothala¬
mus were very few (Fig. lb). Morphogenesis of
the SON was almost completed on FA 16, but a
few neurons were still migrating to this nucleus.
The RCN and PVN were clearly recognized as cell
clusters. The SCN showed itself as a slightly dense
area on the optic chiasma.

General morphology of the neurohypophysis
(ME and PN) did not show much difference as
compared to that on FA 15. The penetration of
blood vessels into the pars tuberalis and the inner
area of the PN was the characteristic phenomenon
at this stage.

FA 17. The cytoplasm of magnocellular
neurons in the SON and PVN was not well-
developed. The parvocellular SCN formed a dis¬

crete structure, locating as a pair close to the
medial optic chiasma.

The external layer of the ME was developed.
The PN was markedly developed on FA 17, and in
some fetuses the central cavity had disappeared.
Pituicytes dispersed all around the PN and inter¬
mingled with axons, so that exclusively fibrous
area was left only at the periphery.

FA 18. On this last day of embryonic life, the
hypothalamic nuclei containing AVP-producing
neurons completed development (Fig. lc). The
adult-type partition of the RCN into lateral and
medial groups was observed.

Further development of the external layer of the
ME was seen, but the layer was thinner than that
in adult animals. Pituicytes further dispersed in
the PN, and among these cells a greater number of
axons was present than on FA 17.

These observations indicate that the mor¬
phogenesis of the HNS is almost completed prior
to the delivery.

Postnatal development. The postnatal develop¬
ment of the hypothalamus was characterized by
the hypertrophy of each neuron (Fig. Id). The
nucleoli in the neuronal nuclei became distinct.
Though the hypothalamic neurons were still
undergoing condensation on PA 2, the intercellu¬
lar spaces became spread after PA 14. However,
the hypothalamic nuclei were obviously distin¬
guished from the surrounding area during the
postnatal maturation.

After birth, the development of the fiber and
external layers of the ME further advanced, and
the growth of the PN was apparent.

2. Development of AVP-producing Neurons

The changes of AVP immunoreactivity of the
hypothalamus and neurohypophysis during fetal
and postnatal development are summarized in
Table 1.

Fetal life

Development of the hypothalamus. On FA 14,
some neurons of the presumptive SON in three out
of the five fetuses were weakly reactive to anti-
AVP serum. Weak AVP immunoreactivity was
also detected in the ventro-caudal part of the RCN
in all animals. The number of immunoreactive

Development of AVP Neurons in Mice

1023

Table 1. Resume of AVP immunoreactivity of the mouse hypothalamus and neurohypophysis
during fetal and postnatal development

Age (days)

Hypothalamus

Neurohypophysis

SON

RCN

PVN

SCN

Ace.

ME

f.l. e.l.

PN

Fetal age

14

+

+

-

-

-

-

-

-

15

+

+

+

-

+

+

-

+

16

+

+

+

+

+

+

-

+

17

+ +

+ +

+

+

+ +

+ +

+

+ +

18

+ +

+ +

+

+

+ +

+ +

+

+ +

Postnatal age

2

+ +

+ +

+ +

+

+ +

+ +

+ +

+ +

14

+ + +

+ + +

+ + +

+ +

+ +

+ + +

+ +

+ + +

30

+ + +

+ + +

+ + +

+ +

+ + +

+ + +

+

+ + +

90

+ + +

+ + +

+ + +

+ +

+ + +

+ + +

+

+ + +

Staining intensity was

arbitrarily graded as:

— , none;

+ , weak; + + ,

moderate;

+ + + ,

intense.

Abbreviations: SON, supraoptic nucleus; RCN, retrochiasmatic nucleus; PVN, paraventricular nucleus;
SCN, suprachiasmatic nucleus; Ace., accessory nuclei; ME, median eminunce; f.l., fiber layer; e.l.,
external layer; PN, pars nervosa.

Accessory nuclei are consisted of the anterior commissural nucleus, nucleus circularis, and anterior and
posterior fornical nuclei.

Fig. 2. Frontal sections of the right supraoptic nucleus during development. Stained with anti-AVP serum, a, FA
15; b, FA 18; c, PA 14; d, PA 90. OC, optic chiasma. Bar: 50 fj.m.

1024

T. Yamashita, K. Kawamoto and S. Kawashima

neurons in the RCN was more than in the SON.
There was no AVP immunoreactivity in other
nuclei of the hypothalamus.

On FA 15, the AVP immunoreactivity of the
SON and RCN neurons became apparent, though
the stainability was still weak (Figs. 2a and 3a).
Each cell nucleus was enclosed in a thin rim of
cytoplasm. There were some neurons which did
not completely settle in the SON. A few neurons
in the PVN and internuclear magnocellular (acces¬
sory) nuclei between the SON and PVN became
first stainable on this day. But no immunoreactive
material was detected in the SCN.

AVP immunoreactivity in the PVN on FA 16
was almost the same as that on FA 15 (Fig. 4a).
Weak AVP immunoreactivity was first detected in
the SCN in one of five FA 16 fetuses. Adult-type
partition of immunoreactive neurons in the RCN
into medial and lateral groups was observed as

early as this fetal day.

Immunostainability of the SON and RCN
neurons increased on FA 17 (Table 1). The SCN
neurons were only weakly stainable with the anti¬
serum in three fetuses on FA 17 (Fig. 5a).

On FA 18, many neurons in the SON and RCN
accumulated the AVP-immunoreactive material in
their cytoplasmic rim. Axonal fibers from the
neurons with the bead-like deposits were encoun¬
tered in these nuclei (Figs. 2b and 3b).

Numbers of immunoreactive neurons (Fig. 8).
During the fetal life, magnocellular neurons in the
SON, RCN and PVN did not grow as large as
those of the adult. The number of immunoreactive
neurons generally increased during the gestation
period. In the SON and RCN, the number of
AVP-immunoreactive neurons rapidly increased
between FA 14 and FA 18. More immunostained
cells were counted in the SON than in the RCN,

Fig. 3. Frontal sections of the right retrochiasmatic nucleus during development. Stained with anti-AVP serum, a,
FA 15; b, FA 18; c, PA 14; d, PA 90. Arrows show the weakly AVP-immunoreactive neurons. IIIV; third
ventricle. Bar: 200 /j.m.

Fig. 4. Frontal sections of the right paraventricular nucleus during development. Stained with anti-AVP serum, a,
FA 16; b, PA 2; c, PA 14; d, PA 90. IIIV, third ventricle. Bar: 50 //m. (in page 1025)

Development of AVP Neurons in Mice

1025

M

4

1026

T. Yamashita, K. Kawamoto and S. Kawashima

Development of AVP Neurons in Mice

1027

Fig. 6. Frontal sections of the median eminence during development. Stained with anti-AVP serum, a, FA 15; b,
FA 18; c, PA 14; d, PA 90. Arrows show the AVP-immunoreactive axonal termini in the external layer of the
median eminence. IIIV, third ventricle; PT, pars tuberalis; bv, blood vessel. Bar: 50 /an.

except on FA 14. In comparison with these nuclei,
the numbers of immunoreactive neurons in the
PVN, SCN and accessory nuclei showed a slight
increase during the fatal life.

Development of the neurohypophysis. With the
maturation of AVP-producing neurons, stainabil-
ity of the ME and PN was also enhanced. AVP
immunoreactivity was first detected on FA 15 in
the ME and the peripheral edge of the PN (Figs.
6a and 7a). AVP-immunoreactive axonal termini
in the external layer of the ME first appeared on
FA 18 (Fig. 6b). Immunoreactive area in the PN
indicated the preferential peripheral localization at
the early stage of development (Fig. 7ab), and it
gradually expanded to the central area. However,
even on FA 17 and 18 strong immunoreactivity was
confined to the peripheral area (Fig. 7cd).

Postnatal life

Development of the hypothalamus. After birth,
the maturation of magnocellular neurons in the
SON, RCN and PVN further advanced. The cells
were hypertrophied and became oval in shape.
The hypertrophy was generally accompanied by
the accumulation of AVP-immunoreactive mate¬
rial (Figs. 2cd, 3cd and 4d).

On PA 2, immunoreactive area of the PVN
became triangular in shape like that of the adult.
With the rapid increase in the number of im¬
munoreactive cells in the PVN between PA 2 and
PA 14, stainability of the cells also increased (Fig.
4bc).

In the parvocellular SCN the increase in AVP
immunoreactivity occurred mainly during postna-

Fig. 5. Frontal sections of the right suprachiasmatic nucleus during development. Stained with anti-AVP serum, a,
FA 17; b, PA 14; c, PA 30; d, PA 90. Arrows show the weakly AVP-immunoreactive neurons. OC, optic
chiasma. Bar: 50 /an. (in page 1026)

1028

T. Yamashita, K. Kawamoto and S. Kawashima

Fig. 7. Frontal sections of the pars nervosa (PN) during fetal development. Stained with anti- A VP serum, a, FA 15;
b, FA 16; c, FA 17; d, FA 18. Arrows show the AVP-immunoreactive area in the periphery of the PN. Weak
non-specific reactivity is seen in the pars intermedia (PI) and pars distalis (PD). Bar: 50 //m.

Fig. 8. Total number of AVP-immunoreactive neurons in the mouse hypothalamus during fetal and postnatal
development. Vertical bars indicate the standard errors of the means. The number of animals was five in all age
groups. For abbreviations, see the footnotes of Table 1.

Development of AVP Neurons in Mice

1029

tal development (Table 1). Adult-like distribution
consisted of the dorso-medial and ventro-lateral
groups was seen on PA 30 (Fig. 5c). Many
neurons were weakly stained with the antiserum
(Fig. 5d).

Bead-like structure of immunoreactive axons
became more frequently encountered in the vicin¬
ity of the neurons as the accumulation in the
perikarya proceeded. For example, the axons
containing AVP-immunoreactive material were
found in the PVN and SCN (Fig. 5b) on PA 2 and
PA 14, respectively.

Most of the neurons of the accessory nuclei were
strongly immunoreactive to anti- A VP serum after
birth (Table 1).

Numbers of immunoreactive neurons (Fig. 8).
The numbers of AVP-immunoreactive neurons in
the RCN and accessory nuclei were stable after PA
2. The numbers in the SON and PVN continued to
increase until PA 14 and maintained the level of
PA 14 thereafter. However, there was a marked
difference in the number of AVP-immunoreactive
neurons between the SON and PVN. On the other
hand, the number of immunoreactive neurons of
the SCN continued to increase until PA 90.

Development of the neurohypophysis. AVP im-
munoreactivity in the fiber layer of the ME and the
PN on PA 14 was stronger than on PA 2, but after
PA 14 it was almost constant (Table 1). Postnatal
development of the external layer of the ME was
marked (Fig. 6c). Many AVP-immunoreactive ax¬
ons with the bead-like deposits grew out from the
fiber layer down to the external layer of the ME,
and their termini were settled on the primary
capillary plexus (Fig. 6d). AVP immunoreactivity
in the external layer of the ME has decreased
concomitantly with the development of the ME
(Table 1).

DISCUSSION

Several investigators have reported the develop¬
mental changes of neurons in the hypothalamic
nuclei by means of [3H]-thymidine autoradiogra¬
phy [28, 29]. In the rat, Ifft [28]* showed that the

* Fetal ages in the cited paper were rearranged for
convenience of comparison with the present study, so
that the day when vaginal plug was observed was
designated as FA 0.

neurons of the SON became visible between FA 11
and FA 15, those of the RCN between FA 11 and
FA 17, those of the PVN between FA 11 and FA
16, and those of the SCN between FA 11 and FA
17 in the matrix layer surrounding the third ventri¬
cle. On the other hand, in the mouse of which
gestation period is about two days shorter than
that of the rat, Okamura et al. [29] reported the
time of origin of AVP-producing neurons by com¬
bined technique of immunocytochemistry and
autoradiography. According to their results,
AVP-producing neurons in the SON and PVN
seem to differentiate between FA 10 and FA 12.
In the SCN the neurons proliferate between FA 10
and FA 14, and most of them appear to be
produced in the latter half of this period. After the
final division in the matrix layer, hypothalamic
neurons migrate to occupy the final loci within 24
hr [30]. In addition, Altman and Bayer [31]
suggested that there is a general lateral-to-medial
internuclear differentiation gradient of the
hypothalamic neurons.

In the present study, AVP-immunoreactive
neurons were first detected on FA 14 in the SON
and RCN but not in other nuclei of the mouse
hypothalamus. Evidently, there was a time lag
between the stage of neuronal proliferation and
the stage of AVP synthesis. Since we could not
observe immunoreactivity in the neurons migrat¬
ing to the SON and RCN, the settlement of cells in
the final loci may have some positive effects on the
initiation of AVP synthesis. Silverman et al. [32],
however, reported that on FA 13 some presump¬
tive SON neurons synthesized the AVP carrier
protein, neurophysin, while they were still migrat¬
ing. Because of the limit of sensitivity of im-
munocytochemical reaction, it is possible that
there was an earlier onset of AVP synthesis in the
hypothalamic neurons and/or that many more cells
contained subdetectable levels of AVP. In fact, a
very small amount of AVP was detected by
radioimmunoassay in the mouse brain on FA 13
(our unpublished observation). If immunocy¬
tochemistry using antibodies to the hormone pre¬
cursor or autoradiography using the labelled oligo¬
nucleotide probe to AVP messenger RNA was
performed, the time of cytodifferentiation of AVP
neurons might be found at earlier embryonic life.

1030

T. Yamashita, K. Kawamoto and S. Kawashima

Magnocellular neurons developed rapidly dur¬
ing the fetal life. Especially in the SON and RCN,
the number of AVP-immunoreactive neurons in¬
creased linearly during the fetal life and reached
the adult level on PA 2. On the other hand, the
PVN appeared to develop later than the SON and
RCN, although the number of AVP-
immunoreactive neurons of the PVN reached the
adult level on PA 14. These results are consistent
with the previous studies [21, 22, 32], Since the
neurons of the SON, RCN and PVN cease to
divide and start migration almost at the same time
[28, 29, 31, 33], it needed to be clarified why the
SON and RCN neurons should be so advanced in
increasing their numbers. The SCN neurons be¬
came immunoreactive in one animal on FA 16.
The later accumulation of AVP in the parvocellu-
lar neurons may be related with their late with¬
drawal from the mitotic cycle as compared to the
magnocellular neurons [28, 29, 31, 33]. This delay
might imply a less significant physiological import¬
ance of the SCN than the SON, RCN and PVN
during perinatal period. Because parvocellular
neurons are known to project their axons to the
forebrain and brain stem [7, 8], but not to the PN
and ME, they may not be related to the regulation
of water and electrolyte metabolism. It is likely
that AVP in the SCN neurons begin to function as
a neurotransmitter after birth, concomitant with
their increase in AVP immunoreactivity.

The adult-type distribution pattern of SON
neurons completes during fetal life, that is, more
AVP neurons are present in the caudal region of
the SON than the rostral region. This particular
distribution pattern has also been shown in the rat
[34].

The occurrence of bead-like deposits along the
axons implies the active transport of neurosecre¬
tory material from the hypothalamus to the PN
and ME. After birth, many large-sized deposits or
Herring bodies could be observed in the axons of
the hypothalamo-neurohypophysial tracts. Axonal
outgrowth from the PVN neurons also occurred
relatively quickly after the appearance of im¬
munoreactivity in the perikarya. AVP-immuno¬
reactive cell bodies were encountered in the SON
and PVN on FA 15, and their termini in the
external layer of the ME, which recieves projec¬

tions of AVP-immunoreactive fibers almost exclu¬
sively from the PVN [11, 12], were detected on FA
18.

Immunoreactive axons were first observed in the
peripheral area of the PN on FA 15. The age of
the fetus at which AVP-immunoreactive axons
were first detected in the present study was older
as compared with the result using neurophysin
immunocytochemistry by Silverman et al. [32],
The findings in the present study were in harmony
with those by Eurenius and Jarskar [17] in that
small fiber bundles were found in the PN, primari¬
ly in the region bordering the intermediate lobe on
FA 14. Their data suggest that axonal outgrowth is
initiated at least in some cells very shortly after
their arrival at the presumptive locus of the
hypothalamic nuclei (the arrival is between FA 13
and FA 14). Our results seem to show that the
transport of the AVP begins after the projection of
axonal termini reaches the PN. At present we
don’t know how long the time lag for the transport
of AVP from the perikarya to the termini last, but
if it is very short, then it is possible that the
hormone is present in the growth cone, the very
front of the axonal growth. Previous studies
suggested that the fetal PN is a heterogeneous
structural entity and that there are distinct territor¬
ies for ingrowing fibers [17, 32]. Present observa¬
tions also demonstrated that in the fetal PN AVP-
immunoreactive axons were exclusively present in
the peripheral area. The fact that the central area
of the developing PN was occupied by proliferating
pituicytes may have some is closely related to the
peripheral distribution of growing fibers.

The present study clarified the developmental
events of AVP-immunoreactive neurons of the
mouse HNS. To sum up, the present results show
that (1) there may be a time lag between the
withdrawal from mitotic cycle and the initiation of
peptide synthesis in AVP neurons, that (2) the
HNS develops rapidly to attain the adult-like pat¬
tern during fetal life, and that (3) the HNS nearly
attains complete maturity by the time of weaning.

REFERENCES

1 Vandesande, F. and Dierickx, K. (1975) Identifica¬
tion of the vasopressin producing and of the oxyto-

Development of AVP Neurons in Mice

1031

cin producing neurons in the hypothalamic mag-
nocellular neurosectetory system of the rat. Cell
Tissue Res., 164: 153-162.

2 Choy, V. J. and Watkins, W. B. (1977) Immunocy-
tochemical study of the hypothalamo-neuro-
hypophysial system. II. Distribution of neurophysin,
vasopressin and oxytocin in the normal and osmoti-
cally stimulated rat. Cell Tissue Res., 180: 467-490.

3 Sofroniew, M. V., Weindl, A., Schinko, I. and
Wetzstein, R. (1979) The distribution of vasopres¬
sin-, oxytocin-, and neurophysin-producing neurons
in the guinea pig brain. Cell Tissue Res., 196: 367-
384.

4 Rhodes, C. H., Morrell, J. I. and Pfaff, D. W.
(1981) Immunohistochemical analysis of mag-
nocellular elements in rat hypothalamus: Distribu¬
tion and numbers of cells containing neurophysin,
oxytocin, and vasopressin. J. Comp. Neurol., 198:
45-64.

5 Kawamoto, K. (1985) Immunohistochemical local¬
ization of vasopressin and oxytocin neurons in the
mouse hypothalamus. J. Sci. Hiroshima Univ., Ser.
B, Div. 1, 32: 25-33.

6 Brownstein, M. J., Russell, J. T. and Gainer, H.
(1980) Synthesis, transport, and release of posterior
pituitary hormones. Science, 207: 373-378.

7 Hoorneman, E. M. D. and Buijs, R. M. (1982)
Vasopressin fiber pathways in the rat brain following
suprachiasmatic nucleus lesioning. Brain Res., 243:
235-241.

8 Zimmerman, E. A., Nilaver, G., Hou-Yu, A. and
Silverman, A. J. (1984) Vasopressinergic and oxy-
tocinergic pathways in the central nervous system.
Fed. Proc., 43: 91-96.

9 Vandesande, F., De Mey, J. and Dierickx, K.
(1974) Identification of neurophysin producing
cells. I. The origin of the neurophysin-like sub¬
stance-containing nerve fibres of the external region
of the median eminence of the rat. Cell Tissue Res.,
151: 187-200.

10 Dierickx, K., Vandesande, F. and De Mey, J.

(1976) Identification, in the external region of the
rat median eminence, of separate neurophysin-
vasopressin and neurophysin-oxytocin containing
nerve fibres. Cell Tissue Res., 168: 141-151.

11 Vandesande, F., Dierickx, K. and De Mey, J.

(1977) The origin of the vasopressinergic and oxyto-
cinergic fibers of the external region of the median
eminence of the rat hypophysis. Cell Tissue Res.,
180: 443-452.

12 Antunes, J. L., Carmel, P. W. and Zimmerman, E.
A. (1977) Projections from the paraventricular
nucleus to the zona externa of the median eminence
of the rhesus monkey: An immunohistochemical
study. Brain Res., 137: 1-10.

13 Lang, R. E., Heil,J., Ganten,D., Hermann, K.,

Rascher, W. and Unger, Th. (1983) Effects of le¬
sions in the paraventricular nucleus of the hypothal¬
amus on vasopressin and oxytocin contents in brain¬
stem and spinal cord of rat. Brain Res., 260: 326-
329.

14 Hawthorn, J., Ang, V. T. and Jenkins, J. S. (1985)
Effects of lesions in the hypothalamic paraventricu¬
lar, supraoptic and suprachiasmatic nuclei on vaso¬
pressin and oxytocin in rat brain and spinal cord.
Brain Res., 346: 51-57.

15 Rinne, U. K. and Kivalo, E. (1965) Maturation of
hypothalamic neurosecretion in rat, with special
reference to the neurosecretory material passing
into the hypophysial portal system. Acta
Neuroveget., 27: 166-183.

16 Pauli, W. K. (1973) A light and electron microscop¬
ic study of the development of the neurohypophysis
of the fetal rat. Anat. Rec., 175: 407-408.

17 Eurenius, L. and Jarskar, R. (1974) Electron mi¬
croscopy of neurosecretory nerve fibres in the neural
lobe of the embryonic mouse. Cell Tissue Res., 149:
333-347.

18 Galabov, P. and Schiebler, T. H. (1978) The ultra¬
structure of the developing neural lobe. Cell Tissue
Res., 189: 313-329.

19 Gash, D., Sladek, C. and Scott, D. (1980) Cytodif-
ferentiation of the supraoptic nucleus correlated
with vasopressin synthesis in the rat. Brain Res.,
181: 345-355.

20 Kawamoto, K., Yamada, Y. and Kawashima, S.
(1984) Immunohistochemical and electron micro¬
scopic studies on the development of the mouse
posterior lobe. Zool. Sci., 1: 980 (Abstract).

21 Silverman, A. J. (1975) The hypothalamic mag-
nocellular neurosecretory system of the guinea pig.
II. Immunohistochemical localization of neuro¬
physin and vasopressin in the fetus. Am. J. Anat.,
144: 445-460.

22 Wolf, G. and Trautmann, B. (1977) Ontogeny of
the hypothalamo-neurohypophysial system in rats —
an immunohistochemical study. Endokrinologie, 69:
222-226.

23 Choy, V. J. and Watkins, W. B. (1979) Maturation
of the hypothalamo-neurohypophysial system. I.
Localization of neurophysin, oxytocin and vasopres¬
sin in the hypothalamus and neural lobe of the
developing rat brain. Cell Tissue Res., 197: 325-
336.

24 Sinding, C., Robinson, A. G., Seif, S. M. and
Schmid, P. G. (1980) Neurohypophyseal peptides in
the developing rat fetus. Brain Res., 195: 177-186.

25 Sinding, C., Seif, S. M. and Robinson, A. G. (1980)
Levels of neurohypophyseal peptides in the rat
during the first month of life. I. Basal levels in
plasma, pituitary, and hypothalamus. Endocrinolo¬
gy, 107: 749-754.

1032

T. Yamashita, K. Kawamoto and S. Kawashima

26 Hsu, S.-M., Raine, L. and Fanger, H. (1981) Use of
avidin-biotin-peroxidase complex (ABC) in im-
munoperoxidase techniques: A comparison between
ABC and unlabeled antibody (PAP) procedures. J.
Histochem. Cytochem., 29: 577-580.

27 Kawamoto, K. and Kawashima, S. (1986) Effects of
glucocorticoids and vasopressin on the regeneration
of neurohypophyseal hormone-containing axons af¬
ter hypophysectomy. Zool. Sci., 3: 723-736.

28 Ifft, J. D. (1972) An autoradiographic study of the
time of final division of neurons in the rat hypothala¬
mic nuclei. J. Comp. Neurol., 144: 193-204.

29 Okamura,H., Fukui,K., Koyama, E., Tsutou, H.
L. O., Tsutou, T., Terubayashi, H., Fujisawa, H.
and Ibata, Y. (1983) Time of vasopressin neuron
origin in the mouse hypothalamus: Examination by
combined technique of immunocytochemistry and
[3H]thymidine autoradiography. Dev. Brain Res., 9:
223-226.

30 Shimada, M. and Nakamura, T. (1973) Time of
neuron origin in mouse hypothalamic nuclei. Exp.
Neurol., 41: 163-173.

31 Altman, J. and Bayer, S. A. (1978) Development of
the diencephalon in the rat. II. Correlation of the
embryonic development of the hypothalamus with
the time of origin of its neurons. J. Comp. Neurol.,
182: 973-994.

32 Silverman, A. J., Goldstein, R. and Gadde, C. A.
(1980) The ontogenesis of neurophysin-containing
neurons in the mouse hypothalamus. Peptides, 1:
27-44.

33 Anderson, C. H. (1978) Time of neuron origin in
the anterior hypothalamus of the rat. Brain Res.,
154: 119-122.

34 Swaab,D.F., Nijveldt, F. and Pool, C. W. (1975)
Distribution of oxytocin and vasopressin in the rat
supraoptic and paraventricular nucleus. J. Endocri¬
nol., 67: 461-462.

ZOOLOGICAL SCIENCE 5: 1033-1042 (1988)

© 1988 Zoological Society of Japan

Molecular- and Immuno-histochemical Study on Expressions of
Vasopressin and Oxytocin Genes Following Sodium Loading

Susumu Hyodo* 1, Mamoru Fujiwara, Moriyuki Sato2
and Akihisa Urano1

Department of Regulation Biology, Faculty of Science, Saitama University,
Urawa, Saitama 338, and 2 Tokyo Research Labolatories, Kyowa
Hakko Kogyo Co., Machida, Tokyo 194, Japan

ABSTRACT — We investigated the effects of drinking hypertonic saline on expressions of vasopressin
(AVP) and oxytocin (OXT) genes in neurosecretory neurons of the supraoptic (SON) and the
paraventricular (PVN) nuclei in rats with the oligonucleotide-mRNA in situ hybridization and the
immunohistochemical avidin-biotin-peroxidase complex (ABC) methods. Autoradiographic hybridiza¬
tion signals that indicate the localization of AVP mRNA were significantly increased in the SON 4 days
after the commencement of the sodium loading, while the signal increase in the PVN was less
conspicuous than that in the SON. Signals for OXT mRNA in both the SON and the PVN were rapidly
increased and attained to a peak level by the second day after the onset of sodium loading. In the
hypothalamic sections adjacent to those used for in situ hybridization, the number of immunoreactive
(ir) AVP neurons was decreased in the PVN by day 7. The percentage of heavily stained AVP neurons
to all ir-AVP ones was decreased in the SON by day 7. Changes in the number of ir-OXT neurons and
the percentage of heavily stained OXT neurons were insignificant. Significant hypertrophy was found in
both AVP and OXT neurons in the SON and the PVN after day 2. The present results thus indicate that
the AVP and OXT mRNAs are increased by sodium loading. It is probable that supraoptic AVP
neurons are more sensitive to changes in extracellular sodium concentration than paraventricular AVP
neurons, and that OXT neurons have some physiological role in the water and salt metabolism.

INTRODUCTION

Arginine vasopressin (AVP) and oxytocin
(OXT), mammalian neurohypophysial hormones,
are synthesized in the hypothalamic magnocellular
neurosecretory neurons, transported mainly to the
neurohypophysis, and are released into the circula¬
tion. Following a rise in plasma osmolality or
plasma Na+ concentration, the number of action
potentials in magnocellular AVP neurons and the
plasma level of AVP were increased [1]. OXT
release was also stimulated by sodium loading [2-
4]. OXT could stimulate electrolyte excretion [2],
It is possible that OXT is involved in the regulation

Accepted January 22, 1988
Received December 18, 1987

1 Present address: Ocean Research Institute, Universi¬
ty of Tokyo, Minamidai, Nakano-ku, Tokyo 164,

Japan.

of water and salt metabolism, although any phys¬
iological significances of OXT for water and salt
metabolism have not been clarified yet.

Chronic water deprivation and sodium loading
elevated amino acid incorporation into the
neurosecretory neurons [5, 6]. These stimuli fur¬
ther induced cytological changes in these neurons,
e.g., hypertrophy, an increase in nuclear and nu¬
cleolar diameters, dilation of endoplasmic reticu¬
lum, and an increase in the number of free ribo¬
somes [7]. These data strongly suggest that bio¬
syntheses of the neurohypophysial hormones are
elevated by water deprivation and sodium loading.
Nevertheless, the intensity of immunoreactivity,
especially that in AVP neurons, was decreased by
these stimuli [8]. An investigation of gene expres¬
sions is thus required for further understanding of
the biosynthetic activity of AVP and OXT

neurons.

1034

S. Hyodo, M. Fujiwara et al.

Since an in situ hybridization (ISH) method can
be an appropriate method for a gene expression
study, we have developed an ISH method using
synthetic oligonucleotide probes which can dis¬
criminate the AVP mRNA from the highly homo¬
logous OXT mRNA [9-11]. Several investigators
including us have now clarified using various hy¬
bridization methods that the AVP mRNA is in¬
creased by sodium loading and water deprivation
[10-18]. The OXT mRNA is also increased by
sodium loading and water deprivation [17, 19].
However, at present, little information has been
available concerning the time courses of changes in
AVP and OXT gene expressions, especially those
of alterations in cellular and immunocytochemical
aspects of neurosecretory neurons. Such informa¬
tion is the indispensable requisite for understand¬
ing how AVP and OXT neurons in the supraoptic
(SON) and paraventricular (PVN) nuclei are in¬
volved in osmoregulation.

In the present study, changes in the levels of
AVP and OXT mRNAs in magnocellular neurons
of the rat hypothalamus were examined after oral
intake of hypertonic saline for 2 to 7 days with the
ISH method using synthetic oligonucleotide
probes. Furthermore, adjacent tissue sections
were immunohistochemically stained, and staina-
bility and sizes of single AVP and OXT im-
munoreactive neurons were determined. Pre¬
liminary results appeared elsewhere [13, 17].

MATERIALS AND METHODS

Animals

Male Wistar-Imamichi rats (Imamichi Institute
for Animal Reproduction, 140-160 g) were
housed in individual cages with 14L: 10D light
schedule. Animals were allowed free access to tap
water and standard laboratory chow (Charles Riv¬
er) for at least 3 days prior to the start of the
experiment. They were then divided into 4 ex¬
perimental groups, each of them included 7 rats,
and were given orally a 2% sodium chloride solu¬
tion as drinking water for 0, 2, 4 and 7 days prior to
sacrifice. A preliminary study showed that effects
of sodium loading on AVP gene expression first
appeared on the second day of treatment [13].

Daily water intakes and urine volumes were mea¬
sured and urine samples were kept at — 20°C to
determine their osmolality and Na+ concentra¬
tion. The animals were killed by decapitation
between 10: 00 to 12: 00 to avoid possible circadian
fluctuations in AVP and OXT gene expressions
[20]. Their hypothalami and pituitaries were im¬
mediately removed and were immersed in a fixa¬
tive solution containing 2% paraformaldehyde,
1% glutaraldehyde and 1% picric acid in 0.05 M
phosphate buffer (pH 7.3). At the same time,
blood was collected and centrifuged. Plasma sam¬
ples were stored at — 20°C. Plasma and urine Na+
concentrations were measured later with an atomic
absorption spectrometer (Hitachi 180-50). Their
osmolality was determined by the freezing point
method.

In situ hybridization and quantitation of auto¬
radiographic signals

The individual tissues were paraffin-sectioned,
divided into several groups, and were processed
for in situ hybridization and immunohistochemis-
try. The precise procedures for tissue preparation
and ISH were described previously [9]. Two types
of 22mer synthetic deoxyoligonucleotide probes,
complementary to the loci of rat mRNAs encoding
AVP-neurophysin (NP) (1-8) and OXT-NP (1-8),
were used in this study. For semiquantitative
expression of hybridization signals, the numbers of
autoradiographic silver grains in 100 /mi X 100 pm
squares settled in each of the SON and the PVN
(Fig. 1) were counted. Then the numbers of grains
in the areas adjacent to the SON and the PVN,
that is, the background levels, were counted, and
were subtracted from the corresponding values in
the SON and the PVN. Thereafter, the single
cellular numbers of grains were calculated by
dividing the specific number of silver grains by the
numbers of immunoreactive neurons within the
100 /mix 100 pm squares.

Immunohisto chemistry

The sections adjacent to those used for grain
counting were immunohistochemically stained by
the avidin-biotin-peroxidase complex (ABC)
method, the detailed procedure of which was
described previously [21]. Specificity tests of im-

AVP and OXT Gene Expressions

1035

Fig. 1 . Schematic diagram showing the loci of 100 /mi x
100 /mi squares in which autoradiographic signals
were quantitated. The numbers of silver grains in
regions 3 (AVP region in the SON), 4 (OXT region
in the SON), 6 (AVP region in the PVN), and 7
(OXT region in the PVN) were counted, and then
the numbers of grains in the adjacent squares (2 and
5) were subtracted to obtain the specific numbers of
signals. The numbers of grains in the regions 2 and
5, which are fibrous neuropil areas, were statistically
in the same range with that in the anterior hypotha¬
lamic nucleus (AH, 1). OT, optic tract; SCN,
suprachiasmatic nucleus.

Fig. 2. AVP immunoreactive (ir) neurons and in situ hybridization of AVP mRNA in the SON of normal (a, b) and
7-day sodium loaded rats (c, d). Note the density of silver grains over the SON of the sodium loaded rat (d) is
higher than that in the control rat (b). On the contrary, stainability of ir-AVP neurons in the sodium loaded rat
(c) is reduced from that in the control rat (a). OT, optic tract. Scale bar, 50 /mi.

1036

S. Hyodo, M. Fujiwara et al.

munohistochemistry were also described previous¬
ly [9, 11]. The ABC method is considered to be
superior than the peroxidase-antiperoxidase
(PAP) method for a quantitative study [21]. In this
study, primary antisera were used as follows: rab¬
bit anti-AVP (UCB-Bioproducts) was diluted 1:
32,000 with phosphate buffered saline containing
0.5% BSA (PBS-BSA, pH 7.6) and rabbit anti-
OXT (a gift from Professor S. Kawashima,
Hiroshima University) was diluted 1:20,000 with

PBS-BSA. These values for dilution gave half-
maximal staining of AVP and OXT neurons in
normal non-treated rats, and are considered to be
theoretically appropriate for quantitative immuno-
histochemistry by an analogy of radioimmunoas¬
say. The values for dilution were determined by
serial dilution experiments. Submaximally stained
and faintly stained sections were first selected by
microscopic observation. By comparison with
them, the medially stained section was singled out

Fig. 3. OXT immunoreactive neurons and in situ hyridization of OXT mRNA in the PVN of normal (a, b) and 2-day
sodium loaded rats (c, d). Note the density of silver grains over the PVN of the sodium loaded rat (d) is higher
than that in the control rat (b). Scale bar, 50 /um.

AVP and OXT Gene Expressions

1037

as the half-maximally stained one.

Intensity of immunohistochemical stainability in
each magnocellular neuron was scored according
to the following criteria: not stained, weakly
stained and heavily stained, by consulting to the
arbitrarily selected standard sections. Im-
munoreactivity was shown by the number of im-
munoreactive (ir) neurons and the percentage of
heavily stained neurons in each of the SON and the
PVN. In addition, after depicting the outline of
individual ir-neurons with a camera lucida, their
cellular areas were determined with a tablet dig¬
itizer-microcomputer system.

RESULTS

Autoradiographic signals of the AVP-NP and
OXT-NP probes were densely localized over the

magnocellular neurons in the SON, the PVN and
several accessory magnocellular nuclei, as was
previously reported [9]. The distribution of AVP-
NP and OXT-NP probes coincided well with the
immunohistochemical localization of AVP and
OXT, respectively. The ventral region of the SON
and the dorsolateral region of the PVN were
predominantly composed of ir-AVP neurons. The
AVP-NP probe was localized in these regions (Fig.
2). While, the dorsal region of the SON and the
ventromedial region of the PVN were rich in
ir-OXT neurons. This distributional pattern coin¬
cided well with the localization of the OXT-NP
probe (Fig. 3).

AVP neurons

The density of autoradiographic signals repre¬
senting the localization of the AVP mRNA was

Days

Days

Fig. 4. Effects of sodium loading on the AVP and OXT mRNA levels in the SON and the PVN. The mRNA levels
are expressed relatively as percent changes to the level of day 0. The numbers of silver grains/ir-cell on day 0 are:
AVP mRNA in the SON, 122.2 + 13.2; AVP mRNA in the PVN, 177.6+13.8; OXT mRNA in the SON, 85.9 +
9.8; OXT mRNA in the PVN, 122.4+10.1. Each point represents the mean + S.E. The number of animals is
given in parentheses. *, p<0.05; **, p<0.01; by the t-test compared to day 0.

1038

S. Hyodo, M. Fujiwara et al.

markedly increased in the SON by the sodium
loading. The increase became statistically signif¬
icant after day 4 of the treatment (Figs. 2 and 4).
On the other hand, the increase in the hybridiza¬
tion signals in the PVN was much smaller than that
in the SON, and was not statistically significant
during the treatment (Fig. 4).

The number of ir-AVP cells was decreased in
the PVN on day 7, while the percentage of heavily
stained cells was significantly reduced in the SON
(Figs. 2 and 5). Significant hypertrophy of ir-AVP
neurons was observed after day 2 of the treatment
in both the SON and the PVN (Figs. 2 and 7). The
hypertrophy of supraoptic neurons was more con¬
spicuous than that of paraventricular neurons.

The density of silver grains in the suprachias-
matic nucleus which also includes ir-AVP was only
slightly higher than the background level, and was
not changed noticeably by the sodium loading.

OXT neurons

The density of hybridization signals for the OXT
mRNA was drastically increased and attained to
peak levels in both the SON and the PVN on day 2
of the sodium loading. The increase in the PVN

was statistically significant (Figs. 3 and 4). The
signal levels in the SON were maintained around
the peak level, while those in the PVN were
somewhat decreased after day 4 (Fig. 4).

The numbers of stained cells and the percent¬
ages of heavily stained cells were not significantly
changed in both the SON and the PVN (Figs. 3
and 6). The number of stained cells was slightly
increased on day 7. On the contrary, the percent¬
age of heavily stained cells was gradually de¬
creased, although it was increased in the PVN on
day 2. Significant hypertrophy of ir-OXT neurons
was observed after the second day of the sodium
loading (Figs. 3 and 7). The temporal pattern of
changes in magnitudes of hypertrophy that was
found in supraoptic and paraventricular OXT
neurons was similar to that observed in paraven¬
tricular AVP neurons (Fig. 7).

Neurohypophysis

Although immunoreactivity of the neurohy¬
pophysis was not analyzed quantitatively, it
appeared that AVP and OXT immunoreactivities
were markedly decreased after day 4 and day 7 of
the sodium loading, respectively, indicating a de-

Fig. 5. Effects of sodium loading on the number of stained AVP neurons (percent change) compared to day 0 (the
numbers per section are: SON, 47.0 + 2.5; PVN, 54.8 + 4.3), and the percentage of heavily stained AVP neurons
per total stained AVP neurons in the SON and the PVN. Each point represents the mean + S.E. (n=7). *, p<
0.05; **, p<0.01; by the t-test compared to day 0.

AVP and OXT Gene Expressions

1039

Fig. 6. Effects of sodium loading on the number of stained OXT neurons (percent change) compared to day 0 (the
numbers per section are: SON, 45.1 + 6.9; PVN, 72.8 + 3.9), and the percentage of heavily stained OXT neurons
per total stained OXT neurons in the SON and the PVN. Each point represents the mean + S.E. (n = 7).

Fig. 7. Effects of sodium loading on sizes of the AVP and OXT immunoreactive neurons in the SON and the PVN.
The values of cell areas (xl0-4mm2) on day 0 are: AVP neurons in the SON, 2.54 + 0.08; AVP neurons in the
PVN, 3.25 + 0.11; OXT neurons in the SON, 2.57 + 0.12; OXT neurons in the PVN, 2.34 + 0.09. Each point
represents the mean + S.E. (n=7). ***, p<0.001; by the t-test compared to day 0.

1040

S. Hyodo, M. Fujiwara et al.

Days

Fig. 8. Changes in oral intake (hatched bars) and urine
volume (open bars) by oral hypertonic saline. Each
point represents the mean±S.E. (n=7). ***, p<
0.001; by the t-test compared to day 0.

pletion of stored pools of neurohypophysial hor¬
mones.

Changes in plasma and urine by oral hypertonic
saline

The oral intake of 2% hypertonic saline was
progressively increased during the experimental

period, and attained to the maximum equilibrium
rate by day 6. The urine volumes were similarly
increased (Fig. 8). The plasma osmolality and the
plasma Na+ concentration were elevated to the
significant level within 2 days after the onset of oral
hypertonic saline, and were further increased
linearly at the same rate (Fig. 9). The urine Na+
concentration was also significantly increased after
day 2 of the treatment. On the contrary, the urine
osmolality was drastically decreased after day 2 of
the treatment (Fig. 9). These results indicate that
the animals under the oral hypertonic saline regi¬
men were exposed to an osmotic stimulus with
hypertonic plasma sodium.

DISCUSSION

The present study showed that the hybridization
signals for the AVP and OXT mRNAs were
increased in the SON and the PVN by oral hyper¬
tonic saline, indicating that the AVP and OXT
mRNA levels in magnocellular neurons were ele¬
vated by hypertonic sodium stimulation. This
result implies that the sodium loading stimulated

cn

E

ID

o 320

i 300

o

E

ID

° 280
ra

E

ID

TO

Q.

''
o L

E

(0

O

Days Days

Fig. 9. Changes in plasma osmolality, plasma [Na+], urine osmolality and urine [Na+] by oral hypertonic saline.
Each point represents the mean + S.E. (n=7). *, p<0.05; **, p<0.01; ***, p<0.001; by the t-test compared to
day 0.

AVP and OXT Gene Expressions

1041

transcription of the genes encoding neurohypophy¬
sial hormones in both the SON and the PVN. In
addition, the sodium loading induced significant
hypertrophy of both AVP and OXT neurons. This
fact indicates that translation rates of AVP and
OXT precursors were also increased by the sodium
loading, since chronic dehydration elicited several
changes which indicate stimulation of protein
syntheses in magnocellular neurosecretory
neurons, that is, an increase in amino acid incor¬
poration [5, 6], increases in nuclear and nucleolar
diameters and dilation of endoplasmic reticulum
[7]. Nevertheless, stainability of ir-AVP neurons
was reduced in both the SON and the PVN,
probably because of a rapid transport of newly
synthesized AVP to the neurohypophysis, as was
discussed in the previous paper [11].

Although the increase in the AVP mRNA level
was observed in both the SON and the PVN, it was
more rapid and marked in supraoptic AVP
neurons. This result agrees with the report of
Sherman et al. [15]. In the present study, the
hypertrophy of AVP neurons was also more
marked in the SON than in the PVN. Similar
differences in response to sodium loading have
been observed between the SON and the PVN by
several investigators. A significant increase in
incorporation of cytidine into RNA [22] and that
of tyrosine into peptides [6] in response to oral
hypertonic saline were demonstrated in the SON,
but not in the PVN. Activation of the electrical
activity of paraventricular AVP neurons in re¬
sponse to an increase in plasma osmolality was
much less than that of supraoptic AVP neurons
[3]. Furthermore, the possibility that supraoptic
neurons are directly osmosensitive and play a part
of an osmoreceptive complex has been indicated
[23, 24]. These results suggest that AVP neurons
in the SON are more responsible for osmotic
regulation than paraventricular AVP neurons.

The OXT mRNA was increased by the sodium
loading. Marked hypertrophy was also observed in
ir-OXT neurons within 2 days after the onset of the
treatment. These results may imply that OXT
neurons have some physiological role in the water
and salt metabolism. This possibility is supported
by the facts that OXT release was stimulated by
sodium loading, and that OXT has been implicated

in the control of renal functions [2], In addition,
the increase in the OXT mRNA was more rapid
than that in the AVP mRNA. Since OXT neurons
released more products than AVP neurons in
response to an acute increase in plasma osmolality
[4], OXT neurons may be more sensitive to an
acute increase in plasma osmolality than AVP
neurons.

A rise in plasma osmolality immediately in¬
creased the firing rates of AVP and OXT neurons
in both the SON and the PVN, and then plasma
AVP and OXT levels were rapidly increased [3].
On the contrary, the present study showed that
statistically significant increases in the AVP and
OXT mRNA levels were detectable after 4- and
2-day sodium loading, respectively. The reason for
this difference in the onsets of secretory activity
and synthetic activity is not clear at present. It may
reflect the presence of abundant storages of these
hormones in the neurohypophysis and/or the
hypothalamic nuclei. Further investigations are
required to clarify signals by which the onset of
gene expression is stimulated.

ACKNOWLEDGMENTS

The authors would like to thank Drs. T. Hirano and T.
Ogasawara and Ms. S. Hasegawa, Ocean Research Insti¬
tute, University of Tokyo, for the use of an atomic
absorption spectrometer and an osmometer, and Prof. S.
Kawashima, Hiroshima University, for providing the
antiserum to oxytocin.

REFERENCES

1 Poulain, D. A. and Wakerley, J. B. (1982) Elec¬
trophysiology of hypothalamic magnocellular
neurons secreting oxytocin and vasopressin.
Neuroscience, 7: 773-808.

2 Balment, R. J., Brimble, M. J. and Forsling, M. L.
(1980) Release of oxytocin induced by salt loading
and its influence on renal excretion in the male rat.
J. Physiol., 308: 439-449.

3 Brimble, M. J. , Dyball, R. E. J. and Forsling, M. L.
(1978) Oxytocin release following osmotic activa¬
tion of oxytocin neurons in the paraventricular and
supraoptic nuclei. J. Physiol., 278: 69-78.

4 Cheng, S. W. T. and North, W. G. (1986) Respon¬
siveness of oxytocin-producing neurons to acute salt
loading in rats: comparisons with vasopressin-
producing neurons. Neuroendocrinology, 42: 174-
180.

1042

S. Hyodo, M. Fujiwara et al.

5 Lepetit, P., Lestage,P., Gauquelin, G., Vitte, P.
A., Debilly, G., Gharib, C., Jouvet, M. and Bobil-
lier, P. (1985) Differential effects of chronic de¬
hydration on protein synthesis in neurons of the rat
hypothalamus. Neurosci. Lett., 62: 13-18.

6 Murray, M. (1967) Effects of dehydration on incor¬
poration of 3H-tyrosine by some hypothalamic
neurons in the rat. Exp. Neurol., 19: 212-231.

7 Rechardt, L. (1969) Electron microscopic and his-
tochemical observations on the supraoptic nucleus
of normal and dehydrated rats. Acta Physiol.
Scand., Suppl., 329: 1-79.

8 Negro-Vilar, A. and Samson, W. K. (1979) De-
hydration-induced changes in immunoreactive
vasopressin levels in specific hypothalamic struc¬
tures. Brain Res., 169: 585-589.

9 Hyodo, S., Fujiwara, M., Kozono, S., Sato, M. and
Urano, A. (1988) Development of an in situ hybri¬
dization method for neurohypophysial hormone
mRNAs using synthetic oligonucleotide probes.
Zool. Sci., 5: 397-406.

10 Nojiri,H., Sato, M. and Urano, A. (1985) In situ
hybridization of the vasopressin mRNA in the rat
hypothalamus by use of a synthetic oligonucleotide
probe. Neurosci. Lett., 58: 101-105.

11 Nojiri, H., Sato, M. and Urano, A. (1986) Increase
in the vasopressin mRNA levels in the magnocellu-
lar neurosecretory neurons of water-deprived rats:
in situ hybridization study with the use of synthetic
oligonucleotide probe. Zool. Sci., 3: 345-350.

12 Burbach, J. P. H., De Hoop, M.J., Schmale, H.,
Richter, D., De Kloet, E. R., Ten Haaf, J. A. and
De Wied, D. (1984) Differential responses to os¬
motic stress of vasopressin-neurophysin mRNA in
hypothalamic nuclei. Neuroendocrinology, 39: 582-
584.

13 Fujiwara, M., Hyodo, S., Sato, M. and Urano, A.
(1985) Changes in vasopressin and oxytocin mRNA
levels in the rat hypothalamus by oral hypertonic
saline. Zool. Sci., 2: 990.

14 McCabe, J. T., Morrell, J. I., Richter, D. and Pfaff,
D. W. (1986) Localization of neuroendocrinologi-
cally relevant RNA in brain by in situ hybridization.
In “Frontiers in Neuroendocrinology”. Ed. by W. F.
Ganong, and L. Martini, Raven Press, New York,
Vol. 9, pp. 149-167.

15 Sherman, T. G., McKelvy, J. F. and Watson, S. J.
(1986) Vasopressin mRNA regulation in individual
hypothalamic nuclei: a northern and in situ hy¬
bridization analysis. J. Neurosci., 6: 1685-1694.

16 Uhl, G. R., Zingg, H. H. and Habener, J. F. (1985)
Vasopressin mRNA in situ hybridization: localiza¬
tion and regulation studied with oligonucleotide
cDNA probes in normal and Brattleboro rat
hypothalamus. Proc. Natl. Acad. Sci. USA, 82:
5555-5559.

17 Urano, A., Hyodo, S. and Sato, M. (1987) In situ
hybridization study of neurohypophysial hormone
mRNAs. In “Proceedings of the 10th International
Symposium of Neurosecretion”. Plenum, New
York/London, in press.

18 Zingg, H. H., Lefebvre, D. and Almazan, G. (1986)
Regulation of vasopressin gene expression in rat
hypothalamic neurons. J. Biol. Chem., 261: 12956—
12959.

19 Van Tol, H. H. M., Voorhuis, D. A. M. and Bur¬
bach, J. P. H. (1987) Oxytocin gene expression in
discrete hypothalamic magnocellular cell groups is
stimulated by prolonged salt loading. Endocrinolo¬
gy, 120: 71-76.

20 Uhl, G. R. and Reppert, S. M. (1986) Supra-
chiasmatic nucleus vasopressin messenger RNA:
circadian variation in normal and Brattleboro rats.
Science, 232: 390-393.

21 Jokura, Y. and Urano, A. (1985) Projections of
luteinizing hormone-releasing hormone and vasoto¬
cin fibers to the anterior part of the preoptic nucleus
in the toad, Bufo japonicus. Gen. Comp. Endocri¬
nol., 60: 390-397.

22 George, J. M. (1973) Localization in hypothalamus
of increased incorporation of 3H cytidine into RNA
in response to oral hypertonic saline. Endocrinolo¬
gy, 92: 1550-1555.

23 Gross, P. M., Sposito, N. M., Pettersen, S. E. and
Fenstermacher, J. D. (1986) Differences in function
and structure of the capillary endothelium in the
supraoptic nucleus and pituitary neural lobe of rats.
Neuroendocrinology, 44: 401-407.

24 Mason, W. T. (1980) Supraoptic neurons of rat
hypothalamus are osmosensitive. Nature, 287: 154—
156.

ZOOLOGICAL SCIENCE 5: 1043-1049 (1988)

© 1988 Zoological Society of Japan

Goitrogenic Action of Manganese on Female Mouse
Thyroid through Three Generations

Mikio Nishida, Jun Kawada, Hiroshi Ishizuka1
and Shigeru Katsura1

Faculty of Pharmaceutical Sciences and 1 School of Dentistry,

The University of Tokushima, Tokushima City,

Tokushima 770, Japan

ABSTRACT — The effect of manganese on mouse thyroids was examined through three generations.
The continuous supply of drinking water containing 200 mg/1 MnCl2-4H20 induced mild goiter only in
female mice.

The manganese treatment did not significantly affect the serum thyroxine (T4) levels in dams or the
plasma levels in neonates, suggesting that the effect was not severe. The syntheses of T4 and
triiodothyronine (T3) were examined in hydrolyzed thyroids. The results showed that ratios of
radioiodide activity in hormonal fractions of the manganese-given groups to those of control groups
were greater than 1 for all three generations of female mice. These results indicated that newly
synthesized hormones were retained in thyroids of the manganese-treated female mice. On the other
hand, in males, the ratio was reduced to less than 1 as the numbers of generations increased.

“Dwarf” neonates born in the third generation of the manganese-treated group showed an ataxial
motion in their gait of walk. However, histological examination of cerebral and cerebellar regions of the
dwarfs revealed no severe changes.

INTRODUCTION

It has been shown that excess manganese in¬
duces an endemic goiter when iodide intake is low
[1, 2]. In laboratory experiments, acute and sub¬
acute administrations of manganese suppressed
thyroidal iodide uptake and affected iodide meta¬
bolism [3, 4]. The authors also demonstrated that
the radioactive manganese accumulated to a high
level in mouse thyroid as well as in other endocrine
organs, such as the pituitary, pancreas and adrenal
[5]. Furthermore, it was found that under ordinary
iodide intake, a 7 week administration of man¬
ganese in the drinking water resulted in thyroid
enlargement in female mice, but not in male mice
[5],

For goiter formation, a prolonged manganese
treatment through an oral route was necessary
because a single [3, 6] or a sustained [4] parenteral
injection of manganese caused very severe reduc-

Accepted February 10, 1988
Received August 3, 1987

tions of iodide uptake and synthesis of thyroid
hormones, but did not induce thyroid enlarge¬
ment. Although prolonged treatment increases
thyroid hormone levels in the gland and reduced
the hormones in the blood [5], the mechanism of
goiter formation by manganese is not well under¬
stood. Thus, we were interested in determining
whether oral administration of manganese for
several generations caused larger goiters, and also
whether hereditary factors were involved in the
manganese-induced goiter. In this study, we ex¬
amined the chronic effects of manganese on
growth, thyroid function and brains of mice in
three consecutive generations.

MATERIALS AND METHODS

Chemicals The chemical reagents used in
this study were purchased from Wako Pure Che¬
mical Industries (Osaka, Japan) and oxytocin was
from Sigma Chemical Co. (St. Louis, MO., USA).
[125I]Na was purchased from New England Nu-

1044

M. Nishida, J. Kawada et al.

clear Co. (Boston, MA., USA). The radioimmu¬
noassay kit, SPAK T4 was purchased from Daiichi
Isotope Institute (Tokyo, Japan). Pronase was a
product of Kaken Pharmaceutical Co. (Tokyo,
Japan), and filter paper No. 51 for chromatogra¬
phy was obtained from Toyo Roshi Co. (Tokyo,
Japan).

Animals Male and female ddY mice
weighing 18-20 g (4 weeks old) were obtained
from a local supplier and maintained on a 12 hr
light-dark cycle in an air conditioned room at 23-
24°C and 50% moisture. Animals were allowed
free access to solid food (Type MF, Oriental Yeast
Co. Tokyo, Japan) and tap water for the control.
The iodide and manganese contents in the food
pellets were 1.04 and 64.5 //g/g, respectively, as
reported previously [5]. Mice were given 200 mg/1
of MnCl2-4H20 in the drinking water for 7 weeks.
These animals were designated as the first genera¬
tion of manganese-treated mice (parents). Some
mice were sacrificed at 11 weeks of age to examine
thyroid enlargement. The remaining mice were
mated to provide the second generation of man¬
ganese-treated mice. During mating and subse¬
quent pregnancy, manganese administration was
continued. The offsprings from these parents with
manganese treatment were designated as the
second generation. The third generation of man¬
ganese-treated mice was obtained by a brother-
sister mating of the second generation with man¬
ganese treatment.

On day 1 when neonates were born, they were
anesthetized with ether and blood was collected
into a heparinized test tube by cutting the carotid
artery. The blood from a litter was pooled and
centrifuged to obtain plasma for hormone analysis.
Another group of neonates was sacrificed on day 5
to collect plasma. Dams were also sacrificed on
day 1 and day 5 after delivery. Blood samples were
taken from the dams. The thyroids were excised
and their weights were recorded.

Intrathyroidal iodine metabolism Mice

treated with manganese and the control were
intraperitoneally given 3-4 /j.C\ [123I]Na per animal
24 hr prior to sacrifice. At autopsy, the thyroids
were excised, weighed and pooled in each group
because an individual thyroid was insufficient for
analysis. The thyroids were added to a test tube

containing 50 mM phosphate buffer (pH 7.4), 1
mM methylmercaptoimidazole and two drops of
ethanol. The mixture was subjected to proteolytic
hydrolysis with Pronase (20: 1 weight ratio). For
this purpose, the test tube was bubbled with N2
gas, sealed with a stopper and incubated at 37°C
for 24 hr. Digestion was terminated by immersing
the tube in boiling water. Samples of the hydroly¬
sates were spotted on paper and subjected to paper
chromatography using two different solvent sys¬
tems (n-butanol : acetic acid : H20=4:l:2, v/v
and n-butanol : ethanol : 2N ammonium hydroxide
=5:1:2, v/v). Radioiodinated compounds were
analyzed by autoradiography. Areas on the paper
showing dark spots were cut and their radioactivi¬
ties were measured by an auto gamma scintillation
counter.

Hormone measurement Serum was pre¬
pared from adult animals, but pooled plasma was
prepared from neonates because of an insufficient
volume of blood from an individual neonate. Cir¬
culating T4 was assayed by SPAK T4 RIA kit. In
this assay system, 25 ju\ of specimen were used in a
single measurement and at least duplicate
measurements were carried out for each specimen.
A cross reaction with T3 was not detectable with
this kit.

Histological study Thyroids were excised
and fixed in Bouin’s fluid. Using ordinary process¬
es, samples were embedded in a paraffin wax,
sliced and sections were stained in hematoxylin-
periodate. Whole brain regions were fixed in 10%
formaldehyde solution and embedded in an epon.
The specimens were stained in luxol fast blue-
cresylechtviolet solution for simultanous staining
of Nissl bodies and the myelin sheath, and in
hematoxylin-eosin solution for ordinary staining.
Stainability and histological changes were ex¬
amined under a light microscope.

RESULTS

Thyroid enlargement by manganese

A morphological study of thyroids of female
mice treated with 200 mg/1 manganese chloride for
7 weeks revealed moderate size goiters with colloid
filled lumens and flattened epithelial cells. Table 1

Manganese as Goitrogen in Mice

Table 1. Manganese effect on body and thyroid weights in three generations of mice

1045

1st generation

2nd generation

3rd generation

C

Mn

C

Mn

C

Mn

Female

No. of
animal

14

12

11

10

9

9

Body (
weight V6/

31.0 + 3.9

32.5 + 2.5

32.0 + 3.0

32.3 + 2.4

32.0 + 3.9

31.2 + 2.5

weghTW)

2. 7 + 0. 3

3. 3 + 0.3

3. 1 + 0.3

4. 3 + 0. 4

2. 8 + 0.5

3. 7 + 0. 4

p<0.001

p<0.001

pCO.001

Male

No. of
animal

9

11

16

11

11

11

Body , ,
weight V6'

40.6 + 3.4

40.9 + 3.0

42.2 + 4.1

43.2 + 3.8

41.5 + 3.4

44.3 + 3.4

3. 3 + 0.4

3. 8 + 1.0

3. 5 + 0. 4

3. 8 + 0. 3

3. 6 + 0.3

3. 6 + 0. 3

NS

NS

NS

C: Control, Mn: Mn treated, NS: Not significant.

shows that, in the first generation, thyroids of
female mice were enlarged by manganese chloride,
but thyroids of male mice were unchanged by the
same treatment. When the mice with manganese
treatment were mated, 9 out of 15 females became
pregnant (60%), whereas in the control group 8
out of 12 became pregnant (67%). In the second
generation, the frequencies of pregnancy of the
manganese-treated female mice and the control
were 7/14 and 10/15, respectively. The body
weight gain and other external appearances were
not changed by manganese treatment. At delivery
time, the average number of neonates bred and
their mean body weight were also similar to those
of control group. When animals of the second
generation were 11-12 weeks old, some were sac¬
rificed to examine their thyroids. The thyroid
weight of the females of the second generation was
slightly greater than that of the first generation. In
the third generation, the size of glands was similar
to that of the first generation but smaller than that
of the second generation.

Manganese effect on circulating T4 level

The results of the circulating T4 levels of non¬
pregnant adult female mice, dams after delivery
(day 1 and day 5) and neonates (day 1 and day 5)

are summarized in Table 2. Just after delivery,
serum T4 levels of dams were within normal ranges
and not influenced by the birthing process. Dams
with manganese treatment maintained normal T4
levels at day 1. However, the T4 levels of dams in
both groups fell to subnormal levels during the
term of lactation (day 5). T4 levels of neonates on
day 1 were very low but increased slightly during
the first five days. This increase in T4 level was also
observed in the manganese-treated group. Thus,
there was no significant effect of manganese on
circulating T4 levels.

Manganese effect on radioiodide uptake into hor¬
monal fraction of thyroid

In order to determine whether manganese inter¬
feres with hormone synthesis, the radioiodide up¬
take into intrathyroidal hormonal fractions was
examined in control and manganese-treated mice
through three generations. Table 3 shows the
radioiodide distribution in T3 + T4 fractions of each
group. 125I activity in the females treated with
manganese was much higher (10-20%) than that
of the control group, resulting in the ratios of
Mn/C being greater than one or at least equal to
one. On the other hand, in males, the ratios of
Mn/C were always less than one. Furthermore, as

1046 M. Nishida, J. Kawada et al.

Table 2.

Manganese effect on

thyroxine level in blood

Group

1st generation

2nd generation

3rd generation

ng/ml

ng/ml

ng/ml

Non-pregnant

Control

38.5 + 5.0
(n=4)

38.9 + 7.9
(n=4)

adult female

Mn-

33.5 + 2.5

40.8 + 8.5

treated

(n=4)

(n=5)

Control

38.0 + 4.7

28.1 + 1.9

(n=6)

(n=4)

Day 1

Mn-

37.5 + 14.2

30.8 + 7.6

treated

(n = 5)

(n=6)

Dam

Control

17.6 + 4.2

16.8 + 1.0

Day 5

(n— 6)

(n=4)

Mn-

21.8 + 8.7

16.5 + 1.6

treated

(n=6)

(n = 6)

Control

6. 4 + 6. 3

9. 4 + 3. 9

(n = 4)

(n=4)

Day 1

Mn-

5. 1 + 2. 9

14.1 + 5.9

treated

(n=5)

(n=6)

Neonate

Control

10.3 + 0.6

14.9 + 1.9

(n=3)

(n=4)

Day 5

Mn-

18.1 + 0.1

17.9 + 1.6

treated

(n=2)

(n=6)

Data show serum hormone levels for adults and plasma hormone levels for neonates.

Each value is the mean + SD.

The numbers within the parentheses for the adult females and dams represent the numbers of mice used in
the study. However, the numbers within the parentheses for neonates represent the numbers of litters in
each group.

the number of generations increased, the ratios
became smaller in male mice.

“ Dwarfs ” and the histological examination of their
brains

Continuous manganese administration did not
affect the growth of parents and the second gen¬
eration of mice when their body weights were
monitored. However, in the third generation of
the manganese-treated group, some smaller neo¬
nates were observed, although the incidence was
not high: 0/11 for the control and 3/17 for the
manganese-treated group (The numerator and de¬
nominator represents numbers of dams which bred

a “dwarf” and total numbers of dams examined,
respectively. In this experiment, the term “dwarf”
is defined for neonates weighing less than 70% of
the average body weight of siblings in the same
litter on day 1). After a half number of normal
siblings were isolated from their mother, dwarfs
were still unable to grow at the rate of normal
mice, suggesting that the dwarfs were not the
results of interrupted lactation but of some other
endogenous dysfunctions.

Figures 1 and 2 illustrate the growth curves and
the photograph of normal and dwarf mice, respec¬
tively. Dwarfs grew slowly and never reached the
level of body weight of normal animals. Since

Manganese as Goitrogen in Mice

Table 3. 125I incorporation into iodothyronine fraction in thyroid

1047

1st generation

2nd generation

3rd generation

c

Mn

Mn/C

C

Mn

Mn/C

C

Mn

Mn/C

% %

% %

% %

Female

8.92

11.05

1.24

6.18

7.05

1.14

5.85

7.09

1.21

(11.56)

(10.73)

(0.93)

(10.60)

(10.10)

(0.95)

(6.09)

(7.68)

(1.26)

Male

5.18

3.82

0.74

9.08

6.12

0.67

7.45

3.42

0.46

(5.88)

(4.95)

(0.84)

(11.50)

(9.57)

(0.83)

(9.87)

(5.32)

(0.54)

C: Control, Mn: Mn treated.

The values represent the percentages of 125I activity in T3 + T4 fractions to total radioactivity on
chromatograms. Values were obtained by using developing solvent system of n-butanol:ethanol:2N
ammonium hydroxide = 5: 1: 2 (v/v) and the values in parentheses were obtained by using a system of
n-butanol : acetic acid : H20=4:l:2 (v/v).

some dwarfs showed ataxial motion, the brains of
dwarfs were subjected to histological examination.
Although the photographs are not shown, both
control and dwarf neonates displayed regular
lamination of nerve cells in the motor area of the
cerebral cortex, and the dwarfs exhibited nerve
cells which were stained lighter than the control.
There was no critical difference in lamination of
nerve cells of cerebellar cortex between the two
groups. In the dwarfs, however, many swollen and
lightly stained Purkinje cells in the cerebellar
cortex were seen as well as an increased number of
nerve fibers in the cerebellar medulla, there was no
difference in stainability of the Nissl bodies of the
nuclei cerebri in normal and dwarf mice.

Fig. 1. Growth curve of normal and dwarf neonates.
The normal group consisted of 13 male and 9 female
mice from 2 litters and for the dwarfs, one male and
one female. The vertical bar represents the standard
deviation.

Fig. 2. A photograph of normal and dwarf neonates.

The photograph shows a dwarf (upper) and a normal (lower) neonate on day 9 after
birth.

1048

M. Nishida, J. Kawada et al.

DISCUSSION

Although symptoms caused by manganese de¬
ficiency [7] and neurological dysfunctions by excess
manganese [8] have been well documented, know¬
ledge of neonatal endocrinology related to excess
manganese ingestion is scant. Previously, we
stated that even with sufficient iodide supply,
excess manganese can be goitrogenic in the female
mouse thyroid, but not in the male mouse thyroid
[5]. The present study provided additional con¬
firming evidence to support the observation men¬
tioned above. Moreover, goiters were always
produced in three generations by continuous admin¬
istration of manganese. However, the size of
goiters did not increase from one generation to the
next as shown in Table 1. Initially, we assumed
that the condition of goiter would become more
severe with each succeeding generation if genetic
factors affected goiter formation by manganese.
Judging from the result in Table 1, there was no
such hereditary influence in the manganese effect.

Histological observations revealed that the goi¬
ter caused by manganese consisted of flattened
epithelial cells with ample colloid in the lumen.
This feature of colloid filled goiter suggests that the
thyroid must be in a hypofunctional state, perhaps
due to the second phase of Marine Cycle [9].
Therefore, it is likely that the use of colloid by
epithelial cells was blocked by manganese in
female mice. This speculation is supported by the
fact that radioiodide activity in the intrathyroidal
T3 + T4 fraction was slightly but steadily higher in
female mice with manganese treatment than in
control mice (Table 3). This result suggests that in
females, the de novo synthesized hormones remain
in the lumen, and in males, the hormones are
promptly released from the lumen.

From the present study, it is not known whether
TSH levels were altered by manganese. We
attempted to use a rat-TSH antibody to measure
mouse TSH level by radioimmunoassay, but the
rat-TSH antibody did not react with the mouse
TSH. Buthieau and Autissier reported that the
serum TSH level was reduced in manganese in¬
jected rats [4], If one could extend this evidence to
mice, excess manganese may depress the thyroid
and the pituitary functions. Na+, K+-ATPase,

which suplies energy to epithelia for colloid en-
docytosis, was not significantly inhibited by man¬
ganese in vitro (authors, unpublished). This evi¬
dence was compatible with the fact that the pro¬
longed oral administration of manganese only
slightly affected the ratio of 125I activity in tissue to
that in serum [5]. Therefore, this feature was
entirely different from the acute parenteral treat¬
ment [3]. At present, the cause of the sex differ¬
ence in the manganese effect on colloid accumula¬
tion is not clear, but in castrated male mice, goiter
was induced by manganese, suggesting levels of
male and female hormones with manganese may
somehow bring about the formation of goiter [5].

Serum T4 levels in non-pregnant females, dams
and the plasma of neonates were not significantly
affected by manganese administration. Perhaps
the manganese effect was mild so that regulatory
processes to maintain homeostasis of the hormone
level could operate properly in all ages of animals
examined. Since the presence of manganese trans¬
ferrin has been demonstrated in sera from several
species [10-12], a similar carrier protein may exist
in the mouse to reduce the toxic effects of an
overload of manganese.

In the third generation of manganese treatment,
dwarfs were born, but there is no information
suggesting that this was specifically due to man¬
ganese treatment. When the manganese content in
milk of the treated dams was measured, it was not
significantly different from the control milk (data
not shown). The course of dwarfism may occur
during the period of pregnancy. Dwarfs could
grow, but showed ataxial motion. We do not know
whether the ataxia in dwarfs mimics Parkinsonism
in humans caused by manganese intoxication [8].
When organ distribution of manganese was ex¬
amined, the level of manganese content in the
entire brain was not high compared with other
organs [5, 16]. However, high amounts of endoge¬
nous manganese was found in the hypothalamus
region of rats [13], indicating that manganese
could be localized in narrow regions to cause
aberrations in the nervous system.

Donaldson et al. [14] have shown that man¬
ganese can exist in vivo as either an oxidant or an
antioxidant, depending on its valency state; Mn2+
may reduce norepinephrine levels in brain regions

Manganese as Goitrogen in Mice

1049

whereas Mn3+ enhances lipid peroxidation and
formation of free radicals. Our recent study on
partition of Mn2+ and total manganese in rat
organs showed that only a very small fraction of
the total manganese remained in the form of Mn2+
in most organs after manganese chloride adminis¬
tration [16]. These data suggest that transvalency
of manganese by biological systems is very active.
Histological study of the brain region of dwarfs
exposed to manganese in early developmental
stages showed no drastic alterations. It is likely
that the effect of excess manganese on the perinat¬
al nerve systems may not be drastic but mild and
latent in nature.

ACKNOWLEDGMENT

The research was supported in part by a Grant-in-aid
for Scientific Research from the Ministry of Welfare of
Japan.

REFERENCES

1 Manescu, S., Negoescu, I., Lupulescu, A. and Di-
aconescu, M.-L. (1960) Investigations on the goi¬
trogenic action of manganese. Rumanian Med.
Rev., 5: 88-90.

2 Savina, P. N. (1969) Importance of manganese in
the formation of endemic goiter. Tr. Tomsk.
Nauch.-Issled. Inst. Vaktsin Syvorotok Tomsk.
Med. Inst., 20: 183-185.

3 Buthieau, A.-M. and Autissier, N. (1977) Action
des ions Mn2+ sur le metabolisme iode thyroidien
du Rat. Com. ren. sean. Soci. Biol., 171: 1024-
1028.

4 Buthieau, A.-M. and Autissieu, N. (1983) Effects
of manganese ions on thyroid function in rat. Arch.
Toxicol., 54: 243-246.

5 Kawada,J., Nishida,M., Yoshimura, Y. and
Yamashita, K. (1985) Manganese ion as a goitrogen

in the female mouse. Endocrinol. Japon., 32: 635-
643.

Kaellis, E. (1970) Effect of manganous ions on
thyroidal iodine metabolism in the rat. Proc. Soc.
Exp. Biol. Med., 135: 216-218.

Hurley, L. S. (1981) Teratogenic aspects of man¬
ganese, zinc and copper nutrition. Physiol. Rev., 61:
249-295.

Mena, I. (1981) Manganese. In “Disorders of
Mineral Metabolism”, vol. 1. Ed. by F. Bronner and
J. W. Coburn, Academic Press, pp. 233-270.
DeGroot, L. J., Larsen, P. R., Refetoff, S. and
Stanbury, J. B. (1984) The Thyroid and its Dis¬
eases. 5th ed. John Willey and Sons, New York, pp.
748.

Foradori, A. C., Bertinchamps, A., Gulibon, J. M.
and Cotzias, G. C. (1967) The discrimination be¬
tween magnesium and manganese by serum pro¬
teins. J. Gen. Physiol., 50: 2255-2266.

Hancock, R. G. V., Evans, D. J. R. and Fritze, K.
(1973) Manganese proteins in blood plasma.
Biochim. Biophys. Acta, 320: 486-493.

Gibbons, R. A., Dixon, S. N., Hallis, K., Russel,
A. M., Sansom, B. F. and Symonds, H. W. (1976)
Manganese metabolism in cows and goats. Biochim.
Biophys. Acta, 444: 1-10.

Donaldson, J. (1973) Determination of Na+, K+,
Mg2+Cu2+, Zn2+ and Mn2+ in rat brain regions.
Can. J. Biochem., 51: 87-92.

Donaldson, J., McGregor, D. and LaBella, F.
(1982) Manganese neurotoxicity: a model for free
radical mediated neurodegeneration? Can. J. Phys¬
iol. Pharmacol., 60: 1398-1405.

Maynard, L. S. and Cotzias, G. C. (1955) The parti¬
tion of manganese among organs and intracellular
organelles of the rat. J. Biol. Chem., 214: 489-495.
Sakurai, H., Nishida, M., Yoshimura, T., Takada,
J. and Koyama, M. (1985) Partition of divalent and
total manganese in organs and subcellular organelles
of MnCl2-treated rats studied by ESR and neutron
activation analysis. Biochim. Biophys. Acta, 841:
208-214.

6

7

8

9

10

11

12

13

14

15

16

.

E§9

■

ZOOLOGICAL SCIENCE 5: 1051-1057 (1988)

© 1988 Zoological Society of Japan

Neuroendocrine Regulation of the Development of Seasonal Morphs
in the Asian Comma Butterfly, Polygonia c-aureum L.: Is the
Factor Producing Summer Morphs (SMPH) Identical to the
Small Prothoracicotropic Hormone (4K-PTTH)?

Tadakatsu Masaki, Katsuhiko Endo* 1 and Kanji Kumagai2

Environmental Biology Laboratory, Biological Institute, Faculty of Science,
Yamaguchi University, Yamaguchi 753, and 2 Biological Institute,

Faculty of Liberal Arts, Yamaguchi University,

Yamaguchi 753, Japan

ABSTRACT — In the butterfly, Polygonia c-aureum L., a physiological mechanism underlying the
photoperiodic control of seasonal-morph development involves a summer-morph-producing factor
(SMPH). Amounts of SMPH present in the brain-extracts of Polygonia pupae as well as in those of the
silkmoth, Bombyx mori, were evaluated by Polygonia pupal assay for SMPH.

The factors responsible for SMPH-activity in the Polygonia and Bombyx brain-extracts were
heat-stable (95°C for 5 min), but they were thought to be peptide hormones since the brain-extracts
became SMPH-inactive by hydrolyzing with trypsin (24 hr at 25°C). According to the Sephadex G-50
gel-filtration pattern of SMPH-activity, the ultimate molecular weight of SMPH was though to be 4500,
which is almost the same size as the small prothoracicotropic hormone purified from the silkmoth, B.
mori (4K-PTTH or bombixin: M. W. 4400). This finding is also supported by evidence in our present
study that 2 (out of 4) Sephadex G-50 fractions judged as being SMPH-active showed 4K-PTTH-activity
by Papilio pupal assay. However, the factor showing SMPH-activity was not thought to be identical
with the one showing 4K-PTTH-activity in P. c-aureum since they were separated by reversed-phase
HPLC.

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 exposure during the larval stages [1,
2]. A physiological system underlying the photo-
periodic control of seasonal-morph development
involves a neuroendocrine system of brain-corpus
cardiacum-corpus allatum, which secretes a sum-
mer-morph producing factor (SMPH) in the early
pupal stage [3, 4]. The factor showing SMPH-
activity is present in the brains of P. c-aureum as
well as in the brains of several other lepidopteran
insects, Bombyx mori , Papilio xuthus and Lycaena

Accepted January 11, 1988
Received October 24, 1987

1 To whom reprints should be requested.

phlaeas daimio [5]. The SMPHs could be extracted
and precipitated in almost the same manner as has
been demonstrated in the small and big prothoraci¬
cotropic hormones (4K- and 22K-PTTHs) of B.
mori [5]; the PTTHs were extracted with 2%
NaCl, but not with acetone and 80% ethanol, and
were precipitated by ammonium sulfate at 80%
saturation [6].

The present study was designed to determine
whether the factors showing SMPH-activity are
identical to the 4K- and/or 22K-PTTHs in lepidop¬
teran insects, P. c-aureum and B. mori.

MATERIALS AND METHODS

Animals Larvae of P. c-aureum and P.
xuthus were held in either transparent plastic con¬
tainers of 0 9x5 cm3 or 19x13x5 cm3 and ex¬
posed to short-day (SD) conditions alternating
periods of 8-hr of light and 16-hr of dark. The

1052

T. Masaki, K. Endo et al.

rearing containers were placed in a cabinet at
constant temperature (20°C or 25°C) and were
illuminated by two 20 W white fluorescent tubes
which were controlled by a 24-hr time-switch.
During the light period, light intensity was pro¬
vided at about 500 lux.

Larvae of Polygonia and Papilio were fed on
leaves of Humulus japonicus and Fagara ailan-
thoides, whereas those of the silkmoth, B. mori ,
were fed on leaves of Morus tiliaefolia. Fresh
leaves were provided daily.

Under the short-day condition, Polygonia larvae
and pupae all developed into butterflies of autumn
morphs, whereas those of Papilio entered diapause
in the pupal stage.

Extraction ofSMPH Brains were obtained
from 0-day-old Polygonia pupae (4 to 12 hr after
larval-pupal ecdysis) and the silkmoth, B. mori , by
dissection in saline (0.9% NaCl) and stored at
— 85°C. Four hundred brains were grouped,
homogenized with a Teflon homogenizer in ice-
cold acetone (a total volume of ca. 4 ml), washed 3
times in 80% ethanol (a total volume of ca. 0.8 ml)
and extracted 3 times with 2% NaCl (a total
volume of ca. 0.6 ml) at 0°C. At each step,
insoluble materials were removed by centrifuga¬
tion at 12,000 Xg for 30 min. The brain-extract of
2% NaCl was heated in a boiling-water bath for 5
min, rapidly cooled and centrifuged at 12,000 Xg
for 30 min. The supernatant was added to solid
ammonium sulfate up to 80% saturation to precipi¬
tate the factor. Then the precipitate was dissolved
in 0.1 M ammonium acetate (pH 7.02) and stored
at — 85°C. The solution was used as crude SMPH.

Trypsin-hydrolysis The crude SMPH was
derived from 800 brains of 0-day-old SD-pupae,
dissolved in 0.2 M ammonium acetate containing
0.01 M calcium chloride (700 pg protein/230 p\)
and incubated with trypsin (7 //g/230 p\) for 20.5 hr
at 25°C. Twenty-four hours later, the incubation
mixture was heated in a boiling water bath (95°C)
for 3 min, rapidly cooled and centrifuged again at
12,000 Xg. The supernatant was bioassayed for
SMPH in addition to a control sample heated
before incubation.

Sephadex G-50 gel-filtration The crude
SMPH from 800 brains of Polygonia pupae or from
250 brains of the silk moth, B. mori, was applied
on to a column (10.2x912 mm) of Sephadex G-50
(Pharmacia, superfine) and eluted with 0.1 M
ammonium acetate (7.3 ml/hr-cm). Sample frac¬
tions consisting of 100 drops each (3.9 ml) were
collected for about 15 hr, lyophilized and stored at
— 85°C.

Reversed-phase HPLC Sephadex G-50
Polygonia fractions showing SMPH-activity and a
highly purified 4K-PTTH of B. mori (bombixin)
[7] — a gift from Professor H. Ishizaki of Nagoya
University — were dissolved in 100/4 of 0.2 M
ammonium acetate, subjected to reversed-phase
HPLC (Shimadzu LC-3A, Tokyo) on a column of
Hi-Pore RP-308 (Toyo Soda, Tokyo). Linear
gradient elution with 10-40% acetonitril in 0.1 M
ammonium acetate was performed over 65 or 90
min. Sample fractions through reversed-phase
HPLC were collected at 4- to 6-min intervals,
lyophilized and stored at — 85°C.

Bioassay of SMPH Five microliters of
samples with/without brain-extracts or the highly
purified 4K-PTTH of B. mori were injected into
the abdomens of 0-day-old female Polygonia
pupae (4 to 12 hr after larval-pupal ecdysis) at
different doses (1- to 100-brain equivalents). The
recipient pupae were allowed to develop at 25°C.

On the day of emergence, butterflies derived
from the treated pupae were classified into one of
the grades 0-4. An average grade score (AGS) for
summer morphs, on which the SMPH-activity of
the sample was evaluated, was obtained from the
response of 6-20 insects. The classification was
based on a gradient used to judge the color of the
ventral side of the wings [4].

Bioassay of 4K-PTTH Five microliters of
samples with/without brain-extract (1- to 100-brain
equivalents) were injected into the abdomens of
diapausing Papilio pupae (15 days after larval-
pupal ecdysis at 25°C). The recipient pupae were
placed at 25°C and observed to see whether or not
they showed any sign of adult-development (eye-
pigmentation). When the pupae exhibited this sign

Summer-morph-producing Hormone

1053

of adult-development within 15 days after treat¬
ment, the test-sample was judged as showing 4K-
PTTH-activity.

RESULTS

Fractional precipitation of SMPH-activity with
ammonium sulfate

An extract from 200 brains of Polygonia SD-
pupae (0-day-old) was made with 2% NaCl and
fractionated by adding solid ammonium sulfate up
to 0.8 saturation. Each precipitate was dissolved in
distilled water and bioassayed using female Poly¬
gonia SD-pupae (0-day-old).

Figure 1 shows that an approximately half of the
SMPH-activity present in the 2%-NaCl extracts
was precipitated by ammonium sulfate between
0.5-0.65 saturation. SMPH-activity was mostly
precipitated by ammonium sulfate at 0.8 saturation
as has been demonstrated in the case of 4K- and
22K-PTTHs of B. mori [6, 7],

Saturation of ammonium sulfate

Fig. 1. Fractional precipitations of SMPH-activity with
ammonium sulfate. Histograms show the percent¬
ages of SMPH-activity recovered from the precipi¬
tates of fractional precipitations.

Effects of heating and trypsin-hydrolysis on the
SMPH-activity

To determine whether or not the factor re¬
sponsible for SMPH-activity in Polygonia brain-
extracts is heat stable and also is resistant to
trypsin hydrolysis, a 2%-NaCl extract was pre¬
pared from 800 brains of Polygonia pupae. The
extract was divided into 4 parts, heated in a

boiling-water bath (95°C) for different time-
periods (0, 1, 3 or 5 min), rapidly cooled and
centrifuged. The supernatants were bioassayed for
SMPH. In addition, an extract was made from 800
brains of Polygonia pupae in the same manner as
above, hydrolyzed with trypsin for 24 hr at 25°C
and bioassayed for SMPH along with a sample
incubated without trypsin.

A high recovery (86%) of SMPH-activity was
obtained from the heat-treated crude SMPH pre¬
paration (Table 1). However, no activity was
detected in the trypsin-hydrolyzed extracts by
Polygonia pupal assay for SMPH (Table 1).

The results indicated that the factor responsible
for SMPH-activity of the Polygonia brain-extracts
was stable upon heating but appeared to be a
peptide hormone since it was hydrolyzed by
trypsin.

Gel-filtration of the crude SMPH preparations
through Sephadex G-50.

Crude SMPH was provided from either 800
brains of Polygonia SD-pupae or from 250 brains
of the silkmoth, B. mori , lyophilized, dissolved in
400 p\ (350 p\ in B. mori) of 0.2 M ammonium
acetate, applied to a column of Sephadex G-50
(Pharmacia, superfine), and eluted with the same
solution (7.3 ml/hr-cm).

SMPH-activity was detected from 5 of 13 gel-
filtrated fractions (tubes no. 11-15), in which
Polygonia and Bombyx materials of M.W. 3500-
5500 were thought to be collected (Fig. 2). Re¬
coveries of the SMPH-activity by the gel-filtration
through Sephadex G-50 reached 40-50% and 56-
62% in the Polygonia and Bombyx crude SMPH
preparations, respectively. Furthermore, two
Sephadex G-50 fractions of the Bombyx crude
SMPH (tubes no. 12 and 13) which showed higher
SMPH-activity than the others were judged as
being 4K-PTTH-active by Papilio pupal assay (Fig.
2).

The results indicated that the molecular sizes of
the Polygonia and Bombyx SMPHs are about 4500
(3500-5500), which are almost the same size as the
4K-PTTH of B. mori (M.W. 4400) but are far
smaller than the size of the 22K-PTTH (M.W.
22,000). However, the SMPHs could not be

1054

T. Masaki, K. Endo et al.

Table 1. Effects of heating and trypsin-hydrolysis on the SMPH-activity in brain-extract of Polygonia
pupae.

Treatments

No.

No. of butterflies classified
into grades:

AGS

0

1

2

3

4

Heating at 95°C

for 0 min

9

0

0

0

0

9

4.0

1 min

9

0

0

0

2

7

3.8

2 min

10

0

0

0

3

8

3.7

5 min

11

0

0

0

3

8

3.7

Trypsin-hydrolysis at 25°C

for 0 hr

6

0

2

3

0

0

1.8

20.5 hr

6

6

0

0

0

0

0.0

20.5 hr (without trypsin)

10

0

0

0

2

8

3.8

Each recipient was injected with the brain-extract of 10-brain (heating) or 20-brain (trypsin-hydrolysis)
equivalents. An AGS (average grade score) for summer morphs was obtained by using a gradient to judge
the color of the ventral-side of the female wings.

i- i i i r i i r i i r i i i

10 15 20

Fraction No.

Fig. 2. Sephadex G-50 gel-filtration patterns of Bombyx (upper panel) and Polygonia crude SMPH preparations
(lower panel). Curved thin lines show absorbance at 254 nm (upper panel) and 280 nm (lower panel),
respectively. Histograms show SMPH-activity detected from the fractions. Double-sided arrow indicates the
fractions judged as being 4K-PTTH-active by Papilio pupal assay.

Summer-morph-producing Hormone

1055

separated from the 4K-PTTHs by gel-filtration
through Sephadex G-50 in both P. c-aureum and
B. mori.

Reverse-phase HPLC in the highly purified 4K-
PTTH of B. mori (bombixin) and the Polygonia
Sephadex G-501 SMPH-active fractions

SMPH-activity of highly purified 4K-PTTH
(bombixin) was bioassayed using short-day Poly¬
gonia pupae. As a result, it was found to show
SMPH-activity as strong as the Polygonia and
Bombyx crude SMPH preparations showed (Fig.
3).

In order to separate the SMPH from the 4K-
PTTH, the highly purified 4K-PTTH (1000-brain
equivalents) was dissolved in 0.2 M ammonium
acetate, applied to on reversed-phase column of
HPLC (Hi-Pore RP-308) and eluted by a linear
gradient of acetonitril concentration of 10-40% in
0.1 M ammonium acetate. The Polygonia SMPH-
active fractions from Sephadex G-50 were lyophi-
lized and applied to the reversed-phase HPLC in

the same manner as above. Sample fractions were
divided into two parts, lyophilized, dissolved in
distilled water and assayed for SMPH- and 4K-
PTTH- activity, respectively.

According to the result of reversed-phase HPLC

Fig. 3. Dose-dependent response curve of SMPH-
activity obtained by highly purified 4K-PTTH of
Bombyx mori (curved thin line with open circles).
Broken lines with letters SD and LD show the
does-dependent response curve obtained from the
brain-extracts of short-day and long-day pupae [5].

E

c

o

CNI

qj

u

c

qj

-Q

o
i r>
_o
<

E

c

LD

CNI

qj

<b

U

c

T3

-Q

O

-Q

<

To 20 30~ To 50 60 70

Retention time (min)

Fig. 4. Reversed-phase HPLC patterns in Bombyx highly purified 4K-PTTH (upper panel) and Polygonia Sephadex
G-50/SMPH-active fractions (lower panel). Solid thin lines show absorbance at 210 nm (upper panel) and 215
nm (lower panel), respectively. Broken lines show the concentration of acetonitril increasing linearly.
Histograms show SMPH-activity in the fractions and double-sided arrows indicate 4K-PHHT-active fractions.

1056

T. Masaki, K. Endo et al.

of the highly purified Bombyx 4K-PTTH, SMPH-
activity was detected from two fractions, where
two small peaks were detected optically at 210 nm.
Furthermore, the two fractions judged as being
SMPH-active showed 4K-PTTH-activity by Papi¬
lio pupal assay (Fig. 4). The fraction showing the
highest SMPH-activity was obtained at the aceto-
nitril concentration of 29-34%, which was signi¬
ficantly higher than the concentration at which
most 4K-PTTH-active fractions were eluted (28-
29%) (Fig. 4: upper panel).

On the other hand, no significant peaks were
detected optically at 215 nm by reversed-phase
HPLC of Polygonia SMPH-active fractions, but 5
fractions showing SMPH-activity were obtained at
the acetonitril concentrations of 28-32% and 34-
43%. Furthermore, one of the fractions showing
SMPH-activity was judged as being 4K-PTTH-
active by Papilio pupal assay. The Polygonia
fraction showing the highest SMPH-activity was
obtained at the acetonitril concentration of 30-
32%, which was almost the same concentration as
the Bombyx fraction showing the highest SMPH-
activity. However, the most active 4K-PTTH
fraction was obtained in P. c-aureum at the aceto¬
nitril concentration of 28-30%, which was a little
lower than the concentration of most SMPH-active
fractions (30-32%) as observed in the case of the
silkmoth, B. mori, (Fig. 4: lower panel).

The results indicated that the Bombyx and Poly¬
gonia factors showing SMPH-activity seemed to be
similar to those of the 4K-PTTH in the size of its
molecules as well as in chemical characteristics.
However, the factor showing SMPH-activity did
not seem to be identical with the 4K-PTTH in P.
c-aureum.

DISCUSSION

The mechanism underlying the photoperiodic
control of seasonal-morph determination in Poly¬
gonia involves a neuroendocrine factor producing
summer morphs (SMPH) [1-4], The SMPH can
be extracted with 2% NaCl from the brains of
Polygonia pupae, but is unsuccessful with acetone
or 80% ethanol [5]. The SMPH is stable to heating
(Table 1), but is thought to be a peptide hormone
since it was precipitated by ammonium sulfate

(Fig. 1) and inactivated by trypsin-hydrolysis
(Table 1). This has been demonstrated in several
other neurohormones of insects, i.e., melanization
and reddish coloration hormone in Leucania sepa¬
rata [8], diapause hormone [9] and PTTHs in B.
mori [7].

In addition, a factor showing the same activity as
Polygonia SMPH is present in the brains of the
silkmoth, B. mori [5]. The Bombyx factor could
be extracted (with 2% NaCl) and precipitated (by
ammonium sulfate at 80% saturation) in almost
the same manner as has been demonstrated in the
case of 4K- and 22K-PTTHs of B. mori [7].

The coincidental behaviors of SMPH and
PTTHs can be explained using two hypothetical
models, as has been demonstrated in a Papilio
system underlying the photoperiodic control of
seasonal-morph determination [10]. The first mod¬
el is based on 3 neurohormones, i.e., one SMPH
and two PTTHs, and secretion of the SMPH is
controlled by larval exposure to appropriate
photoperiod and temperature. The coincidental
behaviors may be explained, in this model, by a
hypothesis that the molecule showing SMPH-
activity is similar to those of 4K- and 22K-PTTHs
in chemical characteristics in P. c-aureum. The
second model involves two neurohormones, i.e.,
one SMPH and one PTTH. One is the SMPH
showing 4K-PTTH-activity, whereas the other
(22K?) is the true PTTH triggering pupal-adult
ecdysis.

From the Sephadex G-50 gel-filtration patterns
of SMPH- and 4K-PTTH-activities (Fig. 2), it may
be concluded that the molecular sizes of the Poly¬
gonia and Bombyx SMPHs (about 4500) are
almost the same as that of the Bombyx 4K-PTTH
(bombixin: M.W. 4400 [7]) but are far smaller than
the size of the Bombyx 22K-PTTH (M.W. 22,000
[7]).

We could not provide any evidence with respect
to the molecular size of Polygonia PTTH trigger¬
ing pupal-adult ecdysis, but it was thought to be
larger than M.W. 15,000. This estimation was
based on the number of days required for adult
development, i.e., most pupae treated with Poly¬
gonia Sephadex G-50 fractions emerged 7 or 8
days after larval-pupal ecdysis as did the controls
or the pupae treated with Bombyx Sephadex G-50

Summer-morph-producing Hormone

1057

fractions. However, the pupae emerged one day
earlier than the controls (day 6 or 7) when they
were treated with one of the Sephadex G-50
fractions containing Polygonia materials of a larger
molecular weight than 15,000.

In addition, the fractions showing the highest
SMPH- and the highest 4K-PTTH-activities were
obtained at different acetonitril concentrations by
reversed-phase HPLC, indicating that the factor
showing SMPH-activity did not seem to be identi¬
cal with the one showing 4K-PTTH-activity in both
P. c-aureum and B. mori. However, several
microheterogenieties were shown to be present in
the molecules of Bombyx 4K-PTTH (bombixin)
[7], We have left open the interesting question of
whether or not one kind of 4K-PTTH molecule
showing lower activity than the others produces
summer morphs in P. c-aureum. The question may
be solved during a course of further study concern¬
ing the roles SMPH plays in the development of
seasonal morphs.

ACKNOWLEDGMENTS

The authors wish to express their sincere gratitude to
Professor A. Okajima and to Professor Y. Chiba of
Yamaguchi University for advice and valuable sugges¬
tions during the course of this work. Thanks are also due
to Professor H. Ishizaki of Nagoya University for his gift
of highly purified 4K-PTTH and to Professor S. Y.
Takahashi of Yamaguchi University for his advice during
the preparation of this manuscript. This work was
supported in party 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¬

terfly, Polygonia c-aureum L. Proc. Japan Acad., 42:
1082-1087.

2 Hidaka, T. and Takahashi, H. (1967) Temperature
condition and maternal effect as modifying factor in
the photoperiodic control of seasonal forms in Poly¬
gonia c-aureum (Lepidoptera, Nymphalidae).
Annot. Zool. Japon. 40: 200-204.

3 Endo, K. (1972) Activation of corpora allata in
relation to ovarian maturation in the seasonal forms
of the butterfly, Polygonia c-aureum L. Dev.
Growth Differ., 14: 263-274.

4 Endo, K. (1984) Neuroendocrine regulation of the
development of seasonal forms of the Asian comma
butterfly, Polygonia c-aureum L. Dev. Growth Dif¬
fer., 26: 217-222.

5 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.

6 Ishizaki, H. and Ichikawa, M. (1967) Purification of
the brain hormone of the silkworm, Bombyx mori.
Biol. Bull., 133: 355-368.

7 Ishizaki, H., Suzuki, A., Isogai, A., Nagasawa, H.
and Tamura, S. (1977) Enzymatic and chemical
inactivation of partially purified prothoracicotropic
(brain) hormone of the silkworm, Bombyx mori. J.
Insect Physiol., 23: 1219-1222.

8 Suzuki, A., Matsumoto, S., Ogura, N., Isogai, A.
and Tamura, S. (1976) Extraction and partial
purification of the hormone inducing cuticular mela-
nization in armyworm larvae. Agric. Biol. Chem.,
40: 2307-2309.

9 Sonobe, H. and Ohnishi, E. (1971) Silkworm Bom¬
byx mori L.: nature of diapause factor. Science, 174:
835-838.

10 Endo, K. and Funatsu, S. (1985) Hormonal control
of seasonal morph determination in the swallowtail
butterfly, Papilio xuthus L. (Lepidoptera: Papilioni-
dae). J. Insect Physiol., 31: 669-674.

mm ■

ZOOLOGICAL SCIENCE 5: 1059-1064 (1988)

© 1988 Zoological Society of Japan

Effects of Water Temperature and Photoperiod on the Beginning
of Spawning Season in the Orange-red Type Medaka* 1

Masahiko Awaji2 and Isao Hanyu

Laboratory of Fish Physiology, Faculty of Agriculture,

University of Tokyo, Bunkyo, Tokyo 113, Japan

ABSTRACT — Rearing experiments were conducted to study effects of photoperiod and water
temperature on the beginning of spawning season in the orange-red type medaka, Oryzias latipes. Fish,
cultured in Tokyo and ready for rapid gonadal growth in spring, was transferred to 7 conditions (16°C
14L, 14°C 14L, 14°C 12L, 14°C 10L, 12°C 14L, 10°C 14L, 8°C 14L), and reared for 4 to 8 weeks.
Changes in gonadosomatic index (GSI) and gonadal histology were examined. Groups at 14 and 16 °C
showed rapid GSI increase irrespective of photoperiod. Active yolk globule accumulation and
spermatogenesis began, and spawning was observed in these groups. After 8 weeks, many regressive
oocytes were noticeable in 14°C groups. In groups at temperature below 14°C, GSIs of both sexes
remained low and spawning was not observable. But yolk globule accumulation proceeded slowly in 10
and 12°C groups. When medaka in early spring were transferred to 4 conditions (16°C 10L, 8L, 6L, 4L),
and reared for 4 weeks, they could mature and spawn in all experimental regimes, which eliminates
possibility for photoperiodic response in spring. From these results, it is clear that temperature rise in
spring only is responsible for the beginning of spawning season in medaka cultured around Tokyo.

INTRODUCTION

Most teleost fishes have their own annual repro¬
ductive cycles. Though some fishes seem to have
endogenous reproductive rhythm [1], the annual
reproductive cycle is generally believed to depend
upon seasonal changes in environmental factors:
photoperiod and water temperature in temperate
zone, and rainfall in tropical zone [1-6].

Medaka, Oryzias latipes , is a small freshwater
teleost native to Japan and her adjacent areas [7].
Two types of medaka, namely the wild and the
orange-red types [8], are well known. Their
annual reproductive cycles have already been in¬
vestigated [9-12], and the spawning season is
understood to extend usually from mid-April to
early September.

There have been several works concerning en¬

Accepted January 29, 1988
Received October 23, 1987

1 Studies on the Reproductive Rhythms of Fishes-
XXIV.

2 To whom reprints should be requested. Present
address: National Research Institute of Aquaculture,
Nansei, Mie 516-01, Japan.

vironmental effects on the beginning of spawning
season in medaka, with attention engaged by
effects of photoperiod and/or water temperature
[10, 13-19]. But obtained results did not always
coincide with one another. As for the orange-red
type medaka, in additions, effects of photoperiod
in combination with water temparature have not
been clearly shown yet, warranting further inves¬
tigations.

This study was conducted to clarify environmen¬
tal factors responsible for the beginning of spawn¬
ing season in the orange-red type medaka. Rapid
and remarkable gonadal growth, i.e. commence¬
ment of yolk globule accumulation in ovaries and
retrieval of meiosis in testes, is leading and inevi¬
table process for the beginning of spawning [9, 10].
Efforts were made to reveal environmental factors
which cause this rapid gonadal growth. Thus
medaka in early spring, which had been reared
under natural condition and was ready for the
rapid gonadal growth, was used as material.
Photoperiod and temperature regimes were
selected to be comparable to, or not too far from,
natural environmental changes in spring.

1060

M. Awaji and I. Han yu

MATERIALS AND METHODS

Two experiments were conducted in 1980 (Ex¬
periment 1) and 1981 (Experiment 2). The orange-
red type medaka Oryzias latipes, born in the
preceding breeding season, were obtained from a
hatchery in Edogawa-ku, Tokyo, kept in a pen
(area 14 m2, depth 40 cm) placed in a circulating
outdoor tank (area 14 m2, maximum depth 50 cm)
at least for 4 months till experiments, and fed
tubifex (freshwater oligochaetes) to satiation.

During experiments, water temperature in the
outdoor tank ranged from 6 to 16°C, and the
daylength [20] increased from 11.3 to 13. 3L.

Experiment 1 was started on February 27 and
ended on April 24. Fish ranging from 21 to 31 mm
in standard length were transferred to following 7
conditions: 8°C 14L (14 hours of light/day), 10°C
14L, 12°C 14L, 14°C 14L, 14°C 12L, 14°C 10L,
16°C 14L. The three 14°C groups were reared for 8
weeks, and the others for 4 weeks. Twenty and 30
fish of each sex were used for 8°C, 10°C, 12°C or
16°C regime and three 14°C regimes, respectively.
Part of fish in each group was taken on March 13
and 27, and the rest at the end.

Experiment 2 was started on March 6 and ended
on April 7. Thirty fish of each sex, with standard
length of 20 to 30 mm, were reared under 16°C 4L,
16°C 6L, 16°C 8L or 16°C 10L. Part of the fish
were sampled halfway also in this experiment.

Each experimental group was kept in a 50 to 60
liter, light proof aquarium with circulation and
some water plants. Photoperiod was controlled
with a 10 W fluorescent lamp connected to a timer,
and water temperature with a thermostat and
heater. Fish were fed with tubifex ad lib.

After taken up, fish were anesthetized with
tricaine methane sulfonate. The whole body with
opened abdominal cavity was fixed in Holland’s
solution-sublimate in Experiment 1 and in Bouin’s
solution in Experiment 2. Standard length, body
weight and gonad weight were measured while in
70% alcohol. GSI (gonad weight X 100/body
weight) was calculated and gonads were embedded
in paraffin by ordinary method. Ovarian sections
in 10 pm were stained with AZAN and testicular
sections in 5 pm with Mayer’s hematoxylin and
eosin.

RESULTS

Experiment 1

Figure 1 shows the changes in GSI under ex¬
perimental regimes. Average GSI of initial con¬
trols was around 2.5% in female and 0.5% in male.
The ovary contained oocytes up to yolk vesicle
stage. The testis had spermatogonia, primary
spermatocytes and residual sperm.

Females exposed to higher temperature (14 or
16°C) had rapid increases of GSI, reaching 6 to 9%
in the fourth week, irrespective of photoperiod.
The beginning of yolk globule accumulation was
observed in the second week, and fully matured
oocytes were noticeable in the fourth week. First
spawning was observed in 14°C 12L group 25 days
after the commencement, followed by spawning in
the other groups at higher temperatures. GSIs of
three groups at 14°C kept high till the eighth week,
but ovaries of these groups contained many regres¬
sive oocytes at the end of the experiment.

Females exposed to lower temperatures (12, 10
or 8°C) had little increase of GSI in spite of long
daylength. GSIs of 10°C 14L and 12°C 14L groups
were 3 to 4% in the fourth week. Their ovarian
histology in the second week was unchanged, but
in the fourth week early yolk globule stage oocytes
were observed. GSI and ovarian histology of 8°C
14L group in the fourth week were equal to those
of initial control. No spawning was observed in
these groups.

GSIs of males exposed to higher temperatures
showed rapid increase, reaching 1.0 to 1.2% in the
fourth week, and three groups at 14°C kept that
value till the eighth week. Increase of spermatogo¬
nia and primary spermatocytes was observed in the
second week, leading to active spermatogenesis
shown by meioses of spermatocytes in the fourth
and eighth weeks.

Males exposed to lower temperature showed a
little increase of GSI, reaching 0.6 to 0.8% in the
fourth week. But active spermatogenesis was not
observed.

Experiment 2

Results of Experiment 1 indicated importance of
rising temperature in the beginning of spawning

Beginning of Spawning Season in Medaka

1061

season in medaka, daylength changes having little
effect on gonads. Experiment 2 was conducted to
examine the possibility of photoperiodic response
in gonadal development.

Figure 2 shows the changes of GSI under ex¬
perimental regimes. Average GSI of initial con¬
trols was around 2.0% in female and 0.5% in male.
Gonadal histology of initial controls was much the

Fig. 1. Changes in the gonadosomatic index (GSI) of the orange-red type medaka reared under various temperature-
photoperiod regimes in Exp. 1.

Fig. 2. Changes in the gonadosomatic index (GSI) of the orange-red type medaka reared under short day regimes in
Exp. 2.

1062

M. Awaji and I. Hanyu

Fig. 3. A: Ovary from the orange-red type medaka reared under 16°C 4L in spring. Tertially yolk globule stage
oocyte with germinal vesicle (arrow) is observable. 50 X. B: Testis from the orange-red type medaka reared
under 16°C 4L in spring. Cysts in various stages of spermatogenesis and the metaphase of first miotic division
(arrow) indicate active spermatogenesis. 200 X.

same as in Experiment 1, except that some ovaries
contained only perinucleolus stage oocytes.

GSIs increased in every group at the raised
temperature, irrespective of daylength. Female
GSIs reached 3 to 4% on the 20th day, and 7 to 8%
after 1 month. At the end of the experiment,
ovaries contained yolk globule stage oocytes (Fig.
3 A) and some ovulatory follicles. Male GSIs were
0.6 to 0.8% on the 20th day, and 0.8 to 1.0% after
1 month. Testes showed some increase in primary
spermatocytes on the 20th day, active sperma¬
togenesis after 1 month (Fig. 3B). Spawning was
observed 1 month after the commencement in 6
and 10L group, and with a little delay in other
groups.

DISCUSSION

In the female medaka under natural condition,
rapid yolk-globule accumulation and consequent
spawning occur in spring after sufficient yolk vesi¬
cle formation and associated ovarian changes in
winter [9, 10]. In the same way, males have
testicular inactiveness and subsequent prolifera¬
tion of spermatogonia in winter, prior to beginning
of active spermatogenesis in spring [9].

In the present study, the authors tried to clarify
environmental factors responsible for the rapid
gonadal growth in spring. The orange-red type
medaka, ready for the rapid gonadal growth and
therefore fully responsive to environmental cues,

was used as material. In Experiment 1, medaka
reached full gonadal maturation under constant
water temperature of 14°C and above, regardless
of photoperiod. This was further confirmed by the
results of Experiment 2, which eliminated possible
involvement of a critical daylength as short as 4
hours. These results clealy indicate that the rising
temperature only is responsible for the rapid
gonadal growth and subsequent beginning of
spawning season in spring.

Effects of photoperiod on the beginning of
gonadal maturation in combination with water
temperature have not been examined well in the
orange-red type medaka. Chan [18] showed nece-
sity of long daylength or appropiate interruption of
dark period for maturation and spawning of the
orange-red type medaka. But her fish had been
kept under artificial conditions till experiments,
and maturational stages of initial controls were
younger than those used in the present study. As
shown already in cyprinids [21-24] and the stickle¬
back [25], ovarian response to environmental ma¬
nipulations has turned out to vary with the matu¬
rational stages in the orange-red type medaka (in
preparation). This coincides with Chan’s results in
the necessity of long daylength for maturation and
spawning of females initially in early or middle of
yolk-vesicle stage. Egami and Hosokawa [10]
transferred the orange-red type medaka main¬
tained under natural temperature and photoperiod
to aquaria at 23 to 26°C in various seasons. From

Beginning of Spawning Season in Medaka

1063

October to February, the gonads developed by the
treatment without photoperiod change, and the
females began to lay eggs earlier than those under
natural condition. Their results agree with ours in
the importance of rising temperature in spring, but
not for the other seasons due to unidentifed
reason.

In the wild type medaka, we found that the
rising temperature only is responsible for the be¬
ginning of spawning season by the same methods
as ones in this study using the fish obtained from
Ushikunuma pond, Ibaraki Pref. (unpublished
data). But Yoshioka [15] indicated that long
daylength is indispensable for maturation and
spawning in spring in the wild type medaka whose
ovary initially contained early yolk-globule state
oocytes. This is inconsistent with the present
results or with our unpublished data on the wild
type medaka. Yoshioka’s material was obtained
from a pond in Hokkaido (northern Japan), in
which changes in photoperiod or temperature is
different from those around Tokyo. Difference in
environmental conditions during immature period
may cause different gonadal response to environ¬
mental manipulations in advanced stages. On the
other hand, Sawara and Egami [19] reported possi¬
ble existence of racial difference in photoperiodic
response of gonads, which may also be a reason of
the inconsistency. To show racial difference clear¬
ly, it must be necessary to use material from
different localities, which is in the same reproduc¬
tive stage and has been reared under the same
environment.

Ovarian regression was observed in 14°C groups
at the end of Experiment 1. The reason of the
regression is not clear. But according to Shiraishi
et al. [26], some females cease spawning after
several weeks of daily spawning under constant
photoperiod-temperature regimes. The same phe¬
nomenon may occur after daily spawning under
some experimental conditions employed here. It is
also possible that 14°C is too low for maintenance
of gonadal maturation.

Observation of yolk globule accumulation in 10
and 12°C groups, and not in 8°C group, suggests
two phases of yolk globule accumulation: fast
accumulation at above 14°C, slow accumulation at
10 to 14°C. In goldfish Yamazaki [27] obtained

similar results by observing slow yolk globule
accumulation at 15°C, and fast accumulation at 20
and 25°C in early spring. It is also known that
female goldfish can mature at 13 to 14°C, but can
not ovulate without temperature rise to 20°C [28].
It will be necessary to ascertain the slow yolk
globule accumulation of medaka at 10 and 12°C
with longer rearing for understanding relationship
between water temperature and maturation or
ovulation.

ACKNOWLEDGMENT

The authors are grateful to Dr. K. Aida for his advice
during this work. Part of this study was supported by a
Grant-in-Aid for Scientific Research from the Ministry of
Education, Science and Culture, Japan

REFERENCES

1 Sundararaj, B. J. and Vasal, S. (1976) Photoperiod
and temperature control in the regulation of repro¬
duction in the female catfish Heteropneustes fossilis.
J. Fish. Res. Board Can., 33: 959-973.

2 Crim, L. W. (1982) Environmental modulation of
annual and daily rhythms associated with reproduc¬
tion in teleost fishes. Can. J. Fish. Aquat. Sci., 39:
17-21.

3 Hanyu, I., Asahina, K., Shimizu, A., Razani, H.
and Kaneko, T. (1983) Environmental regulation of
reproductive cycles in teleosts. Proc. 2nd N. Pac.
Aquaculture Symp., 173-188.

4 Peter, R. E. and Crim, L. W. (1979) Reproductive
endocrinology of fishes: Gonadal cycles and gona¬
dotropin in teleosts. Ann. Rev. Physiol., 41: 323-
335.

5 Peter, R. E. (1981) Gonadotropin secretion during
reproductive cycles in teleosts: Influences of en¬
vironmental factors. Gen. Comp. Endocrinol., 45:
294-305.

6 De Vlaming, V. L. (1972) Environmental control of
teleost reproductive cycles: a brief review. J. Fish.
Biol., 4: 131-140.

7 Yamamoto, T. (1975) Systematics and zoogeogra¬
phy. In “Medaka (Killifish), Biology and Strains”.
Ed. by T. Yamamoto, Keigaku Publ. Co., Tokyo,
pp. 17-29.

8 Yamamoto, T. (1975) Introductory remarks on the
medaka. In “Medaka (Killifish), Biology and
Strains”. Ed. by T. Yamamoto, Keigaku Publ. Co.,
Tokyo, pp. 1-16.

9 Awaji, M. and Hanyu, I. (1987) Annual reproduc¬
tive cycle of the wild type medaka. Nippon Suisan
Gakkaishi, 53: 959-965.

1064

M. Awaji and I. Han yu

10 Egami, N. and Hosokawa, K. (1973) Responses of
the gonads to environmental changes in the fish,
Oryzias latipes. In “Response of Fish to Environ¬
mental Changes”. Ed. by W. Chavin, Charles C.
Thomas, Springfield, pp. 279-301.

11 Nishikawa, S. (1956) Histological observations on
the seasonal spermatogenetic cycle of a medaka,
Oryzias latipes (T & S). Zool. Mag., 65: 203-206.

12 Yamamoto, K. and Yoshioka, H. (1964) Rhythm of
development in the oocyte of the medaka, Oryzias
latipes. Bull. Fac. Fish. Hokkaido Univ., XV: 5-24.

13 Yoshioka, H. (1962) On the effects of environmen¬
tal factors upon the reproduction of fishes. I. the
effects of day-length on the reproduction of the
Japanese killifish, Oryzias latipes. Bull. Fac. Fish.
Hokkaido Univ., XIII: 123-136.

14 Yoshioka, H. (1963) On the effects of environmen¬
tal factors upon the reproduction of fishes. 2. Effects
of short and long day-length on Oryzias latipes
during spawning season. Bull. Fac. Fish. Hokkaido
Univ., XIV: 137-151.

15 Yoshioka, H. (1966) On the effects of environmen¬
tal factors upon the reproduction of fishes. 3. The
occurrence and regulation of refractory period in the
photoperiodic response of medaka, Oryzias latipes.
J. Hokkaido Univ. Education, Sec. II (B), 17: 23-
33.

16 Yoshioka, H. (1970) On the effects of environmen¬
tal factors upon the reproduction of fishes. IV.
Effects of long photoperiod on the development of
ovaries of adult medaka, Oryzias latipes , at low
temperatures. J. Hokkado Univ. Education, Sec. II
(B), 21: 14-20.

17 Yoshioka, H. (1971) On the effects of environmen¬
tal factors upon the reproduction of fishes. The
significance of combinations of light and dark
periods in photoperiodic response of the ovaries of
medaka, Oryzias latipes , in out-of-breeding seasons.
Seibutu kyozai, 8: 76-82.

18 Chan, K. K.-S. (1976) A photosensitive daily
rhythm in the female medaka, Oryzias latipes. Can.
J. Zool., 54: 852-856.

19 Sawara, Y. and Egami, N. (1977) Note on the

differences in the response of the gonad to the
photoperiod among populations of Oryzias latipes
collected in different localities. Annot. Zool.
Japon., 50: 147-150.

20 Rika nenpyo (Chronological Scientific Tables)
(1984), ed by Tokyo Astronomical Observatory,
Maruzen Co., Ltd., Tokyo, pp. 12-16.

21 Asahina, K. and Hanyu, I. (1983) Role of tempera¬
ture and photoperiod in annual reproductive cycle of
the rose bitterling Rhodeus ocellatus ocellatus. Nip¬
pon Suisan Gakkaishi, 49: 61-67.

22 Shimizu, A. and Hanyu, I. (1982) Environmental
regulation of annual reproductive cycle in a spring¬
spawning bitterling Acheilognathus tabira. Nippon
Suisan Gakkaishi, 48: 1563-1568.

23 Razani, H. and Hanyu, I. (1986) Annual reproduc¬
tive cycle of 2-3 years old female goldfish and its
artificial modification by manipulations of water
temperature and photoperiod. Nippon Suisan Gak¬
kaishi, 52: 965-969.

24 Razani, H. and Hanyu, I. (1986) Effects of con¬
tinued short photoperiod at warm temperature and
following changes of regimes on gonadal maturation
of goldfish. Nippon Suisan Gakkaishi, 52: 2061-
2068.

25 Baggerman, B. (1972) Photoperiodic responses in
the stickleback and their control by a daily rhythm of
photosensitivity. Gen. Comp. Endocrinol. Suppl.,
3: 466-476.

26 Shiraishi, M., Tanaka, S. and Deguchi, Y. (1977)
The number of eggs spawned by himedaka, Oryzias
latipes , under artificial photoperiod. Bull. Coll. Agr.
Vet. Med., Nihon Univ., 35: 315-320.

27 Yamazaki, F. (1965) Endocrinological studies on
the reproduction of the female goldfish, Carassius
auratus L., with special reference to the function of
the pituitary gland. Mem. Fac. Fish. Hokkaido
Univ., XIII: 1-64.

28 Yamamoto, K., Nagahama, Y. and Yamazaki, F.
(1966) A method to induce artificial spawning of
goldfish all through the year. Nippon Suisan Gak¬
kaishi, 32: 977-983.

ZOOLOGICAL SCIENCE 5: 1065-1071 (1988)

© 1988 Zoological Society of Japai

Fine Structure of Agranular Cells in the Gummy Shark
( Mustelus manazo) Adenohypophysis

Akira Chiba and Yoshiharu Honma1

Department of Biology, Nippon Dental University, Niigata 951,
and lSado Marine Biological Station, Niigata
University, Niigata 952-21, Japan

ABSTRACT — A light and electron microscopic study of the gummy shark ( Mustelus manazo )
adenohypophysis revealed two types of agranular cells (cavity boundary cells and stellate cells) forming
a loose network in the distal and ventral lobes. Both types of the cells showed a high nucleocytoplasmic
ratio and electron-dense cytoplasm. The cavity boundary cells lining the hypophysial cavities displayed
varying numbers of microvilli, occasional cilia and periluminal vesicles with flocculent content. In some
of the stellate cells abutting on degenerating endocrine cells, phagocytotic activity was evident. These
results suggest that the agranular cells in the shark adenohypophysis may be involved in sustaining the
endocrine cells, absorption or transport of substances from the cells and disposition of dysfunctioning
endocrine cells.

INTRODUCTION

Agranular (chromophobic) cells have been re¬
ported in the adenohypophyses of a wide variety of
vertebrates. These cells in lower vertebrates are
considered to correspond to folliculo-stellate cells
in higher vertebrates and are variously termed as
follicle boundary cells, supporting cells, stellate
cells, chromophobes and so on [1]. The hypoph¬
ysis of selachians is characterized by a well-
developed cavitary system originating from the
recess of Rathke’s pouch in the embryo and con¬
taining a number of lining agranular cells that are
also found dispersed throughout the parenchyma
[2-4], According to Lagios [5], agranular cells are
particularly important as supportive cells for primi¬
tive fishes bearing large cavities in their hypoph¬
ysis. Knowles et al. [2] and Alluchon-Gerard [3, 4]
reported the fine structure of agranular cells in
selachian hypophyses, but their function is not
clearly established.

The present study was designed to elucidate the
fine structure of agranular cells in the gummy
shark hypophysis, with the aim of shedding some

Accepted January 8, 1988
Received November 21, 1987

light on their function(s).

MATERIALS AND METHODS

Five male gummy sharks ( Mustelus manazo
Bleeker), ranging from 40 to 100 cm in total
length, were used in this study. They were caught
in a set net installed off the Sado Marine Biological
Station of Niigata University, located on the west
coast of Sado Island in the Sea of Japan, and
reared in an aquarium at the station prior to use.

For transmission electron microscopy, the brains
with hypophyses were quickly removed from the
skulls immediately after the animals were killed by
decapitation under anesthesia with MS-222 and
immersed in 2% glutaraldehyde in 0.1 M cacody-
late buffer (pH 7.3) containing 1.68% sodium
chloride, as recommended by Saito and Tanaka
[6]. The hypophyses were cut into small pieces,
rinsed in the same buffer, postfixed with 1%
osmium tetroxide, dehydrated and embedded in
Epon 812. Ultrathin sections were double-stained
with uranyl acetate and lead solutions and ex¬
amined with Hitachi H-500 and Jeol 1200 EX
electron microscopes.

For scanning electron microscopy, the gland
fixed with aldehyde was cut into two halves to

1066

A. Chiba and Y. Honma

disclose the hypophysial cavities, dehydrated and
critical-point dried. After sputter-coating with
gold, the specimens were observed with a Hitachi
S-500 scanning electron microscope.

Epon semithin sections stained with azan
trichrome, aldehyde fuchsin (AF), or periodic acid
Schiffs reagent (PAS) were surveyed by light
microscopy.

RESULTS

Light microscopy

Light microscopy of the survey sections showed
a number of agranular cells scattered over the
distal and ventral lobes. These cells were distin¬
guishable from the endocrine cells by their smaller
size, irregular outline and darkly stained cyto¬
plasm. Mainly on the basis of their locations, the
cells could be classified into two cell types, which
coincided with the cavity boundary (pericavity)
cells and the stellate cells in Scy Ilium [3, 4]. The
former lined the cavitary system (hypophysial cleft
in connection with follicular lumina and recess of
the ventral lobe), while the latter were inter¬
spersed among the endocrine cells. These cells
altogether formed a loose network in the paren¬
chyma. The cavities contained varying amounts of
AF- and also PAS-positive colloid.

Transmission electron microscopy

The cavity boundary cells were characterized by
having a thin electron-dense cytoplasm surround¬
ing an irregularly shaped nucleus (Fig. 1). They
were linked to one another by a tripartite junction¬
al complex comprising desmosome, zonulae
occludens and adherens to form the characteristic
barrier between the endocrine cells and the cavi¬
tary system (Fig. 2). This barrier was incomplete,
however, for it was occasionally intercalated with
endocrine cells. The cavity boundary cells pro¬
jected varying numbers of microvilli and one or
more cilia into the cavities in which colloid was
occasionally demonstrated (Figs. 3 and 4). The
colloid was heterogeneous and composed of finely
granular material, dense amorphous material,
vesicles of various sizes, membranous structures
and mitochondria-like bodies, strongly suggesting

that it might represent debris derived from disin¬
tegrating cells (Fig. 3). Usually, a glycocalyx was
spread over the apical surface of the cells. Thin
cytoplasmic processes of the cavity boundary cells
extended between the endocrine cells and/or abut¬
ted on the basement membrane adjacent to the
blood vessels. The cytoplasm had relatively few
inclusions. In addition to the lack of secretory
granules, the presence of abundant free ribo¬
somes, microfilaments and a number of small
vacuoles was characteristic of these cells. The
vacuoles were scattered throughout the entire
cytoplasm, but were mostly concentrated in the
subapical region of the cells together with a num¬
ber of periluminal vesicles having a flocculent
content (Fig. 4). Varying numbers of mitochon¬
dria and Golgi complexes were present as were
electron-dense bodies probably of lysosomal na¬
ture, fragments of granular endoplasmic reticu¬
lum, multivesicular bodies and myelin figures.

The stellate cells had a star-like configuration,
extending their multiple cytoplasmic processes
among the endocrine cells (Fig. 5). They were less
numerous than the cavity boundary cells and were
scattered throughout the parenchyma. Their nu-
cleocytoplasmic ratio was high and no cilia could
be seen. However, cytological characteristics of
the stellate cells were closely similar to those of the
cavity boundary cells. Desmosomes were seldom
encountered between the processes of the adjoin¬
ing stellate cells or between stellate and cavity
boundary cells. In the interior of the parenchyma,
there were intercellular spaces where the processes
of stellate cells and/or cavity boundary cells inter-
digitated with each other. Usually, these spaces
were narrow, but occasionally were dilated and
contained amorphous material and vesicular or
membranous structures. Noticeably, some stellate
cells abutting on disintegrating endocrine cells
projected their intricate cytoplasmic processes into
the endocrine cells; and present within the proces¬
ses were phagocytotic vacuoles, some of which
were filled with structures clearly recognizable as
secretory granules (Fig. 6). The limiting mem¬
brane of the granules seemed to be still intact.
Small parts of the cytoplasm of the distintegrating
endocrine cells appeared to have been engulfed by
the processes of the stellate cells. Empty vesicles

Agranular Cells in Shark Adenohypophysis

1067

Fig. 1. Cavity boundary cells in the distal lobe of the gummy shark ( Mustelus manazo ) adenohypophysis. Their
cytoplasm is electron-dense and contains small vacuoles, a few mitochondria and Golgi bodies. X 10,000.

Fig. 2. A portion of a cavity boundary cell to show the junctional complex between two adjacent cells and a cilium
projecting into the lumen. Notice cell surface coated with glycocalyx. X 24,000.

Fig. 3. Apical portion of a cavity boundary cell showing intra-luminal granular materials, vesicles and mitochondria¬
like bodies. X 18,000.

Fig. 4. Apical portions of cavity boundary cells showing several microvilli, a number of periluminal vesicles with
flocculent content. Note the heterogeneous constitution of the intraluminal colloid. X 22,000.

1068

A. Chiba and Y. Honma

Fig. 5. Stellate cell (S) in the gummy shark adenohypophysis. Note long cytoplasmic processes extending among
endocrine cells. X 13,000.

«-y.M

Fig. 6. Portion of a stellate cell showing phagocytotic activity. Note irregular cytoplasmic processes of the stellate
cells penetrating into disintegrating endocrine cells. Phagocytotic vacuoles with or without secretory gra
micropinocytotic vesicles (arrowhead) and myelin figure are also visible. X 21, 000.

Agranular Cells in Shark Adenohypophysis

1069

and pits were also present along the plasma mem¬
brane of the cytoplasmic processes of the stellate
cells.

Scanning electron microscopy

Scanning electron microscopic (SEM) observa¬
tion of the hypophysial cleft confirmed the nature
of the cavity boundary cells revealed by transmis¬
sion electron microscopy (TEM): a varying num¬
ber of microvilli, few cytoplasmic processes and
occasional cilia projecting from the cells (Fig. 7).
The cilia were absent in the majority of the cells,
while small concavities, possibly representing mi-
cropinocytotic vesicles, were visible here and
there. In some of the cells, small bulbous protru¬
sions with or without microvilli were occasionally
found. The SEM image of the hollow in the
ventral lobe was similar to that of the hypophysial
cleft, though microvilli and cilia were very few.

Fig. 7. Scanning electron micrograph showing apical
surface of cavity boundary cells lining the hypophy¬
sial cleft. A number of microvilli, few cilia and
bulbous protrusions (arrowheads) are noticeable.
X3,600.

DISCUSSION

The present research on the Mustelus hypoph¬
ysis revealed a number of the agranular cells
forming a loose network in the parenchyma of the
distal and ventral lobes as shown to be the case in
the mammalian pars distalis [7] and the adenohy¬
pophyses of chondrostean and holostean fishes [5].
This organization of the cells, depicted by inter¬
connections comprising junctional complexes and
interdigitation, strongly suggests a sustentacular
role for these cells.

Smaller size, lack of secretory granules and
more or less paucity of cell organellae in the
agranular cells suggest a possibility that some of
these cells represent undifferentiated stem cells.
Yoshimura et al. [8] have proposed a stem-cell
hypothesis in the rat. According to Alluchon-
Gerard [9], the embryonic hypophysis of Scy Ilium
includes undifferentiated cells in addition to other
agranular elements, supporting cells and pericavity
cells; however, Knowles et al. [2] have commented
nothing as to the presence of undifferentiated cells
in the hypophysis of the adult Scyliorhinus. In the
light of these results, further examination on the
ontogenetic development of the gland is necessary
to identify the undifferentiated cells in Mustelus
hypophysis.

The cytological features of the agranular cells in
Mustelus, i.e., high nucleocytoplasmic ratio, dark
cytoplasm due to abundant free ribosomes, junc¬
tional complex between adjoining cells and so on,
were essentially identical with those in Scy Ilium [3,
4]. Similar results were obtained in various groups
of vertebrates (cyclostomes, [1, 10]; osteichthyans,
[1, 5, 11, 12]; amphibians, [13, 14]; reptiles, [15,
16]; mammals, [7, 8, 17]).

In accordance with the previous findings on
Scyllium [3, 4], two kinds of agranular cells, the
cavity boundary cells and stellate cells, could be
distinguished in the hypophysis of Mustelus.
However, the “Giant cells” described in the gland
of very young specimens of Scyllium [4] could not
be found in Mustelus. These cells are considered
to be a type of stellate cell unique to the Scyllium
ventral lobe.

The cavity boundary cells in Mustelus often
showed a number of periluminal vesicles with

1070

A. Chiba and Y. Honma

flocculent content and contained a varying number
of vacuoles and lysosome-like bodies in their cyto¬
plasm. These aspects of the cells imply a capacity
of absorption or transport of substances from the
cavities into the cytoplasm and vice versa. In
addition, it should be pointed out that microvilli,
cilia, junctional complex and glycocalyx in the
cavity boundary cells are common characteristics
of epithelial cells, strongly reflecting their deriva¬
tion from stomodeal epithelium in the embryo [5,

9]-

The stellate cells of Mustelus showed a clear sign
of phagocytotic activity. In addition, amorphous
material and membranous structures in the inter¬
cellular spaces of the parcenchyma may be evi¬
dence of the phagocytotic activity of the stellate
cells. It is also suggested that the intercellular
spaces function as a channel for metabolites to be
stored in or transported into the hypophysial cavi¬
ties or into the blood capillaries. To date, no
similar activity of the stellate cells has been noted
in elasmobranchs [2-4, 18]. However, Perryman
[14] demonstrated that the stellate cells of the
anuran hypophysis function as phagocytes both in
vitro and in vivo. Furthermore, Bage and Fer-
nholm [10] have suggested that the stellate cells in
the proadenohypophysis of the river lamprey
transform into phagocytes during late periods of
gonadal maturation. Leatherland and Percy [1]
and Rawdon [11] have also pointed out a phagocy¬
totic role for the agranular cells in the teleost
hypophyses under certain experimental condi¬
tions. On the other hand, Benjamin [19] has an
opinion that agranular cells themselves in the
hypophysis of the teleost ( Pungitius ) are a source
of macrophages.

Recently, S-100 protein, a glial marker, was
immunocytochemically demonstrated in the stel¬
late cells or folliculo-stellate cells in mammalian
hypophyses [20-23]. In elasmobranchs, S-100
protein-like immunoreactivity was demonstrable
in the glial cells of the hypothalamo-hypophysial
region and also of other brain regions, but none of
the epithelial cells in the hypophysis were stained
with bovine S-100 protein antiserum (Chiba et al. ,
in preparation). The most probable explanation
for this negative result is that the S-100 protein¬
like molecule is lacking in the shark adenohypo¬

physis.

Accordingly, the present morphological data
suggest structurally and metabolically supportive
functions for the agranular cells in the selachian
hypophysis.

REFERENCES

1 Leatherland, J. F. and Percy, R. (1976) Structure of
the non-granulated cells in the hypophyseal rostral
pars distalis of cyclostomes and actinopterygians.
Cell Tissue Res., 166: 185-200.

2 Knowles, F., Vollrath, L. and Meurling, P. (1975)
Cytology and neuroendocrine relations of the pitui¬
tary of the dogfish, Scyliorhinus canicula. Proc. R.
Soc. Lond. B, 191: 507-525.

3 Alluchon-Gerard, M. J. (1978) Ultrastructure of
the dogfish adenohypophysis. Cell Tissue Res., 193:
139-154.

4 Alluchon-Gerard, M. J. (1978) Electron micro¬
scopic study of post-thyroidectomy changes in the
ventral lobe of the adenohypophysis in very young
spotted dogfish {Scy Ilium canicula , Chondrich-
thyes). Gen. Comp. Endocrinol., 36: 585-597.

5 Lagios, M. (1973) Follicle boundary cells in the
adenohypophysis of the chondrostean and holostean
fishes: an ultrastructural study of their relationship
to the follicular lumen, to endocrine cells and to the
hypophysial cleft. Gen. Comp. Endocrinol., 20:
362-376.

6 Saito, Y. and Tanaka, Y. (1980) Glutaraldehyde
fixation of fish tissues for electron microscopy. J.
Electron. Microsc., 29: 1-7.

7 Vila-Porcile, E. (1972) Le reseau des cellules folli-
culo-stellaires et les follicules de l’adenohypophyse
du rat (pars distalis). Z. Zellforsch., 129: 328-369.

8 Yoshimura, F., Soji,T., Sato, S. and Yokoyama,
M. (1977) Development and differentiation of rat
pituitary follicular cells under normal and some
experimental conditions with special reference to an
interpretation of renewal cell system. Endocrinol.
Japon., 24: 301-305.

9 Alluchon-Gerard, M. J. (1971) Types cellulaires et
etapes de la differenciation de l’adenohypophyse
chez l’embryon de roussete {Scy Ilium canicula ,
Chondrichthyens). Z. Zellforsch., 120: 525-545.

10 Bage, G. and Fernholm, B. (1975) Ultrastructure of
the proadenohypophysis of the river lamprey, Lam -
petra fluviatilis, during gonad maturation. Acta
Zool. (Stockh.), 55: 25-45.

11 Rawdon, B. B. (1978) Ultrastructure of the non-
granulated hypophysial cells in the teleost Pseudo -
crenilabrus philander (Hemihalochromis philander ),
with particular reference to cytological changes in
culture. Acta Zool. (Stockh.), 59: 25-33.

Agranular Cells in Shark Adenohypophysis

1071

12 Yamamoto, T., Kataoka, K., Ochi, J. and Honma,
Y. (1982) The fine structure of the follicle-forming
cells in the adenohypophysis of the ironfish, a natu¬
ral hybrid between the Funa and the goldfish (Tele-
ostei, Cyprinidae). Arch. Histol. Japon., 45: 355-
364.

13 Cardell, R. R. (1969) The ultrastructure of stellate
cells in the pars distalis of the salamander pituitary.
Am. J. Anat., 126: 429-456.

14 Perryman, E. K. (1983) Stellate cells as phagocytes
of the anuran pars distalis. Cell. Tissue Res., 231:
143-155.

15 Forbes, M. S. (1972) Fine structure of the stellate
cell in the pars distalis of the lizard, Anolis caro-
linensis. J. Morphol., 136: 227-246.

16 Tseng, M. T. and Yntema, C. L. (1976) Fine struc¬
ture of the chromophobe in the pars distalis of
common snapping turtle, Chelydra serpentia. Cell
Tissue Res., 166: 235-240.

17 Shiotani, Y. and Tanaka, Y. (1980) An electron
microscopic study on stellate cells in the rabbit
adenohypophysis under various endocrine condi¬
tions. Cell Tissue Res., 213: 237-246.

18 Mellinger, J. (1962) Cytologie hypophysaire de Scy-
liorhinus canicula (L.) et d’autres poissons elasmo-

branches. Microscopie optique et microscopie elec-
tronique. C. R. Acad. Sci. (Paris), 255: 2294-2296.

19 Benjamin, M. (1981) The origin of pituitary cysts in
the rostral pars distalis of the nine-spined stickle¬
back, Pungitius pungitius L. Cell Tissue Res., 214:
417-430.

20 Cocchia, D. and Miani, N. (1980) Immunocyto-
chemical localization of the brain specific S-100
protein in the pituitary gland of the adult rat. J.
Neurocytol., 9: 771-782.

21 Nakajima,T., Yamaguchi, H. and Takahashi, K.
(1980) S-100 protein in folliculo-stellate cells of the
rat pituitary anterior lobe. Brain Res., 191: 523-
531.

22 Shirasawa, N., Kihara, H., Yamaguchi, S. and
Yoshimura, F. (1983) Pituitary folliculo-stellate
cells immunostained with S-100 protein antiserum
in postnatal and thyroidectomized rats. Cell Tissue
Res., 231: 235-249.

23 Girod,C., Trouillas, J., Raccurt, M. and Dubois,
M. P. (1986) Immunocytochemical localization of
S-100 protein in stellate cells (folliculo-stellate cells)
of the adenohypophysis in the monkey Macaca irus
and Cercopithecus aethiops. Cell Tissue Res., 246:
237-242.

* -. H‘ „

.

'

ZOOLOGICAL SCIENCE 5: 1073-1081 (1988)

© 1988 Zoological Society of Japan

Interpopulation Variations in Clutch Size and Egg Size
in the Japanese Salamander, Hynobius nigrescens

Hisashi Takahashi and Hisaaki Iwasawa

Biological Institute, Faculty of Science, Niigata
University, Niigata 950-21, Japan

ABSTRACT — Interpopulation variations in egg size and clutch size in Hynobius nigrescens were studied
in 32 populations from various breeding ponds in the Niigata area. As the altitude increased, the date of
oviposition was delayed. However, the dates of oviposition in low elevation populations were varied
independent of the altitude. The average number of eggs in an egg sac ranged from 19.5 to 70.2, and the
average egg diameter ranged from 2.17 to 2.92 mm among the populations. Number and size of eggs
among the populations were varied clinally. As the altitude increased, or as the date of oviposition was
delayed, the average number of eggs decreased. The average egg size increased with the altitude.
However, the egg size decreased with the delay in the date of oviposition among 16 lowland populations.
The average total egg volume decreased with the delay in the date of oviposition. Among montane
populations, however, the total egg volume showed a tendency to increase with the altitude. From the
results of multiple regression analyses, the variations in the number of eggs and total egg volume were
explained by the variations in the altitude and the date of oviposition. It was conceivable that the
altitude and the date of oviposition were factors of the variations independent of each other. On the
other hand, the variation in egg size was explained by the variations in the altitude, latitude and
population size.

INTRODUCTION

Salthe [1] and Salthe and Duellman [2] showed
that the interspecific variations in egg size and
clutch size have been found in amphibians, and
proposed that the variations in clutch size and egg
size were related with the difference in the repro¬
ductive mode which might be induced by drastic
environmental changes. However, the variations
in egg size and clutch size have been already
recognized at an intraspecific level in some am¬
phibians [3-6]. These variations were seen in a
different environment. Berven [6] compared mon¬
tane and lowland populations of Rana sylvatica ,
and suggested that the variations in clutch size and
egg size might be due to a different genetic basis.
If the intraspecific variations in egg size and clutch
size correspond with a given environment, the
pattern of the variation in egg size and clutch size
will change with the environment. Most previous

Accepted February 17, 1988
Received November 4, 1987

studies on intraspecific variations in clutch size and
egg size in amphibians have been carried out on
small numbers of populations. It is conceivable,
however, that the variations in egg size and clutch
size are caused by several complex factors. Thus,
comparison among many populations may be a
more effective approach in order to clarify the
factors involved. Moreover, in order to under¬
stand how and why the interspecific variations in
clutch size and egg size occur, we need to know the
intraspecific variations in various environment.

In this paper, data on the interpopulation varia¬
tions in the number and size of eggs at various
altitude and dates of oviposition in a primitive
salamander, Hynobius nigrescens, are presented
and analyzed. Variations in the number of eggs
with altitude are known to some extent in this
species [7].

MATERIALS AND METHODS

The present study was carried out in the Niigata
area in 1986 sampling 32 populations of Hynobius

1074

H. Takahashi and H. Iwasawa

nigrescens (Fig. 1). The breeding ponds of the
populations were located at 5 to 2100 m elevation
and 37° to 38° North Latitude (Table 1). Sixteen
out of the 32 populations were massed at 200 m
elevation and lower locations. It seems that the
variations in reproductive traits among these 16
populations were not induced by the altitude. So
these populations were classified into one group
and named lowland populations. The other 16
populations were named montane populations.

This species breeds in lentic water, and one
female lays a pair of egg sacs. Thus, the mean
number of eggs in an egg sac was used to substitute
for the clutch size in the present paper. The mean
number of eggs in an egg sac in each population
was calculated from the number of eggs in 20 egg
sacs.

Embryos at the blastula or gastrula stage, spher¬
ical and egg-like in appearance, were measured
under a stereo-microscope to the nearest 0.01 mm.
These embryos gradually increase in size with the
advance of development, so the initial (uncleaved)
egg diameter was calculated from the embryo size
using the conversion factors established for this

Fig. 1 . Map of Niigata Prefecture showing sample sites.
Open circle: lowland population, Solid circle: mon¬
tane population.

species [8]. Data on the mean egg size in each
population were taken from the mean egg di¬
ameter in 10 egg sacs. The mean egg diameter in
each egg sac was based on 10 specimens taken
from the central part of the egg sac. Data on the
mean total egg volume in each population are
based on the total egg volume (egg volume X
number of eggs in an egg sacX2) in 10 egg sacs.
The egg volume was calculated from the diameter
using the equation for sphere volume.

Population size was estimated by counting the
number of egg sacs, at the end of the breeding
season. The dates of oviposition were estimated
from the embryonic stages and the degree of
swelling of the egg sacs, while the mean date of
oviposition in each population was calculated from
the dates of oviposition of the egg sacs that were
used for the measurment of egg size. For statistical
analyses, the dates of oviposition must be quanti¬
tated, so the date of oviposition in the first breed¬
ing population was expressed as 1, and dates in
other populations were expressed as 1 plus the
number of days after 1.

Statistical analyses were performed using NEC
statistics program package (ST ATP AC). The
significance of egg size difference among whole
populations, among lowland and among montane
populations were analyzed by the Kruskal-Wallis
test. Spearman rank correlations were employed
to test relationships between two variables. The
multiple regression technique was used to explain
the variations in egg number, egg size and total egg
volume.

RESULTS
1. Breeding period

The relationship between the breeding period
and altitude for each population is illustrated in
Figure 2. The breeding period in each population
closely corresponded with the thawing of the snow.
Populations examined in the present study in¬
cluded various types, from one of the earliest
breeding populations to one of the latest breeding
ones in Niigata area. The date of oviposition
varied up to 4 months among all the populations
examined, and up to 2 months among just the

Reproductive Variation in Salamander 1075

Table 1. Characters in each population

Population

number

Altitude

(m)

Latitude

Mean
date of
oviposition

Number

of

egg sacs

Condition of breeding pondx)

Lowland

1

population

20

37°44'

Mar. 21

48

Shallow pool with constant inflow

2

5

37°48'

Mar. 25

164

Artificial pond near the coast

3

170

37°42/

Mar. 28

72

Farm pond

4

180

37°44'

Apr. 3

ca.1000

Artificial pond

5

60

37°43'

Apr. 3

190

Artificial pond with constant inflow

6

100

37°37/

Apr. 8

130

Large pond with constant inflow

7

40

37°41'

Apr. 14

ca.200

Small artificial pond

8

60

37°08'

Apr. 21

54

Farm pond

9

140

37° 16'

Apr. 24

20

Farm pond

10

100

37°30'

Apr. 24

202

Shallow farm pond

11

150

37°40'

Apr. 25

20

Abandoned rice field

12

140

37°32'

Apr. 29

ca.1500

Large farm pond

13

150

37°40/

May 1

110

Artificial pond

14

140

37°21'

May 3

71

Abandoned rice field

15

80

37° 13'

May 3

84

Farm pond

16

130

37°01'

May 9

246

Large pond owed to spring water

Montane

17

population

300

37°25'

May 4

52

Small farm pond

18

270

37°25'

May 4

90

Abandoned rice field

19

330

37°05'

May 12

32

Farm pond

20

250

37°15'

May 12

136

Farm pond

21

250

37°09'

May 22

200

Farm pond

22

740

37°54'

May 25

30

Small pond

23

520

37° 16'

May 29

594

Middle-sized pond

24

1012

37°37'

June 1

1050

Small pond

25

830

37°01'

June 10

190

Middle-sized pond

26

650

37°39'

June 12

ca.1800

Middle-sized pond

27

630

37°17'

June 13

482

Middle-sized pond

28

1060

37°54'

June 16

ca.800

Middle-sized pond

29

680

36°58/

June 25

1112

Middle-sized pond

30

2080

36°53'

July 9

660

Small ponds in marshy plain

31

2010

36°54'

July 9

714

Small ponds in marshy plain

32

2100

36°53'

July 9

208

Large pond in marshy plain

Small pond: under 25 m2, Middle-sized pond: 25-200m2, Large pond: over 200 m2

lowland populations. In general, the duration of
breeding was longer in lowland (the longest was
about 2 months) and shorter in montane popula¬
tions (the longest was 2 weeks). So far as lowland
populations are concerned, there was no apparent
relationship between the date of oviposition and
altitude (n — 16, ^=0.324). In these cases, the

date of oviposition was determined by the time of
thawing. However, among montane populations
(n=16, rs— 0.847, P<.01), and among all the
populations examined (n=32, r5=0.889, P<.01),
a significant correlation between the date of ovi¬
position and altitude was recognized statistically.

Among the sampling sites in the present study,

1076

H. Takahashi and H. Iwasawa

2200

2000

_ 1000
6

LU

D

Z>

<

500

MAR APR MAY JUN JUL

DATE OF OVIPOSITION

Fig. 2. Breeding period in each population. Open circle: mean date of oviposition used for statistical analysis. Solid
line shows the case that egg sacs had been deposited at first visit to breeding pond. Dash shows beginning or end
of breeding period in each population.

the disappearance of snow was delayed in more
southern lowland populations than northern low¬
land ones (n = 16, 0.76, P<.01).

2. Relationships of clutch size, egg size and total
egg volume to altitude and date of oviposition

Variations in the egg number, egg size and total
egg volume with changes in altitude and date of
oviposition are shown in Figure 3. The maximum
mean number of eggs was 70.2 in population No.
2, and the minimum was 19.5 in No. 26. Rela¬
tionships between the mean number of eggs and
altitude, and between the mean number of eggs
and the date of oviposition are shown in Figure 3a,
b. There was a negative correlation between the
mean number of eggs and altitude among montane
populations and among whole ones. However,
there was no correlation between the mean num¬
ber of eggs and altitude among lowland popula¬
tions. The mean number of eggs varied conspi¬
cuously among lowland populations, while varied
little among higher altitude ones. The mean
number of eggs was significantly correlated with

the date of oviposition; and decreased as the date
of oviposition was postponed.

The maximum mean egg diameter was 2.92 mm
(egg volume: 13.03 mm3) in population No. 26,
and the minumum was 2.17 mm (egg volume: 5.35
mm3) in No. 16. Egg size significantly differed
among all the populations examined, among low¬
land and among montane populations. The rela¬
tionship between mean egg size and altitude is
shown in Figure 3c. There was a positive correla¬
tion between mean egg size and altitude among all
the populations and among the montane ones.
The relationship between the mean egg size and
the date of oviposition is shown in Figure 3d.
Among lowland populations, the mean egg size
decreased as the date of ovipositions was delayed.
Among montane populations, however, the mean
egg size increased as the date of oviposition was
delayed.

The mean total egg volume was negatively
correlated with altitude and with the date of ovi¬
position among all the populations examined (Fig.
3e, f). Particularly among lowland populations,

Reproductive Variation in Salamander

1077

70

r °

a

r °

K

o rs i - 0-140

o : rSr- 0-670**

CL

LLI

CD

60

• : rs=- 0-591*
o+# rs = - 0-798**

• : rsr-0 641**
o+*: rs = - 0 888**

3

50

o o

o

o

_ L

- 8 °

3

°°a9

° O ° ° O

O

0

40

" <8

<P o

LLI

30

20

o o

o

1

r. •

1 u 1

••

O g

% •

° •

L • • • .

•

till

1

|

3.0

o :

rs= -0.144

•

c

o:rs, -0-768** •

d

E

2-8

• :

rs 1 0.776**

. •: rs = 0.723** •

E,

• *

•

•

t

LU

N

26

o

O o o

•

•

o •

00 o

•

C/5

0

2.4

'??

• •
•*

•

•

•

O O o •

°o (f • •

o °% • •

O

°o

•

° o •

LLI

2-2

•

o + *: rs = 0.390**

•

o+* : rs-0.230

o

1

l

1 n 1

o

iiii

i

1=

E

1300

- 0

e

r o

** f

UJ

o : rs =-0-263

O : rsr -0.909'

3

_l

1100

o

• : rs -0-435*
o+*: rs- -0.553**

• ■ is- 0-376

o o+#:rsr- 0.701**

o

o

o

>

900

" ° o o

" ° oo

0

0

LU

_l

700

O O

o

of

••

o o

/• . .

<

1-

o

500

cP

o

\ •
t *

*.* •
•

•

•

* • *

•

•

•

1—

300

1

L

•

l ■. i

1_ 1_ L 1_ l

I _ I _ I _ „_l _ _ 1 _ 1 _ L__ _ I _ l

0 500 1000 2000 APR MAY JUN JUL

ALTITUDE (m) DATE OF OVIPOSITION

Fig. 3. Variations in egg number, egg size and total egg volume with changes in altitude and spawning date. Open
circle: lowland population, Solid circle: montane population, *: significant at<.05, **: significant at<.01, a:
relationship between egg number and altitude, b: relationship between egg number and date of oviposition, c:
relationship between egg size and altitude, d: relationship between egg size and date of oviposition, e:
relationship between total egg volume and altitude, f: relationship between total egg volume and date of
oviposition.

mean total egg volume varied conspicuously, and a
highly negative relationship was observed between
the mean total egg volume and the date of oviposi¬
tion. However, among montane populations,
there was a positive correlation between the mean
total egg volume and altitude.

The relationship between the mean number of
eggs and egg size among populations is snown in
Figure 4. Among all the populations, there was no
significant correlation between egg number and
egg size (n = 32, rs= — 0.307), but, there was a
negative correlation between the mean number of
eggs and mean egg size among montane popula¬

tions (n = 16, rs= — 0.738, P<.01). Among the
lowland populations, a weakly (not significant)
positive correlation (n = 16, r5=0.413) was seen
between mean egg number and mean egg size.

3. Analysis of variations in the egg number, egg
size and total egg volume

As a result of simple correlation, the number of
eggs, egg size and total egg volume related in a
complex manner with the difference in altitude and
the date of oviposition. On the whole, the date of
oviposition was significantly correlated with alti¬
tude. Among lowland populations, however, the

1078

H. Takahashi and H. Iwasawa

90

70

60

a:

m 50

Z

Z>

z

o

o 40

UJ

30

20

l

12

o=s>

25

24

32

29.

30

2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3.0 3.1

EGG DIAMETER ( mm )

Fig. 4. Relationship between mean egg number and mean egg size. Vertical and horizontal lines:
standard deviations, Open circle: lowland population, Solid circle: montane population.

number of eggs, egg size and total egg volume
varied independently from altitude. For the above
reason we estimated that the altitudinal and breed¬
ing-seasonal factors are independent of each other
as far as the variation in the number of eggs, egg
size and total egg volume are concerned. In order
to analyze these variations from the viewpoints of
both the effect of altitude and the breeding season,
each variation was examined by the multiple re¬
gression analysis using altitude and the date of
oviposition as predictor variables (Table 2). These
regression equations were fairly effective in ex¬
plaining the variations in the mean number of eggs

and total egg volume. According to the results of
the test of significance in the partial regression
coefficient in Table 2, both the differences in alti¬
tude and the date of oviposition are significant in
regard to the variations in the number of eggs and
total egg volume among all the populations ex¬
amined. These regression analyses showed that
the mean number of eggs and mean total egg
volume increased with altitude, but decreased with
the postponement of ovipoition. These results also
showed that the altitude and the date of oviposi¬
tion are factors independent of each other.

Variation in egg size cannot be adequately ex-

Reproductive Variation in Salamander
Table 2. Results of multiple regression analyses

1079

Criterion

Multiple

Analysis of

variance

Constant

Predictor

Partial

Standard

Test of
significance
in coefficient

Standard

partial

variable

R

(xio-2)

variable

regression

error

regression

F-ratio

P

r-val

P

coefficient

coefficient

a) Analyses to explain each reproductive characteristic by altitude and date of oviposition

Egg number
(X 102)

0.857

40.4

<.001

64.7

Altitude

0.86

0.38

2.296

<.05

0.422

Date of
oviposition

-45.70

0.07

-6.476

<.001

-1.190

Egg size
(XlO3)

0.534

5.8

<.01

25.2

Altitude

0.19

0.08

2.404

<.05

0.726

Date of
oviposition

— 1.19

1.49

-0.800

>•1

-0.242

Total egg
volume

0.810

27.7

<.001

11.6

Altitude

0.30

0.08

3.881

<.001

0.813

Date of
oviposition

-9.56

1.45

-6.616

<.001

-1.386

b) Analysis to explain egg

size variation by altitude, latitude and population size

Egg size

0.787

15.2

<.001

-69.1

Altitude

0.17

0.04

4.619

<.001

0.657

(XlO3)

Latitude

0.25

0.07

3.724

<.001

0.503

(XlO3)

Population

0.11

0.04

2.562

<.05

0.322

size

plained by the difference in the altitude and the
breeding season (Table 2a). According to the
results of the test of significance in the partial
regression coefficient, altitudinal difference is sig¬
nificant in regard to the variation in egg size among
all the populations examined, but the difference in
the date of oviposition is not significant. In order
to analyze the factor of egg size variation, multiple
regression analysis was performed again to use
altitude, latitude and population size (number of
egg sacs) as predictor variables (Table 2b). This
regression was effective in explaining the variation
and the test of partial regression coefficient re¬
vealed that altitude, latitude and population size
are significant factors in egg size variation.

The results of this regression analysis showed
that mean egg size increased with altitude, latitude
and population size. Similar analyses were per¬
formed for the number of eggs and total egg
volume. However, latitude and population size
were not significant as the variables to explain the
variations.

Effectual results which were obtained by linear

regression models on the variations in the egg
number, egg size and total egg volume meant that
the variations in the egg number, egg size and total
egg volume were clinal variations which corre¬
sponded with gradual environmental changes.

DISCUSSION

1. Breeding seasonal and altitudinal effects on the
number of eggs and total egg volume

We collected many adult specimens of this spe¬
cies in several breeding ponds, and found no
evidence of food ingestion at the time. So the
active period of this species is estimated to be
between the ending of breeding and snowfall in
Niigata area. Therefore, delayed breeding shor¬
tens the active period of salamanders, and, as a
result, a decrease in the total amount of food
ingested throughout the year may occur. It is
conceivable that total ingestion is one of the fac¬
tors which restrict total egg volume. Fraser [9]
showed experimentally in Plethodon cinereus and

1080

H. Takahashi and H. Iwasawa

Kaplan [10] showed in Bombina orientalis that a
difference in the nutritional condition of the
female affected the process of oocyte growth.
Such a phenomenon may be occurring among
different populations of H. nigrescens. It seems
that the decrease in total egg volume is induced by
the nutritive deterioration that is brought about by
the delay of oviposition, and that the number of
eggs decreases to correspond with the decrease in
total egg volume.

It has been shown in several species that the
altitudinal difference induced variations in life
history traits which include variation in the age and
size at maturity [6, 11-14]. These variations in life
histrory traits had different patterns according to
circumstances. In Desmognathus ochrophaeus [11]
and Rana sylvatica [6] , sexual maturation occurred
later in montane than in lowland populations, and
the body size of females increased with the post¬
ponement of sexual maturation. In Rana pretiosa

[12] , sexual maturity was delayed remarkably in a
high elevation population compared with a low¬
land populaton, but the body size at maturity was
similar in both populations. On the other hand, it
was estimated that sexual maturity was delayed in
high elevation populations, and the body size of
females was smaller in high elevation populations
than in low elevation ones in Ambystoma gracile

[13] . These high elevation populations had a 4
months shorter active period than low elevation
ones. These reports showed that sexual maturity is
reached later in higher altitude populations than in
lower altitude ones, and that the growth rate of
most females in high altitude populations is re¬
duced by reason of the decrease in the length of
the active period in high altitude populations. In
northern populations of Plethodon glutinosus,
growth was greatly retarded owing to a longer
hibernating period in comparison with southern
populations which did not hibernate [15]. In
general, the clutch size and the total egg volume
are closely correlated with the body size of females
[1, 2, 16]. It is conceivable that the increase in the
number of eggs and total egg volume with altitude
in the multiple regression analysis is a reflection of
the relative increase in female body size resulting
from the postponement of sexual maturation.
However, if the growth of females is restricted by a

remarkable delay in the breeding season, the body
size of adult females is reduced inspite of the delay
of sexual maturation. After all, it seems that the
relationship between the two variables (the date of
oviposition and altitude) in multiple regression
equations for the number of eggs and total egg
volume means the relationship between the en¬
vironment effect in growth or vitellogenic process,
and the age of maturation of females.

2. Variation in egg size

It has been known that in some amphibian
species the egg size is closely related to the altitude
[3, 4, 6] or latitude [17, 18]. The present study
showed that egg size varied clinally with altitude
and latitude. It is thinkable that the egg size
variation is related to pond temperature in parallel
with altitude and latitude. However, in the breed¬
ing ponds examined in the present study, the
seasonal variation in water temperature within a
pond tended to be greater than geographical varia¬
tions among ponds (unpublished data). That is,
the variation in pond temperature in breeding and
hatching is mostly determined by the difference of
the date of oviposition among populations. So, we
cannot assert that the difference of water tempera¬
ture based on the altitude and latitude is one of the
factors inducing these altitudinal and latitudinal
egg size dines.

On the other hand, egg size was significantly
related with population size. It seems that the
number of egg sacs in one breeding pond is an
indicator of the degree of intraspecific competi¬
tion, of relative food density per larva and of larval
density in the breeding ponds. It is conceivable
that other pond-specific factors are also involved in
egg size determination. However, these factors
could not eliminate the altitudinal and latitudinal
egg size dines. Populations Nos. 1-5 and 7 were
located in the same heap of mountains and at a
similar altitude. Furthermore, these populations
had a similar breeding period. In spite of the
various environmental conditions in their breeding
ponds, mean egg size did not vary conspicuously
among populations.

3. Relationship among reproductive traits

Berven [6] discussed the relationships among the

Reproductive Variation in Salamander

1081

number of eggs, egg size, total egg volume, body
size and age at first reproduction. He proposed
that selection had acted on egg size, and that other
reproductive traits had evolved secondarily. Re¬
sults of the multiple regression analyses in the
present study showed that the number of eggs and
total egg volume were affected by the date of
oviposition. As stated above, it is estimated that
the date of oviposition affects the growth of female
and the vitellogenic process. Several studies have
shown that the differences in growth rate and body
size of amphibians were influenced by environ¬
ment [6, 19, 20]. It is conceivable that the body
size of female is one of important factors which
determine the number of eggs and total egg
volume. Thus, number of eggs and total egg
volume in each population appear to be largely
environmentally determined. On the other hand,
it is conceivable that egg size was not affected by
the date of oviposition, but varied in parallel with
altiudine and latitude. It seems that the egg size
varied in adaptation to the altitudinal and latitu¬
dinal changes in environment. From the negative
correlation between the number of eggs and egg
size among montane populations (Fig. 4), it is
suggested that high altitudinal populations pro¬
duce optimal sized eggs at the expense of the
number of eggs.

REFERENCES

1 Salthe, S. N. (1969) Reproductive modes and the
numbers and size of ova in the urodeles. Am. Midi.
Natur., 81: 467-490.

2 Salthe, S. N. and Duellman, W. E. (1973) Quan¬
titative constraints associated with reproductive
mode in anurans. In “Evolutionary Biology of the
Anurans”. Ed. by J. L. Vial, Univ. Missouri Press,
Columbia, Missouri, pp. 229-249.

3 Pettus, D. and Angleton, G. M. (1967) Compara¬
tive reproductive biology of montane and piedmont
chorus frogs. Evolution, 21: 500-507.

4 Kozlowska, M. (1971) Differences in the reproduc¬
tive biology of mountain and lowland common
frogs, Rana temporaria L. Acta. Biol. Crac., Ser.
Zool., 14: 17-32.

5 Woodward, B. D. (1982) Local intraspecific varia¬
tion in clutch parameters in the spotted salamander
( Ambystoma maculatum). Copeia, 1982: 157-160.

6 Berven, K. A. (1982) The genetic basis of altitudi¬
nal variation in the wood frog Rana sylvatica. I. An
experimental analysis of life history traits. Evolu¬
tion, 36: 962-983.

7 Kurasawa, J. and Iwasawa, H. (1977) Correlation
between the number of eggs deposited in the egg
sacs and the altitude of the spawning place in the
salamander Hynobius nigrescens. Jpn. J. Herpetol.,
7: 27-31. (in Japanese with English summary)

8 Takahashi, H. and Iwasawa, H. (1988) Changes in
the size of egg-form embryos with the advance of
development in the salamander Hynobius nigres¬
cens. Jpn. J. Herpetol., 12: 124-126.

9 Fraser, D. F. (1980) On the environmental control
of oocyte matulation in a plethodontid salamander.
Oecologia, 46: 302-307.

10 Kaplan, R. H. (1987) Developmental plasticity and
maternal effects of reproductive characteristics in
the frog, Bombina orientalis. Oecologia, 71: 273-
279.

11 Tilley, S. G. (1973) Life histories and natural selec¬
tion in populations of the salamander Desmognathus
ochrophaeus. Ecology, 54: 3-17.

12 Licht, L. E. (1975) Comparative life history fea¬
tures of the western spotted frog, Rana pretiosa,
from low- and high-elevation populations. Can. J.
Zool., 53: 1254-1257.

13 Eagleson, G. W. (1976) A comparison of the life
histories and growth patterns of the salamander
Ambystoma gracile (Baird) from permanent low-
altitude and montane lakes. Can. J. Zool., 54: 2098-
2111.

14 Tilley, S. G. (1980) Life histories and comparative
demogrphy of two salamander populations. Copeia,
1980: 806-821.

15 Highton, R. C. (1962) Geographic variation in the
life history of the slimy salamader. Copeia, 1962:
597-613.

16 Kaplan, R. H. and Salthe, S. N. (1979) The
allometry of reproduction: an empirical view in
salamanders. Am. Natur., 113:671-689.

17 Moore, J. (1949) Geographic variation of adaptive
characters in Rana pipiens Schreber. Evolution, 3:
1-24.

18 Matsui, M. (1987) Geographic variations in repro¬
ductive traits and the factors of the variations. In
“Biology of Toad”. Ed. by A. Urano and K. Ishi-
hara, Shokabo, Tokyo, pp. 19-31. (in Japanese)

19 Berven, K. A. (1982) The genetic basis of altitudin¬
al variation in the wood frog Rana sylvativa II. An
experimental analysis of larval development. Oeco¬
logia, 52: 360-369.

20 Harris, R. N. (1984) Transplant experiments with
Ambystoma larvae. Copeia, 1984: 161-169.

■

ZOOLOGICAL SCIENCE 5: 1083-1088 (1988)

© 1988 Zoological Society of Japan

Thyroid Status and Ambient Temperature as Influences
on Weaning in Young Mice

Lee A. Meserve and Mary Ann R. Gonzalez

Department of Biological Sciences, Bowling Green State
University, Bowling Green, OH 43403-0212, USA

ABSTRACT — Depressed thyroid status has been long known to delay normal mammalian development
of physiological and behavioral processes. Investigations of young rats made hypothyroid by chemical
goitrogenesis have demonstrated that such a treatment disrupts normal weaning behavior which can be
restored by thyroxine (T4) injection. The present study was done to determine the influence of
chemically-induced hypothyroidism on weaning in the mouse, an experimental rodent closely related to,
but somewhat metabolically different from the rat. Hypothyroidism was instituted by incorporation of
thiouracil (0.25%) into the maternal diet from the day of mating. Daily from day 6 of life, half the pups
in each litter born to euthyroid or hypothyroid dams were injected with a physiological dose (50 ng/g
body wt) of T4, with the remaining pups receiving vehicle. At 17, 21, or 24 days of age, pups were
allowed to indicate a preference between maternal food or non-maternal (chow) food choices at either
room (24°C) or incubator (35°C) temperature. After spending the night following preference testing in
the home cage, stomach contents of pups were analyzed for proportions of milk and chow. Thyroxine
injection of euthyroid pups did not significantly modify body weight, circulating T4 levels, or behavioral
progression to weaning, suggesting the dose to be physiological. Hypothyroidism effectively eliminated
weaning behavior within the normal time frame, but the behavior was restored by T4 injection.
Elevation of ambient temperature generally increased pup preference for a non-maternal food source at
21 and 24 days, however the increases were small and did not represent accelerated weaning in
euthyroid pups nor normal weaning in hypothyroid mice. In general, the influence of thyroid status on
normal weaning in mice appears to be very similar to that reported in rats.

INTRODUCTION

The influence of insufficiency of thyroid hor¬
mone on pre- and postnatal developmental pro¬
cesses has been known for some time [1]. Howev¬
er, details regarding the presence or severity of
specific developmental deficiencies remain to be
investigated in hypothyroid experimental animals.
The normal process of weaning from total reliance
upon mother’s milk has been well established in
the rat [2], with the first appearance of solid food
in the stomach occurring between days 15-17 [3],
and the disappearance of milk between days 24-28
[4]. Recent studies have also found this precisely
timed developmental process to be delayed by
depressed thyroid status [5, 6] which may result in
part from disrupted temperature control mecha¬

nisms which occur because of hypothyroidism [5].
To our knowledge, similar studies have not been
done in the laboratory mouse, a species in which
adult body weight is about 10% of that in the rat
[7] and in which metabolic rate is about twice that
of the rat [8]. Thus, the present study was carried
out to determine whether hypothyroidism delayed
weaning in the mouse, and if so, whether replace¬
ment of thyroxine (T4) or elevation of ambient
temperature would ameliorate this delay. These
determinations were made by conducting dietary
preference tests after fasting at 17, 21 or 24 days of
age (bracketing the time of normal weaning in
rats), and by measuring the proportion of stomach
contents comprising milk and solid food the next
day, after pups had spent the night with the mother
mouse.

Accepted January 11, 1988
Received December 16, 1987

1084

L. A. Meserve and M. A. R. Gonzalez

MATERIALS AND METHODS

Mice derived from the Swiss-Webster strain
were mated, with successful matings being deter¬
mined by presence of a vaginal plug. On the first
day of pregnancy, females were isolated in opaque
polystyrene cages under controlled temperature
(22 + 10°C) and lighting (12L: 12D) conditions,
and were fed either standard laboratory chow or
Lab Blox Mash (Wayne Animal Foods Division,
Chicago, Illinois) to which 0.25% by weight 2-
thiouracil (Sigma Chemical Company, St. Louis,
Missouri) had been added to induce hypothyroid¬
ism. The dietary regimen of each pregnant or
lactating mouse remained in the same through
testing of young around the time of weaning.

On day 5 after birth of each litter, pups in excess
of a total litter size of 8 were removed from the
nest. The study population consisted of 112 con¬
trol pups from 18 different litters, and 111 chemi¬
cally hypothyroid pups from 27 litters. Both male
and female animals were used in each experiment.
On day 6 of life, and daily thereafter, half the pups
in each litter received a physiological dose of
thyroxine (T4, 50 ng/g body wt) [5] by sub¬
cutaneous injection, and the remaining pups were
administered an equivalent volume of the hor¬
mone injection vehicle (0.3 mM NaOH).

The weaning status of all pups was tested using
two measures; a dietary preference test in the
absence of the natural mother, and a determina¬
tion of stomach contents the morning after the
dietary preference test, the pups having spent the
night in the cage with the natural mother. The
dietary preference test essentially followed the
regime used by Blake and Henning [5] with young
rats. Seventeen, 21, or 24 day old mouse pups
were incubated (35±2°C) and fasted with avail¬
able drinking water for 10 hr, after which a vehi¬
cle-injected and T4-injected pair were placed mid¬
way between chow mash, chow pellets, and a
lactating Nembutal-anesthetized foster dam with
nipples exposed at either room (24°C) or incubator
(35°C) temperature. Behaviors were catalogued
by observation at 10 sec intervals over a total test
time of 5 min, with the summed percentage time
spent eating mash or pellets used to indicate
weaning. After dietary preference testing, pups

were returned to the maternal cage overnight. The
next morning (day 18, 22 or 25) pups were decapi¬
tated and blood was collected from the severed
neck vessels. Serum was extracted by centrifuga¬
tion and frozen at — 20° C for future estimation of
T4 concentrations. After determination of carcass
weight to the nearest 0.1 g, stomachs were rapidly
excised and iced for 20 min to coagulate milk [5],
Total stomach contents were removed and
weighed, then the chow component was weighed
alone. The percentage of total stomach contents
represented as chow was used as an index of
weaning.

Serum T4 concentrations were determined to
measure the effectiveness of thiouracil in depress¬
ing thyroid status, and the degree of T4 replace¬
ment affected by injecting the hormone. Serum
samples were pooled with regard to maternal diet
(with or without thiouracil) and pup injection
(vehicle or T4), but without regard to age, since T4
levels of normal mice are fairly consistent across
this age range [9]. T4 concentrations were deter¬
mined using radioimmunoassay kits generously
donated by Micromedic Systems, Inc. (Horsham,
Pennsylvania). T4 sensitivity of these kits was 0.3
ng/dl, having a within assay variation of 6.5% and
a between assay variation of 7.1%.

Data were statistically analyzed for main effects
of thyroid status and ambient temperature upon
dietary preference, and of thyroid status on stom¬
ach contents and body weight, by analysis of
variance, with individual comparisons of means
performed using Tukey’s test [10]. Differences in
serum T4 concentrations between vehicle- and
T4-injected pups were evaluated with Student’s
t-test. In all cases significance was ascribed to
P<0.05.

RESULTS

The linear growth of euthyroid pups between
days 18-25, expressed by the relationship Y— 2. OX
+ 4.7 (slope=2.0 g/day; r=0.87), was not altered
by daily injection of 50 ng T4/g body weight,
suggesting that this dosage represented a truly
physiological level which had largely turned off
endogenous thyroid hormone production. This
supposition is substantiated by the levels of circu-

Thyroid and Weaning in Mice

1085

lating T4 which, while slightly elevated in T4-
injected normal animals, was not significantly so
(Table 1). Furthermore, administration of this
dosage of T4 increased circulating hormone levels
to over 75% of normal in chemically hypothyroid
pups, and elevated severely depressed body
weights to normal levels.

During the early stages of the weaning process at
17 days of age neither depression of thyroid status
with thiouracil nor administration of T4 by injec¬
tion to euthyroid pups altered the amount of time
spent with the mother or alternate food choices,
regardless of the ambient temperature. In all
cases, pups spent 2% or less of the test time with
food choices alternative to the anesthetized foster
dam. However, T4 administration to thiouracil-fed
pups significantly elevated the time spent with
non-maternal food options to above 3%, and this
influence was augmented to greater than 5% by
elevating environmental temperature (P<0.05,
Fig. 1). After pups spent the night with the natural
mother, analysis of stomach contents revealed that
young of euthyroid mice had begun to consume
chow at this age, at a percentage of total intake
unmodified by T4 injection (approximately 5% of
stomach contents). On the other hand, stomach
contents of 18 day old pups from thiouracil-fed
dams contained no chow, an observation which
was not altered by administration of exogenous T4
(Fig. 2).

By 21 days of age the depressive influence of
thiouracil-induced hypothyroidism on develop¬
ment of weaning behavior had become obvious.

Pups in this group spent less than one-tenth the
time with non-maternal food as did normal pups
(with or without T4 injection) or thiouracil-fed
animals injected with T4 (Fig. 1). As at 17 days,
elevation of ambient temperature only increased
the amount of time spent with non-maternal food
by hypothyroid animals injected with T4. The
general trend was for all incubated pups to spend
more time away from the foster dam than pups at
room temperature, except by normal pups given
T4. On day 22, chow represented over 50% of
stomach contents in normal pups, and was not
influenced by T4 injection. While only traces of
chow were found in the stomachs of vehicle-
injected hypothyroid pups (less than 2% of total
contents), T4 injection resulted in the proportion
of chow approximating that seen in normal animals
(Fig. 2).

For all intents and purposes, the weaning pro¬
cess was completed by 24 days of age in normal
animals, with an average of over 90% of dietary
preference test time spent with food sources other
than the foster dam. This proportion could be
driven to nearly 95% by increasing environmental
temperature to 35°C. The completeness of wean¬
ing was further substantiated by the presence of
100% chow in the stomachs of these pups on day
25 (Fig. 2). Hypothyroidism continued to delay
weaning behavior, with about 5% of the test time
being spent with non-maternal food sources, but at
this age increased ambient temperature significant¬
ly elevated time spent with chow to about twice the
average (Fig. 1). Nonetheless, only trace amounts

Table 1. Body weights and relative circulating thyroxine (T4) concentrations of normal and
thiouracil-fed mice with or without T4 injection

Treatment

Injection

Body Weight (g)

T4 levels4)

(% of control)

18 days

22 days

25 days

Lab Chow

Vehicle

6.8 + 0. l(18)1)a

8.2±0.1(18)d

10.2±0.3(20)f

100.0±7.5(12)h

t43)

6.4 + 0. l(18)a

7.6±0.2(18)d

10.5±0.4(20)f

117.3±9.9(12)h

Lab Chow

Vehicle

4.8±0.2(18)b

6.3+0.2(19)®

8.2±0.2(19)g

24.0 + 2.0(12)*

plus Thiouracil2)

t43)

6.1±0.2(17)c

7.8±0.2(19)d

10.4±0.2(19)f

77.9±4.0(12)j

x) Mean±SEM(n). Values in the same column with the same superscript are not significantly different (P
>0.05).

2) 0.25% of maternal diet by weight from day 1 of pregnancy.

3) 50 ng/g body weight

4) 100% T4 in controls represents a mean concentration of 10 ^g/dl.

1086

L. A. Meserve and M. A. R. Gonzalez

Days of Age

Fig. 1. Percent of dietary preference test time that 17, 21, or 24 day old mouse pups spent with a food choice other
than an anesthetized foster dam. A. Vehicle (0.2 mM NaOH)-injected euthyroid pups. B. Littermates of A
injected daily with 50 ng TVg body weight from day 6 to day of testing. C. Vehicle-injected chemically
hypothyroid pups (0.25% dietary thiouracil from conception). D. Littermates of C injected as in B. Each data
point used to draw the curves is the mean ± standard error of 17-20 pups. At points without bars of error was
smaller than the symbol. Hatched bars represent dietary preference of half these animals at room temperature
(24°C) and open bars represent preference of the remaining pups at incubator temperature (35°C). Note that in
only one case (B., 21 day) did the room temperature value excede the incubator value.

of chow were present in the stomachs of these
animals on day 25. Twenty-four day old hypothy¬
roid pups injected with T4 did not spend an
obviously greater amount of time with chow than
they had spent at 21 days (Fig. 1), but their
stomachs contained 100% chow the next day after
spending the night with their mothers.

DISCUSSION

The experimental procedures performed with
mice in the present study were designed to parallel
as closely as possible the portion of an earlier
investigation concerning thyroid status and wean¬
ing progression done by Blake and Henning in rat
pups [5]. Procedural deletions from the rat study

Thyroid and Weaning in Mice

1087

Days of Age

Fig. 2. Percent of total stomach contents as animal
chow found in the same pups given the dietary
preference test, after spending the night in the home
cage with the natural dam. Circles: pups of euthy¬
roid dam. Triangles: pups of thiouracil-fed dams.
Closed symbols: vehicle-injected pups. Open sym¬
bols: T4-injected pups.

included the determination of rectal temperature
and of incisor length because of the small size of
mouse pups. In addition to those measurements
made in rats, we included observations of the
influence of administered T4 on weaning of pups
from euthyroid dams, and a determination of
circulating T4 levels at the end of the experiment.
It was deemed necessary to include these aspects
to ensure that the T4 dosage of 50 ng/g body weight
represented a physiological dose in mice as well as
rats. Since T4 injection of euthyroid pups did not
significantly modify body weight gain or circulating
T4 levels (Table 1) and since weaning measures of
these animals closely paralleled those of vehicle-
injected littermates (Figs. 1 and 2), the dosage of
exogenous T4 was considered physiological. It is of
interest that T4 injection of 21 day old euthyroid
pups was the only treatment to result in pups
spending more time with non-maternal food
choices at room temperature than at incubator

temperature. While the difference was not statisti¬
cally significant, it may support a previously re¬
ported acceleration of locomotor development
with T4 augmentation in mice [11]. The appro¬
priateness of this T4 dosage was further supported
by replacement injection at this level returning the
growth rate of chemically hypothyroid mice to
normal and significantly improving severely de¬
pressed circulating T4 levels (Table 1).

As reported in rats [5], chemically-induced
hypothyroidism severely retarded the normal
progression of weaning in the present study,
whether assessed by dietary preference test or by
analysis of stomach contents. Indeed, the values
reported here for percent time spent with a non-
maternal food source during dietary preference
testing (Fig. 1, C and D) are nearly identical, at
each age and environmental temperature used to
test chemically hypothyriod and hypothyroid T4-
injected mice, with those measurements made by
Blake and Henning [5] in rats.

Investigators using the hamster to study the
process of weaning have suggested important
maternal influences upon its development [12].
Although there may be informational cues pro¬
vided by the mother mouse as well, these do not
seem to be the primary motivation to weaning
since dietary preference of normal pups shifted
with age to a greater proportion of non-maternal
food choice, even though the maternal food source
was an immobile anesthetized foster dam. Similar
conclusions have been drawn using rats [5]. Fur¬
thermore, elevation of environmental temperature
in an attempt to provide the pups with the body
temperature normally maintained by the maternal
sharing of body heat failed to normalize weaning in
hypothyroid mice or to accelerate its progression
to any great degree in normal mice.

Analysis of stomach contents found T4 injection
of hypothyroid pups to have no effect on the
percent chow consumed at 18 days of age, but to
return chow consumption toward normal at 22
days and completely to normal at 25 days when
stomachs of treated pups contained chow alone as
did those of normal mice (Fig. 2). This is an
interesting finding in light of the dietary preference
results which suggest that T4-injected hypothyroid
mice spent 45% of the test time with non-maternal

1088

L. A. Meserve and M. A. R. Gonzalez

food at 24 days of age, whereas normal pups spent
92% of the test time away from the mother at the
same age (Fig. 1, A and D). This apparent contra¬
diction is likely to be the result of a combination of
a number of factors including the following. The
T4-injected pups may progress in weaning be¬
havior from 24 to 25 days of age rapidly enough to
demonstrate an obvious developmental spurt.
Additionally, the dosage of thiouracil adminis¬
tered may be sufficient to inhibit conversion of
some proportion of injected T4 to triiodothyronine
(T3), the biologically active molecule [13].
Perhaps such a situation is more detrimental to
normal development of locomotor and discrimina¬
tory behaviors, as evidenced by the dietary prefer¬
ence test, than to factors controlling actual con¬
sumption of chow. The latter suggestion is an
attractive one since the concentration of goitro¬
genic material used in this and previous studies [5],
which appears to prevent weaning, has been a
relatively high one. Blake and Henning [6] have
found relatively low doses of the propyl ester of
thiouracil (PTU) to allow weaning to occur in rats,
but with an approximate 3-5 day delay past nor¬
mal. It would be of interest to determine whether
high and low doses of goitrogen or genetic
hypothyroidism [14] have differential effects on
the liver deiodinase enzyme responsible for con¬
verting T4 to T3, and whether such differences play
a role in determining if a rat or mouse weans
normally, weans late, or does not wean at all.

Taken together, the dietary preference tests and
analyses of stomach content performed in the
present study suggest that the influence of thyroid
status on the weaning process in mice is very
similar to that previously reported in rats [5] even
though adult mice are much smaller and have a
significantly higher metabolic rate than rats [7, 8].
Thyroid hormone appears to be required to allow
normal progression of weaning from milk to chow,
regardless of ambient temperature. Further stu¬
dies of the importance of conversion of T4 to T3 for
normal weaning in the mouse are planned.

ACKNOWLEDGMENTS

Supported by the Department of Biological Sciences,
BGSU. The authors thank D. Colon, J. Landis, B.

Murray, L. Parsons, C. Santorelli, and L. Shaffer for

critically reviewing early versions of this paper and L.

Floro for expert secretarial assistance. We thank E.

Harris for donating T4 RIA kits.

REFERENCES

1 Hamburgh, M. (1968) An analysis of the action of
thyroid hormone on development based on in vivo
and in vitro studies. Gen. Comp. Endocrinol., 10:
198-213.

2 Hall, W. G. , Cramer, C. P. and Blass, E. M. (1975)
Developmental changes in suckling of rat pups.
Nature, 258: 318-320.

3 Redman, R. S. and Sweeney, L. R. (1976) Changes
in diet and patterns of feeding activity of developing
rats. J. Nutr., 106: 615-626.

4 Henning, S. J., Chang, S.-S. P. and Gisel, E. G.
(1979) Ontogeny of feeding controls in suckling and
weanling rats. Am. J. Physiol., 237: R187-R191.

5 Blake, H. H. and Henning, S. J. (1983) Weaning in
the rat: a study of hormonal influences. Am. J.
Physiol., 244: R537-R543.

6 Blake, H. H. and Henning, S. J. (1985) Effect of
propylthiouracil dose on serum thyroxine, growth,
and weaning in young rats. Am. J. Physiol., 248:
R524-R530

7 Crispens, C. G., Jr. (1975) Handbook on the
Laboratory Mouse. Charles C. Thomas, Springfield,
Illinois, pp. 80-137.

8 Pasquis, P., Lacaisse, A. and Dejours, P. (1970)
Maximal oxygen uptake in four species of small
mammals. Respir. Physiol., 9: 298-309.

9 Meserve, L. A. (1987) Hypothalamic CRF im-
munoreactivity in genetically hypothyroid ( hyt/hyt )
mice. Proc. Soc. Exp. Biol. Med., 185: 335-338.

10 Zar, J. H. (1984) Biostatistical Analysis. Prentice-
Hall, Englewood Cliffs, New Jersey, 2nd ed., pp.
162-187, 196-190.

11 Murphy, J. M. and Nagy, Z. M. (1976) Neonatal
thyroxine stimulation accelerates the maturation of
both locomotor and memory processes in mice. J.
Comp. Physiol. Psychol., 90: 1082-1091.

12 Swanson, L. J. and Campbell, C. S. (1981) The role
of the young in the control of the hormonal events
during lactation and behavioral weaning in the
golden hamster. Horm. Behav., 15: 1-15.

13 Kaplan, M. M. (1984) The role of thyroid hormone
deiodination in the regulation of hypothalamo-
pituitary function. Neuroendocrinology, 38: 254-
260.

14 Beamer, W. G., Eicher, E. M., Maltias, L. J. and
Southard, J. L. (1981) Inherited primary hypothy¬
roidism in mice. Science, 212: 61-63.

ZOOLOGICAL SCIENCE 5: 1089-1093 (1988)

© 1988 Zoological Society of Japan

A Ghost Shrimp with Four- Articulate Fifth Pereopods (Crustacea:
Caprellidea: Phtisicidae) from Northwest Australia

Akira Hirayama

Biological Laboratory, Department of Liberal Arts, Asia University,

5-24-10 Sakai, Musashino-shi, Tokyo 180, Japan

ABSTRACT — Quadrisegmentum triangulum gen. et sp. nov. was found in a gorgonian host, Isis
hippurus, on Ashmore Reef, Northwest Australia. The new genus is unique in the subfamily Phtisicinae
in having four-articulate fifth pereopods. Genera having six-, five- and three-articulate fifth pereopods
were previously known in this subfamily. Generic relationships within the Phtisicinae are discussed.

INTRODUCTION

The specimens reported herein were collected
by H. K. Larson from a gorgonian host, Isis hip¬
purus Linnaeus, on Ashmore Reef in Northwest
Australia on 24 July 1986 and were sent to me for
identification by A. J. Bruce. They are members
of the subfamily Phtisicinae on the basis of the 3-4
fully segmented pereopods [1, 2] but do not fit any
definition of the known genera in terms of the
numbers of segments in pereopod 5. The new
ghost shrimp is 4-articulate, while the known
genera of this subfamily are 6-, 5- and 3-articulate
[1-6]. These characters are important to the
construction of a phylogeny in the subfamily Phti¬
sicinae because they show successional changes in
the segmentation of pereopod 5. The new ghost
shrimp is described and generic relationships with¬
in the subfamily Phtisicinae are discussed on the
basis of these characters and other generic ones.

All the specimens described herein are depos¬
ited in the collection of the Museum and Arts
Galleries of the Northern Territory, Darwin, Aus¬
tralia.

Quadrisegmentum gen. nov.

Diagnosis

Flagellum of antenna 2, 7-articulate in male and

Accepted February 22, 1988
Received July 8, 1987

5-articulate in female. Mandibles lacking molar
process, provided with 3-articulate palp. Inner and
outer plates of maxilliped subequal in size. Gills
present on pereonites 2-4, small. Pleonal appen¬
dages 2-paired, 2-articulate. Pereopods 3-4, 6-
articulate; pereopod 5, 4-articulate.

Type species. Quadrisegmentum triangulum sp.
nov.

Etymology. The generic name refers to the four-
articulate fifth pereopods. The gender is neuter.

Remarks

The new genus apparently belongs to the sub¬
family Phtisicinae Vasilenko, 1968 [1, 2, 8] with
the following diagnostic characters: Gills present
on pereonites 2-4; mandibles lacking molar, fur¬
nished with 3-articulate palp; pereonites 3-4, 6-
articulate. However, it may be clearly disting¬
uished from other genera within this subfamily by
unique segmentation of pereopod 5; it is 4-
articulate in the new genus and 6-, 5- and 3-
articulate in the known genera.

Within the subfamily Phtisicinae, two-paired
and bi-articulate pleonal appendages suggest that
the new genus is related to six genera: Paraproto
Mayer, 1903 [3], Pseudoprotomima McCain, 1969
[5], Phtisica Slabber, 1769 [3, 4, 8], Protomima
Mayer, 1903 [3, 7], Protoplesius Mayer, 1903 [3]
and Chaka Criffiths, 1974 [6]. All these genera,
including Quadrisegmentum gen. nov., can be di¬
vided into four groups based on the segmentation

1090

A. Hirayama

of pereopod 5: 1) 6-articulate group ( Paraproto ,
Pseudoprotomima), 2) 5-articulate group ( Phtisi -
ca, Protomima, Protoplesius) , 3) 4-articulate
group ( Quadrisegmentum gen. nov.) and 4) 3-
articulate group ( Chaka ). These groups suggest an
evolutionary change in the phylogeny of the sub¬
family Phtisicinae, i.e., the 6-articulate fifth
pereopods evolved to 3-articulate through 5-
articulate and 4-articulate fifth pereopods. Other
generic characters listed to date [1-8] do not
indicate successional or phylogenic relationships
within this subfamily. For example, the setal
formula on the terminal segment of the mandibu¬
lar palp is 1-x-l, where x indicates the number of
short setae held between longer setae at both ends
of a setal row. The x, however, shows irregular
variations at the generic and specific levels [3-8].
Further, the segmentation number of the flagellum
of antenna 2 is irregularly variable in the six
genera: 14-articulate (males) and 10-articulate
(female) in Paraproto , 7-articulate (males) and
5-articulate (females) in Quadrisegmentum gen.
nov., 2 to 5-articulate in Phtisica, 2-articulate
(males) and 5-articulate (females) in Protoplesius,
4-articulate in Pseudoproto and Protomima, and
3-articulate in Chaka. Although the evolutionary
tendency of the flagellum segmentation generally
shows a reduction in the caprellid amphipods [4,7,
8], this phenomenon can not be observed in the
subfamily Phtiscinae. Therefore, I present the
phylogenic assumption that the genera in the sub¬

family Phtisicinae have evolved from the group
with the 6-artiulate fifth pereopods to the group
with the 3-articulate ones through the groups with
5- and 4-articulate fifth pereopods and that the new
genus Quadrisegmentum is at an intermediate state
on the evolutionary line.

Quadrisegmentum triangulum sp. nov.

(Figs. 1-3)

Description of the male holotype (8.0 mm)

Body Not spinose. Length ratios of pereonites

1- 7, 3:4:5:6:9:8:2. Gills present on pereonites

2- 4, small. Pleonal appendages 2-paired, 2-
articulate, serrate on inner margins.

Antennae Antenna 1: Length ratios of
peduncular segments 1-3, 3:4:5; flagellum 10-
articulate, each segment furnished with 1 aesthe-
tasc. Antenna 2: Gland cone of peduncular seg¬
ment 2 distinct; length ratios of peduncular seg¬
ments 3-5, 1:3:3; flagellum consisting of 7 seg¬
ments, distal one rudimentary.

Mouthparts Upper lip symmetrically bilobate.
Lower lip: Mandibular process medium. Maxilla
1: Outer plate provided with 6 tooth-like spines;
palp consisting of 2 segments, distal one provided
with 4 teeth, 1 apical spine and 3 setae on distal
half of outer side. Maxilla 2: Both plates provided
with 5 apical setae. Mandibles similar to each
other; incisor provided with 4 large teeth; lacinia

Fig. 1. Quadrisegmentum triangulum gen. et sp. nov. A: Holotype, male, 8.0 mm. B: Paratype (no. 2), female, 5.2
mm.

New Genus and New Species of Ghost Shrimp

1091

Fig. 2. Quadrisegmentum triangulum gen. et sp. nov., male holotype. A: Head. B: Peduncle of antenna 2. C:
Upper lip. D: Lower lip. E: Maxilla 1. F: Maxilla 2. G: Left mandible. G-l: Palp of left mandible. H: Right
mandible. I: Inner and outer plates of maxilliped. 1-1: Outer plate and palp of maxilliped. J: Coxa 2. K: Coxa 3.
L: Coxa 4. M: Coxa 5. N: Pereonites 6-7. O: Pleonal appendages. P: Gnathopod 1. P-1: Palmar cusp of
gnathopod 1. Q: Gnathopod 2. Q-l and Q-2: Palm of gnathopod 2. R: Pereopod 3. R-l: Propod and dactyl of
pereopod 3. S: Pereoped 4. T: Pereoped 5. U: Pereopod 6.

mobilis consisting of 3 large and several small
plates; accessory setae 3 in number; palp consisting
of 3 segments, of which middle one is furnished
with 1 distal seta, terminal segment of palp pubes¬
cent and bearing 2 setae on distal half. Maxiliped:
Inner plate serrate distally, armed with 3 apically
serrate spines; outer plate subequal to inner plate
in size, provided with 6 inner marginal setae; palp

consisting of 4 segments, of which the penultimate
one is prominently protruded distolaterally.

Coxae Coxae 1-5 coalescent with pereonites,
vestigial; coxae 6-7 absent.

Gnathopods 1-2 Gnathopod 1: Length ratios
of segments from basis to propod approximately
16:3:7:9:10; carpus compressed at 1/3 from pro¬
ximal end; propod triangular; palm defined by

1092

A. Hirayama

Fig. 3. Quadrisegmentum triangulum gen. et sp. nov. , female paratype (no. 2). A: Ventrolateral part of pleonites 3-
4. B: Pleonal appendages. C: Gnathopod 1. D: Gnathopod 2. D-l: Merus, Carpus, propod and dactyl of
gnathopod 2. E: Pereopod 3. F: Pereopod 4. G: Pereopod 5. H: Pereopod 7. FI-1, H-2 and H-3: Spines on
propod of pereopod 7.

prominent, curved, bifid process with 2 accom¬
panying spines; dactyl reaching cusp when closed.
Gnathopod 2 slender, feeble on segments from
basis to carpus; length ratios of segments from
basis to propod approximately 9 : 1 : 3 : 2 : 7; merus
and carpus not prominently overlapping each
other; propod quadrangular; palm extending on
more than half of posterior margin of propod,
truncate and serrate distally, defined by both pal¬
mar and poison projections, each of them armed
with 1 spine; dactyl reaching palmar protrusion
when closed, provided with numerous pits on
grasping margin.

Pereopods 3-7 Pereopods 3-4: 6-articulate,
slender, feeble; pereopod 3, 0.78 as long as
pereopod 4; length ratios of segments from basis to
dactyl approximately 9 : 1 : 7 : 5 : 4 : 3 in pereopod
3, 15 : 1 : 9 : 5 : 5 : 3 in pereopod 4; propod slightly
dilated at 3/7 from proximal end, bearing 4 spines
on distal 4/7 of inner margin, distal spine of them
small. Pereopod 5: 4-articulate, slender; length
ratios of segments 1-4, 4 : 5 : 4 : 2; propod uniform
in width; dactyl falcate. Pereopod 6: 6-articulate,
slender; length ratios of segments from basis to
dactyl approximately 14:1:7:4:7:3; propod pro¬
vided with inner medial tooth bearing 2 paired
spines basally, also with 4 inner spines in addition

to these two paired spines; dactyl falcate.
Pereopod 7 missing.

Description of female paratype (no. 2, 5.2 mm)

Antenna 2: Flagellum consisting of 4 segments
and a rudimentary one. Gnathopod 1 similar to
that of female. Gnathopod 2: Length ratios of
segments from basis to propod 16:2:5:4:11;
merus and carpus obliquely articulate; segmenta¬
tion scar visible between carpus and propod; pro¬
pod semilunar; palm defined by palmar protrusion,
which is amred with 1 spine, slightly rounded,
finely serrate from proximal 2/3 point to distal end
of itself; poison tooth present near palmar spine,
amred with 2 paired spines; dactyl reaching poison
tooth when closed. Pereopods 3-4: Propod not
dilated, provided with 3 small spines; dactyl slight¬
ly dilated proximally. Pereopod 5 similar to that of
male. Pereopod 6 missing. Pereopod 7: Merus,
carpus and propod equal in length, slightly longer
than basis; carpus provided with 2 spines; propod
with 1 medial tooth bearing 2 paired spines basally,
also with 1 proximal and 2 distal spines; dactyl
falcate, reaching opposite proposal tooth when
closed.

New Genus and New Species of Ghost Shrimp

1093

Material examined

Holotype: Male, 8.0 mm, taken from gorgonian
host, Isis hippurus Linnaeus, in 18 m on west islet
of Ashmore Reef, Northwest Australia
(12°14.28'S., 122°59.14'E.); 24 July 1986; coll. H.
K. Larson. Paratypes: One male (no. 1) and two
females (nos. 2-3), collected with the holotype.
Holotype and paratype no. 2 are mounted on slide
glasses in a gum-chloral medium. Collection num¬
ber: NTM Cr. 00447.

Etymology The specific name, triangulum, is
derived from the triangular propod of gnathopod
1.

ACKNOWLEDGMENTS

I wish to thank Drs. H. K. Larson and A. J. Bruce of
the Northern Territory Museum of Arts and Sciences,
Australia, for giving me the opportunity to examine the
present materials. Thanks are also due to Dr. Yoshihide
Suzuki of Asia University for providing me with working
space and facilities.

REFERENCES

1 Vasilenko, S. V. (1968) On the question of systema-
tics and the basic line of development in the family
Caprellidae. Doklady Akademija Nauk USSR, 183:
1461-1464. (In Russian)

2 McCain, J. C. (1968) Familial taxa within the Caprel-
lidea (Crustacea: Amphipoda). Proc. Biol. Soc.
Wash., 82: 837-842.

3 Mayer, P. (1903) Die Caprellidae der Siboga-
Expedition. Siboga Exped., 34: 1-160, pis. 1-10.

4 McCain, J. C. (1968) The Caprellidae (Crustacea:
Amphipoda) of the western North Atlantic. Bull. U.
S. Nat. Mus., 278: 1-147.

5 McCain, J. C. (1969) New Zealand Caprellidae
(Crustacea: Amphipoda). N. Z. J. Mar. Freshwat.
Res., 3: 286-295.

6 Griffiths, C. L. (1974) The Amphipoda of Southern
Africa. Part 3. The Gammaridea and Caprellidea of
Natal. Ann. S. Afr. Mus., 62: 209-264.

7 Arimoto, I. (1976) Taxonomic studies of caprellids
(Crutacea, Amphipoda, Caprellidae) found in the
Japanese and adjacent waters. Spec. Publ. Seto Mar.
Biol. Lab., 3: 1-299.

8 Vasilenko, S. V. (1974) Caprellids of the sea of the
USSR and adjacent waters. Akademija Nauk USSR,
107: 1-288. (In Russian)

'

ZOOLOGICAL SCIENCE 5: 1095-1103 (1988)

© 1988 Zoological Society of Japan

The Drosophila immigrans Species-group of the Subgenus
Drosophila (Diptera: Drosophilidae) in Yunnan, China

Wen Xia Zhang and Masanori J. Tod a1,2

Kunming Institute of Zoology, Academia Sinica, China, and
1 Institute of Low Temperature Science, Hokkaido
University, Sapporo 060, Japan

ABSTRACT — A total of 22 species of the Drosophila immigrans species-group are reported from
Yunnan, China, with description of 4 new species, redescription of 2 known species and new records of
14 known species from China.

INTRODUCTION

The Drosophila immigrans species-group con¬
sists of four subgroups: the immigrans , nasuta,
hypocausta and quadrilineata subgroups [1-3].
This species-group comprises the greater part of
the drosophilid fauna in the Oriental Region [4-6].
Up to the present, 6 species of this group, D.
immigrans Sturtevant, 1921, D. nixifrons Tan, Hsu
et Sheng, 1949, D. hexastriata Tan, Hsu et Sheng,
1949, D. annulipes Duda, 1924, D. spuricurviceps
Zhang et Gan, 1986 and D. ruberrimoides Zhang
et Gan, 1986, have been reported from China [7,
8]. Based upon collections from Yunnan, this
paper adds 4 new and 14 known species to the
faunal list of this group in China. The 14 species
new to China are D. formosana Duda, 1926, D.
shwezayana Toda, 1986, D. burmae Toda, 1986,
D. ruberrima Meijere, 1911, D. pentafuscata Gup¬
ta et Kumar, 1986, D. metasetigerata Gupta et
Kumar, 1986, D. albomicans Duda, 1924, D.
sulfurigaster albostrigata Wheeler, 1969, D.
kohkoa Wheeler, 1969, D. siamana Ikeda et al.,
1983, D. notostriata Okada, 1966, D. flavitibiae
Toda, 1986, D. obscurinervis Toda, 1986 and D.
quadrilineata Meijere, 1911. D. immigrans and D.
annulipes were also collected from Yunnan. In
total, 24 species of the immigrans group (11 spp. of

Accepted September 16, 1987
Received August 4, 1987

2 To whom requests of reprints should be addressed.

the immigrans subgroup, 4 spp. of the nasuta
subgroup, 1 sp. of the hypocausta subgroup, and 8
spp. of the quadrilineata subgroup) have been
recorded from southern China. This number is the
largest on the world, followed by Burma (18 spp.),
Taiwan (17 spp.), India (17 spp.), Borneo (15
spp.), Malaya (12 spp.) and Sumatra (11 spp.),
suggesting that the southwestern part of China is
an evolutionary center of this species-group, espe¬
cially the immigrans and quadrilineata subgroup.

Two known species which were originally de¬
scribed in Chinese [8] are redescribed here in
English, along with 4 new species. All holotypes
and paratypes are deposited in the Kunming Insti¬
tute of Zoology, Academia Sinica, China. The
diagnoses of the subgenus, species-group and spe¬
cies-subgroups were previously given by Toda [3].

D. IMMIGRANS SPECIES-SUBGROUP

Drosophila (Drosophila) spuricurviceps Zhang et
Gan

(Figs. 1-4)

Drosophila (Drosophila) spuricurviceps Zhang et Gan,
1986 [8]: 359.

$ , Body ca. 4. 0-4. 5 mm wing ca. 4. 0-4. 5
mm in length.

Head: Eye purple red, with thin pile. Second
joint of antenna yellow, with 2 stout setae; 3rd
joint yellow. Arista with 4 dorsal, 1 ventral
branches and moderate terminal fork. Palpus
yellow, with 2 stout bristles and a few small hairs.

1096

W. X. Zhang and M. J. Toda

Ocellar triangle black, with a few hairs; ocellars
long. Periorbit yellowish. Frons orange yellow.
Clypeus dark brown. Frons ca. 1/2 as broad as
head, with a few frontal hairs. Face brownish.
Carina low and short, without sulcus. Cheek
yellow, black at base of vibrissa, ca. 1/5 as broad as
greatest diameter of eye. Posterior reclinate orbit¬
al nearer to proclinate than to inner vertical.
Anterior reclinate orbital ca. 2/5 length of poste¬
rior reclinate; proclinate ca. 4/5 length of posterior
reclinate. Vibrissa long, stout and black, 2nd oral
ca. 1/5 length of vibrissa.

Thorax: Mesoscutum dark brown, with 6
obscure black longitudinal stripes; median 2 stripes
between dorsocentrals, sometimes fused to each
other at anterior part; 2 stripes running along line
of dorsocentrals; outer 2 stripes from transverse
suture. Scutellum brownish black, paler at anter¬
ior corner and tip. Thoracic pleura brown, with 3
obscure black longitudinal stripes; upper one nar¬
row, short from lower part of episternum to base
of wing; middle one broad, slightly fused anterior¬

ly with upper one, running from upper part of
propleurite to base of haltere; lower one broad,
running through nearly entire length of sternoepi-
sternum. Sometimes thoracic pleura entirely dark
brown. Humeral plate brownish; humerals 2,
upper one longer. Acrostichal hairs in 6 rows.
Anterior dorsocentral ca. 0.6-0. 7 length of poste¬
rior; cross distance of dorsocentrals ca. 2.0 length
distance. Anterior scutellars parallel, as long as
posterior; posteriors convergent and crossed. Ster-
no-index ca. 0.5-0. 6.

Legs yellow. Fore femur with row of ca. 10
spinules. Preapicals on all tibiae; apicals on fore
and mid tibiae. $ fore tarsus neither modified nor
ornamented. Mid and hind tarsi without row of
minute cuneiform bristles on underside. Fore
metatarsus as long as 2 succeeding tarsal joints
together; mid and hind metatarsi slightly shorter
than rest together. Apical 2 tarsi brown.

Wing hyaline. Veins brown; anterior cross
vein slightly cloud; posterior cross vein cloud;
R4+5 and M parallel. Ci-bristles 2, subequal.

Figs. 1-4. Drosophila (Drosophila) spuricurviceps Zhang et Gan, 1986. 1: Periphallic organs. 2: Aedeagus (dorsal
view). 3: Ditto (lateral view). 4: Ditto (ventral view). Scale-line =0.1 mm.

D. immigrans Species-group from Yunnan

1097

Wing indices: C ca. 4. 0-4. 5, 4V ca. 1.6, 4C ca. 0.6,
5x ca. 1.3, Ac ca. 1.8, C3F ca. 1/3— 1/4. Haltere
milky white.

Abdomen: $ tergites black, except yellow 1st
tergite. tergites yellow, each with uninterrupted
caudal black band. Sternites pale yellow.

Periphallic organs (Fig. 1): Epandrium in up¬
per part dark brown, with 8 bristles; in lower 1/3
pale yellow, with ca. 16-17 bristles; ventrally
truncate. Surstylus with slightly concave row of ca.
8 primary teeth on middle margin; lower part
convex, with small bristles; upper part strongly
convex, bare; medio-inner surface with small bris¬
tles. Cercus setigerous except ventral margin,
brown except anterolateral (black) and ventral
(yellow) margins, caudo-ventrally with flap bear¬
ing V-shaped row of stout bristles on ventral
margin.

Phallic organs (Figs. 2-4): Aedeagus round at
tip, apicoventrally with 1 pair of marginally serrate
flaps, dorsomedially and ventrobasally serrate on
lateral margins. Dorsal process elongated and
flattened flap, apically bifid. Apodeme broad in
lateral view. Novasternum nearly quadrate;
hypandrial plate triangular, with 1 pair of small
submedian spines.

Ovipositor yellow, apically broad and round,
with ca. 22 marginal but no discal teeth.

Specimens examined. China: 6$, l-£, Kun¬
ming, Yunnan Province, 21.11.1987 (M. J. Toda).

Distribution. China: Yunnan.

Relationships. This species is closely related to
D. curviceps Okada et Kurokawa, 1957, but differs
from the latter in the morphology of aedeagus,
surstylus and carina. Zhang and Gan [8] regarded
the presence of V-shaped row of stout bristles on
caudoventral expansion of cercus as an important
diagnostic character by which the present species is
distinguished from D. curviceps, based on the
ignorance of similar character in the original de¬
scription of the latter species [9]. However, one of
us (M. J. T.) confirmed the presence of the similar
structure, i.e. caudoventral expansion with numer¬
ous stout bristles, in Japanese specimens of D.
curviceps.

Drosophila ( Drosophila ) ruberrimoides Zhang et
Gan

(Figs. 5-8)

Drosophila ( Drosophila ) ruberrimoides Zhang et Gan,
1986 [8]: 360.

$ . Body ca. 4. 0-4. 5 mm, wing ca. 4. 0-4. 5 mm
in length.

Head: Eye tannish red, with thick pile.
Second joint of antenna dark brown, with 3 stout
setae; 3rd joint yellow. Arista with 6 dorsal, 4
ventral long branches and moderate terminal fork.
Palpus yellow, with 2 long stout bristles at tip and a
few small hairs. Ocellar triangle black, with sil¬
very-white shinning on outside margin of ocelli in
frontal view, with a few small hairs; ocellars long
and divergent. Periorbit shinning silvery-white in
frontal view. Frons brownish yellow in frontal
view but shinning silvery-white in lateral view,
anteriorly orange-yellow. Clypeus brown. Frons
ca. 1/3 as broad as head, with a few frontal hairs.
Face brown. Carina high, wider in lower part.
Cheek yellow, ca. 1/7 as broad as greatest diameter
of eye. Occiput medially black. Posterior recli-
nate orbital nearer to proclinate orbital than to
inner vertical. Anterior reclinate orbital ca. 1/3
length of posterior reclinate; proclinate ca. 3/4
length of posterior reclinate. Vibrissa and 2nd oral
stout, black and long; 2nd oral as long as vibrissa;
other orals small.

Thorax: Mesoscutum yellow. Scutellum brown¬
ish yellow, quadrate. Thoracic pleura yellow.
Humerals 2, subequal. Acrostichal hairs in 8 rows.
Anterior dorsocentral ca. 2/5 length of posterior;
cross distance of dorsocentrals ca. 3.0 length dis¬
tance. Anterior scutellars divergent; posteriors
convergent. Anterior scutellar as long as poste¬
rior. Sterno-index ca. 0.6; mid sternopleural near¬
ly equal to anterior.

Legs yellow. Fore femur thick, with row of ca.
13-17 spinules. Preapicals on all tibiae, apicals on
mid tibia. $ fore proximal 2 tarsal joints thin and
short, each with thick hair tuft on inner side and 1
black, long bristle at distal end. Mid and hind tarsi
each with row of minute cuneiform bristles on
underside. Fore metatarsus shorter than 2 suc¬
ceeding tarsal joints together; mid and hind meta¬
tarsi slightly shorter than rest together.

1098

W. X. Zhang and M. J. Toda

Figs. 5-8. Drosophila ( Drosophila ) ruberrimoides Zhang et Gan, 1986. 5: Periphallic organs. 6: Aedeagus (lateral
view). 7: Ditto (ventral view). 8: Novasternum (ventral view). Scale-line =0.1 mm.

Figs. 9-13. Drosophila ( Drosophila ) serraprocessata sp. nov. 9: Periphallic organs. 10: Surstylus. 11: Phallic organs
(ventral view). 12: Aedeagus (lateral view). 13: Ditto (dorsal view). Scale-line = 0.1 mm.

Wing hyaline. Veins yellow; anterior cross
vein clear; posterior cross vein cloud; R4+5 and M
parallel. Q-bristles 2, subequal. Wing indices: C
ca. 2. 5-2. 6, 4V ca. 1.4, 4C ca. 0.8-0. 9, 5x ca. 0.8,
Ac, ca. 2.5, C3F ca. 4/5. Haltere yellowish white.

Abdomen: Tergites yellow; 2nd to 5th tergites
each with medially interrupted, caudal, dark
brown band; 6th tergite almost entirely dark
brown.

Periphallic organs (Fig. 5): Epandrium black,
paler in lower part, with 1 bristle near base of
surstylus; toe narrow, with 2 stout black teeth.
Surstylus brown, bell-shaped, with black, stout
primary teeth arranged in U-shaped row. Cercus
slightly oval, black, with many bristles.

Phallic organs (Figs. 6-8): Aedeagus black,
slightly curved ventrad; dorsal process articulated

with aedeagus, with many small spinules on sur¬
face; membranous structure surrounding
aedeagus, with small warts on surface. Anterior
paramere fused to hypandrium, with 1 sensillum.
Novasternum with 1 pair of stout submedian
spines.

Specimens examined. Holotype $ , China:
Kunming, Yunnan Province, 1. VII. 1984 (W. X.
Zhang). Paratype, 1 $ , same data as holotype.

Distribution. China: Yunnan.

Relationships. This species belongs to the D.
immigrans species-subgroup, because of having a
characteristic dorsal process on aedeagus, resem¬
bles somewhat D. ruberrima in body color and $
fore metatarsus, but differs from the latter in the
structure of $ genitalia.

D. immigrans Species-group from Yunnan

1099

Drosophila (. Drosophila ) serraprocessata sp. nov.

(Figs. 9-13)

$ . Body ca. 3. 0-3. 5 mm in length.

Head: Eye tannish red, with pile. Second joint
of antenna yellow, with 3 stout setae; 3rd joint
yellowish. Arista with 7 dorsal, 3 or 4 ventral long
branches and moderate terminal fork. Palpus
yellow, with 1 long stout bristle at tip and a few
small hairs. Ocellar triangle yellow, black on inner
margins of ocelli, with a few small hairs; ocellars
long. Periorbit yellow. Frons yellowish red;
periorbit and anterior part of frons shinning sil¬
very-white in lateral view. Clypeus medially yel¬
low, laterally brown. Frons ca. 1/2 as broad as
head, with a few frontal hairs. Face brownish.
Carina high, upper narrow and lower wide. Cheek
yellow, brown at base of vibrissa, ca. 1/9 as broad
as greatest diameter of eye. Posterior reclinate
orbital nearer to inner vertical than to proclinate
orbital. Anterior reclinate orbital ca. 3/7 length of
posterior reclinate; proclinate ca. 9/10 length of
posterior reclinate. Second oral as long as vibrissa.

Thorax: Mesoscutum brownish yellow.
Scutellum brownish yellow. Thoracic pleura yel¬
low. Humerals 3; upper one smallest. Acrostichal
hairs in 8 rows. Anterior dorsocentral ca. 1/2
length of posterior; cross distance of dorsocentrals
ca. 2.4 length distance. Anterior scutellar as long
as posterior. Sterno-index ca. 0.7.

Legs yellow. Fore femur with row of ca. 11-12
spinules. Preapicals on all tibiae; apicals on mid
tibia. $ fore tarsus anteriorly with long recurved
hairs; proximal 2 tarsal joints with thick hair tuft
on inner side. Mid and hind tarsi each with row of
minute cuneiform bristles on underside. Fore
metatarsus shorter than 2 succeeding tarsal joints
together; mid and hind metatarsi as long as rest
together.

Wing hyaline. Veins brownish yellow; anterior
cross vein clear; posterior cross vein cloud; R4+5
and M nearly parallel. Q-bristles 2, upper one
larger. Wing indices: C ca. 2.8, 4C ca. 0.75, 4V ca.
0.65, 5x ca. 1.0, Ac ca. 2.6, C3F ca. 6/7. Haltere
yellowish.

Abdomen: Tergites yellow; 2nd to 5th tergites
each with narrow, not sharply demarcated, medial¬

ly interrupted, caudal dark brown band; 6th tergite
entirely yellow.

Periphallic organs (Figs. 9 and 10): Epandrium
brown and shinning, with 1 bristle at base of
surstylus and 2 black stout teeth on toe. Surstylus
(Fig. 10) with ca. 14 black primary teeth in con¬
cave row on entire distal margin. Cercus large,
oval, brown, with many long bristles; ventral part
paler, with ca. 3 short bristles.

Phallic organs (Figs. 11-13): Aedeagus
brown, subapically somewhat dilated, dorso-
medially with knob; dorsal process membranous,
hairy on upper half surface, basally with 1 pair of
recurved, thin, bare branches. Hypandrial process
with deep incision and serration on apical margin.
Novasternum nearly quadrate, with 1 pair of sub¬
median spines near bases of hypandrial processes.

Holotype $ , China: Meng-long, Meng-la
Country, Yunnan Province, 3. X. 1985 (W. X.
Zhang).

Distribution. China: Yunnan.

Relationships. This species somewhat resem¬
bles the foregoing species, D. ruberrimoides, in the
morphology of aedeagus and epandrium, but dif¬
fers from the latter in hypandrial process and
surstylus.

Drosophila ( Drosophila ) parustulata sp. nov.

(Figs. 14-16)

$ . Body ca. 3. 5-4.0 mm in length.

Head: Eye tannish red, with thick pile.
Second joint of antenna yellow, with 2 stout setae;
3rd joint yellowish. Arista with 7 dorsal, 5 ventral
branches and small terminal fork. Palpus yellow,
with ca. 3 or 4 long stout bristles and a few hairs.
Ocellar triangle yellowish red, with a few small
hairs; ocellars long. Periorbit yellow. Frons yel¬
lowish red. Clypeus medially yellow, laterally
brown. Frons ca. 1/2 as broad as head, with a few
frontal hairs. Face yellow. Carina high, wider in
lower part. Cheek yellow, ca. 1/7 as broad as
greatest diameter of eye. Posterior reclinate orbit¬
al nearer to proclinate orbital than to inner verti¬
cal. Anterior reclinate orbital ca. 1/3 length of
posterior reclinate; proclinate ca. 2/3 length of
posterior reclinate. Vibrissa and 2nd oral long and

1100

Z. W. Xia and M. J. Toda

Figs. 14-16. Drosophila ( Drosophila ) parustulata sp. nov. 14: Periphallic organs. 15: Phallic organs (ventral view).
16: Aedeagus (lateral view). Scale-line =0.1 mm.

stout, 2nd oral ca. 3/4 length of vibrissa; other
orals short and thin.

Thorax: Mesoscutum and scutellum yellowish
red. Thoracic pleura yellow. Flumerals 2, sub¬
equal. Acrostichal hairs in 8 rows. Anterior
dorsocentral ca. 5/7 length of posterior; cross
distance of dorsocentrals ca. 2.4 length distance.
Anterior scutellars convergent, ca. 3/4 length of
posterior; posteriors convergent and crossed.
Sterno-index ca. 0.65.

Legs yellow. Fore femur with row of ca. 13-14
spinules. Preapicals on all tibiae; apicals on mid
tibia. $ fore tarsus without thick hair. Mid and
hind tarsi each with row of minute cuneiform
bristles on underside. Fore metatarsus as long as 3
succeeding tarsal joints together; mid and hind
metatarsi as long as rest together.

Wing hyaline. Veins brownish yellow; anterior
cross vein clear; posterior cross vein cloud; R4+5
and M nearly parallel. Ci-bristle 1. Wing indices:
C ca. 4.9, 4C ca. 0.4, 4V ca. 1.1, 5x ca. 0.5, Ac ca.
1.4, C3F ca. 5/9. Haltere yellowish.

Abdomen: Tergites yellow; 2nd to 5th tergites
each with narrow, caudal dark brown band.

Periphallic organs (Fig. 14): Epandrium
pubescent dorsocaudally, with ca. 6 bristles in
upper part and ca. 4 at base of surstylus, ventrally
narrow and with ca. 5 short bristles. Surstylus
deeply concave on distal margin, with ca. 5 teeth
on upper distal margin, ca. 7 smaller teeth on
lower distal margin and 2 long bristles. Cercus
pubescent, with many bristles and tuft of several
short bristles at caudoventral corner.

Phallic organs (Figs. 15 and 16): Aedeagus dis-
tally membranous and expanded like trumpet,
laterally with 1 pair of acute, sclerotized projec¬
tions, medioventrally with apically bifid, flattened
appendage. Hypandrial process narrow. Nova-
sternum large, V-shaped, with 1 pair of submedian
spines near bases of hypandrial processes. Anter¬
ior parameres attached to lateral corners of nova-
sternum, apically with 2 sensilla.

Holotype $, China: Meng-men, Meng-la
County, Yunnan Province, 1.X.1985 (W. X.

D. immigrans Species-group from Yunnan

1101

Zhang).

Distribution. China: Yunnan.

Relationships. This species somewhat resem¬
bles D. ustulata Takada, Momma et Shima, 1973 in
the shape of phallic organs, but differs from the
latter in having hypandrial process, aedeagus later¬
ally with 1 pair of acute sclerotized projections and
medioventrally with an appendage, and surstylus
with different arrangement of primary teeth.

D. QUADRILINEATA SPECIES-SUBGROUP

Drosophila ( Drosophila ) nullilineata sp. nov.

(Figs. 17-20)

Body length $ ca. 2. 5-3. 0mm. ca.

3. 0-3. 5 mm.

Head: Eye tannish red, with thin pile. Second
joint of antenna yellow, with 2 stout setae; 3rd
joint yellowish. Arista with 5 dorsal, 2 ventral
branches and moderate terminal fork. Palpus
yellow, with 2 long stout bristles apically and

subapically and a few hairs. Ocellar triangle
yellow, brownish on inner margins of ocelli, with a
few small hairs; ocellars long. Periorbit yellow.
Frons yellowish red, with a few hairs in anterior
part. Clypeus brownish. Frons ca. 1/2 as broad as
head width. Face yellow. Carina high and narrow.
Cheek yellow, ca. 1/6- 1/7 as broad as greatest
diameter of eye. Posterior reclinate orbital nearer
to proclinate orbital than to inner vertical. Ante¬
rior reclinate orbital ca. 1/2 length of posterior
reclinate; proclinate ca. 4/7 length of posterior
reclinate. Vibrissa long, stout and black; 2nd oral
ca. 1/2 length of vibrissa.

Thorax: Mesoscutum entirely yellow without
stripes. Scutellum yellow. Thoracic pleura yellow
with 2 brownish longitudinal stripes; upper one
from upper part of propleurite to base of haltere;
lower one running through nearly entire length of
sternoepisternum. Acrostichal hairs in 6 rows.
Anterior dorsocentral ca. 5/8 length of posterior;
cross distance of dorsocentrals ca. 1.8-1. 9 length
distance. Anterior scutellars divergent, posteriors

Figs. 17-20. Drosophila ( Drosophila ) nullilineata sp. nov. 17: Periphallic organs. 18: Caudo ventral apex of cercus.

19: Phallic organs (ventral view). 20: Aedeagus (lateral view). Scale-line =0.1 mm.

Figs. 21-25. Drosophila (Drosophila) flavimedifemur sp. nov. 21: Periphallic organs. 22: Caudoventral apex of
cercus. 23: Aedeagus (ventral view). 24: Ditto (lateral view). 25: Novasternum. (Scale-line =0.1 mm.

1102

W. X. Zhang and M. J. Toda

convergent and crossed; anterior ca. 1.2 length of
posterior. Sterno-index ca. 0.7.

Legs yellow. Fore coxa brownish in proximal
part and inner margin. Fore femur with row of ca.
9-10 spinules. Preapicals on all tibiae; apicals on
fore and mid tibiae. Fore metatarsus as long as 3
succeeding tarsal joints together; mid and hind
metatarsi as long as rest together. Mid and hind
tarsi without row of minute cuneiform bristles.

Wing hyaline. Veins yellow; anterior and pos¬
terior cross veins clear; R4+5 and M slightly con¬
vergent; R-2+3 slightly curved to costa at tip. Cr
bristles 2, upper one larger. Wing indices: in $ C
ca. 3.6, 4C ca. 0.7, 4V ca. 1.8, 5x ca. 1.75, Ac ca.
3.0, C3F ca. 1/3; in $ C ca. 4.4, 4C ca. 0.6, 4V ca.
1.7, 5x ca. 1.6, Ac ca. 2.4. Haltere yellowish.

Abdomen: Tergites yellow; 2nd to 5th tergites
each with medially interrupted, laterally narrow¬
ing, dark brown caudal band and 1 pair of spots on
lateral margins.

Periphallic organs (Figs. 17 and 18): Epan-
drium entirely pubescent, broad especially in sub-
apical part, with ca. 12 bristles in middle to lower
part. Surstylus with straight row of ca. 7 or 8
primary teeth on middle to lower distal margin;
upper half strongly convex and bare. Cercus oval,
pubescent, separate from epandrium, with many
long bristles and ca. 5 short bristles at caudoventral
apex (Fig. 18).

Phallic organs (Figs. 19 and 20): Aedeagus
slender, longer than apodeme, apically tapering,
transparent and with small spines, serrate on ven-
trosubmedial margins. Anterior paramere broad,
oval in lateral view, separate from novasternum,
with ca. 2 sensilla. Posterior paramere absent.
Novasternum quadrate, pubescent around bases of
1 pair of submedian spines.

Holotype $ , China: Shang-yong, Meng-la
County, Yunnan Province, 27. IX. 1985 (W. X.
Zhang).

Allotype ■£, same data as holotype.

Distribution. China: Yunnan.

Relationships. This species certainly belongs to
the D. quadrilineata species-subgroup because of
having genitalia characteristic to this species-
subgroup, but is unique in having mesoscutum
without longitudinal stripes.

Drosophila ( Drosophila ) flavimedifemur sp. nov.
(Figs. 21-25)

$ . Body ca. 4. 0-4. 5 mm in length.

Head: Eye tannish red, with thick pile.
Second joint of antenna brown, with 2 stout setae;
3rd joint paler. Arista with 4 or 5 dorsal, 2 ventral
branches and moderate terminal fork. Palpus
apically brownish, basally yellow, with 1 long stout
bristle at tip, 1 in subapical part and a few hairs.
Ocellar triangle yellow, black on inner margins of
ocelli, with a few small hairs; ocellars long. Perior-
bit yellow. Frons yellow with 2 anteriorly conver¬
gent brown stripes, with a few frontal hairs on
anterior part of stripes. Clypeus brown. Frons ca.
1/2 as broad as head width. Face yellow. Carina
high, not so wide, brown in lower part. Cheek
yellow, brown at base of vibrissa, ca. 1/6 as broad
as greatest diameter of eye. Posterior reclinate
orbital nearer to proclinate orbital than to inner
vertical. Anterior reclinate orbital ca. 2/3 length of
posterior reclinate; proclinate ca. 7/9 length of
posterior reclinate. Vibrissa long, stout and black;
2nd oral ca. 1/2 length of vibrissa.

Thorax: Mesoscutum yellow, with 7 brown
longitudinal stripes; medial stripe wider, including
2 rows of acrostichal hairs, and running through
nearly entire length of mesoscutum; inner pair of
stripes starting slightly posteriorly, running along
line of dorsocentrals, and continuing to stripes on
scutellum; middle pair paler, from transverse su¬
ture to base of inner postalar; outer pair from
anterolateral corner just above humerus to base of
outer postalar, interrupted at transverse suture.
Scutellum yellow, with 1 pair of brown longitudi¬
nal, posteriorly convergent stripes. Thoracic
pleura yellow, with 3 brown longitudinal stripes;
upper one short, from upper part of episternum to
base of wing; middle one broad, long from upper
part of propleurite to base of haltere; lower one
broad, running through nearly entire length of
sternoepisternum. Humerals 2, upper one longer.
Acrostichal hairs in 6 rows. Dorsocentrals 2 pairs;
anterior ca. 3/5 length of posterior; cross distance
of dorsocentrals ca. 2.0 length distance. Anterior
scutellar slightly longer than posterior; anteriors
parallel, posteriors crossed. Sterno-index ca. 0.5.

Legs yellow. Fore coxa brown in proximal part;

D. immigrans Species-group from Yunnan

1103

mid and hind coxae partly brown. Fore femur
brown on outside; mid femur yellow; hind femur
slightly brownish on outside. Fore femur with row
of ca. 11 stout spinules. Preapicals on all tibiae;
apicals on fore and mid tibiae. Fore metatarsus as
long as 3 succeeding tarsal joints together; mid and
hind metatarsi slightly shorter than rest together.
Mid and hind tarsi without row of minute
cuneiform bristles.

Wing hyaline. Veins brownish yellow; anterior
cross vein clear; posterior cross vein slightly cloud;
R4+5 and M nearly parallel. Ci-bristles 2, sub¬
equal. Wing indices; C ca. 4.4, 4C ca. 0.5, 4V ca.
1.35, 5x ca. 1.2, Ac ca. 2.1, C3F ca. 1/4. Haltere
milky white, with dark brown stalk.

Abdomen: Tergites yellow; 1st tergite with
medially uninterrupted, dark brown caudal band
and 1 pair of dark brown spots on lateral margins;
2nd to 6th tergites each with medially interrupted,
dark brown caudal band and 1 pair of spots on
lateral margins; lateral spots sometimes fused to
caudal band.

Periphallic organs (Figs. 21 and 22): Epan-
drium pubescent, broad especially in subapical
part, with ca. 17 bristles. Surstylus broad, with
straight row of ca. 9 primary teeth on submedial
margin and a few small setae on inner surface;
upper and lower apical part bare. Cercus oval,
pubescent, separate from epandrium, with many
long bristles and ca. 10 short bristles at
caudoventral apex (Fig. 22).

Phallic organs (Figs. 23-25): Aedeagus slen¬
der, apically not so round in ventral view, serrate
on ventrosubmedial margins. Anterior paramere
broad, oval in lateral view, fused to novasternum,
apically pubescent and with ca. 3 sensilla. Poste¬
rior paramere absent. Novasternum quadrate,
submedially pubescent, with 1 pair of submedian
spines on inner margins.

Holotype £, China: Kunming, Yunnan Pro¬
vince, 21.11.1987 (M. J. Toda).

Distribution. China: Yunnan.

Relationships. This species is similar to D.

flavitibiae or D. clarinervis Toda, 1986 in the
coloration of body, but distinguished from the
latters in having entirely yellow mid femur, short
mid sternopleural and different structure of $
genitalia.

ACKNOWLEDGMENT

This work was supported by a Grant-in-Aid for Over¬
seas Scientific Survey from the Ministry of Education,
Science and Culture, Japan (Nos. 60041061, 61043056).

REFERENCES

1 Wilson, F. D., Wheeler, M. R., Harget, M. and
Kambysellis, M. (1969) Cytogenetic relations in the
Drosophila nasuta subgroup of the immigrans group
of species. Univ. Texas Publ., 6918: 207-270.

2 Okada, T. and Carson, H. L. (1983) The genera
Phorticella Duda and Zaprionus Coquillett (Diptera:
Drosophilidae) of the Oriental Region and New
Guinea. Kontyu, 51: 539-553.

3 Toda, M. J. (1986) Drosophilidae (Diptera) in Bur¬
ma II. The Drosophila immigrans species-group of
the subgenus Drosophila. Kontyu, 54: 634-653.

4 Bock, I. R. and Wheeler, M. R. (1972) The Dro¬
sophila melanogaster species group. Univ. Texas
Publ., 7213: 1-102.

5 Throckmorton, L. H. (1975) The phylogeny, ecolo¬
gy, and geography of Drosophila. In “Handbook of
Genetics, Vol. 3”. Ed. by R. C. King, Plenum Publ..
New York, pp. 421-469.

6 Okada, T. (1981) Oriental species, including New
Guinea. In “The Genetics and Biology of Drosophila,
Vol. 3a”. Ed. by M. Ashburner, H. L. Carson and J.
N. Thompson, Jr., Academic Press, London, pp.
261-289.

7 Tan, C. C., Hsu, T. C. and Sheng, T. C. (1949)
Known Drosophila species in China with description
of twelve new species. Univ. Texas Publ., 4920: 196-
206.

8 Zhang, W. X. and Gan, Y. X. (1986) Descriptions of
eight new species of drosophilid flies from Kunming
(Diptera: Drosophilidae). Zool. Res., 7: 351-365.
(In Chinese with English summary).

9 Okada, T. and Kurokawa, H. (1957) New or little
known species of Drosophilidae of Japan (Diptera).
Kontyu, 25: 2-12.

.

.

ZOOLOGICAL SCIENCE 5: 1105-1120 (1988)

© 1988 Zoological Society of Japan

Larval Stages of Coenobita purpureus Stimpson and C. cavipes
Stimpson Reared in the Laboratory and Survival Rates
and Growth Factors of Three Land Hermit
Crab Larvae (Crustacea: Anomura)

Yukio Nakasone

Biological Laboratory, College of Education, University
of the Ryukyus, Okinawa 903-01, Japan

ABSTRACT — The complete larval development of Coenobita purpureus and the glaucothoe of C.
cavipes are described and illustrated based on larvae reared in the laboratory. C. purpureus has five
zoeal and one glaucothoe stages. It seems very likely that the first zoea of C. rugosus described by
Yamaguchi belongs to C. purpureus. The mean points of 50% survivorship of starved larvae of C.
purpureus, C. rugosus and C. cavipes at a seawater temperature of 28±0.9°C were about day 8, 11 and
7, respectively and none of them molted to the next stage; those of fed larvae were about day 19, 19, 16,
respectively. The duration of the fourth stage was shortened in C. rugosus and C. purpureus. The
greatest growth occurred between the second and third zoeal stages in C. purpureus and C. rugosus,
between the fourth and fifth stages in C. cavipes. The first glaucothoes of C. rugosus, C. purpureus and
C. cavipes emerged on the 16th, 17th and 25th days of their larval life, respectively. After the
appearance of glaucothoes, survival rates of the larvae rapidly declined due to both predation of the fifth
zoeae by the glaucothoes and cannibalism of the glaucothoes. The death during the molt to glaucothoes
also was another factor that lowered the survival rates. Some glaucothoes of C. purpureus entered shells
on the 10th day after their appearance.

INTRODUCTION

Borradaile [1] first pointed out that the youngs
of Coenobita rugosus and C. perlatus hatch as
zoeal larvae which are released in the sea. He [2]
illustrated the first zoea of C. perlatus. The first
zoea and glaucothoe of C. rugosus were described
and illustrated by Yamaguchi [3] who studied the
reproductive ecology of this species in Kikaijima
Island of the Amami Islands. Provenzano [4] was
the first to report on the complete larval develop¬
ment in Coenobita, describing and illustrating in
detail the larval development of C. clypeatus.
Recently, Shokita and Yamashiro [5] described
and illustrated the larval development of C. rugo¬
sus and the zoeal stages of C. cavipes.

It is known that some land hermit crabs and the
coconut crab of the family Coenobitidae pass

through five zoeal and one glaucothoe stages in the
larval development [4-6]. The shell use of
glaucothoe has also been observed [4, 6, 7]. The
mean total growth and mean incremental growth
factors of zoeae have been calculated on the basis
of the larvae of two species [8]. However, the
survival rates of zoeal larvae under fed and starved
conditions are little known for any species of the
family.

During my current studies on the ecology and
distribution of land hermit crabs in Okinawa Pre¬
fecture, Japan, I had an opportunity of studying
the larval development and survival rates of C.
purpureus, C. rugosus and C. cavipes under
laboratory conditions. In the present paper, the
complete larval development of C. purpureus and
the glaucothoe stage of C. cavipes are described,
and the larval survival rates and growth factors of
the three species are noted.

Accepted January 20, 1988
Received October 19, 1987

1106

Y. Nakasone

MATERIALS AND METHODS

One ovigerous female of C. cavipes was col¬
lected from Hyakuna, southern Okinawa, at night
on 6 July, that of C. purpureus from the same place
on 21 July and C. rugosus from the same place on 5
August, 1986. They released the larvae shortly
after being placed in a plastic bucket containing
sea water.

Three groups each of 50 zoeae of each species
were reared in plastic bowls each containing 1.2
liters of filtered, moderately aerated sea water.
Newly-hatched Artemia nauplii were given as food
daily and the larvae were transferred to fresh
filtered sea water every four or five days. Water
temperature was 28±0.9°C. The larvae of each
stage were preserved in 5% seawater-formalin.
Dissection was done in glycerin. The total and
carapace lengths of the larvae were measured
before dissection using an ocular micrometer. To¬
tal length was measured from the rostral tip to the
posteromedial end of the telson exclusive of setae.
Carapace length was measured from the rostral tip
to the posteromedial end of the carapace. Draw¬
ings were made with the aid of an ocular micro¬
meter. Three or four larvae of each stage were

examined for confirmation of zoeal characters.

To compare survival rates of larvae of C. rugo¬
sus, C. purpureus and C. cavipes, 6 groups each of
50 zoeae of each species were reared in bowls. For
feeding and starvation experiments, 3 of the 6
bowls were given Artemia nauplii and the other 3
were not. Each bowl was examined daily for live
and dead larvae, which were confirmed under
binocular stereomicroscope.

RESULTS

Coenobita rugosus, C. purpureus and C. cavipes
passed through five zoeal and one glaucothoe
stages, but the glaucothoes all died without molt¬
ing to the first crab stage. For zoeal stages of C.
cavipes and complete larval stages of C. rugosus,
the reader is referred to Shokita and Yamashiro
[5]. The duration of zoeal and glaucothoe stages
and the sizes of larvae of these three species are
summarized in Figure 10 and Table 1.

Description of zoeal stages of C. purpureus

First zoea

Carapace (Fig. 1, A): Long pointed rostrum

Table 1. Comparison of total and carapace lengths (mean + SD) in each stage zoea and glaucothoe of
three species of Coenobita

Species

Size of each stage zoea and glaucothoe (mm)

I

II

III

IV

V

G

Coenobita

TL*

2.70

3.27

4.20

4.57

5.15

4.28

purpureus

±0.03

±0.11

±0.14

±0.05

±0.13

±0.04

(n = 5)

(n=4)

(n=5)

(n=4)

(n=5)

(n = 5)

CL**

1.16

1.39

1.91

2.26

2.34

1.50

±0.04

±0.04

±0.08

±0.02

±0.09

±0.04

(n=5)

(n=4)

(n = 5)

(n=4)

(n = 5)

(n = 5)

Coenobita

TL

2.59

3.10

4.23

4.48

4.85

4.00

rugosus

±0.04

±0.15

±0.07

±0.19

±0.08

±0.02

(n=5)

(n = 5)

(n=5)

(n=5)

(n=5)

(n = 5)

CL

1.16

1.45

1.91

2.02

2.16

1.37

±0.03

±0.07

±0.06

±0.05

±0.10

±0.07

(o — 5)

(n=5)

(n = 5)

(n=5)

(n = 5)

(o = 5)

Coenobita

TL

2.26

2.68

3.25

3.78

5.13

4.33

cavipes

±0.04

±0.05

±0.05

±0.14

±0.15

±0.21

(n = 5)

(n=5)

(n = 4)

(n=5)

(n=4)

(n=4)

CL

1.00

1.19

1.41

1.67

2.31

1.39

±0.02

±0.08

±0.04

±0.08

±0.10

±0.11

(n = 5)

(n = 5)

(n = 4)

(n=5)

(n=4)

(o=4)

*) Total length from rostral tip to posterior margin of telson.

**) Carapace length from rostral tip to posteromedial end of carapace.

Development and Survival of Coenobita Larvae

1107

Fig. 1. Coenobita purpureus. First (A, a) to fifth (E, e) zoeae. Upper, dorsal view (A-E); lower, lateral view (a-e).
Scales show 1.0 mm.

extending beyond tip of antennal scales. Rostrum
and carapace smooth. Eyes sessile.

Antennule (Fig. 2, A): Unsegmented, termi¬
nally provided with 3 aesthetascs (one of them
short) and 3 plumose setae (one of them long),
subterminally with 1 long plumose seta.

Antenna (Fig. 2, F): Biramous; unsegmented
endopod fused with protopod, bearing 3 terminal
plumose setae. Antennal scale distolaterally pro¬
duced into sharp spine, provided with 10 plumose
setae on distal and mesial margins, one of them
short, directly inner to distolateral spine, and with
fine setules on inner margin. Protopod provided
with 1 setiferous spine subterminally.

Mandible: No observation.

Maxillule (Fig. 3, A): Three-segmented en¬
dopod with 2 terminal setae. Basal endite with 2
strong denticulate spines and 2 short simple setae.

Coxal endite with 5 long plumose and 2 short
simple setae.

Maxilla (Fig. 3, F): Endopod unsegmented,
bearing 3 terminal and 2 subterminal setae. Pro¬
ximal lobe of basal endite with 5 setae and distal
lobe with 3 setae. Proximal lobe of coxal endite
with 7 setae and distal lobe with 4 setae. Scaphog-
nathite incomplete, posterior lobe absent, anterior
lobe with 4 plumose setae.

First maxilliped (Fig. 4, A): Basis provided
with 1 hook-like process proximally, setal formula
progressing distally 2, 3, 2. Exopod with 4 nata¬
tory setae. Endopod five-segmented, setal formula
progressing distally 2-2-3-2-5; second to fourth
segments with some setules.

Second maxilliped (Fig. 4, F): Basis with 2
setae. Exopod with 4 natatory setae. Endopod
four-segmented, setal formula progressing distally

1108

Y. Nakasone

Fig. 2. Coenobita purpureus. Antennules (A-E) and antennae (F-J) of zoeal stages I-V. Scales 0.2 mm.

1-2-2-5; second and third segments with some
setules.

Third maxilliped (Fig. 5, A): Uniramous and
rudimentary.

Abdomen (Fig. 1, A): Consisting of 5 somites
and telson. Second somite with 1 conspicuous
middorsal spine, third and fourth somites each
with 1 very small middorsal spinule, fifth somite
with 1 large middorsal spine and 1 large postero¬
lateral spine on each side.

Telson (Fig. 6, A): Bearing 7+7 processes,
outermost one (first) a fixed spine, second a fine
simple hair and third to seventh articulated plu¬
mose setae, fifth and seventh of them provided
with spinules; posteromedian notch semicircular;
posterior margin of telson provided with spinules
on notches.

Second zoea

Carapace (Fig. 1, B): Similar to first stage, but
eyes mobile.

Antennule (Fig. 2, B): Provided terminally
with 4 aesthetascs (one of them short) and 3

plumose setae (one of them long), subterminally
with 3 short and 1 long plumose setae.

Antenna (Fig. 2, G): Similar to first stage.
Basis with 1 tiny spine near junction with scale in
addition to 1 distinct subterminal spine.

Mandible: No observation.

Maxillule (Fig. 3, B): Endopod nearly as in
first stage, but one of two terminal setae plumose.
Basal endite with 4 strong denticulate spines and 2
short smooth setae. Coxal endite as in previous
stage.

Maxilla (Fig. 3, G): Endopod with 5 setae.
Setal formulae of basal and coxal endites as repre¬
sented by 4 + 3 and 7 + 4. Incomplete scaphog-
nathite with 8 plumose setae.

First maxilliped (Fig. 4, B): Setal formula of
basis progressing distally 1, 1, 3, 3. Exopod with 6
natatory setae. Setal formula of five-segmented
endopod progressing distally 4-3-2-1-5.

Second maxilliped (Fig. 4, G): Basis with 3
setae. Exopod with 6 natatory setae. Setal formu¬
la of four-segmented endopod progressing distally
2-3-3-5.

Development and Survival of Coenobita Larvae

1109

Fig. 3. Coenobita purpureus. Maxillules (A-E) and maxillae (F-J) of zoeal stages I-V. Scales 0.1 mm.

Third maxilliped (Fig. 5, B): Now composed
of exopod with 5 natatory setae and lobe-like
endopod.

Pereiopods (Fig. 1, b): Very small buds found
behind third maxilliped in lateral view.

Abdomen (Fig. 1, B): Sixth somite still not
separated from telson.

Telson (Fig. 6, B): 8 + 8 processes including

additional pair of short plumose setae; fifth to

eighth plumose setae with spinules. Median telson
notch indistinct.

Third zoea

Antennule (Fig. 2, C): Two-segmented; distal
segment with 5 aesthetascs (one of them short) in
addition to 2 short plumose setae; proximal seg¬
ment with 3 long plumose and 4 short smooth setae
and bud of inner flagellum small.

1110

Y. Nakasone

Fig. 4. Coenobita purpureus. First (A-E) and second (F-J) maxillipeds of zoeal stages I-V. Scale 0.2 mm.

Fig. 5. Coenobita purpureus. Third maxillipeds (A-E)
of zoeal stages I-V. Scale 0.2 mm.

Antenna (Fig. 2, H): Endopod articulated
with basis, bearing a terminal seta. Scale with 13
plumose setae.

Mandible: No observation.

Maxillule (Fig. 3, C): Endopod and basal en-
dite as in preceding stage, but denticulate spines
on basal endite relatively large. Coxal endite with
6 plumose and 2 short simple setae.

Maxilla (Fig. 3, H): Setation of endopod as in
second zoea. Setal formula of basal endite 5+4,
that of coxal endite 9 + 4. Scaphognathite incom¬
plete, anterior lobe with 10 plumose setae.

First maxilliped (Fig. 4, C): Exopod with 6
natatory setae. Setal formula of five-segmented
endopod progressing distally 3-3-2-2-5.

Second maxilliped (Fig. 4, FI): As in previous
stage.

Third maxilliped (Fig. 5, C): Exopod with 6
natatory setae. Endopod somewhat larger than in
previous stage, bearing 1 short terminal seta.

Abdomen (Fig. 1, C): Sixth somite distinct,
provided with sharp mediodorsal spine. Sixth
somite provided with uropods each consisting of
exopod and endopod; exopod with 8 plumose
setae, endopod naked, much smaller than exopod.

Telson (Fig. 6, C): 8 + 1 + 8 processes; a short

median plumose seta flanking 8 processes; first
process being a short spine, second a fine simple

Development and Survival of Coenobita Larvae

1111

Fig. 6. Coenobita purpureus. Telsons and uropods (A-E) of zoeal stages I-V. Scale 0.3 mm.

hair, third an articulated plumose seta, fourth a
large fused spine, fifth to eighth articulated plu¬
mose setae; third, fifth to eighth and median
plumose setae with spinules.

Fourth zoea

Antennule (Fig. 2, D): Bud of inner flagellum
distinct. Distal segment (outer flagellum) with 4
aesthetascs (two of them long) and 1 long and 1
short plumose setae, proximal segment with 4 very
short smooth and 4 long plumose setae (one of
them subterminal).

Antenna (Fig. 2, I): Unchanged.

Mandible: No observation.

Maxillule (Fig. 3, D): Basal endite with 6
strong denticulate spines and 2 short smooth setae;
coxal endite with 6 pronounced and 2 short setae.

Maxilla (Fig. 3, I): Setation of endopod as in
previous stage. Basal and coxal endites with 5 + 4
and 8 + 4 setae, respectively. Anterior lobe of

scaphognathite with 13 plumose setae.

First maxilliped (Fig. 4, D): Natatory setae as
in previous stage. Setal formula of endopod prog¬
ressing distally 4-3-2-2-5.

Second maxilliped (Fig. 4, I): Exopod with 8
natatory setae; setation of endopod unchanged.

Third maxilliped (Fig. 5, D): Nearly as in pre¬
vious stage, but 1 subterminal plumose seta on
endopod.

Abdomen: Exopod and endopod of uropod
(Fig. 6, D) distinctly articulated with protopod,
exopod provided with 1 sharp posterolateral mar¬
ginal tooth and 10 plumose setae, endopod with 6
plumose setae.

Telson (Fig. 6, D): Unchanged.

Fifth zoea

Antennule (Fig. 2, E): Distal segment (outer
flagellum) bearing 4 terminal (one of them short)
and 2 subterminal (one of them short) aesthetascs

1112

Y. Nakasone

and 3 plumose setae (one of them long); proximal
segment with 4 short smooth and 4 long plumose
setae (one of them subterminal). Bud of inner
flagellum remained undeveloped.

Antenna (Fig. 2, J): Endopod two-segmented,
terminating in 1 seta, scale with 14 plumose setae.

Mandible: No observation.

Maxillule (Fig. 3, E): Unchanged.

Maxilla (Fig. 3, J): Scaphognathite still incom¬
plete, anterior lobe bearing 17 plumose setae.

First maxilliped (Fig. 4, E): Nearly as in pre¬
vious stage, setal formula of endopod progressing
distally 3-3-2-2-5.

Second maxilliped (Fig. 4, J): Unchanged.

Third maxilliped (Fig. 5, E): Similar to pre¬
vious stage but endopod larger.

Buds of pereiopods and pleopods (Fig. 1,
e): More distinct.

Abdomen: Exopod and endopod of uropod
(Fig. 6, E) with 11 and 7 plumose setae, respec¬
tively.

Telson (Fig. 6, E): Unchanged.

Description of glaucothoe stages of C. purpureus
and C. cavipes

The following description is generalized for two
species except differences that are mentioned.

Carapace (Fig. 7, A): Shorter than abdomen
including telson. Rostrum rounded, extending
beyond front. Eyes and eyestalks well-developed.

Antennule (Fig. 7, B): Bearing unsegmented
outer and inner flagella; outer flagellum with 8
aesthetascs and a few setae; inner one with 4
terminal setae in C. purpureus , 5 setae (3 terminal
and 2 at midlength) in C. cavipes ; 2 proximal

C-F, C'-F'.

Development and Survival of Coenobita Larvae

1113

segments each with a few short setae.

Antenna (Fig. 7, C): Composed of 3 peduncu¬
lar segments and six-segmented flagellum; termi¬
nal flagellar segment with 1 long apical and a few
short setae, other segments each with a few setae;
scale reduced to a small knob.

Mandible (Fig. 7, D): Palp three-segmented,
ultimate segment with 6-9 plumose setae in C.
purpureus, 8-11 plumose setae in C. cavipes;
cutting edge of molar process chitinous and brown¬
ish.

Maxillule (Fig. 7, E): Endopod unsegmented
in C. purpureus , two-segmented in C. cavipes ;
bearing 2 short setae, one of them arising from a
process on proximal half of unsegmented endopod
or proximal segment of two-segmented endopod;
basal endite bearing 8 setae (6 plumose and 2
smooth) and 15 spinules in C. purpureus , 13 setae
(4 plumose and 9 smooth) and 12 spinules in C.
cavipes ; coxal endite with 15 plumose setae in C.
purpureus and 16 plumose and 1 smooth setae in
C. cavipes.

Maxilla (Fig. 7, F): Endopod with 2 setae in C.
purpureus, 1 seta in C. cavipes', distal lobe of basal
endite with 2 plumose and 8 smooth setae in C.
purpureus, 10 smooth setae in C. cavipes', proximal

lobe with 7 setae; distal lobe of coxal endite with 7
setae in C. purpureus, 6 setae in C. cavipes',
proximal lobe with 16 plumose and 11 smooth
setae in C. purpureus, 29 smooth setae in C.
cavipes', scaphognathite with posterior lobe com¬
plete, bearing 51 plumose setae in C. purpureus,
75 in C. cavipes.

First maxilliped (Fig. 8, A): Endopod unseg¬
mented; without setae in C. purpureus, bearing 2
terminal setae in C. cavipes', exopod bearing 6
lateral marginal plumose setae in C. purpureus, 1
short terminal and 7 lateral marginal plumose
setae in C. cavipes', basal and coxal lobes bearing
18 and 10 plumose setae respectively in C. pur¬
pureus, 20 and 7 in C. cavipes.

Second maxilliped (Fig. 8, B): Endopod four-
segmented, each segment bearing a few plumose
setae; exopod two-segmented, distal segment
bearing 6 plumose setae in C. purpureus, 7 in C.
cavipes.

Third maxilliped (Fig. 8, C): Endopod five-
segmented, each segment with a few or numerous
setae; exopod three-segmented, ultimate segment
setaless in C. purpureus, bearing 1 plumose seta in
C. cavipes.

Chelipeds (Fig. 8, D): Similar and almost

Fig. 8. Coenobita purpureus (A-E) and C. cavipes (A'-E'), glaucothoes. A, A', first maxillipeds; B, B , second
maxillipeds; C, C', third maxillipeds; D, D\ left chelipeds; E, E\ second pereiopods. Scales 0.2 mm for A-C,
A'-C’, 0.3 mm for D, D\ E, E'.

1114

Y. Nakasone

equal in length; dactylus subequal in length to
palm; each segment with scattered setae.

Second (Fig. 8, E) and third pereiopods: Each
ending in a corneous claw, bearing a few scattered
setae.

Fourth pereiopod (Fig. 9, A): Sparsely setose
on each segment; propodus bearing some corneous
granules or blunt spines in C. purpureus , two rows
of spinules in C. cavipes, dactylus with 1 long and a

few short setae.

Fifth pereiopod (Fig. 9, B): Distal two seg¬
ments sparsely setose, some setae long and curved;
bearing a few corneous granules and spines in C.
purpureus , corneous spines only in C. cavipes.

Pleopods: Present on second (Fig. 9, C) to fifth
abdominal somites; biramous; endopod being a
small lobe, bearing 2 short, curved setae subtermi-
nally; exopod well-developed, bearing 9 plumose

Fig. 9. Coenobita purpureus (A-D) and C. cavipes (A'-D'), glaucothoes. A, A', fourth pereiopods; B, B\ fifth
pereiopods; C, C\ second pleopods; D, D\ telsons and uropods. Scales. 0.2 mm.

Development and Survival of Coenobita Larvae

1115

setae.

Abdomen (Fig. 7, A): Dorsal and lateral
spines absent. Uropodal exopod (Fig. 9, D) of C.
purpureus with 22 plumose and a few smooth setae
and 11 corneous blunt spines on lateral margin,
endopod of same with 11 plumose and a few
smooth setae and 8 corneous blunt spines on
lateral margin. Setation and spination of uropod in
C. cavipes somewhat reduced in number, exopod
with 18 long plumose and a few short smooth setae
and 4 corneous spines, endopod with 8 long plu¬
mose and a few short smooth setae and 2 or 3
corneous spines. Protopod of uropod with 3
plumose and 1 short smooth setae on lateral mar¬
gin; sixth abdominal somite sparsely provided with
smooth setae on posterior margin.

Fig. 10. Survivorship curves for starved (solid circles)
and fed (open circles) larvae of Coenobita purpureus
(A), C. rugosus (B) and C. cavipes (C) at 28±0.9°C.
Horizontal bars (I-V) indicate the duration of zoeal
larvae from first to fifth stages; arrows of bars (G)
show existence of glaucothoes.

Telson (Fig. 9, D): Bearing 9 long plumose
setae on posterior margin, some smooth setae on
lateral margin and dorsal surface.

Pereiopods and pleopods well-developed, func¬
tional.

Survival rates of larvae of three Coenobita species
and duration of each zoeal stage

Survivorship curve of larvae of each species and
the duration of each zoeal stage in both fed and
starved conditions are shown in Figure 10. The
starved first zoeal larvae of C. purpureus , C.
rugosus and C. cavipes all died in 12, 15 and 9 days
after hatching, respectively and none of them
molted to the next stage. The mean points of 50%
survivorship of starved larvae of C. purpureus , C.
rugosus and C. cavipes were about day 8, 11 and 7,
respectively. The larvae of C. rugosus had greater
tolerance for starvation than those of the other two
species. The mean points of 50% survivorship of
fed larvae were about day 19 in both C. purpureus
and C. rugosus and day 16 in C. cavipes. The
death rates of the first and second zoeae of C.
cavipes were higher than those of the other two
species, thus their survivorship curve showed a
different pattern.

The durations of the first to fifth zoeal stages
were 6, 4, 5, 2, 10 days in C. purpureus, 5, 4, 5, 1,
10 days in C. rugosus and 7, 8, 5, 10, 15 days in C.
cavipes. The fourth zoeae of C. rugosus and C.
purpureus molted to the fifth zoeae in one or two
days. But, most of C. rugosus third zoeae directly
molted to the fifth zoeae. The duration used here
indicates the length of time that all survivors in a
certain stage needed till the completion of the
succeeding molt, although the speeds of molting
differed with individuals.

The appearance of glaucothoes obtained in this
study ranged from the 17th to 23rd days of the
larval life in C. purpureus, from the 16th to 19th
days in C. rugosus and from the 25th to 38th days
in C. cavipes. The first glaucothoes emerged on
the 5th day after the appearance of the fifth stage
zoeae in C. purpureus, on the 4th day in C. rugosus
and on the 2nd day in C. cavipes. The survival
rates of larvae (including the fifth zoeae and
glaucothoes) after the appearance of glaucothoes

1116

Y. Nakasone

Zoeal and glaucothoe stages

Fig. 11. Comparison of larval growth rates of three species of Coenobita.

rapidly declined in all the three species, as shown
in Figure 10. Three of the 8 glaucothoes of C.
purpureus used in shell utilization experiment en¬
tered small shells provided in an aquarium on the
10th day after their appearance, one glaucothoe
did not enter shell. For the remaining four it was
not certain whether they had died or hurried in
sand provided in the aquarium.

Total and carapace lengths of each stage zoea and
growth factors

The mean total and the mean carapace lengths
of each stage zoea of the three Coenobita species
are summarized in Table 1 and are plotted in
Figure 11. Both the total and the carapace lengths
of C. cavipes zoeae from the first to fourth stages
were usually small as compared with those of C.
purpureus and C. rugosus. C. cavipes showed
nearly a linear growth in both the total and the
carapace lengths from the first to fourth zoeal
stages. The second stage zoeae of C. purpureus
and C. rugosus rapidly grew when they molted to
the third stage, while the fourth stage zoeae of C.
cavipes did when they molted to the fifth stage.
And finally the last (fifth) stage zoeae of C. cavipes
and C. purpureus attained to the same size. Also,
the fifth stage zoeae of C. rugosus reversely be¬
came much smaller than those of C. purpureus and
C. cavipes which are large-sized species. However,

the glaucothoes of C. rugosus showed almost the
same size as those of C. cavipes in carapace length.
All the glaucothoes obtained of the three species
became smaller than the fifth stage zoeae in both
total and carapace lengths, but this was due to
inclusion of rostrums for measurement.

The zoeal larvae of C. purpureus, C. rugosus
and C. cavipes showed the mean total and the
mean growth factors (cf. Gore [8]) of 1.91, 1.18,
1.87, 1.18 and 2.27, 1.23, respectively. These
values were calculated using total length. Thus,
the greatest total growth factor was 2.2 7* in C.
cavipes. Instar growth factors using total length
ranged from 1.09 to 1.28 in C. purpureus, from
1.06 to 1.36 in C. rugosus, and from 1.16 to 1.36 in
C. cavipes. Both C. rugosus and C. cavipes had the
same greatest instar growth factor (1.36), but the
stages were different: between the second and
third stages in C. rugosus, between the fourth and
fifth stages in C. cavipes, as shown in Figure 11.

DISCUSSION

As shown in Table 2, zoeal characters in each
stage were so similar among the three species that
it was difficult to find the recognition characters for
each species. Also, there were some observational
differences in minor features such as setation of
appendages in the same species between Shokita
and Yamashiro’s [5] and the present studies. The

Development and Survival of Coenobita Larvae 1117

Table 2. Major differences in zoeal characters among Coenobita rugosus [5], C. cavipes [5] and C.

purpureus

Item

C. rugosus

C. cavipes

C. purpureus

Zoea I

Antennular aesthetascs

3(4)

3(3)

3

Setation of antenna

Endopod and scale

3(2) and 10(10)

3(2) and 10(8)

3 and 10

Setation of maxillule

Endopod and coxal endite

3(2) and 7(5)

2(2) and 7(5)

2 and 7

Setation of maxilla

Scaphognathite

5(3)

4(4)

4

Endopod

3+2(3 + 2)

2 + 2(4+4)

3+2

Basal endite

4 + 4(3 + 3)

4+4(3 + 2)

3 + 5

Coxal endite

4+7(3+7)

4+7(2+2)

4+7

Setation of 1st maxilliped

Exopod

4(3)

4(3)

4

Setation of 2nd maxilliped

Exopod

4(3)

4(3)

4

Zoea II

Setation of antenna

Scale

10(10)

10(10)

10

Setation of maxillule

Endopod and coxal endite

2(2) and 7(5)

2(2) and 7(4)

2 and 7

Setation of maxilla

Scaphognathite

6(5)

7(7)

8

Basal endite

4+5(4+3)

4 + 5(4+3)

3+4

Coxal endite

4+7(3 + 3)

4+7(2+3)

4+7

Setation of 1st maxilliped

Exopod

6(3)

6(5)

6

Setation of 2nd maxilliped

Exopod

6(4)

6(4)

6

Setation of 3rd maxilliped

Exopod

5(2)

5(4)

5

Zoea III

Setation of antenna

Scale

13(13)

12(13)

13

Setation of maxillule

Endopod and coxal endite

2(2) and 8(6)

2(2) and 7(7)

2 and 8

Setation of maxilla

Scaphognathite

8(10)

10(7)

10

Basal endite

4+4(4+4)

4+5(4+4)

4+5

Coxal endite

4+7(4+3)

4+7(3 + 6)

4 + 9

Setation of 1st maxilliped

Exopod

6(5)

6(5)

6

Setation of 2nd maxilliped

Exopod

6(5)

6(5)

6

1118

Y. Nakasone

Table 2. (Continue)

Item

C. rugosus

C. cavipes

C. purpureus

Zoea IV

Setation of antenna

Scale

14(14)

15(15)

13

Maxillule

Spines of basal endite

5(4)

6(5)

6

Setation of maxilla

Scaphognathite

11(10)

11(12)

13

Setation of 1st maxilliped

Exopod

6(5)

6(4)

6

Setation of 2nd maxilliped

Exopod

8(5)

7(6)

8

Zoea V

Antennular aesthetascs

4 + l(5 + 2)

4+2(5 + 2)

4+2

Setation of antenna

Scale

15(15)

16(15)

14

Setation of maxilla

Scaphognathite

15(11)

17(10)

17

Setation of 1st maxilliped

Exopod

6(5)

6(6)

6

This table was made on the basis of Shokita and Yamashiro [5].

Figures in parentheses are based on Shokita and Yamashiro’s study [5],

first zoea of C. rugosus reported by Yamaguchi [3]
shares some characters with that of C. purpureus.
As pointed out by Shokita and Yamashiro [5],
Yamaguchi’s description on the coxae of the fifth
legs in the adult male of C. rugosus fits well the
character of C. purpureus , which differs from that
of C. perlatus [9]. In Kikaijima Island where
Yamaguchi studied C. purpureus was the most
abundant species, but the other species were not
found [10]. Judging from these facts, it seems very
likely that the first zoea of C. rugosus described by
Yamaguchi belongs to C. purpureus. The zoeal
characters of the three species were nearly similar
to those of C. clypeatus [4] except minor differ¬
ences such as the setation of appendages.

The general features of the glaucothoe of the
three species are similar, but segmentations and
setations of appendages are different, as shown in
Table 3. It seems possible that these glaucothoe
larvae are separated by the combination of some of
the characters listed in it. Yamaguchi’s glaucothoe
differed from those of C. purpureus and C. rugo¬

sus , both obtained in this study, in that the antenna
has seven flagellar segments as in C. clypeatus [4]
and the antennule has two outer flagellar seg¬
ments.

Figure 10 shows that the larvae of C. rugosus
had greater tolerance for starvation than those of
the other two species. This may suggest that the
larvae of C. rugosus have a more fortunate chance
of survival when they suffered a temporary food
shortage in natural environments. The survival
rates of fed zoeal larvae were higher in C. pur¬
pureus and lowest in C. cavipes. Fed fourth zoeae
of C. rugosus and C. purpureus molted to the fifth
zoeae in one or two days, but those of C. cavipes
took seven to ten days. However, the duration of
the fifth zoeae was reversely longer in C. pur¬
pureus and C. rugosus than in C. cavipes.

According to my interpretation of Provenzano
[4], the mean instar growth factor was 1.17 be¬
tween the first and second stages and 1.11 between
the fourth and fifth stages in C. clypeatus , in which
the greatest instar growth factor was 1.21 between

Development and Survival of Coenobita Larvae

1119

Table 3. Major differences in glaucothoe characters among Coenobita rugosus, C. cavipes [5] and C.

purpureus

Item

C. rugosus

C. cavipes

C. purpureus

Antennular aesthetascs

7(6)

8

8

Antenna

No. of flagellar segments

6(6)

6

6

Mandible

No. of palpal segments and

2(0) and 5-8(8)

3 and 8-11

3 and 6-9

setae on distal segment

Maxillule

Segmentation of endopod

segmented(non)

segmented

non

Setation of maxilla

Endopod

1(1)

1

2

Scaphognathite

51(50)

75

51

Basal endite

10 + 5(7 + 4)

10 + 7

10 + 7

Coxal endite

5 + 21(7 + 19)

6+29

7+27

Setation of 1st maailliped

Endopod

0(2)

2

0

Exopod

5 + 2(9)

7+1

6+0

Basis

Distal lobe

13(12)

20

18

Proximal lobe

9(11)

7

10

2nd maxilliped

Segmentation of exopod

3(1)

2

2

Setae on distal segment

7(4)

7

6

of exopod

3rd maxilliped

Segmentation of exopod

2(2)

3*

3

Setation of telson

9(11)

9

9

Setation of uropod

Endopod

10(13)

8

11

Exopod

17(22)

18

22

*) Bearing one terminal plumose seta on distal segment.

Figures in parentheses are based on Shokita and Yamashiro’s study [5].

the third and fourth stages; the mean growth factor
was 1.17 and the mean total growth factor was
1.85. These values were calculated using total
length. In Birgus latro, the maximum of instar
growth factor was 1.21 between the first and
second stages, the mean growth factor was 1.13
and the total growth factor was 1.64, by my
interpretation of Reese and Kinzie [6]. The
growth factors of five species of Coenobitidae
including the above two and the three species of
Coenobita here studied were 1.18 in mean in¬

cremental growth and 1.91 in mean total growth,
which were nearly similar to the values obtained by
Gore [8]. Also, the instar growth factor ranged
from 1.06 in C. rugosus to 1.36 in both C. rugosus
and C. cavipes. The total growth factor ranged
from 1.64 in B. latro to 2.27 in C. cavipes.

After the appearance of glaucothoes, the surviv¬
al rates of larvae (including fifth zoeae and
glaucothoes) rapidly declined in the three species
studied (Fig. 10). This decline was due to preda¬
tion of fifth zoeae by glaucothoes and cannibalism

1120

Y. Nakasone

of glaucothoes. The death occurred during the
molt to glaucothoes also was one of the factor that
lowered the survival rates. In C. purpureus , the
predation of the fifth zoeae by the glaucothoes was
the chief (first) factor in the decline of the survival
rate and the cannibalism of the glaucothoes was
the next factor. The death during the molt was the
main factor and cannibalism was the next in C.
rugosus. Both the death during the molt and
cannibalism were the chief factors in C. cavipes. It
seems likely that both the predation and cannibal¬
ism result from the narrow rearing bowl and do not
occur in the sea.

ACKNOWLEDGMENTS

I thank Mr. M. Toyama, Cultural Administration Sec¬
tion, Okinawa Prefectural Office of Education, for the
opportunity of studying the larval development of land
hermit crabs, Messrs. E. Matsumoto, N. Miyazato and
M. Zaitsu, graduates of College of Education, University
of the Ryukyus, for rearing the larvae and illustrations. I
express my thanks to two anonymous reviewers whose
comments improved the manuscript essentially. This
work was supported by Agency for Culture Affairs, the
Ministry of Education, Science and Culture of Japan.

REFERENCES

1 Borradaile, L. A. (1899) A note in the hatching-
stage of the pagurine land-crabs. Proc. Zool. Soc.
London, 1899: 937-938.

2 Borradaile, L. A. (1903) Land crustaceans. In “The
Fauna and Geography of the Maidive and Laccadive
Archipelagoes”. Ed. by J. S. Gardiner, Cambridge
Univ. Press, Cambridge, vol. 1, pp. 64-100, pi. 3.

3 Yamaguchi, S. (1938) Spawning and development
of Coenobita rugosus. Bulteno Science, Imp. Univ.
Kyushu, 8: 163-177, 2 pis. (In Japanese with English
summary).

4 Provenzano, A. J., Jr. (1962) The larval develop¬
ment of the tropical land hermit Coenobita clypeatus
(Herbst) in the laboratory. Crustaceana, 4: 207-228.

5 Shokita, S. and Yamashiro, A. (1986) Larval de¬
velopment of the land hermit crabs, Coenobita rugo¬
sus H. Milne Edwards and C. cavipes Stimpson
reared in the laboratory. Galaxea, 5: 267-282.

6 Reese, E. S. and Kinzie, R. A., Ill (1968) The
larval development of the coconut or robber crab
Birgus latro (L.) in the laboratory (Anomura,
Paguridea). Crustaceana, Suppl., 2: 117-144.

7 Reese, E. S. (1968) Shell use: An adaptation for
emigration from the sea by the coconut crab. Sci¬
ence, 161: 385-386.

8 Gore, R. H. (1985) Molting and growth in decapod
larvae. In “Crustacean Issues 2, Larval Growth”.
Ed. by A. M. Wenner, A. A. Balkema, Rotterdam,
pp. 1-65.

9 Nakasone, Y. (1988) Land hermit crabs from the
Ryukyus, Japan, with a new species from the Philip¬
pines (Crustacea, Decapoda, Coenobitidae). Zool.
Sci., 5, 165-178.

10 Saisho, T. and Suzuki, H. (1987) An urgent study
on the distribution and ecology of land hermit crabs,
genus Coenobita, in Kagoshima Prefecture.
Kagoshima Prefectural School Board, 1-64. (In
Japanese).

ZOOLOGICAL SCIENCE 5: 1121-1136 (1988)

© 1988 Zoological Society of Japan

Biochemical Differentiation in Japanese Newts,
Genus Cynops (Salamandridae)

Terutake Hayashi and Masafumi Matsui1

Department of Zoology, Faculty of Science, Kyoto University, Kyoto 606,
and 1 Biological Laboratory, Yoshida College,

Kyoto University, Kyoto 606, Japan

ABSTRACT — Genetic variation was surveyed in 38 populations of Japanese newts of the genus Cynops
using starch gel electrophoresis. C. ensicauda from the Ryukyu Archipelago was shown to be
genetically well differentiated from C. pyrrhogaster from the Japanese main islands. Separation of these
two forms at the species level is supported. C. ensicauda is genetically divided into two groups, each of
which corresponds to previously recognized subspecies. On the contrary, protein variation patterns in
C. pyrrhogaster are not consistent with the previously recognized subspecies or local races. From
available geological information, the electrophoretic clock is calibrated at ID = 13-22 MY in Japanese
Cynops.

INTRODUCTION

Two allopatric newt species of the genus Cynops
are known from Japan. C. pyrrhogaster occurs on
the main islands of Honshu, Shikoku and Kyushu,
while C. ensicauda inhabits the Amami and Okina¬
wa Groups of the Ryukyu Archipelago. A marked
geographic variation in external morphology has
been detected within each species [1, 2]. Some
authors [3, 4] considered morphological variations
of these two species to overlap with each other and
doubted the specific validity of C. ensicauda, treat¬
ing it as a subspecies of C. pyrrhogaster. However,
only a few comparative studies have been made
between these two species [3, 5], and clarification
of the taxonomic relationships of these newts
requires an extensive survey of geographic varia¬
tion in Japanese Cynops from many approaches.

Although North American and European newt
species belonging to Taricha, Notophthalmus , and
Triturus have been studied electrophoretically for
the purposes of population genetics, taxonomy,
and evolutionary biology [6-9], no comparable
studies have been done on Asian newts. Available
data indicate that genetic distance values calcu-

Accepted January 14, 1988
Received December 15, 1987

lated between populations can differentiate named
species or subspecies and, therefore, seem to pro¬
vide rough estimates of the limits of species within
the family Salamandridae. Thus, an electrophore¬
tic analysis should be a useful tool for investigating
taxonomic problems among Japanese newts. The
present study was undertaken mainly in order to
understand the amount of genetic differentiation
between C. pyrrhogaster and C. ensicauda, as
estimated from an electrophoretic analysis of pro¬
tein variation. Moreover, based upon geological
data, we have derived a calibration for the elec¬
trophoretic evolutionary clock in Japanese Cynops
and compare the value with those proposed pre¬
viously [10, 11].

MATERIALS AND METHODS

A total of 610 newts from 23 populations of
Cynops pyrrhogaster in western Honshu and
Kyushu Islands and 15 populations of C. ensicauda
in three islands in the Amami Group and four in
the Okinawa Group, the Ryukyu Archipelago,
were analyzed electrophoretically (Fig. 1 and
Table 1). We used southwestern populations of C.
pyrrhogaster for comparison with C. ensicauda
since they are geographically adjacent to the range
covered by C. ensicauda.

1122

T. Hayashi and M. Matsui

Fig. 1. Geographic localities from which samples of
Cynops were collected. Localities of C. pyrrhogas¬
ter are indicated by circles (1-23) and localities of C.
ensicauda by triangles (24-38).

Samples of liver were removed and maintained
frozen at — 84 °C until used in electrophoresis.
Voucher specimens were fixed in 10% formalin,
later preserved in 70% ethanol and deposited in
Hayashi’s collection at Kyoto University.
Homogenized tissue extracts were analyzed by
standard horizontal starch gel electrophoresis [12-
15], using Connaught starch at a concentration of
11.5%. The enzymes examined and locus designa¬
tions are listed in Table 2. The buffer system
employed in the electrophoretic analysis was 0.155
M tris / 0.043 M citrate, pH 7.0 (1: 15 dilution of
electrode buffer for gel) for all enzymes. Genetic
interpretations of allozymic data were based on
criteria developed by Selander et al. [16]. Enzyme
nomenclature and E. C. numbers follow the most
recent recommendations of the Nomenclature
Committee of the International Union of Bio¬
chemistry [17] and abbreviations and isozyme de¬
signations follow recommendations of Murphy and
Crabtree [18]. Electromorphs were designated by
letters with “a” representing the most rapidly

migrating anodal variant.

The unbiased minimum genetic distance be¬
tween populations (D) recommended by Nei [19]
was computed from observed electromorph fre¬
quencies. According to Nei’s suggestion [19], all
negative values obtained using the collection for
small sample sizes were regarded as being equal to
0. A UPGMA phenogram [20] was constructed
from genetic distances. A contingency Chi-square
test was performed to test for inter-sample electro¬
morph frequency heterogeneity [21], All samples
were also tested for conformance to Hardy-
Weinberg expectations with the Chi-square test.
For statistical tests, P<0.05 was regarded as sig¬
nificant.

RESULTS

A locus was considered polymorphic when two
or more electromorphs were detected. Only one
of the 15 loci resolved ( Ap-A ) was monomorphic
for the same electromorph in all individuals. Table
3 summarizes electromorph frequencies for the
remaining 14 polymorphic loci. Fixed differences
between C. pyrrhogaster and C. ensicauda were
identified at three loci (Acp-A, Iddh-A and M-Me-
A).

Ten of the remaining 11 loci showed significant
heterogeneity in electromorph frequencies (Table
4). At four of these 11 loci, a single electromorph
predominated in all populations ( Ldh-A , Ldh-B,
M-Mdh-A and Pgdh-A ). At Ldh-A and M-Mdh-A
loci, electromorphs other than the common one
were unique to single populations. Among the
remainder of these four loci, some electromorphs
with low to moderate frequency of occurrence
were shared among two or more populations.
Seven other loci had different variants predomi¬
nating in different populations ( S-Aat-A , Est-1 ,
Gpi-A, S-Mdh-A, S-Me-A , Pgm-A, S-Sod-A ).

Within C. pyrrhogaster , three loci were mono¬
morphic (Acp-A, Ap-A and Iddh-A) and all of 12
polymorphic loci showed statistically significant
heterogeneity in electromorph frequencies (Table
4). C. ensicauda had six monomorphic loci (Ap-A,
Iddh-A, Ldh-A, M-Mdh-A, M-Me- A and S-Sod-
A) and significant heterogeneity in electromorph
frequencies was observed at seven of nine poly-

Allozymic Variation in Cynops

1123

Table 1. Species, sample size, and locality data for the animals used for electrophoretic analysis

Species

Population

number

Locality

N

Cynops pyrrhogaster

1

Shigaraki, Shiga

20

2

Miyama, Kyoto

39

3

Kyoto, Kyoto

16

4

Kameoka, Kyoto

8

5

Kumihama, Kyoto

4

6

Tottori, Tottori

19

7

Ningyo Pass, Okayama

20

8

Mt. Daisen, Tottori

8

9

Hirose, Shimane

20

10

Yamaguchi, Yamaguchi

20

11

Sanyo, Yamaguchi

20

12

Yukuhashi, Fukuoka

20

13

Higashiseburi, Saga

20

14

Isahaya, Nagasaki

21

15

Usuki, Oita

5

16

Shiranui, Kumamoto

20

17

Kamijima Isl. , Amakusa Isis.

18

18

Shimojima Isl., Amakusa Isis.

20

19

Minamata, Kumamoto

20

20

Tsuno, Miyazaki

20

21

Miyazaki, Miyazaki

17

22

Tano, Miyazaki

11

23

Kanoya, Kagoshima

18

Cynops ensicauda

24

Naze, Amami-Oshima Isl.

27

25

Mt. Kochi, Amami-Oshima Isl.

20

26

Ukejima Isl.

9

27

Yorojima Isl.

21

28

Kayauchibanta, Okinawajima Isl.

5

29

Mt. Yonaha, Okinawajima Isl.

8

30

Motobu, Okinawajima Isl.

5

31

Mt. Nago, Okinawajima Isl.

10

32

Ginoza, Okinawajima Isl.

10

33

Nakagusuku, Okinawajima Isl.

18

34

Tamagusuku, Okinawajima Isl.

10

35

Chinen, Okinawajima Isl.

9

36

Sezokojima Isl.

4

37

Zamamijima Isl.

30

38

Tokashikijima Isl.

20

1124

T. Hayashi and M. Matsui

Table 2. Enzymes and loci analysed in Japanese Cynops

Enzyme

Enzyme commis¬
sion number

Locus

Acid phosphatase

3. 1.3. 2

Acp-A

Aminopeptidase

3.4.11.1

Ap-A

Aspartate aminotransferase

2.6.1. 1

S-Aat-A

Esterase

—

Est-1

Glucose phosphate isomerase

5.3. 1.9

Gpi-A

L-iditol dehydrogenase

1.1.1.14

lddh-A

Lactate dehydrogenase

1.1.1.27

Ldh-A

Lactate dehydrogenase

1.1.1.27

Ldh-B

Malate dehydrogenase

1.1.1.37

M-Mdh-A

Malate dehydrogenase

1.1.1.37

S-Mdh-A

“Malic Enzyme”*

1.1.1.40

M-Me-A

“Malic Enzyme”*

1.1.1.40

S-Me-A

Phosphoglucomutase

5. 4.2.2

Pgm-A

Phosphogluconate dehydrogenase

1.1.1.44

Pgdh-A

Superoxide dismutase

1.15.1.1

S-Sod-A

Mitochondrial and supernatant loci are denoted by M- and S- prefixes, respectively.
*NADP-dependent malate dehydrogenase

morphic loci. Among Amami Group populations
of C. ensicauda, another monomorphic locus was
recognized ( Est-1 ) and five of eight polymorphic
loci were significantly heterogeneous. Among
Okinawa Group populations, eight loci were
monomorphic and four of seven polymorphic loci
showed significant heterogeneity.

The proportions of polymorphic loci ranged
from 20.0% (populations 1 and 22) to 53.3%
(populations 6, 13, 18 and 19) (x— 37.7%) in C.
pyrrhogaster and ranged from 13.3% (population
32) to 40.0% (population 33) (x=27.2%) in C.
ensicauda (Table 3). The mean number of electro-
morphs per locus was 1.44 (range 1.20-1.60) in C.
pyrrhogaster , and 1.32 (range 1.13-1.47) in C.
ensicauda. The frequencies of genotypes were in
good agreement with Hardy-Weinberg proportion
in most cases, but the significant heterozygote
deficiencies occurred at the S-Aat-A locus in one
population (population 33), at the M-Me-A locus
in five populations (populations 12, 13, 16, 19 and
20) and at the S-Me-A locus in six populations
(populations 2, 7, 8, 9, 14 and 16).

Figure 2 presents a UPGMA phenogram based
on the Nei’s D values, which are shown in Table 5.

The first major dichotomy separates populations of
C. ensicauda from those of C. pyrrhogaster with
the mean D value between them being 0.356
(range 0.239-0.724). The mean intraspecific D
values are 0.035 (range 0-0.133) in C. ensicauda
and 0.060 (range 0-0.336) in C. pyrrhogaster.

The cluster of C. ensicauda is divided into two
distinct regional groups, with the mean D value
between these two groups being 0.078 (range
0.041-0.133). One subcluster is composed of
populations from the Amami Group and another
of populations from the Okinawa Group. The
mean D values are 0.006 (range 0-0.013) within
the former and 0.004 (range 0-0.015) within the
latter.

The cluster of C. pyrrhogaster is also divided
into two distinct groups. One subcluster contains
three populations from southernmost part of
Kyushu (populations 21-23) and another contains
all the remaining populations. The mean D value
is 0.156 (range 0.044-0.336) between these two
groups.

The mean D value between populations of C.
ensicauda and three southernmost populations of
C. pyrrhogaster is 0.532 (range 0.326-0.724), while

Allozymic Variation in Cynops 1125

TableS. Electromorph frequencies and variability estimates for polymorphic loci in 38 populations of
Japanese Cynops

C. pyrrhogaster

Locus

Sasayama race

Hiroshima race

1

2

3

4

5

6

7

8

9

10

Acp-A

c

c

c

c

c

c

c

c

c

c

S-Aat-A

b

a(.069)

a(.107)

a(.125)

b

b(.974)

b

b

a(.139)

b

b(.914)

d(.017)

b(.893)

b(.875)

d(.026)

b(.861)

Est-1

b

b

b

b

b

b(.947)

b(.947)

b

b

b

c(.053)

c(.053)

Gpi-A

c(.975)

b(.013)

b(.063)

b(.125)

c

c

c

c

c

a(.150)

d(.025)

c(.884)

c(.906)

c(.750)

c(.750)

d(.103)

d(.031)

d(.125)

c(.100)

Iddh-A

b

b

b

b

b

b

b

b

b

b

Ldh-A

b

b

b

b

b

b

b

b

b

b

Ldh-B

c

c

c

c

a(.125)

a(.233)

a(.175)

a(.063)

c

c

c(.875)

c(.767)

c(.825)

c(.937)

M-Mdh-A

b

b

b

b

b

b

b

b

b

b

S-Mdh-A

c

b(.016)

c(.937)

c(.937)

c(.875)

a(.026)

a(.025)

c(.438)

c

c

c(.968)

d(.063)

e(.063)

d(. 125)

c(.974)

c(.725)

d(.562)

d(.016)

d(.250)

M-Me-A

b(.941)

b(.677)

b(.438)

b(.375)

b(.375)

b(.895)

b(.667)

b(.875)

b(.643)

b(.333)

c(.059)

c(.323)

c(.562)

c(.625)

c(.625)

c(.105)

c(.333)

c(.125)

c(.357)

c(.667)

S-Me-A

b

a(.219)

b

b

b

b

b(.950)

b(.750)

b(.850)

b

b(.781)

c(.050)

c(.250)

c(.150)

Pgm-A

a(.025)

b(.516)

a(.031)

b(.375)

b(.250)

a(.053)

b(.450)

b(.313)

a(.025)

a(.150)

b(.500)

d(.484)

b(.375)

d(.625)

d(.750)

b(.342)

d(.550)

d(.687)

b(.950)

b(.775)

d(.475)

d(.594)

d(.605)

d(.025)

d(.075)

Pgdh-A

b

b

b

b

b

b(.972)

a(.025)

b

b(.900)

b

c(.028)

b(.975)

b

c(.100)

S-Sod-A

b

b

b

b

b

b

b

b

b

a(.150)

b(.850)

% loci

20.0

40.0

33.3

33.3

26.7

53.3

46.7

33.3

33.3

26.7

polymor¬

phic

number
of alleles
per locus

1.27

1.60

1.47

1.40

1.27

1.60

1.53

1.33

1.40

1.40

Mean

hetero¬

zygosity

.041

.083

.106

.108

.117

.113

.093

.075

.060

.085

1126

T. Hayashi and M. Matsui

Table 3. Continued

C. pyrrhogaster

Locus

Hiroshima race

11

12

13

14

15

16

17

18

19

20

Acp-A

c

c

c

c

c

c

c

c

c

c

S-Aat-A

b

b(.853)

a(.050)

b(.950)

b

b(.775)

b

a(.025)

b(.944)

b(.950)

c(.147)

b(.950)

e(.050)

a(.225)

b(.975)

d(.056)

d(.050)

Est-1

a(.025)

a(.075)

a(.250)

b

b(.800)

a(.575)

a(.306)

a(.100)

a(.250)

b

b(.975)

b(.925)

b(.750)

c(.200)

b(.425)

b(.694)

b(.900)

b(.750)

Gpi-A

a(.100)

c(.975)

b(.025)

c(.955)

c

c

c(.972)

b(.425)

b(.025)

c(.975)

c(.850)

d(.025)

c(.900)

d(.045)

d(.028)

c(.575)

c(.975)

d(.025)

d(.050)

d(.075)

Iddh-A

b

b

b

b

b

b

b

b

b

b

Ldh-A

b

a(.075)

b(.925)

b

b

b

b

b

b

b

b

Ldh-B

c

c

b(.075)

c(.925)

c

c

c

c

c

c

c

M-Mdh-A

b

b

b

a(.050)

b(.950)

b

b

b

b

b

b

S-Mdh-A

c

c

c

b(.048)

b(.300)

b(.050)

c

b(.025)

b(.026)

c(.974)

c(.857)

d(.095)

c(.700)

c(.950)

c(.975)

c(.974)

d(.026)

M-Me-A

b(.684)

b(.472)

b(.769)

b(.735)

b(.500)

b(.529)

b

b(.971)

b(.393)

b(.154)

c(.316)

c(.528)

c(.231)

c(.265)

c(.500)

c(.471)

c(.029)

c(.607)

c(.846)

S-Me-A

b(.975)

b

b

a(.045)

b

b(.800)

b

b

a(.025)

b

c(.025)

b(.955)

c(.200)

b(.950)

c(.025)

Pgm-A

a(.050)

b

a(.075)

a(.045)

b(.800)

b

b

a(.025)

a(.300)

a(.100)

b(.800)

d(.075)

e(.075)

b(.925)

b(.955)

d(.200)

b(.975)

b(.700)

b(.900)

Pgdh-A

b

a(.025)

a(.075)

b

a(.100)

b

a(.118)

a(.125)

b

b(.950)

b(.950)

c(.025)

b(.925)

b(.900)

b(.882)

b(.875)

c(.050)

S-Sod-A

b(.950)

b(.975)

b(.750)

b

b

b(.725)

b(.750)

b(.950)

b(.900)

b

c(.050)

c(.025)

c(.250)

c(.275)

c(.250)

c(.050)

c(.100)

% loci

40.0

46.7

53.3

46.7

33.3

40.0

26.7

53.3

53.3

40.0

polymor¬

phic

number
of alleles
per locus

1.60

1.53

1.60

1.53

1.33

1.40

1.27

1.53

1.60

1.40

mean

.061

.049

.090

.042

.129

.097

.068

.074

.091

.027

hetero¬

zygosity

Allozymic Variation in Cynops

1127

Table 3. Continued

C.

pyrrhogaster

C. ensicauda

Locus

Hiroshima race

Amami Group

Okinawa Group

21

22

23

24

25

26

27

28

29

Acp-A

c

c

c

a

a

a(.917)

b(.083)

a

a

a

S-Aat-A

b(.722)

b(.357)

a(.056)

a(.058)

a(.050)

b(.944)

a(.025)

a(.100)

a(.125)

d(.278)

d(.643)

b(.500)

b(.942)

b(.900)

d(.056)

b(.975)

b(.900)

b(.687)

d(.444)

d(.050)

d(. 188)

Est-1

b

b

b

b

b

b

b

b

b

Gpi-A

b(.971)

b(.955)

b(.333)

c

c

c

b(.029)

c

b(.125)

c(.029)

c(.045)

c(.667)

c(.971)

c(.875)

Iddh-A

b

b

b

a

a

a

a

a

a

Ldh-A

b

b

b

b

b

b

b

b

b

Ldh-B

c

c

c

b(.096)

b(.275)

c

c

c

c

c(.904)

c(.725)

M-Mdh-A

b

b

b

b

b

b

b

b

b

S-Mdh-A

c

c

b(.028)

d

d(.921)

d

d

c(.800)

c(.438)

c(.972)

e(.079)

d(.200)

d(.562)

M-Me-A

b(.500)

c

b(.429)

a

a

a

a

a

a

c(.500)

c(.571)

S-Me-A

b

b

b

b

b(.875)

b

b

b(.300)

b(.313)

c(.125)

c(.700)

c(.687)

Pgm-A

a(.500)

a(.727)

a(.222)

b(.944)

b

b(.944)

b

b(.625)

b

b(.500)

b(.273)

b(.750)

d(.037)

f(.056)

d(.375)

c(.028)

e(.019)

Pgdh-A

b

b

b

a(.042)

a(.025)

a(.333)

a(.350)

b

b

b(.937)

c(.021)

b(.975)

b(.667)

b(.650)

S-Sod-A

c

c

b(.361)

c(.639)

b

b

b

b

b

b

% loci

26.7

20.0

40.0

26.7

33.3

28.6

20.0

28.6

28.6

polymor¬

phic

number
of alleles
per locus

1.27

1.20

1.53

1.40

1.40

1.29

1.20

1.29

1.36

mean

hetero¬

zygosity

.147

.040

.152

.031

.054

.060

.044

.111

.098

1128

T. Hayashi and M. Matsui

Table 3. Continued

Locus

C.

ensicauda

Okinawa Group

30

31

32

33

34

35

36

37

38

Acp-A

a

a

a

a

a

a

a

a

a

S-Aat-A

a(.100)

b(.900)

b

b(.972)

a(.056)

a(.100)

b(.750)

a(.037)

b(.833)

b(.800)

d(.100)

d(.028)

b(.944)

b(.900)

d(.250)

b(.815)

d(.167)

d(.100)

d(.148)

Est-1

b

b

b

a(.028)

b

b

b

b

b

b(.944)

c(.028)

Gpi-A

c

b(.050)

c

c(.861)

c

c

b(.125)

b(.033)

c

c(.900)

d(.139)

c(.875)

c(.967)

d(.050)

Iddh-A

a

a

a

a

a

a

a

a

a

Ldh-A

b

b

b

b

b

b

b

b

b

Ldh-B

c

c

c

c

c

c

c

c

c

M-Mdh-A

b

b

b

b

b

b

b

b

b

S-Mdh-A

c(.500)

c(.850)

c(.550)

c(.719)

c(.556)

c(.600)

b(.333)

c(.696)

c(.550)

d(.500)

d(. 150)

d(.450)

d(.281)

d(.444)

d(.350)

c(.667)

d(.304)

d(.450)

e(.050)

M-Me-A

a

a

a

a

a

a

a

a

a

S-Me-A

b(.100)

b(.450)

b(.300)

b(.306)

b(.278)

b(.300)

b(.125)

b(. 1 17)

b(.400)

c(.900)

c(.550)

c(.700)

c(.694)

c(.722)

c(.700)

c(.875)

c(.883)

c(.600)

Pgm-A

b

b

b

b

b

b(.950)

b

b

b(.950)

d(.050)

d(.050)

Pgdh-A

b

b

b

b(.972)

b

b

b

b(.981)

b

c(.028)

c(.019)

S-Sod-A

b

b

b

b

b

b

b

b

b

% loci
polymor¬
phic

21.4

26.7

13.3

40.0

28.6

21.4

28.6

35.7

26.7

number
of alleles
per locus

1.29

1.33

1.13

1.47

1.36

1.21

1.29

1.43

1.27

mean

hetero¬

zygosity

.086

.053

.047

.081

.079

.079

.119

.073

.096

Allozymic Variation in Cynops

1129

Table 4. Probability levels of electromorph frequency heterogeneity with a contingency Chi-square test

Locus

Total

C. Pyrrhogaster

C. ensicauda

Total

Amami

Okinawa

Acp-A

.001

_

.05

.05

_

S-Aat-A

.001

.001

NS

NS

.05

Est-1

.001

.001

NS

—

NS

Gpi-A

.001

.001

.001

NS

.01

Iddh-A

.001

—

—

—

—

Ldh-A

.001

.001

—

Ldh-B

.001

.001

.001

.001

—

M-Mdh-A

NS

.05

—

S-Mdh-A

.001

.001

.001

.05

.001

M-Me-A

.001

.001

S-Me-A

.001

.001

.001

.01

NS

Pgm-A

.001

.001

.001

NS

.001

Pgdh-A

.001

.001

.001

.001

NS

S-Sod-A

.001

.001

—

—

—

NS indicates statistically insignificant difference at P<0.05.

C.p.sasayamae C.p.pyrrhogaster C.e.ensicauda C.e.popei

.4 L

2. UPGMA phenogram of populations of Japanese Cynops studied. Scale is in genetic distance units of Nei
(1978).

1130

T. Hayashi and M. Matsui

Table 5. Nei’s genetic similarity (below diagonal) and distance (above diagonal) between populations
of Japanese Cynops studied

C. pyrrhogaster

Sasayama race

Hiroshima

race

1

2

3

4

5

6

7

8

9

10

1

0.007

0.017

0.024

0.023

0.004

0.010

0.026

0.022

0.038

2

0.993

0.006

0.008

0.011

0.012

0.009

0.029

0.017

0.024

3

0.983

0.994

-0.005

-0.005

0.017

0.007

0.034

0.026

0.018

4

0.977

0.992

1.005

-0.004

0.024

0.011

0.040

0.031

0.017

5

0.977

0.989

1.005

1.004

0.015

0.003

0.028

0.041

0.028

6

0.996

0.988

0.983

0.976

0.985

0.006

0.025

0.035

0.049

7

0.990

0.991

0.993

0.989

0.997

0.994

0.010

0.026

0.032

8

0.974

0.971

0.967

0.961

0.973

0.976

0.990

0.057

0.076

9

0.978

0.984

0.974

0.969

0.960

0.966

0.974

0.944

0.014

10

0.962

0.977

0.982

0.983

0.973

0.952

0.969

0.927

0.986

11

0.987

0.988

0.980

0.977

0.969

0.975

0.981

0.951

0.997

0.991

12

0.967

0.976

0.973

0.970

0.961

0.954

0.968

0.927

0.997

0.992

13

0.976

0.973

0.961

0.956

0.946

0.963

0.966

0.933

0.990

0.977

14

0.982

0.983

0.970

0.965

0.958

0.968

0.977

0.952

0.998

0.983

15

0.980

0.987

0.988

0.984

0.982

0.972

0.989

0.957

0.994

0.991

16

0.937

0.946

0.938

0.934

0.925

0.925

0.937

0.903

0.969

0.958

17

0.973

0.960

0.939

0.931

0.924

0.957

0.954

0.929

0.979

0.953

18

0.972

0.963

0.945

0.943

0.925

0.955

0.953

0.928

0.978

0.958

19

0.963

0.974

0.978

0.974

0.969

0.955

0.969

0.929

0.986

0.992

20

0.944

0.964

0.974

0.973

0.966

0.934

0.958

0.911

0.983

0.995

21

0.844

0.854

0.860

0.870

0.839

0.833

0.836

0.802

0.853

0.884

22

0.750

0.778

0.805

0.819

0.788

0.746

0.760

0.715

0.775

0.828

23

0.919

0.932

0.936

0.938

0.919

0.908

0.918

0.879

0.945

0.956

24

0.712

0.714

0.719

0.713

0.718

0.698

0.741

0.748

0.732

0.724

25

0.704

0.707

0.709

0.703

0.709

0.691

0.734

0.740

0.729

0.717

26

0.685

0.685

0.690

0.684

0.689

0.668

0.715

0.721

0.709

0.698

27

0.701

0.702

0.707

0.701

0.705

0.685

0.729

0.735

0.725

0.715

28

0.736

0.747

0.742

0.737

0.740

0.724

0.741

0.741

0.751

0.731

29

0.688

0.702

0.692

0.688

0.681

0.667

0.702

0.711

0.732

0.704

30

0.678

0.693

0.680

0.673

0.671

0.658

0.692

0.705

0.725

0.692

31

0.752

0.762

0.752

0.748

0.741

0.734

0.752

0.737

0.787

0.769

32

0.729

0.739

0.729

0.723

0.723

0.711

0.740

0.744

0.765

0.743

33

0.733

0.745

0.733

0.729

0.723

0.715

0.738

0.733

0.770

0.751

34

0.713

0.724

0.714

0.708

0.706

0.694

0.722

0.723

0.751

0.725

35

0.705

0.717

0.706

0.699

0.698

0.685

0.717

0.721

0.746

0.720

36

0.689

0.704

0.688

0.683

0.675

0.668

0.688

0.684

0.735

0.705

37

0.690

0.704

0.690

0.684

0.679

0.670

0.696

0.698

0.736

0.705

38

0.735

0.745

0.737

0.732

0.731

0.718

0.746

0.748

0.768

0.747

Allozymic Variation in Cynops

1131

Table 5. Continued

C. pyrrhogaster
Hiroshima race

11

12

13

14

15

16

17

18

19

20

1

0.013

0.034

0.024

0.018

0.020

0.065

0.028

0.029

0.037

0.058

2

0.012

0.025

0.027

0.018

0.013

0.056

0.040

0.037

0.026

0.037

3

0.020

0.027

0.040

0.030

0.012

0.064

0.062

0.057

0.023

0.027

4

0.023

0.030

0.045

0.036

0.016

0.069

0.071

0.058

0.026

0.027

5

0.032

0.040

0.055

0.043

0.018

0.078

0.079

0.078

0.032

0.034

6

0.025

0.047

0.037

0.033

0.028

0.078

0.044

0.046

0.046

0.068

7

0.019

0.033

0.035

0.023

0.011

0.065

0.047

0.048

0.032

0.043

8

0.051

0.075

0.069

0.049

0.043

0.102

0.074

0.075

0.073

0.093

9

0.003

0.003

0.010

0.002

0.006

0.031

0.022

0.023

0.014

0.018

10

0.009

0.008

0.023

0.017

0.009

0.043

0.048

0.042

0.008

0.005

11

0.006

0.006

0.002

0.005

0.035

0.018

0.017

0.012

0.021

12

0.994

0.012

0.007

0.004

0.027

0.028

0.031

0.009

0.008

13

0.994

0.988

0.009

0.014

0.017

0.002

0.015

0.014

0.035

14

0.998

0.993

0.991

0.004

0.037

0.017

0.017

0.017

0.024

15

0.995

0.996

0.986

0.996

0.030

0.028

0.031

0.007

0.010

16

0.966

0.974

0.983

0.964

0.970

0.026

0.054

0.019

0.043

17

0.983

0.972

0.998

0.983

0.972

0.974

0.016

0.034

0.064

18

0.983

0.969

0.985

0.983

0.970

0.947

0.984

0.043

0.062

19

0.989

0.991

0.986

0.983

0.993

0.981

0.966

0.958

0.009

20

0.980

0.992

0.965

0.976

0.990

0.958

0.938

0.940

0.991

21

0.874

0.858

0.886

0.854

0.849

0.861

0.861

0.891

0.880

0.856

22

0.791

0.791

0.795

0.768

0.779

0.797

0.753

0.784

0.826

0.815

23

0.949

0.951

0.956

0.941

0.941

0.947

0.933

0.935

0.957

0.947

24

0.730

0.733

0.716

0.747

0.747

0.695

0.707

0.707

0.721

0.734

25

0.724

0.728

0.712

0.742

0.740

0.694

0.701

0.701

0.715

0.729

26

0.705

0.710

0.692

0.724

0.727

0.667

0.686

0.686

0.695

0.711

27

0.721

0.726

0.710

0.739

0.742

0.687

0.704

0.706

0.712

0.727

28

0.740

0.734

0.715

0.738

0.739

0.711

0.704

0.703

0.730

0.734

29

0.713

0.717

0.694

0.725

0.715

0.698

0.684

0.692

0.703

0.715

30

0.703

0.705

0.683

0.714

0.704

0.692

0.675

0.674

0.694

0.704

31

0.776

0.778

0.760

0.779

0.771

0.755

0.750

0.753

0.766

0.775

32

0.752

0.753

0.734

0.760

0.753

0.733

0.726

0.725

0.743

0.751

33

0.758

0.758

0.742

0.763

0.755

0.741

0.733

0.734

0.748

0.756

34

0.734

0.736

0.715

0.742

0.735

0.714

0.706

0.705

0.725

0.734

35

0.729

0.731

0.711

0.739

0.730

0.710

0.702

0.701

0.720

0.730

36

0.715

0.717

0.694

0.721

0.718

0.706

0.685

0.693

0.706

0.714

37

0.715

0.717

0.695

0.722

0.711

0.704

0.687

0.688

0.706

0.715

38

0.755

0.757

0.738

0.764

0.758

0.738

0.729

0.728

0.747

0.756

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

T. Hayashi and M. Matsui

Continued

C. pyrrhogaster

C.

ensicauda

Hiroshima

race

Amami Group

Okinawa Group

21

22

23

24

25

26

27

28

29

30

0.169

0.287

0.084

0.339

0.351

0.378

0.355

0.307

0.374

0.388

0.158

0.251

0.071

0.337

0.347

0.378

0.353

0.291

0.354

0.366

0.151

0.217

0.066

0.330

0.344

0.370

0.347

0.298

0.368

0.386

0.139

0.200

0.064

0.338

0.352

0.380

0.355

0.305

0.374

0.396

0.176

0.238

0.085

0.331

0.344

0.372

0.350

0.302

0.384

0.399

0.183

0.293

0.096

0.359

0.369

0.404

0.379

0.323

0.404

0.418

0.179

0.274

0.085

0.300

0.310

0.336

0.316

0.299

0.353

0.368

0.220

0.336

0.129

0.291

0.301

0.327

0.309

0.300

0.341

0.350

0.159

0.255

0.056

0.312

0.316

0.344

0.321

0.286

0.312

0.322

0.123

0.189

0.045

0.323

0.332

0.359

0.335

0.313

0.351

0.368

0.135

0.235

0.052

0.315

0.323

0.350

0.327

0.301

0.338

0.353

0.153

0.234

0.051

0.310

0.317

0.342

0.320

0.309

0.333

0.349

0.121

0.229

0.045

0.333

0.340

0.369

0.342

0.336

0.365

0.381

0.157

0.264

0.061

0.291

0.298

0.323

0.302

0.303

0.321

0.337

0.163

0.249

0.061

0.292

0.301

0.319

0.299

0.303

0.336

0.352

0.150

0.227

0.054

0.363

0.365

0.405

0.376

0.341

0.359

0.368

0.150

0.284

0.070

0.346

0.355

0.377

0.350

0.351

0.379

0.393

0.115

0.243

0.067

0.346

0.355

0.377

0.348

0.352

0.368

0.394

0.128

0.191

0.044

0.327

0.335

0.364

0.340

0.315

0.352

0.366

0.155

0.205

0.054

0.309

0.317

0.341

0.319

0.309

0.335

0.351

0.009

0.027

0.537

0.551

0.604

0.549

0.544

0.558

0.613

0.991

0.069

0.644

0.658

0.723

0.660

0.650

0.657

0.724

0.974

0.934

0.397

0.405

0.443

0.411

0.399

0.408

0.444

0.585

0.525

0.672

0.003

0.005

0.006

0.093

0.054

0.081

0.576

0.518

0.667

0.997

0.013

0.013

0.085

0.043

0.067

0.547

0.485

0.642

0.995

0.987

-0.002

0.104

0.062

0.090

0.578

0.517

0.663

0.994

0.987

1.002

0.106

0.063

0.091

0.580

0.522

0.671

0.911

0.919

0.901

0.900

0.015

0.011

0.572

0.518

0.665

0.948

0.958

0.940

0.939

0.985

-0.003

0.542

0.485

0.642

0.922

0.936

0.914

0.913

0.989

1.003

0.638

0.578

0.722

0.928

0.936

0.915

0.921

0.995

0.989

0.986

0.600

0.537

0.687

0.945

0.900

0.876

0.876

0.989

0.990

0.995

0.613

0.550

0.696

0.929

0.917

0.896

0.895

0.992

0.995

1.002

0.571

0.507

0.667

0.935

0.960

0.942

0.944

0.992

1.001

0.998

0.565

0.500

0.661

0.939

0.949

0.931

0.931

0.993

1.000

1.003

0.575

0.521

0.666

0.884

0.945

0.926

0.926

0.998

0.999

1.001

0.561

0.503

0.657

0.903

0.954

0.934

0.938

0.992

0.999

1.001

0.609

0.555

0.702

0.952

0.938

0.916

0.922

0.994

0.993

0.995

Allozymic Variation in Cynops

1133

Table 5. Continued

C. ensicauda

Okinawa Group

31

32

33

34

35

36

37

38

1

0.285

0.316

0.310

0.338

0.349

0.373

0.371

0.308

2

0.272

0.302

0.294

0.323

0.333

0.351

0.350

0.294

3

0.285

0.316

0.311

0.337

0.349

0.374

0.372

0.305

4

0.291

0.324

0.316

0.346

0.358

0.381

0.380

0.313

5

0.300

0.325

0.324

0.348

0.360

0.394

0.387

0.314

6

0.309

0.341

0.336

0.366

0.378

0.403

0.401

0.331

7

0.285

0.300

0.304

0.326

0.333

0.373

0.362

0.292

8

0.305

0.296

0.310

0.325

0.328

0.380

0.359

0.291

9

0.239

0.267

0.261

0.286

0.293

0.308

0.307

0.263

10

0.262

0.297

0.287

0.321

0.328

0.349

0.350

0.292

11

0.253

0.286

0.277

0.309

0.315

0.336

0.336

0.281

12

0.251

0.284

0.277

0.307

0.313

0.333

0.333

0.278

13

0.275

0.309

0.299

0.336

0.342

0.365

0.363

0.304

14

0.249

0.274

0.271

0.298

0.302

0.327

0.326

0.269

15

0.260

0.283

0.281

0.308

0.314

0.331

0.341

0.278

16

0.280

0.310

0.300

0.337

0.342

0.348

0.351

0.303

17

0.287

0.320

0.311

0.348

0.353

0.378

0.375

0.316

18

0.284

0.321

0.309

0.350

0.355

0.367

0.374

0.317

19

0.266

0.298

0.290

0.322

0.329

0.349

0.349

0.291

20

0.256

0.286

0.280

0.309

0.315

0.337

0.336

0.280

21

0.450

0.511

0.489

0.561

0.570

0.554

0.579

0.496

22

0.547

0.622

0.597

0.678

0.693

0.652

0.686

0.589

23

0.326

0.375

0.363

0.405

0.414

0.406

0.421

0.354

24

0.075

0.056

0.074

0.067

0.063

0.123

0.102

0.049

25

0.066

0.047

0.064

0.057

0.053

0.105

0.087

0.041

26

0.088

0.068

0.088

0.076

0.071

0.132

0.110

0.060

27

0.083

0.064

0.082

0.077

0.072

0.133

0.111

0.057

28

0.005

0.008

0.006

0.002

0.007

0.011

0.008

0.008

29

0.011

0.001

0.007

0.001

0.000

0.011

0.005

-0.001

30

0.014

-0.001

0.005

-0.001

-0.003

0.005

-0.002

0.002

31

0.005

0.000

0.003

0.006

0.007

0.008

0.004

32

0.995

0.001

-0.003

-0.004

0.011

0.004

0.000

33

1.000

0.999

-0.001

0.000

0.006

0.004

0.003

34

0.997

1.003

1.001

-0.004

0.007

0.002

-0.001

35

0.994

1.004

1.000

1.004

0.009

0.002

-0.001

36

0.993

0.989

0.994

0.994

0.991

0.000

0.012

37

0.992

0.996

0.996

0.998

0.998

1.000

0.006

38

0.996

1.000

0.997

1.001

1.001

0.989

0.994

1134

T. Hayashi and M. Matsui

the mean D value between C. ensicauda and 20
remaining populations of C. pyrrhogaster is 0.330
(range 0.239-0.418). Thus, in spite of their ad¬
jacency to the range of C. ensicauda , southernmost
populations of C. pyrrhogaster reach higher level
of genetic differentiation from C. ensicauda than
do remaining conspecific populations.

DISCUSSION

No overlap of electromorphs was detected at
three loci {Acp-A, Iddh-A and S-Me-A) between
C. ensicauda and C. pyrrhogaster, and no indi¬
vidual showed any intermediate condition. Fur¬
thermore in specimens of C. pyrrhogaster from
northeastern Japan, we have not detected any
individual sharing any electromorph with C. en¬
sicauda at the three loci (Hayashi and Matsui,
unpublished). Therefore, it is clear that C. en¬
sicauda and C. pyrrhogaster are genetically distinct
from each other.

Levels of genetic differentiation have been in¬
vestigated at the interspecific and intersubspecific
ranks in other genera of the family Salamandridae
( Taricha [6] and Triturus [22]). Genetic dif¬
ferentiation estimates derived from these data
might provide an indication of the range in values
one might expect between biological species within
this family. Nei’s D values identified between the
two species of Japanese Cynops have the range of
0.239 to 0.724 (x = 0.356) and are smaller than
interspecific values calculated in Triturus species
(range from 0.702 to 1.321, x = 1.117), but nearly
correspond with the values estimated for three
species of Taricha (range from 0.261 to 0.687, x=
0.466). It is also noteworthy that the greatest
genetic difference was observed between geo¬
graphically most adjacent populations of C. pyr¬
rhogaster from southernmost Kyushu and C. en¬
sicauda. From these genetic view points, two
forms of Japanese Cynops, one from the main
islands {pyrrhogaster ) and another from the
Ryukyu Archipelago ( ensicauda ), are judged to be
well differentiated from each other at the species
level, and the designation of the latter form as a
subspecies of the former is unfounded.

Inger [1] studied morphological variations in
newts from Okinawajima and Amami-Oshima Is¬

lands, and considered the differences between
these two populations sufficient to warrant subspe¬
cific distinction. Thus he described the Okinawa
population as a distinct subspecies, Triturus en-
sicaudus popei. On the contrary, Koba [23] and
Nakamura and Ueno [4], without presenting valid
evidence, opposed such a distinction. Our bio¬
chemical analysis shows that populations from the
Amami Group form a group distinct from those
from Okinawa Group with Nei’s D values between
them ranging from 0.059 to 0.140 (x = 0.084).
Although the amount of genetic differentiation
between these two groups is smaller than values
estimated among intersubspecific populations in
Taricha (range 0.104-0.309) [24], the geographic
pattern of biochemical variation is consistent with
the pattern of morphological variations reported
by Inger [1], and seems to support the subspecific
status of C. ensicauda popei (new comb.).

Sawada [2] divided C. pyrrhogaster into six
“local races” (Hiroshima, Sasayama, Atsumi,
Kanto, Tohoku and Intermediate races) from the
analysis of the geographic variations in the pattern
of ventral markings and body proportions. Later,
Mertens [25] gave a name Triturus ( = Cynops)
pyrrhogaster sasayamae for “Sasayama race” and
the other “races” have remained unnamed and
included in a single subspecies C. p. pyrrhogaster.
Our biochemical study contained specimens of two
“local races” belonging to different subspecies
(“Sasayama rac e” = C. p. sasayama, distributed in
northern Kinki and eastern Chugoku Districts
(populations 1-8), and parts of “Hiroshima race”
= C. p. pyrrhogaster, distributed in Kyushu, Shi¬
koku and western Chugoku Districts (populations
9-23)). In the present study, C. pyrrhogaster
showed significant heterogeneity in electromorph
frequencies at all polymorphic loci (Table 4) and it
is clear that local populations are genetically well
isolated and diverged. Especially, southernmost
Kyushu populations (populations 21-23) showed a
high level of genetic differentiation from others,
and consequently populations of Hiroshima race
(i.e. C. p. pyrrhogaster ) did not form a single
group (Fig. 2). This unexpected result indicates
that the widely ranging “Hiroshima race” contains
at least two distinct groups. By contrast, the
“Sasayama race” was found to form a single group.

Allozymic Variation in Cynops

1135

Our data offer an interesting perspective not only
into taxonomic problems but also into genetic
divergence in C. pyrrhogaster. We are currently
conducting an extensive sampling throughout the
range of this species and additional electrophoretic
studies are in progress.

The molecular clock hypothesis has been ap¬
plied to date divergence events of taxa [26, 27].
Since molecular clocks need calibrations before
applied, some calibrations were estimated for elec¬
trophoretic clock, which is a kind of molecular
clocks, from different sources. Nei and
Roychoudhury [11] originally suggested a calibra¬
tion of ID = 5 MY irrespective of animal groups.
Later, Maxson and Maxson [10] calibrated the
electrophoretic clock at ID = 14 MY in plethodon-
tid salamanders based on the correlation with the
albumin clock, and the value has generally been
used [28, 29]. Calibration could be obtained if
geological information of the time of isolation
between two populations is available. Because
newts cannot cross over the sea, a strait must be a
sufficient barrier for gene flow between popula¬
tions. The formation of the straits between the
Japanese main islands and the Ryukyu Archipela¬
go assuredly marked the cessation of gene ex¬
change between populations separated by the sea.
The formation of the strait between the Amami
and Okinawa Groups also must have prevented
newts from gene exchange between isolated
populations.

Kizaki and Oshiro [30, 31] estimated that the
strait between the mainlands and the Ryukyu
Archipelago south of Amami Group was formed
about 8 MYBP during late Miocene times and that
the strait between the Amami and Okinawa
Groups was formed between 1 and 1.5 MYBP
during middle Pleistocene times. If the mean D
value between C. ensicauda and C. pyrrhogaster
(0.356) corresponds to 8 MY, a calibration of 1D =
22 MY is obtained. On the other hand, when the
mean D value between populations of the Amami
and Okinawa Groups (0.078) is compared to the
duration between 1 and 1.5 MY, the electrophore¬
tic clock is calibrated at ID = 13-19 MY. Since
both genetic and geological estimates themselves
may contain a considerable amount of errors, the
difference in these two estimates may not be so

great. At least, calibrations of electrophoretic
clock in Japanese Cynops are regarded as far
greater than Nei’s original estimate (ID = 5 MY)
and are more similar to the calibration used in
plethodontid salamanders (ID = 14 MY). This
result seems to suggest the presence of limited
range of calibration which is specific to the order
Caudata.

ACKNOWLEDGMENTS

Thanks are due to Dr. M. Tasumi, supervisor of the
senior author, for his continued support and encourage¬
ment. Dr. D. M. Green and Mr. T. Hikida reviewed an
early draft of this manuscript. We thank the following
persons who provided specimens and aided in field work:
Messrs. T. Hikida, M. Hinoue, N. Honda, T. Izumi, A.
Mori, H. Okawa, I. Okouchi, H. Ota, S. Tanabe, S.
Tanaka, Y. Utsunomiya, and S. Watanabe, Ms. T. Utsu-
nomiya and Professor T. Seto. For computer assistance,
we thank Messrs. T. Hikida and R. Kitamura. Dr. R.
Murphy provided many helpful comments.

REFERENCES

1 Inger, R. F. (1947) Preliminary survey of the
amphibians of the Riukiu Islands. Fieldiana : Zool.,
32: 295-352.

2 Sawada, S. (1963) Studies on the local races of the
Japanese newt, Triturus pyrrhogaster Boie. I. Mor¬
phological characters. J. Sci. Hiroshima Univ., Ser.
B, 1, 21: 135-165.

3 Kawamura, T. (1950) Studies on hybridization in
amphibians. III. Reciprocal hybrids between Tritu¬
rus pyrrhogaster (Boie) and Triturus ensicauda (Hal¬
lowed). J. Sci. Hiroshima Univ., Ser. B, 1, 11: 71-
79.

4 Nakamura, K. and Ueno, S.-I. (1963) Japanese
Reptiles and Amphibians in Colour, Hoikusha, Osa¬
ka. (In Japanese)

5 Oyama, J. and Nakamura, D. (1939) Uber Artbas-
tarde Triturus pyrrhogaster (Boie) g- und T. en¬
sicauda (Hallowed) £ . Jpn. J. Genet., 15: 78-79.
(In Japanese)

6 Hedgecock, D. and Ayala, F. J. (1974) Evolution¬
ary divergence in the genus Taricha (Salamandri-
dae). Copeia, 1974: 738-747.

7 Kalezic, M. L. (1984) Evolutionary divergences in
the smooth newt, Triturus vulgaris (Urodela, Sala-
mandridae) : electrophoretic evidence. Amphibia-
Reptilia, 5: 221-230.

8 Ragghianti, M. and Wake, D. B. (1986) Genic
variation and its evolutionary implications in the
Italian newt, Triturus italicus. Herpetologica, 42:

1136

T. Hayashi and M. Matsui

206-214.

9 Tabachnich, W. J. (1977) Geographic variation of
five biochemical polymorphisms in Notophthalmus
viridescens. J. Hered., 68: 117-122.

10 Maxson, L. R. and Maxson, R. D. (1979) Compar¬
ative albumin and biochemical evolution in pletho-
dontid salamanders. Evolution, 33: 1057-1062.

11 Nei, M. and Roychoudhury, A. K. (1974) Genic
variation within and between the three major races
of man, caucasoids, negroids, and mongoloids. Am.
J. Hum. Genet., 26: 421-443.

12 Ayala, F. J., Powell, J. R., Tracey, M. L., Mourao,
C. A., and Perez-Salas, S. (1972) Enzyme variabil¬
ity in the Drosophila willistoni group. IV. Genic
variation in natural populations of Drosophila willis¬
toni. Genetics, 70: 113-139.

13 Delorenzo, R. L. and Ruddle, F. H. (1970) Gluta¬
mate oxaloacetate transaminase (GOT) genetics in
Mus musculus ; linkage, polymorphism, and phe¬
notypes of the GOT-2 and GOT-1 loci. Biochem.
Genet., 4: 259-274.

14 Sato, C. (1982) The variation of proteins and en¬
zymes in blood. In “Methodology of Human Ge¬
netics”. Ed. by E. Matsunaga, Kyoritsu-Shuppan,
Tokyo, pp. 114-155. (In Japanese)

15 Shaw, C. R. and Prasad, R. (1970) Starch gel elec¬
trophoresis of enzymes — a compilation of recipes.
Biochem. Genet., 4: 297-320.

16 Selander, R. K., Smith, M. H., Yang, S. Y., John¬
son, W. E. and Gentry, J. B. (1971) Biochemical
polymorphism and systematics in the genus Pero-
myscus. I. Variation in the old-field mouse ( Pero -
myscus polionotus). Stud. Genet. 6. Univ. Texas
Publ., 7103: 49-90.

17 Nomenclature Committee of the International Un¬
ion of Biochemistry. (1984) Enzyme Nomenclature
1984. Academic Press, NY.

18 Murphy, R. W. and Crabtree, C. B. (1985) Evolu¬
tionary aspects of isozyme patterns, number of loci,
and tissue-specific gene expression in the prairie
rattlesnake, Crotalus viridis viridis. Herpetologica,
41: 451-470

19 Nei, M. (1978) Estimation of average heterozygos¬
ity and genetic distance from a small number of
individuals. Genetics, 89: 583-590.

Sneath, P. H. A. and Sokal, R. R. (1973) Numeri¬
cal Taxonomy. W. H. Freeman Co., San Francisco.
Workman, P. L. and Niswander, J. D. (1970)
Population studies on southwestern Indian tribes. II.
Local genetic differentiation in the Papago. Am. J.
Hum. Genet., 22: 24-29.

Kalezic, M. L. and Hedgecock, D. (1980) Genetic
variation and differentiation of three common Euro¬
pean newts ( Triturus ) in Yugoslavia. Br. J. Herpe-
tol. , 6: 49-57.

Koba, K. (1962) Studies on the Snakes of the Genus
Trimeresurus of the Amami and Tokara Islands.
Japan, Japan Society for the Promotion of Science,
Tokyo. (In Japanese)

Hedgecock, D. (1976) Genetic variation in two
widespread species of salamanders, Taricha granulo¬
sa and Taricha torosa. Biochem. Genet., 14: 561—
576.

Mertens, V. R. (1969) Uber die Rassen des Feuer-
bauchmolches ( Triturus pyrrhogaster) und ihre wis-
senschaftlichen Namen. Aquar.-Terrar.-Z., 22:
114-117.

Sarich, V. M. and Wilson, A. C. (1967) Immunolog¬
ical time scale for hominid evolution. Science, 158:
1200-1203.

Thorpe, J. P. (1982) The molecular clock hypoth¬
esis : biochemical evolution, genetic differentiation
and systematics. Ann. Rev. Ecol. Syst., 13: 139-
168.

Larson, A. (1983) A molecular phylogenetic per¬
spective on the origins of a lowland tropical sala¬
mander fauna. I. Phylogenetic inferences from pro¬
tein comparisons. Herpetologica, 39: 85-99.
Larson, A., Wake, D. B., Maxson, L. R. and High-
ton, R. (1981) A molecular phylogenetic perspec¬
tive on the origins of morphological novelties in the
salamanders of the tribe Plethodontini (Amphibia,
Plethodontidae). Evolution, 35: 405-422.

Kizaki, K. and Oshiro, I. (1977) Paleogeography of
the Ryukyu Islands. Marine Sciences Monthly, 9:
542-549. (In Japanese with English summary)
Kizaki, K. and Oshiro, I. (1980) The Origin of the
Ryukyu Islands. In “Natural History of Ryukyu”.
Ed. by K. Kizaki, Tsukiji-Shokan, Tokyo, pp. 8-37.
(In Japanese)

20

21

22

23

24

25

26

27

28

29

30

31

ZOOLOGICAL SCIENCE 5: 1137-1139 (1988)

© 1988 Zoological Society of Japan

[COMMUNICATION]

Effects of Adult Male Cohabitation on Precocious
Puberty in Early Weaning Female Guinea Pigs

Tadashi Kosaka, Masaki Obata, Toru R. Saito
and Kazuaki W. Takahashi1

Toxicology Division, Institute of Environmental Toxicology, Suzuki-cho 2-772,
Kodaira-shi, Tokyo 187, and 1 Department of Laboratory Animal Science,
Nippon Veterinary & Zootechnical College, Kyonan-cho 1-7-1,
Musashino-shi, Tokyo 180, Japan

ABSTRACT — The effects of adult male cohabitation
with infant female guinea pigs weaned at 2 days or 4 days
of age on the onset of puberty were investigated. The
females weaned at 4 days had no growth abnormalities,
while 2 of 20 females weaned at 2 days died due to
developmental abnormalities. Cohabitation with an
adult male significantly advanced the day of vaginal
opening in the females weaned at 4 days. During the first
vaginal opening, 4 of 9 and 7 of 10 females with male
cohabitants in the groups weaned at 2 and 4 days old,
respectively, copulated, became pregnant and delivered
normally. These results indicated that advancement of
female puberty was produced by continuous exposure to
an adult male even in infant female guinea pigs weaned
at 4 days.

INTRODUCTION

Social stimuli that operate primarily through
olfactory mechanisms, including presence of
males, influence reproductive biology in mice [1-
4]. Similar sensory mechanisms are undoubtedly
involved in modifying animal reproduction. Pre¬
vious reports have shown that the presence of an
adult male accelerates the sexual maturation of
female mice [1, 2, 5-9], rats [3], pigs [10] and
human beings [11]. In guinea pigs, our previous
work [12] indicated that a similar advancement in
puberty was caused by adult male cohabitation
from the day of normal weaning (1 or 2 weeks of

Accepted February 15, 1988
Received October 28, 1987

age). It is known that newborn babies of guinea
pigs are well-developed compared with other
rodents, i.e., the babies have a fully developed
coat, walk immediately and eat food; their teeth
are useful and their eyes are open. Thus, it seems
possible that social stimuli can play a useful role in
infant guinea pigs weaned earlier than normal.

The present study, therefore, was designed to
investigate the influence of continuous exposure to
an adult male on the onset of female sexual
maturation in infant female guinea pigs weaned
early.

MATERIALS AND METHODS

Albino guinea pigs of the Hartley strain were
used. They were kept in a controlled semi-barrier
system room with a temperature of 22±2°C, a
photoperiod of 14 hr light/10 hr dark (lights on
from 05:00 to 19: 00 hr) and ventilation 12 times
an hour. The females were housed individually,
either alone or with an adult male cohabitant in an
aluminum cage (260W X 400D X 200H mm), and
given commercial pellets (GB-1, Funabashi Farm
Inc.) and tap water ad libitum.

Infant female guinea pigs were weaned at the
age of 2 days (group A) and 4 days (group B).
Each group was then divided into two subgroups.
One subgroup of females was continuously ex¬
posed to an adult male (5 to 12 months old) with
proven fertility from the time of weaning until the

1138

T. Kosaka, M. Obata et al.

end of the first vaginal opening by placing the male
directly in the cage containing the female; the
other subgroup of females was housed alone.
Vaginal closure membranes were examined once a
day until the end of the second vaginal opening.
Vaginal opening was determined to be positive
when the vaginal membrane was ruptured. Each
animal was weighed at the first vaginal opening. In
the females mating with males during the first
vaginal opening, the gestation period and litter size
were recorded. The day when spermatozoa were
found in vaginal smears was designated as day 0 of
pregnancy. Student’s t test was used to determine
the level of significance.

RESULTS AND DISCUSSION

The results are summarized in Table 1. With
regard to growth after weaning, 2 of 20 females
weaned at 2 days of age (group A) died 2 days after
weaning due to developmental abnormalities. In
the group weaned at 4 days (group B), the mean
age at the first vaginal opening in the females
housed with a male was significantly younger
compared with females housed alone. In females
weaned at 2 days (group A), however, the females
with a male had a slight but not significant
advancement of the first vaginal opening over
females housed alone. The mean body weight at
the first vaginal opening in females with a male was
significantly less than that of females without a

male in both groups A and B (group A, P<0.05;
group B, P<0.01). Mean duration of the first
vaginal opening did not differ between the sub¬
groups of group A, but it was significantly reduced
in females together with a male in group B (P<
0.05) when compared with females housed alone.
During the first vaginal opening, 11 out of 19
females housed with a male (4 of 9 females in
group A and 7 of 10 in group B) copulated and
became pregnant. The mean gestation period in
group A, 72.8 + 1.8 days, was longer than that in
group B, 69.3 + 1.8 days (P<0.05). There was no
significant difference between the mean litter size
in the groups weaned at 2 and 4 days, 2.3 ± 1 .0 and
2.0 ±0.8, respectively.

Concerning the relationship between the time of
vaginal opening and body growth, it is unclear
whether or not precocious puberty caused by the
presence of males is controlled by body growth. In
this study, acceleration of the first vaginal opening
due to male cohabitation was induced without the
female reaching a body weight, as high as that of
females housed alone at the time of first vaginal
opening. Damon et al. [13] indicated that physical
growth and sexual maturation might be a separate
process. The presence of the male, therefore,
which is an olfactory stimuli, and not physical
growth, may lead to the hastening of puberty.

Our previous work [12] reported that female
guinea pigs with an adult male cohabitant from the
time of normal weaning (1 or 2 weeks old)

Table 1. Effects of an adult male cohabitant on the age at vaginal opening, the opening duration and body
weight at the first vaginal opening in female guinea pigs

Age at weaning
(Group)

Cohabitant

No. of
animals
examined

Age at first
vaginal opening
(days)

Duration of first
vaginal opening
(days)

Body weight at first
vaginal opening
(g)

2 days
(group A)

None

9

36.9 + 3.6

5. 1 + 0. 6

312.3 + 11.4

Male

9

33.2 + 5.8

6. 6+1.0

269.4+14.1*

4 days
(group B)

None

10

37.3 + 1.8

7.4 + 0. 7

306.7+ 6.2

Male

10

28.1 + 2.5**

4.8 + 0. 7*

259.0 + 12.0**

All data represent mean + S.E.

*, **: Significantly different from the non-cohabitant subgroup of each group at the 5 and 1% level of probability,
respectively.

Precocious Puberty in Female Guinea Pigs

1139

accelerated puberty. In this paper, the presence of
males resulted in precocious puberty characterized
by advancement of vaginal opening in infant
female guinea pigs weaned earlier (group B) than
normal. In group A, the time of first vaginal
opening in females with and without males ranged
from 16 to 66 and from 29 to 64 days of age,
respectively. In the females housed with a male,
the age at the first vaginal opening was slightly but
not significantly accelerated, that is, the ages at the
first vaginal opening of 5 females (16, 19, 22, 25, 27
days of age) out of 9 were hastened, when
compared with the age at the first vaginal opening
(29 days) in the females housed alone. Precocious
puberty caused by the presence of males might be
not related to the exposure time. Takahashi et al.
[14] reported that the shortening of the rat estrous
cycle caused by the presence of males was
observed in higher incidences when different adult
male rats were frequently exposed to females. In
this study, the exposure of a fresh male to the
females of groups A and B was started at 2 and 4
days of age, respectively. If the critical age of
sensitivity to the fresh stimuli of a male cohabitant
existed around 2 days of age in female guinea pigs,
the above-mentioned slight acceleration of puberty
should be induced. However, during the first
vaginal opening, 7 out of 10 females in the group
weaned at 4-days-old copulated, became pregnant
and delivered normally. These results, therefore,
indicate that weaning at 4 days was the earliest
time at which male cohabitation had an effect on
the precocious sexual maturity in the females.

ACKNOWLEDGMENT

The authors are grateful to Dr. Yasuhiko Shirasu,
Toxicology Division, Institute of Environmental Toxicol¬
ogy, Tokyo, for his valuable comments during this study.

REFERENCES

1 Vandenbergh, J. G. (1967) Endocrinology, 81: 345-
349.

2 Vandenbergh, J. G. (1969) Endocrinology, 84: 658-
660.

3 Vandenbergh, J. G. (1976) J. Reprod. Fertil., 46:
451-453.

4 Parkes, A. S. and Bruce, H. M. (1961) Science,
134: 1049.

5 Eisen, E. J. (1975) J. Anim. Sci., 40: 816-825.

6 Kennedy, J. M. and Brown, K. (1970) Dev. Psycho-
biol., 3: 179-189.

7 Marsden, H. M. and Bronson, F. H. (1965) J. En¬
docrinol., 32: 313-319.

8 Nobunaga,T., Nakamura, K. and Yoshida, Y.
(1965) Japan. J. Anim. Reprod., 11: 7-15.

9 Takahashi, K. W. (1971) Japan. J. Anim. Reprod.,
17: 47-54.

10 Brooks, P. H. and Cole, D. J. A. (1970) J. Reprod.
Fertil., 23: 435-440.

11 McClintock, M. K. (1971) Nature, 229: 244-245.

12 Saito, T. R., Obata, M. and Takahashi, K. W.
(1982) Japan. J. Anim. Reprod., 28: 141-144.

13 Damon, A., Damon, S. T., Reid, R. B., and Vala-
dian, I. (1969) Hum. Biol., 41: 161.

14 Takahashi, K. W. , Saito, T. R. and Amao H. (1986)
Bull. Nippon Vet. Zootec. Coll., 35: 124.

ZOOLOGICAL SCIENCE 5: 1141-1143 (1988)

© 1988 Zoological Society of Japan

[COMMUNICATION]

Maternal Behavior in Virgin Female Rats Following
Removal of the Vomeronasal Organ

Toru R. Saito, Kunie Kamata, Mikio Nakamura
and Minoru Inaba

Department of Pharmacology, Kyorin University School of
Medicine, 6-20-2 Shinkawa, Mitaka, Tokyo 181, Japan

ABSTRACT — Induction of the maternal behavior in
virgin female rats exposed continuously to young foster
pups was studied. The vomeronasal organ (VNO)
ablation accelerated the onset of maternal behavior of
virgin females following continuous exposure to foster
pups. It may be possible that the neural substrates of the
VNO play an inhibitory role in expressing the maternal
behavior in virgin female rats.

INTRODUCTION

The vomeronasal organ (VNO) in most
mammalian species consists of a bilaterally paired
tubular structure lying on either side at the base of
the nasal septum. Peripheral receptor neurons
within the VNO innervate the accessory olfactory
bulbs (AOB) and are anatomically independent of
the olfactory receptor neurons within the olfactory
epithelium, which project to the main olfactory
bulbs (MOB). In turn, the AOB and MOB give
rise to separate afferent pathways which terminate
in different areas of the pyriform lobe and amygda¬
la (AMYG). From the AMYG, vomeronasal
pathways project to the medial preoptic area
(MPOA) and the ventromedial nucleus of the
hypothalamus (VMH) [1].

Functionally, the VNO conveys chemosensory
informations to its parts of the hypothalamus
thought to be important in the regulation of
reproductive physiology and behavioral functions.
Saito and Moltz [2, 3], for example, showed that

Accepted January 21, 1988
Received December 22, 1987

ablation of the VNO reduced the incidence of
sexual behavior in rats and Saito [4] suggested that
the vomeronasal inputs from pups facilitated the
maternal behavior of lactating rats.

It is well known that maternal behavior in the
lactating rat is observed within a few hours of
parturition. The virgin rat, however, does not
show immediate maternal responsiveness toward
foster pups. If virgin rats are exposed continuously
to young foster pups, they begin to show maternal
behavior within 10 days [5], The purpose of the
present experiment was to determine the role of
the vomeronasal system in maternal behavior of
virgin female rats.

MATERIALS AND METHODS

Subject : Eighteen virgin female Wistar rats
(250-300 g) were used in this experiment. The
animals were kept in the room of a temperature of
23-25°C and exposed to a light-schedule of 12 hr
light and 12 hr darkness (light on at 0600). They
received standard laboratory diet and tap water ad
libitum.

Upon reaching 100-150 days of age, these
females underwent either the removal of the VNO
(VOX, n = 9) or a sham surgical procedure
(SHAM, n = 9), as detailed by Saito and Moltz [2,
3] and Saito and Mennella [6]. After the opera¬
tion, each female was housed individually in a wire
mesh cage (50x50x40 cm) faced with Plexiglas.
Shredded paper was supplied as nest material.

Procedure : Two weeks after the operation, 6

1142

T. R. Saito, K. Kamata et al.

foster pups of the same strain, 1 to 2 days old, were
placed in the quadrant of the cage diagonally
opposite to the nest area. Each day, pups were left
with the female for a 24-hr period, after which they
were removed and then replaced by different fresh
pups. After introducing pups, the institution of
maternal behavior of the female was observed for
1 hr every day. To be scored as maternal, a female
was required not only to retrieve all the 6 pups, but
to build a nest, assume a nursing posture, lick the
young, and keep them warm. Observations were
conducted for 8 days.

Histology. At the end of the examination, each
female was sacrificed for histological observations.
The animal was perfused with 10% formalin and
the head was decalcified in a formic acid solution,
embedded in paraffin, sectioned at 40 /an and
stained with hematoxylin and eosin.

Statistical analysis : Data were analyzed with
Fisher’s exact probability test and Mann-Whitney
U test for independent samples.

RESULTS AND DISCUSSION

Figure 1 shows that effects of the vomeronasal
organ ablation on the induction of maternal
behavior. Of the 9 VOX females, all showed the
maternal behavior, while only 2 of 9 females in
SHAM group did within 5 days after exposure of
pups (P<0.001). The remaining 7 females in
SHAM group did not show the maternal behavior
during an 8-day observation period. The average
latency for the onset of maternal behavior in VOX
group was 2.6 days (range: 1-5 days). There was a
significant difference in the latency between VOX
and SHAM groups (P<0.001).

Histological examinations confirmed that the
removal of the VNO was complete in all of VOX
females whereas the VNO was intact in all SHAM
females.

These results are somewhat contrasted to that in
the lactating rats [4]. In lactating females removal
of the VNO severely depressed retrieval, one of
maternal behaviors. Numan [7] showed that the
MPOA was essential for all components of mater¬
nal behavior in the rat. Since the VNO is located
in the nasal septum and has fiber connections with
the AOB which gives projections to the AMYG

Days

I _ I - 1 _ _l _ J _ L _ I _ I _ I

01 2345678

Days

Fig. 1. Latency in time of onset for the display of
maternal behavior in the vomeronasal organ-ablated
(VOX) and sham-operated (SHAM) virgin female
rats. Arrow indicates that the animal failed to act
maternally at the conclusion of the observation
period.

and the MPOA [1], the vomeronasal cues emanat¬
ing from pups may reach the MPOA and depress
the outputs of the MPOA, in this way inhibiting
maternal behavior. In case of the lactating rats,
the hormonal changes which occur around the time
of parturition on the MPOA may act on the VNO
to reduce this inhibitory influence and/or to play a
facilitatory role in displaying maternal behavior.
Further study is needed to clarify these points.

ACKNOWLEDGMENTS

The authors are grateful to Professor H. Moltz, De¬
partment of Behavioral Sciences, The University of
Chicago, Chicago, IL 60637, U.S.A. for his valuable
advice and suggestion on this study. We also thank
Former President Dr. T. Imamichi, Nippon Veterinary
and Zootechnical College, Musashino, Tokyo 180,
Japan, for his helpful discussion.

Vomeronasal Organ and Maternal Behavior

1143

REFERENCES

1 Scalia, F. and Winans, S. S. (1975) J. Comp.
Neurol., 161: 31-55.

2 Saito,T. R. and Moltz, H. (1986) Physiol. Behav.,
37: 507-510.

3 Saito, T. R. and Moltz, H. (1986) Physiol. Behav.,

38: 81-87.

4 Saito, T. R. (1986) Zool. Sci., 3: 919-920.

5 Rosenblatt, J. S. (1967) Science, 156: 1512-1514.

6 Saito, T. R. and Mennella, J. A. (1986) Exp. Anim.
35: 527-529.

7 Numan, M. (1974) J. Comp. Physiol. Psychol., 87
746-759.

ZOOLOGICAL SCIENCE 5: 1145-1148 (1988)

© 1988 Zoological Society of Japai

[COMMUNICATION]

Colony Composition of the Wood-feeding Cockroach,
Panesthia australis Brunner (Blattaria, Blaberidae,
Panesthiinae) in Australia

Tadao Matsumoto

Department of Biology, University of Tokyo,

Komaba, Meguro-ku, Tokyo 153, Japan

ABSTRACT — The wood-feeding and subsocial cock¬
roach, Panesthia australis was investigated with special
reference to colony composition in the eucalypt forests in
New South Wales, Australia in October 1986. Twenty
seven adults and 185 nymphs were collected in the study
areas. The field evidence presented here shows that this
cockroach lives in the family groups and the intimate
adult-offspring relationships may continue for a long
time.

INTRODUCTION

The order Blattaria is of special interest in
discussion of the evolution of eusociality in Iso-
ptera, it is regarded as a modified side branch of
the latter. Of the 4000 species of cockroaches in
the world, a wood-roach, Cryptocercus punctulatus
(Cryptocercidae) is believed to be in the position
nearest to the lower termites in phylogenetic
relationship [1, 2]. There are some morphological
similarities between them, and the wood-roach’s
hindgut fauna of symbiotic cellulose digesting
Protozoa closely resembles those of lower ter¬
mites. The wood roach lives in monogamous pairs
or in intimate family groups. It dwells in the
rotting wood of fallen logs and is frequently
mentioned in discussions of the origins of termite
sociality [3-5], Besides Cryptocercus , some other
blaberid genera (e.g. Panesthia and Salganea ) of
the Panesthiinae are also known as wood-dwelling
and wood-feeding roaches that live in family

Accepted January 18, 1988
Received October 2, 1987

groups [6, 7]. The life history characteristics of
only a few Australian species of the genus Panes¬
thia have so far been reported and available
information are still fragmentary. According to
Shaw [8] both Panesthia australis and P. laevicollis
appear to live in families, and that one usually
finds a pair of adults often associated with from 12
to 20 nymphs in different stages of development
(see also Tillyard [9]). The Panesthiinae occurs
principally in the Indo-Malayan and Australian
regions with a few extending into the Palaearctic
region having ten genera [10]. In Australia the
subfamily is represented by seven genera and
largely restricted to the eastern part of the conti¬
nent.

In 1986, I had an opportunity to study some
wood roaches in New South Wales Queensland.
This paper reports the colony composition and
some life history characteristics of one species,
Panesthia australis Brunner.

STUDY SITE AND METHOD

Social groups of P. australis were collected from
rotten logs in eucalypt forests along road from
Braiwood to Batemans Bay in New South Wales.
Logs were inspected for evidence of roach activity,
and if entry holes were present, the course of the
irregularly-shaped gallery were traced with a chisel
hammer until the end points of the gallery system
were reached. All roaches in one gallery system
were considered to be members of a social unit.
Logs larger than 30 cm in diameter were not

1146

T. Matsumoto

inspected because preliminary check showed that
numerous and complex gallery systems made it
difficult to discern the limits of individual social
units. The gallery systems of families with small

nymphs were shorter and simpler than those of
families with larger nymphs. The wall of galleries
are built out of frass, and many frasses also were
deposited in or near entrances of a gallery. The

Table 1. Group composition and size distribution of Panesthia australis in October 1986. (Figures
indicate number of individuals)

Presence
of adults

Number of nymphs by size class
(body length, mm)

Total Type of
number group

* $

34 32 30 28 26 24 22 20 18 16 14 12 10 8

6

+ +

+ +

2 1 14

1 3 1

2 1 1

16 A
6 A

+

+

+

+

+

+

+

+

+

+

+

+

+

+

1

1 1 1
1

1

2 4
1 1
1 3 1

12 12 111

1 1 3

1 1

2 2

1 1
1 1

2

4 5

1 1

7

9 1 7

2 4

24

13

11

11

10

9

9

6

6

6

5

4

4

3

B

B

B

B

B

B

B

B

B

B

B

B

B

B

+ 1 1
+

3 C

2 C

1

1

- 1 1
2 1

1 1

1 1 2

1 1 3

1 1 1

1

2

2 2

2 1 1 1

1

2

1

2

9 D
6 D
6 D
5 D
4 D
3 D
3 D
3 D
2 D

+ +

+ +

2 E

2 E

A: families with an adult pair a»d nymphs

B: families consisting of a group of nymphs together with an adult female
C: families consisting of a group of nymphs together with an adult male
D: groups of nymphs
E: adult pairs

Colony Composition of Subsocial Cockroach

1147

numbers of adults and nymphs were counted and
their sizes were measured by using calipers. P.
australis adults are about 34 mm long and black in
color. The tegmina and wings are fully developed
but they are dropped after the pair formation. As
the sexual dimorphism is negligible in size and
shape, sex of adults were determined by inspection
of external genitalia. As P. australis is ovovivipor-
ous, the number of nymphs (cluch size) in uterus
could be determined by dissection of adult
females. The number of nymphal instars, and time
required for nymphal development, are unknown.

RESULTS AND DISCUSSION

Forty eight social units (27 individuals of adults
and 185 individuals of nymphs) were collected in
the study areas. Table 1 summarises the composi¬
tion of P. australis social units found in October
1986. Besides 3 solitary females I could classify 29
groups into 5 types (A-E). This means most
individuals belonged to ‘family’ in a wide sense.
Families consisting of a group of nymphs together
with an adult female (type B) comprised 29.1 % of
the total social units (51.9 % of total number of
adults, i.e. 14/27). Families with a male adult and
nymphs (type C) or families with an adult pair and
nymphs (type A) were fewer than type B families.
Table 1 also shows the size distribution of nymphs
in ‘family’ groups, i.e. family groups (type A-C)
and groups of nymphs (type D). Overall mean
brood sizes (number of nymphs) of family groups
were 7.1 (n = 18, SD = 5.3). Total number of
nymphs in ‘family’ type (A-D) was 169 and mean
body length was 16.7 mm (SD = 7.8 mm, range =
7-35 mm).

Besides family groups, I collected 16 individuals
of solitary nymphs and their mean body length was
23.4 mm (SD = 6.6, range = 15-35 mm). This sug¬
gested that small nymphs less than 15 mm body
length do not disperse from groups or they could
not live alone. The nymphs within some families
were not similar in size and color. As the growth
rate of cockroaches is known to be quite slow,
these families possibly contained two or three
broods. For example, one of type B groups had
older nymphs (ca. 26 mm) and younger nymphs
(ca. 11 mm). This also means that adult-offspring

relationship might continue for long time. Thus, it
can be said that P. australis is a subsocial wood
roach in a wide sense. Mean clutch size of
dissected female was 15.7 (n=4, SD = 1.5, range =
14-18). The clutch size of P. australis was lower
than that of Cryptocercus punctulatus (Nalepa
1984; n=33, mean size = 22.7, SD = 10.8, range =
4-42) [5],

The genus Cryptocercus was placed near or in
the Panesthiinae (or Panesthiidae) by many earlier
taxonomists. However, McKittrick [1] revealed
that the resemblance of some panesthiines to
Cryptocercus is due to convergence, and they are
belonged to different superfamilies (Blaberoidea
and Blattoidea). She concluded that there are two
phyletic lineages, one of which evolved ovo vivi¬
parity (Blaberoidea) and the other remained ovi¬
parous (Blattoidea). The Panesthiinae are all
ovoviviparous and are placed in Blaberoidea. The
ovoviviparous cockroaches (Blaberidae) are fre¬
quently subsocial and associations of adults and
young nymphs in 11 genera have been observed in
the field or in laboratory colonies [9]. But in most
cases, the associations are those between mother
and her offspring, not between an adult pair and
their offspring, and the intimate mother-offspring
relationship continues usually for a short time as
aggregation of newly hatched offspring around
their mother [10]. The family groups which
continue for a long time as in Panesthia ([7] and
the present paper) and C. punctulatus [4, 5] are so
far not well-known. The adaptive value, especially
from a sociobiological perspective, of a lasting-
family formation in wood-feeding cockroaches has
still to be considered.

ACKNOWLEDGMENTS

I thank Prof. Y. Ito, Laboratory of Applied Entomol¬
ogy and Nematology, Nagoya University for advice and
encouragement, and Dr. R. W. Taylor, Division of
Entomology, CSIRO for providing accomodation and
transportation. This work was supported by Grants-in-
Aid for the Overseas Scientific Survey on Ecology and
Taxonomy of Australian Social Insects (No. 61042002)
and for Scientific Research (No. 61540471) from the
Japan Ministry of Science, Education and Culture.

1148

T. Matsumoto

REFERENCES

1 McKittrick, F. A. (1964) Cornell Univ. Agr. St.
Mem., 389: 1-197.

2 McKittrick, F. A. (1965) Ann. Entomol. Soc. Am.,
58: 18-22.

3 Cleveland, L. R., Hall, S. R., Sanders, E. P. and
Collier, J. (1934) Mem. Am. Acad. Arts Sci., 17:
185-342.

4 Seelinger, G. and Seelinger, U. (1983) Z. Tier-
psychol., 61: 315-333.

5 Nalepa, C. A. (1984) Behav. Ecol. Sociobiol., 14:
273-279.

6 Deleporte,P. (1985) Actes Coll. Insectes Soc., 2:
27-33.

7 Matsumoto, T. (1986) Proc. 10th Int. Nat. Congr.
IUSSI., Munchen, p. 129.

8 Shaw, E. (1925) Proc. Linn. Soc. N.S.W., 50: 171-
213.

9 Tillyard, R. J. (1926) The Insects of Australia and
New Zealand. Angus & Robertson, Sydney, 560pp.

10 Roth, L. M. (1977) Aust. J. Zool., Suppl. Ser., 48:
1-112.

11 Roth, L. M. and Willis, E. R. (1969) Smithsonian
Misc. Coll., 141: 1-470.

12 Liechti, P. M. and Bell, W. J. (1975) Insectes Soc.,
22: 469-470.

ZOOLOGICAL SCIENCE 5: 1149-1152 (1988)

© 1988 Zoological Society of Japai

[COMMUNICATION]

The Life Span of Wild Populations of the Fish
Oryzias latipes under Natural Conditions

Nobuo Egami, Osamu Terao1, and Yasuhiro Iwao2

2Department of Biology, Faculty of Science, Yamaguchi University,
Yamaguchi 753, 1 Zoological Institute, Faculty of Science,
Tokyo University, Tokyo 113, and National Institute
for Environmental Studies, Ibaraki 305, Japan

ABSTRACT — Wild populations of the medaka, Oryzias
latipes, were collected from irrigation canal or pond near
Imbanuma in Chiba Prefecture and irrigation pond in
Yamaguchi City at different times of the year. The
frequency distribution of the body length of each
population was then examined in order to provide
information on the breeding season, the rate of growth,
and the life span under natural conditions. From the data
it is likely that (1) the breeding season of the fish extends
from May to August, (2) the fish grow during the spring
and summer seasons, and (3) the life span of the wild
medaka is normally one year and a few months. It was,
however, demonstrated that if yearling fish were trans¬
ferred to aquaria and kept under favorable conditions,
some fish could survive more than two years. From the
present results and the life-span data obtained by our
previous observations under laboratory conditions, it is
concluded that life span of this species is quite different
between wild and captivity conditions; the longevity is
markedly modified by environmental conditions.

INTRODUCTION

The medaka, Oryzias latipes (small freshwater
teleost), is native to Japan and its adjacent regions.
This species has come to be widely used as a
laboratory animal in various fields in biology.
Because it is a laboratory animal, the life span of
the species under various conditions should be
examined. As a part of the examination, the
present observation was carried out. Life-span
data for the orange red variety of this fish under

Accepted January 9, 1988
Received November 18, 1987

captivity conditions in laboratory aquaria have
been reported; the mean life span is about 1,000
days, and the longest record 1,838 days [1,2]. On
the other hand, according to some observations
[3-6], the life span of the wild population of this
fish under natural conditions seems to be shorter
than under laboratory conditions; therefore, we
examined the body length distribution of samples
collected randomly from wild populations in var¬
ious seasons and estimated the life span of the fish
from the data. Furthermore, wild fish were
transferred to aquaria, and the subsequent survival
of the fish was examined.

The results suggest that the life span of this
species is quite different between favorable labora¬
tory conditions and natural conditions and that the
longevity is modified by environmental conditions.
The data will be briefly presented in the present
report.

MATERIALS AND METHODS

Two series of collections were carried out. In
the first series, samples were collected from the
irrigation canals of rice fields or small ponds near
Imbanuma, Chiba Prefecture, by O. Terao. The
date of each collection is given in Figures 1 and 2.
All fish caught were preserved in 70% ethanol
after fixation in formalin, and the standard length
of fixed individuals was measured.

In the second series, the sampling was carried
out at a small pond in Yamaguchi City, near

1150

N. Egami, O. Terao and Y. Iwao

Yamaguchi University, by N. Egami and Y. Iwao
and their students in 1985. After an examination
of the parasites of the fish, the total length of each
fish was measured (Fig. 3). Then, some live fish
were transferred to cylindrical aquaria 24 cm in
diameter in order to observe their mortality. The
transferred fish were subjected to disinfection with
methylen-blue, kept in well water, placed under
natural temperature and daylight conditions, and
fed with Tetra-Min food (Tetra Werke, West
Germany).

In both series, fish were caught at random by the
use of a dip net during the daytime. Since the fish
are in the habit of swimming near the surface of
the water in schools, frequently several fish were
caught at one time. Fish smaller than 10-15 mm in
body length could not always be collected by the
net. The ecological conditions of the collection
field and the stomach contents and gonadal condi¬
tions of the fish examined will be reported else¬
where by Terao.

RESULTS AND DISCUSSION

The date of collection, the number of fish
examined, and the frequency distribution of the
body length of each sample are given in Figures 1 ,
2 and 3, respectively. Figure 1 shows the results of
1983 and 1984, and Figure 2, those of 1985, in
Chiba Prefecture (eastern Japan). Figure 3 shows
the results of 1985 in Yamaguchi Prefecture
(western Japan). The standard length of the fixed
samples was measured in the first series (Figs. 1
and 2), while the total length of the live fish was
measured in the second series (Fig. 3).

A perusal of Figures 1, 2 and 3 shows no
significant differences in the body size of matured
fish between these two series. From Figure 1, the
following conclusions may be drawn: (1) The
growth of fish during October to March was almost
nil, since no significant difference in body-length
distribution is observable among the samples
collected in October, November, December, Janu¬
ary and March. (2) Both the yearling and under¬
yearling fish grew from spring to summer, since the
mean body length of both populations increased.
(3) In July, most yearlings disappeared from the
population while, on the contrary, a population of

1 1 2

30 NOV.

’83

90

29 Dec.

’83

84

28 Jan.

’84

86

2 March

’84

78

29 March

’84

155

W

24 April

’84

1 03

26 May

’84

84

1 6 June

’84

1 55y

2 July

’84

1 1 -

1 0 Aug.

’84

124

77V

1 Sept.

’84

121

10

15 20 25 30

5 Oct.

— _ i _

35

’84

Number
of Fish

Standard Length ( in mm )

Fig. 1. Frequency distribution of body length of fish
caught in 1983 and 1984 in Imbanuma.

small fish (less than 20 mm in standard length)
appeared. (4) In August the population of larger
fish (yearling fish) completely disappeared, and a
growth of the smaller fish (under-yearling fish)
took place from July to October. (5) In October
and later, the addition of young fish to the
population did not occur, since no individuals
smaller than 15 mm in standard length were found.

As is shown in Figure 2, a similar, successive
change in body length was observed in the 1985
seasons. In 1985, newly hatched young appeared
earlier than in 1984, but there were no essential
differences in seasonal change in body length
distribution between those two years. From these
facts, it is evident that the breeding season of this
fish is from May to July. A small number of
under-yearlings bred in August.

In the Yamaguchi population (Fig. 3), two
groups of different sizes are clearly distinguishable
in May. It was clear from the observation that the

Life Span of Oryzias latipes

1151

Number

of Fish Standard Length (in mm)

Fig. 2. Frequency distribution of body length of fish
caught in 1985 in Imbanuma.

larger fish were yearling males and females, while
the smaller ones were newly hatched young. The
male secondary sexual characteristics had not
developed in the under-yearling fish in May, but
they became clearly observable in June in some
large individuals. In July the under-yearling fish
grew rapidly and bred, while the yearling fish
disappeared.

It is clear from these data that the succession of
generations took place earlier in Yamaguchi than
in Chiba and that the breeding season is from April
to August.

A perusal of Figures 1, 2 and 3 will show (1) that
the breeding season of the fish extends from
April-May to August, (2) that the fish grow during
the spring and summer seasons, (3) that the
yearling fish and the under-yearling fish are dis¬
tinguishable by their body length, and (4) that the
yearling fish are not able to survive until the end of

108

J\ 8 May ’85

:

1 74

1

1

1

A 2 1 May ’85

189 j.

ifiTV y^^iniTTTmTK

20 5 - - 22 July 'SS

_ : _ .... . rtTTmfta*. _

175

1_ L

i / - 20 Aug.’ 85

i_ i ITITMTITt^ i

_i _ i _ i _ i _ i 1 1 1 ill 1 1 ill M II [rm-Ai _ l_

10 15 20 25 30 35 40

Number

of Fish Total Length ( in mm )

Fig. 3. Frequency distribution of body length of fish
caught in 1985 in the irrigation pond in Yamaguchi.
Hatched areas show males.

summer. Therefore, the life span of the wild fish is
one year and a few months. In other words, the
maximum life span under natural conditions is
shorter than one and a half years.

Kamito [5], Kubo and Sakurai [3], and Awaji
and Hanyu [4] have reported on the life cycle of
this species on the basis of natural observation or
the annual change in body size. As to the life cycle
of wild fish, the present data are in agreement with
their reports in important points.

It was observed in the Yamaguchi populations
that the majority of the yearling adults were slim in
shape and that most yearling fish were infected
with carp-lice ( Argulus japonicus ) or Lernaea
cyprinasea (Table 1).

In order to establish the probable cause of death
under natural conditions, on May 21 and June 15,
1985, 15 and 10 yearling fish respectively were
placed in cylindrical aquaria after the parasites had
been removed. Some fish died within a week after
the transfer; however, those fish which did not die
soon could survive for a long time. On September
18, more than a half of the fish still survived (Table
2); these survivors were plump and healthy in
appearance. Six survivors on September 18 (from

1152 N. Egami, O. Terao and Y. Iwao

Table 1 . Number of fish infected with Argulus and Lernaea

Date of examination

Number of fish examined

Total

Infected

May 8, 1985

40

28

June 15, 1985

48

35

Table 2. Number of fish transferred

to aquaria and survivors

Date of

Number of

Number of

survivors

transfer

fish transferred

Sept. 18, 1985

Sept. 30, 1986

May 21, 1985

15

8

-

June 15, 1985

10

6

2

Fig. 4. Example of fish infected with Lernaea. Besides
a large parasite, small parasites are observable on
the anal fins.

among those caught on June 15) were continuously
kept in captivity, and two of them survived until
the end of September, 1986 (Table 2). From these
facts, it seems highly probable that the fish have
the ability to survive more than two and a half
years if kept under favorable conditions, but they
die of environmental factors (such as infection,
unsufficiency of food, or being caught by a natural
enemy), not of any intrinsic cause, under severe
wild conditions.

We have already reported that the mean life
span of Oryzias latipes under captivity conditions is
about 1,000 days, the longest life span being 1,838
days. The present results show that, under natural
wild conditions, the longest life span is less than

1.5 years. It is highly probable that the main cause
of death under natural conditions is one ecological
factor or other and that the ecological life-span and
the physiological one are quite different in this
species. We must keep in mind, when using this
material as a model animal in biology, that the
ecological and physiological life spans of this fish
are quite different.

ACKNOWLEDGMENTS

The authors wish to thank Messrs. H. Yamazaki, H.
Tajiri, K. Ooya, M. Kumagai, S. Tanaka, S. Ito, and
Miss M. Nakaya of Yamaguchi University for their help
in collecting the fish. Acknowledgement is also due to
Professor A. Shima of Tokyo University for his en¬
couragement to O. Terao.

REFERENCES

1 Egami, N. and Etoh, H. (1969) Exp. Gerontol., 4:
127-129.

2 Egami, N. (1971) Exp. Gerontol., 6: 379-382.

3 Kubo, I. and Sakurai, H. (1951) Japan. J. Ichthyol.,
1: 339-346.

4 Awaji, M. and Hanyu, I. (1987) Nippon Suisan
Gakkaishi, 53: 959-965.

5 Kamito, A. (1928) J. Coll. Agric., 10: 21-38.

6 Terao, O. (1985) The Heredity (Iden), 39(8): 47-50.

Published by

the Japanese Society of Developmental Biologists

Papers in Vol. 30, No. 5. (October 1988)

38. REVIEW: W. H. Clark, Jr. and F. J. Griffin: The morphology and phusiology of the
acrosome reaction in the sperm of the decapod, Sicyonia ingentis

39. P. V. Davydov, O. I. Shubravyi, and S. G. Vassetzky: Larval development of starfishes as
revealed by long-term culture of the embryos

40. T. Matsuzawa, S. Ojima-Kasahara and M. Kitamura: Differential expressions of two types of
alkaline phosphatase during developmental growth of embryoid bodies of mouse terato-
carcinoma OTT-6050 in vitro

41. C. Nislow and J. Morrill: Regionalized cell division during sea urchin gastrulation contributes
to archenteron formation and is correlated with the establishment of larval symmertry

42. Y. Fukui: Cell proliferation during bud formation of the quail uropygial gland

43. M. Yamasaki and A. Tonesaki: Developmental stages of the society finch, Lonchura striate
var. domestica.

44. Y. Matsumoto, T. Kominami and M. Ishikawa: Timers in early development of sea urchin

embryos *

45. N. Tsuchimori, S. Miyashiro, H. Shibai and S. Ikegami: Significance of an increase of in-
teracellular adenosine concentration for dormancy in starfish blastulae

46 T. Yamashita, K. Kawamoto and S. Kawashima: Arginine vasopressin contents of the
hypothalamus and pictuiary during fetal and postnatal development in the mouse

47 M. Maeda: Dual effects of cAMP on the stability of prespore vesicles and 8-bromo cAMP-
enhanced maturation of spore and stalk cells of Dictyostelium discoideum

48 K. A. Tobal, D. K. Ellis and D. I. De Pomelal: Cellular src gene expression associated with
lenthoidogenesis in transdifferentiating cultures.

Development, Growth and Differentiation (ISSN 0012-1592) is published bimonthly by The Japanese Society of
Developmental Biologists, Department of Biology, School of Education, Waseda University, Tokyo 160, Japan.
1988: Volume 30. Annual subscription U. S. $ 110.00 including air speed delivery except Japan. Application to
mail at second class postage rate is pending at Jamaica, NY 11431, U. S. A.

Outside Japan : Send subscription orders and notices of change of address to Academic Press, Inc., Journal
Subscription Fulfillment Department, 6277 Sea Harbor Drive, Orlando, FL 32887, 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, 6277 Sea Harbor Drive, Orlando, FL 32887, 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.

VIII

INTERNATIONAL CONGRESS

OF

PROTOZOOLOGY

TSUKUBA, JAPAN
July 10- 17, 1989

Abstract Deadline: Feb. 28, 1989

PLENARY LECTURES

1 . Evolution of nuclear phenomena in protozoa

2. The contribution of hypotrich ciliates to our understanding of molecular
biology and evolution of ciliates.

3. Cilia in cell motility: membrane-controlled rotary engines

4. Molecular basis of differentiation in African trypanosomes

5. Advances in the biology of intestinal protozoa: Entoamoeba histolytica
and Giardia lamblia

6. The role of protozoa in nature in terms of physiological constraints of
protozoan organization

SYMPOSIA

1. Genome organization and reorganization in ciliated protozoa

2. Protozoan genes: their structure, products and regulation of expression

3. Regulation of cell structure

4. Cell interactions in sexual phenomena

5. Role of cell membranes and cytoskeletal organization in cell motility

6. Cytology and cytochemistry of rumen protozoa

7. Immunology and chemotherapy of Plasmodium

8. Biology of haemoflagellates

9. Cyst-forming coccidia

10. Surviving mechanisms of intracellular protozoa

11. Piroplasmids

1 2. Dynamics of ecology of free-living protozoa

13. Endosymbionts of protozoa

14. Phylogeny and evolution of protozoa

15. Marine protozooplankton

16. Role of protozoa in biotechnology

CONTRIBUTED PAPER - 28 SESSIONS

(PLATFORM OR POSTER PRESENTATIONS)

For more information, contact: Yoshinori Nozawa

Secretary General

VIII International Congress of Protozoology
Department of Biochemist™

Gifu University School of Medicine
Tsukasamachi-40, Gifu 500, Japan

Typesetting on Linotronic 300R, by the courtesy of Iwata Co., Ltd.

NARISHIGE

THE ULTIMATE NAME IN MICROMANIPULATION

OUR NEW MODELS MO-102 and MO-103
MAKE PRECISION MICROMANIPULATION SO EASY!

(Photo: by courtesy of Olympus Optical CO., LTD.)

SOME FEATURES of MO-102 and MO-103:

* The manipulator head is so small that it can be mounted directly on the
microscope stage. There is no need for a bulky stand.

* Hydraulic remote control ensures totally vibration-free operation.

* 3-D movements achieved with a single joystick.

Micromanipulators Microelectrode pullers Stereotaxic instruments

NARismoE) NARISHIGE SCIENTIFIC INSTRUMENT
LABORATORY CO., LTD.

4-9-28, Kasuya, Setagaya-ku, Tokyo 157 JAPAN
Telephone: 03-308-8233 Telex: NARISHG J27781

(i Contents continued from back cover)

identical to the small prothoracicotropic hor¬
mone (4K-PTTH)? . 1051

Reproductive Biology

Awaji, M. and I. Hanyu: Effects of water
temperature and photoperiod on the begin¬
ning of spawning season in the orange-red
type medaka . 1059

Kosaka, T., M. Obata, T. R. Saito and K. W.
Takahashi: Effects of adult male cohabita¬
tion on precocious puberty in early weaning
female Guinea pigs (COMMUNICATION)
. 1137

Morphology

Chiba, A. and Y. Honma: Fine structure of
agranular cells in the gummy shark {Mustelus
manazo) adenohypophysis . 1065

Behavior Biology

Saito, T. R., K. Kamata, M. Nakamura and
M. Inaba: Maternal behavior in virgin
female rats following removal of the vomer¬
onasal organ (COMMUNICATION) . . 1141

Ecology

Takahashi, H. and H. Iwasawa: Interpopula¬
tion variations in clutch size and egg size in
the Japanese salamander, Hynobius nigres-

1073

Meserve, L. A. and M. A. R. Gonzalez:
Thyroid status and ambient temperature as
influences on weaning in young mice . . . 1083
Matsumoto, T.: Colony composition of the
wood-feeding cockroach, Panesthia australis
Brunner (Blattaria, Blaberidae, Panes-
thiinae) in Australia (COMMUNICATION)

. 1145

Egami, N., O. Terao and Y. Iwao: The life
span of wild populations of the fish Oryzias
latipes under natural conditions (COM¬
MUNICATION) . 1149

Taxonomy

Hirayama, A.: A ghost shrimp with four-
articulate fifth pereopods (Crustacea: Cap-
rellidea: Phtisicidae) from northwest Aus¬
tralia . 1089

Zhang, W. X. and M. J. Toda: The Dro¬
sophila immigrans species-group of the sub¬
genus Drosophila (Diptera: Drosophilidae)

in Yunnan, China . 1095

Nakasone, Y,: Larval stages of Coenobita
purpureus Stimpson and C. cavipes Stimpson
reared in the laboratory and survival rates
and growth factors of three land hermit crab

larvae (Crustacea: Anomura) . 1105

Hayashi, T. and M. Matsui: Biochemical dif¬
ferentiation in Japanese newts, genus
Cynops (Salamandridae) . 1121

cens

ZOOLOGICAL SCIENCE

VOLUME 5 NUMBER 5

OCTOBER 1988

CONTENTS

REVIEWS

De Santis, R. and M. R. Pinto: The pathway

of sperm -egg interaction in ascidians : biolo¬
gy and chemistry .............. _ ....... 919

Urano, A.: Neuroendocrine control of anu-
ran anterior preoptic neurons and initiation
of mating behavior . . . 925

ORIGINAL PAPERS

Physiology

Inoda, T., H. Ohtake and M. Morisawa:

Activation of respiration and initiation of
motility in rainbow trout spermatozoa . . . 939
Hidaka, T. and S. Yukiyama: Excitatory and
inhibitory junction potentials recorded from
the red muscle of marine teleost, puffer fish

. . . 947

Negishi, S.: The involvement of microtubules
in the light response of medaka mela-

nophores . . 951

Grundstrom, N., H. Sundgren, J.-O. G.
Karlsson, and IT El wing: A simple and
efficient, method for photometric estimation
of the state of pigment aggregation in fish

melanophores . . . . . . . 959

Endo, Y.: Non-synaptic release of transmit¬
ter-containing vesicles from the enteric
neurons of the rat small intestine ........ 965

Cell Biology

Ishida, H., Y. Shigenaka and M. Imada: Fi¬
brillar system and possible control mecha¬
nism for the cycle of contraction and elonga¬
tion of Spirostomum ambiguum . 973

Developmental Biology

Yamamoto, M. , M. Ishine and M. Yoshida:
Gonadal maturation independent of photic

Indexed in-

Current Contents/LS and AB & ES,

Science Citation Index ,

ISI Online Database ,

CABS Database , INFOBIB

conditions in laboratory-reared sea urchins,
Pseudocentrotus depressus and Hemicentro -
tus pulcherrimus ......................... 979

Yamamoto, M.: Noma l embryonic stages of
the pygmy cuttlefish, Idiosepius pygmaeus
paradoxus Ortmann ...................... 989

Mizuuo, T. , H. Takeda, N. Suematsu, N.
Hironaka and I. Lasnitzki: Absence of
androgen receptors in the prostatic glandular
epithelium derived from testicular feminiza¬
tion mutant (Tfm) mice .... . . . 999

Endocrinology

Tanaka, S., H. Iwasawa and K. Wakabayashi:
Plasma levels of androgens in growing frogs

of Rana nigromaculata . . . . 1007

Oota, Y. and I. Koshimizu: Vascular supply
of hypophysis in the turtle, Geoclemys reeve -
sii ...................................... 1013

Yamashita, T., K. Kawamoto, and S.
Kawashima: Fetal and postnatal develop¬
ment of arginine vasopressi n-im rn uno-

reactive neurons in the mouse . . 1019

Hyodo, S., M. Fujiwara, M. Sato and A. Ura¬
no: Molecular- and immuno-histochemical
study on expressions of vasopressin and ox¬
ytocin genes following sodium loading . 1033
Nishida, M., I. Kawada, H. Ishizuka and S.
Katsura: Goitrogenic action of manganese
on female mouse thyroid through three gen¬
erations _ ............................ 1043

Masaki, T., K. Endo and K. Kumagai:
Neuroendocrine regulation of the develop¬
ment of seasonal morphs in the Asian com¬
ma butterfly, Polygonia c-aureum L. : Is the
factor producing summer morphs (SMPH)

(Contents continued on inside back cover)

Issued on October 15
Printed by Daigaku Printing Co., Ltd.,
Hiroshima, Japan

- ZOOLOGICAL

SCIENCE

Proceedings of the

Fifty— Ninth Annual Meeting of the

Zoological Society of Japan

October 8-10, 1988
Sapporo

Vol. 5 No. 6 December 1988

published by Zoological Society of Japan

. Business Center for Academic Societies Japan

distributed by ysp ^ The Nether|ands

ISSN 0289-0003

ZOOLOGICAL SCIENCE

The Official Journal of the Zoological Society of Japan

Editor-in-Chief:

Hideshi Kobayashi (Tokyo)
Managing Editor:

Seiichiro Kawashima (Tokyo)
Assistant Editors:

Takeo Machida (Urawa)

Sumio Takahashi (Hiroshima)
Kazuyoshi Tsutsui (Hiroshima)

The Zoological Society of Japan:
Toshin-building, Hongo 2-27-2, Bunkyo-ku,
Tokyo 113, Japan. Tel. (03) 814-5675
Officers:

President: Nobuo Egami (Tokyo)
Secretary: Hideo Namiki (Tokyo)
Treasurer: Tadakazu Ohoka (Tokyo)
Librarian: Shun-Ichi Ueno (Tokyo)

Editorial Board:

Howard A. Bern (Berkeley)
Horst Grunz (Essen)

Susumu Ishii (Tokyo)

Roger Milkman (Iowa City)
Tokindo S. Okada (Okazaki)
Hiroshi Watanabe (Shimoda)

Walter Bock (New York)
Robert B. Hill (Kingston)
Yukiaki Kuroda (Mishima)
Hiromichi Morita (Fukuoka)
Andreas Oksche (Giessen)
Mayumi Yamada (Sapporo)

Aubrey Gorbman (Seattle)
Yukio Hiramoto (Chiba)
Koscak Maruyama (Chiba)
Kazuo Moriwaki (Mishima)
Hidemi Sato (Nagoya)

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 more than 1000 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:

Prof. C. Oguro, Managing Editor (for 1989-), 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 1988, The Zoological Society of Japan

[ Publication of Zoological Science has been supported in part by a Grant-in- Aid for Publicaton 1
I of Scientific Research Result from the Ministry of Education, Science and Culture, Japan. J

ZOOLOGICAL SCIENCE 5: 1153-1163 (1988)

© 1988 Zoological Society of Japan

REVIEW

Planarian Regeneration: An Overview
of Some Cellular Mechanisms

Vittorio Gremigni

Dipartimento di Biomedicina Sperimentale, Infettiva e Pubblica-Sezione di
Biologia e Genetica, Universita di Pisa, 56100 Pisa, Italy

INTRODUCTION

Planarian regeneration has always attracted the
attention of naturalists and biologists w'ho in past
centuries tackled the problems related to this
phenomenon by means of various experimental
approaches (see [1] for a comprehensive, historical
review). One of the main stimuli leading to
investigations of planarian regeneration has been
the curiosity as to the mechanisms by which even
very small pieces of a worm (down to about
l/300th of the body according to Morgan [2]; 0.08
mm3 fragments corresponding approximately to 1
X 104 cells according to Montgomery and Coward
[3]) can reconstitute the entire organism. Thus,
this field of research also has the advantage that
most experimental procedures elude even ethical
problems related to vivisection.

Another reason for the continuous interest in
planarian regeneration is probably due to the fact
that, despite the enormous number of papers
published on the topic, several important problems
remain unsolved, and many points are still con¬
troversially debated by different researchers. This
is partly due to difficulties and delays in adopting
modern techniques such as those commonly em¬
ployed in molecular biology and genetics, but it
also reflects, in my opinion, the fact that the
designing and interpretation of useful experiments
must necessarily involve a deep knowledge of the
basis for previous hypotheses and theories and
above all take into account previous data available

Received June 27, 1988

on the topic concerned. Unfortunately, these
elementary scientific rules have not always been
applied.

From a general point of view, the term “Regen¬
eration”, as applied to planarians, refers to both
organismic and cellular repair, and is commonly
applied to phenomena and processes which may
have different origins and mechanisms. Therefore,
I believe that a subdivision would be suitable at
least between: A) Traumatic or Reparative Regen¬
eration and B) Physiological Regeneration.

Traumatic regeneration is by far the more stu¬
died type of regeneration in planarians and is
commonly obtained by cutting a worm at one or
more levels and studying the subsequent steps of
restoration of the missing body parts. This type of
regeneration involves as a first step wound closure
and healing and the formation of a blastema from
which the amputed tissues and organs are rebuilt.
Variations of the classical traumatic regeneration
that do not involve healing and blastema formation
include those following X-ray irradiation or pro¬
longed starvation.

Physiological regeneration, on the other hand,
occurs for example during: a) asexual reproduction
(also called somatic embryogenesis) ; b) normal
tissue renewal for maintainance and c) regrowing
and organ differentiation in species or populations
undergoing seasonal (sexual and asexual) cycles.
During physiological regeneration, neither wound
healing nor blastema formation occurs.

An accurate analysis of different types of plana¬
rian regeneration involves many problems from
the morphogenetic to the cellular and molecular

1154

V. Gremigni

levels (see [1, 4-6] for comprehensive reviews).
This article addresses only a few problems related
to cellular mechanisms in traumatic regeneration
and in particular discusses some of the crucial
points which are presently object of debate.

WOUND CLOSURE AND HEALING

In planarians a wound is commonly defined as
an injury characterized by an interruption of the
body surface including the loss of at least some
epidermal and underlying subepidermal and mus¬
cle tissues and thus opening the unprotected
parenchyma to the external environment. Wound
closure or epidermal healing is the first cellular
event occurring immediately after a transection or
injury, and according to Pedersen [7]; p. 267, can
be considered complete when an uniterrupted
epidermal covering is formed and “...no morpho¬
logical differences between adjacent old epidermal
cells and newly formed epidermal cells can be
discerned”. Wound healing is a more complex
process consisting not merely of a restoration of
the epidermal layer, but also of repair of the
subepidermal membrane and of the underlying
muscle layers. It involves many basic cellular
events such as cell locomotion and migration,
division, differentiation, interaction etc. This pro¬
cess seems to require different times depending
mainly on the different methods and criteria used
by various workers for evaluating when wound
closure and/or healing is complete. Moreover, the
time needed for this process varies according to
many internal and environmental factors such as:
the species or population type, the age of the
worm, the region of the body, the size of the
wound, the season, temperature, light conditions
etc. According to well-documented S.E.M. inves¬
tigations, Pedersen [7] concludes, in agreement
with most of the ultrastructural literature, that the
median time values necessary to complete wound
healing in planarians is between 1 and 3 days, even
though a protective epidermal covering, not due to
newly differentiated epidermal cells, often appears
much earlier [8-10].

According to more recent conventional S.E.M.
and T.E.M. studies and to cytochemical, im-
munoelectronmicroscopical and autoradiographic

investigations [8, 11-16], at transection the wound
surface shows the presence of rounded free cells,
damaged cells, cell fragments and debris, compo¬
nents of which are often ejected by muscle contrac¬
tion. Secreted material is also spread over the
wound region, while epidermal cells in the vicinity
of the wound contain many lysosomes and are
probably involved in autophagic activity. Soon
after transection, the wound is reduced in size by
strong, rapid contractions of the muscle layers
lying beneath the epidermis. Then muscle contrac¬
tions stop, and stretching and flattening of the
poorly differentiated old epidermal cells bordering
the wound occur which contribute to the formation
of a very thin (1.5 //m), initially poorly organized,
heteromorphic epidermal layer. During subse¬
quent stages (between 2 and 3 days after transec¬
tion according to Hori [13]) the epidermis is rapid¬
ly recruited by newly differentiated epidermal cells
coming from the underlying parenchyma. These
cells become intermingled with the former reorga¬
nized tissues. According to most authors, the
newly differentiated epidermal cells are derived
from free immature parenchymal cells which
actively migrate to the wound. Both flattening and
short migration (spreading at the wound) of local
old epidermal cells, and migration of newly dif¬
ferentiating cells from the parenchyma were de¬
monstrated by the brilliant cytochemical and im-
munoelectronmicroscopical investigations on the
cytoskeleton of Pascolini et al. [14, 15] who de¬
tected the presence of actin and adenylate cyclase
activity in migrating epidermal and parenchymal
cells during wound healing. Cell motility was
inhibited and wound closure consequently delayed
after cytochalasin B treatment. Cell migration
appears to depend on the presence of a suitable
substrate composed of secretory material released
by rhabdite-forming cells during wound healing.
Finally the basement membrane is reconstituted by
an interaction between the epidermal cells and
myoblasts [17, 18].

No cells in the wound area appear to divide or to
be involved in DNA or intense RNA synthesis
during the first hours of healing when cell autolysis
is a common pattern and true regeneration has still
to begin. This was clearly shown by Palmberg [16]
through autoradiographic experiments performed

Cellular Aspects of Planarian Regeneration

1155

in Microstomum.

Experiments involving exposure to massive X-
ray doses before transection and colchicine treat¬
ment just after transection showed that wound
closure by old epidermal cells starts as usual but
fails to form a complete epithelial layer. Thus, it
was suggested that pre-existing specialized
epidermal or rhabdite-forming cells alone do not
allow complete healing by morphallaxis and that
lost epidermal cells must be recruited from newly
differentiating parenchymal or blastema cells by
epimorphosis [13].

REGENERATION BLASTEMA

Blastema formation involves by far the most
debated and unresolved problems concerning the
cellular mechanisms of planarian regeneration,
despite the variety of experimental approaches
used to clarify them.

Blastema morphology

The blastema is commonly defined as an accu¬
mulation of undifferentiated or poorly differenti¬
ated cells at the injured area extending between
the wound and the old stump. The first problem
concerning the blastema is to clearly define and
circumscribe as well as possible its borders. It has
been hypothesized that the blastema begins to
form soon after wound closure or healing [6, 8]
even though the first accumulation of blastema
cells probably occurs contemporaneously with
wound healing. A few days after transection, the
blastema usually reaches a large size and becomes
clearly visible even to the naked eye in pigmented
specimens as un unpigmented area posterior to the
wound. But the true boundaries between the
blastema and the old stump are not so easily
distinguishable and can be identified only by elec¬
tron microscopy which reveals small differences
between blastema cells and cells of the old tissues.
This point is not irrelevant if one considers that
some of the controversy concerning blastema cell
origin, activity and fate is possibly due to an
imperfect definition of this crucial structure.

Electron micrographs indicate that the blastema
is first formed by a homogeneous population of
roundish or elliptical undifferentiated cells 7-10

Fig. 1. Electron micrograph of a 36-hrs-old blastema
showing portions of three undifferentiated cells with
the characteristic thin cytoplasm rich in free ribo¬
somes. Some scattered chromatin clumps and a
nucleolus are visible within the nucleus. X7,000.

pm in diameter (Fig. 1) intermingled with some
specialized, often degenerating cells. Later on, an
increasing number of cells showing progressive
signs of differentiation, appear in the distal region
of the blastema, while mostly undifferentiated cells
continue to characterize the proximal region adja¬
cent to the stump. No membrane or specialized
structure has been described to delimit the blaste¬
ma from the old tissues; they are in continuity each
other. The early blastema also appear to be
characterized by a certain looseness in structural
organization i.e. by large intercellular spaces con¬
taining a homogeneous material, perhaps useful
for cell locomotion, that is unusual in contrast to
the neighbouring old stump. Great attention must
therefore be taken in defining the various zones
involved in regeneration, i.e. preblastema, blaste¬
ma, postblastema, prestump, stump etc.

The origin of blastema cells

The crucial point debated in blastema formation
is the origin of blastema cells about which even the
recent literature reveals strongly contradictory
theories. Two main theories have been proposed
and strongly supported or criticized: A) ‘The
Neoblast Theory’ and B) ‘The Dedifferentiation
Theory’ [19]. The neoblast theory, first proposed
in the last century (see [1, 20] for references) has
been clearly reproposed and variously supported
by the French school of Etienne Wolff (see [1, 4, 5]

1156

V. Gremigni

for references). It claims that the blastema is
exclusively built up by specific totipotent,
embryonic reserve cells “The Neoblasts ”, which
remain distributed throughout the body of the
planarian throughout its life and are capable of
migrating and proliferating following a regenera¬
tive stimulus.

The dedifferentiation theory supported among
others by Steinman [21], Lang [22], Hyman [23],
has recently been reproposed by Flickinger [24],
Hay [25] and Coward [26]. This theory claims that
a stock of embryonic, totipotent cells does not
exist in adult planarians or, if it does, it is not
involved in blastema formation. Therefore, regen¬
eration is always due to dedifferentiated cells.

Several variations of these two theories and
conciliatory hypotheses have been proposed by
different authors who approached the problem
[27, 28, 29]. I am among these [30] and would like
to recall the words of Pedersen [17], p. 135 who,
on the basis of the information available at the
time, clearly summarized the state of art on the
point, writing that “•••it is unrealistic to be too
strict in the concept of only one mechanism being
operative in the formation of the blastema. Both
processes, activation of neoblasts and dedif¬
ferentiation of specialized cells, may be operative,
and depending on the circumstances one or the
other mechanism may be dominating’5.

Neoblast structure

The existence in both embryonic and adult
stages of planarian life of undifferentiated cells
more or less uniformly distributed troughout the
body is undeniable. Light and electron microscope
observations confirm these data and define the
morphological characteristics of the so-called
neoblasts even though they do not lead to any
clarity about their origin. Neoblasts are small,
usually ovoid or pear-shaped, cells (6-10 ^mx4-6
fjx n in diameter) with a very high nucleo-
cytoplasmic ratio. They have a very large, round
or ovoid nucleus with a small nucleolus and mostly
diffused or finely clumped chromatin. The cyto¬
plasm is reduced to a strongly basophilic narrow
rim, rich in free ribosomes and containing few
mitochondria. Transmission electron microscopy
is essential for recognizing and characterizing

neoblasts, since many other cells appear similar,
during their cell cycles, at the level of light micros¬
copy. However, when properly studied, these cells
show important differences with truly undifferenti¬
ated cells, i.e. clear signs of cell differentiation or
dedifferentiation [31]. It is noteworthy that cells
reaching up to 15-20 pm and even more in dia¬
meter have often been erroneously interpreted as
neoblasts at the light microscope level.

Nearly all researchers who tackled this point by
means of electron microscopy agree with the view
that cells with the above described morphological
and ultrastmctural features are freely scattered in
developing embryos [32, 33] as well as in the
parenchyma of intact and regenerating adult
planarians [34-38] and in the blastema [17, 37, 39,
40]. What then divides the supporters of the two
conflicting theories? According to suporters of the
neoblast theory, the undifferentiated cells of adult
planarians are embryonic, totipotent cells which
have never undergone differentiation and maintain
the capability of dividing. They constitute a
permanent stock of cells ready to become active
after every type of traumatic or physiological re¬
generation (see [1, 41] for references). By con¬
trast, according to the dedifferentiation theory
supporters, the undifferentiated cells are not
embryonic, reserve cells, but rather dedifferenti¬
ated cells with a very short life span. They would
mostly represent stages in the cycles of secretory,
gastrodermal, muscle cells etc. [31, 42]. There¬
fore, they would not necessarily be totipotent cells.

Cell markers and experimental approaches

Many attempts have been made to resolve this
intriguing conflict and numerous presumptive une¬
quivocal proofs have been claimed by differnt
authors supporting one or the other theory. As a
matter of fact, I believe that morphological
observations alone, even though made at the E.M.
level, cannot completely clarify the origin of blas¬
tema cells. Thus, the best way to answer this
question is, in my opinion, to look for true and
permanent cytological or molecular markers that
can allow one to easily and surely distinguish
between embryonic undifferentiated cells and
differentiated somatic or germ cells.

Some years ago I had the chance to tackle this

Cellular Aspects of Planarian Regeneration

1157

Fig. 2. Scheme of the chromosome cycle of the triplo-hexaploid population from the lake of Iseo.

problem, thanks to the availability in our labora¬
tory of a pseudogamic population of the Dugesia
lugubris-polychroa group provided with a perma¬
nent karyological marker allowing one to easily
recognize male and female germ cells from
embryonic and somatic cells. In these specimens

Fig. 3. Schematic representation of the transection ex¬
periments through which the presence of diploid or
hexaploid cells were alternatively observed in the
regenerative blastemas. a.r. = anterior regenerant;
a.r.b.=anterior regenerant blastema; p.r.=post-
erior regenerant; p.r.b. ^posterior regenerant blas¬
tema; o = ovaries; t = testes.

the zygote (or pseudozygote) as well as embryonic
and somatic cells are triploid (3n = 12 chromo¬
somes), while germ cells before undergoing
meiosis are diploid (male germ cells 2n = 8
chromosomes) or hexaploid (female germ cells 6n
=24 chromosomes) [43]. (Fig. 2).

In an initial series of karyological and ultras-
tructural experiments [40, 44] we demonstrated
that following a transection made just beneath the
ovaries or behind the testes, the blastemas formed
by regenerates which were provided with alterna¬
tively female or male gonads, contained a small
percentage of hexaploid (female, about 1%) or
diploid (male, about 5%) cells along with a high
percentage of triploid (embryonic reserve or soma¬
tic dedifferentiated) cells (Fig. 3). These findings
showed that cells which had already undergone an
important nuclear differentiation (or determina¬
tion) and became germ cells, can contribute to
blastema formation.

In a second series of karyological and cytopho-
tometric (nuclear DNA content) experiments [45,
46] it was demonstrated that germ cells which
contributed to blastema formation can colonize
regions of the body devoid of gonads (and where
gonads will never be formed) during regeneration
(Fig. 4). From these experiments we hypothesized
that germ cells, once reaching the blastema, be¬
have as all other blastema cells taking part in the
rebuilding of any removed part of the body, even
though devoid of gonads. In other words, we
suggested that original germ cells taking part in
blastema formation behave as pluripotent instead
of unipotent cells.

Finally, in a third series of cytophotometric and
karyological experiments [47, 48] we showed that
the original germ cells which took part in blastema

1158

V. Gremigni

A

B

Fig. 4. Schemes of the experiments carried out to obtain cephalic (I) or caudal (II) blastemas from regenerated
planarians. The hatched area represent the regenerated portion of the body. A : dividing cells from a 36-hrs-old
cephalic blastema (ce.r.b.); B: dividing cells from a 36-hrs-old caudal blastema (ca.r.b.); ce.r=cephalic
regenerant; ca.r.=caudal regenerant.

formation are also capable of differentiating into
different cell types, i.e. somatic (pharyngeal mus¬
cle) or germ (male to female) cells, during regen¬
eration (Fig. 5). From the experiments described
above we argued that in planarians; a) blastema
formation and the subsequent regeneration of re¬
moved tissues are due mainly to undifferentiated

cells which, on the basis of other previous inves¬
tigations [49], can be interpreted as reserve,
embryonic cells or neoblasts; b) a small percentage
of premeiotic germ cells can participate in blaste¬
ma formation, thus becoming pluripotent again; c)
a similar process of dedetermination-dedifferen¬
tiation may also occur in some somatic cell types,

Cellular Aspects of Planarian Regeneration

1159

Fig. 5a. Scheme of the transection experiments carried out to obtain a regenerated pharynx from regenerans
containing testes. The histogram represents the distribution of optic absorbance in Feulgen treated nuclei of
muscle cells from the regenerated pharynx. The small peak on the left (mean value about 30 a.u.) is formed by
diploid nuclei. The high peak on the right (mean value about 45 a.u.) is formed by triploid nuclei.

Fig. 5b. Scheme of the experiments in which eggs with 8 bivalents (along with eggs with 12 bivalents as usual) were
laid by animals provided with regenerated ovaries. ph.= pharynx.

provided they did not reach an irreversible stage
od specialization (for example spermatids and
spermatozoa do not dedifferentiate, but eventually
degenerate following transection near the testes
[49, 50]). This suggestion was also based on some
ultrastructural observations [31, 40] but cannot be
proven by the karyological marker available since
the latter does not allow descrimination between
neoblasts and specialized somatic cells; d) all blas¬
tema cells, whatever their source, are capable of

differentiating into any cell type during regenera¬
tion, giving rise to the rebuilt tissues, depending on
the field influence they encounter [51, 52]. There¬
fore, some cells in planarians (germ cells in our
experiments) my undergo transdifferentiation, or
at least transdetermination as suggested by Baguna

[53].

A similar approach based on the use of specific
nuclear or cytoplasmic markers has been proposed
and preliminarly used by Baguna and his cowor-

1160

V. Gremigni

kers (see [9] for general reference). To date they
report the identification of two separate cell
populations: a) neoblasts and b) differentiated
cells, by serial filtration through nylon and discon¬
tinuous density gradients, respectively. Moreover,
they labeled some specialized cells with fluorescent
latex beads that were taken up by pinocytosis and
remained within the cell for long periods of time
[54]. Through this technique, combined with graft¬
ing experiments, the authors obtained evidence
that only neoblasts take part in blastema forma¬
tion. The avenue of research followed is stimulat¬
ing, and it is hoped that in the near future ultras-
tructural investigations will confirm the true nature
of the different cell populations, and that specific
antibodies will be obtained to label different cell
types. The preliminary investigations of Collet et
al. [55] who identified some soluble or particulate
proteins and isolated them from undifferentiated
or differentiated cells appear to be very en¬
couraging.

Blastema cell proliferation

This is another contested area in planarian re¬
generation still debated by some authors, but in
my opinion it is a trivial point. It is widely known
that the regeneration blastema is by definition an
accumulation of undifferentiated cells capable of
dividing and then differentiating. Moreover, it is
well known to every planarian student that when a
careful chromosomal and karyometric analysis is
needed, the best way to proceed is to induce by a
terminal transection the formation of a blastema
where, following colchicine treatment or not, a
great number of dividing cells is encountered.
Mitotic cells have also been identified in sectioned
specimens within the blastema at both the light and
electron microscope levels (see [1, 13, 25]).
However, a few papers refer to the absence of
mitotic activity within the blastema [8, 55, 56]. In
particular, the latter authors suggest that cell divi¬
sion does not occur within the blastema, but rather
in the area they call the “postblastema” where
proliferating, undifferentiated cells accumulate be¬
fore “crossing” into the blastema. My opinion is
that agreement as to the correct definition of
blastema could remove this problem. Nobody
doubts the fact that many undifferentiated cells

grouped near the wound begin to divided soon
after transection and that a number of mitotic cells
continue to be evident a few days later in the area
behind the wound. Why not to define as “blaste¬
ma” this accumulation of proliferating, undiffe¬
rentiated cells which is lacking in intact animals
where free neoblasts are randomly scattered in the
parenchyma ? I suggest that both the body area
located just beneath the newly formed epidermis,
which mainly consists of non-dividing, differentiat¬
ing cells (Distal Area), and the area adjacent to
the old stump, which is mainly composed of un¬
differentiated, dividing cells (Proximal Area), are
part of the blastema.

As far as the nature of cells undergoing mitosis is
concerned, there is no doubt that only undiffe¬
rentiated, and not also specialized cells, are cap¬
able of dividing, but the problem is to know
whether presently undifferentiated cells are solely
neoblasts (see the orgin of blastema cells). The
finding that in the very first few hours following
transection a high mitotic index occurs [56-58] led
to the suggestion that the first cells to be activated
and to proliferate after transection are free neob¬
lasts in the G2 period of the interphase. This
hypothesis appears to be internally consistent
(even though timing of the neoblast cell cycle has
not been yet determined), but does not exclude the
possibility that later on, when a second peak of
mitotic activity occurs, dedifferentiated cells also
undergo mitosis.

Migration of blastema-forming cells

Another problem in planarian regeneration con¬
cerns the migratory capability of blastema cells,
whatever their source might be. The majority of
investigations performed thus far seem to indicate
that cells also coming from distant areas of the
body can take part in blastema formation along
with local cells, particularly in the days following
transection. Among these, the classical studies of
Wolff and Dubois [59, 60] and Dubois [61, 62],
later confirmed by Cecere et al. [63], deserve
citations for the originality of their approaches.

This convinction has been rejected by some
researches (see [28, 53, 58] for references) accord¬
ing to whom neoblasts (or whatever blastema¬
forming cells) only spread randomly over the area

Cellular Aspects of Planarian Regeneration

1161

near the wound following proliferation. The last
authors, however, admit that neoblasts are
“wandering cells” able to move and that the epi-
morphic process in planarian regeneration involves
“cell migration” [56]. Thus, the problem appears
to confined to determining the distance blastema¬
forming cells migrate and the orientation of their
movements.

In our laboratories some clear morphological
and ultrastructural evidence supporting the
hypothesis that cells taking part in blastema forma¬
tion do migrate to the wound was obtained [48].
The same sets of experiments also showed that
blastema formation and regeneration are much
faster when transection is performed in areas of
the body where a great number of undifferentiated
and/or poorly differentiated cells are present.
These results support the view that the first accu¬
mulation of blastema cells has a local origin.

CONCLUSIONS

Some of the points discussed above for trauma¬
tic regeneration could be reproposed as well for
the various types of non-traumatic regeneration
where a blastema is not formed, but where re¬
generating cells might be either neoblasts or de¬
differentiated cells, or both, and might migrate or
not, etc. The great majority of data published until
now favors a unique or major role of the neoblasts
in regeneration, but are mainly based on indirect
evidence, while only some but increasingly more
direct evidence seems to support the hypothesis
that dedifferentiated cells can also take part in
regeneration. My personal convinction is that both
neoblasts and dedifferentiated cells (at least those
not yet completely specialized) can take part in
each type of regeneration process in planarians.
However, perhaps most authors have planned
their experiments and interpreted their data with
the inconscious hope of demonstrating that only
one of the theories proposed is right and that the
other is wrong. I also believe that the avenue of
investigation suggested by the experiments of my
group through the use of permanent cell markers is
presently the most likely to lead to unequivocal
answers to most of the problems still debated on
planarian regeneration. I also feel that paradox¬

ically enough, this type of approach is more useful
for demonstrating the eventual occurrence of cell
dedifferentiation and transdifferentiation, than for
following the fate of embryonic, reserve cells
which presumably have few structures and mole¬
cules that lack in every differentiating or special¬
ized cells. In any case, I believe that every
researcher interested in this topic will anxiously
follow the next steps in the very stimulating prog¬
ram suggested by Baguna etal. [9], i.e. the passage
from theoretical and speculative proposals to the
operative experimental phase involving cellular,
molecular and genetic approaches leading to con¬
sistent and repetitive results.

REFERENCES

1 Brndsted, H. V. (1969) Planarian Regeneration,
Pergamon Press, Oxford, pp. 1-276.

2 Morgan, T. H. (1989) Experimental studies of the
regeneration in Planaria lugubris. Arch. Entwick-
lungsmech, 13: 364-397.

3 Montgomery, J. R. and Coward, S. J. (1974) On
the minimal size of a planarian capable of regenera¬
tion. Trans. Am. Microsc. Soc., 93: 386-391.

4 Wolff, E. (1962) Recent researches on the regenera¬
tion of Planaria. In “Regeneration”. Ed. by D.
Rudnick, The Ronald Press, New York, pp. 53-84.

5 Lender, Th. (1962) Factors in morphogenesis of
regenerating fresh-water Planaria. In “Advances in
Morphogenesis”. Ed. by M. Abercrombie and J.
Brachet, Acedemic Press, New York, Vol. II, pp.
305-331.

6 Chandebois, R. (1976) Histogenesis and Mor¬
phogenesis in planarian regeneration. In “Mono¬
graphs in Developmental Biology”. Ed. by A. Wols-
ky, S. Karger, Basel, Vol. 11, pp. 1-182.

7 Pedersen, K. J. (1976) Scanning electron microsco¬
pical observation on epidermal wound healing in the
planarian Dugesia tigrina. Whilhelm Roux’s Arch.,
179: 251-273.

8 Spiegelman, M. and Dudley, P. L. (1973) Morpho¬
logical stages of regeneration in the planarian Duge¬
sia tigrina : a light and electron microscopic study. J.
Morphol., 139: 155-184.

9 Baguna, J., Said, E., Collet, J., Alaudell, M. C. and
Ribas, M. (1988) Cellular, molecular and genetic
approaches to regeneration and pattern formation in
planarians. Progress in Zool. (in press).

10 Pascolini, R., Lorvik, S. and Camatini, M. (1988)
Migrating cells during wound healing of Dugesia
lugubris s.l.. Progress in Zool. (in press).

11 Skaer, R. J. (1965) The origin and continuous re-

1162

V. Gremigni

placement of epidermal cells in the planarian
Polycelis tenuis (Iijima). J. Embryol. Exp. Mor-
phol., 13: 129-139.

12 Morita, M. and Best, J. B. (1974) Electron micros¬
copic studies of planarian regeneration. II Changes
in epidermis during regeneration. J. Exp. Zool.,
187: 345-374.

13 Hori, I. (1978) Possible role of rhabdite-forming
cells in cellular succession of the planarian
epidermis. J. Electr. Microsc., 27: 89-102.

14 Pascolini, R., Tei, S., Vagnetti, D and Bondi, C.
(1984) Epidermal cell migration during wound heal¬
ing in Dugesia lugubris. Observations based on
scanning electron microscopy and treatment with
cytochalasin. Cell Tissue Res., 236: 345-349.

15 Pascolini, R., Lorvik, S., Maci, R. and Camatini,
M. (1988) Immunoelectron microscopic localization
of actin in migrating cells during planarian wound
healing. Tissue and Cell, 20: 47-53.

16 Palmberg, I. (1986) Cell migration and differentia¬
tion during wound healing and regeneration in Mic-
rostomum lineare (Turbellaria). Hydrobiologia, 132:
181-188.

17 Pedersen, K. J. (1972) Studies on regeneration blas¬
temas of the planarian Dugesia tigrina with special
reference to differentiation of the muscle-connective
tissue filament system. Whilelm Roux’s Arch. 169:
134-169.

18 Hori, I. (1979) Structure and regeneration of the
planarian basal lamina: an ultrastructural study.
Tissue and Cell, 11: 611-621.

19 Slack, J. M. W. (1980) The source of cells for regen¬
eration. Nature, 286: 760.

20 Morgan, T. H. (1900) Regeneration in planarians.
Arch. Entwicklungsmech. , 10: 58-119.

21 Steinman, P. (1908) Untersuchungen fiber das
Verhalten des Verdauungssystem bei der Regenera¬
tion der Tricladen. Arch. Entwicklungsmech., 25:
523-568.

22 Lang, A. (1912) Uber Regeneration bei Planarien.
Arch. Mikrosk. Anat., 79: 361-426.

23 Hyman, L. H. (1951) The invertebrates. Vol. II.
Platyhelminthes and Rhyncocoela. McGraw-Hill,
New York, pp. 1-551.

24 Flickinger, R. A. (1964) Isotopic evidence for a
local origin of blastema cells in regenerating planar-
ia. Exp Cell Res., 34: 403-406.

25 Hay, E. D. (1968) Dedifferentiation and metaplasia
in vertebrate and invertebrate regeneration. In “The
Stability of the Differentiated State”. Ed by H.
Ursprung, Springer-Verlag, Berlin, pp. 85-108.

26 Coward, S. J. (1969) Regeneration in planarians:
some unresolved questions and problems. J. Biol.
Psychol., 11: 15-19.

27 Rose, C. and Shostak, S. (1968) The transformation
of gastrodermal cells to neoblasts in regenerating

Phagocata gracilis. Exp. Cell Res., 50: 553-561.

28 Betchaku, T. (1970) Isolation of planarian neoblasts
and their behaviour in vitro with some aspects of the
mechanism of the formation of regeneration blaste¬
ma. J. Exp. Zool., 174: 253-267.

29 Teshirogi, W. (1986) On the origin of neoblasts in
freshwater planarians (Turbellaria). Hydrobiologia,
132: 207-216.

30 Gremigni, V. (1974) The origin and cytodifferentia-
tion of germ cells in the planarians. Boll. Zool., 41:
359-377.

31 Hay, E. D. and Coward, S. J. (1975) Fine structure
studies on the Dugesia. I. Nature of the neoblast and
other cell types in noninjured worms. J. Ultrastruct.
Res., 50: 1-21.

32 LeMoigne, A., Sauzin, M. J. and Lender, Th.
(1966) Comparaison de l’ultrastructure du neoblaste
et de la cellule embryonnaire des planaires d’eau
douce. C. R. Acad. Sci., 263: 627-629.

33 Le Moigne, A. (1967) Mise en evidence au micro¬
scope electronique de la persistance des cellules
indifferenciees au cours du developpement
embryonnaire de la Planaire Polycelis nigra. C. R.
Acad. Sci., 265: 242-244.

34 Pedersen, K. J. (1959) Cytological studies on the
planarian neoblast. Z. Zellforsch., 50: 799-817.

35 Morita, M. (1967) Observations on the fine struc¬
ture of the neoblast and its cell division in the
regenerating planaria. Sci. Rep. Tohovu Univ.
Biol., 33: 399-406.

36 Morita, M., Best, J. B. and Noel, S. (1969) Elec¬
tron microscopic studies of planarian regeneration. I
Fine structure of neoblasts in Dugesia dorotocepha -
la. J. Ultrastruct. Res., 27: 7-23.

37 Sauzin, M. J. (1966) Etude au microscope electroni¬
que du neoblaste de la Planaire Dugesia gonocepha -
la (Turbellarie, Triclade) et de ses changements
ultrastructuraux au cours des premiers stades de la
regeneration. C. R. Acad. Sci., 263: 605-608.

38 Sauzin-Monnot, M. J. (1973) Etude ultrastructurale
des neoblastes de Dendrocoelum lacteum au cours
de la regeneration. J. Ultrastruct. Res., 45: 206-
222.

39 Le Moigne, A., Sauzin, M. J., Lender, Th. and De-
lavault, R. (1965) Quelques aspects des ultrastruc¬
tures du blasteme de regeneration et des tissues
voisin chez Dugesia gonocephala (Turbellarie, Tric¬
lade). C. R. Seanc. Soc. Biol., 159: 530-538.

40 Gremigni, V. and Puccinelli, I. (1977) A contribu¬
tion to the problem of the origin of blastema cells in
planarians: a karyological and ultrastructural inves¬
tigation. J. Exp. Zool., 199: 57-72.

41 Gremigni, V. (1981) The problem of cell totipoten-
cy, dedifferentiation and transdifferentiation in Tur¬
bellaria. Hydrobiologia, 32: 171-179.

42 Woodruff, L. and Burnet, A. L. (1965) The origin

43

44

45

46

47

48

49

50

51

52

53

54

Cellular Aspects of Planarian Regeneration

1163

of blastema cells in Dugesia tigrina. Exp. Cell Res.,
38: 295-305.

Benazzi, M. (1957) Cariologia di Dugesia lugubris
(O. Schmidt) (Tricladida Paludicola). Caryologia,
10: 276-303.

Puccinelli, I. and Gremigni, V. (1977) Sulla presen-
za nei blastemi di planarie di cellule provenienti dal
territorio gonadico femminile. Accad. Naz. Lincei,
63: 588-592.

Gremigni, V. Miceli, C. and Puccinelli, I. (1980)
On the role of germ cells in planarian regeneration. I
A karyological investigation. J. Embryol. Exp. Mor-
phol., 55: 53-63.

Gremigni, V. Miceli, C. and Picano, E. (1980) On
the role of germ cells in planarian regeneration. II
Cytophotometric analysis of the nuclear Feulgen-
DNA content in cells of regenerated somatic tissues.
J. Embryol. Exp. Morphol., 55: 65-76.

Gremigni, V. and Miceli, C. (1980) Cytopho¬
tometric evidence for cell ‘transdifferentiation’ in
planarian regeneration. Whilelm Roux’s Arch., 188:
107-113.

Gremigni, V., Nigro, M. and Puccinelli, I. (1982)
Evidence of male germ cell redifferentiation into
female germ cells in planarian regeneration. J.
Embryol. Exp. Morphol., 70: 29-36.

Banchetti, R. and Gremigni, V. (1973) Indirect
evidence for neoblast migration and for gametogo-
nia dedifferentiation in ex-fissiparous specimens of
Dugesia gonocephala s.l.. Accad. Naz. Lincei, 53:
477-485.

Fedecka-Brunner, B. (1967) Etudes sur la regenera¬
tion des organes genitaux chez la planaire Dugesia
lugubris. I Regeneration des testicules apres des¬
truction. Bull. Biol. Fr. Belg., 101: 255-319.

Child, C. M. (1929) The physiological gradients.
Protoplasma, 5: 447-476.

Wolff, E., Lender, Th. and Ziller-Sengel, C. (1964)
Le role de facteurs autoinhibiteur dans la regenera¬
tion des Planaires. Rev. Suisse Zool., 71: 75-98.
Baguna, J. (1981) Planarian neoblasts. Nature, 290:
14-15.

Said, E. and Baguna, J. (1985) Cell movement in

intact and regenerating planarians. Quantitation us¬
ing chromosomal, nuclear and cytoplasmic markers.
J. Embryol. Exp. Morphol., 89: 57-70.

55 Collet, J., Auladell, M. C. and Baguna, J. (1984)
Specific protein markers of neoblasts and differenti¬
ated cells in the planarian Dugesia (G) tigrina. J.
Embryol. Exp. Morphol., 82 (suppl): 246.

56 Morita, M. and Best, J. B. (1984) Electron micros¬
copic studies of planarian regeneration. IV. Cell
division of neoblasts in Dugesia dorotocephala. J.
Exp. Zool., 229: 425-436.

57 Said, E. and Baguna, J. (1984) Regeneration and
pattern formation in planarians. I The pattern of
mitosis in anterior and posterior regeneration in
Dugesia (G) tigrina, and a new proposal of blastema
formation. J. Embryol. Exp. Morphol., 83: 63-80.

58 Lindh, N. O. (1957) The mitotic activity during the
early regeneration in Euplanaria polychroa. Ark.
Zool., 10: 497-509.

59 Baguna, J. (1976) Mitosis in the intact and regener¬
ating planarian Dugesia mediterranea n.sp. II. Mito¬
tic studies during regeneration and a possible
mechanism of blastema formation. J. Exp. Zool.,
195: 65-80.

60 Wolff, E. and Dubois, F. (1947) Sur une methode
d’irradiation localisee permettant de mettre en evi¬
dence la migration des cellules de regeneration chez
les Planaires. C. R. Soc. Biol., 141: 903-906.

61 Wolff, E. and Dubois, F. (1948) Sur la migration
des cellules de regeneration chez le Planaires. Rev
Suisse Zool., 55: 218-227.

62 Dubois, F. (1948) Demonstration de la migration
des cellules de regeneration des planaires par la
methode des greffes et des irradiations combinees.
C. R. Acad. Sci., 226: 1316-1317.

63 Dubois, F. (1949) Contribution a l’etude de la
migration des cellules de regeneration chez planaires
dulcicoles. Bull. Biol., 83: 213-283.

64 Cecere, F., Grasso, M., Urbani, E. and Vannini, E.
(1964) Osservazioni autoradiografiche sulla
rigenerazione di Dugesia lugubris. Rendic. 1st. Sci.
Univ. Camerino, 5: 193-198.

.

ZOOLOGICAL SCIENCE 5: 1165-1185 (1988)

© 1988 Zoological Society of Japai

REVIEW

Two Decades since the Naming of Tubulin
— The Multi-facets of Tubulin —

Hideo Mohri and Natsumi Hosoya

Department of Biology, University of Tokyo,
Komaba, Meguro-ku, Tokyo 153, Japan

INTRODUCTION

It was an afternoon in the fall of 1967 when one
of us (H. M.) visited the late Dr. Jean C. Dan’s
laboratory at Ochanomizu University and con¬
sulted with her about naming a new protein, the
main constituent of the microtubule. After a
couple of hours of discussion, we chose the name
tubulin for the protein. This name first appeared
in publications in 1968 [1,2] and opened the doors
to a wide range of research opportunities.

For a long time, many investigators including
ourselves looked for actin and myosin as “contrac¬
tile proteins” in flagella and cilia in order to
establish a similarity on a molecular basis between
muscle contraction and flagellar or ciliary move¬
ment. Although several reports suggested the
presence of myosin-like protein and actin-like pro¬
tein in sperm flagella [3-6], these pioneer studies
were not sufficient to prove that the proteins in
question were really myosin and actin.

A new approach to this problem was taken by
Gibbons in 1963 when he subjected Tetrahymena
cilia to chemical dissection, a combination of de-
membranation with digitonin and dialysis against a
solution of low ionic strength [7]. Thus, a new
ATPase protein, dynein, was isolated and iden¬
tified as the arms projecting from the outer doublet
microtubules in ciliary and flagellar axonemes [7-
10]. Since then, it has been established that
dynein, not myosin, is the force-generating compo¬
nent at least in cilia and flagella causing sliding

Received August 16, 1988

between adjacent outer doublet microtubules [11-

13]-

Then the question arose as to which component
is the counterpart of dynein in such organelles. It
is evident that the most prominent structure in the
axoneme is 9 + 2 microtubules. Microtubules (the
outer doublets) competent for biochemical analy¬
ses were first made available by Gibbons’ chemical
dissection [7]. In parallel with other groups’ work
[14-17], we characterized the protein constituting
the outer doublet microtubules from sea urchin
sperm flagella [2, 18-20]. In comparison to actin
and flagellin, which were considered to be “con¬
tractile proteins” in eukaryotic and prokaryotic
cells respectively at that time, the protein differed
from them in interactions with myosin, by its
binding nucleotides which were GTP and GDP, in
amino acid composition*, in molecular weight, in
monomeric and polymeric forms, etc. On the basis
of these differences, the main constituent of the
microtubule was given the name tubulin [1].

The name tektin had been proposed for structur¬
al proteins of the mitotic apparatus, cilia and
flagella, membrane, etc. [21], but it covered a wide
variety of proteins including actin and other non¬
tubulin proteins. Recently, this term was assigned
to the protein of certain filaments of flagellar
microtubules which has similar properties to in-

* Gross amino acid compositions of tubulin and actin
were rather similar to each other, although those of
tubulin and flagellin were quite different. Star dia¬
grams illustrating the amino acid composition of pro¬
teins first appeared in our paper [1], but it had originally
been invented by Koscak Maruyama.

1166

H. Mohri and N. Hosoya

termediate filament proteins [22]. Another line of
work revealed the ubiquitous localization of col¬
chicine-binding protein in dividing cells, cilia and
flagella, brain tissue, etc. [16, 17, 23]. The colchi¬
cine-binding protein was identified as tubulin.

Tubulin is a heterodimer of a- and /?-tubulin
[24]. According to sequence data of a- and /?-
tubulin (e.g. [25, 26]), its molecular weight is
about 100,000. Establishment of the conditions for
assembly and disassembly of brain tubulin into and
from the microtubule by Weisenberg in 1972 [27]
brought about a great advance in the purification
and characterization of tubulin and microtubule-
associated proteins (MAPs). Tubulin can be fur¬
ther separated from other components by chro¬
matography on a phosphocellulose column [28].
Complete sequences of a- and /?-tubulin were
determined from both Edman degradation [25, 26]
and from the nucleotide sequence of cDNA [291 in
1981.

It is well recognized that microtubules partici¬
pate in diverse cell functions such as cell motility,
intracellular transport, differentiation and mainte¬
nance of cell structures, etc. Furthermore, differ¬
ent microtubule classes exhibit differences in sta¬
bility under physiological conditions and with re¬
spect to physical and chemical treatments. What is
the cause of such functional and biochemical di¬
versity among microtubule classes?

There are various MAPs and proteins which
interact with microtubules. It is certain that these
proteins modify assembly and disassembly of mi¬
crotubules and contribute to the versatile functions
of microtubules. However, the idea that different
classes of microtubules consist of somewhat differ¬
ent tubulin molecules has long been postulated (cf.
[30]), although tubulin was considered to remain
conservative during evolution. Much evidence is
now accumulating to indicate the heterogeneity of
tubulin and the presence of isoforms of both a- and
/?-tubulins by the finding of multiple genes encod¬
ing different tubulin species and post-translational
modifications of tubulin molecules.

The present review will focus on this problem in
order to give insight into the multiple physiological
functions displayed by the fascinating cell struc¬
ture, the microtubule. Several reviews and books
are available for a more detailed description of

tubulin and microtubules [31-38].

HETEROGENEITY OF TUBULIN

Alpha- and (3-tubulin

Before going further into the heterogeneity of
tubulin, it should be noted that the subunits of the
tubulin molecule, a- and /?-tubulin, are not identi¬
cal. Originally, the slower migrating polypeptide
on an electrophoresed urea-polyacrylamide gel
was designated a-tubulin and the faster one, (3-
tubulin [24]. However, the relative electrophoretic
mobility of both subunits varies depending upon
the conditions for electrophoresis [39]. The rela¬
tionship is reversed under certain conditions and in
certain species even under the same conditions
[39-41], as revealed by the two-dimensional elec¬
trophoresis of O’Farrell [42], i.e., isoelectric focus¬
ing and sodium dodecyl sulfate-polyacrylamide gel
electrophoresis (SDS-PAGE), and by immuno-
blotting with antibodies specific to each subunit.
Alpha- and /?-tubulin can be isolated not only by
preparative gel electrophoresis but also by hydrox-
ylapatite column chromatography [43].

The primary structures of a- and /?-tubulin differ
from each other (Fig. 1); the different genes en¬
coding each subunit as will be described in the
following sections. Generally, a-tubulin possesses
tyrosine at the C-terminus encoded by mRNA and
/?-tubulin is slightly shorter in chain length and
more acidic. Exchangeable GTP binds to /?-
tubulin [44]. Species differences, however, exist in
both subunits among metazoans, protists, slime
mold, yeasts and higher plants [45, 46], although
tubulins from different sources such as mammalian
brain and yeast [47] co-assemble into microtubules
in vitro. Accordingly, they are immunologically
different and polyclonal and monoclonal specific
antibodies have been prepared [48-50] and are
now even commercially available. The two sub¬
units also appear different in reconstructed images
from optical diffraction data of electron micro¬
graphs [51, 52].

Gel electrophoresis

As mentioned above, the electrophoretic mobil¬
ities of both a- and ^-tubulin vary among different

Multi-facets of Tubulin

1167

a MET-ARG-GLU-CYS-ILE-SER-ILE-HIS-VAL-GLY-GLN-ALA-GLY-VAL-GLN
8 MET-ARG-GLU-ILE-VAL-HIS-ILE-GLN-ALA-GLY-GLN-CYS-GLY-ASN-GLN

-ILE-GLY-ASN-ALA-CYS-TRP-GLU-LEU-TYR-CYS-

-ILE-GLY-ALA-LYS-PHE-TRP-GLU-VAL-ILE-SER-

LEU-GLU-HIS-GLY-ILE-GLN-PRO-ASP-GLY-GLN-MET-PRO-SER-ASP-LYS-

ASP-GLU-HIS-GLY-ILE-ASP-PRO-THR-GLY-SER-TYR-^-GLY-ASP-SER-

THR-ILE-GLY-GLY-GLY-ASP-ASP-SER-PHE-ASN-

ASP-LEU-GLN-LEU-GLU-ARG-ILE-g^-VAL-TYR-

a THR-PHE-PHE-SER-GLU-THR-GLY-ALA-GLY-LYS-HIS-VAL-PRO-ARG-ALA-VAL-PHE-VAL-ASP-LEU-GLO-PRO-THR-VAL-ILE-
8 TYR-ASN-GLU-ALA-ggg-g^-^g-LYS-TYR-VAL-PRO-ARG-ALA-ILE-LEU-VAL-ASP-LEO-GLU-PRO-GLY-THR-MET-ASP-SER-

a ASP-GLU-VAL-ARG-THR-GLY-THR-TYR-ARG-GLN-LED-PHE-HIS-PRO-GLU
8 VAL-ARG-SER-GLY-PRO-PHE-GLY-GLN-ILE-PHE-ARG-PRO-ASP-ASN-PHE

-GLN-LEU-ILE-THR-GLY-LYS-GLU-ASP-ALA-ALA-

-VAL-PHE-GLY-GLN-SER-GLY-ALA-GLY-ASN-ASN-

a ASN-ASN-TYR-ALA-ARG-GLY-HIS-TYR-THR-ILE-GLY-LYS-GLU-ILE-ILE-
8 TRP-ALA-LYS-GLY-HIS-TYR-THR-GLU-GLY-ALA-GLU-LEU-VAL-ASP-SER-

125

ASP-LEU-VAL-LEU-ASP-ARG-ILE-ARG-LYS-LEU-

VAL-LEU-ASP-VAL-VAL-ARG-LYS-GLU-SER-GLU-

a ALA-ASP-GLN-CYS-THR-GLY-LEU-GLN-GLY-PHE-SER-VAL-PHE-HIS-SER
8 SER-CYS-ASP-CYS-LEU-GLN-GLY-PHE-GLN-LEU-THR-HIS-SER-LEU-GLY

-PHE-GLY-GLY-GLY-THR-GLY-SER-GLY-PHE-THR-

-GLY-GLY-THR-GLY-SER-GLY-MET-GLY-THR-LEU-

a SER-LEU-LEU-MET-GLU-ARG-LEU-SER-VAL-ASP-TYR-GLY-LYS-LYS-SER
8 LEU-ILE-SER-LYS-ILE-ARG-GLU-GLU-TYR-PRO-ASP-ARG-ILE-MET-ASN

a GLN-VAL-SER-THR-ALA-VAL-VAL-GLO-PRO-TYR-ASN-SER-ILE-LEU-THR-
8 SER-ASP-THR-VAL-VAL-GLU-PRO-TYR-ASN-ALA-THR-LEU-SER-VAL-HIS-

-LYS-LEU-GLU-PHE-SER-ILE-TYR-PRO-ALA-PRO-

-THR-PHE-SER-VAL-VAL-PRO-SER-PRO-LYS-VAL-

200

THR-HIS-THR-THR-LEU-GLU-HIS-SER-ASP-CYS-

GLN-LEU-VAL-GLU-ASN-THR-ASP-GLU-THR-TYR-

a ALA-PHE-MET-VAL-ASP-ASN-GLU-ALA-ILE-TYR-ASP-ILE-CYS-ARG-ARG
8 CYS-ILE-ASP-ASN-GLU-ALA-LEU-TYR-ASP-ILE-CYS-PHE-ARG-THR-LEU

a ASN-LEU-ASN-ARG-LEU-ILE-GLY-GLN-ILE-VAL-SER-SER-ILE-THR-ALA
8 ASN-HIS-LEU-VAL-SER-ALA-THR-MET-SER-GLY-VAL-THR-THR-CYS-LEU

225

-ASN-LEU-ASP-ILE-GLU-ARG-PRO-THR-TYR-THR-

-LYS-LEU-THR-THR-PRO-THR-TYR-GLY-ASP-LEU-

250

-SER-LEU-ARG-PHE-ASP-GLY-ALA-LEU-ASN-VAL-

-ARG-PHE-PRO-GLY-GLN-LEU-ASN-ALA-ASP-LEU-

a ASP-LEU-THR-GLU-PHE-GLN-THR-ASN-LEU-VAL-PRO-TYR-PRO-ARG-ALA

GLY

8 ARG-LYS-LED-ALA-VAL-ASN-MET-VAL-PRO-PHE-PRO-ARG-LEU-HIS-PHE

275

-PRO-VAL-

.HIS_pHE_pDn-T.EU-
ILE PHE PR° LEU ALA ARG PHE ASX

■PHE-MET-PRO-GLY-PHE-ALA-PRO-LEU-THR-

SER

ALA"

300

a ILE-SER-ALA-GLU-LYS-ALA-TYR-HIS-GLU-GLN-LEU-SER-VAL-ALA-GLU
8 ARG-GLY-SER-GLN-GLN-TYR-ARG-ALA-LEU-THR-VAL-PRO-GLU-LEU-THR

-ILE-THR-ASN-ALA-CYS-PHE-GLU-PRO-ALA-ASN-

-GLN-GLN-MET-PHE-ASP-ALA-LYS-ASN-MET-MET-

a GLN-MET-VAL-LYS-CYS-ASP-PRC-ARG-HIS-GLY-LYS-TYR-MET-ALA-CYS-
8 ALA-ALA-CYS-ASP-PRO-ARG-HIS-GLY-ARG-TYR-LEU-THR-VAL-ALA-ALA-

CYS-LEU-LEU-TYR-ARG-GLY-ASP-VAL-VAL-PRO-

VAL-PHE-ARG-GLY-ARG-MET-SER-MET-LYS-GLU-

a LYS-ASP-VAL-ASN-ALA-ALA-ILE-ALA-THR-ILE-LYS-THR-LYS-ARG-g“^-ILE-GLN-PHE-VAL-ASP-TRP-CYS-PRO-THR-GLY-
8 VAL-ASP-GLU-GLN-MET-LEU-ASN-VAL-GLN-ASN-LYS-ASN-SER-SER-TYR-PHE-VAL-GLU-TRP-ILE-PRO-ASN-ASN-VAL-LYS-

a PHE-LYS-VAL-GLY-ILE-ASN-TYR-GLU-PRO-PRO-THR-VAL-VAL-PRO-GLY
8 THR-ALA-VAL-CYS-ASP-ILE-PRO-PRO-ARG-GLY-LEU-LYS-MET-SER-ALA

375

-GLY-ASP-LEU-ALA-LYS-VAL-GLN-ARG-ALA-VAL-

-THR-PHE-ILE-GLY-ASN-SER-THR-ALA-ILE-GLN-

a CYS-MET-LEU-SER-ASN-THR-THR-ALA-ILE-ALA-GLU-ALA-TRP-ALA-ARG-
8 GLU -LEU- PHE- LYS- ARG- ILE- SER-GLU-GLN- PHE- THR- ALA-MET- PHE -ARG -

400

LEU-ASP-HIS-LYS-PHE-ASP-LEU-MET-TYR-ALA-

ARG-LYS-ALA-PHE-LEU-HIS-TRP-TYR-THR-GLY-

a LYS-ARG-ALA-PHE-VAL-HIS-TRP-TYR-VAL-GLY-GLU-GLY-MET-GLU-GLU
8 GLU-GLY-MET-ASP-GLU-MET-GLU-PHE-THR-GLU-ALA-GLU-SER-ASN-MET

425

-GLY-GLU-PHE-SER-GLU-ALA-ARG-GLU-ASP-MET-

ASN-ASP-LEU-VAL-SER-GLU-TYR-GLN-GLN-TYR-

a ALA-ALA-LEU-GLU-LYS-ASP-TYR-GLU-GLU-VAL-GLY-VAL-ASP-SER-VAL
8 GLN-ASP-ALA-THR-ALA-ASP-GLU-GLN-GLY-GLU-PHE-GLU-GLU-GLU-GLY

Fig. 1. Comparison of the amino acid sequences of porcine brain
[25] and [26]. Both tubulins show more than one residue at
genes.

-GLU-GLY-GLU-GLY-GLU-GLU-GLU-GLY-GLU-GLU-TYR

-GLU-GLU-ASP-GLU-ALA

a- and /9-tubulins. Data are taken from references
six positions, suggesting the existence of multiple

1168

H. Mohri and N. Hosoya

species. In the early 1970s, resolution of tubulin
into several (not two) distinct bands by isoelectric
focusing was reported in the mammalian brain and
Chlamydomonas flagella [53, 54] and raised the
possibility of the presence of tubulin heterogeneity
within a single species or cell. Chlamydomonas
flagellar tubulin also yielded several bands on
SDS-PAGE [55]. Resolution of a-tubulin into two
distinct bands in the SDS-urea gel system was
observed with mitotic apparatus and cilia of sea
urchin embryos, but not with sea urchin sperm
flagella [56]. Furthermore, hydroxylapatite col¬
umn chromatography of brain tubulin yielded
three peaks referred to as a2 and /9 [43]. We
separated flagellar tubulin of starfish spermatozoa
into an a- and a /9-peak on a hydroxylapatite
column, and found that each of these is further
split into several bands by isoelectric focusing [57].
Amino acid compositions and peptide mappings of
the a- and the /9-tubulins of sea urchin cytoplasmic,
ciliary and flagellar tubulins could be distingiushed
from each other [58]. Development of two-
dimensional electrophoresis by O’Farrell [42] and
one-dimensional peptide mapping after limited
proteolysis by Cleveland et al. [59] further facili¬
tated the detection of heterogeneity among tubu¬
lins. Thus, at least three species of a-tubulin and
two species of /9-tubulin, one of each residing in
the partition wall between the A- and B-tubule of
the outer doublet, were distinguished in starfish
sperm flagella [60].

A similar separation of both a- and /9-tubulin
into several isoforms by isoelectric focusing or two
dimensional gel electrophoresis has been reported
in the vertebrate brain [53, 61-69], with improved
methods yielding 6-7 a-tubulin and 12-14 /9-
tubulin isoforms. This brain tubulin heterogeneity
increases during development and aging, especial¬
ly regarding the number of /9-tubulins [63-65, 67].
The high level of heterogeneity is peculiar to brain
tubulin, since tubulins from other sources such as
the glial cells are less heterogenous (cf. [63, 68]),
and it does not seem to be due to artificial mod¬
ifications such as hydrolysis by proteases during
experiments. Since only several /9-tubulin genes
may exist in the chicken as compared with the 12-
14 /9-tubulin isoforms observed, both the express¬
ion of multiple tubulin genes and post-translational

modification of the tubulin polypeptides would
account for the heterogeneity (cf. [38]). As men¬
tioned above, tubulins from different sources are
not identical as revealed by gel electrophoresis.
For instance, the isoelectric points of /9-tubulins
from chick brain and erythrocytes were pH 5.1 and
5.4, respectively [70]. Even among brain tubulins,
the subspecies found in cold-adapted Antarctic
fishes differed significantly from those of temerate
fishes and mammals, the former being more basic
than the latter [71]. A change in the composition
of a-tubulin was revealed by two-dimensional gel
electrophoresis and peptide mapping during axon¬
al transport [72].

A combination of genetic analysis and gel elec¬
trophoresis revealed that a testis-specific /9-tubulin,
which has multiple functions during spermatogene¬
sis, is present in Drosophila melanogaster and a
mutation in the structural gene for this /9-tubulin
results in the production of non-motile spermato¬
zoa [73, 74], Similar experiments with Chinese
hamster ovary (CHO) cells, Aspergillus nidulans,
and others, indicated that mutants resistant to
anti-microtubule drugs have altered /9-tubulins (cf.

[75] ).

Antigene city

Although an antibody against tubulin from a
certain source widely cross-reacts with microtubu¬
lar structures in diverse plant and animal cells as
revealed by immunofluorescence microscopy (cf.

[76] ), some immunological differences were
observed among tubulins from different sources in
earlier studies [77, 78]. The antiserum to a-tubulin
from porcine brain did not cross-react with
Chlamydomonas tubulin, although the antiserum
to /9-tubulin from Chlamydomonas reacted with
brain tubulin [79]. Quantitative immunological
differences exist among brain tubulins from differ¬
ent species [80].

The introduction of the monoclonal antibody
technique has provided a useful means for reveal¬
ing the heterogeneity of tubulin. Thus, monoclon¬
al antibodies raised against brain tubulin labeled
different bands of /9-tubulin separated on isoelec¬
tric focusing gels into different intensities [49].
Using the monoclonal antibodies to a-tubulin
which differ their epitope specificities, it was de-

Multi-facets of Tubulin

1169

monstrated that tubulins from different sources
and even from different tissues of the same spe¬
cies, such as egg and sperm, were not identical
[81]. One of these antibodies bound only a sub¬
class of cytoplasmic microtubules in fibroblasts
which remained after the treatment of the cells
with colcemid [82]. Monoclonal or polyclonal
antibodies specific for the tyrosinated form and the
acetylated form of a-tubulin were obtained [83—
86], and have been used for localizing the antigens
and characterizing different classes of microtu¬
bules as will be described below.

Amino acid sequence

More direct evidence for the heterogeneity of
tubulin can be obtained from sequence data de¬
rived from purified proteins and cDNA clones. An
early work by Luduena and Woodward [87]
showed that /9-tubulins from chick brain and sea
urchin spermatozoa differ at only one position of
25 residues in the N-terminal region. The result
interpreted as evidence indicating the conserva¬
tiveness of the tubulin molecule during evolution
and led to the idea that tubulin is rather common
among different tissues and organisms in spite of
the multiple functions of microtubules. In their
first complete analysis of the amino acid sequence
of brain tubulin, Ponstingl’s group [25, 26] found
that both a- and /9-tubulin carry more than one
amino acid residue at least at six positions (see Fig.
1), and indicated the presence of at least four
different a-tubulins and two slightly differing /9-
tubulins in porcine brain. On the other hand,
using tubulin mRNAs from embryonic chick brain,
Cleveland and his colleagues [29, 88] succeeded in
the construction and identification of cloned
cDNA sequences for a- and /9-tubulin. They
confirmed not only the conservation of tubulin
sequences among metazoans (this was slightly
modified later), but also the existence of multiple
tubulin genes in one organism.

The list of cloned tubulin genes is now expand¬
ing from vertebrates including the human to in¬
vertebrates, protists, slime mold, fungi, yeasts and
higher plants [38, 89-104]. Some of them are
illustrated in Figures 2 and 3. In general, predicted
amino acid sequences are more homologous
among the organisms evolutionally more intimate

with each other, even though there are ‘silent’
changes in nucleotide sequence that do not affect
the amino acid sequence, and much higher diver¬
gences at the non-coding sequences of tubulin
genes. Based on the largest difference observed
when comparing any given tubulin to tubulins of
yeasts lacking cilia, it has been suggested that
structural constraints caused by interactions with
other components of the ciliary axonemes would
affect the mutation rate of tubulin [47],

Recently, predicted amino acid sequences en¬
coded by a family of seven /9-tubulin genes have
been characterized in the chicken [105]. In com¬
parison with mammalian /9-tubulins, especially in
the C-terminal variable region, these /9-tubulins
are classified into six isotypes [105-107]. Among
them, chick /97 (c/97) belongs to class I isotype
(EEEEDFGEEAEEA) expressed in all tissues.
Both c/91 and c/92 correspond to class II isotype
(DEQGEFEE^EDEA) detected predominantly
in brain, whereas c/94 belongs to class III isotype
(EEEGEMYEDDEEESE-QG-K) found as a
minor neuronal component. Class IV isotype
(EEEGEFEEEAEEE-) which is expressed in
brain or testis includes c/93, and c/95 is classified as
class V isotype (NDGEE AFEDDEEEINE) ,
which is ubiquitous except in the brain. The sixth
isotype (c/96) represents erythroid-specific tubulin
and shows considerable divergence in its overall
sequence. Polyclonal antibodies raised against
synthetic polypeptides corresponding to these C-
terminal sequences revealed that each class of
microtubules in vivo are random copolymers of
available isotypes, although in chicken erythro^
cytes, c/96-rich microtubules are more cold-stable
than c/93-rich microtubules [107, 108]. The C-
terminal region of both a- and /9-tubulin is variable
and binds to, for instance, MAP-2 [109], thus
regulating tubulin assembly.

As described above, a testis-specific /9-tubulin
(/92) was found in Drosophila [73, 74]. Among
four /9-tubulin genes with distinct developmental
expression, complete sequences for Drosophila /91
and /92 tubulins have been determined [110]. Two
(al and «3) of four Drosophila a-tubulin genes are
constitutively expressed, whereas a2 is testis-
specific and a4 is specific to early embryo and
ovary. The last one encodes the C-terminal phen-

1170

H. Mohri and N. Hosoya

Human MRECISIHVGQAGVQIGNACWELYCLEHGIQPDGQMPSDKTIGGGDDSFNTFFSETGAGKHVPRAVFVDLEPTVIDEVRTGTYRQLFHPE

Rat - -

Drosophila - V - V -

Phvsarum ---V - 1 - T-Vr - SV-Y---A - L - -- -

Yeast ---i--- - t - N-Y-NP-TASQNS-GG-S - Q--Y---SI - N---Q---DP--D -

. . . . . . . . ISO

Human QLITGKEDAANNYARGHYTIGKEIIDLVLDRIRKLADQCTGLQGFLVFHSFGGGTGSGFTSLLMERLSVDYGKKSKLEFSIYPAPQVSTA

Rat -

Drosophila - V - 1 - A -

Phvsarum - V--C---V - S - V - G---L - D-CV--S -

Yeast - S - V - LV-E-T-K--RIG-N-S - GA--L---AME-T - Q--V - S

Human VVEPYNSILTTHTTLEHSDCAFMVDNEAIYDICRRNLDIERPTYTNLNRLISQIVSSITASLRFDGALNVDLTEFQTNLVPYPRIHFPLA

Rat - G - - - - -

Drosophila - G - V

Phvsarum - V-S--SL---T-V---L - L-KKA - D - VA-VI--L-T - 1 - M-C

Yeast - V - A--DLA--T - SC - S-E - A-V - E-S - A - V

. . . . . . . . 360

Human - TYAPVISAEKAYHEQLSVAEITNACFEPANQMVKCDPRHGKYMACCLLYRGDVVPKDVNAAIATIKTKRTIQFVDWCPTGFKVGINYQPP

Rat -

Drosophila - M - - -

Phvsarum S - T - SV - SM-A - M - SV-V - C -

Yeast - IV--A--F--SN--Q - Q - Y - A-R---T - I-R--Q--VT---A - I--CDR--

Human TVVPGGDLAKVQRAVCMLSNTTAIAEAWARLDHKFDLMYAKRAFVHWYVGEGMEEGEFSEAREDMAALEKDYEEVGVDSVEGEGEEEGEEY

Rat - - - - - - -

Drosophila - - - L - M--GD - GAE-Y

Phvsarum - VFS-I - - L - AE-S-AG-D - Y

Yeast QHIE-SEI - D - S - S - S - L - R - Q--M-VDYM-AD -

Fig. 2. Comparison of a-tubulin sequences of various organisms. Data shown are from the following sources: human
[92], rat [90], Drosophila al [111], Physarum myxamoeba, unidentified residues being tentatively replaced by
those of Physarum plasmodium [98], and yeast, Shizosaccharomyces pombe al [102].

ylalanine instead of tyrosine and differs from al by
149 residues [111]. Testis-specific tubulins have
also been found in mammalian and chicken testes.
Chicken testicular a-tubulin ends with the C-
terminal serine [112], whereas a mouse testicular
a-tubulin has an extended C-terminal region which
does not terminate with tyrosine [113]. Two
Aspergillus ^-tubulins encoded by ben A and tub C
are also unusually divergent in amino acid se¬
quence, the two tubulins differing by 78 residues
[99]. In Physarum , plasmodial a-tubulin has the
C-terminal methionine as compared with amoebal
a-tubulin ending with tyrosine [114], There are
many other cases revealing the heterogeneity of
predicted amino acid sequences among a- and
/?-tubulin genes in a single organism.

MULTIPLE TUBULIN GENES

Tubulin gene families

As described in the preceding chapter, the exist¬
ence of multiple genes encoding different isoforms

of tubulins has been established over the past ten
years. Since there are several excellent reviews
[38, 89, 115, 116], this topic is dealt with only
briefly in this article.

Before the successful construction of cDNA
clones, - in vitro translation of rat brain tubulin
mRNA indicated that there are different mRNA
species for distinct tubulin isoforms [117, 118]. In
Drosophila , the testis-specific /?-tubulin (f!l) was
found in addition to the ubiquitous /?1 tubulin, and
its structural gene was localized on the third
chromosome by genetic analysis [73, 74]. The
earliest determination of cloned sequences com¬
plimentary to chick brain a- and /?-tubulin mRNAs
was made by Cleveland et al. [88]. Since then,
cDNA clones have been obtained from various
sources either by using enriched mRNA fractions
or by screening cDNA libraries and their se¬
quences have been identified.

In mammals, there are 10-20 genes each for a-
and /?-tubulins (cf. Table 1 of [38]), although many
of them have been shown to be pseudogenes which
have multiple deletions and/or termination codons

Multi-facets of Tubulin

1171

Human

Chicken

Drosophila

Chlamvdomonas

Yeast

MREIVHIQAGQCGNQIGAKFWEVISDEHGiDPTGTYHGDSDLQLDRISVYYNEATGGKYVPRAILVDLEPGTMDSVRSGPFGQIFRPDNF

- S -

- E--N .

- N -

- ---- - -

- G -

- v__

- E - F- -

-----R -

- M -

. . Y- . .

- !--S - Y - A - T-CG - L-FN - HD-I-KE-LN--F---SS--T---S-N - W-I-A--NSAI-NL . Y

. . . . . . . - . 180
Human VFGQSGAGNNWAKGHYTEGAELVDSVLDVVRKEAESCDCLQGFQLTHSLGGGTGSGMGTLLISKIREEYPDRIMNTFSVVPSPKVSDTVV

Chicken - S - M -

Drosophila - S-G -

Chlamvdomonas - T - 1 - VC - M-L -

Yeast I - S - V - M--I-R---G--S - 1 - F---K--L---M-A - L - T -

Human EPYNATLSVHQLVENTDETYCIDNEALYDICFRTLRLTTPTYGDLNHLVSGTMECVTTCLRFPGQLNADLRKLAVNMVPFPRLHFFMPGF

Chicken - - - K - A--SG -

Drosophila - K - A--SG -

Chlamvdomonas - A--CMVL - K - F - I-AV-SGI-C - LI - V—

Yeast - HS - F - Q---K-NQ-S - N---SV-SG---S--Y - S - L - V-Y

Human APLTSRGSQQYRALTVPDLTQQVFDAKNMMAACDPRHGRYLTVAAVFRGRMSMKEVDEQMLNVQNKNSSYFVEWIPNNVKTAVCDIPPRG

Chicken - E - M--S - 1 -

Drosophila - E - M - 1 - F - C -

Chlamvdomonas T - E - MW - C-A - AS-L - T - SS - K-

Yeast - AI - SF-S - E - M-E - A---N - F---KV-V---EDE-HK--S---D - Q - SVA-Q-

Human

Chicken

Drosophila

Chlamvdomonas

Yeast

LKMAVTFIGNSTAIQELFKRISEQFTAMFRRKAFLHWYTGEGMDEMEFTEAESNMNDLVSEYQQYQDATAEEEEDFGEEAEEEA

- SA - D-QGE-E--G--DEA

- SA - V - E - D--GE-D-DE--GGGDE

- SA - M - V - S - GE-EG-E- -A

-D--A - A - S - VGD--S - K - S - L--S - E--V-DD-EVD-NGDFG-PQNQDEPITENFE

Fig. 3. Comparison of /7-tubulin sequences of various organisms. Data shown are from the following sources: human
D-/71 [91], chicken /?2 [29], Drosophila pi [93], Chlamydomonas [95] and yeast, Saccharomyces cerevisiae [101].

within the exon sequences [119]. In the mouse, for
instance, six a-tubulin genes and five /7-tubulin
genes are functional ones [120, 121]. As already
mentioned, seven /7-tubulin genes give rise to six
isotypes [105] and 4-5 o-tubulin genes [88] are
expressed in the chicken. Both a- and /7-tubulin
genes constitute dispersed multigene families,
since they are not closely linked to each other and
are located on different chromosomes [122]. The
same is true in Drosophila which expresses all of
four genes for both a- and /?-tubulins [123, 124], In
contrast, some genes within the same family which
consists of more than 10 genes are clustered in the
sea urchin [94], and the majority of multiple genes
in Trypanosoma are arranged in a tightly packed
cluster of alternating a- and /?-tubulin genes [125],
In Leishmania , both a- and ^-tubulin genes are
found in separate tandem repeat clusters, the latter
also existing as dispersed genes [126, 127].
Another protozoan parasite, Crithidia , however,
exhibits a dispersed arrangement of multiple tubu¬
lin genes [128]. Physarum , Aspergillus and
Chlamydomonas possess 2-4 different DNA sequ¬

ences for a- and /7- tubulin, respectively, which are
also dispersed (cf. [38]). It appears that at least
two a-genes are functional among eight a-tubulin-
like sequences in Naegleria [129]. On the other
hand, the presence of a single gene was reported
for Tetrahymena a-tubulin ([130], listed as /? in
Table 1 of [38]), Candida /7-tubulin [100] and yeast
/?-tubulin [101], although two a-tubulin genes,
essential and dispensable, have been found in
yeast [102, 131]. The genome of higher plants also
contains a /7-tubulin gene family consisting of 5-6
distinct genes and/or pseudogenes [41, 104].

In some cases, different genes code for the same
tubulin. For instance, two Chlamydomonas /?-
tubulin genes are somewhat different, yet encode
an identical protein [95]. Two mammalian genes
encode the class IV /?-tubulin isotype [116]. Further¬
more, the class II /?-tubulin isotype is expressed by
c/?l and c/72 through different pathway in the
chicken [132].

Differential gene expression

The finding of multiple tubulin gene families

1172

H. Mohri and N. Hosoya

raises the question as to how these genes are
expressed, coordinately or selectively, during cell
differentiation, embryonic development or life cy¬
cle. As mentioned earlier, tubulin heterogeneity
appears most extensively in mammalian brain and
increases during development and aging. It is most
likely that at least some of the heterogeneity is due
to the expression of different genes and the rest to
post-translational modification. In the mouse, mal
a-tubulin mRNA decreases during embryogenesis
and postnatal development of the brain, whereas
ma2 mRNA remains relatively constant [133].
Among /?-tubulin isotypes, neurospecific m/22 and
constitutive m/?5 /?-tubulin mRNAs are expressed
very early in development. On the contrary,
another neurospecific m/?4 mRNA is expressed
later in the developing brain [134]. In the rat, Tal
a-tubulin mRNA is abundant in regions with
neurons undergoing neurite extension, whereas
expression of T26 a-tubulin mRNA is relatively
constitutive [135]. As described in the preceding
chapter, chick c/71 and c/?2 genes are expressed
predominantly in the brain during development,
other /7-tubulin isotypes being constitutive or spe¬
cific to other tissues or cells [105-107, 136]. Simi¬
larly, the expression of human h5/7 is significant in
cells of neural origin, but h/?2 is expressed in
numerous human cell lines [137].

The existence of testis-specific tubulin (/72) was
first demonstrated in Drosophila [73, 74]. Muta¬
tion in the structural gene for pi causes various
defects not only in the sperm axoneme but also in
the meiotic spindle and in the cytoplasmic microtu¬
bules, suggesting multiple functions for this spe¬
cific /?-tubulin [115, 138]. On the other hand, the
expression of /71 is ubiquitous during early
embryogenesis and predominant in neural deriva¬
tives during later stages, whereas p3 is expressed at
mid-embryogenesis with /73 mRNA being re¬
stricted to the mesoderm [139, 140]. Among four
Drosophila a-tubulin genes, a2 is male-specific in
adults and a\ is maternal and found in nurse cells,
eggs and early embryos, while the other two genes
are constitutively expressed [111, 115]. Testis-
specific o-tubulins also exist in the mouse [113,
120] and in the chicken [112]. In the former, an
a-tubulin gene appears to be expressed only in
post-meiotic cells [141]. The antiserum against

novel synthetic peptides corresponding to the uni¬
que C-terminal sequence of a mouse testicular
a-tubulin specifically binds to the manchettes and
meiotic spindles of the mouse testis, but not with
neuronal microtubules [113]. In the chicken, c/73
/?-tubulin is testis-specific and c/?6 is erythroid
specific, assembling into marginal band microtu¬
bules in nucleated erythroid cells [136, 142].

Cilia and flagella are organelles specializing in
cell motility. Since they are formed at restricted
stages during embryogenesis or the life cycle, and
also regenerate after removal or resorption into
the cell body, these organelles provide us with
good material for elucidating the mechanisms of
specific induction and regulated expression of spe¬
cific genes. Transit elevation in translatable
mRNAs for tubulin and other ciliary and flagellar
components was observed within short periods
after amputation of cilia and flagella in sea urchin
embryos [143], in Tetrahymena [144] and in
Chlamydomonas [145-147]. A similar phe¬
nomenon was observed after the initiation of dif¬
ferentiation from amoebae to flagellates in Naeg-
leria [148]. Specific tubulin synthesis also occurs in
the differentiation of a parasitic protozoan Leish-
mania from nonmotile amastigotes to motile pro-
mastigotes [149]. Deflagellation induces the coor¬
dinately regulated transcription of the two a- and
two /?-tubulin genes of Chlamydomonas [95, 150,
151], and the stabilization of tubulin transcripts
[152]. Coordinate regulation of mRNAs for a- and
/?-tubulin as well as calmodulin was observed in the
differentiation of Naegleria [129]. In sea urchin
embryos, the accumulation of three /?-tubulin
mRNAs (/? 1, p2 and /73) from a multiple gene
family are coordinated with ciliogenesis, whereas
another /7-tubulin mRNA, /74, is not induced [153].
The above three /7-tubulin mRNAs are later en¬
riched in the pluteus ectoderm. In contrast, /74
mRNA is enriched in the endomesoderm after the
gastrula stage. Post-translational modification of
tubulin for cilia and flagella will be described in the
following chapter.

Aspergillus nidulans has two /?-tubulin genes,
ben A and tub C. The former is involved in both
vegetative growth and asexual sporulation and the
latter is involved mainly in asexual sporulation
[99]. A heat-sensitive mutation at the structural

Multi-facets of Tubulin

1173

gene of ben A blocks both nuclear division and
nuclear migration at a restrictive temperature
[154]. The tub A a-tubulin gene must also be
concerned with both processes. Such temperature-
sensitive mutations in a- and /9-tubulin also affect
spindle formation in CHO cells [155]. In slime
mold, Physarum, an a-tubulin gene is expressed in
the plasmodial phase, whereas another a-tubulin
gene is transcribed in the amoebal phase of the life
cycle [98, 114]. In a similar manner, one of the two
/91 isotypes appears in the myxamoeba, the /92
isotype being found only in the plasmodium.
Another /91 isotype is expressed both in the myx¬
amoeba and in the plasmodium [41]. Among
higher plants, tissue-specific and developmental
stage-specific accumulation of /9-tubulin isotypes
was reported in the carrot. In the floret except for
the pollen, /91, /92 and /93 isotypes are present,
whereas the /94 isotype is detected in the stamen
and pollen. The /95 isotype is differentially ex¬
pressed in the stem and leaves, and the /96 isotype
is found only in seedlings [41]. Such a specific
expression of both a- and /9-tubulin genes has also
been reported in Arabidopsis thaliana [104].

Concerning the regulation of tubulin gene ex¬
pression, Ben-Ze’ev et al. [156] reported that
tubulin synthesis in mouse 3T6 cells is depressed
by colchicine which elevates the level of unassem¬
bled tubulin subunits in the cell. Since then, such a
feedback control for the synthesis of both a- and
/9-tubulin has been ovserved in various vertebrate
cells by using antitubulin drugs or by direct mi¬
croinjection of purified tubulin [38, 157-160].
Change in the rate of tubulin synthesis is closely
associated with change in the amount of tubulin
mRNAs. Autoregulatory control of tubulin gene
expression appears to be the result of a cytoplas¬
mic event which modulates stability of the mRNAs
bound to polysome, and not due to changes in
transcription rates. Recently it has been shown
that changes in the stability of polysome-bound
/9-tubulin mRNAs are specified by the first 13
translated nucleotides which encode the first four
/9-tubulin amino acids, Met-Arg-Glu-Ile [161].
Either the binding of tubulin subunits to the N-
terminal coding sequences of a translating J3-
tubulin RNA or the interaction of free subunits
with the nascent tubulin polypeptides emerging

from the ribosome would stimulate tubulin mRNA
degradation. Similar autoregulation has been
documented in Chlamydomonas [152, 162, 163]
and sea urchin embryos [153, 164], although en¬
hanced transcription of tubulin genes after decilia-
tion or deflagellation occurs independently of the
amount of unassembled tubulin molecules [146,
164-166].

As mentioned already, there are multiple func¬
tional genes for both a- and /9-tubulin isotypes and
each isotypic class is expressed in a unique manner
during development. However, in studies where
antibodies were raised against synthetic poly¬
peptides corresponding to the C-terminal variable
domain sequences of different vertebrate /9-tubulin
isotypes, it was revealed that each type is present
in vivo an apparently uniform distribution in all
classes of microtubules in cultured cells, with a
similar level of each isotype present being in the
assembled form [107]. In contrast, microtubules
enriched in the c/93 isotype were found to be less
cold-stable than those enriched in c/96, which is
preferentially expressed in erythroid cells and
thrombocytes [108]. In the nematode, Caeno-
rhabditis elegans , most cells have microtubules
with 11 protofilaments, but the touch receptor
neurons have microtubules with 15 protofilaments
as compared with the 13 protofilaments in usual
microtubules [167]. The 15 protofilament microtu¬
bules are lost by colchicine-treatment or by gene
mutation concomitant with the loss of touch sensi¬
tivity. Diversity in functions of microtubules,
therefore, would be manifested by interactions of
tubulin with other specific microtubule proteins.
Recently, the microinjection of biotin-labeled
tubulin before fixation and the staining with anti¬
tubulin and a fluorescent secondary antibody were
combined in order to provide a quantitative analy¬
sis of dynamics at the single microtubule level
[168]. Such a technique would be also useful for
analyzing functions and dynamic states of specific
tubulins. The problem of spatial organization of
microtubules for morphogenesis was reviewed by
Kirschner and Mitchison [169].

1174

H. Mohri and N. Hosoya

POST-TRANSLATIONAL MODIFICATION
OF TUBULIN

Phosphorylation

Results showing the phosphorylation of brain
tubulin by a cAMP-dependent protein kinase
firmly associated with tubulin preparation was first
described by Goodman et al. [170], although sub¬
sequent work indicated that the preferred sub¬
strates of protein kinase are MAPs which co-purify
with tubulin (cf. [171]). Tetrahymena axonemes
also contain such protein kinase(s), which pre¬
ferentially phosphorylate tubulin among the ax-
onemal proteins [172]. It was reported that only
/?-tubulin was specifically phosphorylated in vivo at
a serine residue near the C-terminus in the
mammalian brain [173, 174]. Only certain species
of /?-tubulin separated by isoelectric focusing were
phosphorylated [175]. In differentiated neuroblas¬
toma cells, the phosphorylated isotype was iden¬
tified as pi (class III isotype) [176]. However,
there have been reports revealing the phosphoryla¬
tion of brain a-tubulin as well as /?-tubulin [177—
179], On the contrary, in the axonemes of
Chlamydomonas grown in the presence of 32Pi, the
phosphorylation of only a few components of
a-tubulin was observed [180]. Both a- and /?-
tubulin were found to be phosphorylated in the
platelets [181, 182]. In the uterine smooth muscle
of a parturient rat, it was observed that phosphoryl¬
ation of /?-tubulin switched to ^-tubulin during
labor [183].

Several kinds of protein kinases distinct from
cAMP dependent protein kinase engage in the
phosphorylation of tubulin. Thus, Ca2+-
calmodulin-dependent protein kinase (kinase II)
which phosphorylates both a- and /?-tubulin at
their C-terminal region was obtained from brain
tissues [178, 179, 184]. The phosphorylation
occurred mainly at serine residues, although
threonine residues were also phosphorylated. The
phosphorylation of tubulin causes reduction in its
ability to polymerize, and to bind to MAP-2, and
an enhancement in disassembly of microtubules
[179, 184]. The phosphorylation also enhances the
association of tubulin with phospholipid vesicles
[185]. Such a system would participate in the

release of neurotransmitters from the synapto-
somes [178, 184], Casein kinase II, which appears
to be present in brain, phosphorylates a serine
residue at the C-terminal region of /?-tubulin [186].
Tubulin phosphorylated by this enzyme retains its
ability to polymerize. Casein kinase I was reported
to phosphorylate multiple sites of tubulin [187].
The presence of tyrosine kinases which phosphory¬
late tubulin, including the src oncogene product,
has also been reported [188-190]. Among them,
insulin receptor kinase phosphorylates the C-
terminal tyrosine of a-tubulin resulting in the dis¬
appearance of the ability of tubulin dimer to
assemble into a microtubule. On the other hand, it
phosphorylates tyrosine residues located in other
parts of the molecule when a-tubulin is not tyrosin-
ated at the C-terminus, and the phosphorylated
tubulin retains its assembling capacity [188]. An
endogenous protein kinase in the growth cone of a
neonatal rat brain phosphorylates tyrosine and
serine residues of both a- and /?-tubulin, the result¬
ing phosphotyrosine exceeding phosphoserine
[189].

As can be seen from the statements above, the
phosphorylation of tubulin in situ would affect the
assembly and disassembly of microtubules and
their multiple functions. The phosphorylation of
/?-tubulin increases with microtubule assembly dur¬
ing neurite growth in neuroblastoma cells [191]. It
is possible that the phosphorylated /?-tubulin iso¬
type in neuroblastoma cells has its assembly and/or
function regulated by phosphorylation, whereas
the other isotypes are regulated by MAPs [176]. In
the platelets, stimulation by agonists such as
thrombin results in a reduction of phosphate incor¬
poration into tubulin [182]. It has been suggested
that phosphorylation of a- and /?-tubulin at their
C-terminal region by different kinases would reg¬
ulate tubulin assembly and its interactions with
MAPs and other cellular components [186, 188].

Apart from the phosphorylation of tubulin itself,
phosphorylation of MAPs also affects their in¬
teractions with microtubules (for review, see
[192]). Thus, phosphorylation of MAP-2 by
cAMP-dependent protein kinase inhibits microtu¬
bule assembly [193]. Extensive phosphorylation of
MAP-2 reduces its binding to formed microtubules
[194], A calmodulin-dependent protein kinase

Multi-facets of Tubulin

1175

appears to be more active than the cAMP-
dependent kinase in the phosphorylation of MAP-
2 [195], MAP-1, tau and other MAPs can also be
phosphorylated, resulting in the modulation of
their association with microtubules (cf. [192]).
Ca2+-calmodulin-dependent phosphorylation of a
62K protein induces disassembly of microtubules
in sea urchin mitotic apparatus, probably at ana¬
phase [196]. cAMP-dependent phosphorylation of
certain protein(s) has been reported to be re¬
sponsible for the initiation of sperm flagellar motil¬
ity [197,198]. Bi-directional pigment granule
movements along microtubule arrays in mela-
nophores are regulated by phosphorylation and
dephosphorylation of a 57K protein [199].

Tyrosination

Another post-translational modification is tyro¬
sination and detyrosination of a-tubulin. Incor¬
poration of tyrosine or phenylalanine into the
C-terminus of a brain protein was first described by
Barra et al. [200]. The protein was subsequently
identified as a-tubulin [201, 202]. An enzyme
which adds a tyrosine residue to a C-terminal
glutamate of a-tubulin, tubulin-tyrosine ligase
(TTLase), has been found in extracts of all verte¬
brate tissues and cultured cells [203-206] and has
been highly purified from the brain [207]. TTLase
is also present in sea urchins, but not in Tetrahyme-
na [208]. The proportion of a-tubulin which can be
tyrosinated by the ligase increases after treatment
of a-tubulin with carboxypeptidase A. On the
other hand, a carboxypeptidase which releases
tyrosine from the C-terminus of a-tubulin, tyrosyl-
tubulin carboxypeptidase (TTCPase), has been
obtained from brain tissue [209-211]. It was first
postulated that the primary post-translational
modification of a-tubulin is tyrosination. Howev¬
er, subsequent identification of the codon for the
C-terminal tyrosine in a-tubulin mRNAs indicated
just the reverse situation, i.e., the primary mod¬
ification of a-tubulin is detyrosination by TTLCase
leaving glutamic acid as the C-terminal residue and
TTLase restores the original state of a-tubulin (cf.
[38]). Unassembled tubulin dimer is the substrate
for TTLase, although both tyrosinated and detyro-
sinated dimers can assemble in vitro [203, 204, 212,
213], whereas TTLCase removes tyrosine pre¬

dominantly from tubulin dimers of assembled mic¬
rotubules [210]. TTLase has a binding site on
/?-tubulin [213].

The proportion of tyrosinated and non-
tyrosinated tubulins and the level of the enzymes
participating in the modification vary among differ¬
ent materials. In the mammalian brain, non-
tyrosinated tubulin occupies 25-50% of the total
soluble tubulin while tyrosinated tubulin accounts
for a lesser amount [204, 208, 214-216]. Further¬
more, some percentage of tubulin remains non-
tyrosinable even after treatment with carboxypep¬
tidase A. The fraction of tyrosinated tubulin
changes with the stage of development and aging
of the animal. Even among nerve tissues, cyto¬
plasmic tubulin from the squid nerve is fully tyro¬
sinated as a result of very little activity of TTCPase
and definite activity of TTLase, in contrast to the
situation in mammalian brain [216]. On the other
hand, membrane-bound tubulin from both squid
ganglion and mammalian brain contain only small
amount of tyrosinated tubulin [216, 217]. Lack of
C-terminal tyrosine appears to be a characteristic
of membrane tubulin (cf. [218]). In the giant axon
of the squid restoration of membrane excitability
which had once been destroyed was achieved by
perfusing TTLase in combination with tyrosine,
tubulin and some other factors [219, 220]. Non-
tyrosinated tubulin is absent in toad oocytes and
early embryos, whereas the toad brain contains a
higher proportion of the non-tyrosinated form
[221]. TTCPase is not detectable in the former,
but is comparable to that found in mammalian
brain in the latter. Little tyrosinable tubulin is
present in Tetrahymena where no TTLase activity
has so far been detected [208]. It was reported that
MAP-2 bound to microtubules made with detyro-
sinated tubulin to a lesser extent than to those
made with tyrosinated tubulin [222]. Replacement
of the C-terminal tyrosine by phenylalanine was
observed in the infant rat brain after induction of
hyperphenylalaninemia [223].

As a useful tool for analyzing the behavior of
tyrosinated (Tyr) and detyrosinated (Glu) a-
tubulin within cells, a rat monoclonal antibody
(YL1/2) raised against yeast tubulin [83] was found
to specifically react with the tyrosinated form of
brain a-tubulin [84], The antibody also reacted

1176

H. Mohri and N. Hosoya

with the synthetic peptide Gly-(Glu)3-Gly-(Glu)2-
Tyr, but did not react with Gly-(Glu)3-Gly-(Glu)2.
In vitro assembly of the microtubule was not
affected by the antibody. When injected into 3T3
cells, the antibody bound to microtubules without
affecting their distribution at low concentration,
but inhibited saltatory movement of organelles and
mitosis at higher concentrations [84, 224]. Polyc¬
lonal antibodies specific for the C-termini of either
Tyr a-tubulin or Glu a-tubulin have also been
prepared against the corresponding synthetic
polypeptides [85]. Investigations with these anti¬
bodies revealed a distinct, but overlapping dis¬
tribution of subclasses of microtubules in cultured
cells [85, 225]. For example, microtubules with
Tyr a-tubulin are present both in the spindle and
asters, but those with Glu a-tubulin are restricted
to half spindles and the interzone at telophase
during mitosis. In contrast to proliferating cells,
the level of Glu a-tubulin is higher in stable
microtubules of differentiated cells such as those in
sperm axonemes [226]. Based on the staining
pattern of a trypanosome with the YL1/2 during
the cell cycle, it has been suggested that Tyr
a-tubulin is a marker of newly formed microtu¬
bules, the detyrosinated one accumulating in ‘old’
microtubules [41]. A similar conclusion that the
turnover rate of Tyr microtubules is higher than
that of Glu microtubules has been reached using
these antibodies in an epithelial cell line [227]. It
appears that Tyr tubulins predominant in the tubu¬
lin pool assemble to give Tyr microtubules which
are converted into Glu microtubules by TTCPase.
When the Glu microtubules disassemble, the re¬
sulting Glu tubulins may be efficiently retyrosin-
ated by TTLase [228].

Acetylation

In Chlamydomonas, the post-translational con¬
version of an a-tubulin (al) confined to the cell
body into a major flagellar a-tubulin (a3) was
observed during regeneration after deflagellation
[146, 229]. The a-tubulin translated in vitro from
mRNA isolated from cells during regeneration was
identified as al, while there was no detectable
synthesis of a3. Similar post-translational mod¬
ification of flagellar a-tubulin was also described in
the alga Polytomella [230] and in the trypanosome,

Crithidia fasciculata [231, 232]. The modification
was reversible and the flagellar a3 tubulin was
deacetylated to the cytoplasmic al form during
flagellar resorption, which is accompanied by mic¬
rotubular disassembly [233]. From these findings,
a close relationship between the post-translational
modification and flagellar assembly was post¬
ulated.

Soon after it was revealed that the post-
translational modification of a-tubulin in Chlamy¬
domonas was due to acetylation of the e-amino
group of lysine(s) [234]. The same was true in the
flagellar axonemal microtubules of Trypanosoma
brucei [41]. Furthermore, both al and a3 tubulin
appear to exist in both tyrosinated and detyrosin¬
ated forms, the al being the true substrate for
TTLase. As mentioned above, specific monoclon¬
al antibodies were raised against acetylated a-
tubulin from sea urchin sperm axonemes [86]. The
antibodies bound to the axonemal a-tubulin of
different organisms, but did not recognize non-
acetylated a-tubulin such as that of unfertilized sea

Fig. 4. Immunofluorescence of the mitotic apparatus of
the starfish, Asterina pectinifera. Mitotic apparatus
was double-stained with a polyclonal antibody
against sea urchin tubulin (a: Polyscience Inc.) and a
monoclonal antibody specific for acetylated a-
tubulin (b: named 6-11B-1, a kind gift from Dr. G.
Piperno, Rockefeller University, NY).

Multi-facets of Tubulin 1177

1 2345 12345 12345

Fig. 5. Analysis of the acetylation of tubulin from several sources. Bands represent the following samples: egg
extract and sperm flagellar fraction of starfish, Asterias amurensis (1,2), egg extract of sea urchin, Hemicentrotus
pulcherrimus (3), egg extract and sperm axoneme fraction of sea urchin, Anthocidaris crassispina (4, 5). Samples
were subjected to SDS-PAGE, followed by electrophoretical blotting to nitrocellulose membrane, a: Coomassie
brilliant blue (CBB) staining of gel, b: immunostaining of nitrocellulose with a monoclonal antibody against
a-tubulin (Amersham), c: immunostaining of nitrocellulose with 6-11B-1 (see Fig. 4). T: tubulin.

urchin eggs (Figs. 4 and 5). The antigen, however,
was not restricted to the flagellar and ciliary ax-
onemes. Microtubules containing acetylated a-
tubulin have been found in basal bodies, cen-
trioles, mitotic spindles, midbodies and certain
subclasses of cytoplasmic microtubules by im¬
munofluorescence microscopy using one of the
monoclonal antibodies [41, 86, 235, 236]. In the
life cycle of Physarum , acetylated a-tubulin is
localized in the flagella of the flagellate and in the
microtubule organizing center (MTOC) of the
amoeba, but is not detected at the plasmodial stage
[237]. Acetylated a-tubulin is abundant in neural
tissues [238, 239].

Thus, the acetylation of a-tubulin is not confined
to flagellar and ciliary assembly. In addition to
axonemal microtubules, acetylated a-tubulin is
present in cytoplasmic microtubules resistant to
depolymerizing drugs such as nocodazole or col¬
chicine, but not in cold-resistant microtubules [41,
235, 236, 238]. Tubulin acetylation is also en¬
hanced by the treatment of PC12 cells with nerve
growth factor which reduces the susceptibility of
microtubules to depolymerizing agents [238].
Furthermore, it has been reported that tubulin
polymer is a better substrate than tubulin dimer for

partially purified a-tubulin acetylase [240]. The
acetylation of a-tubulin may contribute to the
stabilization of certain classes of microtubules,
especially those forming special arrays such as in
elongated structures. In some cell types, acety¬
lated tubulin and detyrosinated tubulin reside in
the same subclass of microtubules. However, this
is not the case in other cell types [241, 242].
Post-translational modification, therefore, is not
always a prerequisite for microtubule stability. It
is not certain whether acetylation increases the
lipid solubility of tubulin.

Glycosylation and other modifications

It was reported that tubulin preparations from
the mammalian brain contain appreciable amounts
of carbohydrates [243, 244]. That [14C]glu-
cosamine is incorporated into tubulin in vivo and
recovered as a mixture of glucosamine and galacto-
samine, was shown in the mouse brain [245].
Binding of concanavalin A and other agents con¬
sidered to have specificity for complex carbohy¬
drates to microtubules was also demonstrated
[246, 247]. Ciliary membrane tubulin was found to
be periodic acid-schiff (PAS) positive, whereas
axonemal tubulin was not significantly positive

1178

H. Mohri and N. Hosoya

[248]. Furthermore, a dramatic increase of tubulin
glycosylation was observed in diabetic rats, reduc¬
ing the ability of tubulin to undergo normal in vitro
assembly [249]. In spite of these findings, specific
glycosylation of tubulin as a post-translational
modification still remains unproven.

On the other hand, we detected the presence of
lipids in the outer doublet fraction in sea urchin
spermatozoa [250]. Microtubules isolated after the
incubation of chick embryos or Hela cells with 32Pi
contained labeled phospholipids [251]. In addi¬
tion, it was reported that brain tubulin contained
phospholipids and assembled into membranous
forms [252], Since Feit and Barondes [253] pro¬
vided the evidence for tubulin localization in mem¬
brane fractions from nervous tissue, membrane-
associated tubulin has been obtained from various
sources such as thyroid [254] and cilia [255] (for
review, see [218]). Tubulin can be found in
lipid-soluble or water-soluble forms. Membrane

tubulin appears to form a high molecular weight
complex with lipids and other proteins. Even
though tubulin is a constituent of membrane, the
association of specific lipids with tubulin remains
to be elucidated.

Recently, methylation of tubulin and high
molecular weight MAPs by protein carboxy-
methyltransferase was reported [256]. The enzyme
is rich in the brain and testes. Its roles in sperm
motility [257] and other functions have been sug¬
gested. There is a report that reductive methyla¬
tion of a highly reactive lysine residue of a-tubulin
(Lys-394 of bovine brain tubulin) renders tubulin
assembly incompetent [258]. Some evidence was
obtained which indicates that tubulin is a zinc-
containing protein [259], without any further sup¬
port. However, it is certain that zinc ions modify
tubulin polymerization, bringing about a zinc-
induced sheet [260].

We have followed changes in the cytoskeleton

2 3

I M I M |

M

>

' •

A06 J* A

.

V

Fig. 6. Analysis of isoforms of tubulin at immature (I) and maturing (M) stages of starfish oocytes, Asterina
pectinifera. Samples were subjected to isoelectric focusing in a one-dimensional slab minigel system. Proteins
were then electrophoretically blotted to a nitrocellulose membrane and stained with anti a-tubulin (1) and anti
/?-tubulin (2) monoclonal antibodies (Amersham). Proteins were further analyzed by subsequent SDS-PAGE.
After isoelectric focusing, the slab gel was sliced into strips, applied to SDS-PAGE as the second dimension using
the same minigel system, and stained with CBB (3). Large arrow heads indicate a- and ^-tubulin. Small arrow
heads indicate major proteins which did not change isoelectric points at these two stages.

Multi-facets of Tubulin

1179

during the maturation of starfish oocytes [261]. In
immature oocytes with germinal vesicles, the mi¬
crotubule network is widely distributed in the
cortex and within the cytoplasm. When the
oocytes are subjected to treatment with 1 -methyl
adenine, the germinal vesicle breaks down and the
first meiotic spindle is formed concomitant with a
reduction in the number of cytoplasmic microtu¬
bules. Such a change is repeated during the second
meiotic division. In a preliminary experiment, the
total amount of tubulin varied only slightly during
maturation, but both a- and /?- tubulin exhibited a
characteristic cyclic change. Namely, among the
four isoforms of a-tubulin distinguishable by
isoelectric focusing, two basic isoforms dis¬
appeared cyclically during meiotic divisions. Such
a cyclic shift to more acidic forms also occurred in
/?-tubulin (Fig. 6). It appears that this is not due to
degradation and synthesis of these isoforms, but is
a post-translational modification. However, phos¬
phorylation, tyrosination and acetylation (see Fig.
5) so far do not seem to be the cause, but the
change was found to be Ca2+-dependent. Further
experiments are needed to elucidate this peculiar
phenomenon.

SUMMARY AND PERSPECTIVE

During the two decades since the naming of
tubulin, it is apparent that much information has
been accumulated in relation to tubulin and micro¬
tubules. Thus, the multiple functions of tubulin
even in a single cell has been established, even
though tubulin has been a relatively conservative
protein during evolution. There are multiple gene
families for both subunits of tubulin, and each of
the genes are dispersed in the genome and are
transcribed coordinately or selectively during de¬
velopment or a life cycle. Even though different
tubulin isotypes copolymerize to microtubules,
certain tubulin isotypes are expressed in special
tissues and/or at specific stages of embryogenesis,
and form special classes of microtubules. These
different classes of microtubules exhibit a different
stability to cold and various drugs and have specific
interactions with MAPs and other cellular compo¬
nents which have not been treated in this article.

Furthermore, post-translational modification

gives rise to more tubulin isoforms than the num¬
ber of expressed genes. It appears that phosphoryl¬
ation of the C-terminal region by different protein
kinases regulates the assembly of tubulin and
interactions with MAPs, etc. Conversely, phos¬
phorylation of MAPs and other proteins would
also modify tubulin assembly and various functions
of microtubules. Most, but not all of the newly
formed tubulin species have the C-terminal tyro¬
sine. When assembled into microtubules, Tyr
tubulin is detyrosinated by TTCPase, and if such
microtubules are degraded, Glu tubulin in the
dimer pool is retyrosinated by TTLase. The activi¬
ties of these enzymes would affect the proportion
of Tyr and Glu microtubules within cells. Acetyla¬
tion of tubulin may contribute to the stability of
certain classes of microtubules. It is conceivable
that glycosylation and binding to specific lipids are
concerned with the various phenomena shown by
membranes. Tubulin would be modified by other
as yet unknown processes.

Presently and in the coming years, specific anti¬
bodies against the known sequences of individual
tubulin isoforms will be utilized to distinguish
different classes of microtubules in various cell
types and to follow their temporal and spatial
changes immunocytochemically, although this is
not almighty. Such antibodies would also reveal
the nature of specific interactions of different
tubulins or microtubules with MAPs and other
cellular components. The genetic approach, such
as the isolation of new tubulin mutants or the
introduction of in vitro engineered genes into cells
will further contribute to understanding the struc¬
ture and function of specific tubulins and microtu¬
bules. Studies of genes other than tubulin genes
are also needed to elucidate the mechanisms of
tubulin gene regulation or construction of specific
microtubule arrays. In combination with such
experiments, the injection of fluorescently-labeled
or biotinylated tubulin into cells will provide a
powerful tool for investigating the dynamics of
individual microtubules. Video-enchanced con¬
trast microscopy has proven to be efficient in
observing a single microtutule and/or phenomena
associating with microtubules. Together with the
isolation and characterization of new proteins or
enzymes which interact with microtubules, re-

1180

H. Mohri and N. Hosoya

search along the above-mentioned lines will grad¬
ually solve the unanswered problems concerning
the multiple functions of microtubules.

REFERENCES

1 Mohri, H. (1968) Nature, 217: 1053-1054.

2 Yanagisawa, T., Hasegawa, S. and Mohri, H.
(1968) Exp. Cell Res., 52: 86-100.

3 Engelhardt, V. A. (1946) Adv. Enzymol., 6: 147-
191.

4 Burnasheva, S. A. (1958) Biokhimia, 23: 558-563.

5 Nelson, L. (1962) In “Spermatozoan Motility”.
Ed. by D. W. Bishop, Am. Assoc. Adv. Sci.,
Washington, pp. 171-187.

6 Nelson, L. (1966) Biol. Bull., 130: 378-386.

7 Gibbons, I. R. (1963) Proc. Natl. Acad. Sci. USA,
50: 1002-1010.

8 Gibbons, I. R. and Rowe, A. J. (1965) Science,
149: 424-426.

9 Mohri, H., Hasegawa, S., Yamamoto, M. and
Murakami, S. (1969) Sci. Pap. Coll. Gen. Educ.,
Univ. Tokyo, 19: 195-217.

10 Ogawa, K., Mohri, T. and Mohri, H. (1977) Proc.
Natl. Acad. Sci. USA, 74: 5006-5010.

11 Summers, K. E. and Gibbons, I. R. (1971) Proc.
Natl. Acad. Sci. USA, 68: 3092-3096.

12 Sale, W. S. and Satir, P. (1977) Proc. Natl. Acad.
Sci. USA, 74: 2045-2049.

13 Hata, H., Yano, Y., Mohri, T., Mohri, H. and
Miki-Noumura, T. (1980) I. Cell Sci., 41: 331-340.

14 Stephens, R. E., Renaud, F. L. and Gibbons, I. R.
(1967) Science, 156: 1606-1608.

15 Renaud, F. L., Rowe, A. I. and Gibbons, I. R.
(1968) J. Cell Biol., 36: 79-90.

16 Shelanski, M. L. and Taylor, E. W. (1967) J. Cell
Biol., 34: 549-554.

17 Shelanski, M. L. and Taylor, E. W. (1968) J. Cell
Biol., 38: 304-315.

18 Mohri, H., Murakami, S. and Maruyama, K.
(1967) J. Biochem., 61: 518-519.

19 Mohri, H. and Yanagisawa, T. (1967) Zool. Mag.,
76: 276-278.

20 Mohri, H. and Shimomura, M. (1973) J.
Biochem., 74: 209-220.

21 Mazia, D. and Ruby, A. (1968) Proc. Natl. Acad.
Sci. USA, 61: 1005-1012.

22 Linck, R. W., Albertini, D. F., Kenney, D. M.
and Langevin, G. L. (1982) Cell Motility, Suppl. 1:
127-132.

23 Borisy, G. G. and Taylor, E. W. (1967) J. Cell
Biol., 34: 525-533.

24 Bryan, J. and Wilson, L. (1971) Proc. Natl. Acad.
Sci. USA, 68: 1762-1766.

25 Ponstingl, H., Krauhs, E., Little, M. and Kempf,
T. (1981) Proc. Natl. Acad. Sci. USA, 78: 2757-
2761.

26 Krauhs, E., Little, M., Kempf, T. . Holer
Warbinek, R., Ade, W. and Ponstingl, H. (1981)
Proc. Natl. Acad. Sci. USA, 78: 4156-4160.

27 Weisenberg, R. C. (1972) Science, 177: 1104-
1105.

28 Weingarten, M. D., Lockwood, A. H., Hwo, S.-
Y. and Kirschner, M. W. (1975) Proc. Natl. Acad.
Sci. USA, 72: 1858-1862.

29 Valenzuela, P., Auiroga, M., Zaldivar, J., Rutter,
W. I., Kirschner, M. W. and Cleveland, D. W.
(1981) Nature, 289: 650-655.

30 Fulton, C. and Simpson, P. A. (1976) In “Cell
Motility.” Ed. by R. Goldman, T. Pollard and J. L.
Rosenbaum, Cold Spring Harbor Press, New
York, pp. 987-1005.

31 Mohri, H. (1976) Biochim. Biophys. Acta, 456:
85-127.

32 Goldman, R., Pollard, T. and Rosenbaum, J. L.
(1976) Cell Motility Book C. Cold Spring Harbor
Press, New York.

33 Kirschner, M. W. (1978) Int. Rev. Cytol., 54: 1-
71.

34 Dustin, P. (1978) Microtubules. Springer- Verlag,
Berlin.

35 Roberts, K. and Hyams, J. S. (1979) Microtu¬
bules. Academic Press, London.

36 Sakai, H., Mohri, H. and Borisy, G. G. (1982)
Biological Functions of Microtubules and Related
Structures. Academic Press, New York.

37 Borisy, G. G., Cleveland, D. W. and Murphy, D.
G. (1984) Molecular Biology of the Cytoskeleton.
Cold Spring Harbor Press, New York.

38 Cleveland, D. W. and Sullivan, K. F. (1985) Ann.
Rev. Biochem., 54: 331-365.

39 Perry, G. W. and Wilson, D. L. (1982) J.
Neurochem., 38: 1155-1159.

40 Cyr, R. I., Sotak, M., Bustos, M. M., Guiltinan,
M. I. and Fosket, D. E. (1987) Biochim. Biophys.
Acta, 914: 28-34.

41 Gull, K., Hussey, P. J., Sasse,R., Schneider, A.,
Seebeck, T. and Sherwin, T. (1986) J. Cell Sci.,
Suppl. 5: 243-255.

42 O’Farrell, P. H. (1975) J. Biol. Chem., 250: 4007-
4021.

43 Lu, R. C., and Elzinga, M. (1977) Anal.
Biochem., 77: 243-250.

44 Geahlen, R. L. and Haley, B. E. (1979) J. Biol.
Chem., 254: 11982-11987.

45 Clayton, L., Pogson, C. I. and Gull, K. (1979)
FEBS Lett., 106: 67-70.

46 Little, M., Ludueha, R. F., Langford, G. M.,
Asnes, C. F. and Farrell, K. (1981) J. Mol. Biol.,

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

Multi-facets of Tubulin

1181

149: 95-107.

Little, M. (1985) BioSystems, 18: 241-247.
Piperno, G. and Luck, D. J. L. (1977) J. Biol.
Chem., 252: 383-391.

Gozes, I. and Barnstable, C. J. (1982) Proc. Natl.
Acad. Sci. USA, 79: 2579-2583.

Thompson, W. C., Asai, D. J. and Carney, D. H.
(1984) J. Cell Biol., 98: 1017-1025.

Crepeau, R. H., McEwen, B. and Edelstein, S. J.
(1978) Proc. Natl. Acad. Sci. USA, 75: 5006-5010.
Amos, L. A. and Baker, T. S. (1979) Nature, 279:
607-612.

Feit, H., Slusarek, L. and Shelanski, M. L. (1971)
Proc. Natl. Acad. Sci. USA, 68: 2028-2031.
Witman, G. B., Carlson, K. and Rosenbaum, J. L.
(1972) J. Cell Biol., 54: 540-555.

Piperno, G. and Luck, D. J. L. (1976) J. Biol.
Chem., 251: 2161-2167.

Bibring, T., Baxandall, J., Denslow, S. and Wal¬
ker, B. (1976) J. Cell Biol., 69: 301-312.
Kobayashi, Y. and Mohri, H. (1977) J. Mol. Biol.,
116: 613-617.

Stephens, R. E. (1978) Biochemistry, 17: 2882-
2891.

Cleveland, D. W., Fischer, S. G., Kirschner, M.
W. and Laemmli, U. K. (1977) J. Biol. Chem.,
252: 1102-1106.

Kobayashi, Y. (1982) J. Biochem., 92: 1305-1308.
Feit, H., Neudeck, V. and Baskin, F. (1977) J.
Neurochem., 28: 697-706.

Marotta, C. A., Harris, J. L. and Gilbert, J. M.
(1978) J. Neurochem., 30: 1431-1440.

Gozes, I. and Littauer, U. (1978) Nature, 276:
411-413.

Dahl, J. L. and Weibel, V. J. (1979) Biochem.
Biophys. Res. Commun., 86: 822-828.

Von Hungen, K., Chin, R. C. and Baxter, C. F.
(1981) J. Neurochem., 37: 511-514.

George, H. J., Misra, L., Field, D. J. and Lee, J.
C. (1981) Biochemistry, 20: 2402-2409.

Wolff, A., Denoulet, P. and Jeantet, C. (1982)
Neurosci. Lett., 31: 323-328.

Moura Neto, V., Mallat, M., Jeantet, C. and
Prochiantz, A. (1983) EMBO J., 2: 1243-1248.
Field, D. J., Collins, R. A. and Lee, J. C. (1984)
Proc. Natl. Acad. Sci. USA, 81: 4041-4045.
Murphy, D. B. and Wallins, K. T. (1983) J. Biol.
Chem., 258: 7870-7875.

Detrich, H. W., Ill and Overton, S. A. (1986) J.
Biol. Chem., 261: 10922-10930.

Brown, B. A., Nixon, R. A. and Marotta, C. A.
(1982) J. Cell Biol., 94: 159-164.

Kemphues, K. J., Raff, R. A., Kaufman, T. C.
and Raff, E. C. (1979) Proc. Natl. Acad. Sci.
USA, 76: 3991-3995.

74 Kemphues, K. J., Raff, E. C., Raff, R. A. and
Kaufman, T. C. (1980) Cell, 21: 445-451.

75 Oakley, B. R. (1985) Can. J. Biochem. Cell Biol.,
63: 479-488.

76 Weber, K. , Osborn, M. , Franke, W. W. , Seib, E. ,
Scheer, U. and Herth, W. (1977) Cytobiologie, 15:
285-302.

77 Fulton, C., Kane, R. E. and Stephens, R. E.
(1971) J. Cell Biol., 50: 762-773.

78 Yanagisawa, T., Yazaki, I. and Mohri, H. (1973)
Sci. Pap. Coll. Gen. Educ., Univ. Tokyo, 23: 155—
158.

79 Piperno, G. and Luck, D. J. L. (1977) J. Biol.
Chem., 252: 383-391.

80 Morgan, J. L., Holladay, C. R. and Spooner, B. S.
(1978) Proc. Natl. Acad. Sci. USA, 75: 1414-1417.

81 Asai, D. J., Brokaw, C. J., Thompson, W. C. and
Wilson, L. (1982) Cell Motility, 2: 599-614.

82 Thompson, W. C., Asai, D. J. and Carney, D. H.
(1984) J. Cell Biol., 98: 1017-1025.

83 Kilmartin, J. V., Wright, B. and Milstein, C.
(1982) J. Cell Biol., 93: 576-582.

84 Wehland, J., Willingham, M. C. and Sandoval, I.

V. (1983) J. Cell Biol., 97: 1467-1475.

85 Gundersen, G. G., Kalnoski, M. H. and Bulinski,
J. C. (1984) Cell, 38: 779-789.

86 Piperno, G. and Fuller, M. T. (1985) J. Cell Biol.,
101: 2085-2094.

87 Luduena, R. F. and Woodward, D. D. (1973)
Proc. Natl. Acad. Sci. USA, 70: 3594-3598.

88 Cleveland, D. W., Lopata, M. A., MacDonald, R.
J., Cowan, N. J., Rutter, W. J. and Kirschner, M.

W. (1980) Cell, 20: 95-105.

89 Cowan, N. J. and Dudley, L. (1983) Int. Rev.
Cytol., 85: 147-173.

90 Lemischka, I. R., Farmer, S. R., Rancaniello, V.
R. and Sharp, P. A. (1981) J. Mol. Biol., 151:
101-120.

91 Hall, J. L., Dudley, L., Dobner, P. R.. Lewis, S.
A. and Cowan, N. J. (1983) Mol. Cell. Biol., 3:
854-862.

92 Cowan, N. J., Dobner, P. R., Fuchs, E. V. and
Cleveland, D. W. (1983) Mol. Cell. Biol., 3: 1738-
1745.

93 Rudolph, J. E., Kimble, M., Hoyle, H. D., Subler,
M. A. and Raff, E. C. (1987) Mol. Cell. Biol., 7:
2231-2242.

94 Alexandraki, D. and Ruderman, J. V. (1981) Mol.
Cell. Biol., 1: 1125-1137.

95 Youngblom, J., Schloss, J. A. and Silflow, C. D.
(1984) Mol. Cell. Biol, 4: 2686-2696.

96 Kimmel, B. E., Samson, S., Wu,J., Hirschberg,
R. and Yarbrough, L. R. (1985) Gene, 35: 237-
248.

97 Conzelmann, K. K. and Helftenbein, E. (1987) J.

1182

H. Mohri and N. Hosoya

Mol. Biol., 198: 643-653.

98 Singhofer-Wowra, M., Little, M., Clayton, L.,
Dawson, P. and Gull, K. (1986) J. Mol. Biol., 192:
919-924.

99 May, G. S., Tsang, M. L.-S., Smith, H., Fidel, S.
and Morris, N. R. (1987) Gene, 55: 231-243.

100 Smith, H. A., Allaudeen, H. S., Whitman, M. H.,
Koltin, Y. and Gorman, J. A. (1988) Gene, 63:
53-63.

101 Neff, N. F., Thomas, J. H., Grisafi, P. and Bot-
stein, D. (1983) Cell, 33: 211-219.

102 Toda, T., Adachi, Y., Hiraoka, Y. and Yanagida,
M. (1984) Cell, 37: 233-242.

103 Guiltinan, M. J., Ma, D.-P., Barker, R. F., Bus¬
tos, M. M., Cyr, R. J., Yadegari, R. and Fosket,
D. E. (1987) Plant Mol. Biol., 10: 171-184.

104 Oppenheimer, D. G., Haas, N., Silflow, C. D. and
Snustad, D. P. (1988) Gene, 63: 87-102.

105 Monteiro, M. J. and Cleveland, D. W. (1988) J.
Mol. Biol., 199: 439-446.

106 Sullivan, K. F. and Cleveland, D. W. (1986) Proc.
Natl. Acad. Sci. USA, 83: 4237-4331.

107 Lopata, M. A. and Cleveland, D. W. (1987) J.
Cell Biol., 105: 1707-1720.

108 Joshi, H. C., Yen, T. J. and Cleveland, D. W.
(1987) J. Cell Biol., 105: 2179-2190.

109 Vallee, R. (1980) Proc. Natl. Acad. Sci. USA, 77:
3206-3210.

110 Michiels, F., Falkenburg, D., Muller, A. M.,
Hinz, V., Otto, U., Bellman, R., Glatzer, K. H.,
Brand, R., Bialojan, S. and Renkawitz-Pohl, R.
(1987) Chromosoma, 95: 387-395.

111 Theurkauf, W. E., Baum, H., Bo, J. and Winsink,
P. C. (1986) Proc. Natl. Acad. Sci. USA, 83:
8477-8481.

112 Pratt, L. F., Okamura, S. and Cleveland, D. W.
(1987) Mol. Cell. Biol., 7: 552-555.

113 Hecht, N. B., Distel, R. J., Yelick, P. C.,
Thanhauser, S. M., Driscoll, C. E., Goldberg, E.
and Tung, K. S. K. (1988) Mol. Cell. Biol., 8: 996-
1000.

114 Monteiro, M. J. and Cox, R. A. (1987) FEBS
Lett., 217: 260-264.

115 Raff, E. C. (1984) J. Cell Biol., 99: 1-10.

116 Cleveland, D. W. (1987) J. Cell Biol., 104: 381-
383.

117 Marotta, C. A., Strocchi, P. and Gilbert, J. M.
(1979) J. Neurochem., 33: 231-246.

118 Gozes, I., Baelselier, A. and Littauer, U. Z.
(1980) Eur. J. Biochem., 103: 13-20.

119 Wilde, C. D., Crowther, C. E., Cripe, T. P., Lee,
M. G.-S. and Cowan, N. J. (1982) Nature, 297:
83-84.

120 Villasante, A., Wang, D., Dobner, P., Dolph, P.,
Lewis, S. A. and Cowan, N. J. (1986) Mol. Cell.

Biol., 6: 2409-2419.

121 Wang, D., Villasante, A., Lewis, S. A. and
Cowan, N. J. (1986) J. Cell Biol., 103: 1903-1910.

122 Cleveland, D. W., Hughes, S. H., Stubblefield, E.,
Kirschner, M. W. and Varmus, H. E. (1981) J.
Biol. Chem., 256: 3130-3134.

123 Sanchez, F., Natzle, J. E., Cleveland, D. W., Kirs¬
chner, M. W. and McCarthy, B. J. (1980) Cell, 22:
845-854.

124 Kalfayan, L. and Wensink, P. C. (1981) Cell, 24:
97-106.

125 Seebeck, T., Whittaker, P. A., Imboden, M. A.,
Hardman, N. and Braun, R. (1983) Proc. Natl.
Acad. Sci. USA, 80: 4634-4638.

126 Landfear, S. M., McMahon-Pratt, D. and Wirth,
D. F. (1983) Mol. Cell. Biol., 3: 1070-1076.

127 Spithill, T. W. and Samaras, N. (1987) Mol.
Biochem. Parasitol., 24: 23-37.

128 Tittawella, I. and Normark, S. (1987) FEMS Mic¬
robiol. Lett., 43: 317-320.

129 Shea, D. K. and Walsh, C. J. (1987) J. Cell Biol.,
105: 1303-1309.

130 Callahan, R. C., Shalke, G. and Gorovsky, M. A.
(1984) Cell, 36: 441-445.

131 Schatz, P. J., Solomon, F. and Botstein, D. (1986)
Mol. Cell. Biol., 6: 3722-3733.

132 Sullivan, K. F., Lau, J. T. Y. and Cleveland, D.
W. (1985) Mol. Cell. Biol., 5: 2454-2465.

133 Lewis, S. A., Lee, M. G.-S. and Cowan, N. J.
(1985) J. Cell Biol., 101: 852-861.

134 Denoulet, P., Edde, B. and Gros, F. (1986) Gene,
50: 289-297.

135 Miller, F. D., Naus, C. C. G., Durand, M.,
Bloom, F. E. and Milner, R. J. (1987) J. Cell
Biol., 105: 3065-3073.

136 Havercroft, J. C. and Cleveland, D. W. (1984) J.
Cell Biol., 99: 1927-1935.

137 Lewis, S. A., Gilmartin, M. E., Hall, J. L. and
Cowan, N. J. (1985) J. Mol. Biol., 182: 11-20.

138 Fuller, M.T., Caulton, J. H., Hutchens, J. A.,
Kaufman, T. C. and Raff, E. C. (1987) J. Cell
Biol., 104: 385-394.

139 Raff, E. C., Fuller, M. T., Kaufman, T. C., Kem-
phues, K. J., Rudolph, J. E. and Raff, R. A.
(1982) Cell, 28: 33-40.

140 Gasch, A., Hinz, U., Leiss, D. and Renkawitz-
Pohl, R. (1988) Mol. Gen. Genet., 211: 8-16.

141 Distel, R. J., Kleene, K. C. and Hecht, N. B.
(1984) Science, 224: 68-70.

142 Murphy, D. B., Wallins, K. T., Machlin, P. S.,
Ratrie, H. and Cleveland, D. W. (1987) J. Biol.
Chem., 262: 14305-14312.

143 Merlino, G. T., Chamberlain, J. P. and Klein-
smith, L. J. (1978) J. Biol. Chem., 253: 7078-
7085.

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

Multi-facets of Tubulin

1183

Marcaud, L. and Hayes, D. (1979) Eur. J.
Biochem., 98: 267-273.

Weeks, D. P. and Collis, P. S. (1976) Cell, 9: 15-
27.

Lefebvre, P. A., Silflow, C. D., Wieben, E. D.
and Rosenbaum, J. L. (1980) Cell, 20: 469-477.
Minami, S. A., Collis, P. S., Young, E. E. and
Weeks, D. P. (1981) Cell, 24: 89-95.

Lai, E. Y., Walsh, C., Warded, D. and Fulton, C.
(1979) Cell, 17: 867-878.

Fong, D. and Chang, K.-P. (1981) Proc. Natl.
Acad. Sci. USA, 78: 7624-7628.

Silflow, C. D. and Rosenbaum, J. L. (1981) Cell,
24: 81-88.

Brunke, K. J., Young, E. E., Buchbinder, B. U.
and Weeks, D. P. (1982) Nucl. Acids Res., 10:
1295-1310.

Baker, E. J., Schloss, J. A. and Rosenbaum, J. L.
(1984) J. Cell Biol., 99: 2074-2081.

Harlow, P. and Nemer, M. (1987) Genes Dev., 1:
1293-1304.

Oakley, B. R. and Morris, N. R. (1981) Cell, 24:
837-845.

Abraham, I., Marcus, M., Cabral, F. and Gottes-
man, M. M. (1983) J. Cell Biol., 97: 1055-1061.
Ben-Ze’ev, A., Farmer, S. R. and Penman, S.
(1979) Cell, 17: 319-325.

Cleveland, D. W., Lopata, M. A., Sherline, P. and
Kirschner, M. W. (1981) Cell, 25: 537-546.
Cleveland, D. W., Pittenger, M. F. and Feramisco,
J. R. (1983) Nature, 305: 738-740.

Pittenger, M. F. and Cleveland, D. W. (1985) J.
Cell Biol., 101: 1941-1952.

Pachter, J. S., Yen, T. J. and Cleveland, D. W.
(1987) Cell, 51: 283-292.

Yen, T. J., Gay, D. A., Pachter, J. S. and Cleve¬
land, D. W. (1988) Mol. Cell. Biol., 8: 1224-1235.
Baker, E. J., Keller, L. R., Schloss, J. A. and
Rosenbaum, J. L. (1986) Mol. Cell. Biol., 6: 54-
61.

Lefebvre, P. A. and Rosenbaum, J. L. (1986)
Ann. Rev. Cell Biol., 2: 517-546.

Gong, Z. and Brandhorst, B. P. (1987) Mol. Cell.,
Biol. 7: 4238-4246.

Weeks, D. P., Collis, P. S. and Gealt, M. A.
(1977) Nature. 268: 667-668.

Seyfert, H.-M. (1987) Eur. J. Cell Biol., 43: 182-
188.

Chalfie, M. and Thomson, J. N. (1982) J. Cell
Biol., 93: 15-23.

Schultze, E. and Kirschner, M. (1987) J. Cell
Biol., 104: 277-288.

Kirschner, M. and Mitchison, T. (1986) Cell, 45:
329-342.

Goodman, D. B. P., Rasmussen, H. , DiBella, F.

and Guthrow, C. E., Jr. (1970) Proc. Natl. Acad.
Sci. USA, 67: 652-659.

171 Sloboda, R. D., Rudolph, S. A., Rosenbaum, J.

L. and Greengard, P. (1975) Proc. Natl. Acad. Sci.
USA, 72: 177-181.

172 Murofushi, H. (1973) Biochim. Biophys. Acta,
327: 354-364.

173 Eipper, B. A. (1972) Proc. Natl. Acad. Sci. USA,
69: 2283-2287.

174 Eipper, B. A. (1974) J. Biol. Chem., 249: 1407-
1416.

175 Forgue, S. T. and Dahl, J. L. (1979) J.
Neurochem., 32: 1015-1025.

176 Luduena, R. F., Zimmermann, H.-P. and Little,

M. (1988) FEBS Lett., 230: 142-146.

177 Reddington, M., Tan, L. P. and Lagnado, J. R.
(1976) J. Neurochem., 27: 1229-1236.

178 Burke, B. E. and DeLorenzo, R. J. (1982) J.
Neurochem., 38: 1205-1218.

179 Wandosell, F., Serrano, L., Hernandez, M. A. and
Avila, J. (1986) J. Biol. Chem., 261: 10332-10339.

180 Piperno, G. and Luck, D. J. (1976) J. Biol.
Chem., 251: 2161-2167.

181 Ikeda, Y. and Steiner, M. (1979) J. Biol. Chem.,
254: 66-74.

182 Berry, S., Dawicki, D. and Steiner, M. (1988)
Biochem. Biophys. Res. Commun., 151: 1250-
1255.

183 Joseph, M. K., Fernstrom, M. A. and Soloff, M.
S. (1982) J. Biol. Chem., 257: 11728-11733.

184 Yamauchi, T. and Fujisawa, H. (1983) Biochem.
Biophys. Res. Commun., 110: 287-291.

185 Hargreaves, A., Wandosell, F. and Avila, J. (1986)
Nature, 323: 827-828.

186 Serrano, L., Diaz-Nido, J., Wandosell, F. and Avi¬
la, J. (1987) J. Cell Biol., 105: 1731-1739.

187 Singh, T. J., Akatsuka, A., Huang, K.-P., Mur¬
phy, A. S. N. and Flavin, M. (1984) Biochem. Bio¬
phys. Res. Commun., 121: 19-26.

188 Wandosell, F., Serrano, L. and Avila, J. (1987) J.
Biol. Chem., 262: 8268-8273.

189 Cheng, N. and Sahyoun, N. (1988) J. Biol. Chem.,
263: 3935-3942.

190 Akiyama,T., Kadowaki, T., Nishida,E.,
Kadooka,T., Ogawara, H., Fukami,Y., Sakai,
H., Takaku, F. and Kasuga, M. (1986) J. Biol.
Chem., 261: 14794-14803.

191 Gard, D. L. and Kirschner, M. W. (1985) J. Cell
Biol., 100: 764-774.

192 Olmsted, J. B. (1986) Ann. Rev. Cell Biol., 2:
421-457.

193 Jameson, L. and Caplow, M. (1981) Proc. Natl.
Acad. Sci. USA, 78: 3413-3417.

194 Murthy, A. and Flavin, M. (1983) Eur. J.
Biochem., 137: 37-46.

1184

H. Mohri and N. Hosoya

195 Yamamoto, H., Fukunaga, K., Goto, S., Tanaka,
E. and Miyamoto, E. (1985) J. Neurochem., 44:
759-768.

196 Dinsmore, J. H. and Sloboda, R. D. (1988) Cell,
53: 769-780.

197 Tash, J. S., Hidaka, H. and Means, A. R. (1986)
J. Cell Biol., 103: 649-655.

198 Hayashi, H., Yamamoto, K., Yonekawa, H. and
Morisawa, M. (1987) J. Biol. Chem., 262: 16692-
16698.

199 Rozdzial, M. M. and Haimo, L. T. (1986) Cell, 47:
1061-1070.

200 Barra, H. S., Rodriguez, J. A., Arce, C. A. and
Caputto, R. (1973) J. Neurochem., 20: 97-108.

201 Barra, H. S., Arce, C. A., Rodriguez, J. A. and
Caputto, R. (1974) Biochem. Biophys. Res. Com-
mun., 60: 1384-1390.

202 Arce, C. A., Rodriguez, J. A., Barra, H. S. and
Caputto, R. (1975) Eur. J. Biochem., 59: 145-149.

203 Raybin, D. and Flavin, M. (1975) Biochem. Bio¬
phys. Res. Commun., 65: 1088-1095.

204 Raybin , D . and Flavin , M . (1977) J . Cell Biol ., 73 :
492-504.

205 Thompson, W. C. (1977) FEBS Lett., 80: 9-13.

206 Preston, S. F., Deanin, G. G., Hanson, R. K. and
Gordon, M. W. (1979) J. Mol. Evol., 13: 233-244.

207 Murofushi, H. (1980) J. Biochem., 87: 979-984.

208 Kobayashi, T. and Flavin, M. (1981) Comp.
Biochem. Physiol., 69B: 387-392.

209 Argarana, C. E., Barra, H. S. and Caputto, R.
(1978) Mol. Cell. Biochem., 19: 17-21.

210 Thompson, W. C., Deanin, G. G. and Gordon, M.
W. (1979) Proc. Natl. Acad. Sci. USA, 76: 1318-
1322.

211 Barra, H. S., Argarana, C. E. and Caputto, R.
(1982) J. Neurochem., 38: 112-115.

212 Arce, C. A., Hallak, M. E., Rodriguez, J. A.,
Barra, H. S. and Caputto, R. (1978) J.
Neurochem., 31: 205-210.

213 Wehland, J. and Weber, K. (1987) J. Cell Biol.,
104: 1059-1067.

214 Rodriguez, J. A. and Borisy, G. G. (1978)
Biochem. Biophys. Res. Commun., 83: 579-586.

215 Barra, H. S., Arce, C. A. and Caputto, R. (1980)
Eur. J. Biochem., 109: 439-446.

216 Kobayashi, T. and Matsumoto, G. (1982) J.
Biochem., 92: 647-652.

217 Nath, J. and Flavin, M. (1978) FEBS Lett., 95:
335-338.

218 Stephens, R. E. (1986) Biol. Cell., 57: 95-110.

219 Matsumoto, G. and Sakai, H. (1979) J. Membrane
Biol., 50: 15-22.

220 Sakai, H., Matsumoto, G., Endo, S. and
Kobayashi, T. (1981) In “Nerve Membrane
Biochemistry and Function of Channel Proteins”.

Ed. by G. Matsumoto and M. Kotani, Univ. Tokyo
Press, Tokyo, pp. 185-202.

Barra, H. S., Modesti, N. M. and Arce, C. A.
(1987) Comp. Biochem. Physiol., 87B: 151-155.
Kumar, N. and Flavin, M. (1985) Eur. J.
Biochem., 128: 215-222.

Rodriguez, J. A. and Borisy, G. G. (1979) Scien¬
ce, 206: 463-465.

Wehland, J. and Willingham, M. C. (1983) J. Cell
Biol., 97: 1476-1490.

Gundersen, G. G. and Bulinski, J. C. (1986) J.
Cell Biol., 102: 1118-1126.

Gundersen, G. G. and Bulinski, J. C. (1986) Eur.
J. Cell Biol., 42: 288-294.

Webster, D. R., Gundersen, G. G., Bulinski, J. C.
and Borisy, G. G. (1987) Proc. Natl. Acad. Sci.
USA, 84: 9040-9044.

Gundersen, G. G., Khawaja, S. and Bulinski, J. C.
(1987) J. Cell Biol., 105: 251-264.

Brunke, K. J., Collis, P. S. and Weeks, D. P.
(1982) Nature, 297: 516-519.

McKeithan, T. W., Lefebvre, P. A., Silflow, C. D.
and Rosenbaum, J. L. (1983) J. Cell Biol., 96:
1056-1063.

Russell, D. G., Miller, D. and Gull, K. (1984)
Mol. Cell. Biol., 4: 779-790.

Russell, D. G. and Gull, K. (1984) Mol. Cell.
Biol., 4: 1182-1185.

L’Hernault, S. W. and Rosenbaum, J. L. (1985) J.
Cell Biol., 100: 457-462.

L’Hernault, S. W. and Rosenbaum, J. L. (1985)
Biochemisty, 24: 473-478.

LeDizet, M. and Piperno, G. (1986) J. Cell Biol.,
103: 13-22.

Piperno, G., LeDizet, M. and Chang, X.-J. (1987)
J. Cell Biol., 104: 289-302.

Diggins, M. A. and Dove, W. F. (1987) J. Cell
Biol., 104: 303-309.

Black, M. M. and Keyser, P. (1987) J. Neurosci.,
7: 1833-1842.

Sale, W. S., Besharse, J. C. and Piperno, G.
(1988) Cell Motil. Cytoskel., 9: 243-253.

Maruta, H., Greer, K. and Rosenbaum, J. L.
(1986) J. Cell Biol., 103: 571-579.

Schulze, E., Asai, D.J., Bulinski, J. C. and Kirs-
chner, M. (1987) J. Cell Biol., 105: 2167-2177.
Bulinski, J. C., Richards, J. E. and Piperno, G.
(1988) J. Cell Biol., 106: 1213-1220.

Falxa, M. L. and Gill, T. J. (1969) Arch. Biochem.
Biophys., 135: 194-200.

Margolis, R. K., Margolis, R. U. and Shelanski,
M. L. (1972) Biochem. Biophys. Res. Commun.,
47: 432-437.

Feit, H. and Shelanski, M. L. (1975) Biochem.
Biophys. Res. Commun., 66: 920-927.

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

Multi-facets of Tubulin

1185

246 Huttich, K. , Muller, H. and Unger, E. (1977) Acta
Histochem., 58: 324-331.

247 Prus, K. and Mattisson, A. (1979) Histochemistry,
61: 281-289.

248 Stephens, R. E. (1977) Biochemistry, 16: 2047-
2058.

249 Williams, S. K., Howarth, N. L., Devenny, J. L.
and Bitensky, M. W. (1982) Proc. Natl. Acad. Sci.
USA, 79: 6546-6550.

250 Mohri, H., Murakami, S. and Maruyama, K.
(1967) Abst. 7th Intern. Congr. Biochem. Tokyo,
pp. 1-13.

251 Daleo, G. R., Piras, M. M. and Piras, R. (1974)
Biochem. Biophys. Res. Commun., 61: 1043-1050.

252 Feit, H. and Shay, J. W. (1980) Biochem. Bio¬
phys. Res. Commun., 94: 324-331.

253 Feit, H. and Barondes, S. H. (1970) J.
Neurochem., 17: 1355-1364.

254 Bhattacharyya, B. and Wolff, J. (1975) J. Biol.
Chem., 250: 7639-7646.

255 Stephens, R. E., Oleszko-Szuts, S. and Good, M.
J. (1987) J. Cell Sci., 88: 527-535.

256 Ohta, K., Seno, N., Yoshida,T., Hiraga, K. and
Tuboi, S. (1987) Biochimie, 69: 1227-1234.

257 Bouchard, P., Gagnon, C., Phillips, D. M. and
Bardin, C. W. (1980) J. Cell Biol., 86: 417-423.

258 Szasz, J., Yaffe, M. B., Elzinga, M., Blank, G. S.
and Sternlicht, H. (1986) Biochemistry, 25: 4572-
4582.

259 Morisawa, M. and Mohri, H. (1972) Exp. Cell
Res., 70: 311-316.

260 Larsson, H., Wallin, M. and Edstrom, A. (1976)
Exp. Cell Res., 100: 104-110.

261 Hosoya,N., Shirai, H., Hosoya, H., Nagahama,
Y. and Mohri, H. (1987) Zool. Sci., 4: 1046.

-

1187

Proceedings of the

Fifty-Ninth Annual Meeting of the

Zoological Society of Japan

October 8-10, 1988
Sapporo

Suffixal letters of abstract number refer to the abbreviated
subfields of zoology

PH : physiology

GE : genetics

BI : biochemistry

DB : developmental biology

EN : endocrinology

MO : morphology

BB : behavior biology

EC : ecology

TS : taxonomy and systematics

ZOOLOGICAL SCIENCE 5: 1189 (1988)

© 1988 Zoological Society of Japan

[THE ZOOLOGICAL SOCIETY PRIZE]

Structure and Function of Nonspiking Interneurons

Mituhiko Hisada

Zoological Institute, Faculty of Science, Hokkaido University, Sapporo 060

Our study of nonspiking interneurons has arisen
out of an attempt to understand how the central
nervous system of the crayfish selects and com¬
mands a certain behavioral act.

In contrast to the general notion that the neural
system is comprised mainly of spiking, action-
potential generating neurons, we have found that a
large proportion of the total neurons in a ganglion
of the crayfish is the local nonspiking type of
neuron.

Nonspiking neurons can exert motor effects
according to their membrane potential change.
Current injection through a microelectorode into
the nonspiking neurons results in an extensive
change in motoneuron activity. The transmission
is chemical. Some nonspiking neurons can mod¬
ulate the motoneuron activity even when they are
hyperpolarized. This indicates that they have a
continuous release of transmitters at their “rest¬
ing” potential.

Although we were able to classify the nonspik¬
ing neurons with regard to their postsynaptic effect
and mode of transmitter release, no exact struc¬
tural correlate to these functional types has so far
been found. There is a high possibility that their
individual branches function rather independently,
forming numerous local circuits. Input and output
synapses found distributed in close proximity
support this notion. This is an important aspect to
be incorporated in the evaluation of any neural
network involving this type of neuron. Yet
another important aspect is that by gathering many
synaptic inputs from various branches, nonspiking
neurons could exert a smooth sustained output
suited for control of the relative threshold of the
postsynaptic cells by modulating their resting
membrane potentials.

Having gained this basic knowledge about non¬
spiking neurons, we hypothesize that the nonspik¬

ing neurons act as a heterosynaptic facilitatory
element to a given reflex pathway, i.e., a gating or
a biasing element in the selection of the behavioral
output. This possibility is tested vigorously in the
interaction of the abdominal movement system
and uropod steering system. Intracellular analysis
revealed that the uropod motoneurons are con¬
tinuously depolarized while the abdominal system
is in operation. This depolarization is found to be
mediated by nonspiking neurons. Subthreshold
excitatory postsynaptic potentials from the des¬
cending pathway can elicit spike activity in the
motoneurons only while this continuous back¬
ground depolarization takes place via nonspiking
neurons. Nonspiking interneurons are thus in¬
volved, not merely in the direct control of
behavioral acts but also in the modulation and
coordination of single and multiple action.

Injection of current subthreshold to the spike
initiation into some dendritic branches of an
ordinary spiking motoneuron in the crayfish mod¬
ulates the activity of neighboring motoneurons.
This indicates that the nonspiking mode of signal
transmission is far more widely used in the central
nervous system of arthropods than generally be¬
lieved. This finding together with the presence of a
large number of nonspiking neurons necessitated a
radical revision of our understanding of how the
central nervous system operates in these animals.
It remains to be seen whether the nonspiking mode
of transmission is a primitive form of intercellular
communication. Yet at present, it appears reason¬
able to assume that the nonspiking neurons are the
products of an evolutionary process during which
the behavior to be executed has become more and
more complex, though the total number of
neurons has remained limited. The hope is that in
time we will be able to answer many of these
unsettled questions.

ZOOLOGICAL SCIENCE 5: 1190 (1988)

© 1988 Zoological Society of Japan

[THE ZOOLOGICAL SOCIETY PRIZE]

The Intracellular Symbiont of Insects as a Genetic Element

Hajime Ishikawa

Zoological Institute, Faculty of Science, University of Tokyo, Tokyo 113

Since the discovery of intracellular symbionts as
Blochmann bodies in several insect tissues in the
late 19th century, studies of symbionts have
centered upon their structure and nutritional
biochemistry in relation to the physiology of host
insects. Until quite recently, it had scarcely drawn
biologists’ attention that the intracellular symbiont
serves as a genetic element with a genome distinct
from that of the host. It is our work on aphid
symbionts that brought about a breakthrough in
symbiology in this respect.

Intracellular symbionts of aphids are harbored
by the mycetocyte, a cell differentiated specifically
for this purpose. The aphid embryo growing
parthenogenetically in its mother’s ovariole is
infected with symbionts at an early stage of
development, and the symbionts are thereby trans¬
mitted from generation to generation of the host
insect without ever having a free-living stage.
Probably because of this, the symbionts cannot
successfully replicate themselves extracellularly.
Conversely, aposymbiotic aphids are completely
sterile, suggesting that the host depends on the
symbionts for its reproduction. Such a close
interdependence between the host and symbionts
is rarely observed in other organisms, including
insects that contain endosymbionts.

The primary symbiont of the pea aphid, Acyr-
ihosiphon pisum, possesses a genome a few times
larger than that of E. coli. The symbiont’s
genomic DNA is characteristic in that its G + C
content is as low as 30%. While the symbiont
synthesizes several hundred proteins when incu¬

bated in vitro, it synthesizes but one protein,
symbionin, when present inside the host cell. One
possibility that has been raised is that symbionin is
coded for by the host genome and is synthesized by
the symbion’s ribosomes. Implicit in this assump¬
tion is the belief that the mRNA of symbionin is
transferred across the boundary from host to
symbiont in order to be translated. Symbionin is
an acidic protein with a molecular mass of 63 kDa,
and is produced primarily by the symbionts har¬
bored by the embryo inside the mother’s ovariole.
Symbionin may either be a storage protein to be
consumed over the course of post-embryonic
development of the host insect or an enzyme
involved in the recycling of nitrogen compounds
essential to the host.

Since most of the aphid symbiont’s genes are not
expressed inside the host cell, like pseudogenes,
they are neutral to Darwinian selection and thus,
are fully susceptible to mutation. Actually, it has
been suggested that the aphid symbiont’s genome
is undergoing evolution much more rapidly than
that of the host genome. According to the Serial
Endosymbiosis Theory, cell organelles containing
DNA are looked upon as intracellular symbionts
which have become specialized through the loss of
much of their genetic coding capacity. If at
present, a symbiont tends to become specialized
functionally through evolution in response to its
intracellular environment, we may well suppose
that similar events in the past eventually led to the
production of cell organelles such as mitochondria
and plastids.

ZOOLOGICAL SCIENCE 5: 1191 (1988)

© 1988 Zoological Society of Japan

[THE ZOOLOGICAL SOCIETY PRIZE]

Developmental and Molecular Analyses of Ooplasmic Factors
Required for Pole Cell Formation

Masukichi Okada

Institute of Biological Sciences, University of Tsukuba, Tsukuba 305

It was in the second decade of this century that
germ plasm, a locally specialized ooplasm, was
observed to participate in the establishment of the
germ line in some developmental systems. The
aim of our present studies is to attain an under¬
standing of the role of the germ plasm in germ line
formation on a molecular level.

The germ line cells, which are called pole cells in
Drosophila melanogaster, are formed at the early
syncytial blastoderm stage. Pole cell formation is
inhibited if an early cleavage embryo is uv-
irradiated posteriorly at its germ plasm. However,
we found that the irradiated embryo restored its
ability to form pole cells if it was microinjected
with intact germ plasm immediately after irradia¬
tion. This showed that a germ plasmic factor could
be transplantable without losing its ability to form
pole cells. Furthermore, this raised the possibility
that one could use this uv-irradiation-micro-
injection procedure for monitoring the factor
during its isolation from embryos. Testing every
subcellular fraction using this procedure, we scruti¬
nized embryo homogenates and found that the
ability to restore pole cell formation was retained
in the postmitochondrial fraction sedimented from
a homogenate of early cleavage embryos and
oocytes.

The postmitochondrial fraction was analyzed in
order to find the component(s) responsible for the
restoration activity. Digestion of the fraction with
a protease did not affect the restoration activity of
the fraction, but RNase treatment dramatically
decreased the activity. Encouraged by these
results, we extracted RNA from the post¬
mitochondrial fraction to test its restoration activi¬
ties. Only poly(A) +RNA could restore pole cell

forming ability to uv-irradiated embryos. For
further analysis, a cDNA library was generated
from the poly(A) + RNA and was screened in order
to select a cDNA of the RNA responsible for the
restoration activity. Since we had found a stage
specific mode of occurrence of poly(A)+RNA with
restoration activity, cDNA clones that hybridized
with poly(A)+RNA present in young cleavage
embryos but not in blastoderm embryos were
selected, using the colony hybridization technique.
DNA extracted from each of those clones was
tested as to whether it hybridized with
poly(A) + RNA that had restoration activity.
Finally, such a cDNA clone (pDE20.6) was
obtained. The cDNA insert of pDE20.6 hybri¬
dized with a 1.5 kb RNA on a northern blot. No
southern blot signal was detected from nuclear
DNA, nor was an in situ hybridization signal found
on salivary gland chromosomes. Furthermore, a
search in a computerized data base for nucleotide
sequences homologous to pDE20.6 cDNA pro¬
duced only mitochondrial large ribosomal DNA of
Drosophila yakuba. The homology was 98%.

Tracing the properties of a cytoplasmic factor to
restore pole cell formation to uv-irradiated
embryos, we have encountered the mitochondrial
IrRNA. On the other hand, our biological studies
suggest that more cytoplasmic factors involved in
the pathway of pole cell formation and germ cell
differentiation should be found in germ plasm.
Thus, our conclusion for the present is that germ
plasm is not simply a germ cell determinant but a
complex of cytoplasmic factors encoded by
mitochondrial as well as nuclear genes. The role of
every factor must be elucidated before we can
reach the goal of our studies.

1192

Physiology

PH 1

BACTERIAL CHEMOTAXIS AS AN ORIGIN OF
NEURONAL TRANSDUCTION.

T. Shinozawa-*- and S. Fukunaga^ . Dept, of
Natural Science, Naruto Univ. of
Education. Naruto. Tokushima. _ _

There are several analogical points in
the mechanism of bacterial chemotaxis and
neuronal transduction. In both systems, the
adsorption of some chemicals (ligands)
triggers the transduction and several amino
acids act as triggers in both systems.
Therefore, we supposed that the origin of
neuronal transduction is the bacterial
chemotaxis. The effect of some chemicals
affecting the neuronal system were tested
on the bacterial chemotaxis. Chemotaxis of
Escherichia coli was assayed by the
modification of Adler's method using 1 ;ul
capillary and also by the swarming method
on the semisolid agar plate. More than lO-^
M of methamphetamine (called "HIROPON" in
Japan) attracted EU_ coli . Acetylcholine
was not an attractant or a repellent for
this bacteria. However, more than 10~^ M of
this chemical inhibited the chemotaxis of
this bacteria for aspartate. These data
support the above idea concerning the
origin of neuronal transduction system and
now more precise experiments are under
investigation. (Present Address: 1; Dept,
of Polymer Science, Faculty of Technology,
Gunma University. 2; Hitoyoshi Educational
Institution for Agricultural Technology)

PH 2

AN ANTIGEN FOR A MONOCLONAL ANTIBODY WHICH
DEPRESSES THE RESPONSE EVOKED BY GLUTATHI¬
ONE IN HYDRA , PROBABLY A RELEVANT RECEPTOR
K. Ohta , K. Hanai and H. Morita. Dept. of
Biol., Fac. of Sci., Kyushu Univ., Fukuoka

There are at least 5 components of the

behavioral response (R1-R5) which are
evoked at different concentrations of GSM
( S-methy lglutathione ) . We have prepared
several monoclonal antibodies (Mabs) each
of which specifically depresses some of
the components. Specific effects on the
response and specific structures stained
by these Mabs suggest that these Mabs bind
to receptors mediating the response. As we
could only detect the antigen correspond¬
ing to R5 , we analysed the R5 antigen in
detail. Mab J245 is one of these antibod¬
ies and depresses R2 , R4 , and R5 . We exam¬
ined the interactions between the antigen
for J245 and ligands. We labelled the
solubilized membrane fraction with GSPAP
( S- ( p-azidophenacyl ) -glutathione ) which
specifically depressed R5 . We analysed the
labelled antigen with HPLC after extrac¬
tion by the immunoprecipitation . We could
detect a 250kDa protein as the R5 antigen.
GSM or L-glu , a competitive inhibitor of
the response, dose-dependently reduced the
amount of 35S-GSPAP bound to the antigen.
The dissociation constants for GSM and L-
glu were 55 yM and 90 yM, respectively.
These values were consistent with those
estimated from the behavioral experiments.
The antigen may be a glutathione receptor
mediating R5 .

PH 3

GENETIC ALTERATION OF MULTIPLE TASTE RECEPTOR
SITES FOR SUGARS IN Drosophila.

T. Tanimura.

Dept. Biol., Fac. Sci., Fukuoka Univ., Fukuoka

Several lines of evidence suggest that there
are multiple taste receptor sites for sugars in
Drosophila. Mutants of Drosophila with an altered
taste sensitivity to a particular group of sugars
can provide a direct evidence for the presence of
multiple sugar receptor sites. If a gene coding
the receptor molecule has been identified, we
would know the nature of the molecule through
cloning the gene. Employing the two-choice
behavioral preference test utilizing two kinds of
food-colors, autosomal mutants with a reduced
sensitivity to glucose were isolated through a
mutagenesis with ethylmethane sulfonate. A free
recombination and selection procedure was used to
eliminate lethal mutations and to isolate dominant
mutants. Three mutants were isolated and genetic
analyses showed that they belong to a same
complementation group. The gene was mapped on the
left arm of the third chromosome. Two-choice
preference tests with different combinations of
two kinds of sugars suggested that the taste
sensitivity to both glucose and sucrose is
reduced, while the sensitivity to fructose do not
change in the mutant. Experiments on the time
course of feeding of a single sugar solution also
supported the idea. Electrophysiological analyses
of the mutant would reveal the precise role of the
gene in the taste receptor mechanisms.

PH 4

RESPONSES OF TARSAL SUGAR RECEPTOR IN
DROSOPHIRA .

A . Shiraishi 1 , T.Koga^-, T . Tanimura^ .
iDept. of Biol., Fac. of Sci., Kyushu
Univ., Fukuoka and ^Dept. 0f Biol., Fac.
of Sci., Fukuoka Univ., Fukuoka.

Two strains, C~anton-S and mutant,

Tre, in Drosophira were used through¬
out^ this work. Flies after emergences
were raised on 0.1 M sucrose and tap
water during three or four days and used
in the experiment. Ambient temperature
during the experiment was between 22°C
and 23° C . Relative humidity was above 80
%. Tip-recording method was adopted to
record impulses from the largest chemo-
sensory hairs located at the ventral side
of tarsus. Stimulus sugar solution was
dissolved into 0.001 M LiCl to maintain
electric conductivity. Two kinds of
impulses were recorded on stimulation by
NaCl . The large spikes were presumed to
be from salt receptor. The small spikes
did not suppressed by higher concentration
of NaCl but, presumed to be from water
receptor. Responses on stimulations by
various sugars were also recorded from the
largest hair. Therefore, it was concluded
that one hair contained three kinds of
chemoreceptors , sugar, salt and water
receptor, respectively. Response of
tarsal sugar receptor of the mutant,

Tre , was suppressed only to treharose
stimulation .

Physiology

1193

PH 5

TASTE CELL ADAPTATION AND CALCIUM.

M. Ozaki and T. Amakawa. Dept, of Biol.,
Col. of Gen. Educ., Kobe Univ. , Nada,
Kobe.

Following the report in visual and
olfactory cells, also in vertebrate sugar
taste cells, it has been reported that
cyclic nucleotide acts as intracellular
messenger for the sensory-transduction.

Here we report that Ca2 + may cause
negative feed back to the transduction
cascade in sugar receptor cell of the
blowfly, Phormia regina , promoting
adaptation of the cell. Increase of
intracellular Ca2+ concentration and
application of IP^ into the cell accele¬
rated adaptation. Moreover, phorbol ester
accelerated the adaptation. Therefore,
sugar reception may trigger not only the
transduction cascade but also the inositol
phospholipid pathway as adaptation
cascade. That is, sugar reception couples
with PIP- hydrolysis producing IP-, IP-, ->
mobilizes intracellular Ca2+, ana the Ca2+
may promote the adaptation probably via
protein kinase C. _

Involvement of Ca2+ in the adaptation
mechanism has been documented in visual
cells, ^ut in this case, it is thought
that Ca2+ permeates the stimulus-regulated
sodium channel into the cell and acts via
guanylate cyclase.

PH 6

PHYSIOLOGY AND MORPHOLOGY OF SEX PHEROMONE
RESPONSIVE RELAY NEURONS IN DEUTOCEREBRUM
OF THE MALE AMERICAN COCKROACH
T.Hanada and T.Shibuya. Inst. of Biol.
Sci.,Univ. of Tsukuba, Ibaraki .

Responses to periplanone A and B of
deutocerebral relay neurons of male
american cockroach Periplaneta americana
were recorded intracellular ly. The relay
neurons responded to these phromone
components were devided into three types of
macroglomerular (MG) neuron and two types of
ordinary-glomerular (OG) neuron according to
their arborization patterns in the
glomeruli. MG relay neurons which arborized
both whole part and ventral part of MG
responded to both periplanone A and B
strongly, but two types differed in pattern
of the dose-response relationship. In these
two types of neuron, the dose-response
curves to periplanone A were parallel to
that of periplanone B, but shifted at the
higher concentration. Other types of MG
relay neuron which has arborization in the
dorsal part responded to periplanone A
preferentially. Some OG relay neurons also
responded to pheromone components with
excitatory or inhibitory response. But
pheromone responses of these neurons were
more gentle in increase of frequency of
spikes than general odor responses. These
results suggest pheromone information is
mainly processed in MG labelled line system
for their mating behavior, but other system
exists for utilization of pheromone
information in some aspects of behavior.

PH 7

ANTENNAL CONTACT CHEMORECEPTOR CELLS
RESPONDING TO INTERMALE RECOGNITION
PHEROMONE IN THE MALE LOBSTER COCKROACH
A.Watanabe and T.Shibuya. Inst, of Biol.
Sci., Univ. of Tsukuba, Ibaraki.

In adult males of the lobster cockroach
Naupheta cJnerea, sexual recognition occurs
through antennal contact. To a conspecific
adult female, a male responds with
wing-raising display as a courtship
behavior. A male responds to other males
with aggressive actions, though the
releaser pheromone of wing-raising display
(wing-raising stimulant) exists in
cuticular wax of both sexes. The inhibition
of wing-raising display and release of
aggressive behavior are due to intermale
recognition pheromone called nauphoetin
( octadecyl ( Z ) -9-tetracosenoate ) .

Spike responses to nauphoetin were
obtained by means of tip recording method
from chaetic sensilla. These excitatory
responses were observed in 4.72% of tested
chaetic sensilla. Wing-raising stimulant
did not elicite spikes in chaetic sensilla
responsive to nauphoetin. Excitatory
response elicited by wing-raising stimulant
was not suppressed by nauphoetin. In these
contact pheromones, response spectra of
chaetic sensilla did not overlap at all.
These results suggested that peripheral
input of these contact pheromones was sent
in parallel to central nervous system. It
was inferred that the inhibition of
wing-raising display by nauphoetin depended
upon central processing in the brain.

PH 8

TASTE SYSTEM OF MEXICAN SALAMANDER,
AXOLOTL .

T. Nagai. Dept. Physiol., Teikyo Univ.
Sch . Med , , Tokyo 1 73 _

The axolotl, Ambystoma mexicanum, is a

neotonous salamander, and is normally
aquatic for its entire life cycle. It is
easily maintained in the laboratory. It
has a large non-distendible tongue devoid
of hypoglossal innervation. From glosso¬
pharyngeal nerve bundle, the afferent
responses were recorded and integrated, to
salts ( K C 1 , NaCl, C a C 1 - ) , citric acid,
quinine-HCl, sucrose, and amino acids
(Phe, Arg, Gly, Ala, Glu). The responses
to salts were robust, but not to amino
acids even at 50 mM, although axolotl is
carnivore. In contrast, a mechanical
stimulation onto a single taste bud
induced a brisk response. The induced
units were larger in amplitude than
chemosensitive units, suggesting the
presence of fibers with a 1 arger diameter.
A tentative study on taste-induced
behaviour was done in avoidance reaction
to bitter taste. The axolotl in my
laboratory, usually fed on dog food
pellets, rejected the pellet containing
quinine-HCl of more than 10 mM. A Co-Lys
complex applied to the distal cut end of
glossopharyngeal nerve stained the cell
bodies (20 - 3b urn in diameter) and fibers
in the ganglion. I ntracel 1 u 1 ar recording
as well as staining is in progress to more
study the morphology and physiology of the
ganglion cells.

1194

Physiology

PH 9

LOBULE STRUCTURE AND SOMATOTOPY OF THE
MEDULLARY FACIAL LOBE IN THE CHANNEL
CATFISH.

T. Hayama and J. Caprio. Dept, of Zool.
and Physiol., Louisiana State University,
Baton Rouge, LA, USA _ ,

Correlation of somatic organization
with lobule structure in the facial lobe
(FL), a primary medullary gustatory nucleus
of the freshwater channel catfish,

Ictalurus punctatus , was examined. Histo¬
logical examination revealed six longitudi¬
nal columns (i.e. lobules) extending ros-
trocaudally in the FL. Elec t r ophy s i o 1 o-
gical mapping indicated that each of the 6
lobules receives segregated input from dis¬
crete portions of the external body sur¬
face; four barbels (the medial mandibular,
lateral mandibular, and maxillary and nasal
barbels), the face-flank and the pectoral
fin. The proximal-distal axis of each of
the barbels and the an t e r o-po s t e r i o r body
axis are represented in a po s t e r o-an t e r i or
lobule axis, indicating that the fish in
the FL is facing caudally. Taste responses
within the three barbel lobules examined
were generally observed in the dorsal
regions, while tactile responses were
throughout the lobules. The findings
indicate that the FL of the channel catfish
is highly organized to localize tactile and
taste stimuli on the external body surface
and further suggest that the lobule of the
FL is differentiated into two portions, a
dorsal taste and tactile, and a ventral,
primarily tactile region.

PH 10

CATION SELECTIVITY OF CYCLIC NUCLEOTIDE¬
GATED CONDUCTANCE IN ISOLATED OLFACTORY
RECEPTOR CELLS.

N. Suzuki. Zool. Inst., Fac. of Sci.,
Hokkaido Univ. , Sapporo.

Cation selectivity of cyclic nucleo¬
tide-gated conductance to alkali metal and
organic ions was studied in enzymatically
isolated bullfrog olfactory receptor cells
by means of whole-cell clamp technique.
Under voltage-clamp at -75 mV, 0.5 mM cGMP
in Cs+-internal solut ion ; ( mM ) CsCl 95,
CaCl2 0.7, MgCl2 2, EGTA-2Na 11, HEPES 10,
NaOH 6, pH 7.2, was injected into olfacto¬
ry receptor cells in Ca^+-free external
solution; (mM) NaCl 110, KC1 2, CoCl2 3,
MgCl2 1, glucose 10, HEPES 5, NaOH 1.76,
pH 7.2. Current-voltage relations for the
c-GMP-gated conductance were determined
by exchanging external solutions, in
which Na ions were replaced by other
cations. After corrections for non-specif¬
ic leak and voltage-dependent currrents,
reversal potential measurements showed
that the selectivity among alkali metal
ions was weak with a sequence, Li> Na > K
> Rb > Cs, and permeability ratios of
1.03, 1.0, 0.95, 0.78, and 0.75. Organic
cations, choline, tetramethylammonium and
tetraethylammonium , were less permeable
and permeability ratios to Na+ were 0.39,
0.19 and 0.17. Results suggest that the
cyclic nucleotide-gated conductance is the
pore with relatively high electric field
and its size is ~ 7 Angstroms in diameter.

PH 11

ODORANT-ACTIVATED CATIONIC CONDUCTANCE IN
THE OLFACTORY RECEPTOR CELLS OF NEWT
T. Kurahashi and T. Shibuya. Inst. Biol.
Sci., Univ. of Tsukuba, Ibaraki .

Some properties of the odorant-activated
conductance in the solitary olfactory
receptor cells was compared with those of
the olfactory receptor potential in situ.

(1) The sensitivity to odorants was
locarized at the apical dendrite. This is
well agreed with the results derived from
the elec trophysiologi cal and biochemical
study for the site of olfactory
transduction .

(2) The conductance showed time
dependent-inactivation which is comparable
to the adaptive properties of the olfactory
response .

(3) Permeability to each alkali metal
ion was almost identical, but little to the
anion, indicating the conductance to be a
cation-selective conductance. The
permeability ratio between K and Na, almost
1, well explained the reversal potential
value of receptor potential in in situ
cell, 0 mV.

These observations strongly suggest that
the olfactory receptor potential arises
from the activation of cation-selective
conductance .

PH 12

ACTIVE POTENTIAL OF DISSOCIATED NERVE CELLS
OF HYDRA.

H.Kusaka, T.Itayama and Y.Sawada. Res. Inst.
Elect. Comm., Tohoku University, Sendai.

Hydra is one of the simplest animal which
has a neural network system. In order to
understand the information processing mecha¬
nism in this system, it was planned to study
activation potential of a neural network
system placed on a multi-micro electrode
plate fabricated by a micro electronics
technics .

Hydra cells are dissociated by being
pipetted after soaked in a medium of high
osmotic pressure. Nerve cells are sucked by
a micro-pipet and placed on the electrodes
covered with the dissociation medium. Each
electrode is 30 microns wide and is separa¬
ted from each other by 30 microns. Only a
small circular part of 5 micron diameter at
the tip of each electrode is deinsulated by
use of exima laser. The resistance of the
electrode is in general several tens M-ohm,
and the capacitance between the neighboring
electrodes is several f-farad.

Under a favorite condition signals which
might be single electric pulses or an pulse
train are observed. In the present stage,
however, network formation has not been
observed on the electrode plate. Also it is
difficult to obtain the reproducible data.
The most important things we have to achieve
may be to construct an electrode facilita¬
ting cell adhesion and to develop a technic
to place nerve cells precisely on the spot.

Physiology

1195

PH 13

POSSIBLE ROLE OF PROCTOLIN AS NEURO¬
MODULATOR IN THE COCKROACH LEG MUSCLE.

H. Washio, S. Ishikawa, A. Ohmori and M.
Maruyama . Mitsubishi Kasei Inst. Life Sci.
Machida, Tokyo.

The modulatory action of the exogenous
proctolin and octopamien applied upon
neurally-generated contraction was studied
on the coxal depressor muscle 177d of the
cockroach, Periplaneta americana.

Proctolin increased the basal tension
marckedey and decreased the relaxation
rate of the contraction, while octopamine
increased the contraction and relaxation
rates and decrased the amplitude of the
tetanus tension. Stimulation of one of the
DUM neurons intracellularly during
production of tetanus of the muscle
reduced the amplitude of tetanus tension
and induced a rise in the basal tension of
this mucle.

The biological responsiveness to the
extract of DUM neurons was reduced by
dilution in a way which paralleles changes
in the responsiveness to different
concentrations of authentic proctolin. The
single peak of proctolin-like bioactive
materials which were partially purified
from the DUM cell extract coincided with
that of authentic proctolin. These rsults
implies the possible involvement of
proctolin in the DUM neuron with the
increase in the basal tension of the coxal
muscle in the cockroach together with
biogenic amine, octopamine.

PH 19

AMINE-CONTAINING NEURAL PROCESSES IN THE
GILL OF APLYSIA.

M. Kurokawa, K. Kuwasawa and M. Otokawa*.
Dept, of Biol., Tokyo Metropolitan Univ. ,
"Biol. Lab., Hosei Univ. , Tokyo. _

The aminergic constituent elements of the

gill of A. kurodai and A. j uliana were examined
using histochemical and HPLC techniques. In
both whole-mounted preparations and frozen
sections treated with glyoxylic acid,
yellowish green fluorescent fibers were
observed in the branchial nerve and in the
branchial ganglion. Fluorescent network
structures were observed in the branchial
vessels and pinnule. The structures contained
cell bodies having one or a few thin processes
and a dendritic process. Immunoreactive
fibers to rabbit antiserotonin antiserum
were observed in the branchial nerve and
around neurons in the branchial ganglion (BGNs) .
HPLC analysis showed that there was a greater
content of serotonin than dopamine in the
branchial nerve, whereas the branchial
ganglion, efferent vessel and pinnule had
greater contents of dopamine than serotonin.
These results suggest the presence of the
central serotonergic innervation to the gil 1 ,
and dopaminergic neurons in the gill periphery.
It was shown that serotonin depressed EJPs
induced by an identified central motor neuron
in the gill muscle cells, and that dopamine
produced excitatory effects on BGNs.

The results suggest that the aminergic
cellular organization of the gill involves
dopaminergic sensory neurons, and serotonergic
presynaptic inhibitory fibers originating
from the abdominal ganglion.

PH 15

EFFECTS OF THE Ca2t IONOPHORES, X-537A AND
A23187, ON THE NEUROMUSCULAR TRANSMISSION IN
THE FISH RED MUSCLE.

H. Igata1 and T. Hidaka. Dept. Biol., Fac.
Gen. Edu. , and ‘Fac. Sci., Kumamoto Univ.,
Kumamoto.

Effects of Ca2* ionophores, X-537A and A23187
were investigated on neuromuscular transmis¬
sion using the red muscle, M. levator pinnae
pectoral is of freshwater teleost (carp, Cypr-
inus carpio) and marine teleost (puffer fish,
Takif ugu rubripes and T. poeci lonotus) . Both
T(-537A and A23187 had similar actions. On the
application of these agents, the amplitude of
excitatory junction potential (ejp) and inhi¬
bitory junction potential (ijp) and the fre¬
quency of miniature excitatory junction po¬
tential (mejp) increased transitorily and then
decreased gradually after maximum increase
as reached. Finally, ejp and ijp disappeared.

Therefore, to confirm if these effects were
presynaptic or postsynapt ic, effects of these
agents on ACh potential and nerve action po¬
tential were investigated.

From the results, it is suggested that X-537A
and A23187 increase the intracellular Ca2+
concentration in the nerve terminals, and
consiquently accelerate ACh release, which
leads to increase in the amplitude of ejp and
ijp and the frequency of mejp. The decrease in
these junction potentials might be explained
by the desensitization of ACh receptor by
these agents. Final di sappearence of ejp and
ijp would be attributed in part to block of
nerve action potential by X-537A.

PH 16

DISTRIBUTION OF THE EXCITATORY AND INHIBI¬
TORY ACH RECEPTORS IN THE FISH RED MUSCLE.
T. Hidaka and T. Kita1. Dept. Biol., Fac.
Gen. Edu., and ‘Fac. Sci., Kumamoto Univ.,
Kumamoto.

The excitatory junction potential (ejp),
the inhibitory junction potential (ijp)
and the diphasic junction potential were
elicited by the nerve stimulation in the
red muscles of the freshwater and marine
teleosts. In the resting muscle, the mini¬
ature excitatory and inhibitory junction
potentials were observed. The previous
studies showed that the neuromuscular
transmissions were mediated by ACh and the
nature of receptors was nicotinic. In the
present study, it was intended to record
ACh potential elicited by the iontophore-
tic application of ACh and to reveal the
distribution of ACh receptors in the red
muscle of carp.

The ACh pipette was in position on the
surface of the muscle fiber within 100 n m
from the recording electrode. ACh poten¬
tials and junction potentials could be
recorded from the same neuromuscular junc¬
tion. Three types of ACh potentials, depo¬
larizing, hyperpolarizing and diphasic
ones, were recorded. The directions of the
potential change of ACh potentials and
junction potentials were same in some
muscle fibers and were different in other
fibers. Thus, it was demonstrated that the
excitatory and inhibitory ACh receptors
were closely distributed in this muscle.

1196

Physiology

PH 17

EFFECT OF ALCOHOL COMPONENT TO INITIA¬
TION OF MATING BEHAVIOUR IN MALE BEET
ARMYWORM .

F. Mochizuki1 and T. Shibuya2. 1 Shin-
Etsu Chemical. Co. LTD., Tokyo and 2Inst.
of Biol. Sci., Univ. of Tsukuba , Ibaraki .

Sex pheromone components of the beet
armyworm.Spodoptera exiqua , were
recently identified as [ Z9 , El 2-1 4 : Ac ]
and [Z9-14:OH], The impulse responses of
a single sensillum to both components
were extracellularly recorded. A single
sensillum responded with different
amplitude of impulses to each component.
Large impulses (about ImV) of "A" cell
in the sensillum appeared mainly to
[ Z9 , El 2-1 4 : Ac ] and small (about 0 . 5mV)
of "B" cell to [Z9-14:OH], Concentration
of threshold to the former was lower
about ten times than to the latter. To
the mixture of both components, impulse
frequency of "A" cell was decreased with
increase of [Z9-14:OH] concentration and
impulse initiation was delayed at high
ratio. The results may be shown some in¬
teraction between the sensillum cells to
mixture stimulation. Then responses to
[ Z 9 , El 2-1 4:0H] which has characteristic
of both [Z9,E12-14: Ac ] and [Z9-14:OH]
were recorded. The substance initiated
impulses of "B" cell mainly. Moreover,
mixture of ( Z9 ,E1 2-1 4 : Ac ] and [Z9,E12-
1 4 : OH ] acted as an attractant in wind-
tunnel-test. These results indicate that
[-0H] may be neccessary for excitation
of "B" cell.

PH 18

FUNCTION OF TEGULA ON THE SWALLOWTAIL
BUTTERFLY, Papilio xuthus .

T. Satoh and K. Aoki. Life Sci. Inst.,
Sophia Univ. , Tokyo

In normal insect flight, it has been
known that the centrally generated flight
pattern is subject to beat-by-beat
control of sensory inputs. We examined
such a sensory control system of wingbeat
putting a highlight on the tegula.

Methods were video recording of tethered
flight with the high speed CCD camera and
cobalt chloride filling of both tegula
and thoracic ganglion through the Nib
root. The intact wingbeat frequency was
6 to 8 Hz. When both forewings were cut
off 1/4, 1/3 and 1/2 of the wing length,
the animal showed graded and significantly
higher wingbeat frequency than frequency
of the previous length. The animal showed
apparently abnormal wingbeat pattern
after partial lesion of tegulae. The
tegula (3 - 3.5 mm long and 2 mm wide)
covers wing hinge of the forewing from
the top side. It has clustered hairs
along the lateral side. Whole mount
preparation of the tegula after cobalt
chloride filling shows each bipolar
sensory nerve extends the dendrite into
a hair. Central arborization was observed
in prothoracic and ptero- thoracic ganglia.

PH 19

INNERVATION AND ACTION OF DUM-NEURON IN
TfiE TERMINAL GANGLION ON FEMALE CRICKET, '
TBLEOGRYLLUS COMMODUS , TO RETRACTOR.

H.Ai and N.Ai. Dept, of Biol., Tokyo Gaku-
gei Univ., Koganei, Tokyo. _ _ _ _

In adult female cricket, five clusters
of DUM are oriented on the dorsal region.
Their innervation has been investigated by
using back-fill staining respectively . . In
the experiment , DUM located in posterior
portion of AG-V( abdominal ganglion-V) is
usbd. The DUM-neuron cluster is formed five
cells constantly. One of them is innervated
to posterior tergosternal muscle (M-4 )as re¬
tractor of ovipositor. Others are similarly
innervated to the common oviduct. Process
of DUM runs along the 7th lateral root of
AG-V. From the cut end of periphery located
to M-4,NiCl2 was. flowed by back-fill method.
One DUM and five motoneuron were cleraly
stained, and motoneurons were oriented on
the basal part of N-7 root. Intracellular
activities of DUM were recorded with burst
or spike train. But their time courses are
slightly long than that of motoneuron. In
retractor muscle M-4, junction potentials
and spontaneous oscillatory ones were re¬
corded intracellularly . However, when the
anterior connective was stimulated electri¬
cally and externally, only oscillatory
potentials were gradually disappeared, but
not junctional potentials .Repetitive stimu¬
lation to connective may yield the DUM's
activity, and DUM-neuron would be risen
long-lasting depression effect to muscle
activity for ovipositor movement.

PH 20

ULTRASTRUCTURAL AND IMMUNOHISTOLOGICAL
INVESTIGATION OF THE SPERMATHECAL DUCT OF
THE FEMALE CRICKET, GRYLLUS BIMACULATUS .

K . Yasuyama 1 , T . Kimura 2 and T . Yamaguchi 2 .
rDep. of Biol., Kawasaki Medical School,
Kurashiki and 2Dep. of Biol., Fac . of Sci.,
Okayama Univ. , Okayama. _ _____

The ultrastbueture of the spermathecal

duct(SD) and the distribution of the
serotonin- and proctolin-immunoreactive ( 5-
HTi and PROi) nerve fibers in its muscular
envelop were investigated. The wall of SD
consists of three cellular types: epithe¬
lial cells, secretory cells having a cavity
formed by an invagination of apical cell
membrane, and ductule cells carrying a thin
cuticular ductule. The secretory and duc¬
tule cells are restricted to the median
region where SD is highly convoluted, and
are absent from the proximal region near
the junction with the genital chamber and
from the distal region near the junction
with the spermatheca. When an isolated SD
was incubated in 5 mM ionic lanthanum solu¬
tion, lanthanum entered the cuticular duc¬
tule and reached not only the secretory
cavity, but also the lumen of SD. This fact
suggests that the cuticular ductule may be
an ionic pathway connecting the hemocoel
with the secretory cavity and lumen of SD.
5-HTi fibers are distributed in the proxi¬
mal region, while PROi fibers are extensi¬
vely distributed throughout SD. These
results reveal the presence of the morpho¬
logical differences among the proximal,
median and distal regions of SD.

Physiology

1197

PH 21

PH 23

Effect of some drugs and the bioactive
substance derived from the spermatophore
on the mechanical activity of the
spermathecal duct in the female cricket.
T. Kimura1 , K. Yasuyama2 and T.
Yamaguchi 1 * * . ^ept . of Biol., Fac. of
Sci., Uni v. of Okayama, Okayama and
2Dept. Biol., Kawasaki Med. Sch . ,
Kurashiki .

The u 1 t r as t r uc t ur a 1 and immunocyto-
chemical study of the spermathecal duct
(SD) showed the presence of a noticeable
difference among the proximal, median and
distal regions of SD (Yasuyama e_t a 1 . .
1988). Proctolin (Pr:>10_9M) increased
both the frequency and amplitude of
rhythmic small contractions of the distal
region near the spermatheca in a dose
dependent manner, and octopamine (OA:
>10“6 *M) and 5-HT (>10_4M) decreased both
of them, but the bioactive substance
derived from the spermatophore (BSS) had
no effect on this region. On the
contrary, OA and 5-HT decreased the
frequency of rhythmic large
contractions of the proximal region near
the genital chamber as well as its
amplitude, and BSS increased only its
frequency enormously. However, Pr never
changed the mechanical activity of this
region. Although the median region did
not show spontaneous contractions, Pr of
low concentration (<10"19M> induced the
rhythmic contractions, while Pr of high
concentrations (>10~8M) suppressed the
mechanical activity of the median region.

NONSPIKING GIANT INTERNEURONS (NGIs) IN
THE CRAYFISH BRAIN AND THEIR RESPON¬
SIVENESS TO MULTIMODAL SENSORY INPUTS.
Y.Okada and T . Yamaguchi . Dep. of Biol.,
Fac. of Sci., Okayama Univ. , Okayama.

Five pairs of NGIs (G1-G5) were
identified morphologically and function¬
ally in the crayfish brain: G1 , G2 and G3
are contained in one cluster, and G4 and
G5 are separate (Okada and Yamaguchi,
1988). Illumination of the contra- and
ipsilateral eyes, respectively elicits de¬
polarizing and hyper po 1 ar i z i ng graded
potentials in each of G1 , G2 and G3. These
three NGIs are also sensitive to a body
tilt around the length axis of the body:
they respond with depolarizing and hyper-
polarizing potentials to the body rolling
to the ipsilateral and contralateral
sides, respectively. In the sinusoidal
body rolling with the amplitude of 8 and
the frequency ranging from 0.05 to 0.5 Hz,
the higher the frequency of body rolling
is, the larger the response occurrs. The
dorsal illumination increases significant¬
ly the amplitude of response to body
rolling. Depolarizing and hyperpolar izing
current injection into a NGI increases and
decreases the firing rate of the eye-down
fibers, respectively. These results
suggest that the NGIs integrate the visual
and gravitational inputs to control the
compensatory eyestalk movement.

PH 22

MORPHOLOGY AND RESPONSE PROPERTIES OF LOCAL
NONSPIKING INTERNEURONS IN THE COCKROACH CERCAL
SYSTEM

J. Okuma, H. Morishita and Y. Kondoh Wako Research
Center, Honda R&D Co. LTD., Wako, Saitama 351-01,
Japan

1. Local nonspiking interneurons in the cereal sys¬
tem of the cockroach, Periplaneta americana, have
been studied by intracellular recording and
staining. Eleven pairs of local nonspiking inter¬
neurons were identified morphologically and charac¬
terized physiologically by their response properties
to low frequency airborne sound. 2. The interneurons
form two different morphological categories: those
having their dendritic fields on both side of the

ganglion (neurons 101 to 107) and those having their
dendritic fields on one side of the ganglion
ipsilateral to the cell body (neurons 201 to 204).

3. The stimulus of a low frequency sound modulated

by a sine wave (5-150Hz) and a Gaussian white noise
(cutoff frequency of 150Hz) was applied to the

cockroach from different directions. Over the

frequency band 20-100Hz the gain of neuron 101

approximates a 20dB/decade slope in gain in time
derivative system. Moreover, a first-order wiener
kernel of neuron 101 well matched the time deriv¬

ative of a pulse. The wave form of the kernel was

able to reverse when the stimulus was applied from
the opposite direction. At frequencies below 10Hz
there is a significant increase in gain. This may
correspond to a nonlinear component predicted from
the 2nd-order kernel. Thus, neuron 101 would encode
the time derivative of the signal (i.e., the dis¬
placement of the signal) in a frequency range of 20-
100Hz and the direction of the stimulus source.

PH 29

THE RELATIONSHIP BETWEEN FUNCTIONAL
DIFFERENCE OF STATOCYST INTERNEURONS AND
PROJECTION IN THE OPTIC LOBE.

H. Nakagawa and M. Hisada, Zool. Inst.,
Fac. Sci., Hokkaido Univ., Sapporo _

In the crayfish Procambarus clarkii, we found that
the positional and vibration information was carried by
two different groups of statocyst interneurons. The
position sensitive interneurons project to the optic
lobe, while the vibration sensitive ones do not. A
^descending interneuron (Bj) which reliably responded to
substrate vibration was identified in the
circumesophageal connective. The interneuron received
input from bilateral statocysts. The frequency response
of Bj was examined by artificial bending of statocyst
hairs at various frequencies and was compared with that
of the position sensitive descending interneuron Cj
(1982). Cj showed directional tonic response to the
stimulation at low frequency (<lHz), while Bj showed
reliable response with a fixed latency to the
stimulation at high frequency (>5Hz). The results
suggest that Cj and Bj carries different information,
positional and vibration information, respectively. The
morphology of Bj in the brain was revealed by combining
electrical stimulation of statocyst nerve and
intracellular injection of NiC^ followed by silver
intensification. Bj projects to the bilateral
deutocerebrum and the tritocerebrum ipsilateral to the
axon. However, unlike position sensitive descending
statocyst interneurons such as C j ( 1 987), Bj does not
project to the optic lobe. Several other brain
interneurons were also analyzed for their vibratory and
equilibrium sense and also tested for the presence of
projection in the optic lobe. Again position sensitive
ones were found to project to the optic lobe .while
vibration sensitive ones did not.

1198

Physiology

PH 25

IDENTIFIED LOCAL NONSPIKING INTERNEURONS
INVOLVED IN THE ABDOMINAL POSTURE CONTROL
SYSTEM.

T.Toga, M.Hisada. Zool. Inst., Fac. Sci.,
Hokkaido Univ., Sapporo.

We identified the local nonspiking interneurons
which facilitated the abdominal extension movement of
the crayfish, Procambarus clarkii. The interneurons
were located in the fourth abdominal ganglion. They had
bilaterally extending structure connected by a thick
process. These local bilateral interneurons (LBs) could
be further classified into two morphological types. One
type of LBs extended their dendrites in both
hemiganglia, and there were a few dendrites near the
midline of the ganglion. The other type had many
dendrites near the midline, and had asymmetric
distribution of dendrites in both hemiganglia. These
morphological types, however, could not be related to
any difference in their physiological characteristics.
Intracellular injection of depolarizing current into a
LB cell decreased the spontaneous spike discharge rate
of the slow extensor inhibitor in the second roots and
the slow flexor excitors in the third superficial roots
of the fourth ganglion. The result suggests that LBs
are involved in the abdominal posture control system.
In order to examine the membrane potential change of LBs
during the abdominal posture movement, we adopted the
dorsal whole animal preparation. During the fictive
abdominal extension movement, LBs showed continuous
depolarization. The LB cells may receive excitatory
input from extension command fibers, and facilitate the
extension movement by inhibiting the neuronal activity
operating antagonistically to the movement.

PH 26

FUNCTION OF LOCAL NONSPIKING INTERNEURON AS
THE LOW-PASS FILTER IN THE CRAYFISH MOTOR
CONTROL SYSTEM.

M.Takahashi , T.Takahata , M.Hisada1 and
K.Naka . ^Zool. Inst., Fac. of Sci., Univ.
of Hokkaido, Sapporo and ^Natl. Inst, for
Basic Biol., Okazaki. _

The signal transmission characteristics
between the premotor local nonspiking
interneurons and uropod motoneurons were
studied by intracellular injection of
sinusoidally or white-noise modurated
current into the interneuron neurites
through glass microelectrodes in the
terminal abdominal ganglion of crayfish.
The spike activity of motoneurons was
recorded extracellularly from their axons.
We found that the effect of current
injection became weaker with the increase
in current stimulus frequency. The cutoff
frequency was about 10Hz and the
attenuation constant was -10 db/oct,
indicating that interneurons would operate
as the low-pass filters. When the current
was injected into the motoneurons
themselves, the cutoff frequency was about
100Hz with the attenuation constant of
-5dB/oct. This observation indicates that
the motoneurons would also work as the low-
pass filters, but with the higher cutoff
frequency than the nonspiking interneurons.
The result stands with the idea that local
premotor nonspiking interneurons integrate
discrete post synaptic potentials into the
DC signal so that they can control motor
output continuously and in a graded way.

PH 27

CONNECTIONS BETWEEN MOTOR- AND LOCAL
NONSPIKING INTERNEURON

M.Sato1 and M.Hisada2. ^Biology, Rakuno
Gakuen Univ., Ebetsu, Hokkaido. 2Zool.
Inst., Fac. of Sci., Hokkaido Univ.
Sapporo.

The connection between a local
nonspiking interneuron ( LNSN ) and an
adductor exopodite motor neuron (Add MM) in
the 6th abdominal ganglion of the crayfish
was investigated using a double marking
method ( whole-mount silver intensifi¬
cation and HRP marking) for both light-
(LM) and electron microscopy (EM). LNSN
was marked with HRP, and Add MN with
silver. Then the resin embedded specimen
was sectioned 50pm thick. On the LM
observation of these two neurons, numerous
contacts of two neurons were found.
However on the EM, direct connection is
found to be scarce, and that synapses from
the LNSN to the Add MN were recognized
only two in the present study.
Electrophysiological ly, Add MN received a
signal from LNSN. This response of Add MN
was a low depolarizing change (l-2mV), when
a 5nA depolarizing current was injected
i n t r ace 1 1 u 1 ar 1 y to LNSN. The low response
appears to reflect the scarcity of synapse.
Alternatively, the depolarizing input to
LNSN may induce a inhibitory input onto Add
MN through other uncertain neurons.

PH 28

BEHAVIORAL SIGNIFICANCE OF A BURSTING OR
NON-BURSTING DISCHARGE PATTERN IN A
CRAYFISH ANAL MOTONEURON.

A. Muramoto, Fukushima Biomed. Inst, of
Envir. & Neopl. Pis., Fukushima.

An identifiable anal motoneuron (AML) of
the crayfish, Procambarus clarkii which is
capable of driving rhythmic anal contrac¬
tions can fire spontaneously in two differ¬
ent discharge patterns, bursting (phasic)
and non-bursting (tonic) . In this study,
the possible significance of these two
different firing states on behavior was
elucidated .

Simultaneous recordings of AML output
and anal rhythm in isolated abdominal
preparations proved that both bursting and
non-bursting patterns are able to generate
the same anal contractions under conditions
where there is no defecatory response. On
the other hand, intact, normally behaving
animals expel thick feces or a thin fecal
strand. In the former case, vigorous
contractions of the anus with big amplitude
could be seen and in the latter case, the
anus contracted weakly with almost constant
amplitude. Such powerful contractions
followed phasic firing in the AML, whereas
its tonic firing participated with weak
contractions .

These results provided evidence that the
different discharge patterns in the moto¬
neuron might actually influence different
behavior .

Physiology

1199

PH 29

NEUROANATOMICAL STUDIES OF OKN IN THE
MIDBRAIN OF THE MEDAKA, ORYZIAS LATIPES.

K. Aoki and Y. Kasuya, Life Sci. Inst.,
Sophia Univ. , Tokyo.

Electrophysiological studies concerning
visual events have presented the data in
the fish brain of medaka. 1) Fifty-six
movement-sensitive units were recorded in
the midbrain of the anesthetized medaka
stimulated by a rotating striped drum. In
the 56 units obtained, 24 units exhibited
directionally selective response. These
units were recorded mainly from around the
area pretectalis, the tractus opticus, the
nucleus of the posterior commissure, the
optic tectum, midbrain tegmentum, corpus
glomerulosum and immediately above the
area ventralis lateralis. 2) OKN (opto¬
kinetic nystagmus) were evoked by electri¬
cal stimulation in the midbrain. The
neurons evoked eye movement by electrical
stimulation were located on the dorsal
part of the midbrain: the area pretec¬
talis, around the posterior commissure,
the optic tectum and midbrain tegmentum.
3) Several kinds of eye movements were
evoked by electrical stimulation in the
midbrain. In 8 cases of evoked eye move¬
ments, horizontal conjugate movements were
observed in 6 cases. After the electrical
stimulation, cobalt-lysine complex was in¬
jected extracel lular ly at. the stimulus
sites. The process of neurons stained by
cobalt-lysine in the midbrain extended
into the forebrain and the cerebellum.

PH 30

VISUAL STIMULUS SELECTION AND ATTENTION IN THE
GOLDFISH ( CARAS SIUS AURATUS)

K. Ohnishi. Dept, of Physiol., Nara Medi. Uni.,
Kashihara _

The goldfish processes and stores only one aspect
(pattern) in the visual discrimination between the
compound stimuli containing heterogeneous aspects,
color and pattern (Ohnishi, 1987). The present
study was performed to examine whether such instru-
mentally guided visual stimulus (or aspect) selec¬
tion results from the fish's selective attention to
the aspect (pattern). It is thought that the fish
actively attends to the presented visual stimuli in
the discrimination process but does not in the rein¬
forcement process. Therefore, if the fish processes
and stores it in the discrimination process, atten¬
tion will actively take part in the stimulus selec¬
tion. Furthermore, since the goldfish's telen¬
cephalon is known as an attention or arousal center
during learning, it is presumed that telencephalic
ablation makes great effects on the stimulus selec¬
tion. Two runways training apparatus used in the
present study was designed to separate spatially
and temporally the presentations of the discrimi¬
native stimuli (DS) and the stimuli at reinforce¬
ment (SR). When the fish were trained in the case
that DS were the compound stimuli (green-vertical
vs blue-horizontal) and SR were the neutral stimuli
(gray), the fish showed no learned performances.
While, in the case that DS were the simple stimuli
(black vs white) and SR were the compound stimuli,
the normal fish showed learned performances but
the operated fish did not, and in memory tests,
the both groups could discriminate the compound
stimuli and pattern ones but not the color ones.
These data suggest that the stimulus selection may
be not mediated by the fish's selective attention.

PH 31

DESCENDING PATHWAYS TO THE BULBOSPINAL
REGION IN THE JAPANESE TOAD.

K.Takei1 , T. Matsushima 2 , Y.Oka2, M.Satou2
and K.Ueda2. ’Dept, of Physiol., Keio Univ.
School of Med., Tokyo and 2Zool. Ins., Fac.
of Sci., Univ. of Tokyo, Tokyo. _

As a step in elucidating the neural
network involved in the prey-catching
behavior in the Japanese toad, we studied
neuroanatomically the descending pathways to
the hypoglossal motoneurons which control
the tongue movements. We labeled
retrogradely origins of projections to the
hypoglossal motoneurons after the injection
of horseradish peroxidase and/or cobalt ic
lysine complex into the dorsomedial nucleus
(DMN) of the hypoglossal nucleus (XII) . The
labeled cell groups in the mesencephalon
were found in the pretectal, the tectal, the
subtectal and the tegmental regions. The
labeled cell groups in the rhombencephalon
consisted of the rhombencephalic reticular
formation and the nucleus ventralis nervi
octavi. Similar results were obtained after
the application of these tracers at the
level of the rostral spinal cord. These
findings indicate that the origins and
pathways of cells descending to the DMN of
the XII and the spinal cord are considerably
overlapped.

PH 32

QUANTITATIVE EM ANALYSIS OF THE SYNAPTIC
INPUTS TO THE TOAD VAGUS MOTONEURON
LABELED WITH COBALTIC LYSINE COMPLEX.

Y. Oka, M. Satou and K. Ueda. Zool . Inst.,
Fac. of Sci., Univ. of Tokyo, Tokyo.

As a morphological basis for the study
of the neuronal circuitry involved in the
sexual calling behavior of the toad, we
studied the synaptic organization of the
vagus motoneurons, which control the
laryngeal muscles (effector organ of
calling) , using a computer-aided
quantitative analysis of cobaltic lysine-
labeled neurons at the electron
microscopic level. The synaptic terminals
on the motoneurons could be classified
into 4 types, the Type S (containing
spherical and larger synaptic vesicles) ,
Type F (containing flat and smaller
vesicles), Type C (having subsynaptic
cistern) and Type G (having granular
vesicles). The percentage of the total
number of synapses was 63.3% (Type S),
33.9% (Type F), 1.3% (Type C) and 1.5%
(Type G) . The ratio of the occurrence of
the Type S (probably excitatory) synapses
to that of the Type F (probably
inhibitory) synapses increased as a
function of the distance from the center
of the vagus motor nucleus. The synaptic
density remained almost constant over the
whole neuronal somatodendritic membrane.
Origins of these synaptic terminals should
be clarified to know the functional
significance of this synaptic organization

1200

Physiology

PH 33

The retinoid composition and the structure
of the aeschnid compound eye before and
after eclosion.

T.Seki1, S.Fujishita1 and E.Eguchi2.

'Dept. Health Sci . , Osaka Kyoiku Univ.,
Osaka, 2 Dept . Biol., Yokohama City Univ.,
Yokohama . _

A compound eye of an adult true dragonfly
is composed of two regions: dorsal large
facet region (LFR) and ventral small facet
region (SFR) . The LFR contains only

retinal and the SFR both retinal and 3-OH
retinal; the 3-OH retinal is predominant
in the SFR. Lew (1933) has reported that
the LFR is formed during the eclosion from
the non-functional tissue which occupied
the posterior corner of the nymphal eye
(X-tissue) , while the nymphal compound eye
is retained at the posterior portion of
the SFR of the adult aeschnid eye. The
retinoid compositions in the functional
nymphal eye and the X-tissue were then
examined separately. The former contained
both retinals at the molar ratio similar
to that in the SFR of the adult eye, but
the latter no retinal at all. Retinoids
contained in the X-tissue were retinyl and
3-OH retinyl esters, which were 1 1 -cis
isomers mainly.

Rhabdomal structures were observed by an
electron microscopy. The diameter of the
rhabdom in the functional nymphal eye (~ 5
M.m) was larger than that in the adult LFR
3 p.m) . The X-tissue was also observed
and it was found, against our
expectations, that the rhabdom has already
been formed; the diameter was about 1 ^m .
The rhabdom in the X-tissue is supposed to
have contained no visual pigments.

PH 3 4

RETINOID METABOLISM IN THE COMPOUND EYE OF

DROSOPHILA - FORMATION OF 1 1 -CIS 3-

HYDROXYRETINOL WITH BLUE-LIGHT.

T. Ishizaki, A. Terakita and K. Ozaki.
Dept, of Biol., Fac. of Sci., Osaka Univ.,
Toyonaka , Osaka 560. _ _

When the fruit fly. Drosophila, is kept

in the dark, all-trans 3 -hydroxyret inal
(AT-3-OH RAL) which keeps away from opsin
is accumulated in their eyes. Recently,
we found that, by exposing a fly to blue-
light, the amount of accumulated AT-3-OH
RAL was greatly decreased accompanied by
the formation of 11 -cis 3-hydroxyretinol
(11 -3-OH ROL). When the fly was returned
to the dark again, thus formed 1 1 -3-OH ROL
disappeared. We therefore expected that
11 -3-OH ROL might be a direct precursor of
11 -3-OH RAL, a chromophore of rhodopsin.
In order to examin this point, we prepared
the fly which lacks chromophore of visual
pigment but maintains the accumulation of
AT- 3 -OH RAL. With blue light, AT-3-OH RAL
in this fly was converted into 11 -3-OH
ROL, as was seen in the normal fly. Dur¬
ing the subsequent dark period, 11 -3-OH
ROL decreased and 1 1 -3-OH RAL increased
instead. All those changes followed stoi¬
chiometric relationships. The finally
formed 11 -3-OH RAL was shown to be a rhod¬
opsin chromophore, since it was changed to
all-trans and backed to 1 1 -cis with blue
and red light respectively. In conclu¬
sion, AT-3-OH RAL stored in the eye of
Drosophila is first converted to 1 1 -3-OH
ROL with blue light, and then into 1 1 -3-OH
RAL to be a chromophore of visual pigment.

PH 35

3 -HYDROXYRETINOL IN THE RETINA WITH
RHODOPSIN-PORPHYROPSIN PIGMENT SYSTEM.

T. Suzuki Dept, of Pharmacol., Hyogo Coll,
of Med., Nishinomiya, Hyogo 663.

The crayfish, Procambarus clarkii, has
both rhodopsin and porphyropsin in the
retina. The proportion of porphyropsin
varies with seasons, being at a high level
in winter and a low level in summer. 3,4-
Didehydroretinyl ester, which is found in
the retina of winter crayfish, is probably
the chromophore precursor of porphyropsin.

We found a new retinoid in the crayfish
retina which had a high proportion of
porphyropsin. This retinoid was identified
to be 3-hydroxyretinol from its UV-spectrum
and fluorescence. The 3-hydroxyretinol was
present exclusively in 11 -cis configuration
and esterified form. This compound accoun¬
ted for 3-9% of total retinyl esters in the
retina. 3-Hydroxyretinal , which is known
as a visual pigment chromophore in insects,
has not been detected in the crayfish. When
the animals were kept under 25°C,LD condi¬
tions, both 3-hydroxyretinol and dehydro¬
retinol disappeared. Dehydroretinol and
3-hydroxyretinol accumulated in the retina
when the animals were kept under 10°C,DD
conditions. These results suggest that the
3-hydroxyretinol is an intermediate in the
conversion of retinol to dehydroretinol and
that the low temperature stimulus triggers
the synthesis of 3-hydroxyretinol.

PH 36

HISTOCHEMICAL LOCALIZATION AND CHROMOPHORE
OF PHOTOPIGMENTS IN MULTIPLE PHOTORECEPTIVE
SYSTEM IN ONCHIDIUM VERRUCULATUM
Y. Shimatani 1 , N.Katagiri?, Y.Katagiri1. and
T. Suzuki3 , 'Dept. of Physiol., 2Dept. of
Anat., Tokyo Women's Med. Coll., Tokyo,
3 Dept. of Pharmacol., Hyogo Coll, of Med.,
Nishinomiya

The localization of rhodopsin and
r;etinochrome in the stalk eye (SE), dorsal
eye (DE) and dermal photoreceptor cells
(DP) of marine gastropod Onchidium was
investigated using a histochemical
fluorescence technique (Ozaki et al., Cell
Tissue Res., 233, 1983) and a microscopic

high speed video system with high sensitive
image intensifier (NAC, MHS-200).

Retinochrome was localized in the
photoreceptor cell cytoplasm of SE and DP,
which were occupied by a large mass of
photic vesicles. Rhodopsin was localized in
the microvillar layer of SE. Both pigments
could not be demonstrated in DE by this
method. They may not be reduced: to
N-retinyl protein.

Chromophore of photopigment in SE and
DE was extracted as retinal oximes after
dark-adaptation, then analysed using high
performance liquid chromatography. All-
trans- and 1 1 -cis-retinal were detected in
both SE and DE. Total amount of retinoid
was 0.41 pmol/SE in average (n=17) and
0.17 pmol/DE (n=50). Proportion of 1 1 -cis-
and all-trans -isomers was 41% and 59% in
SE, 60% and 40% in DE.

Physiology

1201

PH 37

DOES CEPHALOPOD HAVE COLOR VISION ?
LOCALIZATION OF THREE VISUAL PIGMENTS IN'
THE RETINA OF Watasenia scint illans

1Y. Kito^M. Seidou^S. Matsui, 2M.
Michinomae and3K. Yoshihara. 1Dept. Biol.,
Osaka Univ. Toyonaka , 2Dept . Biol., Konan
Univ., Kobe, 3SUNBOR, Mishima-gun. Osaka.

Three visual pigments were found in the

eye of the firefly squid, W. scintillans
(1). The chromophores of major pigment
( Amax: 484 nm) , second one (Amax: about
500 nm) and third one (A max: about 470
nm) were retinal (A1 pigment), 3-dehydro-
retinal (A2 pigment) and 4-hydroxyretinal
(A4 pigment) (2). In this paper, it was
determined by microdissection and HPLC
analysis that A2 pigment located at the
proximal area of the long rhabdome of
ventral retina and A4 pigment at the
distal area. A1 pigment was not found in
the long rhabdome. Ultrastructure of

the long rhabdome revealed a complex of
two cells which suggested that A2 and A4
pigment contained each in different
photoreceptor cell. Primary structures of
retinal binding site of three pigments
were identical ( 3 ) . Extraocular photo¬
sensitive vesicles contained only retinal
(possibly A1 pigment). Hepatopancreas

contained only 11-cis retinylester . Eggs
contained retinol and a little amount of
4-hydroxyretinol . These facts suggest
that 3-dehydroretinal and 4-hydroxy¬
retinal were synthesized in the specific
photoreceptor cells of the retina.
(1) S. Matsui et al., 1988, J. Gen.
Physiol., 92 , 55-66 . (2) S. Matsui et

al. , 1988, Biochim. Biophys. Acta, 370-

374 (3) M. Seidou et al., 1988, Biochim.

Biophys. Acta, in the press.

PH 39

X-RAY DIFFRACTION OF THE LIVE RETINA OF
THE FIREFLY SQUID. WATASENIA SCINTILLANS
M. Seidou1, Y.Kito , T . Hamanaka2 , JT!

7 o i

Wakabayashi and Y.Amemiya . Dep. of

O

Biol., Osaka Univ., Toyonaka, Dep. of
Biophs. Egn. , Osaka Univ., 3PF, Lab. of

High Energy Phys., Tsukuba. _

The specific area of ventral retina
locating just behind a row of eye
photophores have very long outer segments
(~500 /zm) and few pigment granules are
spread in the outer segment layer. These
features of the firefly squid retina give
a favarable sample for x-ray diffraction.
By using conventional apparatus for x-ray
analysis, it took more than ten hours to
get a diffraction pattern from
glutaraldehyde fixed retina. In this
paper, by use of synchrotron radiation
at Photon Factory and the imaging plate
(Fuji film, Tokyo), it was made possible
to get the pattern from the live retina
within five minutes exposure.

The experiment was carried out by
using the mirror-monochrometer optics
(BL-15A1) at Photon Factory. The

specific area of the retina was cut into
a piece (3mm x 1mm) and measured in the
dark. The diffraction pattern recorded
on the Imagining Plate showed reflection
spots from hexagonal lattice with about
570 A. unit cell constant at low angle.
This means that rhabdome is made of
microvilli arranged in hexagonal lattice.
Diffused spots existed at the vertexes of
a hexagon corresponding to spacing of
about 40 A. This pattern probably
originates from internal structure of the
microvilli membrane. In both diffraction
patterns, x-ray intensity of diffraction
spots along vertical axis were stronger
than that of ones along other axes.

PH 38

LIGHT AND ELECTRON MICROSCOPIC INVESTIGATION
OF CELLULAR ARCHITECTURE IN THE RETINA OF
THE FIREFLY SQUID, WATASENIA SCINTILLANS
'H.Masuda, 1 M. Michinomae, ZM. Seidou andzY.

Kito l)Dep. of Biol., Fac. of Sci., Konan
Univ., Kobe Hyogo. 2)Dep. of Biol., Fac. of
Sci., Osaka Univ., Toyonaka Osaka.

The cellular architecture of the retina was
observed with light and electron micro¬
scopes. The rhabdome in the small area of
the ventral retina was more than two times
as long as in the other parts of retina.
Histological analysis of the ventral retina
revealed that two cell types, long retinula
cells (500-600 pm) and short retinula cells
(100-200 pm), were found in there and it

represented mosaic of both long and short
retinula cells. There also appeared to be
two types of retinula outer segment (OS).
Type 1 were long, while Type 2 were short OS
Type 1 OS well developed from basement

membrane to distal end of the retina. Type 2
occupied the proximal 200 pm of OS. The

microvilli of a single OS form two parallel

rows oriented perpendiculaly. The micro¬
villi of Type 1 OS appeared only in distal
400 pm area. On the other hand, micro-
dissection and HPLC analysis revealed that
3-dehydroretinal (A2) located at the proximal
area of the ventral retina and 4-hydroxy¬

retinal (A4) at the distal area. As a result,
A4 and A2 pigment were synthesized in Type 1
and Type 2 OS, respectively. Therefore, the
firefly squid is supposed to be capable of
color vision.

PH 40

MOLECULAR STRUCTURE OF SQUID RETINAL-
BINDING PROTEIN.

K.Ozaki^, A.Terakita1 1 , R.Hara1 , T.Hara1 ,
M.Ozaki , M. Nishimura3 , I . H. -Nishimura3 and
H.Mori . 'Dept. Biol. , Fac. Sci. , Osaka Univ.,
Toyonaka, 2Dept.Biol., Col . Gen. Educ . , and
3Dept. Biol. , Fac. Sci. , Kobe Univ. , Kobe and
4Meiji Inst. Health Sci., Odawara. _

From the squid retinal cDNA library,

clones which carried the cDNA inserts
(^2.8 kbp) covering the complete coding
region of the squid retinal-binding pro¬
tein ( RALBP ) mRNA were isolated. In the
cDNA, only one open reading frame (1032
bp) was long enough to encode the entire
RALBP polypeptide and was located upstream
of an extensive nontranslated region at 3'
end. Northern hybridization analysis
revealed that the cDNA prove hybridized
with a single retinal RNA species about
2800 nucleotides long, suggesting that the
cDNA is originated from the intact mRNA of
RALBP. The primary structure of the pro¬
tein was deduced from the nucleotide se¬
quence and verified with the amino-acid
compositions of polypeptides derived from
the purified RALBP. The translation prod¬
uct was highly hydrophilic and had an
abundance of acidic amino-acid residues.

A hydrophobic region ( 'v 20 amino-acid
residues) was found in the middle of the
polypeptide chain, where high possibility
of (3-sheet structure was also suggested.
Since this region reveals week homology
to the bovine cellular retinal -binding
protein, we consider that the structure
might be essential to retinal binding.

1202

Physiology

PH 41

AMINO ACID SEQUENCE OF OCTOPUS RHODOPSIN
AND ITS HOMOLOGY TO OTHER VISUAL PIGMENTS
M. Tsuda1 , N. G. Abdulaev2 , and T. Miyata3 .
^ept. of Phys., Sapporo Medical College,

2 Shemyakin Inst. of Bioorganic. Chem.,
Moscow, USSR, 3Dept. of Biol., Kyushu Uni¬
versity, Fukuoka

We determined the nucleotide sequence
of cDNA encoding octopus rhodopsin and on
the primary structure of this protein.
Amino acid sequence of octopus rhodopsin
was compared with those of other visual
pigments. Octopus rhodopsin, like mamalian
and Drosophila pigments, has lysine (306)
corresponding to position 296 of bovine
rhodopsin, the site of covalent attachment
of 11-cis retinal. The polypeptide chain
is divided into several alternating hydro-
phobic and hydrophilic sequences, the lat¬
ter being much more extended than those in
bovine and human opsins. A characteristic
feature of the octopus rhodopsin polypep¬
tide chain is the unexpected long C-termi-
nal adjacent to the last hydrophobic
strech. This tail contains a large number
of prolines and glutamines clustered into
eleven blocks of repetitive sequences of
Tyr-Pro-Pro-Gln-Gly . Octopus rhodopsin was
shown to interact with mamalian G-protein.
This implies that a binding site for G-
protein should be conserved in both inver¬
tebrate and vertebrate rhodopsin. Close
homology of cytoplasmic loops in visual
pigments is indicative of G-protein inter¬
acting with one or more of these loops.

PH 43

VISUAL PIGMENTS OF LAMPREY

O.Hisatomi , T.Iwasa"' , F.Tokunaga”' , M.
Ishikawa2 and A.Tonosaki2. 1 Dept, of
Phys., Fac. of sci., Tohoku Univ. , Sendai,
2Dept. of Anat. , Yamagata Univ., Sch. of
Med., Yamagata .

The similarity in the amino acid

sequence among visual pigments in human
suggests that there was a common ancestor.
It is important to clarify when rhodopsin
and color pigments differentiated from this
protein. Lamprey, which is one of the most
primitive vertebrates, is the most suitable
object for studying the origin of the color
pigments .

The visual pigments of lamprey was
purified by Con A column chromatography.
The purified pigment had the absorption
maximum at 503 nm and had high affinity to
anti-bovine rhodopsin antibodies, suggest¬
ing that it was the pigment in the short
cell and was regarded as lamprey rhodopsin.
The lamprey rhodopsin was slowly bleached
in the presence of 100 mM NH2OH in the
dark, but did not show chloride effect.

Genomic DNA was extracted from lamprey
livers. The extracted DNA was digested by
restriction enzymes (BamHI, EcoRI , Hindlll,
PstI) and analyzed by Southern blot
hybridization with cDNA's of bovine
rhodopsin and human red and blue visual
pigments as probes. The patterns of the
bands of DNA fragments hybridized by the
probes were quite different among the used
probes, suggesting that lamprey has
homologous genes to those of rhodopsin and
blue and red visual pigment genes.

PH 42

PH 44

METABOLISM OF RETINOIDS' DURING DEVELOP¬
MENTS OF XENOPUS LAEVIS AND RHACOPHORUS
ARBOREUS.

M. Azuma, T. Seki and S. Fujishita. Dept,
of Health Sci. , Osaka Kyoiku Univ. , Osaka.

Retinoids in the eggs and larvae of
Xenopus and Rhacophorus were analysed by
high-performance liquid chromatography
( HPLC ) . Xenopus eggs included all-trans
retinal and 3-dehydroret inal (1:1). The
retinals were converted to all-trans
retinyl ester and 3-dehydroret inyl ester
during development until stage 46. The
proportion of all-trans 3-dehydroret inal
scarcely changed, but that of all-trans 3-
dehydroret inyl ester increased after stage
42. 1 1-Cis retinal and 3-dehydroret inal
emerged in the eyes at stage 40 (1:1).
The proportion of 1 1-cis 3-dehydroret inal
increased after stage 43. Rhacophorus
eggs mostly contained all-trans retinal,
which was converted to all-trans retinyl
ester during development. 1 1-Cis retinal
appeared in the eyes at stage 33, but 11-
cis 3-dehydroret inal did at later stages.
These larvae also contained all-trans 3-
dehydroret inyl ester. 1 1-Cis retinal and
3-dehydroret inal must be supplied from
all-trans retinyl ester and 3-dehydro¬
ret inyl ester via retinol and 3-dehydro-
retinol. The result that the proportions
of 1 1-cis 3-dehydroret inal are generally
greater than those of all-trans 3-dehydro¬
ret inyl ester, suggests that 3-dehydro-
retinol is more efficiently used than
retinol for visual pigment chromophore.

PURIFICATION OF A CONE VISUAL PIGMENT,
IODOPSIN, AND MONOCLONAL ANTIBODIES TO IT.
T.Yoshizawa1 , Y.Shichida1, Y.Fukada1
Y.Taniguchi1 , O.Kuwata1, M.Takeichi1
and S.Horiuchi2.

^ept. of Biophys., Fac. of Sci., Kyoto
Univ., Kyoto and 2Dept. of Biol., Col. of
Gen. Educ. , Osaka Univ., Toyonaka, Osaka.

Spectroscopic studies of iodopsin have
been carried out using a crude extract of
iodopsin in the presence of rhodopsin or
its opsin. For biochemical investiga¬
tions, it is essential to separate
iodopsin from rhodopsin or its opsin. We
first purified iodopsin, to which mono¬
clonal antibodies (mAbs ) were prepared.

Membrane proteins extracted from outer
segments of chicken visual cells with 2%
digitonin were subjected to a Con A- and
a DEAE-Sepharose columns to separate iod¬
opsin from rhodopsin and other proteins.

The iodopsin thus purified was used for
immunization and screening, and four kinds
of mAbs were obtained. Immunoblotting and
immunoprecipi tat ion- spectroscopic analyses
revealed them to be highly specific for
iodopsin. Immunohistochemical studies
proved that iodopsin was localized at a
majority of the cone outer segments of
chicken retina.

The absorption maximum of iodopsin was
shifted about 20nm to the blue in the
presence of one of the mAbs, showing that
the binding of the mAb induced a conforma¬
tional change of iodopsin.

Physiology

1203

PH 45

LOCALIZATION OF IODOPSIN IN THE RETINA
AND THE PINEAL OF CHICKENS AND QUAIL.

T. Oi^hi , A. Kawata^ ■£> . Kojima , Y.
Fukada^ and T. Yoshizwa , Dept. of Biol.,
Fac . of Scj., Nara Women's University,
Nara, and ^Dept. of Biophys., Fac. of
Sci., Univ. of Kyoto, Kyoto.

Localization of iodopsin in the retina
and the pineal of chickens and Japanese
quail was studied immunohistochemically
using monoclonal antibodies against chick¬
en iodopsin. Paraffin sections (4pm) were
stained with iodopsin monoclonal anti¬
bodies with avidin-biotin system (Vector) .
In both chickens and quail, outer segments
of cones (double cones and some single
cones) were stained, but those of rods and
some single cones were not stained. To
identify types of cones by the color of
oil droplets, we made whole mount prepara¬
tions of the retina fixed in 10% formalin
with 0.3% Triton X-100. Cones with green
oil droplets (double cones) and red oil
droplets (single cones) were positively
stained, but cones with yellow, colorless
and blue oil droplets were not stained.
In the pineal body, degenerated outer
semgments of pinealocytes were stained.
But intensity of staining was less in the
pineal than in the retina, indicating less
amount of iodopsin exists in the pineal.
Localization of iodopsin in the brain is
now under investigation.

PH 46

ISOLATION OF A cDNA CLONE USING
MONOCLONAL ANTIBODIES TO CHICKEN IODOPSIN.
O.Kuwata, Y.Fukada, Y.Shichida,
T.Yoshizawa, K.Yasuda and Y.Shimura.
Dept, of Biophys., Fac. of Sci.,

Kyoto Univ., Kyoto.

Among cone pigments for color vision,
only chicken iodopsin, a red-sensitive
one, has been considerably characterized
by biochemical and spectroscopic studies.

As the first step of investigation into
the relationship between the structure and
the physiological function of iodopsin in
a cone cell, we have prepared monoclonal
antibodies (mAbs) against purified
iodopsin. They are so specific for
iodopsin as to enable us to isolate a cDNA
of iodopsin.

Chick retina has been confirmed to
contain iodopsin by immunoblotting
analysis. Double-stranded cDNAs for
poly(A)+ RNAs of the chick retina were
synthesized, and cloned into the
expression vector, Agtl 1 to construct a
cDNA library of the chick retina. Plaques
bearing the products of each expressed
cDNA were screened using the iodopsin-
specific antibodies as a probe.

A northern blotting analysis revealed
that one of the positive cDNA clones
specifically hybridized with an mRNA
(about 2.2kb) in the chick retina.
Analyzed sequence of the cDNA clone
(562bp) suggested that it might be
consisted of two smaller cDNA fragments.

PH 47

EXTRACTION AND PURIFICATION OF FOUR KINDS
OF CHICKEN CONE PIGMENTS
T.Okano, Y.Fukada, I.D. Artamonov and
T.Yoshizawa .

Dept. of Biophys., Fac. of Sci.,
Kyoto University, Kyoto.

In order to elucidate the transduction
process and the regulation mechanism in
cone cells which are responsible for color
vision, we first attempted to isolate all
the visual pigments in separate forms.

By use of CHAPS in the presence of
phosphatidylcholine, a mixture of rhod-
opsin and cone pigments was solubilized
from outer segments of chicken visual
cells. The mixture was successively chro¬
matographed on Con A- and DEAE-Sepharose
columns. Then we obtained fractions of
rhodopsin (96% purity), iodopsin (95%),
green pigment (74%), and a mixture of blue
and violet pigments with admixture of iod¬
opsin. By means of partial bleaching, ab¬
sorption maxima of the difference spectra
between the pigments and retinal-oxime
were estimated as follows; rhodopsin: 503
nm, iodopsin: 571 nm, green: 508nm, blue:
455nm, violet: 425nm. When the spectra of
five kinds of chicken visual pigments were
plotted on a scale of "fourth root of
wavelength", they were in good agreement
in the long wavelength side with each
other except for violet pigment. Then the
absorption maximum of violet pigment was
estimated at 41 5nm from a spectrum of the
pigment corrected by other pigments.

PH 48

THE PHOTOBLEACHING PROCESS OF IODOPSIN
IN THE PRESENCE OF NITRATE ION
Y.Imamoto, T.Okano, H.Kandori, Y. Fukada,
Y.Shichida and T.Yoshizawa
Dept, of Biophys., Fac. of Sci.,
Kyoto Univ. , Kyoto

There are many differences in chemical
nature or photochemical behaviour between
iodopsin and rhodopsin. One of the
noticeable differences is the thermal
reaction of bathointermediates produced by
irradiation at liquid nitrogen tempera¬
ture; bathoiodopsin was converted to
original iodopsin, while bathorhodopsin
was converted to a next intermediate,
lumirhodopsin. Another is the dependency
of the absorption spectrum of iodopsin on
the concentration of Cl-. The present
paper reports some differences in photo¬
chemical behaviour at low temperature
between original iodopsin (Cl- binding
form) and N-iodopsin (prepared by
substitution of Cl- to NO^-).

On irradiation at -T85°C, both of
iodopsin and N-iodopsin were converted to
batho-intermediates . Prolonged irradi¬
ation of each pigment produces a photo¬
steady state mixture composed of original
pigment, batho-, and iso-pigments. Most
of bathoiodopsin was thermally reverted to
iodopsin, while N-bathoiodopsin was
converted to the next intermediate and
bleached at room temperature. These
results suggested that the reversion of
bathoiodopsin to iodopsin requires Cl-.

1204

Physiology

PH 49

PRIMARY INTERMEDIATE OF CHICKEN IODOPSIN
AT ROOM TEMPERATURE.

H.Kandori, Y.Imamoto, T.Okano, Y.Fukada,
Y.Shichida and T.Yoshizawa.

Dept, of Biophys., Fac. of Sci.,
Kyoto University, Kyoto.

Picosecond laser photolysis of chicken
iodopsin was carried out at room tempera¬
ture. Immediately after excitation, a
bathochromic photoproduct was observed,
which was stable on picosecond time scale.
This may correspond to "bathoiodopsin" ,
which was discovered by low temperature
spectrophotometry (Yoshizawa and Wald,
1967). Our finding suggests as follows :

(1) Bathoiodopsin is the primary inter¬
mediate under physiological temperature.

(2) The primary photochemical event of
iodopsin is an isomerization of the
chromophore .

(3) The primary photochemical reaction
would be faster or simpler in iodopsin
than in rhodopsin, because no precursors
of bathoiodopsin were observed within our
time resolution (< 20 psec) and no appar¬
ent absorption change due to an intense
excitation was observed in iodopsin.

(4) The relaxation process of batho¬
iodopsin at room temperature is different
from that at low temperature. Batho¬
iodopsin seems to proceed to bleaching at
room temperature, while bathoiodopsin at
low temperature reverted to the original
pigment (iodopsin) on warming.

PH 50

ACTIVATION OF GTP-BINDING PROTEIN BY A
PHOTOBLEACHING INTERMEDIATE OF IODOPSIN.
Y.Fukada, T.Okano, I .D. Artamonov and
T.Yoshizawa. Dept, of Biophys., Fac. of
Sci., Univ. of Kyoto, Kyoto.

Electrophysiological studies have
suggested that the visual transduction
process in cone cells is similar to that
in rods. For complete understanding of the
molecular mechanism in cones, activation
of GTP-binding protein (transducin: T) by
a photobleaching intermediate of a cone
visual pigment was investigated.

One of the cone pigments, iodopsin was
purified from chicken retinas, and then
reconstituted into phosphatidylcholine
liposomes. Since we have not succeeded to
isolate a cone-specific T, we purified T
from bovine retinas in which rod cells are
predominant. Measurements of [^HJGppNHp
(GTP analogue) binding to T in the
presence of the reconstituted iodopsin
showed that T was remarkably activated
only when iodopsin was irradiated. The
degree of the light-dependent activation
was similar to those observed in the
presence of bovine or chicken rhodopsin.
These observations indicate that the
active site of the visual pigment, which
is responsible for the activation of T, is
conserved in iodopsin, and suggest that
the first step of the visual transduction
process in a cone cell should be the
activation of T by a photobleaching
intermediate of the visual pigment.

PH 51

THREE DIMENSIONAL ARCHITECTURE OF THE
CYTOSKELETON IN THE RETINAL PIGMENT
EPITHELIUM.

Jiro USUKURA, Dept, of Anat. Nagoya Univ. Nagoya
The Retinal pigment epithelium ( RPE ) supports in a
crucial way the life of neural retina through many
complex functions such as phagocytosis and pigment
migration. Cytoskeletal mechanisms play strategic
roles in these processes. The purpose of current
study is to reveal three dimensional architecture of
the cytoskeleton by freeze-etching using rapidly
frozen frog retina and to identify its components by
immunocytochemistry. The cytoskeleton of RPE was
mainly composed of actin filaments and vimentin.
Actin filaments were mostly localized at the apical
region and beneath the lateral membrane of the
junctional area. In apical microvilli, many actin
filaments extended longitudinally as making bundles.
Large globular substances were attached on actin
filaments, and frequently linked them each other.
Actin filaments were closely associated with
pigment granules and microvillar membrane. On the
other hand, vimentin was localized predominantly in
basal regions of the cell body. Myosin and -actinin
were also detected by immunoassay, though it was
difficult to demonstrate their morphological
counterparts because of dispersed localization. No
prominent morphological changes in the cytoskeleton
were induced by light conditions.

PH 52

THE INFLUENCE OF MONOCHROMATIC STIMULI TO
THE BRAIN ON VISUALLY INDUCED PHOTOTACTIC
BEHAVIOR OF ORB WEAVING SPIDERS.

S. Yamashita. Biol. Lab., Kyushu Inst, of
Design, Fukuoka.

On a dark background, tethered orb
weaving spiders^ Argiope amoena and
Nepbrila clavata , which walked on a y-maze
globe showed a negative phototaxis to
light aimed at the eyes. When the brain
was illuminated through the cuticle
covering it, the spiders showed a positive
phototaxis to light aimed at the eyes
(Yamashita and Tuji, 1987). The spectral
effectiveness of monochromatic lights
shown on the brain in changing phototactic
behavior from negative to positive was
tested with various wavelengths. The
spectral effectiveness function was found
to have a maximum at about 420-440 nm.

The photosensitive neurons in the brain
reported by Yamashita and Tateda (1983)
have a maximum sensitivity at about 420-
440 nm, suggesting that the cerebral
photosensitive neurons control the visual
phototactic behavior of the spiders.

Physiology

1205

PH 53

STRUCTURE OF THE COMPOUND EYE AND THE
OPTIC NEUROPIL OF THE TIGER BEETLE
(CICINDELA CHINENSIS).

Y. Toh and A. Mizutani. Dept, of Biol.,

Fac . of Sci., Kyushu Univ., Fukuoka. _

The visual system of the adult tiger
beetle consists of ommatidial layer and
underlying three optic neuropils (lamina,
medulla, and lobula) . Each ommatidium
contains eight retinular cells, which are
classed based upon the level of their
rhabdomeres into one distal cell, six
proximal cells and one basal cell.
Rhabdomeric microvilli of the distal
cells, four proximal cells and the basal
cell align vertically, whereas those of
two proximal cells align horizontally. No
twist of rhabdoms occur in the dorsal area
of the compound eyes. Eight retinular
axons from each ommatidium project into an
optic cartridge in the lamina, where they
synapse with several lamina monopolar
neurons. In addition to afferent synapses
reciprocal synapses between retinular
axons and between monopolar neurons as
well as feedback synapses from monopolar
neurons upon retinular axons occur.
Dendrites of most lamina monopolar neurons
are limited in the individual cartridge,
but some monopolar neurons extend
horizontally long collaterals. Moreover,
dendrites of several medulla neurons
widely spread in the lamina. These
structures are discussed in relation to
possible visual function of this insect.

PH 54

MOVEMENT SENSITIVE NEURONS IN THE LOBULA
OF THE VISUAL SYSTEM OF THE TIGER BEETLE
(CICINDELA CHINENSIS ) LARVA.

A. Mizutani and Y. Toh. Dept, of Biol.,

Fac. of Sci., Kyushu Univ., Fukuoka. _

The larva of the tiger beetle possesses
six ocelli on either side of the head, two
of which (St.l and St . 2 ) are larger than
the others. St.l orients dorso-
posteriorly, whereas St. 2 orients dorso-
anteriorly. Two layered optic neuropils
(lamina and medulla) occur beneath each
ocellus. The optic neuropils of St.l and
St. 2 occur side by side, and they
converge upon the common third optic
neuropil (lobula). In the present study
responses of visual interneurons in the
lobula to moving objects over the head
have been intracellularly recorded, and
the recorded cells have been
morphologically identified by ionophoretic
fluorescent dye injection. Most of
recorded cells extend distally dendritic
processes towards the medulla, and
proximally a single axon towards the
brain. The cells exclusively responding to
transverse movement of the object over the
head possess extensive dendritic
arborizations in only one of two medullae,
which suggests movement detection by one
ocellus. The cells exclusively responding
to antero-posterior movement of the object
possess extensive dendritic arborizations
in both St.l medulla and St . 2 medulla,
suggesting co-operation of the two (St.l
and St. 2) ocelli for movement detection.

PH 55

PARTICIPATION OF THE COMPOUND EYES IN THE
SPECTRAL SENSITIVITY OF THE OCELLI OF A
NOCTUID MOTH.

S. Yamazaki1, S. Yamashita2 and M.
Sasaki1. i-Lab. of Entomol., Fac. of Agr . ,
Tamagawa Univ., Tokyo and 2Biol. Lab.,
Kyushu Inst, of Design, Fukuoka.

The spectral sensitivity of the
ocelli in the noctuid moth, Anadevidia
peponis , was investigated by recording
ERG ' s . Under conditions of constant
darkness, the peak sensitivities were
observed at about 340 nm in the ultra¬
violet and at about 480-540 nm in the
visible. On a light background, the
sensitivity to each monochromatic light
decreased, as compared with the sensitiv¬
ity for the dark-adapted ocellus. The
decrease in sensitivity was large in the
visible, but small in the ultraviolet.

The decrease in the retio of visible to
ultraviolet sensitivities ( v I s/uv) under
the light was not observed in moths
whose compound eyes were removed or whose
ocellar nerves were cut. On the other
hand, when a small light spot for back¬
ground illumination was presented to only
a portion of a compound eye, VIS/UV for
the dark-adapted ocelli decreased. These
observations show that the decrease in
ocellar VIS/UV under light is mediated by
signals from the compound eyes.

PH 56

HISTOLOGICAL IDENTIFICATION OF THE RED
RECEPTOR AND ITS CENTRAL PROJECTION IN THE
CABBAGE BUTTERFLY'S COMPOUND EYE
M. Shimohigashi and Y. Tominaga. Biol.
Lab., Fac. Sci., Fukuoka Univ., Fukuoka

The ommatidium of the compound eye of a
cabbage butterfly, Pieris rapae , has nine
retinula cells (Rl-9). The aim of this
study is to clarify which type of these
retinula cells undertakes the red color
reception. By intracellular recording,
five different types of spectral
sensitivity curves were obtained. Each
curve had a peak at near 340 nm (UV) , 420
nm (violet) , 480 nm (blue) , 560 nm (yellow-
green) and 620 nm (red), respectively.
Intracellular dye injection with cobalt-
lysine clearly indicated that the basal
retinula cell (R9) was sensitive to red
color, and in addition, one distal cell
( R 5 ) to UV and another (R3) to yellow-
green. The electron microscope observation
of tissues partially injected with HRP gave
the following results concerning the
central projection of retinula cells. (1)
The red receptor extended the long fiber
to the medulla. (2) The UV receptors also
had the long fibers, which possessed en
passant synapses with the lamina neurons.
(3) The violet, blue and yellow-green
receptors had the short fibers, which
terminated at the lamina cartridges.

1206

Physiology

PH 57

HISTAMINE AS A PHOTORECEPTOR TRANSMITTER
IN THE COCKROACH ( PERIPLANETA AMERICANA)
OCELLUS

J. T. LIN, M. MIZUNAMI, Y. TOH and
H. TATEDA Dept, of Biol., Fac . of Sci.,
Kyushu Univ., Fukuoka. _

To identify the neurotransmitter at the
synapses from retinular axons to the
second-order neurons ( L-neurons ) in the
cockroach ocellus, the effects of serval
putative neurotransmitters ( GABA, ACh and
histamine ) of insect photoreceptors on
L-neurons were examined. Responses of
L-neurons to application of putative
transmitters were intracellularly
recorded. Ionophoretically applied
histamine induced a dose-dependent
hyperpolarization as did the ocellar
illumination, whereas GABA and ACh had no
significant effects. By perfusion,
histamine hyperpolarized the L-neurons,
and all the tested histamine antagonists
( Hi: promethazine, mepyramine; H2 :
cimetidine, ranitidine ) completely,
reversibly inhibited the photoresponses of
L-neurons except cyclizine. Curare is
well known specific antagonist to ACh
receptors, but it inhibited both
photoresponses and histamine-induced
responses of L-neurons. Therefore, curare
can be considered as a histamine
antagonist in the cockroach ocellus.

GABA, picrotoxin, ACh, carbachol, eserine,
neostigmine, and oC-BGTX, had no
significant effects. These results suggest
that histamine is a photoreceptor
transmitter in the cockroach ocellus.

PH 58

DEVELOPMENTAL PROCESS OF VISUAL FUNCTION
DURING LOWERING OF THE NEURONAL ACTIVITY
IN THE FLY.

K. Mimura, Nagasaki Univ., Fac. of Liberal
Arts, Nagasaki.

The development of some of the visual
function in the flesh-fly depends on the
visual experience in the early and short
period of post-emergence (Mimura 1986,
1987, 1988). This report attempted to in¬
vestigate the detailed developmental proc¬
ess of the visual function (pattern dis¬
crimination) by short-term suppression of
neuronal activity at various stages of
post-emergence. The suppression was
brought on by lowering the environmental
temperature to 10 °C or less. Results were
as follows: (1) Very short-term (20-30min)
and normal neuronal activity immediately
after visual experience is necessary for
actuating the development of pattern dis¬
crimination. (2) Normal neuronal activity
for 5h after the first visual experience
is needed for the sufficient development
of pattern discrimination.

Additionally, it was demonstrated that
by applying some inhibitors of protein
synthesis, phototaxis is not impeded but
pattern discrimination is. This finding
illustrates the developmental difference
between innate and acquired functions.

PH 59

DISCRIMINATION OF COLOURED PAPERS IN THE
WALKING BLOWFLY, LUCILIA CUPRINA.

T . Fukushi Dept, of Biol., Miyagi Coll,
of Educ., Sendai.

An individual blowfly is trained by
presented it with a droplet of sugar
solution on a patch of coloured paper
(colour mark 1X1 cm) . After having
consumed the sugar droplet, the fly
starts a systematic search. While
searching, it is confronted with an
array of colour marks consisting of 4
colours and displayed on the test
cardboard .

The flies trained to one of blue,
green, yellow or red prefer this colour
significantly when confronted with an
array including these 4 colours. The
blue and yellow marks are learned
easily. Do the flies discriminate
between blue and yellow by the hue of
colour or the brightness? To answer this
question, the following experiments are
carried out.

(1) The test array including white
instead of green is used. (2) The flies
are trained to blue or yellow with
reduced light intensities - of reflection
and tested on the standard test
cardboard. (3) The test array including
blue (or yellow) , white and two grades
of grey is used. The results show that
the fly discriminates between blue and
yellow by means of the hue of colour.

PH 60

SIGN CHANGE OF PHOTOTAXIS IN WHITE-EYE SILK¬
WORM MUTANT (W- 1 )

I. Shimizu. Res. Sect.of Environ. Biol., Labo.
for Plant Ecological Studies, Fac.of Sci.,
Kyoto Univ., Kyoto.

A white-eye mutant(w-l) of the silkworm
( Bombyx mori) lacks ommochome pigments due

to a defect -of kynurenine 3-hydroxylase.

Their laravl stemmata and adult compound eyes
are pale white. The phototactic behavior of
the mutant varied with light and dark
adaptations. Light-adapted newly-hatched

larvae showed positive phototaxis at low
intensities of a test yellow light, and
negative phototaxis at intense test lights.
After dark-adaptation they showed negative
phototactic response from low to high
intensities of test light. Green light was
effective in both positive and negative

responses. The half-time of sign reversion in
the phototaxis was approximately 20-30 min in
dark and light-adaptation. Mature w-1 larvae
showed ordinary positive phototaxis to the
target of light after light-adaptation. On the

other hand dark-adapted larvae showed a
series of complex but stereotyped behavior. At
first they orientated themselves to the

opposite of the light-target and turned toward
the light-source again and moved their heads
from side to side for a while. Finally they
turned to the opposite side again and moved
away. F-l hybride between the mutant (wl/wl)
and a black-eye race showed strong positive
phototaxis irrespective of light and dark
adaptation:the phototactic character of the

mutant is apparently recessive.

Physiology

1207

PH 61

TRANSRETINAL RESISTANCE CHANGES INDUCED BY
LIGHT AND BY DEPLETION OF CALCIUM FROM THE
EXTERNAL MEDIUM IN THE OCTOPUS RETINA.

K. Ohtsu. Ushimado Marine Laboratory,
Fac. of Sci., Okayama Univ., Okayama.

In octopus retinae, conductance of
photoreceptor cell membranes increases
upon illumination, resulting in a
vitreous-negative electrical change, ERG,
in a transretinal recording configuration.
A similar electrical change, transretinal
voltage (TRV), was generated without
i 1 luminaion by incubating the retina in
Ca2+-free artificial sea water (ASW) . The
TRV, however, disappeared in a few min
after the addition of 10 mM Ca2+. It
seems likely that photoreceptor membrane
resistance changes are reflected in
transretinal electrical resistance ( TRR ) .
It is therefore to be expected that the
TRR decreases upon illumination or
depletion of Ca2+ from an external medium
and increases after the addition of Ca2+.
Unexpectedly, however, the TRR increased
transiently upon dim illumination and
decreased after the addition of Ca2+,
whereas, an intense light stimulus
decreased transiently the TRR. This
inconsistency might be attributed to
decreased extracellular space because the
TRR is partially a function of the
resistance of the retinal extracellular
space. The results here suggest that the
rhabdomeric microvilli may have swelled
upon illumination or depletion of Ca2+.

PH 62

EFFETS OF [Ca2+] AND IBMX ON
PHOTORESPONSES OF DETACHED FROG RODS.

K. Azuma , Dept. of Biol., Osaka Med.
Coll., Takatsuki , O^aka 569 _ _

Effects of [Ca^+] and IBMX on the
light sensitivity of rods detached from
the frog Rana catesbeiana retinas were
investigated by sucking the inner
segments into pigettes.

Reducing [Caz ] and the presence of
IBMX influenced amplitudes and time
courses of the photoresponses. Low Ca (90
juM) and IBMX ( 50 juM) induced increases in
the amplitudes of the responses. Low Ca
caused a shortening in the falling phase
of the responses and IBMX slowed rising
and falling phases of the responses.

At normal Ca, the onset of adaptation
light produced a steady outward change of
membrane currents which tended to recover
to the initial dark level at the offset of
the light. At low Ca and in the presence
of IBMX the offset of the light induced a
large inward change of membrane current
which transiently overshot the initial
dark level. In the overshoot state, which
was caused by a transient increase of
membrane current, flash lights evoked
photoresponses which were about 1.2 -1.4
times larger than those in the initial
dark (called hypersensitivity).

Thus hypersensitivity of detached rods
was observed at low Ca and in the presence
of IBMX. These results indicated that the
hypersensitivity was caused by the rod
photoreceptor itself.

PH 63

INTRACELLULAR POTASSIUM ACTIVITY OF THE
HORIZONTAL CELLS IN THE STINGRAY RETINA.
M.Fu jimoto, H.Yanase and J.Katayama.
Zool . Inst., Fac. of Sci., Hiroshima Univ.
Hiroshima .

The horizontal cells in the stingray
retina are very large. Making use of this
characteristics, intracellular activities
of potassium in the horizontal cells were
studied by means of ion-exchanger micro¬
electrodes. Double-barrelled electrodes,
having a tip diameter of 1-2 pm, were used
and the distal portion of one barrel was
filled by K+ exchanger resin (Corning no.
477317). The electrodes were calibrated in
a set of 7 (KCl+NaCl) solutions of 300 mM
constant ionic strength; the K+ concentra¬
tions were 2, 5, 10, 20, 50, 100 and 200
mM. The average membrane potential and
activity of potassium ions in HI
horizontal cells (luminosity type) were
-13.8+1.8 mV and 126.1+8.5 mM (n=24),
respectively. The membrane potentials of
H3 horizontal cells ( chromaticity type)
were often larger than those of HI cells.
During light-evoked hyperpolarization, the
intracellular K+ activity decreased. On
the other hand, light-evoked depolariza¬
tions in H3 cells were found to be accom¬
panied by increase in the K+ activity.
These light-evoked changes in intra¬
cellular K+ activities are incompatible
with those deduced from the light-evoked
change in Na+ activity.

PH 64

EFFECT OF DOPAMINE ON BIPOLAR CELLS IN THE
CARP RETINA.

T.SAITO1 and M.YAM^DA . 1Inst.of Biol. Sc.,
Univ. of Tsukuba, Electrotechnical
Lab . , Tsukuba-shi , Ibaragi, 305

Bipolar cells in the carp retina are elec¬
trically coupled, which makes the recept¬
ive field size of these cells much larger
than their dendritic field, we have used a
long narrow slit of light(170um width and
4mm length) of different positions to char¬
acterize the receptive field profile of ON
bipolar cells. Stimulation at the receptive
field center produced a large depolarizing
response followed by a hyperpolarizing one.
Both responses decresed in their amplitude
with the slit displacement from the center.
The spatial distribution of the hyperpola¬
rizing response was much larger than that
of the depolarizing one, suggesting that
the hyperpolarizing response is mediated by
horizontal cells. The peak amplitude of the
depolarizing response was plotted against
the slit displacement from the center in
semilogarithmic ordinates. The decay of the
response could be fitted by a straight line
whose slope allowed the measurement of the
length constant value(52-98 urn). Dopamine
(50 urn) hyperpolari zed the membrane poten¬
tial, enhanced the depolarizing response
and reduced the hyperpolarizing response.
Dopamine also produced a narrowing of the
receptive field surround, while had little
effect on the central receptive field size.

1208

Physiology

PH 65

STRETCH RECEPTOR ORGANS IN THE THORAX OF
THE PILL BUG ( ARMADI LLI PI UM VULGARE) .

A. Niida, K. Sadakane, and T. Yamaguchi.
Dept. of Biol., Fac. of Sci., Okayama
Univ. , Okayama.

Bilateral arrangement of four types of
stretch receptor organs (SROs) responsive
to displacement of the tergites was found
in the thorax of pill bugs. The first type
is the most anterior SRO whose dendrite
spreads over a non-spec i al i zed muscle
lying between the second and third
segments. The second and third types are
well-developed muscle receptor organs with
receptor cells, each of which the
dendritic field extends on to a distinct
receptor muscle, and the receptor muscle
for the former lies between the third and
fourth segments and that for the latter
does in the fourth segment. The fourth
type is located in each segment posterior
to the fifth segment: a pair of sensory
cells spread the dendrites on a single
receptor muscle lying in their own
segment. Both of the sensory cells have
slowly adapting property, but one member
has a low threshold and optimal
sensitivity to rather small bending of the
thorax, while other member has the
opposite properties. This suggests that in
a sequence of bending, optimum response
ranges are shared with the two sensory
cells .

PH 66

THE CHARACTERISTICS OF RESPONSES AND THE
CENTRAL PROJECTIONS OF AN ANTENNAL COLD
RECEPTOR ON THE COCKROACH, PERI PLANETA
AMERICANA L.

M. Nishikawa, F. Yokohari, and T. Ishibashi
Biol. Lab., Fac. Sci., Fukuoka Univ.,
Fukuoka

The antennal cold receptors occur in
two types of sensilla, the olfactory- ther-
moreceptive sensillum (OT type) and the
hygro- thermorecept ive sensillum (HT type) .
The following report shows the characteris¬
tics of responses of HT type cold recep¬
tors .

1. The static activity at constant temper¬
atures (T) was lower at higher tempera¬
tures. The static activity was linearly
correlated with constant temperatures in
the range of 20-30°C. This is remarkably
different from that of the OT type which
did not show any high correlation.

2. The response to constant rate of temp¬
erature falling (-dT/dt) was higher at
higher rate of temperature falling. At the
same time the sensitivity to temperature
drops was lower at higher instantaneous
temperatures .

3. The response of the cold receptor does
not depend on the temperature difference
(-^T) .

The central projections of the recep¬
tors in the sensilla were examined with
extracellular cobalt-lysine staining meth¬
od. Some neurites of the receptors of the
both sensilla terminated at the same area.

PH 67

RESPONSES TO HYGRO-STIMULATION IN THE
ANTENNAL LOBE OF HONEY BEE( Apis mellifera) .
T. Itoh, F. Yokohari and Y. Tominaga.
Dept. of Biol., Fac. of Sci., Univ. of
Fukuoka, Fukuoka. _ _

About 50 hygro-thermoreceptive sensilla

have been found on an antenna of the honey
bee ( Apis mellifera ) . Each of them contains
at least three sensory cells, a dry, a
moist and a cold cell. Their axons are sent
toward the brain via the antennal nerve. In
the brain, most of antennal nerve fibers
enter the antennal lobe.

Responses to hy gr o-stimulation applied to
the antenna were recorded extracellularly
from the dorsal region of the antennal
lobe. Two types of responses were recorded,
dry responses and moist responses. Dry
responses were phasic or tonic in their
response time course. Their impulse
frequency was increased with relative
humidity decrease. Moist responses were
phasic-tonic. Their impulse frequency was
increased with relative humidity increase.

Extracellular cobalt injection via a
recording electrode revealed two types of
olfactory lobe neurons. One type is a
local interneuron which has wide branching
pattern restricted in the antennal lobe.
The other is output neuron which innervates
single glomerulus in the dor so-postero
region of the antennal lobe and sends an
axon toward protcerbrum. We consider that
humidity information may be processed via a
neuronal system resembling olfactory
information processing system.

PH 68

MOTOR OUTPUTS OF THE DESCENDING OCELLAR
INTERNEURON OF THE AMERICAN COCKROACH.

T. OHYAMA and Y. TOH . Dept, of Biol., Fac.
of Sci., Kyushu Univ., Fukuoka.

Two types of descending ocellar
interneurons have been reported in the
American cockroach in the previous study
(Ohyama & Toh, 1987). They responded to
ocellar illumination with a few off-spikes
and to air puffs to cerci with continuous
spike discharges. The continuous spike
discharges caused by cereal stimulation
were suppressed by ocellar illumination.

In the present study, motor outputs of the
descending ocellar interneuron were
examined by recording activities of the
peripheral nerves of the metathoracic
ganglion during intracellular stimulation
of the descending ocellar interneuron. The
nerves were named after Pipa and Cook
(1959). Intracellular stimulation of the
descending interneuron elicited continuous
spike discharges in the N2 and N6
ipsilateral to the axon of the descending
interneuron, but did not in the ipsilateral
N3 and N5, and contralateral N2, N3 and N5.
These data show that some of the
metathoracic motoneurons are activated by
the descending ocellar interneuron, though
such motoneurons have not been
morphologically identified. It has been
reported that the N2 and N6 innervate
flight and leg muscles (Pipa & Cook, 1959).
The present data together with previous
data are suggestive of ocellar control of
flight or locomotion activities through the
descending ocellar interneuron.

Physiology

1209

PH 69

DIRECTIONAL SENSITIVITY OF CRICKET CERCAL
SENSORY INTERNEURONS TO AIR-CURRENT
STIMULUS

M. Kanou. Biol. Inst., Fac. of Gen.
Educ., Ehime Univ., Matsuyama

Directional sensitivity of cereal
sensory interneurons (CSIs) of the
cricket, Gry 1 1 u s bimacula tus D, was
investigated using a unidirectional air-
current stimulus. The stimulating
apparatus was a wind tunnel consisted of a
pair of push-pull driven loud-speakers.
The directional characteristics of CSIs
was recorded intracellularly and was
expressed by the frequency threshold
curves to air-current stimuli from
different orientations. Recordings were
obtained at 6 positions at 30 or 120
intervals around the animal in the
horizontal plane.

CSIs of 8-1 ( MGI : medial giant inter¬
neuron), 9-2, 9-3, 10-2 and 10-3 were

investigated. The optimal directions of
CSIs to air-current were different from
each other. The optimal directions and
other directional characteristics of a
pair of (right and left) CSIs were
symmetry with respect to the body axis.
On the other hand, the optimal directions
of CSIs in the same type (8-1 , 9-2 and 9-3
in PT-type; 10-2 and 10-3 in T-type: Kanou
and Shimozawa 1984, J. Comp. Physiol.)
were symmetry with respect to the
transverse axis of the body.

PH 71

EFFECT OF TEMPERATURE AND LACTATE
CONCENTRATION ON THE ISOLATED HEART
ACTIVITY OF THERMALLY ACCLIMATED GOLDFISH.
H.Tsukuda. Dept, of Biol., Fac. of Sci.,
Osaka City Univ. , Osaka

The frequency and the amplitude of
ventricular contraction were recorded for
the isolated heart of 10°C- of 25°C-
acclimated goldfish. Total activity of the
heart was evaluated by the power calculated
from the frequency and the amplitude.

Lactate induced the frequency to
increase in 25°C-fish heart. In general,
5mM lactate was effective in the mainte¬
nance of heart activity, while 1 OmM as well
as 20mM lactate showed negative effect.

Increase in temperature enhanced
decrease in the heart activity for one hr.
The decrease was faster in 10°C-fish than
in 25°C-fish, especially at higher
temperatures. The frequency increased in
every case with increase in temperature
between 5° and 30 °C. The frequency of
10°C-fish heart was significantly higher at
10°C and lower at 30°C than that of 25°C-
fish heart. The amplitude of 10°C-fish
heart decreased with increase in
temperature and that of 25°C-fish heart
showed the maximum at 15°C. The maximum
power was attained at 20 °C irrespective of
acclimation temperature, but the power of
10°C-fish heart was higher at 5° and 15°C
and lower at 20°, 25° and 30°C than that of
25°C-fish. These results indicate adaptive
effect of thermal acclimation.

PH 70

COMPARISON OF FREEZING-TOLERANCE BETWEEN
TWO SPECIES OF DROSOPHILIDAE , CHYMOMYZA
COSTATA AND C. FUSCIMANA. 2 1

K. Shimadal and A. Riihimaa . iInst. Low
Temp. Sci., Hokkaido Univ., Sapporo and
^Dep. Genet., Univ. Oulu, Finland.

Many species of Drosophilidae inhab it -

ing in temperate and frigid regions enter
reproductive diapause at the adult stage.
The overwintering or laboratory cultured
diapausing adults are usually susceptible
to freezing. Unlike those species, C. cos -
tata and C. fuscimana have a larval dia¬
pause. The diapausing larvae of both spe¬
cies are potentially tolerant to freezing
at temperatures below -80°C. They exhibit¬
ed the freezing- tolerance when they were
acclimated to 4°C for 1 month, initiated
to freeze with ice inoculation at nearly
below 0°C and cooled at a constant rate of
0.1°C/min. The slow freezing conditions
were essential for their post-thawing sur¬
vival. Between the two species, C. fusci¬
mana larvae were more sensitive to cooling
velocity. As natural cryoprotectants , tre¬
halose and proline were detected in C_. cos-
tata larvae. During the cold acclimation,
the” trehalose and proline contents were
increased to 22 and 175 nmole/mg fresh
weight, respectively. The remarkable in¬
crease in proline content seems to be in¬
volved with the occurrence of freezing-
tolerance in C. costata . In C. fuscimana ,
its proline content is still unknown.

PH 72

EFFECT OF THERMAL ACCLIMATION ON HEART
LACTATE DEHYDROGENASE IN GOLDFISH.
T.Yonehana and H.Tsukuda. Dept, of Biol.,
Fac. of Sci., Osaka City Univ., Osaka.

Goldfish were acclimated to 10° or 25°C
for more than two months before isolation
of heart. Enzyme materials were prepared
by Sephadex gel filtration of the centri¬
fuged supernatant of heart homogenate.

Vmax and Km values were calculated by
Lineweaver and Burk's plot.

The enzyme activity increased with
increse in temperature between 10° and
40°C, and the higher concentration of
substrate showed inhibition effect at lower
temperatures. There was no difference
between apparent Km values as function of
temperature in 10°C- and 25°C-fish. The
ratio of pyruvate reduction to lactate
oxidation at 10°C was higher in 10°C-fish
than in 25°C-fish. In 10°C-fish, the Q1 0
value calculated from the Vmax was lowest
in temperature range from 10° to 15°C for
the lactate oxidation and 15° to 20°C for
the pyruvate reduction, while in 25°C-fish,
that was lowest in temperature range from
20° to 25°C for the pyruvate reduction as
well as the lactate oxidation.

Thus, the Q1 0 value of either tempera¬
ture fish was lowest at temperature near
the each acclimation temperature. This
seems to be one of acclimation phenomena
to temperature in ectotherms and important
ability of them for survival.

1210

Physiology

PH 73

THE EFFECT OF BREEDING TEMPERATURE ON WING-
LENGTH OF THE WATERSTRIDER , GERRIS PALUDUM.
T.Harada and K. Taneda. Dept, of Biol. Fac.
of Sci., Kochi Univ., Kochi.

Gerris paludum is known to show an alary
dimorphism (macroptera and brachyptera ) .
Our field investigation showed that there
were differences among the generations in
the component ratio of the alary di¬
morphism. G^ paludum is usually tri-
voltine in Kochi. The frequency of macro-
pterous specimens in the 2nd generation
(late summer generation) was higher than
that in either of the other generations.
This finding suggests that wing-length is
correlated with breeding temperature. First
instar nymphs were bred at high (30°C) or
low temperatures (20°C) until they became
adults. The frequency of macropterous
specimens was higher in the high
temperature breeding group than in the low
temperature breeding group. The 5th instar
nymphs, which had grown and just molted in
the field, were bred at the high or low
temperature. There was little difference in
the frequency distribution of wing-length
between breeding temperatures. Thus, it
seems that the breeding temperature at the
5th instar nymph stage does not affect the
wing-length. The ecological significance of
the present findings was also discussed.

PH 74

CHEMICAL COMPOSITION OF REFLECTING
PLATELETS IN FISH IRIDOPHORES AND THEIR
LIGHT-REFLECTING PROPERTIES.

R.Fujii, N.Oshima, M.Aihara, H.Kasukawa
and N.Arakawa. Dept, of Biol., Fac. Sci.,
Toho Univ., Funabashi.

Chemical nature of thin reflecting
platelets in iridophores of some teleosts
was studied by thin-layer chromatography on
crystalline cellulose. The platelets from
blue damselfish (Chrysiptera c y anea ) are
composed exclusively of guanine, as judged
by the limits of sensitivity of this
method. By contrast, chromatograms of the
plalelets from carp ( Cy pr inus car pio )
revealed that hypoxanthine is present in
addition to the main component, guanine.

In tilapia ( Sarotherodon niloticus) , trace
of unknown fraction was detected. Judged
from Rf value and fluorescence, it might be
xanthine. In the blue-green damselfish
( Chr omis viridis) , the platelets from
dorsal skin are composed almost purely of
guanine, while those from ventral skin
contained hypoxanthine besides guanine. In
neon tetra ( Paracheir odon innesi ) , plate¬
lets either from dorsal or ventral skin
contained small fractions of two sub¬
stances, presumably hypoxanthine and
xanthine. It is suggested that very thin
platelets, such as those from blue
damselfish, are composed of guanine, while
usual thicker ones contain other purines,
though they are closed related to their
common and main component, guanine.

PH 75

MOTILITY OF THE IRIDOPHORES IN THE NEON
TETRA -- I. THE LIGHT-REFLECTING PROPERTY.
H.Nagaishi, K.Miyaji, N.Oshima and R.Fujii.
Dept, of Biol., Fac. of Sci., Toho Univ.,
Funabashi .

The color change in lateral stripe of
the neon tetra, Par ache _ir odon in n e j; i_ ,
originates from the physiologically active
iridophores sensitive to light. Like the
iridophores of the blue damselfish, they
contain the stacks of guanine plates, and
reflect the brilliant color due to the
multilayered thin-film interference
phenomenon of the "non-ideal" type. When
illuminated from direction of the dorsal
side, the stripe reflected iridescent
color, but the illumination from the
direction of ventral side did not induce
the reflection from the stripe. The fact
may imply that the every guanine plate in
the iridophores arranges regularly forming
a constant angle with the surface of the
skin. The peak of wavelength of reflected
light from the iridophores was shifted
toward the longer wavelengths by illumi¬
nation or application of norepinephrine and
so on. In that case, the relative
reflectance decreased. From the optical
analysis, it was also suggested that the
shift of the spectral peak may be caused by
the changes in the angle of inclination of
guanine plates within the iridophores.

PH 76

MOTILITY OF IRIDOPHORES IN THE NEON TETRA
— II. REGULATION OF MOTILITY.

N.Oshima, H.Nagaishi and R.Fujii. Dept, of
Biol., Fac. of Sci., Toho Univ., Funabashi.

The color of stripe of the neon tetra,
Paracheirodon innesi , changes from green
in the daytime to blue-violet at night.
This phenomenon may be due to the responses
of iridophores to light, since similar
changes were observed in the fish denervated
by reserpine. However, the color of stripe
changes rapidly to yellowish green, when
the fish are subjected to the stresses or
eat baits. Therefore, the possible involve¬
ment of endocrine and nervous system in the
iridophore motility was examined. The skin
forming the stripe was excised carefully,
and various substances were exogenously
applied. The responses were microphoto-
graphed or recorded photoelectrically .

K+ and norepinephrine (NE) caused the
shift of the spectral peak of light
reflected from the iridophores toward
longer wavelengths. Adenosine and ATP, co¬
transmitters, were found to reverse the
action of NE. MCH and tf-MSH also caused the
shift of the spectral peak toward longer
and shorter wavelengths, respectively.
Melatonin showed only a little effect at
high concentrations. These results suggest
that, although the neon tetra iridophores
are fundamentally under the direct control
of light, neural and hormonal regulation
may contribute to the dynamic or minute
changes in color of stripe.

Physiology

1211

PH 77

RECEPTOR MECHANISMS OF MOTILE IRIDOPHORES
OF THE FLOATING GOBY , CHAENOGOB I US ANNULARIS.
T. Iga, N. Maeno and T. Asari. Dept, of
Biol., Fac. of Sc i . , Shimane Un i v . , Matsue

Recently, we found that iridophores of
the floating goby, Chaenogob i u s annularis
(brackish form), are motile. These irido¬
phores responded to some stimuli with intra
cellular translocation of the platelets
within the cells. High K solution induced
dispersion of the platelets within the
cells. This response was lost in irido¬
phores in denervated scale preparations.
Noradrenaline caused dispersion of the
platelets. The action was inhibited by
alpha adrenergic antagonists, but not
blocked by beta adrenolytic agents, indi¬
cating that noradrenaline acts on the alpha
adrenoceptors of the iridophore membrane
to induce platelet dispersion. Isoprotere¬
nol brought about platelet aggregation and
its action was antagonized by propranolol,
but not by phenoxybenzam i ne , suggesting
that the iridophores possess beta adreno¬
ceptors which function in platelet aggre¬
gation within the cells. Forskolin was
effective in inducing aggregation of the
platelets. Melatonin enhanced aggregation
of the platelets. Based on these results,
it was concluded that the motile irido¬
phores are under the dual control of the
sympathetic adrenergic nerves and hormones,
and that an increase in intracellular cAMP
induces platelet aggregation within the
cel Is.

PH 7°,

BETA-1 AND BETA- 2 ADRENERGIC RECEPTORS
COEXIST IN THE MELANOPHORE OF GOBY,

TRIDENTIGEE OBSCURES.

H. Katayama. Mukaishima Mar. Biol. Stat. ,
Fac. of Sci., Hiroshima Univ. , Mitugi-gun.

Effects of several 3-adrenergic agonists
and antagonists on denervated melanophores
were examined to study subtype of 3-adren-
ergic receptors of the melanophore. Pigment
of melanophores was made to aggregate with¬
in the cells by bathing a split caudal fin-
ray isolated from the goby in physiological
solution containing lOOyM verapamil and 10
yM phentolamine. The 3-agonist isoprenaline
(ISO), epinephrine (E) , norepinephrine (NE) ,
salbutamol (SAL) or dobutamine (DOB) caused
redispersion of aggregated pigment of the
melanophores in the presence of verapamil
and phentolamine. The order of effective¬
ness of the 3-agonists was ISO > E ^ NE ^SAL
> DOB . The pigment dispersing effect of all
these 3-agonists was completely inhibited
by propranolol. The 3-1 specific antagonist
metoprolol completely inhibited the effect
of the 3~1 agonists NE and DOB, and it par¬
tially inhibited the effect of nonspecific
agonists ISO and E, while it did not inhib¬
it the effect of SAL, a 3~2 specific ago¬
nist. Butoxamine , a 3-2 antagonist, par¬
tially inhibited the effect of SAL and E.

It is reasoned from these results that NE
and DOB act on 3-1 adrenergic receptors of
the melanophore, ISO and E stimulate both
types of receptors, while SAL selectively
stimulates 3-2 receptors, and hence 3-1 and
3-2 receptors coexist in the melanophore.

PH 79

ALPHAz-ADRENOCEPTORS AND INHIBITORY GTP-
BINDING PROTEIN IN FISH MELANOPHORES.

F. Morishita and K. Yamada. Zool. Inst.,
Fac. Sci., Hiroshima Univ., Hiroshima.

Study was made on the effect of per¬
tussis toxin (IAP), on alpha2-adrenoceptor-
mediated pigment-aggregation response of
scale melanophores of black goldfish,
Carassius auratus , to ascertain the in¬
volvement of inhibitory GTP-binding protein
(Gi) in the response. IAP (2yg/ml) markedly
inhibited the aggregation response of
melanophores to norepinephrine (NE) , where¬
as it had no effect on NE-treated aggre¬
gated melanophores. When the homogenate of
the scales was treated with 32P-NAD and IAP-,-
a protein band at the molecular mass of 43
kDa was labeled in SDS-polyacrylamide gel.
The labeling was markedly reduced when NE
and GTPyS, an analogue of GTP, or nicotin¬
amide were added to the reaction mixture.
These results indicate that the observed
band correspond to the substrate for IAP,
i . e . Gi .

These findings suggest that Gi is in¬
volved in alpha2-adrenoceptor-mediated
response of goldfish melanophores to NE and
that the bindig of NE with the adrenoceptor
brings about dissociation of Gi , which in¬
hibits adenylate cyclase activity and
causes aggregation of pigment within the
melanophores .

PH 80

BODY COLOR CHANGE AND BEHAVIOR IN UPSIDE-
DOWN CATFISH ( SYNODONTIS NIGRIVENTRIS ) .
H.Nishi, H.Nagaishi, R.Fujii and N.Oshima.
Dept, of Biol., Fac. of Sci., Toho Univ.,
Funabashi .

As its name denotes, upside-down catfish
( Synodont i s nigriventris) normally swim
upside down, and can rest in many postures.
Besides those interesting behaviors, unique
morphology may be noticed: Having melano¬
phores, its abdominal skin is dark like the
dorsal skin. Such coloration should be
advantageous for survival, since it is dif¬
ficult for the predators to find the fish
from above. Although the melanophores on
both dorsal and ventral skin were likewise
controlled quite orthodoxly, there was a
little difference in the effective
concentrations of exsogenously applied
neural and hormonal substances.

Excitement darkening and blanching were
observed in this fish species. When
electric shock was applied to the white-
adapted fish, it darkened very rapidly, in
comparison with the adaptation to the black
background. This response was suppressed by
the previous injection of propranolol, a
beta-blocker. Moreover, under the stresses,
the fish stopped swimming. These responses
to the stress and unplesant factors appear
to be reasonable, since they are nocturnal
in their habits and pale at night by the
action of melatonin.

1212

Physiology

PH 81

ULTRASTRUCTURES OF CONTRACTILE BUNDLES IN
EPITHELIAL CELLS OF THE SPONGE.

A.Matsuno, H.Ishida, M.Kuroda and Y.Masuda
Dept, of Biol., Fac . of Sci., Shimane Univ.
Matsue and ^ Dept . of Biol., Kawasaki Medi¬
cal School, Kurashiki.

Epithelial cells in the oscule and body
wall of the sponge bear bundles of thin
filaments ( 4-6 nm in diameter ) . These
bundles originated from a desmosome of one
end of the cell and ran through to the one
of the opposite end. Thin filaments con¬
nected directly with tono-f ilaments of the
desmosome. We supposed that those thin
filaments in the bundles are constructed
by actin and able to contract. Contraction
of the bundles subsequently lead transform
of the cell-shape. Consequently, the oscule
and body wall show dynamic contraction.

Actin filamets in cells were idenfied
using 1 ) Observations by f luorescence-
phallacidin staining and 2) Electron micro¬
scopic examinations of arrowhead decoration
with rabbit heavy meromyosine. 1)Fluo-
rescence-phallacidin staining revealed
bundles of actin filamets running across
the cell. They ran radially from the
central area in some cases. 2) From E.M
examinations, HMMwere found irregularly on
the thin filaments, but did not show
ordered arrowhead.

PH 83

BIOLOGICAL ACTIVITIES OF GANGLION EXTRACTS
FROM THE MOLLUSC FUSINUS FERRUGINEUS.
T.Kanda1, I . Takabatake^ , Y.Fujisawal, T.
Ikedal, Y.Muneokal and M. Kobayashil . -'-Fac.
of Integrated Arts and Sci., Hiroshima
Univ., Hiroshima and ^Dep . , Biol., Fac.,
Sci., Shimane Univ., Matsue. _

Acetone extract of the ganglia of a

prosobranchi mollusc, Fusinus ferrugineus,
was forced through a C-18 cartridge. The
retained material was applied to a column
(2.6 X 40 cm) of Sephadex G-15. Fractions
of 4 ml each were collected, and their
biological activities were first assayed on
twitch contractions of the radula retractor
muscle of Fusinus . Three peaks (E^ , E2 and
E3) of contraction-potentiating activity
and one peak (I) of contraction-inhibiting
activity were found at fraction No. 22, 45,
50 and 29, respectively. All of the activ¬
ities of these peaks were destroyed by
incubating the fractions with the protease
subtilisin. Using several molluscan
muscles, the actions of the substances in
these peaks were examined in more detail.
The obtained results suggested that the
substance in peak El may be a novel excita¬
tory peptide of which molecular weight is
more than 1500, and that the substances in
peak E2 and peak E3 are FMRFamide-like
peptides. The results also suggested that
the substance in peak I consists of at
least two peptides having molecular weight
of 800-900, and that one of them is a
catch-relaxing peptide or a peptide related
to it .

PH 82

BIOLOGICAL ACTIVITIES OF METHANOL EXTRACTS
OF THE DEMOSPONGES, SUBERITES SERICEUS AND
CHOANITES FICUS.

Y. Fujisawa, K.Kanda, T.Ikeda and Y.Muneoka.
Fac., Integrated Arts and Sci., Hiroshima
Univ. , Hiroshima. _

S. sericeus and C. ficus are symbiotic
demosponges which cover the shell surface
of the mollusc Fusinus ferrugineus . In the
present experiments, we first confirmed that
methanol extract of the sponges has various
kinds of activity on molluscan muscles. In
the second series of the experiments, the
extract was forced through C-18 cartridges
and the retained and flowthrough materials
were respectively assayed on the ABRM of
Mytilus . The results showed that both of
the materials have some kinds of activity,
and that these activities vary with the
season. In the third series of the experi¬
ments, the materials were fractionated with
a column (2.6 X 40 cm) of Sephadex G-15,
and the activities of the fractions were
assayed on the ABRM. From the fractions of
the retained material, we , obtained two
peaks of contraction-potentiating activity
and one peak of contraction-inhibiting ac¬
tivity. From the fractions of the flow¬
through, we obtained two peaks of contrac¬
tion-potentiating activity and two peaks of
contractile activity. We also obtained con¬
traction-inhibiting activity at the posi¬
tion of fractions 27-36 .However , the peak
position of this activity was not detected
clearly, since most of these fractions also
showed the foregoing contractile activity.

PH 84

FURTHER ANALYSIS OF THE SMOOTH MUSCLE
CONTRACTION- INDUCING FACTOR IN THE
DIGESTIVE TRACT OF THE SEA URCHIN.

N. Takahashi1 , N. Sato2, I. Oota3 , M.
Takahashi1 and K. Kikuchi2 . 'Marine
Biomed. Inst., 2 Dept, of Pathology and
3Dept. of Physiology, Sapporo Med. Coll.,
Sapporo. _

According to our paper submitted in
1987, a contraction-inducing factor (s)
(CF) of gonadal smooth muscle is present
in the digestive tract of the sea urchin,
Strongylocentrotus intermedius . Data
obtained so far indicate this substance to
possibly be a glycoprotein: by the Molish
reaction, a glycoprotein was positive at
the CF fraction of gel filtration and two
reactions of purified CF with coomassie
blue and PAS in SDS-PAGE were also
positive. Our previous study indicated
the molecular weight of CF to be from
2,000 ~ 4,000 and this factor to resist
the action of various proteases. In the
present study, CF fractions treated with
L-glycosidase and cellulase for 1 2 hours
at 37 °C were found to be less capable of
causing contraction. CF is thus concluded
to be glycoprotein and apparently, its
glyco-component induces contraction of
smooth muscle.

Physiology

1213

PH 85

SIMULTANEOUS MEASUREMENTS OF TENSION AND
INTRACELLULAR FREE CALCIUM IN SINGLE
MOLLUSCAN SMOOTH-MUSCLE CELLS.

N. Ishii^, A.W.M. Simpson2 and C.C.
Ashley2. 1 Zool. Inst., Fac. of Sci.,
Univ. of Tokyo, Tokyo and 2Univ. Lab. of
Physiol ., Oxford, U.K. _

In order to learn the role played by

calcium ions in the 'catch' contraction,
we have measured intracellular free-
calcium concentration ( [ Ca^ + ] ^ ) in
intact, single smooth-muscle cells iso¬
lated from the anterior byssus retractor
muscle of Mytilus edulis. Smooth-muscle
cells were enzymatically dispersed and
loaded with a fluorescent Ca2+ indicator
fura-2, as previously described (Ishii et
al., 1988). [Ca2+]i and tension were

simultaneously measured on an epifluores-
cence microscope by employing a dual¬
excitation fluorimeter (PTI Inc.) together
with a sensitive force transducer. A
brief electrical stimulation evoked a
transient elevation of [Ca2+]i from a mean
resting value of 138 nM to ca. 500 nM,
which was associated with a catch con¬
traction (half relaxation time, ca. 200
s). However, [Ca2+]i quickly declined
towards its resting level (T-| /2/ ca. 10 s)
and was unaffected by 10 uM 5-hydroxy-
tryptamine which brought about a rapid
relaxation. The results suggest that both
maintenance and relaxation of the catch
tension are independent of [Ca2+Ji.
Reference: Ishii, N., Simpson, A.W.M. &
Ashley, C.C. (1988). Biochem. Biophys.
Res. Commun. 1 53, 683-689.

PH 87

SHORTENING VELOCITY OF BARNACLE STRIATED
MUSCLE

H. Iwamoto, A. Goto, A. Muraoka and H. Sugi.
Dept. Physiol., Sch. Med., Teikyo Univ, Tokyo

The unloaded shortening velocity of the
adductor scutorum muscle from a barnacle,
Tetraclita squamosa, was measured at various
timings during a contraction elicited by 8 s AC
stimulation at 5°C. During the period of
stimulation, the velocity was fairly constant.
During relaxation, the velocity declined much
faster than the force, in marked contrast to the
frog skeletal muscle in which the velocity is
constant over the period of a twitch. The rate of
rise in force after a quick reduction in length
behaved in a similar manner. The results suggest
that the barnacle muscle maintains its force
economically during relaxation with a reduced
rate of crossbridge cycling, as vertebrate
vascular and molluscan smooth muscles.

Effect of changing the intracellular Ca level on
the unloaded shortening velocity was examined
by altering the external Ca concentration from 2
to 100 mM. The velocity increased with
increasing external Ca, but its concentration
dependence was smaller than that of the force.
Therefore, the reduction in the intracellular Ca
level alone does not account for the dissociation
of the force and velocity during relaxation.

PH 86

PROPERTIES OF A HIGH-MOLECULAR WEIGHT CAL¬
CIUM BINDING PROTEIN ISOLATED FROM PLASMA
MEMBRANE-ENRICHED FRACTION OF A MOLLUSCAN
SMOOTH MUSCLE

T . Yamanobe1 , T.Mimura1 and H.Sugi2.

1Central Lab. Analyt. Biochem. and zDept.
Physiol., Sch. Med., Teikyo Univ., Tokyo.

In the anterior byssal retractor muscle
(ABRM) of Mytilus edulis, Ca ions are accu¬
mulated 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 charac¬
teristic of the plasma membrane-enriched
fraction of ABRM (Yamanobe, Mimura and Sugi,
1985). The protein exhibits Ca dependent
changes in the ultraviolet circular dichroic
spectrum. The downward shift of the spect¬
rum by Ca indicates Ca-induced increase in
the a -helix content of the 450K protein.
And the protein exhibits distinct changes
in fluorescence and UV absorption by the
change in pCa from 7 to 5.

These results are consistent with the view
that the 450K protein may be involved in the
regulation of mechanical activity of the ABR
M in physiological range of pCa. Experiments
are progress to elucidate the physiological
role of the 450K protein in the contraction-
relaxation cycle of the ABRM.

PH 88

STRUCTURE AND PROTEIN COMPOSITION OF
Z-SEGMENTS.

M. Kuroda, T. Matuno and H. Ishida.
Dept, of Biology, Fac. of Sci., Shimane
Univ. Matsue.

Major body of contractile proteins
were removed by extracting myofibils with
low concentrations of lactic acid. When
the acid-extracted fibrils were viewed
under a phase microscope, scaffolding
structure of myofibrils with un-extracted
Z-disks were detected (called as Z-
segments). The structure was quite
similar to Kl-extracted ghosts from
myofibrils. Immunoblotting combined with
SDS-PAGE showed that actin, alpha-actinin
and desmin were major constituents of the
stucture. Myosin and a M-line components
persisted even after the repeated
extraction of the segments with lactic
acid. The result of immunof louresence
microscopy indicated that actin and
alpha-actinin exclusively localized at
Z-disks of the segment. On the other
hand, anti-desmin stained both Z-disks
and inter-Z-disk spaces where no obvious
structure was shown under phase
microscope. Electron microscopic images
of the segments revealed branching
filaments which interconnected the
adjacent Z-disks.

1214

Physiology

PH 89

PURIFICATION AND LOCALIZATION OF SPECIFIC
PROTEIN COMPONENT DECREASED IN ATROPHIED
SOLEUS MUSCLE IN RAT,

Y. ATOM I, S.YAMADA, H.HATTA and R.STROHMAN1
Dept, of Sports Sci . College of Arts and
Sci. Univ. of Tokyo, 1 Dept. Zool. Univ.
of CAL.

Hypoactivity atrophy in rat skeletal
muscle was induced by the tail suspension
and the protein components of the atrophi¬
ed soleus muscle were compared with those
of the control muscle by one- and two-dim-
entional gel electrophoresis. A protein
component, referred to as 24-kDa protein
significantly decreased in quantity in the
suspended soleus. 24-kDa protein was pur¬
ified by gel filtration and ionchromatogr-
aphy. Immunoprecipitation of 24-kDa prot¬
ein with its antibody in vitro culture of
muscle cells provided an evidence showing
that 24-kDa protein is synthesized in the
muscle cells and the rate of its synthesis
is significantly smaller in the atrophied
muscle, however, that in the stretched and
suspended muscle was twice higher compared
with the control one after 1 week suspen¬
sion. FITC-staining with anti 24-kDa pro¬
tein antibody in glycerinated myofibril
showed the localization at Z line of myo¬
fibrils. From these results, 24-kDa pro¬
tein which fractionated in low salt buffer
and might bind to myofibril could be used
as a clue to solve the mechanism of soleus
muscle atrophy.

PH 90

IMAGE ANALYSIS OF MYOFILAMENT LATTICE IN FROG
SKELETAL MUSCLE FIBERS SLOWLY STRETCHED DURING
CONTRACTION.

T. Tsuchiya, S. Suzuki, Y. Oshimi, T. Takei and
H. Sugi. Department of Physiology, School of
Medicine, Teikyo University, Tokyo.

Frog single muscle fibers were fixed with 2%
OsO^ in resting, during isometric tetanus and
during tension enhancement after stretch (from 2.3
to 2.6 um in sarcomere length) to examine whether
the force enhancement after slow stretch of the
isometrically tetanized skeletal muscle correlated
with the distortion of myofilament lattice. The
transverse section was observed with electron
microscope (JEOL 100CX, 2000FX) and the array of
myofilament lattice was examined using image
analyzer (Toshiba Tospix II). Fourier transform
and rotation averaging of the transverse section
image indicated that the distortion of myofilament
lattice was larger in the fiber stretched in
tetanus than in the isometrically tetanized fiber
and in the relaxed fiber. The isometrically
tetanized fiber showed the best hexagonal array in
the above three states. The geometrical
measurement of displacement of actin filament
from the position stabilized by electrostatic
repulsion force in myofilament lattice also
showed similar results. It is concluded that the
distortion of myofilament lattice may be
associated with the force enhancement of the
tetanized fiber by slow stretch.

PH 91

EFFECTS OF TEMPERATURE ON THE ELECTRICAL
AND MECHANICAL PROPERTIES OF FISH SWIM-
BLADDER MUSCLE.

T . Kobayash i 1 , K.Ooiwa1* A. Goto1, Y.Tani-
guchi3 and H.Sugi1. 1Dept. of Physiol.,
Sch. of Med. Teikyo Univ. , Tokyo and
=Dept . of Cell Biol., Sch. of Med., Tokai
Univ . , I sehara . _ _

The effects of temperature on the mem-

brane potential and force of a swimbladder
muscle in a sound producing fish, Sebast i s-
cus marmoratus were studied. The muscle
fiber bundles exhibited extremely rapid
twitches lasting 15 msec at 20 °C . The iso¬
metric twitch force was maximum at about 12
°C, and decreased with decreasing tempera¬
tures. With repetitive stimuli of 200 Hz,
the twitches summated to produce a larger
force, but steady tetanic force was not
maintained due to extremely rapid spontane¬
ous relaxation. In the electrical
properties, two types of muscle fibers were
found. The fibers of one type gave propa¬
gated action potential with small or no
overshoot, while the fibers of other type
never showed it, but an abortive spike was
seen. The amplitude of action potential of
some fibers decreased monotoni cal ly , but
that of other fibers did not with decreas¬
ing temperatures. On the other hand, the
isometric force in Ca2+ activated chemical¬
ly skinned fibers was not affected by
temperature changes. These results suggest
that there are multi-fibers in swimbladder
muscle, and part of the E-C coupling
processes is inhibited at low temperature.

PH 92

IDENTIFICATION OF TWO CARDIO-ACCELE RATORY
NEURONS IN AN ISOPOD, BATHYNOMUS DOEDERLEINI .
K. Tanaka and K.Kuwasawa. Dept, of Biol.,
Tokyo Metropolitan Univ., Tokyo.

The 1st and 2nd cardio-acceleratory axons
( CAl and CA2 ) arise from the 3rd roots of,
respectively, the 2nd thoracic ganglion
( TG2 ) and the drd thoracic ganglion (TG3).
While impulse activities of candidate cell
bodies for CAl or CA2 located at the center in
a posterior half of the ganglion were intra-
cellularly recorded, impulse activities of
single CAl or CA2 axon were extracellularly
recorded and each of the axons was stimulated
at the periphery. Impulses recorded from
the candidate cell bodies show a one-to-
one manner to, respectively, peripheral
impulses and antidromic stimulus pulses.
EPSPs and IPSPs were recorded from them
in response to stimulation of connectives
anterior or posterior to TG2 or TG3 .

Cobalt ions and lucifer yellow dye were
iontophoretically injected into candidate
cell bodies for CAl and CA2 . Each of the
cell bodies is 50 pm in diameter. Their
axons run from the ventral side to the
dorsal side and then to basal regions of
the 3rd roots arising from their own
ganglia, and bifurcate there. In either
one of the two neurons, one branch runs to
the anterior ganglion and another one, into
the 3rd root of the own ganglion.

From these electrophysiological and
anatomical results, we identified the
candidate cell bodies with CAl and CA2
neurons themselves.

Physiology

1215

PH 93

SYNCHRONIZATION MECHANISMS OF AURICULAR
BEATS IN AN OYSTER CRASSQSTREA GIGAS
H. Yamagishi , H. Uesaka7 and A . Ebara
iInst. of Biol. Sci., Univ. of Tsukuba ,
Tsukuba. 2FACOM Dep . FACOM HITAC Ltd.
Tsukuba Bra., Tsukuba. _

Two auricles and a ventricle of the
oyster beat coordinately i n situ when
perfusing the pericardial cavity. We have
reported that coordination of the beat
rhythms between auricle and ventricle was
achieved by their mechanical interaction.
To understand the mechanism of
synchronization of the auricular beats,
interaction between two auricles was
examined .

Two auricles were linked with each
other through several myocardial bundles
at the portion near their venous ends.
Myocardial cells in the two auricles were
electrically coupled and action potentials
in the two auricles were synchronized by
their electrical interaction.

Even when electrical interaction
between the two auricles was eliminated by
cutting the junction between them, the
auricular beats became synchronized by
repetitive ventricular stretching. This
could be predicted from the mechanical
interaction between the auricles and
ventricle .

Ventricular stretching may be
fundamental in synchronization of
auricular beats in the heart of many
bivalves where two auricles are separately
connected with the ventricle.

PH 95

NEURAL REGULATION OF THE HEART BEAT IN A
PULMONATE ACHAT INA FULICA.

M.Kobayashi, Y.Furukawa and K.Hori. Fac .
of Integrated Arts and Sci. , Hiroshima
Univ. , Hiroshima.

Nine neurons regulating the heart beat
were identified in the central nervous
system of Achatina fulica.. A pair of
neurons in the cerebral ganglia (CG) was
found to have excitatory connections with
some heart excitatory neurons including a
periodically oscillating neuron (PON) and
three tonically autoactive neurons (3 TANs)
in the suboesophageal ganglion. Firings
of the CG neurons or the application of
serotonin (5-HT) produced a slow depolariz¬
ing potential in PON which was found to be
due mainly to the decrease of background
K+- conductance . The depolarization was de¬
pressed by a 5-HT antagonist, methysergide .
Similar phenomena were also observed be¬
tween the CG neurons and 3 TANs. By con¬
trast, the application of the peptide Phe-
Met-Arg-Phe-NH2 (FMRFamide) induced a
hyperpolarization in PON and 3 TANs and
inhibited their activities . Action poten¬
tials of PON and 3 TANs were broadened by
firings of the CG neurons and 5-HT appli¬
cation, whereas FMRFamide shortened them.
These results suggest that the excitatory
neurotransmitter of the two CG neurons to
PON and 3 TANs is 5-HT and FMRFamide may
act inhibitorily at these synapses. The
broadening of action potentials may be ef¬
fective for more release .of transmitters
resulting in the enhancement of heart beat.

PH 94

HISTOCHEMICAL ANALYSIS OF AMINERGIC NERVOUS
SYSTEMS IN THE PERICARDIUM AND THE HEART OF
LIOROPHURA JAPONICA.

S.Matsumura and K.Kuwasawa. Dept, of Biol.,
Tokyo Metropolitan Univ., Tokyo.

PH 96

HIBERNATION-RELATED CHANGES IN CARDIAC
FUNCTION AND SERUM COMPONENTS.

N.Kondo' and J.Kondo . ^Dept. of Muscle
Physiol. Mitsubishi kasei Inst, of Life
Sci., Machida, Tokyo and ^Mitsubishi Kasei
Corp. Res. Center, Yokohama, Kanagawa.

The pericardium and the heart receive ex¬
citatory innervation from both the lateral and
ventral nerve cordsClnc and vnc). Bath-applied
5-HT (threshold, 10"°M) produced excitatory
effects on the pericardium and the heart.

Yellowish-green fluorescent fibres were
observed in whole-mount preparations of the
pericardium and the heart. The fibres
possessed swelling structures in their routs.
Yellowish-green fluorescent cells were
observed in frozen sections of Inc and vnc.

Serotonergic nervous plexuses were
revealed by a immunohis tochemical technique
using rabbit anti-serotonin antiserum in
whole-mount preparations of the pericardium
and the heart. Cell body-like swelling
structures were 3.9±1.4pm and 4.4±1.6pm in
diameter, respectively in the pericardium
and the heart in the immunoreact ive neural
plexuses. In paraffin sections of Inc and
vnc, immunoreact ive cells ( 4-10pm in
diameter ) exist in clusters.

Results suggest that the pericardium and
the heart receive serotonergic excitatory
innervation from the nerve cords either
directly or through the plexuses . Functions
of the neural plexuses remain to be examined.
The presence of the neural plexuses may show
phylogenically primodal innervation of the
pericardium and the heart in Mollusca.

From our previous finding that the
changes in cardiac properties observed in
hibernating chipmunks occur before entering
into hibernation, the possible involvement
of as yet unknown substance(s) in
hibernation has been suggested (Kondo,
1987). Therefore, we attempted to search
component (s) in serum associated with the
changes in cardiac function in hibernating
and nonhibernating states by using high-
performance liquid chromatography (HPLC)
and electrophysiological techniques. In
serum from hibernating animals, the
fraction corresponding to M.W.140K was
found to be much less than that from non¬
hibernating animals during summer season.

In many of sera from nonhibernating animals
during winter season, this fraction was
also reduced. In cardiac muscles from these
animals, the electromechanical responses
were similar to those observed in hiberna¬
ting preparations which were characterized
by a reduced amplitude of action potential
plateau and a frequency-dependent decrease
in contraction. The present results
indicate the existence of hibernation-
related component (s) in serum associated
with changes in cardiac function.

1216

Physiology

PH 97

PH 99

THERMORECEPTION AND THERMOTACTIC BE¬
HAVIOR IN PARAMECIUM CAUDATUM
T. Tominaga and Y. Naitoh, Inst. Biol.
Sci., Univ. Tsukuba, Ibaraki 305.

Thermotactic distribution of the
specimens of Paramecium caudatum in an
area, half of which has a temperature
different from the other half, depends
on the kinetic activities of the
specimens in each half of the area and
on the phobic responses exihibited by
the specimens upon their encountering a
border between the two halves. To un¬
derstand mechanisms of the phobic
responses at the temperature border, we
examined membrane potential responses of
the specimen to the heat applied locally
to the cell surface. Tip of a 3M KCl-
filled glass microcapillary electrode
was placed close to the cell surface,
then 70 KHZ current was passed through
the electrode to produce the heat at its
tip. Thus a limited area of the cell
surface could be heated up. When an¬
terior end of the cell was heated, a
transient hyperpolarization took place.
Upon cessation of heating a transient
depolarization was observed. On the
contrary, heating of the posterior end
produced a membrane depolarization fol¬
lowed by a Ca spike. Upon cessation of
heating a transient hyperpolarization
was seen. Degrees of these electric
responses increased with increasing both
intensity and area of heating.

CONTROL OF THE ORIENTATION OF CILIA ON THE
MEMBRANE FRAGMENT OF TRITON-GLYCEROL-
EXTRACTED PARAMECIUM CAUDATUM.

M. Noguchi and Y. Nakamura. Dept, of
Biol,, Fac. of Sci., Toyama Univ., Toyama.

We examined the direction that the
cilia pointed when observed from above.

In order to observe the cell surface from
above, ciliated sheets of surface struc¬
ture were formed by mechanical fragmenta¬
tion of the cell model. The ciliated
sheets which stuck to the glass surface
were perfused successively with reactiva¬
tion solutions. Cilia gradually changed
their orientation clockwise from 12
o'clock to 5 o'clock (with the anterior of
the cell defined as 12 o'clock), when Ca^+
concentration was decreased gradually from
10-6 M to 10“^ m. Cyclic AMP changed the
orientation of cilia from 12 o'clock to 4
o'clock in the presence of 10“^ m Ca^+.

The top view of the reorienting cilia
revealed that the orientation of cilia is
controlled by Ca^+ and cAMP within the
range of 12 to 5 o'clock. When the
ciliated sheets were perfused with a tryp¬
sin containing solution in the presence
either EGTA or cAMP, posteriorly oriented
(4-5 o'clock) cilia gradually changed
their orientation counterclockwise and
finally pointed toward the direction
between 12 and 2 o'clock depending on the
Ca2 + concentration. This suggests that
some component of the Ca^+ sensitive
mechanism(s) retains its function after
trypsin digestion.

PH 98

THE MECHANISM OF RHEOTAXIS IN PARAMECIUM
CAUDATUM.

K. Itogawa and K. Taneda. Dept, of Biol.
Fac. of Sci., Kochi Univ., Kochi.

The motile behavior of a specimen showing
rheotaxis in the current of a test solution
was recorded on videotape. The picture was
analyzed with reference to the angular
change in the direction of the longitudinal
axis of the specimen. It was found that the
motile tracks of the specimens consisted of
two phases, rheotactic and non-rheotactic .
The angular rate of upstream orientation
(orientation rate) was measured only during
the rheotactic phase. The orientation rate
increased with increasing flow rates of the
surrounding medium. Culture age was not a
factor in the orientation rate. The CNR
mutant ( cauda tum-non-reversal ) showed con¬
spicuous rheotaxis. Postural change in Ni-
immobilized specimens flowing down in the
current was also recorded on videotape. The
most probable mechanism of rheotaxis in
Paramecium was discussed on the basis of
the results. The model in which the up¬
stream torque is produced by the two forces
(gyration and flowing) acting at the
different centers was best supported by the
results. The other additional force of the
torque seems to be caused by the asym¬
metrical shape of the cell.

PH 100

DISTRIBUTION OF IONIC CHANNELS IN THE
MEMBRANE OF NOCTILUCA MILIARIS
K. Oami and Y. Naitoh, Inst, of Biol.
Sci., Univ. of Tsukuba, Ibaraki 305.

Marine dinof lage 1 late Nocti luca mi liaris
exhibits spontaneous membrane potential
responses wlric-h regulate movement of
food gathering tentacle (TRPs). The
TRPs include a depolarizing Na spike and
a hyperpolarizing Cl spike. We examined
distribution of ionic channels in the
membrane responsible for the spikes by
measuring regional impedance change and
localized current. A decrease in the
impedance occurred in association with a
depolarizing Na spike restrictedly at
the region around the cytostome and the
tentacle. An impedance decrease during
hyperpolarizing Cl spike was observed
anywhere on the cell surface. An inward
membrane current was always associated
with a Na spike when the external¬
recording electrode was placed around
the cytostome and the tentacle. The
current was outward when the electrode
was placed on the other membrane area.
Diphasic membrane current was recorded
anywhere on the cell surface in associa¬
tion with the Cl spike. We concluded
that depolarization-sensitive Na chan¬
nels were localized at the region around
cytostome and the tentacle, while
hyperpolarization-sensitive Cl channels
were present anywhere on the cell sur¬
face .

Physiology

1217

PH 101

CONTRACTION OF VORTICELLA IN RESPONSE TO
A MECHANICAL STIMULUS I. LOCALIZATION OF
MECHANORECEPTION MECHANISM
K. Katoh and Y. Naitoh, Inst. Biol.
Sci., Univ. Tsukuba, Ibaraki 305.

A fresh water protozoan Vorticel la ex¬
hibits a contraction (coiling of the
stalk plus shrinkage of the cell body)
in response to a mechanical stimulation.
The response is of all-or-none type.
Mechanical threshold for the contraction
was the smallest at the distal end of
the cell body and largest at the
proximal end of the stalk. The mechani¬
cally induced contraction was indepen¬
dent of the external ions. That is, the
contraction was seen in K+-rich solu¬
tion, Ca2+-rich solution, distilled
water and even in an EDTA-containing
solution. Thses results indicate that
activation of ionic channels in the cell
membrane is not involved in the
mechanoreception for the contraction.
When the cell was extracted with 0.01 %
Triton X-100 at 20°C, the mechanical
threshold increased rapidly with time,
and became infinite in 7 min. External
application of Ca2+ ions (10-4M)
produced the contraction in the 7 min-
extracted cells. This fact indicates
that contractile mechanism remained un¬
disrupted by the extraction. We con¬
cluded that the mechanoreception
mechanism for the contraction resides in
membrane systems of the cell.

PH 102

DIRECTIONAL EFFECT IN INDUCED ROTATION OF
THE FLAGELLAR BEAT PLANE IN SPERM WITH IM¬
POSED HEAD VIBRATION.

Chikako Shingyoji, Jun Katada, Keiichi Takahashi
and I. R. Gibbons. Zool. Inst., Fac. Sci., Univ.
Tokyo, Tokyo, and Pacific Biomed. Res. Center,
Univ. of Hawaii.

When the head of a sea-urchin sperm is held
by a suction pipette and vibrated laterally, the
flagellum becomes to beat in phase with the im¬
posed vibration. Rotating the plane of vibration
around the head axis induces rotation of the
beat plane ("winding"). Termination of the
vibration after the winding is followed by a
spontaneous "unwinding" of the beat plane
(Gibbons et al., Nature, 325: 351-352, 1987). We
studied the effect of the direction of winding on
the induced rotation and unwinding of the beat
plane in the spermatozoa of the sea urchin,
Hemicentrotus pulcherrimus. The beat plane could
be wound up for up to 10 revolutions in both
clockwise (cw) and counterclockwise (ccw) direc¬
tions. However, a nearly complete unwinding was
observed only after a ccw winding; the number
of unwinding cycles was almost equal to that of
the winding cycles after a ccw winding but was
much less after a cw winding (less than a few
cycles even after 10 cycles of winding). The
maximum angular velocities of spontaneous un¬
winding were, however, almost the same in both
directions. The results indicate that the
mechanism controlling the flagellar beat plane has
rotational directionality, the nature of which is,
however, still unknown.

PH 103

RESPONSE OF REACTIVATED SPERM FLAGELLA TO
IMPOSED HEAD VIBRATION.

Jun Katada, Chikako Shingyoji and Keiichi
Takahashi. Zool. Inst., Fac. Sci., Univ. Tokyo,
Tokyo.

Gibbons et al. (Nature, 325: 351-352,1987) have
shown that, by holding the sperm head with a
suction pipette and vibrating the pipette laterally
in a plane that rotates around the head axis, the
flagellar beat plane can be rotated through
several cycles of revolution. Spontaneous unwind¬
ing of the beat plane occurs on cessation of the
vibration. To study the mechanism of the rotation
of the beat plane, we applied rotatory vibration
to the heads of reactivated spermatozoa of the
sea urchin, Hemicentrotus pulcherrimus. In the
conventional Triton-extracted sperm, we could not
rotate the beat plane by more than 360 degrees.
Substitution of Nonidet P-40 for Triton X-100 in¬
creased the proportion of the sperm showing
rotation of the beat plane, but the plane rotated
through only a few cycles. Demembranation with
the zwitterionic detergent 3-[ (3-cholamidopropyl)-
dimethyl-ammonio]-l-propanesulfonate (CHAPS) as
the sole detergent did not give good results be¬
cause of a low reactivation rate, but the sper¬
matozoa demembranated with a solution containing
0.04% CHAPS and 0.01% Nonidet P-40 showed a
high reactivation rate, and the reactivated sper¬
matozoa responded to the imposed rotatory vibra¬
tion by rotating their flagellar beat plane
through more than 10 cycles of revolution. It is
expected that the new method of demembranation
will be useful for the study of the mechanism
controlling the flagellar beat plane.

ph m

EFFECTS OF THE LENGTH OF A FLAGELLUM ON ITS
MOVEMENT IN DEMEMBRANATED SEA-URCHIN SPERM.
Y. Taniguchi, Y. Mogami and S. A. Baba.
Dept, of Biology, Ochanomizu Univ., Tokyo.

We analyzed the movement of a flagellum
of various length to clarify the length
effect on flagellar movement. Sperm from
Hemicentrotus pulcherrimus were

demembranated with Triton X-100, severed by
pipetting and reactivated with ATP of 0.2-
1.0 mM under either potentially asymmetric
(PA) or symmetric (PS) conditions. The
beat frequency of severed flagella
decreased linearly with the flagellar
length up to about half the wavelength of
full length flagella as reported with live
sperm by Baba, S. A. , Mogami, Y. , Yogo, A.
and Imamura, Y. (Zool. Sci. 4: 972, 1987).

The relation between the beat frequency of
5-pm-long flagella determined from the
regression line for flagella shorter than
the critical length and the ATP
concentration is found to be similar in
form to the Michaelis-Menten law of enzyme
kinetics. Under PA conditions Kn was
determined to be 0.23 mM , which was not
much different from 0.25 mM for the full
length flagella. However, VBax was 45.5 Hz
and therefore significantly higher than
38.5 Hz for the latter (under PS conditions
reliable values for K» and Vmax were not
obtained). The dependence of kinetic
parameters on the flagellar length
described here may provide criteria in
simulation experiments with many proposed
sliding-bending conversion models.

1218

Physiology

PH 105

EFFECTS OF THE SEMINAL PLASMA MOTILITY
INHIBITOR ( SPMI ) AND POLY AMINO ACIDS ON
MOTILITY OF THE DEMEMBRANATED SEA URCHIN
SPERM FLAGELLA.

M. Okuno. Dept. Biol., Coll. Arts & Sci.,
Univ. Tokyo, Tokyo.

In order to get deeper insights, into
the mechanism of the inhibition of SPMI on
flagellar movement, we looked at the
effects of poly amino acids on motility of
the demembranated ' sea urchin sperm
flagella. The present experiments showed
that only the basic poly amino acids,
especially the high molecular weight (15 KD
or more) poly-L-Ly sine (PLL) inhibited the
motility of the demembranated flagella.
When the concentration of PLL was changed,
the flagellar movement stopped suddenly at
a. critical concentration (approximately
0.5 pg/ml). The motility and the beat
frequency did not change much at the
concentrations lower than the critical
concentration, however, the bend angle
decreased with the concentration. The
rigor bends of flagella incubated with PLL
were preserved even when MgATP was added.
The sliding disintegration of doublet
microtubules was also inhibited by PLL.
However, ATPase activity of the axoneme"was
not inhibited by PLL. Since the feature of
the several effects of SPMI on flagellar
movement were very similar t'o those of PLL,
it was likely that both, of the molecules
interacted with the flagellar axoneme and
inhibited the motile mechanism of flagella
in a similar manner.

PH 107

MEASUREMENT OF INTRACELLULAR CA2+ IN THE
RAINBOW TROUT SPERMATOZOA USING FLUO¬
RESCENCE MICROSCOPE/VIDEO-CAMERA SYSTEM
M. Morisawaj Y. Kudci T. Inoda and S.
Tanimoto . 1Misaki Mar. Biol. St. Univ
Tokyo, Miura,„ Mi t sub i sh i -Kase i Inst. Life
Sci., Tokyo, JFac. Sci. Univ, Toho , Chiba.

2 +

Using a Ca -indicator fura-2, we foun<jl
an elevation of intracellular CaZ
concentration in response to the initiation
of trout sperm motility. One volume of
trout semen was immersed in 20 valumes of
artificial seminal plasma (£SP) containing
10-1 5uM fura-2 and 40mM K and incubated
for 1 hour. After washing out the fura-2
with ASP, fura-2 loaded sperm were
suspended in the medium containing 150mM
NaCl and 5mM KC1 . Spermatozoa wer^
completely immotile in the medium by K
and the fluorescence was showed clearly in
head region but slightly in flagellum.
Intracellular Ca^ in immotile sperm was
around 300 nM, when calcium concentration
was measured by a system for real-time
quantitative monitoring of intracellular
free calcium concentration (Kudo, Y &
Ogura, A., Br . J . pharmac , 1986). Calcium

concentration increase! to around lOOOnM
when extracellular Ca^ increased, and then
spermatozoa initiate motility. These
results sugges£+ that the increase in
intracellular Ca^ is indispensable for the
initiation of flagellar motility in rainbow
trout spermatozoa.

PH 106

CHANGES IN MEMBRANE POTENTIAL AND
INITIATION OF SPERM MOTILITY IN RAINBOW
TROUT. - 2 3

S. Tanimoto1, Y. Imae and M. Morisawa .
TFac. of Sci., Toho Univ., Chiba, Inst, of
Mol. Biol., Fac. of Sci., Nagoya Univ.,
Nagoya, 3Misaki Mar. Biol. St., Fac. Sci.
Univ. Tokyo, Miura.

+ 2 +

We have reported that K efflux and Ca"
influx across the plasma membrane occur in
response to the initiation of sperm

motility in rainbow trout. The fact
implies the generation of membrane

potential, thus, we measured the potential
using the JH-TPMP and the TPP electrode
methods. The negative potential became
large accompanied wi|h the initiation of
sperm motility in K free medium, and

reached plateau at 3-5 min. This level

maintained for at least 20 min, suggesting
that there is no significant ion flux any
more after negative potential was

generated. The negative potential is
generated in LiCl or choline chloride
solution as well as NaCl solution. Thus
the generation of potential primarily
depend on the external K and independeg|
of other monovalent cations. External Ca
which accelerates sperm motility reduced
the K+ dependent negative pot^ijtial. These
suggest that K efflux and Ca" influx are
related to the potential generation and the
initiation of sperm motility in rainbow
trout.

PH 108

EFFECTS OF Ca2+, CYCLIC AMP, CAFFEINE AND
MgATP2- ON FLAGELLAR MOVEMENT OF
DEMEMBRANATED HAMSTER SPERMATOZOA.

S. Ishijima. Biological Laboratory, Tokyo
Institute of Technology, O-okayama,
Meguro-ku, Tokyo.

Mammalian "spe-rm alter their beating
pattern in the female reproductive tract
(Hyperactivation) after acquiring motility
in the male reproductive tract. The
plasma membrane plays an important role on
initiating the hyperactivated motility but
the changes in beating pattern during
hyperactivation are finally based on the
changes in axoneme. To understand the
mechanisms of hyperactivation, I examined
the effects of Ca2+, cAMP, caffeine and
MgATP^- on flagellar movement of
demembranated hamster spermatozoa.
Demembranated hamster sperm beat with
frequency of about 3.5 Hz in the range of
4-10 mM MgATP2- at 20°C. Percent
motility, which was nearly 100% at low
Ca2+, decreased with increasing Ca2+,
whereas it increased with increasing cAMP.
When Ca2+ decreased and cAMP increased,
bending in midpiece became large, which
yielded similar beating pattern typically
observed in living and hyperactivated
spermatozoa. These results suggest that
Ca2+ and cAMP regulate waveform of hamster
sperm flagella, and thereby that the low
concentration of Ca2+ and/or the high
concentration of cAMP may cause the
hyperactivated motility.

Physiology

1219

PH 109

EFFECTS OF CALCIUM ON THE VOLUME OF
NEMATOCYSTS ISOLATED FROM ACONTIA OF
A SEA ANEMONE, CALLIACTIS TRICOLOR.

M. Hidaka. Dept, of Biology, Univ. of the
Ryukyus, Okinawa.

The size of nematocysts isolated from
the acontia of the sea anemone Calliactis
tricolor were measured in solutions
containing different concentrations of
Ca^+. The isolated nematocysts were
attached to a glass slide coated with
polylysine. Polyethyleneglycol (PEG) was
used to increase the osmotic pressure of
the test solutions. The volume of
nematocysts was calculated by assuming
that the nematocysts were ellipsoid. When
nematocysts were immersed in 50 mM CaC^
containing 40% PEG their volume decreased
by about 20%. There was no fold on the
capsule wall of the shrunk nematocysts.

The nematocysts re-expanded when immersed
in 5 mM EGTA and 40% PEG. Nematocysts did
not change their volume when immersed in
distilled water (DW) or 1M sodium citrate.
Nematocysts treated with DW or 1 M sodium
citrate for 30 minutes also shrank in 50
mM CaCl2- Nematocysts stained with 0.1%
toluidine blue could be discharged when
treated with DW or 5 mM EGTA. The stained
intracapsular fluid did not dissociate in
the water came from oucside of the capsule
but was excluded partially -or completely
from the capsule through the tip of the
thread.

PH 110

SEPARATION OF ACIDIC CELLS FROM ASCIDIAN
BLOOD CELLS.

H. Michibata \ J. Hirata 2, H. Sakurai 2, and Y.
Hamaguchi3. 1Biol. Inst., Fac. of Sci., Toyama Univ.,
Toyama, 2Fac. of Pharmaceut. Sci., Univ. of
Tokushima, Tokushima, and 3Biol. Lab., Tokyo Inst,
of Technol., Tokyo.

High levels of vanadium content in ascidian
blood cells have been considered in association with
the fact that the intracellular pH is extremely low.
The intracellular pH of the vanadocyte that contains
high levels of vanadium has however been disputed
recently. One of the reasons for this confusion
seems to ascribe that pH measurements have been
carried out without cell fractionation.

In the present experiments, we have therefore
examined whether there are acidic blood cells in
ascidian, Ascidia ahodori, if there are, which blood
cell type is acidic and whether the acidic cell con¬
tains a high amount of vanadium actually, using cell
fractionation technique on a Ficoll density gradient
centrifugation and neutron activation analysis for
vanadium determination. Intracellular pH was meas¬
ured with a microelectrode after the fractioned sub¬
populations were homogenized under either aerobic
or anaerobic conditions.

Consequently, the subpopulation of signet ring
cells indicated pH2.85 to 2.94, under both aerobic and
anaerobic conditions and had high levels of vanadium.
These results suggest that vanadium is accumulated
by ascidian blood cells in association with the intra¬
cellular low pH.

PH 111

STRUCTURAL CHANGES IN THE SKIN OF JAPANESE
TREEFROG DURING THE WATER INFLUX.

I. Kamishiina, and H. Nakashima. Dept, of
Biol., Facult. of Sci., Okayama Univ.
Okayama 700 _

Japanese treefrogs take water exclu¬

sively through the abdominal skin. The
water absorption through the excised skin
was enhanced or depressed by administra¬
tion of 3-adrenergic receptor agonists or
by prolactin pretreatment, respectively.
Keeping the animal under dehydrated condi¬
tion also caused enhancement of the water
influx through the skin. This enhancement
was hindered by administration of 0-adre-
nergic antagonist. The epithelial cells in
the skin showed morphological changes in
accordance with the water influx rate.

When excised skin was activated to per¬
meate water, granullar cells, the outer¬
most cells of the epithelium, showed dense
and amorphous cytoplasm. Granules which
characterized the cells were distended and
appeared empty, and finally became dis¬
cernible in the dense cytoplasm. The acti¬
vation of the skin also caused conspicuous
modifications on the cell surface showing
interdigitating microvilli and deep inden¬
tations of the cell membrane. Basal cells
of the epithelium also became to have long
microvillous processes, and they showed
wide intercellular space among them.

On the other hand, when the water per¬
meation through the skin was depressed,
neither the granular cells nor the basal
cells showed no microvillous processes.

PH 112

ROLE OF CALCIUM IN WATER PERMEABILITY
CONTROL IN FISH GILLS - EFFECTS OF
LANTHANUM AND CALCIUM IONOPHORE.

J. Ohtsu and J. Yamada. Dept, of Biol
and Aquacult., Fac. of Fish., Hokkaido
Univ., Hakodate.

Effects of lanthanum and calcium
ionophore, A23187, on the osmotic water
permeability of isolated gills of tilapia
(Oreochromis niloticus ) were examined
under various calcium conditions. Gill
arches incubated in various solutions were
weighted and the rate of weight increase
was used as an index of the water
permeability. Calcium in incubation media
lowered the permeability depending on its
concentration. Addition of lanthanum to
the incubation media inhibited, and of
calcium ionophore accelerated, the
lowering effect of calcium. Use of sodium
sulfide, an inhibitor of respiratory
enzymes, caused no difference of the
permeability regardless of calcium
concentrations in the incubation media.

The results seemed to show that calcium
ions affect the osmotic water permeability
of the branchial epithelium cells by
entering into the cytoplasm.

1220

Physiology

PH 113

EFFECTS OF AMBIENT AND DIETARY CALCIUM
LEVELS ON THE FORMATION AND RESORPTION OF
THE PHARYNGEAL BONE IN TILAPIA
(OREOCHROMIS NILOTICUS)

Y. Takagi and J. Yamada. Dept, of Biol,
and Aquacult., Fac. of Fish., Hokkaido
Univ., Hakodate.

The formation and resorption of the
pharyngeal bone of tilapia were studied by
histomorphometry . When fish were reared
in a low Ca water with a Ca-deficient
diet, increased bone resorption and
decreased bone formation were observed.

The mineral apposition rate of newly
formed bone was not changed. Similar but
less apparent changes were observed in
fish reared in the low Ca water with a
normal Ca diet. The former group showed
hypercalcemia, while the latter
maintained normocalcemia . The results
indicate that when ambient calcium is
deficient, fish can mobilize bone calcium
to maintain the plasma level by increasing
bone resorption and decreasing bone
formation.

PH 115

THE PLATLET-LIKE STRUCTURES (PLS) AND COAG¬
ULATION OF THE LAND SLUG HEMOLYMPH.

E.Furuta1, K.Yamaguchi2 and A.Shimozawa1
'Dept, of Anat. and2Lab. of Med. Sci. Dokkyo
Univ. Sch. of Med., Tochigi.

Soon after injection of a high concent¬
ration of foreign materials or a sham in¬
jection into the hemocoel, many PLS were
observed in the collected hemolymph of land
slug, Incilaria spp. In vitro , they attach¬
ed themselves to the foreign particles with
long fine fibers. The hemolymph could coag¬
ulate when exposed to air and coagulation
was induced by either LPS or 1 , 3-J3-D-glu-
can, both of which resulted in gel forma¬
tion. The glucan caused it in presence of
Ca , ljjTjit the reaction of the LPS pathway
was Ca -independent. Moreover, certain
serine protease inhibitors, such as DFP or
STT impaired clotting and interacted with
the clamping of the PLS. After treatment
with the glucan, the LPS, or foreign parti¬
cles, most of the PLS projected many long
fibers into each other. But, without treat¬
ment or after injection of EGTA or serine
protease inhibitors , the PLS either did not
possess the long fibers.

These results suggest that the coagula¬
tion of land slug hemolymph' is induced by
serine protease and that the enzyme is able
to be activated via two independent path¬
ways by the LPS or the glucan. Furthermore,
it suggests that this system deeply invo¬
lves the PLS and plays an important part in
host-defense reactions.

ph m

CALS I UM- DEPENDENT AGGREGATION OF HAEMOCYTES
OF POMACEA CANALICULATA (GASTROPODA:
PROSOBRANCHlAj

A.Shozawa, C.Suto and N.Kumada. Dept, of
Med. Zool Nagoya Univ. Sch. of Med., Nagoya.

Aggregation response of snail haemo-
cytes is considered to be important in
wound healing and similar response are
observed at the site of encapsulation. But
biochemical regulation of this phenomenon
remains unclear.

Pomacea haemocytes rapidly aggregate
when they are removed from the animal. We
have detected that the response depends on
extracellular Ca^4- and the aggregates can
be redispersed by adding EDTA.

In this study, we tried to inhibit the
aggregation using calmodulin antagonists
(W-7, W-5) and proteinkinase-C inhibitor
(H-7). Aggregation was assayed with a
platelet aggregometer . The aggregation was
clearly inhibited by W-7 in dose dependent
manner. W-5 did not inhibit the response.
Therefore, calmodulin- dependent Ca2+
transport may have an important role in the
cellular response. On the other hand, we
considered that proteinkinase-C might not
affect the aggregation, because the haemo¬
cytes stimulated by phorbol myristate
acetate rarely aggregated. However, the
weak inhibition of the aggregation by H-7
was observed. This indicates that acti¬
vation of proteinkinase-C may also involve
in the cellular response.

PH 116

ALLOTOLERANCE INDUCED IN THE FRY OF INBRED
STRAIN OF MEDAKA BY SCALE-TRANSPLANTATION.
T. Matsuzaki and A. Shima. Zool. Inst.,
Fac. of Sci., Univ. of Tokyo, Tokyo.

One- to 45-days-old fry of inbred strain
of the Medaka (H04C) were grafted a piece
of scale from— adults of HB32C strain in a
posterior abdomen of the recipient. The
state of the grafts was continuously
observed under a stereoscopic microscope,
and rejection was considered as being
complete when all of the melanophores on
the grafted scale had been destroyed. The
graft transplanted to fry younger than 2
weeks of age showed considerable breakdown
of melanophores during the first two days.
Three to 5 days after transplantation, the
state of grafts were relatively stable for
the younger as well as older fry.
Therefore we considered the breakdown of
melanophores until the first 2 days might
be caused by mechanical injury during the
operation. Tolerability of the recipients
to the grafts of HB32C was estimated by
comparing the number of recipients carrying
intact melanophore ( s) on the graft at 4
days, with that of 12 days after trans¬
plantation. The younger recipients became
tolerant while only a small number of the
older fry became so. Almost all tolerized
recipients were prone to accept a graft of
secondary transplantation from HB32C.
Among the recipients that had rejected the
primary graft, significant number of
younger fry could take a secondary graft.

Physiology

1221

PH 117

GROSS REACTIVITY OP LARVAL ADRENAL GRAND
OP RANA CATESBEIANA, WITH THE THYMOCYTE
SPECIFIC MONOCLONAL ANTIBODY Rc-T-1.
K.Sugiyama and N.Yamagata Dept -Biology ,
Faculty of Science, Hirosaki Univ. ,
Hirosaki-shi. _ _ _

Nevertheless a' reactivity of mAb Rc-T-1

was exclusively found on the thymocytes
in adult frogs, it was also detectable in
the larval mesonephric region during meta
-morphosis. In the aspects of morphology
,this immunoreactive tissue was consider¬
ed as the cortex of metamorphic adrenal
gland, in which many of foliclic structu¬
res were included. The cells prepared
from this tissue were partially similar
to those of larval thymus, both in morph¬
ological and immunological features, as
examined by indirect fluorescent staining
. These observatios lead us to speculate
an altered posibility concerning hemopoi¬
esis, that the thymocytes might be dreived
from the stem cells via adrenal gland,
in which unfated stem cells would be pri¬
marily affected to be thymocytes. These
similarities, however, may suggest solely
an existence of the same antigen, as a
steroid hormon receptor.

PH 118

CROSS IMMUNITY BETWEEN HYMENOLEPIS MICRO¬
STOMA AND H. NANA IN MICE.

K. Onitake’T, J. Andreassen2 and A. Ito3.
"'Dept, of Biol., Fac. of Sci., Yamagata
Univ., Yamagata, 2pep-t. 0f Parasit., Inst,
of Popu. Biol., Univ. of Copenhagen, Den¬
mark and 3pept. of Parasit., Gifu Univ.

Sch. of Med., Gifu. _

When BALB/c mice inoculated with 5 cysti-
cercoids of Hymemolepi s microstoma (day 0)
were challenged with H. nana , they showed
strong resistance to challenges with both
eggs and cysticercoids of H. nana from day
20 onward. The immunity - shown not to be
premunition - became complete from day 30
onward: No tissue cysticercoeds nor lumenal
adults of H. nana were established in these
mice. However, when H. nana challenge was
done on day 30 to mice, which were infected
with H. microstoma (day 0) but from which
the established immatrue H. microstoma were
removed by an anthelimintic on day 10, the
immunity evoled was directed exclusively
against the lumenal phase of the cysti-
cercoid challenge but not against the tis¬
sue cysticercoids of the egg challenge.

When mice experienced the prepatent infec¬
tion with H. microstoma twice, the immunity
to cysticercoid challenge became complete,
but against egg challenge the oncospheres
developed into tissue cysticercoids in the
intestinal wall but thereafter completely
failed to develop into lumenal adult tape¬
worms. It is concluded that a correlation
exists between the presence of stage speci¬
fic antibodies and resistance to the diff¬
erent antibodies take part in the host.

PH 119

HEAT SHOCK PROTEINS IN MOUSE MACROPHAGES
CULTURED IN VITRO; A PRELIMINARY REPORT.

C. Tachi . Zool. Inst., Fac. Sci., Univ.
Tokyo, Tokyo.

Macrophages (M0) , as one of the crucial

members of the defense mechanisms in
animal's body, are exposed to variety of
stimuli which affect the cells adversely,
leading, in all likelihood, to the produc¬
tion of the heat shock proteins (HSP). In
turn, HSP might act to modify the func¬
tions of M0. We reported previously
(Tachi, C., 1987; Tachi, C. & Tachi, S.,
in press) that M0 exist in two different
cellular shapes, i.e., rounded and elon¬
gated, in the blastocyst-M0 co-culture,
and that M0 in direct contact with the
embryos are invariably rounded. Further¬
more, the alkaline conditions of the
medium caused rounding, while the acidic
medium, or Ca-ionophore added to M0 in
the alkaline medium, induced the
elongation of M0 (Tachi, C. & Zor, in
press). Since HSP90 is associated with
actin in cells (Koyasu et al., 1 986), it
might be involved in the morphological
changes of M0. Immunohistochemical detec¬
tion of HSP90 in M0 under different envi¬
ronmental conditions, was attempted using
anti-mouse HSP90 antisera raised in
rabbit (kind gift from Dr I. Ihara, Tokyo
Met . Inst . Med . Sci ., Tokyo ) . It was suggested
that in the rounded M0, the rate of HSP90
production is much enhanced, while in the
elongated M0, it is reduced.

PH 120

FRESH HUMAN SERUM KILLED MANY CELL CULTURES
EXCEPT HUMAN CELLS

Kazuhiko Kaji. Integrated Res., Tokyo
Metro. Inst. Gerontol. Tokyo

Usually bovine serum has been used for
cultivation of many cell strains derived
from different species. The choice of
species from which serum is obtained has
been left out of consideration. In testing
the effect of human serum on cultured
cells, we found that fresh human serum
killed many cell cultures from various
species except cells of human origin.
Though all of 11 fresh human sera tested
from different male and female adult donors
(20 —"4 5 yrs.) strongly supported cell growth
of human fibroblasts and endothelial cells,
these sera killed bovine endothelial cells
and smooth muscle cells, rat smooth muscle
cells, mouse fibroblasts and chick
fibroblasts, with different degrees of the
strength depending on the donors.
Surprisingly, these sera also killed
Paramecium tetraurelia. Human umbilical
cord serum (2 of 3 tested ) showed the same
cell killing effect. The killing effect on
cultured cells started very rapidly. Within
10 min after addition of the serum, bubbles
came out from the surface of cells. Then,
the cells were detached from the substratum
and died in 2 hrs . As human plasma also
killed cells like serum, the killing
factor (s) were not present in platelets.
Heat treatment (56 °C, 30 min) of the serum
inactivated the killing activity.
Pretreatment of human serum with anti human
IgG or IgM neutralized the killing
factor(s). These results suggest that
human adult and sometimes even fetus serum
contain natural immunoglobulines against
many other species including Paramecium and
that the immunoglobulines and complement
(heat unstable) attack cells. On the
other hand, bovine adult and fetus sera did
not show any killing effect on many cell
strains so far tested.

1222

Genetics

GE 1

Sex chromosome of sea urchins.

K.Saotome, Yokohama City Institute of
Healthy Yokohama. _

The problem about sex chromosome of
sea urchins still remain unsolved.

When chromosome preparation was made
by air-drying method from Hemicentrotus
pulcherrimus embryos, two kinds of karyo-
types were observed. One had a heteromor-
phic pair consisting of large and middle
acrocentric chromosomes, and the other
carried a middle acrocentric homorogous
pair, although the remaining meta- ,
submeta-, and subtelocentric 20 pairs were
the same between the two. As the ratio of
two karyotypes is about 1:1 in 98 meta¬
phase plates obtained through four
spawning seasons, heteromorphic chromosomes
may be not polymorphism but sex chromo¬
some. To clarify this possibility, the
following chromosomes were analyzed; l)male
chromosome from andro-merogons , 2 ) chromo¬
some from hybrid (Pseudocentrotus
depressus % X H . pulcherrimus £ ) , and 3)
female chromosome from parthenogenetic eggs.
Both chromosome numbers in andro-merogons
and parthenogenetic eggs were haploid chro¬
mosome number (n=21) . There were two karyo¬
types each having large or middle acro¬
centric chromosome in merogons and hybrids,
whereas only one karyotype having middle
acrocentric chromosome: was observed in
parthenogenetic eggs. These results suggest
that _H . pulcherrimus is male heterogamety .

GE 3

CHROMOSOMAL STUDY OF RAM AND BOAR
SPERMATOZOA USING ZONA-FREE HAMSTER OVA*

'H. Tateno, * 1 K. Mikamo, 2H. Kano, 3 4M.
Akaike. 'Dept, of Biol. Sci., Asahikawa
Med. Col., Asahikawa, 2Res. Center of Nihon
Nohsan Kogyo K.K., Funabashi, 3Dept. of
Meat Animal Reprod., Obihiro Uni. of agri.
& Vet. Med., Obihiro. _ _

Recently, direct chromosome analysis

became possible for bull spermatozoa
(Tateno and Mikamo 1 987) by the use of the
interspecific in vitro fertilization sys¬
tem. In the present study, this method was
modified in order to utilize it for ram and
boar sperm chromosomes. Important points of
modification are as follows. (1) Frozen
semen samples were used in the ram, while
in the boar fresh ones or preserved ones in
Galf's solution at 1 5 °C were studied. (2)
Medium TYH was used throughout the process
of preparing ram spermatozoa and TCM1 99 for
boar spermatozoa. (3) Ram spermatozoa were
treated with 10-15 pM ionophore, while boar
spermatozoa with 4-15 pM due to a greater
individual variation in sperm sensitivity
to ionophore. The time required for sperm
capacitation was 1.5-2 h in the ram and 1-
1.5h in the boar. As observed in the bull,
process of the capacitation was reflected
well in spermatozoal movement patterns in
both species.

The success rate of the method for ram
spermatozoa was more than 50%. However, it
was between 10 and 40% in the boar. This
fluctuation may be due to difference in the
quality of individual samples.

GE 2

RESTRICTION ENZYME BANDING OF THE
MARSUPIAL I A ( DAS YURO IDES BYRNE I )

K.Ookatal, S . Konnoz , and Y,Kato*
developmental and Reproductive Biology
Center., Yamagata, 2Bio Science
Laboratory., Yamagata. _

We have examined the effects of restric¬
tion enzymes on Marsupialia (Dasyuro ides
by rne i ) metaphase chromosomes. Fixed chro¬
mosomes were treated with restriction en¬
zymes followed by Giemsa staining. HaeQI or
Alul digestion produced a C-banding pat¬
tern, whereas HindEor Hinfl revealed a G-
banding pattern. Propidium iodide ( a dye
specific for DNA and RNA ) also stained the
HaeUf- 1 reated chromosomes in a C-banding
pattern. This result supports the presum-
tion that the restriction enzyme digestion
makes chromosomal DNA fragmented and a re¬
lease of the shorts fragments results in
diminished staining by Giemsa. Moreover,
other factors, e.g.the i naccesab i 1 i ty of
DNA to a restriction enzyme, may be also
important for restriction enzyme - i nduced
banding.

GE 4

SPONTANEOUS INCIDENCE OF CHROMOSOME ABERRA¬
TIONS IN MAMMALIAN ZYGOTES

Y. Kamiguchi and K. Mikamo, Dept, of Biol.
Sci., Asahikawa Med. Col., Asahikawa.

The primary incidence of spontaneous
chromosome aberrations in 1 -cell zygotes of
the golden hamster (GH) was studied, before
the selective death of chromosomally abnor¬
mal zygotes takes place due to their genic
imbalance. The results were compared with
previous data in the Chinese hamster (CH)
and the mouse (M).

(1) A total number of 451 zygotes
obtained from 39 mated females were karyo¬
typed successfully. (2) Total aberration
frequency of GH (13.5%) was similar to that
of M (13.7%), while it was higher than that
of CH (8.0%). (3) Incidence of aneuploidy

in GH (1.4%), which consisted of 0.7% of
hyperploidy and 0.7% of hypoploidy, was the

lowest as compared with other two species.

(4) Polyploidy in GH (4.0%) occurred more
frequently than in CH (1.5%) and M (2.5%).
In GH, frequency of digynic triploidy was
about 3 times higher than diandric tri¬
ploidy, while the latter was much higher
than the former in CH. (5) Incidence of
structural aberrations in GH (8.2%) was far
higher than that of CH (2.8%) and M (1.8%).
In GH, structural aberrations of male
origin was about 5 times higher than those
of female origin. Contrary to GH, struc¬
tural aberrations of male and female origin
occurred at about one to one ratio in CH.

Genetics

1223

GE 5

KARYOSYSTEMATICS OF THE MUSTELINE SPECIES: ON THE TAX¬
ONOMIC ALLOCATION OF THE LEAST WEASELS FROM HOKKAIDO.

Y. Obara, Dept, of Biol., Fac. of Sc i . , Hirosaki Univ.,
H i rosaki .

The Japanese least weasel, Mustela nival is namiyei ,
is distributed only in Hokkaido and Tohoku of Northern
Honshu. Taking account of a Robertsonian fusion, dupli¬
cation of C-heterochromat i n and two translocations, the
Tohoku population (2n=38) was shown to highly corres¬
pond in the G-banded karyotype to the Japanese ermine,
M. erminea ni ppon (2n=44), also having a close affinity
with the Hokkaido population (2n=42) in spite of their
extensive alteration of karyotype (Obara 1985,1987).

In the present study, G- and C-banded karyotypes of
the least weasels from Hokkaido were compared in detail
with those of the Japanese ermine. Their karyological
relationships could be well explained by a Robertsonian
fusion rearrangement and duplication of C-heterochroma-
tin. Taken together the present and previous findings,
the karyological differentiation in the Japanese least
weasel was discussed in some detail in connection with
speciation, intraspecific differentiation, migration
routes and taxonomic allocation of this species.

GE 6

ROBERTSONIAN FUSION AND NOR VARIATION IN THREE SPECIES
OF VESPERTILIONID BATS.

T. Ono and Y. Obara. Dept, of Biol., Fac. of Sci.,
Hirosaki Univ., Hirosaki.

The distribution patterns of NORs in Nyctalus furvus,
N. lasiopterus aviator and Vespertilio orientalis were
examined using G- and NOR-banding techniques, with spe¬
cial attention to the behaviour of the centromeric NOR
following Robertsonian fusion rearrangements. The karyo¬
typic relationships among these species could be well
explained by the Robertsonian fusion system. The number
and distribution patterns of NORs were all different
from one another. The steady and active interstitial
NORs were detected in common on the secondary constric¬
tion of the long arm of the acrocentric pair No. 15 which
has not taken part in the Robertsonian fusion in all
species studied, while the centromeric NORs locating
on the short arm close to the centromere were found in
the acrocentric pair No. 8 of N.l. aviator. This acro¬
centric pair just corresponded in the G-banding pattern
to the long arm of the large metacentric pair of V.
orientalis which has been formed through a Robertsonian
fusion rearrangement. In V. orientalis the NORs could be
detected not in this metacentric pair, but in the small
acrocentric pair No. 23. These findings strongly suggest
that these vespertilionid species may be descendants
from a common ancestor which had a distinct secondary
contriction on the long arm of the pair No. 15, and fur¬
ther the positional variations of the NORs may also be
regarded, in addition to the Robertsonian system, as
one of the speciation factors.

GE 7

GENETIC RELATIONSHIP BETWEEN JAPANESE
AND CONTINENTAL SPECIES OF APODEMUS

K. Tsuchiyal , M. Sakaizumi2, S. Wakana3,
H. Suzuki4 and K. Moriwaki3.
iMiyazaki Med. College, Miyazaki, 2The
Tokyo Metropol. Inst. Med. Sci., Tokyo,
3Cent. Inst. Exp. Anim. , Kawasaki,

4The Jikei Univ. , School of Med. , Tokyo
and 3Natl . Inst. Genet., Mishima.

Three Apodemus species are recognized
in Japan. Among them the chromosomal
constitution of A. giliacus distributed
in Hokkaido is similar to that of a
Korean species A. peninsulae .

Biochemical, ribosomal DNA(rDNA) and
mitochondrial DNA(mtDNA) analysis
revealed that the genetic difference
between A. giliacus and A. peninsulae was
very small, throwing doubt on the
validity of A. giliacus as an independent
species. Major genetic differences
suggested a long divergence times among
A. speciosus , A. argenteus and A.
peninsulae .

Sequence divergence of rDNA and mtDNA
classified the seven Apodemus species
examined into six groups; speciosus ,
argenteus , peninsulae , agrarius , semotus
and f lavicollis-sylvaticus . No species

pairs showed close connections among
these six groups.

GE 8

Polymorphism of mitochondrial DNA in Jpanese
Apodemus based on restriction cleavage
patterns

S. Wakana1, H. Suzuki2, M. Sakaizumi3, K.
Tsuchiya4, & K. Moriwaki5.

'Central Inst. Exp. Animals., Kawasaki, 2The
Jikei Univ. School of Med. Sci., Tokyo, 3The
Tokyo Metropol. Inst. Med. Sci., LMiyazaki
Med. College, 5 N a 1 1 . Inst. Genet., Mishima.

The cleavege patterns of mitochondrial
DNAs (m tDN'As) from two Japanese species of
Apodemus ; Apodemus speciosus and Apodemus
argenteus ,were investigated using 10
restriction endonucleases. We found that
Apodemus speciosus collected in twenty areas
of Japan had much polymorphisms of mtDNA and
those with sequence divergence (P0) were 0.12.

Apodemus argenteus collected from eleven
parts of Japan had a few polymorphisms and
those with sequence divergence were 0.09 . The
different values in variations of two species
mtDNA suggested that A. speciosus had arrived
in Japan before A. argenteus.

1224

Genetics

GE 9

GEOGRAPHIC DISTRIBUTION OF RIBOSOMAL DNA
VARIANTS IN APODEMUS SPEC I OSUS
H. Suzuki ^ , K.Tsuchiya"! M. Sakai zumi 3 ,
S.Wakana4, S.Sakurai, K.Moriwaki5.
1Inst. Med. Sci.. The Jikei Univ. School
of Med., Tokyo, 2Miyazaki Med. College,
'"The Tokyo Metropol. Inst. Med. Sci.,
4Central Inst. Exp. Animals, 5Natl.
Inst. Genet., Mishima.

Southern blot analysis for ribosomal
DNA ( rDNA) of Japanese field mice
( Apodemus speci osus ) using thirteen
restriction enzymes revealed that the
main construction of the spacer regions
near the 18S and 28S rRNA genes was
similar , including those from Hok’kaido
and nine isolated island populations,
Tokara, Tanegashima, Yaku, Tsushima,
Dogo (Oki), Dozen (Oki), Sado, Miyake,
and Izu-Ohshima.

The polymorphic EcoRI and Kpnl sites,
however, were unique for each isolated
island population. Concerning these site
variations, distinct patterns were
obtained between the two Oki populations
and between Shikine and Izu-Ohshima in
spite of their close geographic
proxi mi ty .

Geographically restricted minor faint
bands were occasionally observed in
addition to major bands. In some cases
appearance was geographically
restricted, suggesting a limited gene
flow in the wild population of A .
speci osus .

GE 10

GENETIC DIFFERENTIATION OF MEDAKA
( ORYZIAS LATIPES) AND HISTORY OF THE
WATER SYSTEM IN THE SETO INLAND SEA AREA.

1 2

M. Sakaizumi , A. Kikuta and K.
Tsuchiya . The Tokyo Metropol. Inst.
Med. Sci., Tokyo, ^Okayama Univ. Med.
School, Okayama and '’Miyazaki Med.
College, Miyazaki.

We have already reported that the wild
populations of medaka around the Seto
Inland Sea are similar but distinct from
other populations separated by mountains.
In addition, a genetic difference has
been found between populations of the
eastern and the western parts of this
area, judging from allelic distribution
of the Ldh-A locus. It is supposed that
in the last glacial period two major
water systems existed in the eastern and
western regions, as deduced from a survey
of the sea bottom, leading us to
investigate the correlation between
allelic distribution and geological
evidence .

We investigated the frequency and
distribution of the Seto Inland Sea area-
specific alleles, Acp— and Ldh-A— , in 78
locations. The detailed distribution of
these alleles suggested a close
connection between regional genetic
differentiation observed at present and
historical changes of the water system in
this area.

GE 11

MEIOTIC ANALYSIS OF CHROMOSOMAL
POLYMORPHISM IN ORYZIAS MINUTILLUS FROM
THAILAND.

S. Tsuyuki1 and H. Uwa2. ^ept. of Biol.
Kanazawa Univ., Kanazawa and 2Dept. of
Biol., Shinshu Univ., Matsumoto.

In Oryzias minutillus, the chromosomal
polymorphism has been detected among
specimens from three localities in
Thailand: Bangkok, 4M*+1SM+12A; Chiang
Mai, 6M* + 1 SM+8A; Phuket, 21 A (M*: "large"
metacentrics ) . Meiosis was examined in
parental fish and their FI hybrids.
Chromosomes in first meiotic metaphase
were divided into two categories; vivalent
(V) and "large" vivalent (V*) chromosomes.
Diploid number and chromosome constitution
in meosis were summarized as follows. 1)
Parents: Bangkok, 2n=34, [4V*+13V], Chiang
Mai, 2n=30, [6V*+9V], and Phuket, 2n=42,
[21V]. 2) FI hybrids: Bangkok x Phuket,
2n=38, [4V*+13V], and Bangkok x Chiang
Mai, 2n=32, [6V*+9V].

Number and constitution of meiotic
chromosomes of FI hybrids were well
consistent with each other if we suppose
that 8 and 4 chromosomes of acrocentric
type in phuket and Bangkok paired with 4
and 2 "large" metacentrics from Bangkok
and Chiang Mai, respectively.

Therefore, we conclude that these
chromosomes are homologous in each other
and the chromosomal polymorphism among
three localities is associated with
Robertsonian fusion.

GE 12

KARYOTYPE AND PHYLOGENY OF HAINAN MEDAKA,
ORYZIAS CURVINOTUS, FROM HONG KONG.

H^ Uwa 1 and D. Dudgeon2. 1Dept. of Biol.,
Shinshu Univ., Matsumoto and 2Dept. of
Zool., Univ. of Hong Kong, Hong Kong.

Oryzias curvinotus , was nominated by
Nichols and Pope- (1 927) from Hainan. This
species is distinguished from 0^_ lat ipes
by the posteriorly situated dorsal-fin,
though no difference is found in the
number of anal-fin rays between two
species (Uwa and Parenti, 1988). The
Hainan medaka is distributed in S China
including Guangdong, Hong Kong, and Hanoi.

The karyotype of CL_ curvinotus from
Hong Kong is 2n=48, NF=82, 4M+1 3SM+5ST+2A.
NORs are located on the short arms of a
submetacentric chromosome pair (NORs:SM).
These data indicate 0^_ curvinotus is a
member of the biarmed chromosome group of
Oryzias : O. mekongensis (2n=48, NF=58,

NORs : A ) , 0. sp. ( 2n = 48 , NF = 64, NORs:A), 0.
latipes ( 2n=46-48 , NF=68-70, NORs:SM), and
0. luzonensis (2n=48, NF=96, N0Rs:SM). In
this group, the arm number (NF>48)
indicates the degree of karyotype
evolution by pericentric inversion (Uwa et.
al. , 1 983; Uwa, 1 986).

According to this hypothesis, 0.
curvinotus belongs to a subgroup with
NORs -chromosomes of SM type, and seems to
be more related to 0^_ latipes than 0.
iU Y. ° II® IL s_L.§_ • This conjecture seems to
coincident with results of interspecific
hybridizations among these three species.

Genetics, Biochemistry

1225

GE 13

ISOLATION AND CHARACTERIZATION OF A
REPETITIVE SEQUENCE (OLR1) IN THE JAPANESE
MEDAKA, ORYZIAS LATIPES .

K.Naruse, H.Mitani and A.Shima. Zool.Inst.,
Fac. of Sci.. Univ. of Tokyo, Tokyo. _

In order to characterize Medaka genome at
DNA level, we isolated and cloned 1 k base-
pairs DNA fragment containing repetitive
sequence (0LR1) from our genomic library of
the Medaka, Oryzias latipes . 536 base-pairs
of 0LR1 DNA were sequenced. 0LR1 was very AT
rich. GC content of 186 base-pairs of 5' end
was 26.8 % and that of 350 base-pairs of 3’
end was 30.5 %. About 6000 copies of 0LR1-
related sequences were found in the Medaka
genome. OLRl-related sequences were trans¬
criptionally active in the cultured Medaka
cells ( 0L32 ) .

We examined the distribution of OLRl-
related sequences among 9 fish species,
Oryziatidae (7 species) and Poeciliidae (2
species). Genomic DNA gel blot and slot blot
analyses showed that OLRl-related sequences
were conserved in the following 3 species;

O . curvinotus , 0 . luzonensis , and 0 .
mekongnensis . These 3 species are classified
by Uwa (1985) into the biarmed type in
karyotype analysis. In the following 6
species, the sequences could not be detected
; 0^ .javanicus , 0 . minutillus , 0 .
cerebensis , 0 . melastigma , P . sphenopus and

P . formosa.

We believe that 0LR1 is a useful DNA
marker to study the phylogenic relationship
among genus Oryzias .

GE 14

USEFULNESS OF SPECIFIC-LOCUS MUTATION ASSAY
SYSTEM ESTABLISHED IN THE MEDAKA ORYZIAS
LATIPES FOR STUDYING REGULATION OF GENE
EXPRESSION DURING EMBRYONIC DEVELOPMENT AND
SPERMATOGENIC MATURATION.

A.Shima and A.Shimada. Zool.Inst., Fac. of
Sci., Univ. of Tokyo, Tokyo.

We have recently established the tester
Medaka Oryzias latipes which are multiple
recessive homozygous for the 3 loci. By the
use of this tester, we are examining the
relationship between gamma-ray doses and
frequencies of mutations induced in male
germ cells of various spermatogenic stages.
Because we chose such marker genes whose
mutant phenotypes can be recognized early
during development, and also because we are
now able to culture somatic cells from indi¬
vidual mutant embryos almost all of which
would eventually die before hatching, we can
save, as cultured cells, the mutants which
would be concealed due to dominant lethals.
These cultured cells would be useful for
somatic cell genetic analysis of the marker
as well as lethal genes of the Medaka. On
the other hand, the dose-response analysis
revealed the spermatogenic stage-specific
responses with regard to induced mutageni¬
city. The specific-locus mutation rate
(per locus) was highest in sperm, middle in
spermatids and extremely low in spermato¬
gonia. These results indicate the increase
with differentiation of the mutagenicity of
male gametes, affording interesting subject
to study differentiation and DNA repair.

GE 15

THE NEW BODY COLOg MUTANT IN THE MEDAKA.
H.Tomita. Lab. of Freshwater Fish Stocks,
Fac. of Sci., Nagoya Univ., Nagoya. _

The mutant(mo) in the medaka (Oryzias
latipes ) was found in the offsprings of
brown type in orange-red stocks in the
market. The BmoR fish is light brown
in a body color. It has few visible
leucophores, light colored melanophores ,
and ordinary iridocytes.

The eyeballs of this mutant larvae
are black color because of small quantity
of guanine in iridocytes, while normal
ones show metallic reflection. These
black eyeballs gradually change to normal
ones following to growth. The other
iridocytes in skin have also small quanti¬
ty of guanine in this mutant young, and
quantity of guanine increases to normal
at the adult stages.

The larvae of this mutant (BmoR)
has light colored melanophores and dilute
leucophores. In this mutant, most leuco¬
phores disappear till 7 mm body length.

The mo Ulotant is easily detected in
black eyeballs at the larval stage and
in few visivle leucophores and light
colored melanophores at the adult stage.

The mo mutant is recessive and autosomal.
The mo alleles may be not linked to the
b, the ci , the gu, the _i, and the ml-3
alleles .

BI 1

THE cAMP-DEPENDENT PROTEIN KINASE IN THE
SEA URCHIN SPERMATOZOA

E. YOKOTA1’2 I. M ABUCHI 1 , A, KOBAYASHI3
and H. SATO2.

Dept. Biol., Coll. Arts and Sci., Univ.
Tokyo, Tokyo, ^Sugashima MBL, Nagoya Univ.,
Toba, JDept. Agric. Chem. , Okayama Univ.,
Okayama . _

It is considered that c AM P-d e p e nd e n t
protein kinase (cA kinase) plays an
important role in the flagellar movement of
spermatozoa. We investigated the
extr actabi 1 ity and properties of the cA
kinase of sea urchin spermatozoa. The cA
kinase activity was detected in a Triton X-
100-soluble fraction of whole sperm.
Further extration of the isolated axonemes
with Triton X-100 again released a
considerable amount of the cA kinase
activity. When the axonemes extracted three
times with Triton X-100 were treated with
0.6 M KC1, only a small amount of the cA
kinase activity was released. However, when
these high s a 1 1 -e x t r a c t ed axonemes were
further extracted with a low salt solution,
the cA kinase activity was again released.
The flagellar movement of Triton model
freed from Triton-soluble materials by
sucrose density gradient centrifugation was
induced by the addition of ATP and
inhibited by K-26, an inhibitor of cA
kinase. From these results, cA kinases
remained in the axonemes after the first
extraction with Triton X-100 may be
involved in the regulation of flagellar
movement in the Triton model.

1226

Biochemistry

BI 2

TROPOMYOSIN IN THE SEA URCHIN EGG CORTEX.
S.Maekawa, M.Toriyama and H. Sakai. Dept, of
Biophys. and Biochem. , Fac. of Sci., Univ.
of Tokyo, Tokyo. _

A heat stable actin binding protein was
purified from the Triton-treated cortex
fraction of fertilized sea urchin eggs and
identified to be tropomyosin(TM) . Egg TM
showed characteristics typical of nonmuscle
TMs such as low molecular weight, low Lys/
Arg ratio, high Mg2+ requirement in binding
to F-actin, short periodicity of Mg^+para-
crystal, in addition to the common propert¬
ies of all TMs, namely, stability to high
temperature, anomalous migration on SDS urea
gel, dissociation from F-actin under high
salt condition and very acidic isoelectric
point. Co-sedimentation assay indicated
that the binding of egg TM to F-actin was
non-competitive with 260K act in-binding pro¬
tein previously purified. It was suggested
that both the proteins play a cooperative
role in the formation of actin filament-
based cytoskeletal structure in the cortex.

BI 3

AN N-TERMINAL SEGMENT OF DEPACTIN
PARTICIPATES IN INTERACTION WITH ACTIN.

K. Sutoh and I. Mabuchi2. ^Dept. of
iochem. and Biophys., Fac. of Sci., and
Dept, of Biol., Coll, of Arts and Sci.,
University of Tokyo, Tokyo.

One-to-one complex of actin and
depactin, an ac t in-depo 1 y mer i zing protein
isolated from starfish oocytes, was cross-
linked with 1-e t h y 1 -3- [ 3-( d ime t h y 1 amin o )
propyl ]carbodiimide to introduce covalent
bonds at their contact site. Locations of
cross-linking sites were identified along
the depactin sequence by the end-label
fingerprinting, which employed site-
directed antibodies against the N- and C—
termini of depactin as end-labels.
Mapping with these end-labels have
revealed that the N-terminal segment of
depactin (residues 1-20) contains sites in
contact with the N-and C-terminal segments
of actin, both of which participate in
interaction with depactin [Sutoh, K. &
Mabuchi, I. (1986) Biochemistry 25: 6186-
6192] .

1P* ' ,8DENVKEEIRAFK19

BI 4

AMINO ACID SEQUENCE OF PROFILINS OF SAND
DOLLAR ( Clypeaster j aponicus ) AND SEA
URCHIN ( Anthocidaris crassispina ) EGGS.

T. Takagi1 , 12 Mabuchi2 and H. Hosoya .
^Biol. Inst., Fac. of Sci., Tohoku Univ.,
Sendai, 2Dept. of Biol., Coll. Arts and
Sci., Univ. of Tokyo, Tokyo, 2Dept. of
Ultrastruct. Res., The Tokyo Metropolitan
Inst., of Med. Sci., Tokyo _

The amino acid sequences of profilins
obtained from eggs of sand dollar ( C.
j aponicus ) and sea urchin (A^_ crassispina )
have been determined. Both were composed
of 139 amino acid residues and the N-
termini were blocked by acetyl group. Both
sequences are very homologous ( 1 4% diffe¬
rence ) . When the amino acid sequences of
profilins so far sequenced are compared
(Human platelet. Calf spleen, Acanthamoeba
Ia/b and II, Sand dollar, Sea urchin),
homologous sequences were found in about
30 residues of both N- and C-terminal
parts. On the other hand, frequent amino
acid replacements were observed in the
middle region. These results suppose to
indicate that both N-and C-terminal parts
of profilins participate in actin binding.

The amino
following.

acid sequence

of

: CL

1 aponicus is

S W

D

S

Y

I

D

N

L

V

A

Q

T

X

D

A

S

G

T

A

H S

D

R

A

C

I

I

G

L

D

G

G

A

P

W

T

T

A

G

H A

N

A

L

X

L

Q

G.

T

E

G

A

N

I

A

X

C

F

X

S K

D

F

S

A

F

M

A

G

G

V

H

A

E

G

L

X

Y

Q

F L

R

E

E

D

A

X

L

V

L

A

X

X

X

G

E

G

A

I

T L

Q

A

S

X

T

A

I

V

I

A

H

C

P

E

G

G

Q

Q

G N

T

N

K

G

V

S

V

I

A

E

Y

L

E

S

L

G

M

BI 5

THREE ISOFORMS OF 45K PROTEIN FROM SEA
URCHIN EGGS.

M. Ohnuma and I. Mabuchi. Dept, of Biol.,
Coll, of Arts and Sci., Univ. of Tokyo.

45K protein purified from sea urchin egg
severs actin filament and forms a 1 to_ 1
complex with actin in the presence of Ca^ .
This complex caps the barbed end of the
actin filament in a Ca^+- independent manner
and thereby causes depolymerization of
actin .

By means of isoelectric focusing in the
presence of 8M urea, it was found that the
45K protein consisted of three components
which migrated very closely with each
other. These three components were digested
with trypsin, <x -chymotryps in or
Staphylococcus aureus V8 protease and
digested peptides were analysed by SDS-
PAGE. The digestion patterns of the three
components were similar to each other.
Therefore, these three components were
likely to be isoforms of the 45K protein.
When the 45K protein fraction was loaded
onto a hydroxyapatite column and eluted by
0-0.3 M phosphate gradient, the most basic
component was eluted at a little higher
phosphate concentration than other two
components. Actin binding activity of the
three components was examined. A
nitrocellulose membrane onto which the 45K
protein had been transferred was incubated
with rabbit skeletal muscle actin either in
the presence or absence of Ca^ and the
bound actin was detected by immunobl^tting
using antibodies against actin. All the
three components showed the actin binding
activity. The activity of the most basic
component seemed to be weaker than those of
the other component.

Biochemistry

1227

BI 6

BI 8

THE STRUCTURE OF MAGNESIUM-PARACRYSTAL OF
F-ACTIN FORMED IN THE PRESENCE OF MUSCLE
20K PROTEIN.

H. Abe*, T. Obinata*, and I. Mabuchi**.
*Dept. of Biol., Fac. of Sci., Chiba Univ. ,
Chiba and **Dept. of Biol., Coll, of Arts
and Sci., Univ. of Tokyo, Tokyo.

We have already isolated and charac¬
terized an actin regulatory protein, with
apparent mol wt of 20 kDa from embryonic
chicken skeletal muscle. The 20K protein
posesses properties similar to cofilin
isolated from porcine brain; it binds to
both G- and F-actins in a molar ratio of
1:1 and regulates actin assembly in vitro
in a pH-sensitive manner. In this study,
we induced actin paracrystals by addition
of 50 mM MgC^ in the presence of 20K
protein, and examined them by electron
microscopy. Several paracrystalline
patterns were observed in a single bundle
in the presence of 20K protein at slightly
acidic pH, but no bundle was formed at weak
alkaline pH. Major patterns were 1 ) an
array of parallel filaments with a spacing
wider than that of pure actin paracrystal
(type A), 2) network-like pattern (type B),
and 3) a fine mosaic-like pattern (type C).
By comparing the optical diffraction
pattern from the image of type A with that
of Mg-paracrystal of F-actin alone, it was
demonstrated that periodicities of genetic
helices of the actin filaments in this
bundle is significantly larger than that in
the pure actin paracrystal'.

DUAL REGULATION OF ACTIN ASSEMBLY BY ACTIN
INHIBITORY PROTEINS AND MYOSIN AT THE
EARLY PHASE OF MYOFIBRILLOGENESIS .

T. Obinata, S. Ohshima and H. Abe. Dept of
Biol., Fac. of Sci., Chiba Univ., Chiba.

In embryonic chicken skeletal muscle, a
large amount of actin exists as monomer
(G-actin) without being polymerized. Two
proteins ( 1 9K and 20K proteins) which
inhibits polymerization of actin in the
embryonic muscle tissue were isolated and
characterized (Abe & Obinata, 1988). In
this study, we examined how these actin
binding proteins are involved in the
regulation at the early phase of myofibril
formation. It was observed that the
amount of 1 9K and 20K proteins in muscle
decreased during development. The inhibi¬
tory action of the two proteins on actin
was removed by myosin or heavy meromyosin.
Actin filaments were formed on the surface
of myosin filaments when myosin filaments
were added to the mixture of actin and 1 9K
protein in the presence of 0.1 M KC1 and 1
mM MgCl_ . SDS-PAGE analyses of whole
muscle lysates have demonstrted that
myosin-actin ratio is very low at young
embryonic stages but increases markedly as
muscle develops. Based on these obser¬
vations, we propose that actin assembly is
dually controlled in the developing muscle
by the inhibitors and the accelerator
(myosin); this mechanism may enable the
ordered assembly of actin and myosin in
the early phase of myof ibrillogenesis .

BI 7

PURIFICATION OF A PROF ILIN -LIKE PROTEIN
FROM EMBRYONIC CHICKEN SKELETAL MUSCLE.

S. Ohshima, H. Abe and T. Obinata. Dept, of
Biol., Fac. of Sci., Chiba Univ., Chiba.

We have been interested in the regula¬
tion of actin assembly during myofibrillo-
genesis, and several actin-binding proteins
have already been detected in developing
skeletal muscle. In this study, a 16 kDa-
mol-wt protein that inhibits the polymeri¬
zation of actin was purified from embryonic
chicken skeletal muscle by a combination of
poly-L-proline affinity, ion exchange
( DE-52 and SP-Sephadex ) , and gel filtration
column chromatography. This protein was
present in a monomeric form in a physi¬
ological salt solution, when examined by
gel filtration with HPLC. The mol wt
estimated by SDS-PAGE was about 16 kDa. It
was observed that this protein binds to
actin at a 1:1 molar ratio by cross-linking
analyses with EDC . Effect of 1 6K protein
on actin polymerization was examined by
viscometry. The rate and extent of actin
polymerization induced by 50 mM KC1 was
remarkably inhibited by this protein, while
in the presence of micromolar level of
MgC^ in addition to KC1 , decrease in the
rate and extension of lag-phase of actin
polymerization were observed only slightly,
but the steady state level of viscosity was
scarcely affected. All these properties of
1 6K protein is very similar to those of
prof ilin.

BI 9

STUDIES ON C-PROTEIN ISOFORMS OF CHICKEN
AND MAMMALIAN STRIATED MUSCLES BY TWO
DIMENSIONAL ELECTROPHORESIS AND PEPTIDE
PAPPING.

T. Kojima, K. Sano, K. Shinbo and T.
Obinata. Dept of Biol., Chiba Univ., Chiba.

We have established a method to charac¬
terize C-protein variants by a combination
of 2D-electrophoresis and immunoblotting .

As previously reported, six C-protein
variants have been distinguished in
developing chicken striated muscles; four
of them are of slow muscle-type and the
others are of cardiac- and fast muscle-
types. We examined C-protein of rat and
rabbit by similar procedures. Since one of
the monoclonal antibodies (FC-18) to
chicken C-protein exhibited cross-reaction
with the mammalian protein, fast skeletal-
and cardiac-type C-proteins of the mammals
were identified on 2D-gel with this
antibody. The size and pi of mammalian
C-proteins were approximately the same as
the counterparts of chicken. The structure
of the C-protein isoforms in chicken and
rabbit was compared by peptide mapping
according to Cleaveland (1977). The four
slow-type variants, two embryonic- and two
adult-forms, of chicken exhibited high
homology although specific fragments were
detected in each isoform, while cardiac-
and fast-type isoforms differed remarkably
from the slow-type variants. Chicken and
mammalian C-proteins differed significantly
even in the same type of muscle.

1228

Biochemistry

BI 10

CHARACTERIZATION OF cDNAs OF a-ACTININ
EXPRESSED IN EMBRYONIC CHICKEN SKELETAL
MUSCLE

T. Endo and T. Obinata. Dept, of Biol.,
Fac. of Sci., Chiba Univ., Chiba. _

Peptide mapping by limited proteolysis
revealed that during development of chicken
fast skeletal muscle ( pectoralis ) ,
molecular species of a-actinin switched at
least twice, at embryonic stage and after
hatching. To demonstrate the presence of
these embryonic and perihatching forms of
a-actinin in addition to the adult form,
cDNA library in Agtl 1 was constructed from
mRNA of 1 2 d old embryo skeletal muscle.
Using an antibody to adult pectoralis a-
actinin, we screened this library and
obtained 13 positive plaques. Four of
these clones were further characterized.
Restriction maps of these clones were very
similar to that of adult pectoralis a-
actinin cDNA . Partial nucleotide sequences
of the clones 1 , 2, and 3 were identical
with that of the adult a-actinin cDNA.
However, we cannot rule out the possibility
that other regions are distinct between
these clones and the adult cDNA, as
represented by the fact that smooth muscle
and nonmuscle a-actinin cDNAs differ^only
in the sequences encoding EF-hand Ca^+-
binding site. Furthermore, putative 5'-
untranslated region of the clone 4 had no
homology to the corresponding region of the
adult cDNA. Thus, at least this clone is
likely to be specific for the embryonic
stage.

BI 11

IN VITRO BINDING OF CHICKEN MYOFIBRIL
I -PROTEIN TO MYOSIN.

K. Ohashi and M. Yodono. Dept, of Biol.,
Fac. of Sci., Chiba Univ., Chiba _

I-protein, which is localized at the

A- I junctions of myofibrils, can bind to
myosin filaments and has inhibitory action
on the actomyosin ATPase activity in low
ionic strength solutions. In the presence
of I-protein, reconstituted myosin filaments
assembles into huge bundles.

SDS polyacrylamide gel electrophoresis
and Sephacryl S-400 gel filtration showed
that I-protein molecules were soluble in
PBS as homodimers of 50 kDa subunit. Thus,
each I-protein molecule is expected to have
two binding sites to myosin.

From the trypsinized fragments of I-
protein, 30 kDa peptide was purified using
DEAE cellulose column chromatography. This
peptide was shown to be present as a mono¬
meric form in PBS, using Sephacryl S-300
gel filtration. This peptide also bound to
reconstituted myosin filaments in a solution
of 20 mM KC1 and 10 mM potassium phosphate
buffer, pH 6.5. In this case, no large
bundle of myosin filaments was observed.

The amount of I-protein which bound to
LMM paracrystals in a solution of 50 mM KCl
and 10 mM potassium phosphate buffer, pH
6.5 is much less than that of I-protein
which bound to myosin filaments in a same
solution .

BI 12

BINDING OF NATIVE CONNECTIN TO MYOSIN
FILAMENTS

T. Murayama, Y. Nakauchi , S. Kimura and K.
Maruyama. Dept, of Biol., Fac. of Sci.,
Chiba Univ. , Chiba. _ _ _

Connectin is a long elastic filamentous
protein responsible for positioning myosin
filament at the center of a sarcomere and
for passive tention generation of stretch¬
ed muscle fibers. Thus connectin links a
myosin filament to adjacent Z lines.

We have previously observed that native
connectin forms an aggregate of myosin
filaments at low ionic strengths. In the
present study, the binding of connectin to
myosin was quantitatively investigated
using a sedimentation technique and densi-
tometric estimations of proteins. In the
presence of 60 mM KCl and 5 mM phosphate
buffer, pH 7.0, myosin was not sedimented
by a centif ugation for 10 min at 2000 g,
whereas myosin-connectin complex was easi¬
ly precipitated. As much as 0.56 mg of
connectin was bound to 0.1 mg of myosin.
Thus 1.5 moles of connectin bound to one
mole of myosin suggesting that some 150
connectin filaments bound to a single myo¬
sin filament of 0.5 pm in length. This
value was much more than the' ratio found
in muscle (12:1). C protein did not affect
the binding of connectin to myosin filam¬
ents .

BI 13

ANTI -CONNECT IN MONOCLONAL ANTIBODIES BIND
DENSE BODIES OF THE BODYWALL MUSCLE CELLS
OF CAENORHABDITIS ELEGANS .

K . Maruyama-*- , T . Matsuura^ , A.Matsuno^, H.
Ohmuro2 and M.Shibata^. 1-Dept. of Biol.,
Fac. of Sci., Chiba Univ., Chiba, 2Dept. of
Biol., Fac. of Sci., Shimane Univ., Matsue,
3Div. of Bioin'f-ajrm. Eng., Tokyo Metropol.
Inst, for Med. Sci., Tokyo and 4Med. and
Biol. Labs, Co. Ltd, Ina.

Several kinds of monoclonal antibodies
specific to vertebrate skeletal muscle
connectin ( 3000 kDa elastic protein) were
observed to cross-react with 400 kDa prote¬
in of the bodywall muscle of the nematode,
Caenorhabditis elegans. Immunofluorescence
microscopy revealed that the 400 kDa prote¬
in was localized in the bodywall muscle
cells. Immunoelectron microscopy further
showed that the antibodies specifically
bound to the dense bodies of the bodywall
muscle cells.

The 400 kDa protein in question was not
present in the bodywall muscle cells of the
unc-22 deficient mutants of C. elegans.
Therefore, it is suggested that the product
of the unc 22 gene (400 kDa protein) is
probably a connectin-like elastic filament
that links a myosin filament to a dense
body, since D. G. Moerman et al. (Genes &
Development, 2, 93-105, 1988) reported that
the unc 22 gene product is present on the
myosin filaments of the C. elegans bodywall
muscle cells.

Biochemistry

1229

BI 19

LOCALIZATION OF (3-ACTININ IN CHICKEN BREAST
MUSCLE MYOFIBRILS AS REVEALED BY IMMUNO¬
FLUORESCENCE MICROSCOPY.

S. Shimaoka, H. Yamamoto and K. Maruyama.
Dept, of Biol., Fac. of Sci., Chiba Univ. ,
Chiba . _

0-Actinin is an actin filament capping
protein in rabbit and chicken skeletal
muscles. In 1977, we reported that it was
localized to the M line region of chicken
breast muscle myofibrils (K. Maruyama et al
J. Biochem. , 81, 215-232, 1977). However,
two recent reports claimed that it is at
the Z lines (J. F. Casella et al., J. Cell
Biol., 105, 371-379, 1987) and at the free
ends of the actin filaments (T. Funatsu et
al., J. Biochem., 103, 61-71, 1988).

In view of these discrepancies, we have
reinvestigated the localization of (J-act-
inin. Affinity purified polyclonal anti¬
bodies against 0-actinin, <31(35 kDa) sub¬
unit and 011(32 kDa) subunit were used for
immunofluorescence microscopy. Anti-0 -
actinin and anti- JSI subunit antibodies
stained both M line region and Z line. On
the other hand, anti- 01 I subunit antibodies
reacted with Z lines. Therefore, it is at
least true that 0-actinin is localized to
Z lines. Further work is in progress for
elucidating the localization at the other
ends of actin filaments in a sarcomere of
chicken breast muscle myofibrils.

BI 15

ANALYSIS OF PHO S PHO RYL ATED NONMUSCLE
MYOSIN WITH NATIVE GEL ELECTROPHORESIS.
H.Takano-Ohmuro. K.Kohama* and T. KAMINUMA
Tokyo Met. Inst, of Med. Sci., Tokyo, and
*Dept. of Pharmacol., Fac. of Med., Univ.
of Tokyo, Tokyo _

It has been reported that functions of
smooth muscle myosin and nonmuscle myosin
are modulated by their regulatory myosin
light chain (RLC) phosphorylation ijn vitro.

We have reported that 1) smooth muscle
myosin phosphorylated at its RLC with
myosin light chain kinase (MLCK) increases
its mobility according to its phosphoryla¬
tion degree on native pyrophosphate gel
electrophoresis (PPi PAGE), while smooth
muscle myosin phosphorylated with protein
kinase C does not increase its mobility,

2) in the case of striated muscle myosin,
l) is not observed. and 3) analysis of
myosin phosphorylation state in contract¬
ing smooth muscle can be carried out by
PPi PAGE without myosin purification.
These results confirmed that PPi PAGE is
an excellent tool to analyze myosin
phosphorylation state of smooth muscle
both i_n vitro and in vivo.

We analyzed phosphorylated nonmuscle
myosins from intestinal brush border,
platelet, and brain with PPi PAGE. We
found that these myosins phosphorylated
with MLCK jji v i t r o also showed mobility
increase on P P i P A G E . We are further
planning to apply this method to study
myosin phosphorylation functioning in
nonmuscle cells in vivo.

BI 16

PHOSPHORYLATION OF DROSOPHILA MYOSIN LIGHT
CHAINS CAUSES INCREASE IN MYOSIN ADENOSINE
TRIPHOSPHATASE ACTIVITY

S. Takahashi-*- , H. Ohmuro2 and K. Maruyama 1
I-Dept. of Biol., Fac. of Sci., Chiba Univ.,
Chiba and 2Div. Bioinfom. Eng., Tokyo
Metropol, Inst, for Med. Sci., Tokyo. _

Drosophila flies cannot fly for several
hours after emergence. Meanwhile, the
myosin ATPase activities markedly increase.
Previously, we suggested that this ATPase
activation was due to phosphorylation of
myosin light chains, as known in vertebrate
smooth muscle myosin.

In the present study, we have prepared
a crude myosin light chain kinase (MLCK)
sample from Drosophila flies. Myosin from
just emerged flies were incubated with this
MLCK in the presence of ATP (32p labeled) .
Phosphorylations of myosin LC1 and LC2 were
detected by autoradiography of the two-
dimensional electrophoresis patterns. The
Ca-ATPase activity increased from 1.2 to
1.7 pnoles/mg/min and the Mg-ATPase action
was elevated from 0.06 to 0.13 pmoles/mg/
min .

From these in vitro studies, it is
suggested that Drosophila myosin light
chains are phosphorylated by MLCK during
several hours after emergence and as a
result ATPase activity is activated to
facilitate active flight.

BI 17

THE 1200 KDA FILAMENTOUS PROTEIN OF
CRAYFISH STRIATED MUSCLE.

D. H. Hu and K. Maruyama. Dept, of Biol.,
Fac. of Sci., Chiba Univ., Chiba.

The 1200 kDa protein was isolated and
purified to homogeneity from crayfish claw
muscle by a method to purify native connec-
tin from chicken breast muscle. This prot¬
ein was a long filament at physiological
icpnic strengths under electron microscope.
The amino acid composition was very similar
to those of chicken breast muscle connectin
(3000 kDa) and of honeybee thoracic muscle
prejectin (360 kDa) . The 1200 kDa protein
cross-reacted with polyclonal as well as
monoclonal antibodies against vertebrate
muscle connectin. Immunofluorescence using
anti-1200 kDa protein serum showed that it
was localized at the A-I junction of cray¬
fish muscle sarcomeres.

The 1200 kDa protein was present not
only in crayfish claw and body muscles but
also in honeybee and beetle thoracic and
leg muscles. However, it was absent in
Pecten adductor muscles. Thus this protein
may be called arthropod connectin. It is
to be noted here that the protein is ident¬
ical with honeybee thoracic muscle preject¬
in: the molecular weight of prejectin was
estimated to be 1200 kDa, not 360 kDa, in
the present study.

1230

Biochemistry

BI 18

HEPARIN BINDING GROWTH FACTOER AND NEW MUSCLE
FIBER FORMATION IN WORK-INDUCED MUSCLE HYPER¬
TROPHY

S.Yamada’ and R.C.Strohman2 ’Dept, of Sports and
Scis, Univ. of Tokyo, 2Dept. of Zool. Univ. of
Califor. Berkeley _ _

A polyclonal antibody to bovine basic FGF
stains muscle cells both in vivo (rat) and in
vitro (chicken). Staining patterns in vivo
reveal that FGF is concentrated in the muscle
extracellular matrix, presumably bound to
heparin-rich components of the fiber basal
lamina. The intensity of staining increases
during tenotomy-induced hypertrophy and this is
especially marked in the fiber basal lamina area.
The role of increased levels of extracelluar
matrix FGF is not known but is most probably
related to satellite cell activation during
fiber growth and new fiber formation during
hypertrophy. Many mononucl eaeate cells appear
in the extracellular matrix space during hyper¬
trophy and some are identified as sattelite
cells based on staining with antibodies to
troponine T and myosin. Small new fibers are
also observed to arise in the muscle inter¬
stitial space. In vitro, chicken muscle cells
secrete an FGF-like activity which stimulates
satellite cell growth and which is inhibited by
heparin. Heparin-FGF interactions in regulating
new fiber formation during hypertrophy will
discussed.

BI 19

LACTATE DEHYDROGENASE ACTIVITY CHANGES
IN MOUSE BY VOLUNTARY WHEEL EXERCISE.

T. A. Nomaguchi and Y. Sakurai.

Dept, of Biol., Tokyo Metropol. Inst, of
Gerontol., Tokyo.

LDH activity and the isozyme patterns in
various tissues of exercised young (3 mo.)
and old (24 mo.) mice were determined.

C57BL/6 male mice were exercised for one
week with a wheel (20 cm in diameter), and
the wheel activity in young and old mice
was 7197.4 ± 1004.7, 3430.8 ± 1084.6

counts a day, respectively. LDH activity
(U/mg protein) in homogenates of brain,
thymus, heart, lung, liver, spleen, kidney
and femoral muscle was assayed at pH 7.5
in a medium containing 50 mM tris-HCl, 1
mM KCN , 0.15 mM NADH and 0.7 mM pyruvate.

LDH activity did not change in all
tissues between young and old mice in
control. The activity in the exercised
mice, however, increased in brain and
muscle of the young and in liver of the
old, but decreased in kidney of the old,
respectively with significance.

Cummulative percentage of LDH isozymes
determined by densitometry changed in lung
(LDH 5), liverfLDH 1) and spleen(LDH1, 2,
4, 5) of the young, and in thymus(LDH2)

liver(LDH1, 2, 4) and spleen(LDH2, 5) of
the old with exercise. LDH isozyme ratios
(H/M ) also significantly increased in
liver of the young and in spleen of the
both animals, but decreased in brain,
liver and kidney of the old.

BI 20

SOME PROPERTIES OF THE PURIFIED LDH-A4 AND
-b4 ISOZYMES FROM THE MUSCLE OF ORYZIAS
LATIPES AND REASSOCIATION OF THE RESPECTIVE
SUBUNITS IN VITRO.

T. Sasaki, I. Iuchi, K. Yamagami and Y.
Hyodo-Taguchi* . Life Sci. Inst., Sophia
Univ., Tokyo and Div. Biol., Nat. Inst.
Radiol. Sci., Chiba*

The muscle LDH-A4 and -B4 isozymes were
purified from the' inbred (H04C) and the
outbred fish of Oryzias latipes , and some
of the enzymological as well as the protein
chemical characteristics of the isozymes
and the respective subunits were examined.

The B4 enzyme was markedly more heat-
stable and of higher affinity to both the
substrates, i.e., lactate and pyruvate,
than the A4 enzyme.

The molecular weights of the A and the B
subunits were re-examined and were es¬
timated at 36.0 kDa and 36.5 kDa , respec¬
tively, which were larger than the values
reported in our previous paper [Zool. Sci.,
4: 1024 (1987)]. Comparison of amino acid
compositions of the LDH subunits between
the medaka and the quinnat salmon,
Oncorhynchus t shawy t scha , revealed that
there was a great similarity between the
values for the corresponding subunits.

The results of the in vitro hybridiza¬
tion of the purified A and B subunits
strongly suggested that the middle two of
the six isozyme bands of the inbred (H04C)
fish were A2B2 conformers.

BI 21

PURIFICATION AND SOME PROPERTIES OF 1 4-NM
FILAMENT PROTEIN (49K PROTEIN) FROM
TETRAHYMENA THERMOPHILA
O. Numata , I. Takagi and Y. Watanabe.

Inst, of Biol. Sci., Univ. of Tsukuba,
Tsukuba .

Purified 4HK. protein from Tetrahymena
pyrif ormis (T. p. ) has been shown to re¬
semble intermediate filament proteins of
vertebrate origin in several respects. On
the other hand, 49K protein from T.
thermophi la (T. t. ) has not so far been
purified. Here, we attempted to isolate
the 49K protein from acetone powder of T.
t. by using FPLC, and succeeded in iso¬
lating 49K protein. From the comparison
between T. p. and T. t. 49K proteins,
there are strong resemblances as indicated
by the same assembly conditions, the same
filament structures and the indistinguish¬
able immunogenicities . In addition, the
N-terminal 31 amino acid residues of T.
t. 49K protein shows 93.5% sequence iden¬
tity with that of T. p. 49K protein.
Furthermore, the N-terminal amino acid
residues of Tetrahymena 49K proteins show
35.5% sequence identity with chicken
desmin residues 226-256 and hamster
vimentin residues 228-259 which correspond
to the junction region between coil lb Of -
helical domain and non-0r-helical spacer II
segments .

Biochemistry

1231

BI 22

UNIQUE PROPERTIES OF TETRAHYMENA ACTIN.

M. Hirono, R. Tanaka and Y. Watanabe, Inst,
of Biol. Sci., Univ. of Tsukuba, Tsukuba.

Actin from Tetrahymena pyrif ormis has been
purified by monitoring the presence of the
actin gene product with an antiserum
against an N-terminal peptide deduced from
the nucleotide sequence of the Tetrahymena
actin gene that we cloned previously. This
highly purified Tetrahymena actin shares
many essential properties with ubiquitous
actin, including iondependent polymer¬
ization to microfilaments, binding with
muscle HMM to form arrowheads, and the
activation of the Mg2+ATPase of muscle
myosin S-1 . On the other hand, some proper¬
ties of this purified Tetrahymena actin
clearly differ from those of muscle actin:
(_i) Tetrahymena actin has an 8-fold less
ability to activate the Mg2+ATPase of mus¬
cle myosin S-1 than muscle actin; ( ii )
Tetrahymena actin did not bind to phal-
loidin at all; ( iii ) Tetrahymena actin did
not inhibit DNase I activity at all; ( iv )
Tetrahymena actin did not bind to skeletal
muscle <X-actinin; (v) Tetrahymena actin did
not bind to smooth muscle tropomyosin at
all. In general, Tetrahymena actin whose
primary structure is diverged to a great
extent has very unique properties as com¬
pared with actins known so far. This
uniqueness may provide important clues for
elucidating problems concerning the
relationships between the structural and
functional domains in an actin molecule.

BI 23

Ca2+ /CALMODULIN-DEPENDENT PHOSPHORYLATION
OF CILIARY /3-TUBULIN IN TETRAHYMENA.

J. Hirano-Ohnishi and Y. Watanabe.
Institute of Biological Sciences,
University of Tsukuba, Tsukuba.

The ciliary axoneme is the minimal
structure which displays Ca2+-dependent
modulation of ciliary movement. We found
that ciliary /3-tubulin was exclusively
phosphorylated by an endogenous
Ca2+/calmodulin-dependent protein kinase(s)
in Tetrahymena axonemes, and virtually no
other substrate which was phosphorylated
Ca2+/calmodulin-dependently was detected.
The phosphorylation of /3-tubulin also
occurred in the outerdoublet microtubule
fraction, suggesting that the responsible
enzyme(s) was tightly associated with
outerdoublet microtubules. In experiments
with several types of permeabilized cilia
retaining ciliary motility, Ca2+-dependent
phosphorylation of /3-tubulin also occurred
exclusively. From these results, it is
inferable that the phosphorylation of
tubulin is involved in Ca2+-dependent
ciliary reversal.

BI 24

CLONING AND SEQUENCING OF THE cDNA FOR
TETRAHYMENA CALCIUM BINDING PROTEIN
(TCBP).

T.Takemasa1, K.Ohnishi2, T.Kobayashi3, T.Takagi3,
K.Konishi3, and Y. Watanabe1.

1 Inst. Biol. Sci. Univ. of Tsukuba, Ibaraki. 2 Natl. Inst, of
Health, Japan. 3 Biol. Inst., Fac. of Sci., Univ. of Tohoku,
Sendai. _

We succeeded in cloning and sequencing of
Tetrahymena cDNA containing the entire coding
region for a novel calcium binding protein (TCBP).
DNA and RNA blot analyses indicate that TCBP is
encoded by a single species gene from which is
derived a 0.9kb long transcript. The deduced TCBP is
composed of 218 amino acids and its molecular
weight is estimated as 24,702 dalton. The primary
protein sequence of TCBP and its predicted secondary
structure suggest that the protein is likely to contain
four homologous calcium binding domains that
conform to the helix-loop-helix (or EF hand) structure
found in various Ca-binding proteins, but one of
which is predicted not to actually bind calcium. Aside
from these regions, there is no overall structural
identity or apparent similarity with calbindin,
calmodulin, or troponin C. It is suggested that the
secondary structure of TCBP is significantly different
from the other proteins of this class, which bind four
Ca atoms.

BI 25

SUBSTRATE SPECIFICITY AND PHOSPHORYLATED
INTERMEDIATE OF Ca2+-ATPase OF
SARCOPLASMIC RETICULUM FROM SCALLOP
M. Abe , J. Nakamura , T. Watanabe2 and
K. Konishi' 'Biol. Inst. Sci., 2Dept. of
Biol., Tohoku Univ., Sendai.

We have studied some characteristics
of sarcoplasmic reticulum (SR) from
scallop striated adductor muscle.

Scallop SR hydrolyzed not only ATP but
also CTP, ITP and GTP. Ca ion was
transported by scallop SR coupling with
the hydrolysis of these nucleotides. Both
the velocity of the Ca ion transport and
the extra splitting of the NTP grade were
the following sequence: ATP > GTP > CTP
■"ITP like rabbit skeletal muscle SR.

1 00K protein of scallop SR was phospho¬
rylated by [ r - 2P]-ATP in the presence of
1 m M Ca2 and 5mM M g 2 + at 0°C. The
phosphorylated protein was acid stable and
sensitive to hydroxy lamine. These results
indicate that the phosphorylated protein
is the intermediate of Ca2+-ATPase of SR.

Although the amount of the
phosphorylated protein of scallop SR was
about a half of that of rabbit SR, the
ATP hydrolysis activity was about five
times higher than that of rabbit SR at
1 2°C. Ca2+ transport of scallop SR was
also higher than that of rabbit SR. These
results suggest that the rate of
dephosphorylation from the phosphorylated
intermediate is much higher than that of
rabbit SR at such a low temperature.

1232

Biochemistry

BI 26

AMINO ACID SEQUENCE OF SARCOPLASMIC Ca-
BINDING PROTEIN (SCP) FROM EARTHWORM,
EISENIA FOETIDA.

T.Furukohri , T. Suzuki', and K. Nakamura2.

' Dept . of Biol., Fac. of Sci., Kochi
Univ., 2Dept. of Life Sci., Hiroshima
Women's Univ., Hiroshima.

Amino acid sequence of sarcoplasmic Ca-
binding protein (SCP) from earthworm,
Eisenia f oetida , was determined. Earthworm
SCP consists of 175 amino acid residues,
and the N-terminus is acetylated alanine.
The molecular weight is calculated to be
19553. The established sequence showed
that earthworm SCP is composed of 3
calcium binding domains, domain I, domain
III and domain IV, same as polychaete
SCPs. The sequence of domain I, III and IV
of earthworm SCP shows a high degree of
homology to polychaete SCPs , but the
region of sequence number 35 to 92 is very
different to corresponding sequence region
of those SCPs. The complete amino acid
sequence of earthworm SCP is following:

Ac - AF VERKLKTY
ADRFVESEKL
VVSDGTKHKL
ALAGPLPLFF
QILGIDPKLA
VTAGTEFFTS

FKRIDFDKDG

DAARGADLKL

TEPVFVETVK

SAVDANADGK

AASFSAIDTN

EDEKSPSQHF

AITRNDFESM

VQVWEQYLKG

KQLHSAKLKE

IQTEEYAIFF

HDGDISLDEF

WGPLV.

BI 28

ON THE HETEROGENEITY AND TISSUE SPECIFICITY
OF SPINY LOBSTER MUSCLE TROPOMYOSIN.

M.Itoh, H.Koyama and T . Ishimoda-Takagi .
Dept, of Biol., Tokyo Gakugei Univ., Tokyo.

Heterogeneity and tissue specificity of
muscle tropomyosin (TM) of the spiny lobster,
Panulirus japonicus , were investigated. TM
preparation purified from abdominal muscles
of the spiny lobster contained three major
TM isoforms (designated as a, b and c compo¬
nents) and several minor TM components. To
investigate the detailed distribution of the
major TM isoforms in spiny lobster muscles,
TM was extracted from each of classified
muscle tissues and examined electrophoretic-
ally. Flexor-type muscles contained a large
amount of a component, a small amount of c
component and a very little amount of b com¬
ponent. Extensor-type muscles contained a
significant amount of a and c components and
a small amount of b component. Leg muscle
contained only b component, and this compo¬
nent was also involved exclusively in intes¬
tine. Cardiac muscle contained a single TM
isoform which was distinguishable from other
TM isoforms. Similar heterogeneity and tis¬
sue specificity of TM were also observed in
another species of the spiny lobster, P.
cygnus. In order to investigate the correla¬
tion of distribution of TM isoforms with
functional roles of muscles, the leg muscle
was divided into fast and slow muscles and
difference in the distribution of TM isoforms
was examined. However, no difference in the
distribution of TM isoforms was observed in
fast and slow muscles of the leg muscle.

BI 27

TISSUE-SPECIFICITY OF CRUSTACEAN TROPO¬
MYOSIN

J. Miyazaki and T. Hirabayashi. Inst, of
Biol. Sci., Univ. of Tsukuba, Tsukuba

Many tropomyosin ( TM ) isoforms, which
show tissue-specific distributions, have
been found in vertebrates. However, we
previously reported that TM isoforms of a
horsehoe crab showed no tissue-specificity.
To find either tissue-specific or non¬
specific distributions of TM isoforms, we
examined them in five species of crusta¬
ceans. We identified spots of TM isoforms
by preparation of TM from Procambarus
clarki , electrophoresis in the presence of
urea, and cross-reaction with an antiserum
against TM from P. clarki . We found 4 to 6
isoforms in muscle tissues and several
peptides in nonmuscle tissues which cross-
reacted with the antiserum. The latter
seems to represent nonmuscle TM isoforms.
The former TM isoforms showed tissue-
specific distributions. Among them there
was an isoform which was found in cardiac
muscle as a single major component and in
other tissue in an extremely reduced or
undetectable amount. Therefore, we can
expect some tissue-specific function in the
heart isoform, however we can not expect it
in other isoforms, because of their complex
distributions. We think that most TM
isoforms have the same function, but their
modes of expression are controlled by
tissue-specific mechanisms.

BI 29

ON THE HETEROGENEITY AND TISSUE SPECIFICITY
OF BIVALVE MUSCLE TROPOMYOSIN.

T. Ishimoda-Takagi , M.Ueda and T.Inoue.

Dept, of Biol., Tokyo Gakugei Univ., Tokyo.

Heterogeneity and tissue specificity of
tropomyosin (TM) in adductor and cardiac
muscles of twa_species of the bivalve,
Scapharca br ought onii and Atrina pectinida,
were investigated. Two major isoforms of TM
(refered to as a and B components) were re¬
cognized in both species. However, distribu¬
tion of the TM isoforms in adductor and car¬
diac muscles of these two species was dif¬
ferent from each other. In the translucent
portion of S. brought onii adductor muscle,
only B component was detected, whereas, a
large amount of a component and a small
amount of B component were detected in the
opaque portion of adductor muscle. Cardiac
muscle of S. brought onii contained B compo¬
nent as a major TM component and a component
as a minor one. In the translucent portion
of A. pectinida adductor muscle, only a com¬
ponent was detected, and a component was
also detected as a major TM component in the
opaque portion of adductor muscle and cardi¬
ac muscle. The opaque portion of adductor
muscle and cardiac muscle further contained
B component as a minor TM component. From
these results and our previous results ob¬
served in scallop and eulamellibranch mus¬
cles, we concluded that tissue-specific dis¬
tribution of TM isoforms in muscle tissues
were not necessarily observed in bivalve
muscles, although two major TM isoforms were
commonly involved in bivalve muscles.

Biochemistry

1233

BI 30

TYPING OF COLLAGENS FROM TESTS OF SEA
URCHIN.

K. Shimizu and K. Yoshizato. Develop. Biol.
Lab., Dept, of Biol., Fac. of Sci., Tokyo
Metropolitan Univ. , Tokyo. _ _

We have already reported about
biochemical and immunological characteriza¬
tions of collagens from tests of sea
urchin, Asthenosoma i j imai (Zool. Sci.., 4,
1020). Types of sea urchin collagens (SUC)
and properties of their subunits, 1-chains,
were investigated in this report. SUC were
fractionated with selective salt
precipitation. SUC, same as mammal collagen
type I, were precipitated at 0.7 M NaCl in
HC1 , pH 3.0 and between 2.2 M and 3.0 M
NaCl in 0.05 M Tris HC1, pH 7.5. Four d-
chains with 143 k, 126 k, 116 k and 103 kDa
were observed in SDS-polyacrylamide gel
electrophoresis ( SDS-PAGE ) . Chains with 126
kDa and 103 kDa were strongly stained with
anti SUC sera in immunoblots. The 143 kDa
i-chains was a glycoprotein because it was
positive in SDS-PAGE with PAS stain. A
mixture of rf-chains of SUC was separated to
four fractions on CM-cellulose ion exchange
chromatography. These results suggested
that SUC had four «r-chains and that they
consists of two species of collagens. As
chains with 143 kDa and 126 kDa were
solubilized earlier than the others in the
pepsin extraction process, one of the two
collagens has molecules with 143 kDa and
126kDa as «f-chains; the other has 116 kDa
and 103 kDa molecules. One of the two
collagens might be a new species of type 1
collagen or a new type not reported
hitherto .

BI 31

PURIFICATION AND CHARACTERIZATION OF A
NUCLEAR PROTEIN KINASE Nil FROM SEA
URCHIN EMBRYOS.

Y.Masuyama, l.Yasumasu.

Dept, of Biol., Sch . of Educ . , Waseda
Univ . , Tokyo . _

Polyamine- stimulated protein kinase was
estimated in isolated chromat in from sea
urchin embryos and the activity was changed
during early development. The enzyme
phosphorylated endogenous and exogenous
non-histone protein (acidic protein) but
not histone. The protein kinase extracted
from isolated nuclei of sea urchin embryos
with 0 . 4M NaCl, was purified by FPLC anion-
exchange, casein-phosvit in-Sepharose and
FPLC gel chromatographies. 0.4M NaCl
extract contained only one type of nuclear
protein kinase, Nil • NI activity was not
found in 0 . 4M NaCl extract. The molecular
weight of partial purified Nil enzyme were
about 100,000 as determined by gel
chromatography. The enzyme consists of
three subunits of molecular weights 42,000,
39,000 and 28,000 as determined by
SDS/polyacrylamide gel electrophoresis and
they were close to mammalian a , a ’ and
/? , respectively. Partial purified Nil
enzyme activity was stimulated by
polyamine. Heparin inhibited Nil activity
but its inhibition was not so strong as the
other reported N II or casein kinase II in
mammal, Drosophila or yeast.

BI 32

PURIFICATION AND PROPERTIES OF SEA URCHIN
EGG ARGININE KINASE.

Y.Yazawa, Dept, of Nutritional Physiol.,
Hokkaido Univ. of Education at Asahikawa,
Asahikawa , Hokkaido .

Arginine Kinase(AK) was isolated and
purified from sea urchin unfertilized
eggs with ammonium sulfate fractionation,
DEAE-Tovopearl column chromatography, gel
filtration, and hydroxylapatite column
chromatography. The M.W. of 43K was esti¬
mated from SDS-PAGE and 86k was estimated
from gel filtration under the physiologi¬
cal conditions. The pH-activity curve for
the forward reaction showed optimum at pH
6.0 and the specific activity was 2.57
piol Arg/mg.min. It was very low compared
with them of arginine kinases isolated
from nonvertebrate muscles. In the rever¬
se reaction, a plateau was shown between
pll8 . 0 and 9*6 and the specific activiti¬
es were 1.14 umol Pi/mg. min.

BI 33

ISOLATION AND PROPERTIES OF PALOLO-WORM
CREATINE KINASE

Y.Yazawa and Y. Yamamoto, Dept, of Nutri¬
tional Physiol., Hokkaido Univ. of Educ¬
ation at Asahikawa, Asahikawa, Hokkaido.

Creatine kinase (CK) was purified from
palolo— worm with ammonium sulfate fract¬
ionation, DEAE-Toyo pearl column chromat¬
ography, gel filtration and hydroxylapa-
tite column chromatography. The prepara¬
tion, CK, was homogeneous from SDS-PAGE.
The M.W. of palolo worm was estimated to
be 43K from SDS-PAGE and about 86k from
gel filtration under the physiological
conditions. This enzyme was composed of
378 amino acids and the star diagram
was very similar to that of rabbit ske¬
letal CK . Rabbit antiserum to palolo
worm CK was prepared. It cross-reacted
with chicken skeletal CK but sis not
cross-react with chicken brain type CK.
Enzymatic properties of palolo worm CK
were investigated and compared with th¬
ose of rabbit skeletal CK.

1234

Biochemistry

BI 34

PURIFICATION AND PROPERTIES OF CRAYFISH
ARGININE KINASE.

Y.Yazawa and Y.Ota, Dept, of Nutritional
Physiol. , Hokkaido Univ. of Education at
Asahikawa, Asahikawa, Hokkaido.

Arginine kinase was purified from tail
muscle of crayf ish( Procambarus ) with am¬
monium fractionation and DEAE-Toyopearl
column chromatography. The preparation
was shown to be homogeneous and the M.W.
was estimated to be 4 OK from SDS-PAGE.
Procambarus AK was composed of 348 ami¬
no acids and it was very similar to th¬
at of Cambarus AK (352 amino acids) re¬
ported previously by us. Rabbit antise¬
rum to Cambarus AK cross-reacted with
Procambarus AK by immunoblotting. Its
enzymatic properties were investigated
under the variable conditions. Effects
of pH, bimetal ions and KC1 concentrat¬
ions on AK activity were studied.

BI 35

DISTRIBUTION AND INTRACELLULAR LOCALIZATION
OF CATHEPSIN E IN MAMMALIAN BLOOD CELLS
S. Yonezawa, Dept, of Zoology, Fac . of Sci.,
Hokkaido Univ., Sapporo.

The intracellular localization of rat
neutrophil cathepsin E was examined by a
method involving disruption of cells in a
Parr cell disruption bomb and density cen¬
trifugation of the cavitate on discontinu¬
ous Percoll gradients, which allows the
segregation of three particulate fractions;
azurophil granule-rich (PI) , specific gra¬
nule-rich (P2) and plasma membrane-rich (P3).
A combined application of immunochemical
and electrophoretic methods revealed that
cathepsin E was associated with P3 and the
soluble fraction, strongly suggesting that
the enzyme is a nonlysosomal acid protease
in rat neutrophils (Arch. Biochem. Biophys.
267 (1988) in press) . A survey of acid
proteases in polymorphonuclear leukocytes
(PNL) , mononuclear leukocytes (MNL) and
erythrocytes (RBC) from other mammals with
anti-cathepsin D and anti-cathepsin E anti¬
bodies demonstrated that the distribution
of cathepsin E in mammalian blood cells was
species-specific: Cathepsin E was detected
in PNL, MNL and RBC from rats and rabbits,
in PNL and MNL from guinea gigs, and in RBC
from humans, but not in RBC from guinea
pigs, pigs or cows, nor in PNL or MNL from
humans, pigs or cows. The results would
preclude a possibility that the function of
cathepsin E may be directly or closely re¬
lated to specialized biological functions
of these cells.

BI 36

Z-PHE-ARG-MCA HYDROLYZING ACTIVITY IN THE
TADPOLE TAIL OF BULLFROG, RANA CATESBEIANA.

K. Kobayashi and S.Horiuchi.

Life Sci. Inst., Sophia Univ., Tokyo.

Z-Phe-Arg-MCA is the substrate of lyso-

somal thiol proteinases, cathepsins B and

L. A Z-Phe-Arg-MCA hydrolyzing activity
was detected in a tadpole tail. The
enzyme activity in a tail extract showed
the maximal velocity at pH 7.6. Dithio-
threitol activated the enzyme(s), and
leupeptin, E-64 and iodoacetic acid
inhibited it. The enzyme(s) is probably
a thiol proteinase. The activity in fin
and skin was 29.1 mU/mg at premetamorphic
stage (St.IX-XV), 28.1 mU/mg at the onset
of metamorphosis (XIX) and 22.3 mU/mg at
the climax of metamorphosis (XXI). On the
other hand, the activity in the other
tissues was 1.5 mU/mg (IX-XV), 1.8 mU/mg
(XIX) and 3.6 mU/mg (XXI). Tissue
fractionation studies showed that the most
of Z-Phe-Arg-MCA hydrolyzing activity was
found in nuclear fraction, and distribution
pattern of activity was different from
those of lysosomal acid proteinases. The
enzyme(s) was extracted by sodium phosphate
buffer (pH 6.0) containing 0.1 1 Triton
X-100 and subjected to acetone fraction¬
ation (0-70%). Gel filtration of acetone
fraction on a Sephadex G-100 column showed
that the most of the activity was eluted

at the void volume and the remaining
activity was eluted at the position of
molecular weight of 20,000-30,000.

BI 37

ENDOGENOUS INHIBITORS AGAINST THE THIOL
PROTEINASE IN METAMORPHOSING TADPOLE TAIL
OF RANA CATESBEIANA.

S.Fujita, K.Kobayashi and S.Horiuchi.

Life Sci. Inst., Sophia Univ., Tokyo.

Endogenous thiol proteinase inhibitors
were found in metamorphosing tadpole tail.
These inhibitors specially adsorbed on CM-
Papain Agarose, and further purified by
gel filtration and ion exchange column
chroinatography with Superose 12, Mono S
and Mono Q in FPLC system.

Molecular weight of these inhibitors were
determined as 140,000 and up to 2,000,000
by gel filtration, and smaller one
contained two kinds of the inhibitors
which were pi 9.6 and 5.9.

20jug of the inhibitor (pi 9.6) inhibited
the activity of 0.125pg of the thiol
proteinase in tadpole tail. Also the
activities of ficin and papain were
inhibited by this inhibitor, but the
activities of tadpole tail cathepsin D-
like proteinase, bovine cathepsin D,
pepsin, chymotrypsin and trypsin were not
inhibited .

It seems that these inhibitors involved
in the regulation of proteolysis in
metamorphosing tadpole tail resorption.

Biochemistry

1235

BI 38

PURIFICATION AND CHARACTERIZATION OF
NUCLEIC ACID-STIMULATED PROTEASE FROM
XENOPUS EMBRYOS.

Is. Miyata, M. 2 Sendai and1 H. K. Kihara

1 . Lab, of Research for Biosynthesis
and Metabolism, Keio Uni., School of
Medicine, Tokyo.

2. Dep. of Chemistry, College of
Humanities and Science, Nihon Univ.
Tokyo .

A thiol protease was purified from
Xenopus embryos. This protease had
a molecular weight of 43,000-44,000
and was composed of two subunits with
molecular weight 30,000 and 13,000.
The optimal pH of the reaction
catalyed by the protease was around pH
3.8 but, in the presence of RNA, the
optimal pH was shifted from pH 3.8 to
pH 4.2. The protease was activated by
addition of total RNA of Xenopus
embryos or nucleotides. In contrast,
on addition of tRNA, the activation of
the protease was not observed. The
results obtained suggest that the
protease may participate in the
degradation of proteins in reactions
that are activated by nucleic acids or
nucleotides .

BI 39

EXPRESSION OF SUBSTANCES IMMUNOLOGICALLY RELATED
TO EMBRYONIC CHICK PEPSINOGEN IN STOMACHS OF
ADULT LOWER VERTEBRATES.

S.Yasugi1, T.Matsunaga1 and T.Mizuno2. 1Zool. Inst.,
Fac. Sci., Univ. of Tokyo, Tokyo and 2Fac. Pharmac.
Sci., Teikyo Univ., Kanagawa. _

We have shown that the stomach gland cells of
embryonic and adult chick express embryo-type
(ECPg) and adult-type (ACPg) pepsinogen, respec¬
tively, and that all vertebrate species examined
contained substances immunoreactive to the anti-AC
Pg antiserum. In the present study we studied the
occurrence of substances immunologically related
to ECPg, an avian homologue of mammalian pro-
chymosin, in the stomach gland cells of 21 adult
vertebrate species by immunohistochemistry . Gland
cells of higher vertebrates did not show reac¬
tivity against anti-ECPg antiserum, but those of
teleosts and elasmobranchs reacted to the anti¬
serum. Zymogram and immunoblot of stomach extracts
revealed that anti-ECPg-reactive substances
possess true peptic activity. The results suggest
that ACPg- and ECPg-type pepsinogens originated as
early as with evolution of teleosts and elasmo¬
branchs, and that ECPg-like pepsinogens are
expressed in the stomachs of adult fish whereas
their expression is confined to the embryonic or
young periods in higher vertebrates.

BI 40

<X,(*-TREHALASE FROM THE ART EMI A EMBRYOS
Z.Nambu1 and F.Nambu-Akiyama2 . ^Dept. of
Biol. School of Nursing and Med. Technol.
Univ. of Occupat. and Environm. Health,
Japan, Kitakyushu, 2Dept. of Chem. School
of Med. Univ. of Occupat. and Environm.
Health, Japan, Kitakyushu. _

Changes of the o(,o(-tr ehalase activity as

well as its properties were investigated in
the developing Artemia embryos and further,
the enzyme was purified from the nauplius.

The optimal pH of the enzyme changed
from 8.0 to 7.5 before the emergence of the
prenauplius and remained unaltered there¬
after. Most of the enzyme activities was
present in a soluble form and a part of it
was recovered in a pigment layer after cen¬
trifugation. On the other hand, microsomal
trehalase activity decreased during the
development. The total activity of the
embryos was leveled up at the emergence of
the prenauplius as well as during the de¬
velopment of the nauplius.

The soluble enzyme from the nauplius was
purified more than 360 fold by the follow¬
ing 5-step procedures, acetone treatment,
DEAE-Seph arose CL-6B , Sephadex G-75, Con A-
Sepharose and Sephadex G-75 column chroma¬
tography. The enzyme had a molecular weight
of 64,000 on SDS-ur ea-PAGE. However, the
active enzyme had a slightly higher molec¬
ular weight on SDS-PAGE. The apparent Km
of the enzyme was 24 mM in phosphate buffer
pH 7.5, The enzyme was highly specific for
trehalose, only cellobiose being hydro¬
lyzed to a limited extent.

BI 41

CONFORMATIONAL CHANGES OF 21S DYNEIN ATPASE
COUPLED WITH ATP HYDROLYSIS. (I)

K.Inaba and H.Mohri. Department of Biology, Univ.
of Tokyo, Tokyo. _ _ _ _

Conformational changes of 21S dyne in ATPase from

sea urchin sperm flagella were examined by proteo¬
lytic digestion under physiological conditions. In
the presence of 2mM ATP (or ADP) plus lOOyM Vi,
dyne in heavy chains were digested by trypsin into
quite different polypeptides from those obtained in
other cases (none, 2mM ATP, 4mM AMPPNP, 4mM AMPPCP,
2mM ADP , lOOyM Vi) . In the presence of 4mM ATPyS ,
however, the digestion pattern was similar to that
in the presence of ATP (or ADP) and Vi to a certain
extent. In all other conditions 165kDa and 135kDa
polypeptides were the main products and remained
stable, whereas in the presence of ATP (or ADP) and
Vi 300kDa, 200kDa, 150kDa/148kDa and 105kDa/96kDa
were produced. The latter digestion pattern was not
observed in the absence of any divalent cations
even upon the addition of ADP and Vi. The addition
of ADP and Vi to the products (165kDa and 135kDa) ,
which were obtained by the tryptic digestion in the
absence of ADP and Vi , caused further degradation
of the two polypeptides into smaller ones (150kDa/
148kDa and 105kDa/96kDa) . We further performed
quantitative analysis of the degradation of 165kDa
and 135kDa by means of the estimation of the first-
order rate constants for the degradative reactions
of these polypeptides . The results obtained above
indicated that, in the ATP hydrolysis cycle, dynein
changes its conformation only slightly in dynein-
ATP state and remarkably in dynein-ADP-Pi state,
which is presumably responsible for force
generation.

1236

Biochemistry

BI 42

CONFORMATIONAL CHANGES OF 21S DYNEIN ATPASE
COUPLED WITH ATP HYDROLYSIS. (II)

K.Inaba and H.Mohri. Department of Biology, Univ.
of Tokyo, Tokyo. _

Tryptic digestion of 21S dynein heavy chains

from sea urchin sperm flagella produced the stable
two polypeptides, 165kDa and 135kDa, which are
likely to be the main components of Fragment A. The
addition of ADP and Vi to the products caused
further degradation of two polypeptides (165kDa and
135kDa) into smaller ones, lmmunos taming of the
digestion products with polyclonal anti-165kDa and
anti-135kDa IgG, which were prepared from anti-21S
dynein serum by blot -purification, indicated that
165kDa and 135kDa are degraded into lS0kDa/148kDa
and 96kDa, respectively. Photoaffinity labeling
with [a-32p] 8-N3ATP and photosensitized cleavage
in the presence of ATP and Vi revealed that both
adenine- and vanadate -binding sites are located on
165kDa. The results indicated that the
conformational change of 165kDa region by binding
and hydrolyzing ATP spreads to 135kDa region. ATPase
activity of 21S dynein was highly activated by
tryptic digestion in the absence of ADP and Vi.
However, in the presence of ADP and Vi during
tryptic digestion, ATP ase activity was gradually
decreased to 50% of the initial activity. When ADP
and Vi were added to once activated preparation,
ATPase activity was also decreased. Digestion by
trypsin of two components of 21S dynein dissociated
by low salt treatment, a and 8, resulted in
producing polypeptides similar to each other but a
little different from those of intact 21S dynein.
Digestion of 21S dynein by chymo trypsin and
thermolysin also showed peculiar patterns in the
presence of ADP and Vi.

BI 44

EFFECTS OF EDC , A ZERO-LENGTH CROSSLINKER,
ON THE MICROTUBULE-DYNEIN COMPLEX.

T. Shimizu1*2, M. Hato1 , S.P. Marchese-
Ragona2, K.A. Johnson2, and N. Hosova3.
TRes. Inst. Polym. Text., Tsukuba, 2Dept.
Molec. Cell Biol., Pennsylvania State
Univ., PA, USA, and 3pept . Biol., Coll.
Art. Sci., Univ. Tokyo, Meguro, Tokyo.

The microtubule-dynein (MT-D) complex
consisting of 22S dynein from Tetrahymena
cilia and MAP-free MTs was subjected to
treatment with various concentrations of
EDC at 28°C for one hour. Following
crosslinking of the complex, nearly all of
the ATPase activity cosedimented with the
MTs even in the presence of ATP. EM
observation by negative staining revealed
that the MT-D complex treated with 1 mM
EDC was not dissociated by ATP although
the dynein decoration pattern was dis¬
torted. The complex treated with 3 mM EDC
exhibited normal MT-D pattern even after
the addition of ATP. The ATPase activity
of the MT-D complex was enhanced about 30-
fold to 12 umoles/min/mg dynein (150 s-1)
by the treatment with 1-3 mM EDC. These
results indicate the ATPase activation was
caused by the close proximity of the
dynein ATPase sites to the MTs and provide
further support for the functional inter¬
action of all three dynein heads with the
MT.

BI 43

ROTATION AND TRANSLOCATION OF
MICROTUBULES IN VITRO INDUCED BY
DYNEINS FROM TETRAHYMENA CILIA.
Y.Y.Toyoshima 1 , R.D.Vale3. 1Dept. of Biology,
Ochanomizu University, Tokyo, 2Dept. of Pharmacology,
Univ. of California San Francisco, USA. _

Dynein, the force-generating enzyme that powers the
movement of cilia and flagella, has been characterized
biochemically, but no simple system has been available for
examining its motile porperties. Here we describe a
quantitative in vitro motility assay in which dynein
adsorbed onto a glass surface induces linear translocation
of purified bovine microtubules. Using this assay, we show
that both 22S and 14S dyneins from Tetrahymena cilia
induce movement but have distinct motile properties. A
unique poperty of 14S dynein, which has not been
described for other motility proteins, is its ability to
generate torque that causes microtubule to rotate during
forward translocastion. In the axoneme, 14S dynein-
induced torque may play an important role in generating
three-dimensional ciliary beating patterns. To explore
whether the multi-headed structure of 22S dynein was
required to generate motiliy, we digested the three-headed
22S dynein into two-headed and single-headed fragments
using chymotrypsin. The single-headed fragment did not
induce movement, even though it was capable of binding
to mictotubules. The two-headed fragment, on the other
hand, induce microtubule movement at velocities similar
to those described for intact 22S dynein. These findings
indicate that the intact three-headed structure of 22S
dynein is not required for generating microtubule
movement in vitro.

BI 45

AMINO ACID SEQUENCE ESSENTIAL FOR STIMULAT¬
ING MICROTUBULE ASSEMBLY.

S.Kotani, G.Kawai1-, H. Aizawa1-, S. Yokoyama1- ,
H. Murof ushil and H. Sakai1-. Fac. of Comp.
Sci. and Syst . Eng., Kyushu Inst, of Tech.,
Iizuka, -'-Dept . of Biophys. and Biochem. ,

Fac. of Sci., Univ. of Tokyo, Tokyo.

We have reported isolations of micro¬
tubule binding fragments of two microtubule-
associated proteins(MAPs ) , 190-kDa MAP and
tau [Aizawa et al . , J. Biol. Chem. 262,
3782-13787 (1987); Aizawa et al . , J. Biol.
Chem. 263, 7703-7707 (1988)]. We compared
the amino acid sequence of the two frag¬
ments and identified a common sequence of 20
amino acids. A synthetic polypeptide corre¬
sponding to the common sequence decreased
the critical concentration of tubulin for
polymerization. The polypeptide-induced
polymers were morphologically normal micro¬
tubules and cold unstable. The common se¬
quence was thus identified as the active
site of 190-kDa MAP and tau for the promo¬
tion of microtubule assembly. Proton nucle¬
ar magnetic resonance study suggested that
several amino acid residues such as Val,

Lys, Gly are close to tubulin molecule when
the polypeptide binds to tubulin. These re¬
sults are against the idea that polycationic
nature of the MAPs are essential for their
activity of binding to microtubules and
stimulating tubulin assembly.

Biochemistry

1237

BI 46

INTERACTION OF TETRAHYMENA CILIARY DYNEIN
WITH MICROTUBULES II.

S.Ohba and T.Miki-Noumura

Dept of Biology, Ochanomizu University,
Tokyo _

We reported previously interaction of
dynein with microtubules, using binding of
dynein isolated from Tetrahymenacilia to
singlet microtubules reassembled from por¬
cine brain tubulin. Vallee et al recently
reported that MAPsIC has ATPase and is
capable to translocate microtubules in
vitro, as similar to dynein( 1 987 ) . We
attempted here to study the interaction of
dynein with PC-tubulin, which was deprived
of MAPs. The interaction was analysed by
measuring turbidity with spectrophotometer,
and density of gel bands in electrophore¬
sis. On some occasion, a microtubule-
dynein complex could be visually confirmed
by electron and dark-field microscope.

PC-microtubules bound with dynein failed
to depolymerize by diluting with buffer
solution to below critical concentrations
of PC-tubulin. In the control experiment,
dilution induced complete depolymerization
of PC-microtubules into tubulin. Contrary
to the MAPs-microtubules, PC-microtubules
in the complex, were not depolymerized by 0
C incubation. The different responce of
PC-microtubules from the MAPs-microtubules
in the complex suggests a different binding
of dynein to PC-microtubules. We have
actually found that dynein bound to the PC-
microtubules in ATP-insensibive manner at
some region. Further study on this point is
carried out at present.

BI 47

COMPARISON OF TUBULIN ISOTYPES FROM VARIOUS
ORGANISMS.

S. Tanaka^ , E.Masuyama^, K. Nakamura^ , S.
Wada , M.Okuno2, and H.Mohri2 Dept, of
Living Sciences, Hiroshima Women's Univ.,
Hiroshima and 2Dept. of Biology, College of
General Education, Univ. of Tokyo, Tokyo.

Tubulins were isolated electrophoretic-
ally from sperm flagella of oyster and sea
urchin, and from cilia of Tetrahymena.
Analysis using isoelectric focusing (IEF)
and immuno-b 1 ott ing have been made to
compare the isotype constituents among
these organelles. The results have shown
that there are at least five similar
tubulin isotypes with same isoelectric
points in these species. However, the most
dominant isotype involved in beta-tubulin
of each species has shown a marked
different mobility in a IEF pattern.

A comparison between the isotypes of
the axonemal tubulins from sperm flagella
and cilia within a single organism has also
been made using sea urchin and oyster as a
sample. In both species, a subtle but
significant difference of IEF pattern has
been observed in a basic region of alpha-
tubulins.

During these analysis, we have found
that the stains-all staining is useful for
the determination of alpha- and beta-
tubulins after SDS-PAGE or IEF. That is,
isotypes from beta- tubu 1 ins are stained
blue, while those in alpha-tubulins red by
the staining with a stains-all after IEF.

BI 48

"NANOMETRY" OF K I NES IN-DEPENDENT MOTION OF
POLYSTYRENE MICROBEADS ON MICROTUBULES.
S. Kamimura. Dept, of Biol., College of
Arts and Sciences, Univ. of Tokyo, Tokyo.

Kinesin-dependent motion of polystyrene
microbeads (diameter, 0.33 pm) under
phase-contrast microscope was analyzed in
nanometric precision using a specific
photosensing system ("photosensing nano-
metry", Appl. Optics, 1987, 26:3425).

Kinesin was prepared from the crude
extract of porcine spinal cords using the
binding affinity to microtubules with AMP-
PNP/PPPi. Microbeads incubated with the
kinesin fraction, which surfaces were
covered with about 3000 molecules of 130kd
peptide of kinesin, showed active gliding
motion on demembranated sea urchin sperm
flagella. I could not observe the micro¬
beads to move stepwise reflecting the
lattice structure of microtubules. The
motion at 5 y M ATP was relatively smooth
during 20-50 nm displacement along the
microtubules. After the smooth motion the
microbeads sometimes jumped by about 10 nm
at a rate more than 0.5 pm/s.

From the estimated number of kinesin
molecules between the microtubules and the
microbeads (about 10), and the low rate of
ATP hydrolysis by kinesin (less than 1/s),
it is assumed that during a cycle of ATP
hydrolysis the kinesin glide smoothly on
the microtubule surface, and that the next
ATPase cycle follows a small jumping
motion .

BI 49

PURIFICATION AND CHARACTERIZATION OF
SYMBIONIN , AN APHID ENDOSYMBIONT-SPECIFIC
PROTEIN

E.Hara and H.Ishikawa

Zool.Inst., Fac.of Sci., Univ. of Tokyo,
Tokyo.

Aphid endosymbionts in_ vivo synthesize
only one protein, symbionin, while the
isolated symbionts _in vitro synthesize
several hundred proteins. In the present
study, we purified symbionin from aphid
extracts and characterized the molecule by
electron microscopy and immunoblotting . The
purification was performed using ammonium
sulfate precipitation, hydroxylapatite and
DEAE-Sephacel column. Sephacryl S-400 gel
filtration suggested that symbionin is 12-
mer of 63kDa subunit. Electron microscopic
observation indicated a symbionin molecule
consists of two doughnuts which are stacked
on each other. Probably, each doughnut
contains 6 subunits. On immunoblotting,
polyclonal anti-symbionin antiserum cross-
reacted with not only symbionin, but also
SLP ( symbionin-like protein) and several
other proteins that are supposed to be
degradation products of symbionin.

1238

Biochemistry

BI 50

EFFECT OF HEAT TREATMENT ON THE INTRA¬
CELLULAR SYMBIOSIS IN APHID
C.Ohtaka and H.Ishikawa

Zool.Inst., Fac.of Sci., Univ.of Tokyo,
Tokyo .

Aphid endosymbionts in vivo are apparent¬
ly regurated by the host so that they may
produce only one protein, symbionin. It
seems that as the control of the host over
the endosymbiont is somehow weakened, some
of the symbiont's genes begin to be
expressed. Of interest is how this control
mechanism works .

To know this we determined effects of
heat treatment on the symbiosis system of
the pea aphid. Newborn nymphs of the aphid
were treated at 37 °C for hours. After the
treatment they were reared at 15 °C. The
resultant adults at 20 days were signifi¬
cantly smaller in size than controls, did
not produce progeny at all, and their body
color was unusually yellow. In addition,
endosymbionts of these insects at 20 days
neither contained norsynthesized symbionin.
While nymphs immediately after the treat¬
ment contained symbionin at a normal amount,
the symbionin content in the treated
insects were decreased over the course of
post-embryonic development. As was expected
from these results, the heat-treated
insects at 20 days did not contain normal
symbionts in the mycetocyte, which was
demonstrated by electron microscopic
observation .

BI 51

PARTIAL CHARACTERIZATION OF CULTIVABLE
BACTERIA RELATED TO APHID ENDOSYMBIONTS.
S.Sato and H.Ishikawa. Zool . Inst., Fac.
Sci., Univ. Tokyo, Tokyo. _

It has long been known that the aphid
has intracellular symbionts in its myceto¬
cyte. This system will be one of the
most suitable models to examine the
Symbiosis Theory as to the origin of
eukaryotic cells containing organelles.

To study this symbiotic system closely, it
is desireble to grow the symbionts
separately from its host. However,
a close interdependence between the
partners in this system has not permitted
this so far.

Recentry we found that there are at least
two kinds of cultivable bacterial organisms
in the symbiont fraction isolated from
the pea aphid Acyrthosiphon pisum. Southern
blot analysis of the genomic DNA indicated
that one of the cultivatable bacteria is
closely related to a symbiont.

The bacterium , Gram-positive, has a genome
with low GC-content ( ca . 30%) and grows
vigorousely extracel lulary under aerobic
condition. To estimate its phylogenic
position, its rRNA genes were cloned and
partialy sequenced. The result suggested
that the bacterium is different from any of
known bacteria.

/

BI 52

STRUCTUAL ANALYSIS OF NON-TRANSCRIBED
SPACER OF THE APHID rDNA
0. Y.Kwon,K.Ogino and H . Ishikawa . Zool .
Inst., Fac. Sci ., Univ. Tokyo , Tokyo .

With respect to molecular properties of
ribosomal RNA , the aphid is one of the few
exceptional groups among many eukaryotes.
The 18S aphid rRNA has molecular weight
significantly iarger than that of common
eukaryotic 18S rRNA. Also, the 28S rRNA from
aphids lacks the hidden break and has
larger size compared with those of other
insects. Evolutional ly these characters of
aphid rRNA are interesting.

We constructed a gene library of
Acyrthosiphon pisum and obtained four rDNA
clones that were different sizes.
Restriction mapping indicated that the
difference of the clones is due to variety
of the number of tandemly-repeated sequence
of 240bp in the non-transcribed spacer.

The repeated sequence was highly spacies-
specif ic .

By Sl-nuclease mapping, the initiation
site of rDNA transcription was determined
to be located on the last unit of the
repeated sequence about 1 . 2kbp upstream
from the 5 ' -end of the 18S rRNA sequence.
Sequence analysis of the flanking region
of the initiation site revealed presence
of several consensus nucleotides in spite
of the spacies-specif icity observed in the
repeated sequence.

BI 53

GLYCOGEN AND PHOS PHOR YLASE IN THE MALE
REPRODUCTIVE SYSTEM AND SPERMATOPHORE OF
THE SILKMOTH.

M. OSANAI , Y. KUNITOMO AND H. KASUGA.
Dept. Biol., Tokyo Metro. Inst. Geront.,
Itabashi, Tokyo and *Dept. Seric. Sci.,
Fac. Agr., Tokyo Univ. Agr. Tech., Fuchu.

The a cti-v i ty of glycogen
phosphory lase and the concentration of
its substrate, glycogen in the male
reproductive system of the silkmoth,
Bombyx mori were determined before and
after ejaculation. Their changes with
time were also determined in the
spermatophore formed in the female bursa
copulatrix. Both large quantities of
glycogen and high activities of
phosphory 1 ase were found in the testis,
ampulla d. deferentis plus d. deferens,
g. lacteola, v. seminalis and
spermatophore. Most of the glycogen and
phosphory 1 ase were present in the cell
secretions transferred, not in the cells.
Amount of glycogen and phosphory 1 ase
activity are transferred without
significant loss from the male
reproductive tract to the spermatophore.
After formation of the complete
spermatophore, the activity of
phosphory 1 ase keeps a constant niveau
until 300min after the beginning of
mating, while the amount of glycogen
decreases slowly with time. This
decrease coincides with the increase of
alanine in the spermatophore after
mating .

Biochemistry

1239

BI 54

CHARACTERIZATION OF THE TESTIS SPECIFIC a-
HYDROXY ACID OXIDASE FROM DROSOPHILA
VIRILIS.

Y.Taka , T.Aotsuka . Dept, of Biol., Fac.
of Sci., Tokyo Metropolitan Univ., Tokyo.

A testis specific a-hydroxy acid oxi¬
dase (aHAOX-T) was found in Drosophila
species in the virilis species group in
common, but it was not found in the
members of other species groups except for
D.sordidula in the robusta species group.

aHAOX-T was purified over 900-fold from
D .virilis adult males by ammonium sulfate
fractionation, ultracentrifugation, DEAE-
Sephdex chromatography, hydroxyapatite
chromatography, and gel filtration HPLC.

The native molecular weight determined
by gel filtration HPLC was about 240K. The
subunit molecular weight determined by SDS
electrophoresis was about 40K. Thus, the
native form is likely to take a hexameric
structure .

aHAOX-T oxidized various a-hydroxy
acids, and oxidized L-lactate most
efficiently among them.

Km value for L-lactate estimated by the
double reciprocal plot of Lineweaver and
Burk was 1.06mM. The pH optimum for the
oxidation of ImM L-lactate was 8.1.

BI 55

GUANOSINE 3 5 ' -MONOPHOSPHATE-DEPENDENT PRO¬
TEIN KINASE FROM SILKWORM PUPAE.

S. Y.Takahashi, Inst. of Biol, Fac. of Lib. Arts,
Yamaguchi Univ. , Yamaguchi , 753, JAPAN

Guanosine 3' : 5 ' -monophosphate-dependent

protein kinase (G-kinase) has been purified
to homogeneity from silkworm pupae and par¬
tially characterized. Some of their propert¬
ies were compared to those of the kinase
from other sources. The kinase is highly de¬
pendent on cGMP and contained specific cGMP
binding activity. The kinase has an apparent
molecular weight of 145,000 as determined
by gel filtration on Superose 12 column. On
SDS-PAGE , a single protein band of 72,000
molecular weight was observed. The enzyme
was strongly inhibited by GT I in a non-com¬
petitive manner not only for ATP but also
for cGMP. GTP was also inhibitory on G-ki-
nases from other tissues from silkworm but
did not inhibited the cAMP-dependent pro¬
tein kinase and the highly purified G-ki-
nase from monkey cerebellum. The abitlity
of several proteins to serve as substrates
in the kinase reaction was studied at a
fixed concentration of 0.5 uM cGMP. Various
histones and protamine as well as silkworm
vitellin may be phosphorylated by G-kinase
but vitellogenin was not effectively phos¬
phorylated by the kinase. G-kinases phospho¬
rylated vitellin with 30-40% of their pro¬
tamine or histone H2B phosphorylating ac¬
tivity, indicating that silkworm vitellin
was a preferential substrate of the kin¬
ases .

BI 56

DISCRETE LOCALIZATION OF TWO ALKALINE PHOS¬
PHATASE ISOZYMES IN THE CELL SURFACE OF
SILKWORM MIDGUT EPITHELIA .

M.Azuma, S.Takeda, T. Suzuki and M.Eguchi.
Lab. of Appl. Genet. Dept, of Appl. Biol.
Kyoto Inst, of Techn. , Kyoto.

Silkworm alimentary canal (midgut) has
two types of alkaline phosphatase (ALP)
isozymes; membrane -bound (m-ALP) and solu¬
ble (s-ALP). To elucidate their functions
in midgut cells, we have performed immuno¬
chemical analysis. Purified m-ALP and s-ALP
have a single polypeptide of Mr=58,000 and
61,000, respectively, and their enzymatic
properties are clearly different. The anti¬
body raised against purified m-ALP or s-ALP
was mono-specific for each ALP and did not
crossreact with each other.

Regional difference exists in both ALPs.
The m-ALP antigen is equally distributed
along the length of midgut except the most
anterior portion, where the m-ALP is defi¬
cient. The s-ALP is mostly spread in the
posterior midgut, followed by lesser amounts
in the middle and anterior, respectively.
Immunohistochemistry showed m-ALP was lo¬
cated at the brush border of columnar cells
and s-ALP at the apical surface of goblet
cells. The s-ALP may function as a certain
cation-stimulated ATPase, since goblet cells
in lepidopteran insects play a indispensa¬
ble role in ion transports and water regu¬
lations. Here we propose each ALP can be
utilized as a typical cell marker for each
type in silkworm midgut epithelia.

BI 57

XANTH0MMATIN GRANULES FOUND IN THE LARVAL
EPIDERMAL CELLS OF^THE SILKWORM.

H.Sawada1 ,M.Tsusue 1 jK.Dohke1 and S.Sakurai 2
1 Biol .Lab .Kitasato Univ.Sagamihara,2 Lab. of
molecular genetics Jikei Univ. I.N.S. Tokyo.

Three pigments were found in the
larval epidermal tissues from many strains and
mutants of the silkworm. They are under
genetical and hormonal controls. Sepiapterin is
detectable as a complex with a polypeptide,
localized in granules in the cells. Although
xanthommatin was found in many strains, larger
amount of the pigment was detected in quail
mutant. We obtained lem/lem, q/q mutant by cross
mating experiment, and found that sepiapterin
and xanthommatin are localized in different
granules in the epidermal cells. We purified the
pink granules from the epidermal cells by
ultracentrifugation. In the granules,
xanthommatin combined with protein to make
insoluble lump. Tween, deoxycholate and other
detergent could not solubilize the lump, but
only SDS could solubilize it, remaining still
insoluble precipitate. Amino acid analysis
showed that the protein is composed of large
amount of hydrophilic amino acid. Different from
free xanthommatin, the pigment was not oxydized
by hydrogen peroxide. On the contrary,
solubilized pigment showed yellow color. Pronase
also acted on the pink granules to yield yellow
supernatant and pink precipitate. Based on these
experiments, we speculate that an immino group
of xanthommatin may be combined with proteins to
make the reduced form of the pigment stable.

1240

Biochemistry

BI 58

MONOCLONAL ANTIBODY PRODUCTION AGAINST SEPIA-
PTERIN REDUCTASE FROM FAT BODY OF SILKWORM,
BOMBYX MORI.

T.Iino] K.Dohke^and M.Tsusue?

^Dept.of Gen. Educ., Nihon Univ., Tokyo.

2

Biol. Lab., Kitasato Univ., Sagamihara.

In the previous report, we have shown that
sepiapterin reductase from fat body of silkworm
is quite similar to that from mammalian sources.
In order to demonstrate whether the immunolo¬
gical difference is present between insect and
mammalian sepiapterin reductase, we tried to
make monoclonal antibody against sepiapterin
reductase from fat body of silkworm. The spleen
cells of BALB/c mice immunized twice against
purified sepiapterin reductase from fat body of
silkworm were fused with mouse myeloma cells
P3-Ag8-Ul (P3U1 ) . The hybridoma cells, which were
positive to sepiapterin reductase by the ELISA
and by immunoblotting method, were selected for
the preparation of the monoclonal antibody.

The monoclonal antibody which was designated as
"l-C-8" did not react with partially purified
sepiapterin reductase from rat erythrocytes and
liver. Although the catalytic property of
sepiapterin reductase from fat body of silkworm
is identical with that from rat enzyme, immu¬
nological studies show no apparent homology
between the two enzymes.

BI 59

ISOMERIZATION OF DIHYDRO-FORM OF 6-LACTOYL
PTERIN (SEPIAPTERIN) BY SEPIAPTERIN REDUCTASE
T. Sueoka and S. Katoh. Department of Bio¬
chemistry, School of Dentistry Meikai Uni-
versity, Sakado, Saitama 350-02

Besides the NADPH-oxidoreductase activity,
sepiapterin reductase (SPR) has an isomerase
activity against 6-lactoyl (Cl ' -keto) 5, 6, 7, 8
-tetrahydropterin, by which activity it is
converted into the C2 ' -keto isomer in the
absence of NADPH. As sepiapterin (6-lactoyl
7 , 8-dihydropterin) is also a good substrate
of SPR in the presence of NADPH, it is ex¬
pected to be isomerized by SPR. Although the
expected C2'-keto product of the dihydro-
ptenn has been believed to be too labile to
exist in practice, we have recognized herein
evidences of the actual occurrence of iso¬
merization of sepiapterin by SPR.

When sepiapterin was incubated with large
amount of SPR purified from rat erythrocytes
in the absence of NADPH, a novel colorless
compound was formed. It was detected ampero-
metrically (at 500 mV) and fluorometrically .
It was converted into dihydrobiopterin when
it was further incubated with SPR and NADPH.
By 270 MHz 1 H-NMR analysis in D20, instead
of a doublet (1.39 ppm) of the C3 ' -methyl
protons of sepiapterin, it showed a singlet
signal at lower range (2.05 ppm) . From these
results, it was identified as 6-1 ' -hydroxy-
2'-oxopropyl 7 , 8-dihydropterin. Thus SPR
also isomerizes sepiapterin. This finding
strongly supports our idea that the isomer¬
ase activity is involved in the total funct¬
ion of the NADPH-dependent reductase of SPR.

BI 60

EFFECT OF N-ACETYL SEROTONIN ON THE
ISOMERASE ACTIVITY OF SEPIAPTERIN REDUCTASE
S. Katoh and T. Sueoka. Department of Bio¬
chemistry, School of Dentistry Meikai Uni¬
versity, Sakado, Saitama 350-02

N-acetyl serotonin (NAS) was found by us

in 1982 as a specific and strong inhibitor
(Ki=0.17 piM) of the NADPH-dependent reduct¬
ion of sepiapterin by sepiapterin reductase
(SPR) . As a matter of course, NAS inhibites
the formation of tetrahydrobiopterin from 6-
pyruvoyl tetrahydropterin on the pathway of
its biosynthesis since SPR is necessary for
this process. Recently we demonstrated that
SPR has also an isomerase activity and this
new activity might be involved in the total
function of the NADPH-dependent reductase
activity of SPR in vivo.

However, it was inconsistent that the
isomerase activity was insensitive against
NAS if it was assayed in the absence of
NADPH. When lOOyM NAS was added with 6-
lactoyl tetrahydropterin, SPR and NADPH, both
consumption of the substrate and formation
of tetrahydrobiopterin were inhibited. If
100 uM NAS was added to the reaction system
containing 31 pM 6-lactoyl tetrahydropterin,
and 212 mU SPR in the presence of IpM NADP+
and 50 mM dithioerythritol at pH 8.6, NADP+
stimulated isomeriation of the substrate
(first rate constant: k,) as well as the suc¬
cessive reducing reaction of the isomerized
product, 6-1 ’ -hydroxy- 2 ' -oxopropyl tetrahydro¬
pterin to tetrahydrobiopterin (k2) was inhib¬
ited. The rate of inhibition of k2 (64%) was
larger than that of k, (29%) . It seemed, at
present, that NAS might inhibit the binding
of the pterin to the coenzyme bound SPR.

BI 61

LEVEL CHANGES OF CATECHOLAMINES DURING THE PUPAL
STAGE OF PAPILIO XUTHUS.

Y. Ishizaki and Y. Umebachi. Dept, of Biol., Fac.
of Sci., Kanazawa Univ., Kanazawa.

The pale yellow pigment, Papiliochrome II, in
the wings of Ft xuthus is a, combination of kynure-
nine and N-/3-alaayldopamine (NBAD) . In order to
elucidate the biosynthesis of this pigment, level
changes of free form of dopamine (DA), /3-alanine,
and NBAD in haemolymph and wings during the pupal
stage were already investigated.

This time, bound forms of catecholamines in
haemolymph and wings have been examined in further
details placing the focus on N-/3-acetyldopamine
(NADA). Before and after acid-hydrolysis of the
PCA-extract of haemolymph and wings, catecholamine
derivatives were separated by HPLC (ODS) and deter¬
mined with electrochemical detector (Coulochem,
Model 5100A). In haemolymph, there are bound-for¬
ms of DA, NBAD, and NADA during the pupal stage.

The bound-form of NADA has been identified as (3-
glucoside by HPLC, TLC, and hydrolysis with /3-glu-
cosidase. The quantity was on a level of 10 - 30
fig per 100 pi haemolymph, and at the middle pupal
stage (about 6th day), there was a broad peak of
it. On the other hand, in the wings, bound forms
of DA and NBAD were not found. But, the bound
form of NADA was found after emergence of butter¬
fly. The presence of a small quantity of NADA
glucoside in haemolymph around the 6th day and in
the wings after emergence may mean an inavtivation
of dopamine.

Biochemistry

1241

BI 62

COMPARISON OF CHEMICAL PROPERTIES OF PUPARIAL
CUTICLE BETWEEN YELLOW AND EBONY OF DROSOPHILA
MELANOGASTER.

T. Tagi, M. Aoki, Y. Hoda and Y. Umebachi. Dept,
of Biol., Fac. of Sci., Kanazawa Univ., Kanazawa.

Some chemical properties of the puparial case
have been compared among three kinds of mutant
strains yellow Q_) , black (_b) and ebony (e) . The
puparial case (exuvia) of is yellowish brown,
hard, and light. On the other hand, those of _b
and are whitish, hard (though less hard than in
y) , and heavier than y_. Protein content is higher
in _b and e^ than in y_. The quantity of/S-alanine in
cuticle is largest in y_, very small in _b. and zero
in ej. On the other hand, the quantity of ketoca-
techol which is released by hydrolysis in 1 N HC1,
at 100*0 for 5 hr, is larger in e^ than in y_. In
_b, it was between ^and _g_.

In order to investigate the cross-linking of
cuticle with soluble protein fraction, the extract
of cuticle in 4 % HCl-methanol was used. In the
extract too, the above-mentioned tendencies of ft-
alanine and ketocatechol in y_, _b, and e^ have prov¬
ed to be kept. Furthermore, the fraction of cate¬
chol derivatives in y_ and e_ was separated with
affinity chromatography which contains boronic
acid as ligand, and the peak of catechol derivati¬
ves was estimated. Here too, catechol fraction
was bigger in e^ than in y_. All these results su¬
ggest that the proportion of /3-sclerotization in
cuticle protein is higher in e_ than in y_. But th¬
is does not necessarily mean that y_ has only a li¬
ttle activity for /3-sclerotization. Because the
puparial cuticle of y_ to which dopamine was given
releases much more ketocatechol on hydrolysis.

BI 63

FURTHER INFORMATION ON CHEMICAL PROPERTIES OF THE
DEEP YELLOW PIGMENT OF PAPILIO MACHAON.

Y. Umebachi. Dept, of Biol., Fac. of Sci., Kanaza¬
wa Univ., Kanazawa.

When the deep yellow scales of _P. machaon are
extracted with 70 % ethanol, the extract is yellow,
and its main pigment is Papiliochrome II. But the
remaining scales are still deep yellow. The latt¬
er pigment can be extracted with 1 % HCl-methanol .
The extract contains the deep yellow pigment (Pa¬
piliochrome M) to which some protein combines.

The decomposition and degradation of the pigment
was investigated with amino acid analysis and two-
dimensional thin-layer chromatography. The pigme¬
nt decomposes to kynurenine and N-/3-alanylnorepi-
nephrine (NBANE) by being heated to 10'2 -10*' N HC1
at 106 "C for 2-3 hr. On hydrolysis in 1-2 N HC1
at 106 °C for 2-3 hr, NBANE decomposes to ft -alanine
and norepinephrine. With hydrolysis in 3-6 N HC1,
the latter substance degrades or changes further.
All the hydrolysates do not contain 3-hydroxykynu-
renine, dopamine, and ketocatechol. These results
suggest that Papiliochrome M is formed from kynu¬
renine and N-/3-alanyldopamine. Molar ratio of ft-
alanine to kynurenine was 2.13±0.09 (9). Among
total amino groups of ft -alanine, only 42.5 ±4.4
(5) per cent was free, and the remainder (about 60
per cent) seems to be bounded to some other group.
The above-mentioned 1 % HCl-methanol extract can
be further divided into two fractions: water-inso¬
luble and water-soluble. The latter can be furth¬
er divided into two pigment fractions. One is ad¬
sorbed to DEAE-cellulose and the other is not ad¬
sorbed. The latter seems to be the major pigment.

BI 64

CLONING AND SEQUENCE ANALYSIS OF CDNA
ENCODING THE PRECURSOR OF THE HORNET
VENOM PEPTID , MAS TOP ARAN X .

Il.Hirakawa , E.Munekata , T.Yasui1, ^
K.Saito , T.Yasuhara^ and T. Nakaj ima“ .

1 Institute of Applied Biochemistry ,
^niversity of Tsukuba , Tsukba, 305 and
“Faculty of Pharmaceutical Sciences,.
University of Tokyo, Bunkvo-ku, 113.

Mastoparans are tetracec a peptides

isolated from v/asp and hornet which is
known to cause degranulation of mast
cell and endogenous histamine release.

The cDMA clones encoding mastoparan
X, isolated from hornet, Vespa
xathoptera , were identified from a cDHA
library of hornet abdomen by means of a
synthetic oligonucleotide probe. The
nucleotide sequence from the cDUA clones
showed hydrophobic moiety at N-terminus
typical for signal sequence, and mas¬
toparan X sequence is located C-terminal
end of the precursor with a glycin at C-
terminus which is characteristic for C-
terminal structure. Generally, biologi¬
cally active peptides are processed at
the site of paired basic amino acids,
but prosequence of mastoparan X
possessed no such a site, whereas it has
a region with 7 amino acid residues
preceeding the sequence of mastopan X as
in the prostructure honeybee melittin.
The conversions of promelittin and
promastoparan may be occurred in same
mechanisms .

BI 65

PURIFICATION AND CHARACTERIZATION OF AN
ANTIOXIDANT IN THE LARVAL HEMOLYMPH OF THE
ARMYWORM , Pseudaletia separata.

Y.Hayakawa Biochem. Lab., The Inst, of Low
Temp. Sci., Hokkaido Univ., Sapporo. _

During the course of study on the
chemical nature of larval hemolymph of the
armyworm, an antioxidant was found in the
hemolymph of this insect. The antioxidant
becomes undetectable 2 days after the final
molting, whereas it's concentration remains
at relatively higher level until 4th day of
last" instar larvae parasitized by the para-
sitoid, Apantales kariyai , which prevents
the pupation of the host insect. The anti¬
oxidant was purified using a reversed-phase
HPLC. The purified antioxidant shows
neither the nature of peptide nor the
nature of amino acid but should contain
amino residue(s) because it reacts with
ninhydrin reagent. The antioxidant also
exhibits the three absorption maxima at
227nm, 268nm and 374nm, and the spectra was
largely altered by H202 in the presence of
a small amount of peroxidase. The complete
structure of the antioxidant awaits further
investigation. The physiological function
of the antioxidant is unknown. It may be
possible that the antioxidant- is responsi¬
ble for the metabolic fate of juvenile
hormone (JH) in the hemolymph of the host
insect since it was found that the anti¬
oxidant disappears from the hemolymph
almost simultaneously with the disap¬
pearance of JH from the unparasitized
larval hemolymph.

1242

Biochemistry

BI 66

STIMULUS-INDUCED PROTEASE RELEASE FROM
HEMOCYTES OF THE ASCIDIAN, Halocvnthia roretzi.
K. Azumi and H. Yokosawa. Dept, of Biochem., Fac. of
Pharmaceutical Sci. , Univ. of Hokkaido., Sapporo

During the research for understanding the roles
of hemocytes on defence mechanism of H. roretzi. we
found that protease release from hemocytes was
induced by calcium ionophore, A23187, or phorbol 12-
myristate 13-acetate, PMA. First, we isolated this
protease by three chromatographic operations using
phenyl-Sepharose, hydroxylapatite, and DEAE-5PW.
The molecular weight was determined to be 11,000 by
gel filtration in the presence of 0.01% Brij-35. The
optimum pH was 8.5. The enzyme showed narrow
substrate specificity (it hydrolyzed only Suc-Leu-Leu-
Val-Tyr-MCA), and it was inhibited strongly by o-
phenanthroline and EDTA, and weakly by
diisopropylphosphorofluoridate. Next, we analysed
mechanisms of protease release induced by A23187 or
PMA. The protease release was inhibited by H-7
(protein kinase C inhibitor) and W-7 (calmodulin
inhibitor). Specific groups of hemocytes isolated from
the whole hemocytes by discontinuous bovine serum
albumin density gradient centrifugation released the
protease in response to A23187 or PMA.
Lipopolysaccharide also induced the protease release,
but 81-3 glucan didn’t. Our preliminary study showed
that hemocytes contained at least two kinds of factors
stimulating the protease release.

Thus, H. roretzi hemocytes would function in
defence mechanism, releasing at least a protease in
response to extraneous and endogenous stimuli.

BI 67

THE HEMOLYMPH PLASMA OF AN ASCIDIAN,
HALOCYNTHIA RORETZI , INCLUDES AMIDOLYTIC
ENZYME AND ITS INHIBITOR: PURIFICATION OF
THE INHIBITOR.

T.Abe, F.Shishikura and K. Tanaka, Dept, of
Biol., Hihon Univ. Sch. of Med., Tokyo.

The factor inhibiting amidolytic enzyme
in the hemolymph plasma was obtained from
the same plasma of Halocynthia roretzi .

Purification of the factor was performed
as follows; precipitation with 80% satura¬
tion of ( NH^ ) 9SO4 , the gel filtration with
Toyopearl HW-65F column, and the chromatog¬
raphy with DEAE Toyopearl 650M column and
CM Toyopearl 650M column.

The activity of inhibition was assayed
using hemolymph plasma or enzyme fractions
as enzyme sourses, and Boc-Leu-Gly-Arg-MCA
as a substrate ( Tris-HCl/pH8 . 0 , 30min, 25C)

The activity was observed in a pass¬
through fraction of the DEAE chromatography
A highly active fraction was obtained in a
single peak on the further chromatography
with CM column eluted with NaCl gradient.
This fraction after the treatment with 2-
mercaptoethanol , however, showed three
bands in SDS-PAGE.

The inhibitory fraction of the gel fil¬
tration did not inhibit trypsin (lOug/ml)
to Boc-Phe-Ser-Arg-MCA although the same
amount of the fraction inhibited the
amidase in the ascidian hemolymph plasma.

BI 68

PURIFICATION AND CHARACTERIZATION OF
HEMAGGLUTININ IN THE HEMOLYMPH OF THE
SILKWORM, BOMBYX MORI
K. Amanai, S. Sakurai and T. Ohtaki
Depertment of Biology, Faculty of Science,
Kanazawa University, Kanazawa 920

As reported previously, the protein with
hemagglutinating activity in the hemolymph
of Bombyx mori may have an important role
in morphogenesis underlying the pupal and
adult development. The hemagglutinin was
partially purified by precipitation with
307, saturated ammonium sulfate and gel
filtration on Sephacryl S-300. From the
elution pattern of activity from Sephacryl
S-300, a 350kDa protein was estimated as
hemagglutinin. For identification of the
protein responsible for the
hemagglutinating activity, antibody was
prepared. Antibody raised against a 350kDa
protein inhibited hemagglutinating
activity. For further purification
affinity chromatography was performed
using glucuronic acid or galacturonic acid
as ligand. After that, we obteined two
proteins with molecular weight of about
88kDa and 90kDa. Western blotting analysis
using antibody raised against a 350 kDa
protein revealed that Bombyx hemagglutinin
was a tetramer composed of 2 different
subunits with molecular weight of about 88
kDa and 90kDa.

BI 69

SEPARATION OF THE COMPONENTS IN PROPHENOL-
OXIDASE ACTIVATING SYSTEM IN PLASMA OF THE
SILKWORM, Bombyx mori

M, Ochiai and M. Ashida. Institute of Low
Temp. Sci., Hokkaido Univ., Sapporo. _

The prophenoloxida se ( proPO ) activating
system of insect hemolymph is a cascade,
which is triggered with elicitors such as
8 - 1 , 3 -glucan ( 8 -G ) or pept idoglycan. The
system is considered to play some important
roles in the defence reaction in insect
hemolymph. So far, the interaction of
elicitor with its recognition protein is
known to initiate the sequential
activations of two serine enzyme zymogens,
proBAEEasefa precursor of an esterase
hydrolyzing benzoy larginine ethyl ester)
and proPPAE(a precursor of proPO activating
enzyme ) .

In the present study, we tried to
fractionate the plasma components involved
in the activation of proBAEEase by 8-G.
After column chromatography of silkworm
plasma on Sephadex G-150 two fractions
(fraction A and fraction B) were obtained.
Fraction A and B were shown to contain
proBAEEase activating factor contaminated
with 8-1,3-glucan recognition protein(8-
GRP) and proBAEEase, respectively. BAEEase
activity appeared in the mixture of
fraction A, B and 8-G, but not in the
mixture of fraction B, purified 8-GRP and
8-G. These results suggest that in addition
to 8-GRP and proBAEEase other factor(s) is
involved in the activation of proBAEEase in
silkworm plasma by 8-G.

Biochemistry

1243

BI 70

IMMUNOLOCALIZATION OF PROPHENOLOXIDASE
AMONG HEMOCYTES OF THE SILKWORM, Bombyx
mori . 1 ?

Ijl.Ashida ,M.Ochiai and T. Niki
Biochem. Lab. The Institute of Lo^ Temp.
Sci . , Hokkaido Univ., Sapporo, and Lab. of
Cellular Technol., Faculty of Technol.,
Takushoku Univ., Hachi -oj i -shi . _

Silkworm ( Bombyx mori ) hemocytes were
fixed with 3% glutaraldehyde immediately
after collection and embedded in Lowicryl
K4M resin. Thin sections of the hemocytes
were stained by an indirect immunogold
staining method using rabbit anti-
prophenoloxidase/lgG as a primary antibody
and colloidal gold coated with goat anti-
rabbit/lgG as a secondary antibody.

Electron micrographs of the sections
revealed that only plasmatocytes and
oenocytoids have prophenoloxidase both in
cytoplasm and nucleus whereas granulocytes,
spherulocytes and prohemocytes do not have
appreciable amounts of the proenzyme.
Cytoplasmic inclusions of oenocytoids also
contain the proenzyme.

Once materials in the granules of
granulocyte were discharged into the plasma
and formed coagulums, they cross-reacted
with anti-prophenoloxidase/lgG, suggesting
that prophenoloxidase was trapped in the
coagulums by unknown mechanisms. This
obsevation is discussed in relation to the
dispute concerning the presence of
prophenoloxidase or phenoloxidase in the
granulocyte .

BI 71

ACTIVATION OF LACCASE-TYPE PROPHENOLOXIDASE
IN THE CUTICLE OF INSECT. V.

PROCESSING DURING PRO-LACCASE ACTIVATION.
H. I . Yamazaki . Biol., Lab., Atomi Gakuen
Women's Univ.. Saitama. _

Cuticular phenoloxidase that is laccase-
type seems to play an important role in the
sclerotizing process of insect cuticle.
The existence of the pro-laccase was shown
in tf-chymo tryp t i c digestion from newly
ecdysed pupal cuticle in Bombyx mori .

However , ac t i vat i on process of the laccass
type phenoloxidase in native cuticle was
not yet obvious. And it was attempted to
elucidate the occurrence of pro-laccase
peptide and the processing mechanism of the
peptide in the course of hardening and
darkening by i mmunob 1 o t t i ng analysis.

A piece of cuticle was punched out from
each pupal stage and subjected to SDS-PAGE.
The separated peptides on the polyacryl¬
amide gel were transferred to nitrocellu¬
lose filter. The transferred peptides was
visualized by immunodetection.

A major peptide that was reactive to
anti-laccase antiserum was detected in the
newly ecdysed pupae. The molecular weight
of peptide was 150 kDa. After 6 hours
later, 92 kDa and 52 kDa peptides were
visualized, on the other hand, 150 kDa
peptide was reduced evidently. Moreover,
52 kDa and 38 kDa peptides were found 12
hour after ecdysis, at the time 92 kDa
peptide became faint.

BI 72

BIOSYNTHESIS OF PUPAL CUTICULAR PROTEINS
OF Bombyx mori .

H.Nakato, S.Izumi and S.Tomino

Dept. Biol. Tokyo Metropol. Univ., Tokyo

Biosynthesis of pupal cuticular proteins
(PCPs) was studied at protein and mRNA
level during larval-pupal transformation
of Bombyx mori .

Immunohistochemical study with the
affinity-purified antibody indicated that
PCPs localize only in the cuticular layers
of pupal integuments.

A cDNA clone for a PCP component was
isolated from a library constructed from
poly (A) RNA of the pupal epidermal cells.
Northern blot analysis of epidermal mRNA
demonstrated that the developmental
profile of PCP-mRNA closely reflects that
of the hemolymph concentration of ecdy-
steroids, i.e. the level of PCP-mRNA
rises at each larval molt, while marked
elevation of the level was noted immediate¬
ly after the larval-pupal ecdysis .
Experiments employing the isolated abdomen
of the last instar larvae proved the
positive involvement of ecdysteroids in
the induction of PCP-mRNA synthesis in the
epidermal cells.

Partial primary structure of a PCP
component was deduced from the nucleotide
sequence of a cloned cDNA. PCP exhibited
a compactly folded secondary structure with
four repeating domains, each of which
comprising -Trp-Asn-Ala-Pro-Ala-His-Gln-
Asp/Ser- sequence.

BI 73

PURIFICATION OF V I TELL I N BINDING PROTEIN
FROM OVARIAN MEMBRANE OF Locusta migratoria
K. Yamasaki and H . M i yakawa . , Dep t . 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
However, the VBP in membrane fraction has
not been purified, so the chemical character
istics is yet unknown. An attempt to iso
late VBP in membrane fraction was carried
out and some successful results were obtain
ed .

VBactivity was determined electro-washing
method and nonspecific protein binding act
ivity was estimated using FITC-bovine serum
albumin. Membrane fraction from ovary was
fractionated by solubilization with the use
of octyl-B-glucoside, sodium dodecylsul f ate
and urea. Five preparations was obtained
and designated as OGs , OGp , OGs-CM , OGp-SDSs
and OGp-SDSp . OGp-SDSp was purified 28.2
times as compared with starting material.
Polysaccharides were found in all fractions
by pheno 1 -su 1 f ur i c acid method and positive
for PAS reaction on SDS-PAGE. Fraction OGs
was highly nonspecific one, BSA could bind
0.4 times compared with vitellin. Recovery
of total activity in fractions obtained
finally was 97.5 %.

1244

Biochemistry

BI 74

STRUCTURE AND EXPRESSION OF VITELLOGENIN
GENE IN THE SILKWORM, Bombyx mori
K.Yano, M.Toriyama, S.Izumi and S.Tomino
Dept. Biol. Tokyo Metropol . Univ. Tokyo

Vitellogenin of the silkworm, Bombyx mori
is synthesized in the female fat body at
larval-pupal ecdysis and its synthesis is
regulated in a sex- and stage-specific
manner at the level of mRNA. The B. mori
vitellogenin is composed of each two
molecules of heavy chain (Mol. wt. 180,000)
and light chain (Mol. wt. 42,000), each of
which is coded for by distinct mRNA.

We have cloned a portion of mRNA sequence
for each vitellogenin subunit and analyzed
their structures. Despite the lack of
homology between nucleotide sequences of
the cloned regions of two cDNAs , the light
chain cDNA hybridized with the light chain
mRNA as well as the heavy chain mRNA. The
heavy chain cDNA also cross-hybridized with
mRNA for each subunit.

Genomic clones for vitellogenin subunits
were isolated from a B. mori gene library.
Southern blot hybridization indicated that
the genomic regions complementary to cDNAs
for two vitellogenin subunits localize in
close vicinity on the cloned DNA fragment.
The result of RNA blot analysis with the
genomic DNA as probe together with that of
sequence analyses suggested that mRNA for
each vitellogenin subunit is most likely to
be derived from a common pre-mRNA by
processes involving alternate splicing.

BI 75

MONOCLONAL ANTIBODY AGAINST APOLIPOPHORIN-
II INHIBITS THE UPTAKE OF DI ACYLGLYCEROL
FROM FAT BODY BY LIPOPHORIN IN LOCUST,
LOCUSTA MIGRATORIA

T.Hiraoka and Y.Hayakawa. The Inst, of Low
Temp. Sci., Hokkaido Univ., Sapporo. _

A major insect haemolymph lipoprotein
"lipophorin" has two apoproteins,
apoprotein I (Apo-I, 250K) and apoprotein
II (Apo-II, 85K). It has been reported
that Apo-I constitutes the surface of
lipophorin together with phospholipids, on
the contrary, Apo-II exists mainly inside
the lipophorin particle. Lipophorin serves
to transport lipids such as diacy lglycerol
(DG) from sites of storage (e.g., fat body)
to sites of utilization (e.g., flight
muscle). However, the mechanisms of
loading and unloading such lipids by
lipophorin is still unknown. In order to
investigate the mechanisms, we prepared
monoclonal antibodies screened by western-
blot i m m u n o s t a i n i n g method. The
preparations obtained were incubated with
lipophorin and fat body in the presence of
adipokinetic harmone to test if the
antibodies inhibit the loading of DG from
fat body by lipophorin. The antibody
specific for Apo-II considerably inhibited
DG uptake by lipophorin. These results
suggest that Apo-II is , at least in part,
localized on the surface of lipophorin
particle and may contribute to DG uptake by
lipophorin from fat body.

BI 76

EFFECT OF ADIPOKINETIC HORMONE ON PROPER¬
TIES OF LIPOPHORIN OF THE SILKWORM, BOMBYX
MORI .

K. Miura and I. Shimizu. Lab. for Plant
Ecological Studies, Fac. of Sci., Kyoto
Univ. , Kyoto.

Adult hemolymph of the silkworm, Bombyx
mori contained much low density lipophorin
(LDLp) , which was rich in a third apopro¬
tein, apolipophorin III (apo-III) , in ad¬
dition to high density lipophorin (HDLp) .
Adipokinetic hormone (AKH) injection into
adult moths resulted in an increase of
LDLp content with concomitant depletion of
HDLp and free apo-III. By AKH injection
diacy lglycerol content in hemolymph in¬
creased by about two and a half times while
protein content did not change. Diacyl-
glycerol content of LDLp from injected
moths was about three times that of HDLp
from control moths. These results suggest
that AKH injection induced an accelerated
loading of diacylglycerol from fat body by
HDLp and that HDLp was transformed to LDLp
by accepting additional diacylglycerol and
apo-III. SDS-PAGE analysis revealed that
LDLp from injected moths was associated
with an appreciable amount of a protein
with a mol. wt of 18 Kd (18K protein) in
addition to much apo-III, suggesting the
involvement of 18K protein in AKH-induced
formation of LDLp. AKH-in jection into
larvae and pupae caused neither a forma¬
tion nor a density shift of pre-existing
HDLp.

BI 77

INSECT APOLIPOPHORIN III

C. Katagiri1 , M. Sato2, Y. Katsube2 and H.
Chino 1

^nst. of Low Temp. Sci., Hokkaido Univ.,
Sapporo, and 2Inst. for Protein Res.,
Osaka Univ., Suita.

Adipokinetic hormone secreted from the
corpus cardiacum is known to stimulate
lipid transport from fat body to muscle.
When adipokinetic hormone is released,
lipophorin expands in size. In the
activated lipophorin, diacylglycerol is
found to be three times more. Further, a
third apolipoprotein class, apolipophorin-
III (apoLp-III), is observed to be
associated. When we subjected apoLp-III to
a small-angle X-ray scattering study, the
scattering profile of apoLp-III exhibited
prolate ellipsoidal structure with a
homogeneous internal electron density.
Simulation studies using distance
distribution function, P(r), proposed the
possible structure of apoLp-III; a prolate
ellipsoid with a=b=18A and c=40A. Circular
dichroic measurements at various
temperatures showed that apoLp-III returns
to the native ot-helix-rich structure after
demonstrating a random coil structure in
boiling water. Such structural
reversibility of apoLp-III may contribute
to the expansion of the lipophorin
particle and to the acceleration of the
loading and unloading of diacylglycerol by
lipophorin .

lo

o

o

lo

'5b

o

T3

<u

c/a

a>

CO

o

'>

"O

o

T3

o

N

co

G

G

X

>>

T3

<D

X

U

c/a

cO

S-H

<D

C

<D

D-

u

D,

S

co

D-

O

<U

X

C

CT\

<4-1

V5

0)

o

CO

£

00

CO

lo

’d-

<L>

a

Os

O

•v

<u

G

00

CO

u

6

CL,

,1)

<4-4

0)

Z

G

O

00

«o

M

c

o

u

>

C/3

G

o

-o

O

G

1

u

c

CO

X

G

<D

CO

E5

X

H

’Q

on

G

O

10 Morisawa, S. and Morisawa, M. (1988) Induction of
potential for sperm motility by bicarbonate and pH
in rainbow trout and chum salmon. J. Exp. Biol.,
136: 13-22.

Biochemistry

1245

BI 78

SPECIFIC BINDING OF CRUSTACEAN EYESTALK
HORMONE TO HEART MEMBRANE RECEPTOR AND
ELEVATION OF cGMP LEVEL

T.Ezure, M.Ishida, K.Tozawa and T.Ohoka
Dept, of Biol., Fac. of Sci., Tokyo Metro-
pol. Univ. , Tokyo.

Sinus gland extracts of some crustacean
species were tested for the activity to
elevate the cellular cGMP level in an in
vitro system. In Procambarus and Sesarma
heart, sinus gland extracts of the same
species were effective to elevate cGMP
level, but extracts from other species
showed only a slight or almost no effect.

It was suggested the presence of species
specific receptor which act to stimulate
guanylate cyclase activity.

To elucidate the specific binding mecha¬
nism of membrane bound receptor, hypergly¬
cemic hormone of Procambarus was purified
by gel filtration on Asahipack GS 320H and
yBondasphere phenyl column. Active fraction
was subjected to iodination, ^^I-labeled
hormone was prepared. Binding activity was
estimated according to the method of Bruns ,
using GF/C filter impregnated in polyethyl-
eneimine solution. Specific binding of hor¬
mone to receptor of heart membrane was
demonstrated, and dissociation constant of
hormone and receptor was estimated to be
about 0.3 yM.

Solubilization of receptor from membrane
fraction was unsuccessful.

BI 80

PROTEIN KINASES
LIVER NUCLEI.

K. Asami. Div.
Radiological Sci

IN THE REGENERATING RAT

of Biol., Natl. Inst, of
, Chiba-shi.

Phosphorylation of histone Hi seems
to be prerequisite to DNA synthesis in the
regenerating liver, since it occurs at the
time of DNA synthesis and since X
irradiation prior to partial hepatectomy
inhibits both DNA synthesis and the
phosphorylation. We reported the presence
of the nuclear protein kinase (s), the
activity of which increased at the time of
the phosphorylation (24 h after
hepatectomy) and was partially sensitive
to X irradiation. The radiation-sensitive
enzyme was not activated with cAMP and did
not require Ca ions. In the present
studies, further characterization of the
nuclear enzymes was intended. First, the
activities against different substrates
were compared. The histone Hi kinase
activity was very low in the irradiated
nuclei, while the casein kinase activity
was not affected with X irradiation.
Secondly, the enzymes were fractionated
with FPLC using Superose 12 column. The
nuclear extract from the regenerating
liver contained a histone HI kinase and
two casein kinases, NI and Nil. Histone HI
kinase was not observed in normal liver
nuclei nor in the irradiated liver nuclei.
Casein kinases, NI and Nil, were found in
both normal and regenerating liver nuclei
and were not influenced with X rays.

BI 79

ENDOGENOUS SUBSTRATE PROTEINS FOR CYCLIC
GMP-DEPENDENT PROTEIN KINASE IN INTESTINE
OF PROCAMBARUS CLARK I I

M.Shibata, and T.Ohoka. Dept, of Biol., Fac.
of Sci., Tokyo Metropol. Univ., Tokyo.

Crustacean hyperglycemic hormone (CHH)
was known to elevate the guanylate cyclase
activity, as well as cGMP level in several
tissues of crayfish.

To elucidate the mechanism of action of
CHH, cGMP-dependent phosphorylation was
invastigated in Procambarus intestine.

Several phosphory lated proteins were
detected by SDS-PAGE and autoradiography,
after incubation of Lubrol-solubilized mem¬
brane fraction with [y-32p]ATP in the pres¬
ence of cGMP and cGMP-dependent protein
kinase, the latter was a purified prepara¬
tion from silkworm pupae. Non-solubilized
membrane fraction was hardly phosphory lated ,
but after solubilization, the phosphoryl¬
ation rate increased to 20-fold. The molec¬
ular weight of these substrate proteins
were estimated to be 57, 50, 46, 37 and 18
kilo daltons . Solubilized fraction was fur¬
ther fractionated by DE 52 chromatography
and gel filtration on Sephacryl S-300 HR
column.

18 K protein was found in break through
fraction of DE 52, 57 K and 50 K proteins
were effectively separated from 37 K, after
elution by 0.3 M NaCl and gel filtration.

Almost of these phosphorylatable proteins
were revealed to be minor component of the
tissue, detected by silver staining method.

BI 81

AP4A-BINDING PROTEIN IN NUCLEAR MATRIX OF
SEA URCHIN EMBRYOS.

M.MORIOKAl & H.SHIMADA2.

1 . Zool . Inst . , Fac . of Sci., Univ. of Tokyo,
Tokyo.

2 . Zool . Inst . , Fac . of Sci., Univ. of Hiroshima
Hiroshima.

We previously suggested that AP4A is
involved in the molecular mechanism regu¬
lating the extremely high rate of DNA
replication in the embryonic cells of sea
urchin through its binding to the nuclear
matrix. This hypothesis was based on our
findings that the intracellular level of
AP4A periodically decreased at the onset
of S phase in every cell cycle with the
concomitant increase in the nuclear AP4A-
binding activity. In this experiment, we
have attempted to purify the AP4A-binding
protein from the nuclear matrix prepared
from 8 cell-stage embryos.

We found that two distinct AP4A-binding
proteins, 23kDa and 70kDa, in the nuclear
matrix. The 23kDa component was easily
solubilized from the nuclear matrix by
gentle sonication in the buffer with low
ionic strength and partially purified
through (NH4)2S04 precipitation, DEAE-
cellulose column chromatography and Sepha-
dex G-100 gel filtration, successively.

The 70kDa component was too tightly
associated with the nuclear matrix to be
released by gentle sonication.

1246

Biochemistry

BI 82

EXPRESSION OF MURINE ENDOGENOUS RETROVIRAL
ENVELOPE GLYCOPROTEIN IN SEMINAL VESICLE.
N.Watanabe, H . Amanuma2 and Y . Ikawa1 .
^ab. of Molecular Oncology and 2Lab. of
Gene Technology and Safety, Tsukuba Life
Science Center, RIKEN, Tsukuba.

We found an expression of endogenous

retroviral envelope glycoprotein in
seminal vesicle and epididymis of DBA/2,
NZB and NZW mice. The glycoprotein is
secreted into seminal fluid. In Western
blotting analysis using an antibody again¬
st the murine leukemia viral envelope
glycoprotein, several bands with molecular
weight of about 68,000 dalton were detec¬
ted in SDS lysate of seminal vesicle and 2
bands with rather smaller molecular weight
were detected in that of epididymis.
Smaller amount of the similar glycoprotein
was detected in several other inbred and
wild mice. The expression of the glyco¬
protein was found not to be regulated by
hormones. The expression phenotype
inherits to their offsprings in incomplete
recessive way. In RNA blot analysis using
type specific oligonucleotide probe, the
envelope glycoprotein was found to be
xenotropic virus-origin and smaller
amounts of full-size genome was detected
in both seminal vesicle and epididymis.
In an SI nuclease protection assay, the
env region of cloned NZB virus genome was
homologous to RNA expressed in DBA/2
seminal vesicle. On the other hand, LTR
and pol region sightly differed from this
RNA species.

A possible role of such an expression
of xenotropic envelope glycoprotein is
discussed in terms of its protecting func¬
tion against retroviral infection.

BI 83

RECOMBINATIONAL PROTEINS; AI-REC AND
DNA-LIGASE , DURING MEIOTIC PROPHASE OF
SPERMATOCYTES.

A . Higashitani , S.Tabata and Y.Hotta.

Dept. of Biol., Fac. of Sci., Univ . of
Nagoya, Aichi.

During the periods of the homologous
pairing and chiasma formation, appearance
of the several proteins that may involve
in the DNA recombination has been
reported. In this presentation, we will
report the presence of two proteins during
meiotic prophase in mouse spermatocytes;
ATP-independent rec protein (Ai-rec) and
DNA ligase. Ai-rec catalyzed "D-loop
formation", which is believed to be the
first step of DNA recombination without
exogenous ATP, while E . coli RecA protein
and m-rec protein reported earlier (Hotta
et al. Chromosoma 93:140-151,1985) require
ATP. The activity of Ai-rec was
significantly suppressed by addition of
anti-E . coli RecA antibodies. In parallel,
DNA ligase, which should function at the
last step of DNA recombination, having a
molecular weight of 67,000 daltons was
partially purified from spermatocytes.
These two proteins were abundant in early
prophase of meiosis and may play
significant roles in meiotic chromosomal
recombination .

BI 8^

DETECTION OF MELANOCYTE SPECIFIC GENES IN
VERTEBRATES BY N O N - R A D I 0 A C T I V E
HYBRIDIZATION METHOD.

H.Kondoh-^ , K.Ito"*, Y.Koshida-^ , T. Merita ^ ,
T.Takeuchi2, H. Yamamoto2 and S.
Shibahara3, 1Dept. of Biol., Coll, of Gen.
Educ. , Osaka Univ., Toyonaka , 2Biol.
Inst., Tohoku Univ., Sendai, 3Fac. of
Med., Tohoku Univ., Sendai.

Enzymatic detection systems of non¬
radioactive DNA probe for hybridization
have several advantages. The systems
require no special safety precaution, can
be stored for a long time, and have rapid
detection procedure. The biotin-labeled
probes for genomic Southern blots were
prepared by nick translation, random
priming, and Photobiotin method. Both the
types of membranes for blotting and the
labeling conditions with Photobiotin were
examined. The DNA probes biotinylated by
these methods were very highly sensitive
to detect the unique sequences in genomic
DNA. In order to detect the target
sequences in the genomes of M.musculus ,
G . callus , X . laevis , and C . carassius ,
Photobiotin-labeled Tyrs-33 probe and pMT4
probe were hybridized to detect the
sequences homologous to the mouse
tyrosinase cDNA , and mouse b-locus related
cDNA , respectively. With the Tyrs-33
probe, some specific bands were observed
in these animal genomes, but with pMT-4
probe, specific bands might be seen only
in mouse genome.

BI 85

SEARCH FOR URICASE-GENE IN CHICKEN EMBRYO
H.Ogawa1, M. If.o2 , M.Nakamura2 and Y.Takagi 2
Dept . of Biol., 2Biv. of Mol. Genetics,
Fujita-Gakuen Health University, Aichi.

We tried to isolate cDNAs for uricase in
various species and compare their
structures in order to understand . the
inactivation mechanism of the genes
encoding the enzymes responsible for purine
catabolism. A cDNA clone for rat uricase
was isolated from a rat liver cDNA library
in Agtll, using an antibody against the
enzyme. The nucleotide sequence of the cDNA
encoded a polypeptide of 280 amino acids
with a molecular mass of 32,226 Da. Nine
highly homologous regions were found
between the predicted amino acid sequences
of rat uricase and soybean nodulin
uricase. Next, we tried to detect uricase
in the homogenates at various embryonic
stages of chicken, using the antibody
against rat uricase, although uricase
activity is not found in adult chicken. A
35 kDa-protein cross-reacting with -the
antibody was present in the homogenates
prepared from 4-12 days after the
fertilization. Then, we tried to isolate
cDNA for- uricase from 10 days-chicken
embryonic cDNA library, using rat uricase-
cDNA as a probe. One positive clone was
isolated and contained a region highly
homologous with the conservative sequences
of rat uricase-cDNA. These results suggest
that uricase could be expressed at some
embryonic stages of chicken.

Biochemistry

1247

BI 86

ANALYSIS OF PHENOTYPIC EXPRESSION OF THE
WHITE LOCUS IN MALPIGHIAN TUBULES OF DRO¬
SOPHILA MELANOGASTER.

S . Yagi-1- and H.Ogawa^. 1Dept.of Biol., Coll.
G. Educ.,Univ. of Tokushima, Tokushima and
2Dept.of Biol.,Sch.of Hygiene ,Fuj ita-Gakuen
Health Univ. , Aichi.

In order to probe the w locus function,
contents of 3-hydroxykynurenine (3-OHK) of
apricot (wa) and eosin (we) in larval Malpi¬
ghian tubules were compared with those of a
wild type , Oregon R, seven mutants , brown (bw)

, cardinal brown (cd bw) , vermilion brown (v
bw) , cinnabar brown (cn bw) , scarlet brown (st
bw) , lightoid brown ( ltd bw) and w by HPLC.

The large and weak fluorescent granules
are abundant in Malpighian tubules of Oreg¬
on R,bw, cd bw and 3-OHK actively transferr¬
ed into and accumulated in tubules , while w
and w^ at w locus, as v bw, cn bw,st bw and
ltd bw, result no accumultion of 3-OHK. On
the other hand, w£ lead to a little level
for accumulation of the compound. The small
and strong fluorescent granules are observ¬
ed tubules of v bw and cn bw. Larvae of v
bw and cn bw were given 3-OHK. 22h later,
3-OHK accumulation in Malpighian tubules
exceeds that of Oregon R. However, no acc¬
umulation of 3-OHK in tubules after feeding
the compound is detected in st bw, ltd bw, w
and w£ larvae, which are absent the granule-
s. In wf;, the small and weak fluorescent
granules contain in Malpighian tubules. 3-
OHK contents in tubules after administrat¬
ion of the compound was increased twice
this amount and reached 10 % of Oregon R.

BI 87

STRUCTURE AND EXPRESSION OF RAT RENIN GENE
M. Tada, A. Fukamizu, M.-S. Seo, S. Takahashi and K.
Murakami

Institute of Applied Biochemistry, Gene Experiment Center,
University of Tsukuba, Ibaraki, 305 JAPAN

By use of the mouse cDNA as a probe to screen a
Sprague-Dawley rat kidney cDNA library and a
genomic library, the structure of the rat renin
precursor as well as its genomic organization
have been determined. The mRNA transcript
corresponding to the rat renin cDNA clone was 1.6
kilobase in length and was expressed by 15 times
after the chronic administration of both sodium
depletion and captopril. Nucleotide sequence
analysis of a full-length cDNA clone shows it is
1434 nucleotides, encoding a 402-amino acid
precursor, which is 85%, 82% and 68% identical to
the mouse Ren-1 and Ren-2, and human renins,
respectively. The single rat renin gene is
approximately 11 kilobases long, consisting of
nine exons and eight introns. The 5'-flanking
region of the gene encompasses several potential
regulatory elements, including 2 TATA boxes and
estrogen, glucocorticoid, and cAMP-responsive
elements. Sensitive RNase protection analysis
indicated that a small amount of the renin mRNA
was expressed in the extra-renal tissues such as
adrenal, brain, liver, lung and testis.

BI 88

CHANGE IN OSMOTIC PRESSURE OF THE
HAEMOLYMPH DURING DEVELOPMENT IN THE
SILKWORM, Bombyx mori . : CHANGE IN CATION

CONCENTRATION

S. NAKAYAMA. Tezukayama Coll., Nara .

Changes in osmotic pressure and sodium,
potassium and amino acid concentratons in
haemolymph were studied during development
in the silkworm, Bombyx mori.

Osmotic pressure increased from day 0
in the fifth instar to day 6 in the fifth
instar, then declined and continued to
decrease until the spinning stage. During
the larval-pupal ecdysis, osmotic pressure
increased somewhat. Measurement of amino
acid revealed that the change in the amino
acid content during development was
correlated to osmotic pressure.

Potassium and sodium contents were
determined with an ion chromatography.
Their concentrations began to decline at
day 4 in the fifth instar. On the other
hand, osmotic pressure did at day 6. These
results suggested that decrease of
potassium and sodium contents is possibly
prerequisite for decrease of osmotic
pressure during fifth instar development
of Bombyx mori .

BI 89

GANGLIOSIDES IN RAT PRIMARY CULTURE NEURONS.
M. Ogisol ,M. Ohtal , S . Hirano^ , and K.Itoh2.
1-Dept. of Physiol., Toho Univ. Sch. of Med.,
Tokyo, 2Gunma Univ. Sch. of Med., Maebashi.

The physiological roles of gangliosides
during neural process formation were exa¬
mined in rat primary culture neurons. Addi¬
tion of colchicine to the medium caused
inhibition of neural process formation and
reduction of protein content. However,
ganglioside content did not decrease much,
suggesting that ganglioside synthesis
occurs at a normal level during inhibited
process formation. During the maturation of
primary culture neurons, changes in gangli¬
oside composition were observed. Expression
of tetrasialoganglioside GQlb was supressed
by co-culture of neurons with confluent
astroglial cells. GQlb is known as a candi¬
date for the promotion of neural process
formation. On the other hand, GQlb content
increased in parallel with the maturation
of neurons in mixed primary culture. The
difference suggests cell-to-cell inter¬
action between neurons and glial cells at
the ganglioside level. The addition of a
phorbol ester, PMA, to mixed culture
neurons caused morphological changes in
neuronal network formation. Significant
differences were observed in subspecies of
monosialoganglioside GMl . Thus, ganglio¬
sides may have a physiological function in
neural development.

1248

Biochemistry, Developmental Biology

BI 90

LIPID COMPOSITION IN THE TADPOLE TAIL OF
RANA CATESBEIANA DURING METAMORPHOSIS.

M. Ryuzaki. Dept, of Biol., Kitasato Univ.
School of Medicine, Sagamihara. _

The lipid composition of simple lipids (S’

L) and phospholipids (PL) obtained from the
larval tail in Rana catesbeiana at Taylor
and Kollros (TK) stage V, X, XX, XXI and
XXII-XXIII were analyzed in order to clarify
the relationship between the regressive
process of the tail and these lipid composi¬
tions during metamorphosis in this frog. In¬
dividual SL and PL were separated by thin-
layer chromatography on layers of Silica Gel
HR. The weight percentage (wt%) of total SL
to total lipid (TL) was ca. 33% at TK stage
V and X and increased gradually until ca. 70
% at TK stage XXII-XXIII; that of total PL
to TL was ca. 66% at TK stage V and X and
decreased gradually until ca. 28% at TK
stage XXII-XXIII. The wt% of free fatty acid
(FFA) and triglyceride (TG) to TL were ca.
12-15% and 4% at TK stage V and X and in¬
creased gradually until ca. 38% and 13% at
TK stage XXII-XXIII, respectively. That of
the other SL; i.e. cholesterol and choles¬
terol ester did not significantly change
during metamorphosis. The wt% of PL classes;
i.e. phosphatidylethanolamine , phosphatidyl¬
choline, phosphatidylserine , phosphatidyl-
inositol, sphingomyelin and lysgphosphati-
dylcholine to total PL also did not signifi¬
cantly change during metamorphosis. The ac¬
companying increase in the wt% of total SL
to TL was due primarily to the increase of
FFA and TG during metamorphosis.

BI 91

INTERRELATIONS BETWEEN MEMBRANE DAMAGE AND
CELLULAR RECOVERY IN CELLS IRRADIATED BY
RADIATIONS.

M.Kato, S.Yonei and H.Tomiyama. Dept, of
Zool., Fac. of Sci., Kyoto Univ., Kyoto.

Although membranes are considered as one
of the sites of radiation damage in cells,
the molecular mechanism of membrane damage
and its role to cell death caused by radi¬
ations remain unsolved. We observed, with
the aid of fluorescent probes, that a low
dose of X-rays readily induces significant
changes in the structure of erythrocyte
membranes, i.e. the change in membrane
fluidity. On the other hand, we found the
modification of radiation effects on cell
survival of E.coli by specific membrane¬
binding agents such as procaine. These
agents sensitize the radiation lethality
in DNA-repair proficient cells, but not in
repair deficient cells. Further studies in
E.coli irradiated by UV showed that the
removal of thymine dimers from DNA was re¬
duced by the addition of procaine during
post-irradiation incubation. When the ex¬
tent of liquid holding recovery was mesur-
ed as a function of the temperature at the
time of LHR and the Arrhenius plot was
made, two distinctive phases for the LHR
were denonstrated in UV-irradiated RecA
derivative of E.coli ole28Ei .

It is therefore concluded that some com¬
ponents involved in excision repair of DNA
is associated with cell membrane.

BI 92

BIOTRANSFORMATION OF METHYLMERCURY ( MeHg ) i
IN VITRO IN THE TISSUES OF WILD AND LABORA¬
TORY ANIMALS.

S.Omata1 , T. Y. Toribara2 , E . Cernichiari2
and T. W. Clarkson2 . 1 Dept . Biochem. , Fac.

Sci., Niigata Univ., Niigata and 2Envir.
Hlth. Sci. Ctr. , Rochester Univ., Rochester
U.S.A. _

In order to assess MeHg-demethylating act¬
ivity, finely chopped tissues were incubat¬
ed with (203)Hg-MeHg and a buffered salt
solution at 30°C for 24 hr. and then ex¬
tracted with acidified benzene. The MeHg-
demethylating activity was highest in the
liver of penguins, moderate in a skua and
harbor seals, and lowest in SD rats. The
activity did not change in the presence of
penicillin, GSH or Na-selenite ( up tp
200/1 of Se/Hg). MeHg accumulation in the
liver of the seals was about 20-fold that
in the penguin liver. GSH S-transf erase
activity in the liver of the rats was high¬
est, those in the penguins and seals being
only 6 % and 23 % of that in rat liver, re¬
spectively. The Se-dependent GSH peroxidase
activities in the liver of the wild animals
were similar, while the Se-independent en¬
zyme activities in the liver of the birds
were very low. Artif actitious demethyla-
tion was detected under some experimental
conditions. We are very grateful to mem¬
bers of the Argentine Antarctic Inst., and
Prof. T. Suzuki, Tokyo Univ. Sch. Med., for
providing the animal tissues.

DB 1

OBSERVATIONS OF EGGS AND SPERM DURING EARLY
EVENTS OF FERTILIZATION IN THE FISH (Oryzias).
T. Iwamatsu^ and K. Onitake2, ^Dept. Biol.,
Aichi Univ. of Educ., Kariya 448 and 2Dept. Biol. ,
Fac. of Sci., Yamagata Univ., Yamagata 990

Preliminary experiments showed that electron
microscopy might not Be practical investigate
the aeries of rapid reactions of the medaka
egg following sperm attachment, because of
the difficulty in fixing eggs within 5 sec.
Therefore, fertilizing egg and sperm were
observed in a live state. The mean time re¬
quired for sperm to stop moving after attach¬
ment to the egg surface was 4 sec. This in¬
dicates that membrane fusion between both
gametes may not occur within the first sec
after sperm attachment, when sperm head con¬
tinues to rotate. Stop of sperm movement was
quickly (mean 5 sec) followed by exocytosis
of small cortical alveoli at the cytoplasm
facing a micropyle. When 10 juM inositol tris-
phosphate was microin jected into an unferti¬
lized egg, the first exocytosis occurred after
a mean time of 5 sec. This time to first exo¬
cytosis was not significantly different from
that after stop of sperm movement. Measure¬
ment of the time sequence of exocytosis at
various points distant from the micropyle in¬
dicated that the mean propagating velocity
of the exocytosis wave increased, as the dis¬
tance from the micropyle increased up to 150
jam . These results indicate that the time lag
between sperm attachment and exocytosis may
reflect the metabolic chain-reactions in¬
volving the increase in cytoplasmic Ca2 + in
the cortical cytoplasm.

Developmental Biology

1249

DB 2

ELECTRICAL RESPONSE OF THE MEDAKA EGG
MICROINJECTED WITH EGTA , BY INSEMINATION.
S.Ito and K.Shimamoto. Dept. of Biol., Fac.
of Sci. , Kumamoto Univ. , Kumamoto.

An increase of intracellular free cal¬
cium ions is thought to be a principal
ionic signal responsible for activation of
the medaka egg, as in other species. Upon
fertilization, the medaka egg generates a
two step-depolarization followed by a slow
hyperpolarization. Increasing extracellular
Ca2+ enhanced the magnitude of the second
depolarization (Ito and Shimamoto, 1987).

We suppressed the rise in calcium by micro¬
injection of EGTA ( 1 nl : 20mM dissolved in 10
mM Hepes-KOH , pH 7.0) into the egg cortex
before insemination and observed the effect
on fertilization potential and cortical
alveoli breakdown. Injection of EGTA into
the egg immersed in 1 8mM CaCl2 dissolved in
158mM sucrose generated only a few of tran¬
sient step-like depolarization at any time
of insemination , but never induced a spike¬
like depolarization and breakdown of corti¬
cal alveoli. These transient depolarizat¬
ions seem to depend on sperm-egg interact¬
ion, since they were not observed when the.
egg was activated by current injection or
A23187. Examination by light microscopy
revealed that condensed sperm nuclei were
embeded in the egg cortex near the micro-
pyle, suggesting that EGTA- in jected egg was
polyspermic under the unactivated condition.

DB 3

CORTICAL ALVEOLI BREAKDOWN IN THE EGGS OF
THE ROSE BITTERLING.

T.Ohta, T.Iwamatsu and Y . Yoshimoto* . Dept,
of Biol. Aichi Univ. Educ . , Kariya. *Dept.
of Lib. Arts,- Kansai Med. Univ., Maikata.

Cortical changes in the eggs of the rose
bitterling, Rhodeus ocellatus ocellatus ,
were investigated . In a physiological sa¬
line solution (PSS) , the eggs showed a
slight breakdown of cortical alveoli (CABD)
which began at the vegetal pole and pro¬
gressed slowly. On the other hand, CABD be¬
gan quickly in freshwater or in an isotonic
sucrose solution, in which CABD occurred
from the vegetal pole to animal pole. Sper¬
matozoa did not move at all in the PSS.
Therefore, the eggs were not fertilized in
the PSS. When inseminated eggs in the PSS
were transferred into freshwater, rapid
CABD occurred at the animal pole which
took about 20 min to complete. During CABD,
an increase of free Ca++ions was ascertained
by the luminescence of injected aequorin.
Some layers of cortical alveoli existed in
the cortical cytoplasm of inactivated eggs.
Many intramembranous particles were observ¬
ed on the PF surface of cortical alveoli,
while they were very scarce on the EF sur¬
face. The surface of the eggs was covered
with microvilli and microplicae. With the
advance of CABD, the components of corti¬
cal alveoli were released through the wide
openings and the egg surface became smooth.

DB 9

THE MAJOR CORTICAL ALVEOLUS GLYCOPROTEIN
(HYOSOPHORIN) FROM THE EGGS OF MEDAKA,
ORYZIAS LATIPES.

K. Ki ta j ima 1 , S. Inoue^, and Y. Inoue1.
*Dept. of Biophys. and Biochem., Fac. of
Sci., Univ. of Tokyo, Tokyo, ^School of
Pharmaceutical Sci., Showa Univ., Tokyo.

"Hyosophor in" is a major cortical
alveolus glycoprotein which we first
isolated from rainbow trout eggs. Now we
have isolated from medaka eggs, a novel
type of s i a 1 o g 1 y c o p r o t e i n s which possess
the following properties typical to
"hyosophor in".

(1) High molecular weight (Mf =-'100K)
s i a 1 o g 1 y c o p r o t e i n s (H-SGP) and a low
molecular weight (Mr = 7K) sialoglyco-
peptide (L-sgp) isolated from the un¬
fertilized and the fertilized medaka eggs,
respectively, have identical carbohydrate
and amino acid compositions.

(2) Carbohydrate contents of H-SGP and
L-sgp are as high as 90% (w/w).

(3) L-sgp of which amino acid sequence
was determined as As p -A 1 a -A 1 a - Se r -As n* -
G 1 n - Th r - Va 1 - S e r (* the g 1 y c o s y 1 a t i o n
site), was shown to be the least structur¬
al unit of H-SGPs.

Finally, we have succeeded in localiz¬
ing SGPs in isolated cortical alveoli by
the immunof 1 uorescense staining technique.

DB 5

BACTERICIDAL ACTION OF FERTILIZATION ENVE¬
LOPE EXTRACT FROM EGGS OF THE FISH, ONCO-
RHYNCHUS MASOU AND SALVELINUS LEUCOMAENIS

S. Kudo and M. Inoue. Dept, of Anat . and
Lab. of Drug Resist, in Bact., Gunma Univ.
Sch. of Med., Maebashi.

The vitelline envelope (VE) and fertili¬
zation envelope (FE) in eggs of the fish,
Oncorhynchus masou and Salvelinus leucomae-
nis were purified by homogenization in 5 mM
Tris-HCl buffer (pH 7.0) containing 2 mM
EDTA and by repeated washing with the same
buffer. In order to extract the outermost
layer material, the purified VEs and FEs
were processed overnight at 4°C in 5 mM
Tris-HCl buffer (pH 7.0) containing 8 mM
2-mercaptoethanol , 2 mM EDTA, 0.3 M a-lac-
tose, 0.3 M glucose and 0.9% NaCl. The
solution extracted from purified VEs or FEs
was dialyzed against 5 mM Tris-HCl buffer
(pH 7.0), followed by centrifugation, and
the clear supernatant was lyophilized.

These extraxts from the FEs of both fish
species had a strong bactericidal action on
Vibrio anguillarum, but not on Aeromonas
hydrophi la and Escherichia coli, and were
shown to contain two kinds of lectins, one
of which agglutinated human B-type erythro¬
cytes and the other several kinds of fish
spermatozoa. The extracts of purified VEs
from eggs of both fish had no bactericidal
action on the bacteria examined, nor any
agglutination effect on human erythrocytes
and fish spermatozoa. The bactericidal
action was du'e to bacteriolysis.

1250

Developmental Biology

DB 6

PERMANENT BLOCKAGE OF POLYSPERMY IN THE
FLOUNDER, PARALICHTHYS OLIVACEUS, AND
ALASKA POLLACK, THERAGRA CHALCOGRAMMA ,
EGGS.

T.S. Yamamoto and W. Kobayashi. Zool.
Inst., Fac. of Sci., Hokkaido Univ.
Sapporo .

Fine structure of the micropyle was
studied in unfertilized and developing
flounder and Alaska pollack eggs. In
both species, the egg envelope of
unfertilized eggs consisted of concentric
lamellae and showed a wavy inner surface.
In the developing eggs, it decreased in
thickness and showed a relatively smooth
inner surface. These observations
indicate that the egg envelope may be
compressed centrifugally after formation
of perivitelline space.

The micropyle in unfertilized eggs
consisted of a shallow vestibule and a
narrow canal. As evident from SEM
observations, however, the micropylar
canal in developing eggs was closed.
Sections through the micropyle suggested
that the closure is due to the swelling
of the canal wall as well as the
compression of the egg envelope. The
closure of micropylar canal prevents
excess sperm from penetrating into the
perivitelline space and is involved in
the permanent blockage of polyspermy.

DB 7

NECESSITY OF BREAKDOWN OF SPERM NUCLEAR
ENVELOPE FOR MALE PRONUCLEUS FORMATION.
M. Yamashita1 ' 2 , H. Onozato2,

T. Nakanishi2 and Y. Nagahama1.

1Lab. of Reprod. Biol., Natl. Inst, for
Basic Biol., Okazaki, 2Natl. Res. Inst,
of Aquaculture, Inland Station, Mie.

A male pronucleus is not formed in the
eggs of gynogenetically developing crucian
carp, ginbuna, Carassius auratus langs-
dorf ii . To investigate this, we observed
ultrastructural changes in the incor¬
porated sperm nuclei, which revealed that
the sperm nuclear envelope was not broken
down during the fertilization process.

When nuclear envelope-free spermatozoa
were microinj ected into the activated
eggs, they metamorphosed into male pronu¬
clei in the ginbuna eggs. The resultant
male pronuclei synthesized DNA. The mem¬
brane-free spermatozoa were also trans¬
formed into male pronuclei in a cell-free
system prepared from the activated ginbuna
eggs, and the resultant male pronuclei
condensed into chromosomes when matura¬
tion-promoting factor extracted from the
mature ginbuna eggs was added to the sys¬
tem.

These results indicate that the failure
of the male pronucleus to form in ginbuna
is due to the persistence of the sperm
nuclear envelope, demonstrating the neces¬
sity of sperm nuclear envelope breakdown
for complete formation of the male pronu¬
cleus .

DB 8

ROLE OF SPERM GLYCOSIDASES IN THE ASCIDIAN,
HALOCYNTHIA RORETZI

M. Saito*, M. Mochii**, H. Sawada*, M. Hoshi*.

*Dept. of Life Sci., Fac. of Sci., Tokyo Inst, of
Technol., Tokyo, **Dept. of Dev. Biol., Natl.

Inst, of Basic Biol. , Okazaki.

Sperm glycosidases are suggested to be the vitel¬
line coat binding protein in ascidians. For
further study of the role of sperm glycosidases in
fertilization, H alocynthia roretzi, one of the most
suitable ascidians for the biochemical analyses,
was used. In the sperm of this animal, (3-D-N-
acetylhexosaminidase (3-D-HexNAc'ase) activity was
the most potent and a-L-fucosidase (a-L-Fuc'ase)
activity was the second. B-D-HexNAc'ase was puri¬
fied and enzymologically characterized. p-Nitro-

phenyl-B-D-N-acetylgalactosaminide, a substrate of
B-D-HexNAc'ase, did not block sperm binding to the
vitelline coat, but it did activate the sperm mo¬
tility. p-Nitrophenyl-a-L-fucoside, a substrate of
a-L-Fuc'ase, inhibited the sperm binding and thus
fertilization, but it did not affect sperm motili¬
ty. Since major N-glycosides of, the vitelline coat
of H. roretzi have terminal 3-D-N-acetylgalactos-
amine and a-L-fucose (Ichikawa et al., unpublished
data), they may serve as the sperm receptor as well
as an activator of the sperm bound to the vitelline
coat.

DB 9

VITELLINE COAT LYSIN IN SPERM EXTRACT OF
THE ASCIDIAN, HALOCYNTHIA RORETZI.

H. Sawada, N. Menrai, S. Takizawa, and
M. Hoshi. Dept, of Life Sci., Fac. of
Sci., Tokyo Inst, of Technol., Tokyo.

We previously reported that at least
two sperm trypsin-like proteases, acrosin
and spermosin, and one chymotrypsin-like
protease (multi-functional protease) were
involved in sperm penetration through the
vitelline coat of the egg, H. roretzi.
Howe'ver, either of these isolated
proteases did not or only weakly show the
lysin activity toward the vitelline coat.
Then, we attempted to search the potent
lysin activity in sperm extract by using
125I-labeled vitelline coat as a
substrate .

In sperm extract, a lysin activity,
which could digest the major component
(60k) and also minor components (80k, 55k,
and 45k) of the vitelline coat, was
detected by SDS-PAGE followed by
autoradiography. The lysin activity
showed an optimum pH about 9.0, and was
inhibited by strong inhibitors toward
fertilization such as leupeptin, antipain,
chymostatin, and soybean trypsin inhibitor.
Furthermore, we also demonstrated that 60k
component was specifically localized on
the vitelline coat by histochemical
technique using anti-60k rabbit antiserum.
Thus we conclude that a vitelline coat
lysin, most probably a protease, exists in
sperm, and may participate in sperm
penetration of the vitelline coat.

Developmental Biology

1251

DB 10

BINDING SITE OF A SPERM ACROSOMAL PROTEIN
IN THE VITELLINE COAT OF ABALONE EGG.

N. Usuil and K. Haino-Fukushima2. iDept. of
Anat. , Teikyo Univ. Sch. of Med. and 2Dept.
of Biol., Fac. of Sci., Tokyo Metropolitan
Univ. , Tokyo. _

Two major proteins of molecular weights
15,500 and 20,000 from the acrosomal vesicle
of the abalone Ealiotis discus are responsible
for lysis of the egg vitelline coat (VC) .
Our previous study showed that 2 OK VC lysin
dissolved the outer electron-dense layer of
the VC, whereas 15. 5K lysin led to extreme
loosening of thick main part, a feltwork of
fine filaments. These VC lysins, at least
15. 5K lysin, seem to dissolve the VC by a
stoichiometric, non-enzymatic mechanism.

To examine the mechanism of lysin action,
binding of 15. 5K lysin to the VC main part
was studied by immunoelectron microscopy:
The VCs were isolated and dissolved halfway
by treatment with crude acrosomal contents .
After brief fixation, the specimens were
labeled with anti-15. 5K lysin antiserum
then protein A-gold complex. The antigenic
sites (gold labels) were found only on the
outer and inner surfaces of the VC possess¬
ing unloosened main part, but the labels
were densely distributed among attenuated
fine filamentous feltwork of dissolved VC
main part. This lysin seems to bind to and
react with fine filaments making up the VC
main part and remain associated with the
structure even after VC lysis. This morpho¬
logical evidence suggests a stoichiometric
lysin action of 15. 5K acrosomal protein.

DB 11

ACTION OF THE VITELLINE COAT LYSIN OF
TEGULA PFEIFFERI .

K. Haino-Fukushimal and N. Usui2. ^^Dept . of
Biol., Fac. of Sci., Tokyo Metropolitan
Univ. and 2Dept. of Anat., Teikyo Univ. Sch.
of Med., Tokyo. _ _

Action of the vitelline coat (VC) lysin

of a top shell, Tegula pfeif feri , has been
extremely investigated in biochemical
field. But, there was scarcely any morpho¬
logical observation of this action.

We tried an electron microscopical obser¬
vation on dissected eggs and isolated VCs of
T. pfeif feri before and after treatment of
VC lysin. The results showed that the VC was
composed of an homogeneous basal layer hav¬
ing a low electron density and conical
structures adhering to the layer at quali¬
fied intervals. Moreover, the conical
structure is formed of an asterisk structure
closed to the basal layer and a deposit of
granules piled on the structure. After
treatment of VC lysin, the basal layer
changes twofold in surface area and the dia¬
meter of granules forming a conical struc¬
ture increases from 100 nm to 150 nm, too.

Ultrastructural immunocytochemical local¬
izations of the reacted VC lysin and the
released product were revealed with anti-
VC lysin- and anti-soluble product antisera
using a protein A-gold technique. The
results showed that VC lysin bound to the
deposit of granules and soluble products
were released from the surface of basal
layer between conical structures.

DB 12

DETERMINATION OF THE STRUCTURE OF LYSIN
BINDING SITES ON VITELLINE COAT: THE TWO
DISTINCT PROTEIN CORES WHICH COMPOSED
LYSIN BINDING SITE.

M.Shitara & K .Haino-Fukushima . Dept. of Biol.,
Fac. of Sci., Tokyo Metropolitan Univ., Tokyo.

Vitelline coat lysin of marine Mollusca,
Tegula pfeif feri irreversibly binds to the
vitelline coac(VC) , bringing about the lysis
of VC. We previously reported that the lysin
binding sites on VC (LBS) required both N-
glycosidic bond and certain amino acid
sequence which was missing by the Pronase
digestion. Besides, LBS containing minimum
VC fragment was recovered when solubilized
VC was degraded in successin by V8 protease
digestion and TFMS deglycosylation .

In this study, we obtained two distinct
protein cores both composed complete LBS,
from slow hydrolysis of the VC fragment by
chymotrypsin and gel filtration with Sepha-
dex G-100 column equibriumed 4M urea. One of
these core was abandant in Thr , Ser and Hyp
residues and supposed to include N-glycos-
idic bond and the rest of sugar chain by
TFMS glycosylation. The other core was esti¬
mated much smaller than the former and rich
in hydrophobic amino acid residues which
were lost by the Pronase digestion. Neither
the former nor the latter recovered complete
LBS without each other. In addition, both
of them werefractionated by 6N HC1 acid
hydrolysis of 37°C and gel filtration with
Sepharose 4B column.

DB 13

OOGENESIS OF TETILLA JAPONICA
Y.Watanabe and H.Fukuda. Dept, of Biol.,
Fac. of Sci., Ochanomizu Univ., Tokyo.

On the surface of the ovulated eggs of T.

japonica, there radiate numerous fiber
bundles, which are composed of collagen
fibers. We examined the process and period
of collagen fiber formation in oogenesis.
Small pieces of tissue from female sponge
in reproductive period were fixed and
observed with light microscope and
electron microscope. There were two stages
of growth in oogenesis. The initial growth
involved the uptake of soluble material
and small fragments of nurse cells. During
the major period of growth, oocyte
actively phagocytized nurse cells directly
and accumulated yolk granules. Collagen
fibrogenesis began at the end of major
growth period. At this period, amoeboid
cells migrated and surrounded the oocytes.
With EM observation, these cells secreted
fibrils to the side of oocyte. Compared
with archeocyte which secrete mesohyl
collagen of TL_ j aponica , collagen
fibrogenesis of these cells were more
active. The collagen synthetic cells of
gemmule in fresh water sponges presented
the feature of collumnar epitherial cells.
In contrast to it, collagen fibrogenetic
cells of the fiber bundles were amoeboid,
and their morphology were similar to
lophocyte, which act as mesohyl collagen
fibrogenesis in Chondrosia renifermis ,
reported by Garrone.

1252

Developmental Biology

db m

ORIGIN AND FORMATION-PROCESS OF CORTICAL
GRANULE AND YOLK GRANULE IN OOCYTE OF
FRESHWATER PLANARIANS .

W. Teshirogi and T. Shirakawa. Dept, of
Biol., Fac. of Sci., Hirosaki Univ.,
Hirosaki

It has been pointed out by Gremigni
(1979) that the oocyte of species in
Dugesiidae does not have cortical gra¬
nule, but has yolk granule, while that
of species in Planariidae and Dendrocoe-
lidae have cortical granule, but do not
have yolk granule. This time , we have
reexamined the relation between cortical
granule and yolk granule using Japanese
freshwater planarians i.e. Polycelis
auriculata and Phagocata vivida (Plana¬
riidae), Bdellocephala brunnea (Dendro-
coelidae), and Dugesia japonica japo-
nica (Dugesiidae).

The results of the reexamination cer¬
tainly supported the view of Gremigni
as mentioned above. That is to say, the
cortical granule and yolk granule do not
exist together in the oocyte of the same
species of planarians. The cortical gra¬
nule is originated from Golgi complex.
Furthermore, it has been made clear that
glycogen granules participate in the
formation of the cortical granule in Bd.
brunnea. The yolk granule also is orTg-
inated from Golgi complex, and the grow¬
ing yolk granule takes in nuclear satel¬
lite materials and mitochondria.

DB 15

ULTRASTRUCTURAL OBSERVATIONS OF VITELLOGE¬
NESIS AND EGG-SHELL SYNTHESIS IN OOCYTES OF
THE TICK, HAEMAPH Y SAL I S LONGICORNIS.

Y. Yano, T. Mori, S. Shiraishi and T. A.
Uchida. Zool . Lab. Fac. of Agr . Kyushu
Univ . , Fukuoka _

Yolk granules first appeared in oocytes
with polarity (diam. ~ 100 pm) when the tick
detached from the host after engorgement. A
germinal vesicle with a vacuolated nucleo¬
lus was situated towards the funicle cells.
Secretion of substances forming the primary
egg membrane (egg-shell) into the extracel¬
lular space between well— developed micro¬
villi coincided with yolk deposition: the
substances seem protein judging from both a
positive reaction to the tannic acid fixa¬
tion and a negative test to the performic
acid— phosphotungstic acid staining. Active
pinocytotsis suggests that yolk precursors
are taken from the haemocoel . Multivesi—
cular bodies, free ribosomes, rough endo¬
plasmic reticulum, mitochondria, glycogen
and lipid droplets were pronounced through¬
out the cytoplasm. In advanced oocytes
(diam. ~ 300 pm) just before ovulation,
large polygonal yolk granules which were
formed by coalescence of small round ones
occupied most of the cytoplasm: the nucleus
was adjacent to the egg membrane and a wide
interchorionic space was formed between
them. Finally, the egg membrane became a
single— layered egg-shell ( a/3 pm in thick¬
ness) after the microvilli had been re¬
tracted .

DB 16

Time course of appearance of the SF sub¬
stance and vitellogein in the liver of the
estrogen-treated adult male medaka .

K. Murata, T.S. Hamazaki, I. Iuchi, and K.
Yamagami. Life Sci. Inst., Sophia Univ.,
Tokyo.

It has been reported that the spawning
female-specific ( SF ) substance (major com¬
ponent of the egg envelope) and vitello¬
genin ( Vg ) were produced in the liver of
the spawning female medaka by the ad¬
ministration of estrogen. In the present
study, the first appearance of these sub¬
stances in the liver of the estrogen-
treated male was examined using anti-FI an¬
tibody (anti -chorion glycoprotein antibody
) and anti-Vg antibody as probes. The
adult male fish were kept in the tap water
containing 100 ng/ml of estradiol-1 7/3 at
15°C. Both substances, SF substance and
Vg, were detected in the liver 8 to 12 hr
after the beginning of the estrogen treat¬
ment by the immunoelectrophoret ical and
immunohistochemical analyses. Moreover, de
novo synthesis of the SF substance in the
liver was examined by preliminary experi¬
ments of _in vitro incorporation of
[14C]amino acid mixture into the sliced
liver of the estrogen-treated female fish.
The hepatic cells were found to synthesize
and release the SF substance into the cul¬
ture medium. This is a direct evidence of
an intrahepatic synthesis of the SF sub¬
stance in the spawning female fish.

DB 17

LIGHT AND SCANNING ELECTRON MICROSCOPIC
STUDIES ON FERTILIZATION OF PELMATOHYDRA
ROBUSTA.

C.Kanai and K.Noda. Lab. of Electron Micro-
sopy, Tokyo Metropolitan Inst, of Gerontol.

Sperm enter the egg only through the de¬
pression of the egg of Pelmatohydra robusta
(Noda and Kanai,1981). The depression was
examined morphologically in unfertilized
and fertilized eggs. All of the specimens
were fixed in glutaraldehyde and postfixed
in osmic acid following by dehydration.
Materials embedded in Epon-812 were used
for histology. Cr it ica 1 -point dried and
non-coating specimens were examined under a
scanning electron microscope (Hitachi, S-
800) at an accelating voltage of 0.6 - IkV.

The entrance of the depression is usual¬
ly constricted and covered with a thin coat
in eggs fixed immediately after burst-out.
The egg-coat prevent sperm entering the de¬
pression. After burst-out, the entrance of
the depression gradually spreaded and the
egg-coat closing the entrance of the de¬
pression begin to be torn. Sperm can enter
the depression through the tear of the coat
within 1 5 min, when the insemination was
done immediately after burst-out. The en¬
trance was completely opened within 30 min
after burst-out. After fertilization, the
bottom of the depression rose and was accu¬
mulated by fibrous structures.

Two depressions were sometimes observed
in single egg. Sperm could enter such egg
through both the depressions and initiate
normal cleavage.

Developmental Biology

1253

DB 18

ELECTRON MICROSCOPIC OBSERVATION OF
MATURATION AND FERTILIZATION IN THE
BIVALVE, Ljternula limicol^. 11^

K.Hosokawa and Y . D . Noda^ . _ Biol. Lab.

Tokyo Dent. Coll. Chiba, ^Biol. Inst.
Fac. of Sci . Ehime Univ. _ _

The ovotestis of a bivalve, La ternula
limicola (monoecious) matures concurrently
with a spawning season ( Aug. - Sept.).
The shells of the bivalve were removed
with care and the material was placed in
80 % SW containing O.lM acetylcholine.

Fully matured oocytes (prophase stage of
the primary oocyte) and sperm were spawned
from the openings of the oviduct and
spermiduct that ran parallel with each
other. To observe the interaction between
the oocyte and sperm, light and electron
microscope were used. The sperm passed
through the investments and a micropyle —
like pore were reached the surface of the
oocyte and penetrated with a rush into the
oocyte. The ocyte was self-fertilized. As
the absence of a typical acrosoine
(Kubo,1977), the sperm in this species
lack the acrosorae reaction. Although the
oocyte was inseminated after fixation with
2.5 % glutar ardehyde (pH 7.4), numerous

sperm were able to reach the surface of
that oocyte. Upon fertilization, it seems
that the sperm take up a positive
behaviour towards the oocyte. In this
observation, gamete membrane fusion could
not be morphologically svbstaintiated .
But, we have confirned that the oocyte
carried out the self- fertilization.

DB 19

THE CYCLIC AMP-ADENYLATE CYCLASE SYSTEM
IN STARFISH FOLLICLE CELLS 9
1 M. Mita , N. beta1 and Y. Nagahaffia~
Dept. of Biochem., Teikyo Univ. Sch. of
Med., Tokyo, ^Lab. of Reprod. Biol., Natl.
Inst, for Basic Biol., Okazaki. _

Resumption of meiosis in starfish
oocytes is induced by 1 -me thy 1 aden i ne (1-
MeAde) produced by ovarian follicle cells
under the influence of a gonad-stimulating
substance (GSS). Our previous studies
have shown that GSS stimulates cAMP
production in starfish follicle cells,
coincident with an increase in 1-MeAde
production. These results suggest that
cAMP is important in mediating 1-MeAde
production by starfish follicle cells in
response to GSS.

The present study was designed to
investigate the possible function of GTP-
binding proteins (G-proteins) and
adenylate cyclase (AC) in GSS-induced 1-
MeAde production in follicle cells of
As ter i na pectini fera. GSS stimulated AC
activity, but not phosphodiesterase
activity. GSS-induced AC activation was
further enhanced by GTP . GTP analogs
(GTP-yS and Gpp(NH)p) also stimulated AC
activity. In addition, both cholera toxin
and pertussis toxin stimulated AC activity
These results suggest that G-proteins are
important in mediating the response of
starfish follicle cells to GSS, and
further that at least two types of G-
proteins, such as Gs and Gi , participate
in the GSS-induced AC activation.

DB 20

PARTICIPATION OF TRYPSIN-LIKE ENZYME IN
OOCYTE MATURATION IN STARFISH.

M. Takagi Sawada1’2. H. Sawada, M.
Hoshi , and T. Someno . ^Dept. of Life
Sci., Fac. of Sci., Tokyo Inst. of
Technol. , Tokyo, 2Biosci. Lab., Abe
Trading Co. , Osaka, JRes. Labs. ,
Pharmaceut. Grp. , Nippon Kayaku Co. ,
Tokyo .

It is known that oocyte maturation in
starfish is strongly inhibited by
chymotrypsin inhibitors including
chymostatin. In contrast, trypsin-like
enzyme has been considered not to play an
essential role in oocyte maturation, since
inhibitors of trypsin-like enzymes,
antipain and leupeptin ( ac e t y 1 - Leu- Leu -
Arg-H) weakly block germinal vesicle
breakdown (GVBD). In order to elucidate
the physiol igical role of trypsin-like
enzyme, we synthesyzed subsite-substituted
9 leupeptin analogs, and examined the
effects of these analogs on oocyte
maturation in a starfish, Aster ina
pectinif era .

IC50 of Z-Leu-Leu-Arg-H for GVBD was
17-fold lower than that of leupeptin,
which indicates that similar inhibition
level of chymostatin. Z-Phe-Ser-Arg-H , the
strongest inhibitor, showed 100-fold
strong inhibition compared with that of
leupeptin. Thus we conclude that trypsin-
like enzyme possesing a narrow subsite
specificity participates in starfish
oocyte maturation.

DB 21

A CALYCULIN- INDUCED OOCYTE MATURATION IN
ASTERINA PECTINIFERA.

H. Tosuji1, T. Nakazawa1, Y. Kato2 and N.
Fusetani2. xDept. of Biol.. Fac. of Sci.,
Toho Univ., Funabashi and 2Lab. of Mar.
Biochem., Fac. of Agric., Univ. of Tokyo,
Tokyo . _ _

Calyculin (CAL) A is a lipophilic

extract from sponge Discodermia calyx. It
has been known a novel antitumor
metabolite. In the present study, CAL A
was found to induce the oocyte maturation
of Asterina pectinif era . In order to
elucidate the mechanism of oocyte
maturation. several characteristics of
CAL A effect were analyzed in comparison
with those of 1-MA. CAL-induced oocyte
maturation occurred more slowly than the
1-MA- induced one. (It was required about
70 min for 50% of GVBD.) The oocyte
maturations with both inducers were
inhibited by Pronase E. Each effect of
CAL and 1-MA on the oocyte maturation did
not block mutually. These results suggest
that CAL receptor may be different from 1-
MA receptor. The increase in protein
kinase (histone kinase) activity was also
slowly in proportion to the rate of GVBD.
Additionally, this oocyte maturation was
inhibited by arachidonic acid (AA) as same
as the 1-MA- induced one, but the protein
kinase activity was not inhibited by
arachidonic acid. It suggests that AA
inhibit the induced GVBD after the protein
phosphorylation. This protein kinase
activity existed in cytosol fraction.

1254

Developmental Biology

DB 22

INHIBITION OF STARFISH OOCYTE MATURATION:
EFFECTS ON SIGNAL RECEPTION & TRANSDUCTION.
M. Yoshikuni1, K. Ishikawa2 & Y. Nagahama1
^ab. of Reprod. Biol., Natl. Inst, for
Basic Biol., Okazaki, 2Dept. of Biol., Fac.
of Sci., Shizuoka Univ., Shizuoka.

Many inhibitors of oocyte maturation in
starfish have been reported. This has made
it difficult to find the real pathway of
signal transduction of the 1 -methyladenine
stimulus. In this study, we classified the
inhibitors into two types by means of a
radio-receptor assay developed in our lab*.
One type suppressed the specific binding of
[ 3H ] 1 -methyladenine to the receptor on the
oocyte surface. This type included various
compounds, such as atropine, caffeine,
chloroquine, dansylcadaverine, dibucaine,
propranolol, verapamil, etc. This type of
inhibitor was found to act on signal
reception. Another type included calmodulin
inhibitors (W-5, W-7), protease inhibitors
( chymostatin , leupeptin, tosyl-L-arginine
methylester ) , C-kinase activators ( phorbol
ester, dioctanoylglycerol ) , and SH reagents
( phenylarcine oxide, N-ethy lmaleimide ) .
These reagents didn't suppress 1 -methyl
adenine binding but still inhibited oocyte
maturation. These results suggest that
calmodulin, protein phosphorylation and
proteolysis play important roles in the
pathway of signal transduction, after
1 -methyladenine binding, to produce
maturation-promoting factor. *M. Yoshikuni
et al., (1988) PNAS 85, 1874-1877.

DB 23

A G-PROTEIN RESPONSIBLE FOR 1 -METHYL¬
ADENINE INDUCED MATURATION IN STARFISH.

K. Chiba, L.A.Jaffe1, T.Kishimoto,

T.Katada, H.Tadenuma and M. Hoshi .

Dept. Life Sci., Fac. of Sci., Tokyo Inst,
of Tech, Tokyo. 1Dept. of Physiology,

Univ, of Connecticut Health Center, U.S.A.

1 -Methyladenine ( 1-MA) reinitiates

maturation of starfish oocytes; formation
of maturation-promoting factor (MPF) and
the following germinal vesicle breakdown
(GVBD) . Microinjection of pertussis toxin
(PTX) , an inhibitor of G-protein, inhibits
1-MA induced GVBD in Aster ina pectinifera
and Patiria miniata . We raised a question
whether PTX blocks the formation of MPF or
its action or the both. When MPF (cyto¬
plasm taken from maturing oocytes) was
injected into the PTX-prein j ected oocytes,
more than the half of them underwent GVBD
(21/24 in A . p . and 25/46 in P.m. ) . The
MPF-rescued oocytes were fertilized
normally in terms of the elevation of
fertilization envelope (FE) and cleavage.

PTX catalyzed the ADP-r ibosylation of a
39-kDa G-protein of cortices isolated from
A . p . oocytes, when they were incubated
with ^P-labeled nicotinamide adenine
dinucleotide .

These results indicate that a G-protein
of 39-kDa mediates the transduction of the
external signal (1-MA) into an intra¬
cellular factor (MPF) , but this G-protein
is not involved in FE elevation and
cleavage .

DB 24

Poly ( A ) +RNA extracted from starfish eggs
induces oocyte maturation.

K. Tachibana1-1'3, M.Ishiura2, T.Uchida1 and
T.Kishimoto1. 1 Biol . Lab. , Tokyo Inst. Tech.,
Tokyo, 2 Dept .Cell Biol ., Natl . Inst . Basic
Biol ., Okazaki , 3 Dept . Biol ., Nagoya Univ.,
Nagoya . _ _ _ _

We have extracted RNA from starfish

eggs and found that the egg poly(A)+RNA
can induce germinal vesicle breakdown
(GVBD) upon injection into immature
oocytes of starfish and Xenopus . The
molecular size of the poly (A)+RNA
responsible for GVBD was estimated to be
approximately 22S by sucrose density
gradient centrifugation. Since this size
is similar to the reported size of sea
urchin cyclin mRNA , we have performed
northern blot analysis; starfish RNA was
extracted from immature and maturing
oocytes at various periods after 1-methyl-
adenine addition and probed with a
synthetic oligonucleotide cording for the
conserved region between sea urchin and
surf clam cyclin cDNA. This analysis
suggests that the starfish RNA contains
cyclin homologous transcript and that the
contents of cyclin homologous poly(A)+RNA
increase gradually during oocyte
maturation. Incubation of starfish egg
poly(A)+RNA and antisense cyclin
oligonucleotide with or without RNase H
completely destroyed its GVBD-inducing
activity. These results indicate that the
poly(A)3'RNA responsible for GVBD is the
cyclin homologous mRNA.

DB 25

RANDOM DISPLACEMENT OF I -CELLS INTO BUDS AND
MAINTENANCE OF THE I -CELL POPULATION IN HYDRA.
T. Fujisawa. Natl. Inst, of Genetics, Mishima.

It has been believed that the I— cell populat¬
ion in hydra is maintained by regulating its
self-renewal probability (Bode & David, 1978).
Recently, however, Heimfeld (1986) reported
that the rate o-f I-cell growth was relatively
unchanged even when its density was low and
that the recovery of I-cells in the adult
tissue was attained by increased retention. In
the present study, I have re-examined this by
using a different experimental scheme. Polyps
of the low 1-cell density was produced by in¬
troducing a small piece of tissue from a nor¬
mal polyp into an epithelial polyp which
lacks an I~cell lineage and it was examined
the growth rate of I-cells and the mechanism
of 1-cell displacement into buds. The results
showed that unlike the Heimfeld’ s results the
growth rate of I-cells was 2-fold higher at
low density than normal. Furthermore, it was
suggested that displacement of I— cells from an
adult into buds was a random process. These
results do not support for the Heimfeld’ s
hypothesis but rather support, although in¬
directly, for the prevalent concept of self-
renewal regulation.

Developmental Biology

1255

DB 28

DB 26

NERVE NET FORMATION IN REGENERATING HEAD
OF HYDRA: STUDIES USING MUTANTS AND
CHIMERAS

H.Mizumoto 1'2and 0. Koizumi -1-
IPhysiol . Lab . , Fukuoka Women's Univ.,
Fukuoka, 2Dept.Biol., Fac.Sci., Kyushu
Univ . , Fukuoka

BASAL DISK FORMATION IN HYDRA,

Pelmatohydra robusta. INVESTIGATION WITH
A MONOCLONAL ANTIBODY WHICH RECOGNIZES
BASAL DISK GLAND CELLS SPECIFICALLY.

Y. Kobayakawa and E. Ohotsuka. Biological
laboratory. College of General Education,
Kyushu University 01, Fukuoka.

We

of nerv
hydra i
Sci . , 4
step i
cells a
second
cells a
In

the ner
and ch
f luores
antiser
budding
def icie
mutant ,
strains
normal
compose
nerve c
examine
compar i
parenta
that t
control
epithel

rep
e ne
n th
(6)
s t
t th
is

t th
the
ve n
imer
cenc
urn .

mu
nt m
mh-
, w
str
d of
ells
d.
son
1 st
he
led
ia 1

or te
t du
e la
, 1
he

e re
the
e ap
pre
et
as

e wi
Thr
tant
utan
1 , a
hich
ain
th
fro
Evid
of

rain
ner
by e
cell

d the t
ring he
st annu
039, 1

appear e
gener at
appea
ex .

sent st
regener
by me
th a
ee muta
, L4 ,
t, reg-
nd thes
were
(105)
e epit
m dif f e
ence o
the ch
s shows
ve ne
nvironm
s .

wo step formation
ad-regeneration in
al meeting (Zool.
987 ) . The first
ance of ganglion
ing tip and the
ranee of sensory

udy ,
atio
ans
pept
nt s
a

16,
e c
pro
and
heli
rent
btai
imer
the
t

ents

we

of
ide ,
trai
reg
a ma
hime
duce
mu
al
str
ned
as

cl

form

Pr

n m
i
RF
ns :
en er
lti-
r ic
d f
tant
cell
ains
f ro
and
ear
atio
ovid

amined
utants
mmuno-
amide ,
a slow
ation-
headed
hydra
r om a
s and
s and
, were
m the
their
cases
n is
ed by

A monoclonal antibody, AE03,
specifically recognized granules which are
located only in apical cytoplasm of basal
disk gland cells in hydra, Pelmatohydra
robusta . The boundary between the basal
disk and the peduncle was distinct as a
space between two neighboring cells, one
of which possessed AE03-positive granules
while the other did not. Thus, we
confirmed the usefulness of this
monoclonal antibody as a marker to
identify differentiating basal disk gland
cells in their differentiation. We
examined the regeneration of the basal
disk in hydra which amputated feet and the
formation of the basal disk during budding
with respect to the appearance of AE03-
positive granules. During foot
regeneration, the reappearance of AE03-
positive granules occurred as the first
indication of the regeneration of the
basal disk 12 h after amputation, and the
reappearance was affected by the feeding
condition. During budding AE03-positive
granules appeared after the constriction
between the bud and the parental body
column which began after the completion
of the head and body formation.

DB 27

MORPHOLOGICAL STUDY ON THE MECHANISM OF DE¬
LAMINATION EXHIBITED BY EMBRYO OF PELMATO¬
HYDRA ROBUSTA.

K.Noda and C.Kanai. Lab. of Electron Micro¬
scopy, Tokyo Metropolitan Inst, of Gerontol.

We obtained small eggs by reducing the
population of oogonia surrounding oocyte by
dissection. There was a large difference in
the average diameter between experimentally
dwarfed (328ju) and normal (479p) eggs.

We used these small and large eggs for
investigating the trigger events of delami¬
nation. Histological observation of embryos
with their maximum diameter revealed that
large embryos are in the process of delami¬
nation, while smaller ones than 360/j are
still in the blastula stage. There is large
difference in thickness of the wall between
large and small embryos with their maximum
diameter. In small blastula, blastmeres are
cuboidal or even squamous in shape. They
always reduce their diameter to obtain co¬
lumnar shape of blastmeres before they
beginn to delaminate.

We conclude, therefore, that columnar
shape of the blastmeres is one of the
events which trigger delamination.

It is also suggested that whole tension
on the surface of embryo determine the
maximum diameter of embryo, and initiate
for blastmeres with some special conditions
to change from cuboidal to collumar shape.

The numbers of times of cleavage had no
direct relation to the trigger events of
delamination.

DB 29

WOUND CLOSURE IN PLANARI AN EPIDERMIS. III.
THE EPIDERMAL CELL MOVEMENT.

S. Ishii. Div. Cell Sci., Cent. Res. Lab.,
Fukushima Med. Col. , Fukushima.

Epidermal cell movement involved in wound
closure was further examined by SEM using
regenerants of D. j aponica , B. brunnea and
P. vivida. There seems to be a correlation
of the cell movement with circumstances of
the substratum cells move. Conclusive re¬
marks on the movement and related events
are: (1) Sheet shifting. Rapid cellular mo¬
bilization occurs in the surrounding area
of the wound where many epidermal rugae
have developed by contraction. An increas¬
ed density of cell population in this area
causes an outward shifting of the epider¬
mis, probably by gliding of cells, that
continues until late blastema stages. (2)
Cell extension . The epithelioid is formed
by a spreading of specially stimulated mar¬
ginal cells of the wound. The rapid spread¬
ing of it may be analogous to that of the
leading lamella of the vertebrate epider¬
mis. The sheet gradually degenerates and
often detaches. The adhesion of it to the
substratum is very poor because of the in¬
terference by cell debris derived from the
injury. (3) Cell migration . An additional
growth of the epithelioid exclusively owes
to cell migration. Migration also occurs
on the re-exposed surface of the wound if
it is favorably conditioned; the narrower
the gap of wound, the faster the migration.

1256

Developmental Biology

DB 30

MARGINAL STRATIFICATION OF THE PLANARIAN
WOUND EPIDERMIS.

I. Hori. Dept. Biol., Kanazawa Med. Univ.,
Uchinada, Ishikawa.

The closure of anterior and posterior
wounds was studied in worms transected be¬
hind auricles.

Quick response of epidermal cells after
amputation depended on muscular contraction
near the open wound. Early epithelization
involved two dynamic changes of epidermal
cells at the wound margin. First, the cells
became stratified so that the old cells
from dorsal and ventral epidermis simulta¬
neously spread to cover the wound surface
and joined by the first intention near the
wound center. Secondly, these cells were
passively' stretched by partial relaxation
of wounds. This was a causative factor for
extreme elongation of the wound epidermis.
Structural connections between cells and
the basal lamina were removed in response
to the rapid relaxation. The disappearance
of hemidesmo somes was obviously linked to
the possibility of cell locomotion. Basal
protrusions were reduced in number, in pro¬
portion to the degree of elongation of each
cell. The wound epidermis adhered tightly
to some kind of parenchymal cells.

Further development of the wound epider¬
mis was established not by replacement with
the second epidermis but by successive re¬
newal of parenchyma-derived precursor cells
along the course of cell differentiation
within the regeneration blastema.

DB 31

MORPHOLOGICAL CHANGES IN EARLY REGENERATING
SCALES IN THE JAPANESE FLOUNDER.

S. Kikuchi' , N. Makino2 , A. Shimozawa3 and
H. Nakamura3. ’Kominato Mar. Lab., Fac. of
Sci., Chiba Univ., 2Chiba Pref. Fish Farm.
Ctr. and 3Dept. of Anat., Dokkyo Univ.,
School of Med. _ _

Regeneration of the scale on the Japanese

flounder, Paralichthy s olivaceus was ob¬
served. Scales were plucked from the scale
pocket and regenerating scales were ob¬
served 0, 2, 5, 7, 10 and 14 days after op¬
eration histologically, HE staining and
alizarin red for observation of calcifica¬
tion. Within 2 days, the wound was re¬
stored and the delicate layers of flat
scale formation cells (SFC) was observed in
the dermis. Between the layers, a very
thin scale plate was partially developed.
After 5-10 days, SFC spread in two definite
layers and a calcified scale plate was
formed between them. The calcification was
observed to progress from the outer surface
of the plate. This process of calcifica¬
tion resulted in several projections on the
scale. In the peripheral region of the
scale, these projections arranged them¬
selves in circular lines which made the
characteristic pattern of the scales.
Grooves, another characteristic pattern of
these scales, were formed in the anterior
region of early regenerating scales. The
edge of these scales were surrounded by
tall SFC. Even after 14 days of regenera¬
tion, no ctenium was observed on the ocu¬
lar side of the fish.

DB 32

A STUDY ON TISSUE SPREADING IN THE
WOUND CLOSURE OF FETAL RAT SKIN IN VITRO.
S.Ihara, Y.Motobayashi^ and E.Nagao. Dept,
of Plast. Surg. and Dept, of 'Biochem.,
Kitasato Univ. Sch. of Med. _ _

A wound closure was observed within
2 days when the open-wounded fetal rat
( Spraque-Dawley ) skin piece was excised at
1 6 day of gestation and cultured in an
organ culture system (using Dulbecco's
modified Eagle medium containing 10% fetal
calf serum) . The skin pieces from later
fetal stages failed in this capacity. Thus
the rapid wound closure may represent the
mode of fetal wound healing. From
phase-contrast microscopic survey, it was
found that the wound was closed by
spreading, but not contraction, of the
wound-surrounding skin. The histological
sections showed that, at the closing wound
edge, some large-sized cells (possibly
differentiated k e r a t i n o c y t e s ) were
spreading out in a parallel direction with
the edge. The small-sized (basal) cells
did not directly enveloped the wound, but
they were adjacent to the marginal zone of
the large cells. In the present system,
the wound closure was completly inhibited
by cycloheximide ( 5pg/ml) or cytochalasin B
(lOpg/ml), but neither by hydroxyurea
(2mM), colcemid ( 50ng/ml ) nor LACA (a
proline analog, lOOpg/ml). These findings
suggest that cy toskele tons , especially
microfilament system, play a major role in
an extensive spreading of skin, which
results in the wound closure.

DB 33

CELL CYCLE ASPECTS ON ARCHENTERON
F0RAMTI0N IN SEA URCHIN EMBRYO.

H. Mi zoguchi1 , and S. Tanaka2. 1 Lab. of
Biol. , Jun. Col. of Rissho Univ. ,
Saitama, 2 Dept. of Develop. Biol. ,
Mitsubishikasei-Inst. of Life Sciences,
Tokyo. _ _

In order ftr~ inquire whether cell

proliferation at blastula stage of sea
urchin embryo is related to archenteron
formation, effects of 5-bromodeoxyuridine
(Br4lT) were investigated. Detection of S
phase cells was done by BrdU label,
assayed by immumocytochemistry using
monoclonal antibody against BrdU and
streptavidin-biotin peroxidase system.

Treatment of the embryos with BrdU,
starting at the swimming blastula stage,
inhibited archenteron formation. BrdU-
caused inhibition of archenteron
formation was canceled about 90% by
excess thymidine. Addition of BrdU to
the embryo culture, at the mesenchyme
blastula and gastrula stage, resulted in
the formation of normal archenteron.
Inhibition of archenteron formation
treated by BrdU and recovary of
archenteron formation by adding thymidine
were only detected at the swimming
blastula stage.

It seems that S phase cells in vegetal
hemisphere of swimming blastula were
participated to archenteron formation.

Developmental Biology

1257

DB 3L\

SPATIOTEMPORAL EXPRESSION OF COLLAGENASE
DURING PRIMARY MESENCHYME CELL INGRESSION
IN SEA URCHIN BLASTULAE.

H. Katow, Biol. Lab., Rikkyo Univ., Tokyo

The basal lamina of sea urchin blastulae,

which divides the epithelium from the blas-
tocoel, is mainly composed of such large
molecules as collagen. The blastocoel is
the very place where the morphogenetic
movement occurs. The ingression of the
primary mesenchyme cells (PMCs) is the
prelude of such movement. In order to in¬
gress the PMCs into the blastocoel, they
must pass through the basal lamina.

In this study polyclonal antibodies
raised against collagenase from Clostridium
hystor icum (EC. 3.4.24.3) was applied to
blastulae of Clypeaster j aponicus . Pre¬
sumptive PMCs were localized using a PMC
specific monoclonal antibody, P4. Location
of above two antibodies was examined immu-
nohistochemically .

The enzyme was not seen in the basal
lamina of both swimming and mesenchyme blas¬
tulae. However, in the vegetal plate blas¬
tulae, in which the PMC ingression was in
progress, the enzyme was seen in the basal
lamina of immediately above the presumptive
PMCs.

Although further examinations are still
under way, it is conceivable that the
enzyme is expressed only around the period
of the PMC ingression, and that the expres¬
sion is regulated spatiotemporal ly .

DB 35

ARE THE MESOBLASTIC CELLS GUIDED TO
LOCOMOTE WITH ECM ON UNDERSURFACE OF EPI-
BLAST IN THE CHICK EMBRYO AT THE PRIMITIVE
STREAK STAGE ?

R. Toyoizumi and S. Takeuchi . Zool . Inst.,
Fac . Sci., Univ. Tokyo, Tokyo.

The mesoblastic cells in the chick
embryo at the primitive streak stage(H.H.,
St . 4 ) migrated radially from anterior part
of primitive streak either on or through
the network of ECM fibers. To know whether
the orientation of their locomotion was
determined with the distribution of ECM, we
examined the relation between the orientat¬
ion of the cells and the distribution of
f ibronectin(FN) which was one of the most
important components in ECM adhesive to the
cells .

With the aid of immuno-gold method for
scanning electron microscopy, FN was found
to be rich in the ECM fibers which run
radially from an anterior center of primi¬
tive streak along the undersurface of epi-
blast . Some of filopodia were recognized to
attach to the fibers.

The orientation of each cell was defin¬
ed as an average of angles of filopodia on
the assumption that they were more oftenly
protruded to the direction of cell locomot¬
ion. The anterior mesoblastic cells had
significant preference to the toward direc¬
tion to locomote.

The ECM fibers aligned radially along
undersurface of epiblast could be a factor
giving the mesoblastic cells the direction
to locomote.

DB 36

WHEN DOES THE EPIBLAST ESTABLISH EPITHELIAL
STRUCTURE IN THE CHICK EMBRYO ?

S. Takeuchi. Zool. Inst., Fac. Sci., Univ.
Tokyo , Tokyo .

To throw light on the basic mechanisms of
cellular behaviour during morphogenesis, F-
actin in epiblastic cells of chick embryos
during primitive streak(PS) formation was
surveyed cytochemical ly .

The chick embryos at the stages from 2 to
5 (Hamburger & Hami lton , 1 95 1 ) were fixed in
3%neutral formalin in PBS supplemented with
ImM MgCl 2 , 0 . ImM CaCl2,and 0 . 05%Tr itonX-100 .
The embryos stained In toto with Rhodamine
phalloidin were observed with an epi-fluor-
escence microscope.

Up to the stage 3, no F— actin was recog¬
nized in the epiblastic cells at central
area but those at the margin of blastoderm,
which moved extensively outwards (epiboly) .
In some tiers of marginal cells, F-actin
was aligned along cell border. The meso¬
blastic cells seemingly invaginat ing ( flask
cells) were stained along cell membrane but
had no bundles running through cell body.
Since after St . 4 , clusters of epiblastic
cells circumscribed with F-actin began to
be observed along PS or around presumptive
Hensen's knot. After Hensen's pit was form¬
ed ( St . 5 ) , all of epiblastic cells were re¬
cognized to have F-actin along cell border.

F- actin equipped in the epiblast along

cell border is considered to play somehow
essential role in embryonic morphogenesis .

DB 37

PARTIAL PURIFICATION OF SELF-NONSELF REC¬
OGNITION MOLECULES ON THE VITELLINE COAT
OF ASCIDIAN EGGS.

K.Kawamura, T.Kameda and M.Nakauchi. Dept.
of Biol., Fac. of Sci., Kochi Univ., Kochi.

Gametes of Ciona intestinalis can discrim¬
inate between self and nonself. The allo-
recognition takes place on the vitelline
coat of eggs. When unfertilized eggs were
treated with acid seawater (pH. 3.2) for
15 minutes, the conditioned seawater could
inhibit sperm-egg binding. This inhibitory
effect was directed only toward nonself
sperm, showing that the acid extracts have
an allorecognition activity. The activity
was heat-stable. Trypsin strengthened non-
specifically to some extent the inhibitory
effect of extracts on sperm-egg binding.

On the other hand, the inhibitory effect
was lost completely after protease diges¬
tion. The allorecognition activity could
be separated into chloroform/methanol (C:M=
2:1) -soluble and CM-insoluble fractions,
both of which were nondialyzable (molecular
cut, 2,000) and contained amino acids. The
former fraction was occupied mainly by glu¬
tamic acid. It had three major peaks after
orderly-phase HPLC. Any peaks did not show
the activity solely, but peak H restored
the activity when recombined with peak I.
The peak H was sensitive to protease and
was easily denatured in the elution solv¬
ent (isopropanol+hexane+water) . The peak I
was highly hydrophobic and could be further
purified in the linear gradient of hexane
from 90% to 45%.

1258

Developmental Biology

DB 38

VARIATIONS IN THE STRUCTURES OF SPERM
ACTIVATING PEPTIDES FROM THE EGG JELLY OF
SEA URCHINS IN ORDER ECHINOIDA
K. Yoshino1, M. Yamaguchi1, N. Suzuki1,

K. Nomura2 and H. Kajiura3

1; Noto Marine Lab Kanazawa Uni v . , Uchiura
2; Dept, of Biochem., Tokyo Metropolitan
Inst, of Gerontol., Tokyo
3; Natl. Inst, for Basic Biol., Okazaki

It has been reported that the egg jelly
of sea urchins belonging to order Echinoi-
da contain Sperm Activating Peptide I ( SAP-
I and its derivatives of which structure
can be written as Xl-Phe-X2-Leu-X3-Gly-X4-
Gly-Val-X5 in general. In order to con¬
firm generality of the structure, we iso¬
lated twenty different Sperm Activating
Peptides from the egg jelly of sea ur¬
chins, Tripneustes gratilla , Pseudo bole tia
maculata , Echinometra mathaei (Type A and
Type B) and Heterocentrotus mammillatus .
Although their structures were quite vari¬
able, the residue at position 2 was aro¬
matic and those at positions 4 and 9 were
hydrophobic nature. Some peptides con¬
tained a rare amino acid, dichlorotyrosine
at the position 2 ( Gly- [diCl-Tyr ] -Asp-Leu-
Asn-Gly-Gly-Gly-Val-Gly and Gly- [ diCl -Tyr ]
-Asn-Leu-Asn-Gly-Gly-Gly-Val-Gly ) . These
peptides increased cyclic GMP concentra¬
tion in own sperm and stimulated Hemicent-
rotus pulcherr imus sperm respiration at
half-maximal stimulations of about 10-10M,
the value was almost the same as that of
SAP-I for H. pulcherr imus spermatozoa.

DB 39

EFFECT OF SPERM ACTIVATING PEPTIDE (SAP-I)
ON LIPID METABOLISM IN HEMICENTROTUS PUL¬
CHERR IMUS SPERMATOZOA

T. Harumi 1 , M. Mita2, N. Ueta2 and N.
Suzuki1. 1Noto Marine Lab., Kanazawa
Univ., Uchiura and 2Dept. of Biochem., Sch .
of Med., Teikyo Univ., Tokyo. _ _

I t has been reported that decreased
respiration rates and motility of H. pul-
cherrimus spermatozoa in artificial sea
water (ASW) at pH6 . 6 could be reversed by
addition of SAP-I. Recently, it was re¬
ported that sea urchin spermatozoa utilize
phosphatidylcholine (PC) as energy source
for motility.

In the present study, we examined
whether SAP-I stimulates lipid metabolism
in H. pulcherr imus spermatozoa in slightly
acidified ASW. When spermatozoa were in¬
cubated with SAP-I at pH6 . 6 for lhr, sperm
PC content decreased, while there was no
significant difference in the PC content
between with and without SAP-I at pH8 . 2 .
After incubation at pH8 . 2 , free fatty acid
(FFA) content in spermatozoa significantly
increased, although the FFA content remain¬
ed unchanged at pH6 . 6 even with SAP-I. In
the presence of SAP-I, sea urchin spermato¬
zoa metabolized exogenously added [14C]-
oleic acid at pH6 . 6 to almost the same
level as that at pH8 . 2 . This was dependent
on SAP-I concentrations.

These results indicate that SAP-I ac¬
celerates lipid metabolism, that is, from
PC to FFA and from FFA to CO2, in sea ur¬
chin spermatozoa in acidified sea water.

DB 40

MACROMOLECULAR COMPONENTS OF THE EGG JELLY
OF THE SEA URCHIN HEMICENTROTUS PULCHER-
RIMUS

M. Yamaguchi, T. Shimidzu arid N. Suzuki.
Noto Marine Lab., Kanazawa Univ., Uchiura.

Two macromolecular components, fucose
sulfate glycocon jugates ( FSG ) and sialo-
proteoglycan (SPG), were purified from the
egg jelly of the sea urchin H. pulcherrimus
by sequential chromatographies on Sepharose
4B , Sepharose CL-2B and DEAE Bio-Gel col¬
umns. FSG contained 250-300jug protein and
lOOpg fucose even after chromatography on a
Sepharose CL-2B column in 7M urea or a TSK
gel G6000PW column in 0.1% SDS . Most of
protein (about 80%) was released from FSG
after reduction-alkylation and separated
from the fucose moiety by HPLC on a TSK gel
G6000PW column in 0.1M sodium phosphate
( pH6 . 8 ) containing 0.1% SDS. The released
proteins were consisted of four components
with the molecular mass of 340K, 310K, 145K

and 140K. When [ 1 4C ] -iodoacetic acid was
used for reduction-alkylation, all four
components contained radioactivity. This
indicates that disulfide bonds play an im¬
portant role in f ucose-protein conjugation.
On the other hand, SPG purified by gel
filtration and ion exchange chromatography,
followed by CsCl density gradient centri¬
fugation, contained only 10% protein by
weight. Thus it should be called sialo-
proteoglycan .

DB 41

INDUCTION OF THE ACROSOME REACTION BY SAP-I
AND FUCOSE-RICH GLYCOCON JUGATE ( FRG )

M. Yamaguchi, M. Kurita and N. Suzuki, Noto
Marine Lab., Kanazawa Univ., Uchiura.

FRG prepared from the egg jelly of the
sea urchin Hemicentrotus pulcherrimus in¬
duced the acrosrome reaction in H. pulcher¬
rimus spermatozoa. FRG was tested for the
acrosome reaction inducing ability with or
without synthetic SAP-I. In the absence of
SAP-J , FRG showed about half of the ac¬
tivity of the original crude jelly. SAP-I
increased the rates of the acrosome reac¬
tion induced by FRG to nearly the equal
rates with the crude jelly. SAP-I promoted
the acrosome reaction with FRG at half-
maximal concentration of 4xlO_10M. The
concentration was almost the same value of
the peptide to cause half-maximal stimula¬
tion in sperm respiration. When FRG was
digested with pronase E, the activities of
FRG to induce the acrosome reaction were
reduced to about 50% of non-digested FRG
activities. SAP-I is known to increase '
intraceller pH, and concentrations of Ca2+
and cyclic nucleotides in sperm cells as do
monensin, A23187 and IBMX . They also in¬
creased the rates of the acrosome reaction
with FRG or pronase-digested FRG. However,
the rates did not reach those of FRG or
digested FRG with SAP-I. These results in¬
dicate that SAP-I acts as a specific co-
factor for induction of the acrosome reac¬
tion by FRG.

Developmental Biology

1259

DB 42

BINDING OF [3H] NITRENDIPINE TO PROTEINS IN
PLASMA MEMBRANES OF SEA URCHIN SPERM.

K. Mikami-Takei and I. Yasumasu. Dept, of
Biology, Sch. of Educ . , Waseda Univ., Tokyo.

A plasma membrane fraction of sea urchin
sperm was prepared after the method
described by Podell et al. (1984) * . The
electrophoretic profiles of sperm membranes
from three sea urchin species were similar
and exhibited several characteristic bands.
The membrane fraction bound r H] nitrendi¬
pine, an antagonist of voltage-dependent
calcium channel, though the binding seemed
to be of low affinity and needed concent¬
rations of about 1CK times greater than
those in muscle and nerve cells. Another
calcium channel antagonist, diltiazem,
enhanced [h] nitrendipine binding in sperm
membranes in concentrations of 10 to 10()pM.
The binding was not decreased by excess
amount of nifedipine. Electrophoretic
analysis showed there were several bands
that bound^ [JH] nitrendipine . Diltiazem
enhanced [JH] nitrendipine bindings in four
bands (MW 210, 148, 125 and 89 KDa) , and
inhibited bindings to ^.he other bands.
Nifedipine inhibited [n] nitrendipine
bindings in these four bands and increased
bindings of a few other bands that expla¬
ined the failure of measurement of specific
binding of [ JH] nitrendipine by this drug.

It is probable that there are calcium
channels among proteins of the four bands.
*Podell, Moy and Vacquier, BBA 778 (1984)
25-37

DB 43

INVOLVEMENT OF WGA BINDING PROTEIN IN THE
INDUCTION OF THE ACROSOME REACTION OF TWO
SPECIES OF SEA URCHINS, STRONGYLOCENTROTUS
INTERMEDIUS AND HEMICENTROTUS PULCHERRIMUS
Y. Sendai and K. Aketa. Akkeshi Mar. Biol.
Stat. , Hokkaido Univ. , Akkeshi Hokkaido

When the acrosome reaction (AR) of sea
urchin sperm is induced by the egg jelly,
the fluxes of specific ions (Ca2+ and Nan-
influxes and H+ efflux) occur. In the
preceding studies, we showed that the wheat
germ agglutinin-binding protein (WGAbp) of
Strongylocentrotus intermedius ( S . i ■ ) sperm
membrane may be important for the induction
of the ionic events associated with the AR
(Sendai et al . , 1986, 1987). We now find
that WGA binds to the acrosomal region of
Hemicentrotus pulcherrimus (H. p . ) sperm as
well as S^ i^_ sperm and inhibits the AR.

The WGAbp of F sperm exhibits the same
molecular mass (260 kD under non-reducing
condition) as that of Sj_ sperm. The
polyclonal antibody directed against the
WGAbp of i^_ sperm cross-reacts with the
WGAbp of F p^. When FL_ p^ sperm are
treated with this antibody ( IgG or Fab
fragments), the AR occurs. In addition, the
egg jelly of S^ can induce the AR of
H . p . sperm.

These results suggest that the WGAbp of
the present two species of the sea urchins
is a common regulator protein and/or a
channel protein involved in the AR-inducing
system( s) .

DB 44

CHYMOTRYPSIN-LIKE PROTEASES FROM SEA URCHIN
SPERM AND POSSIBLE PARTICIPATION OF 720kD
PROTEASE IN THE ACROSOME REACTION.
K.Matsumura and K. Aketa. Akkeshi Mar. Biol.
Stat., Hokkaido Univ., Akkeshi Hokkaido.

Chymotrypsin inhibitors and substrates
inhibit the egg jelly-induced acrosome re¬
action (AR) of Strongylocentrotus inter¬
medius . It is suggested that chymotrypsin-
like protease participates in the activa¬
tion of Ca2+ channels at AR (Matsumura and
Aketa, 1986,1987) .

In the present study, three types of
chymotrypsin-like proteases were partially
purified from sperm by a procedure in¬
cluding DEAE-Sephacel and hydroxylapatite
chromatographies and Sephacryl S-300 gel
filtration. Molecular weights of these pro¬
teases were estimated to be 27kD, 65kD and
720kD by gel filtration. All of these pro¬
teases hydrolyzed Suc-Leu-Leu-Val-Tyr-MCA,
a chymotrypsin substrate with inhibitory
effect on AR. Chymostatin inhibited all of
these proteases and AR. Leupeptin, even at
higher concentrations, had no effect on
both 720kD protease and AR, while it inhib¬
ited 27kD and 65kD proteases. The optimal
pH for 65kD and 720kD proteases were 7.0-
7.2 and that of 27kD was 8.2. Sixty five
kD protease had strong hydrolysis activity
toward Suc-Ala-Ala-Pro-Phe-MCA, a chymo¬
trypsin substrate with no effect on AR,
while 720kD had no activity toward this
substrate. These results suggest that 720kD
protease participates in AR of sea urchin
sperm.

DB 45

ASTERINA CO-ARIS ; PARTIAL PURIFICATION
AND ITS EFFECTS ON SPERM.

Y.Okita, T.Amano, T.Qkinaga, and M.Hoshi

Dept, of Life Sci. , Fac. of Sci. , Tokyo

Inst, of Technol. , Tokyo

Two jelly components, a high molecular
weight glycoprotein (ARIS) and a group of
steroidal saponins (Co-ARISs), are
required to induce the acrosome resction
(AR) in normal sea water in the starfish
Asterias amurensis.

Although ARIS is quite species-spesific,
all of the three Asterina Co-ARISs
served as Co-ARIS for Asterina sperm.
The activity of Co-ARIS in the jelly coat
of Asterina eggs behaved like Asterias
Co-ARIS through a purification procedure
using ion-exchange chromatography and
High Performance Liquid Chromatography on
an ODS column. However the activity to
increase the internal pH (pHi) of sperm
was eliminated by the procedure. The AR
induced by ARIS and partially purified
Co-ARIS was not proceeded by a pHi
increase in Asterina as reported in
Asterias.

Ca^-channel inhibitors and K + -
inhibitors blocked the jelly induced AR
in both species. From these data, we
conclude that basic mechanisms underlying
the induction of AR are common in the
family Asterd.idea: ARIS and Co-ARIS in
the jelly coat stimulate Ca-channels and
modulate the membrane potential of the
spermand eventually induce the AR.

1260

Developmental Biology

DB 46

ARIS REGULATES Ca- INFLUX TO INDUCE SPERM
HISTONE DEGRADATION OF THE STARFISH,
ASTERINA PECTINIFERA.

T. Amano, Y. Okita and M. Hoshi .

Dept, of Life Sci . , Fac . of Sci . , Tokyo
Inst, of Technol., Tokyo.

ARIS (Acrosome Reaction-Inducing
Substance) and diffusible components (Mg)
in the egg jelly are essential for
inducing the degradation of sperm histones
as well as the acrosome reaction in the
starfish, A. pectinif era .

Even if the absence of the jelly,
sperm underwent the histone degradation
but not the acrosome reaction in Na+-free
seawater. This reaction is Ca2 + -
dependent. Once sperm were treated with
egg jelly in Ca2+-def icient seawater, they
lose the capacity to degrade histones in
response to the jelly in normal seawater.
Histone degradation, however, was induced
in such jelly-pretreated sperm if they
were transferred into Na+-free seawater.
Similarly Ca2+-ionophore A23187, that
induced histone degradation hardly in
normal seawater, induced it in such
pretreated sperm. When sperm were treated
with A23187 and Mg together, they degraded
histones in the normal seawter . These
results suggest that ARIS regulates Ca2+-
influx and eventually induce the histone
degradation in cooperation with Mg.

DB 47

THE MORPHOLOGICAL CHANGES IN MITOCHONDRION
OF SEA URCHIN SPERM AT FERTILIZATION.

A. Hino, Dept, of Biology, Fac. of Sci.,
Nagoya University, Aichi.

It was reported that in the sea urchin

Pseudocentrotus depressus , almost all
sperm were acrosome-reacted when they pass
through the egg jelly layer (Kato, '87).

In this paper, another morphological
change in sperm which were interacted with
egg jelly or egg surface were studied.

After the fixation with Janus green B
containing glutaraldehyde solution, light
microscopic observation showed the shift
and swelling of mitochondrion of the sperm
which adhered on the egg surface. Such
mitochondrial round-up was also observed
in the sperm which were in egg jelly layer
but in lower percentage comparing to the
attached sperm. The intact sperm mito¬
chondrion was located around the flagellar,
but in reacted sperm, the mitochondrion was
translocated to the one side of the sperm
nucleus. With TEM observation, the inner
compartment of interacted mitochondrion
was completely fulled the space bounded
by the outer membrane. In vigorouly
swimming sperm, the cristae of mitochondria
was tightly folded and contorted. These
morphological changes in the mitochondria
might be reflected that the changes in the
respiration when sperm interact with egg.

DB 48

ULTRAMICROSCOPIC OBSERVATION OF ACROSOME
REACTION IN HAMSTER SPERMATOZOA ON HOMO
AND HETEROLOGOUS EGG ZONAE PELLUCIDA .
N.Uto1 , I.Ohta2 and O.Serizawa3. ^ept. of
Biol., 2 Central Lab. for EM and 3 Inst, for
Exper. Animals, Hamamatsu Univ. School of
Med . , Hamamatsu .

It is well established that zonae of
mature unfertilized eggs have the ability
to induce the acrosome reaction (AR) of
capacitated spermatozoa. And as shown in
previous report, the capacity of zonae to
induce AR is species-specific, that is,
the zona of hamster egg was found to have
stronger capacity to induce AR of hamster
spermatozoa than those of mouse. In the
latter case, many spermatozoa detached
from egg surface during subsequent incu¬
bation but some of remained spermatozoa
were induced AR in low degree. However,
the progression of AR on foreign zona tend
to retard and not followed by perfect AR.
Various morphological changes of acrosome
caps were observed (e.g., swollen, col¬
lapsed) . Hamster spermatozoa did not
penetrate the zona of mouse.

Electron mioroscopic observation indi¬
cated that the spermatozoa on foreign zona
were midway of AR or stopped at that point
In most of the spermatozoa, the acrosome
cap increased in volume and fusin of two
membrane (plasma membrane and outer mem¬
brane) occured partially but not caused
to exocytosis of cap. It may be depend
on significant heterogeneity of two zona.

DB 49

ACTIVATION OF SEA URCHIN EGGS BY 1-FLUORO-
2 , 4-DINITROBENZENE

H. Nakamura. Akkeshi Mar. Biol. Stat.
Hokkaido Univ., Akkeshi, Hokkaido.

When the eggs of the sea urchin, Strong-
ylocentrotus intermedius , were treated with
1-f luoro-2 , 4-chrnitrobenzene (FDNB) dissolved
in normal, K+-, Ca2+- or Mg2+ free arti¬
ficial sea water (ASW), they formed the fer¬
tilization envelope (activation envelope) in

2 4:o 5 minutes. No cytolysis was observed
even by prolonged exposure to FDNB. Mini¬
mum concentration for the activation of
more than 98% of eggs was 100 jiM. In ASW
containing both FDNB and L-cysteine, no
activation was observed. When the eggs were
pretreated with iodoacetamide , the effect
of FDNB was not observed. These results
suggest that FDNB effect is exerted by F-
released in consequence of the reaction be¬
tween FDNB entered into the egg and intra¬
cellular SH-groups. l-chloro-2 , 4-dinitro¬
benzene, another derivative of FDNB, was
not effective. Blocking of SH-groups might
not be responsible for the activation.

Lowed pH, 6.2, and pretreatment with
cytochalasin B made eggs insensitive to Ca
ionophore, ionomycin. In the same con¬
ditions, FDNB also did not activate eggs.

On the other hand, ionomycin activated eggs
pretreated with TMB-8 (inhibitor of intra¬
cellular Ca mobilization) , but FDNB did
not. These results suggest that the action
site of FDNB in the process of cortical re¬
action exists before that of ionomycin.

Developmental Biology

1261

DB 50

THE FUSION OF SPERM AND EGG MEMBRANE IS NOT
THE CAUSE OF AN INITIAL ELECTRICAL CHANGE
DURLNG FERTILIZATION

K.Takemoto and H.Kuroda, Sugashima Marine
Biol. Lab., Sch.Sci., Nagoya Univ. , Toba

The fertilization potential ( FP ) of sea
urchin eggs is known to consist of several
components. A step-like depolarization (SD)
is the initial component of it. McCulloh
and Chambers (1986) suggested that sperm-
egg fusion caused SD. We previously showed
that in low Na (8.6mM) artifitial sea water
(8.6 Na ASW) an egg was not activated by
sperm and did not produce FP. In this medi¬
um, a train of SDs lacking other components
of FP was observed. We examined the hypoth¬
esis of McCulloh and Chambers using DNA-
specific fluorochrome ( Hoechst33342 ) trans¬
fer technique. Eggs were loaded with the
dye and its membrane potential (MP) were
measured with conventional microelectorode .
Confirming the change of MP, the egg was
fixed with glutaraldehyde. If the sperm-egg
continuity is established before fixation,
the sperm head must be stained with the dye
pre-loaded in the egg. In normal ASW, the
egg was fixed 2-3min after the onset of FP.
The sperm head stained with the dye was ob¬
served by fluorescence microscope. In 8.6
Na ASW, the eggs were fixed after occur¬
rence of several SDs. Although several
sperms were observed to attach on the egg
surface, no sperm was stained with the dye.
This results show that the sperm-egg fusion
does not cause SD in 8.6 Na ASW and do not
support the above hypothesis.

DB 52

EFFECT OF Li+
FORMATION IN
E . Tazawa ^ , A .
■^Biol . Inst . ,

ON FERTILIZATION MEMBRANE
Li+ TREATED EGGS.

Fujiwara2 and I.Yasumasu,

_ , Yokohama City Univ., Yokohama

and ^Dept. 0f Biol., Waseda Univ., Tokyo.

Incubation of unfertilized eggs of sea
urchin with 100 mM Li + for 3 min at 20°C
blocked fertilization-induced accumulation
of 1,4,5-inositol triphosphate but did not
inhibit fertilization membrane formation.
Fertilization membrane formation was block¬
ed in Li+ treated eggs, when nifedipine was
added to egg suspension 30 sec after
insemination. Nifedipine blocked <+->Ca^ +
inflax into eggs, which was acturated around
30 sec after insemination in normal and Li+
treated eggs. Fertilization membrane
formation was blocked in Li+ treated eggs,
when they were transfered to Ca^+ free
artificial sea water. In intact eggs,
fertilization membrane formation was not
blocked when they were transterred to

Ca2 + free artificial sea water or exposed
tonifedipine 30 sec after insemination. In
Li* treated eggs, Ca2+ induced fertiliza-
tion membrane seems to depend solely on Caz+
inflax .

DB 51

EFFECTS OF TMB-8 ON OXYGEN CONSUMPTION
ENHANCED BY TREATMENTS WITH ACTIVATING
REAGENTS IN SEA URCHIN EGGS.

M. K. Kojima, S. Washida and S. Nakamura.
Dept . of Biol . , Fac . of Sci . , Toyama Univ . ,
Toyama .

We reported that O2 consumption of unf.
sea urchin eggs is enhanced not only by
treatment with procaine, but also by sub¬
threshold stimulation with Ca ionophore
A23187, insufficient to induce visible cor¬
tical changes. In the present study, there¬
fore, effects of NH4OH on O2 consumption
was first examined and it became clear that
when unf. eggs are exposed to 5-50 mM NH4OH
sea water, their O2 consumption rises 2.5-
3.5 times than that of the controls. As the
next step, it was determined whether or not
TMB-8, an antagonist of intracellular Ca
release, has inhibitory effects on increase
of O2 consumption induced by treatments
with activating reagents, such as procaine,
A23187 and NH4OH. It is interesting to note
that an increase of O2 consumption is can¬
celled by treatments combining 150 mM TMB-8
with 1 ;uM A23187, but not by treatments
combining 150 mM TMB-8 with 10 mM procaine
or 10 mM NH4OH . These results suggest that
Ca ionophore A23187 induces intracellular
Ca release and as a result, rise of respi¬
ration occurs, while weak bases, such as
procaine and NH4OH can enhance 02 consump¬
tion by stimulation of some metabolic
changes which do not directly connect to
processes of intracellular Ca release.

DB 53

INHIBITION OF FERTILIZATION MEMBRANE,
FORMATION AND Li+ AMOUNT
IN EGGS KEPT WITH Li*

I.Yasumasu1 , E. Tazawa2 and A.Fujiwara1
^ept. of Biol., Sch . of Educ . , Waseda
Univ., Tokyo, 2Bio. Inst., Yokohama City
Univ. , Yokohama. _

In unfertilized eggs of sea urchin,
incubated with Li* , the Li+ amount increased
in relation with the time of incubation and
the concentration of external Li+ . The
increase in the amount of Li+in eggs kept
with Li+ was strongly blocked by ouabain in
a concentration range for the inhibition of
Na+ , K * ATPase in sea urchin eggs. These
suggest that Li* is transported into eggs
across plasma membrane in place of K+in the
reaction catalyzed by Na+ , K+ATP ase . In K+
free artificial sea water containing Li+ ,
the amount of Li+ in egg became markedly
higher than in those kept in normal sea
water . In eggs thus treated in the presence
of Li+ , blockage of fertilization membrane
formation, which was observed when Li*
treated eggs were transferred to Ca2+ free
ASW 30 sec after insemination, was evident¬
ly observed in eggs containing Li+ at above
10 pmol/egg volume.

1262

Developmental Biology

DB 54

FERTILIZATION MEMBRANE FORMATION IN SEA
URCHIN EGGS INDUCED BY SEVERAL Ca2+
RELEASING SUBSTANCES, SUCH AS RYANODINE
AND QUERCETIN

A.Fujiwara, K.Taguchi and I.Yasumasu
Dept, of Biol. Sch. of Educ . , Waseda
Univ . , Tokyo . _ _

In sea urchin eggs, fertilization
membrane formation was induced by
ryanodine, miconazole, quercetin, clotri¬
mazole, doxorubicin, halothane caffeine and
chloroform. These compounds are known to
induce Ca2+ release through Ca2+ dependent
Ca2+ channels but failed to cause Ca2+
release through IP3 dependent Ca2+ channels,
in sarcosome of skeletal and smooth muscle.
In sedimentable fraction obtained from sea
urchin eggs, which underwent Ca2+ uptake at
the expence of ATP, 45Ca2+was released by
IP3 , Ca2* and these compounds. These
compounds caused Ca2+ release from the
sedimentable fraction which had been
treated with IP3 but did not induce the
release from the fraction having been
treated with Ca2+. Ca2+ release was not
induced by Ca2+ but was induced by IP3 in the
Ca2t treated fraction. In IP3 treated
fraction, IP3 did not cause Ca2+ release but
Ca2+ induced its release. Unusual Ca2+
release through Ca2+ dependent Ca2+ channels
in intracellular Ca24- stores probably have a
potency to induce fertilization membrane
formation in sea urchin eggs.

DB 56

INNER MOTION OF LIPIDS IN CELL MEMBRANE
AFTER FERTILIZATION OR ARTIFICIAL ACTIVA¬
TION OF SEA URCHIN EGGS.

K.Hirano. Tsukuba Res. Lab., Hamamatsu
Photonics KK, Tsukuba.

To investigate microviscosity of the
cell membrane, the fluorescence dopolari-
zation was measured by the fluorescence
spectrophotometer or under the microscope
using continuous light irradiation. The
cell membrane of unfertilized eggs was
stained with 5- ( N-hexadecanoyl ) -aminof luo-
rescein (AF) or RH421 . These eggs were
activated by insemination or incubation in
Ca + + ionophore-rcontaining seawater.

The whole cell measurements by the
fluorescence spectrophotometer showed
a little increase of the fluorescence
anisotropy after the activation in AF-
stained eggs but not in RH421 -stained eggs.
The local measurement by the microscope
system showed remarkable changes in the
anisotropy different with the directions
of the exiting polarization. Cytoplasmic
membranes of activated eggs stained with
perylene continued the level with unferti¬
lized eggs.

These results suggested that the invi-
ronment adjacent to the cell membrane
become a little viscous after the activa¬
tion, and the inner motion of the lipids
in the microvilli is suppressed more than
that in the basal membrane from 6-8 min
after the formation of the fertilization
or activation envelope.

DB 55

ANALYSIS OF CORTICAL GRANULE EXOCYTOSIS OF
THE SEA URCHIN EGG USING A HIGH-SPEED
VIDEO SYSTEM.

T. Mohri and Y. Hamaguchi, Biol. Lab.,
Tokyo Inst, of Technol., Tokyo.

During cortical granule exocytosis
which is observed upon fertilization, the
membrane of cortical granules fuses with
the egg plasma membrane and then cortical
granules extrude the contents into the
perivitelline space.

Cortical granule exocytosis of the sea
urchin egg is considerably rapid
phenomenon because we could not analyse
the process in detail with a normal speed
video system (60 fields/sec). But we
could observe the extrusion process of
cortical granules using a high-speed video
system (200 fields/sec Nac microscopic
high speed video, MHS-200: Nac
Incorporated) although we could not detect
when the membranes began to fuse. When
the egg of the sea urchins ( Hem i cent rotus
pulcherrimus and Clypeaster j aponicus ) was
fertilized and microinj ected with Ca-EGTA
buffer at high concentration of Ca2 + and
when the isolated cortices of sand dollar
(C. j aponicus ) eggs were perfused with a
solution containing high Ca2+, it took 30-
40 msec for each cortical granule to
extrude the contents.

I 25

MICROAUTORADIOGRAPHXC LOCALIZATION OF I-
GLYCOPEPTIDES SUGGESTS THE GRADUAL RELEASE
OF INDUCING SUBSTANCE FROM ORGANIZER.

K.Takeshimal, M.R.M.Diaz2, T .C . Takahashil ,
S.Osawa2, N.Takahashi3, K.Takata1.

1 : Radioisotope Ctr. , Nagoya Univ., Nagoya,
2: Dept, of Biol., Fac. of Sci . , Nagoya
Univ., Nagoya, 3:Dept. of Biochem. , Nagoya
City Univ. Sch. of Med., Nagoya.

Our previous studies have led us to the
idea that the organizer substance is a
molecule with Con A-like characteristics.
The treatment of organizer with glycopep-
tides obtained from ovalbumin (OVA) in¬
hibited neural induction in Cynops com¬
petent ectoderm. We prepared l^I-labelled
OVA and its glycopeptides , and examined the
binding sites of these ligands.
Autoradiography revealed that silver grains
were dense in the cell-cell interspaces,
but sparse either on the outer-surface or
in the cytoplasm of the organizer explant.
This pattern of localization was not ob¬
served in ectoderm and in ventral mesoderm
explants. A control experiment using 125i-
bbvine serum albumin (BSA) did not show the
particular distribution in the explants.

The release of 1251-OVA from the organizer
explant was more gradual than that of 125I-
BSA up to 9 hours after labelling. Results
suggest the gradual release of inducing
substance from the surface as well as from
the intercellular spaces of the organizer
explant.

Developmental Biology

1263

DB 58

MONOCLONAL ANTIBODIES RECOGNIZING THE INTE¬
RIOR CELL SURFACE OF THE COMPETENT ECTODERM
E . Asakura1 , S .Osawa1 , T .C .Takahashi2 ,

K .Takeshima2 , and K.Takata2, 1 Dept, of

Biol., Fac. of Sci., Nagoya Univ. , Nagoya,

2 Radioisotope Center, Nagoya Univ. , Nagoya

Neural induction is one of the most drama¬
tic events in the development of amphibian
embryos. Similar to the living organizer, a
plant lectin, Concanavalin A (Con A) , was
found to neuralize competent ectoderm
explants in vitro . We therefore suppose
that the receptor of neuralizing signal
exists on the interior cell surface of the
competent ectoderm, and has oligosaccharide
moiety with Con A-affinity.

A fraction from the competent ectoderm
purified by Con A-affinity chromatography
was used as immunogen for the preparation
of monoclonal antibodies. The assays for
screening hybridomas were 1) enzyme-linked
immunoassay to detect immunoglobulin secre¬
tion, 2) indirect immunofluorescence stain¬
ing of the ectoderm sections. 38 clones
were obtained based on these criteria. Four
lines of the hybridomas were grown as an
ascitic tumor in mice. Further immunohisto-
logical staining with these ascites showed
that there were at least 3 types of mono¬
clonal antibodies. One recognized both in¬
terior and exterior cell surface, another
stained cytoplasm as well as interior cell
surface, and the third was specifically di¬
rected to the interior cell surface. These
clones could produce candidate antibodies
against the receptor of neuralizing signal.

DB 59

ANALYSIS OF OLIGOSACCHARIDES IN CYNOPS
PRESUMPTIVE ECTODERM BY TWO-DIMENSIONAL
MAPPING TECHNIQUE.

T.Takabatake1 , S. Osawa1, T .C. Takahashi2 ,

K. Takeshima2 , N. Takahashi 3 and K.Takata2.
xDept. of Biol., Fac. of Sci., Nagoya
Univ., Nagoya, 2Radioisotope Center,

Nagoya Univ., Nagoya, 3 Dept, of Biochem. ,
Nagoya City Univ. Sch. of Med., Nagoya.

Previous studies showed that the neural
inducing effect of concanavalin A (Con A)
was ascribed to glycoproteins of the
presumptive ectoderm. To determine the
types of oligosaccharide in the glycopro¬
teins which are concerned in this signal
transfer, we studied N-linked oligosac¬
charides of st. 11 presumptive ectoderm.

Several peaks of oligosaccharides
released from ectoderm by glycopeptidase
treatment in the HPLC elution profile were
observed. Most of these peaks were also
observed in analysis of the adsorbed frac¬
tions obtained by using Con A affinity
chromatography. These oligosaccharides
with Con A affinity could be the candi¬
dates for the factors participating in
neural induction by Con A. Two of the
oligosaccharides obtained were plotted on
the map (5.97,5.87) (7.49,8.01) using two-
dimensional mapping technique (*) .

By comparing with 113 kinds of known
oligosaccharides on the map, these two
were thought to be oligo-mannose-type
having structural differences from the
known ones. * N. Takahashi et al.,1988
Anal .Biochem. 171,73-90 .

DB 60

INTERNALIZATION OF CON A MOLECULES IS NE¬
CESSARY FOR ITS DORSALIZING EFFECT ON THE
VENTRAL MESODERM.

M.R.M. Diaz1. S. Osawa1, T.C. Takahashi2,
K. Takeshima2 and K. Takata2. 1Dept. of
Biol., Fac. of Sci., Nagoya Univ. 2Radi-
oisotope Ctr., Nagoya Univ,, Nagoya. _

Addition of Con A resulted in dorsali-
zation of the ventral mesoderm. To deter¬
mine whether its mode of action is similar
to that in neural induction, the inhibito¬
ry effect of methyl-a-D-mannoside , a mole¬
cule which can abolish Con A effect, was
tested. No significant decrease in dorsal-
ization was observed. Also, explants
brought into contact with Con A immobilized
on Sepharose beads did not differentiate
into dorsal tissues. Apparently, Con A is
not capable of inducing dorsalization un¬
less internalized, possibly by pinocyto-
sis. Lowering of culture temperature from
20 °C to 6°C (thereby lowering pinocytotic
activity) decreased the incidence of dor-
ization. Microautoradiography using
I-Con A showed dense silver grains in
the cytoplasm of the 20 °C explants but
sparse in the 6°C ones. In both cases,
silver grains were observed on the surface,
with no significant difference in density.
SEM and TEM showed prevalence of pinocyto¬
tic pits and vesicles in the 20 °C explants
but not in the 6°C ones. Colloidal gold
conjugated-Con A was found localized in ve¬
sicles in the cytoplasm. Results suggest a
possible role of Con A-like molecules in
triggering the machinery for dorsalization.

DB 61

MORPHOLOGICAL CHANGES AND NEURAL-INDUCING ACTIVITY
OF NEWLY-MESODERMALIZED ECTODERM CELL
A.Miyagawa, A. S. Suzuki and T.Kuwana Dept. of Biol.,
Fac . of Gen . Edu . , KumamotoUniv . , kumamoto .

The presumptive ectoderm of Cynops gastrula was
mesodermalized by swimbladder of Carassius sp. and
acquires neural-inducing activity. In the present
study, morphological changes of mesodermal
ectoderm(mE) cells and interfaces between competent
ectoderm and mE were examined by TEM and SEM.

There was inner space in the control combinants
of both competent ectoderms even 3hrs after
combination and the inner space became larger with
cultivation. The mE cells shifted to be globular
shape and the mE cells kept to be in contact with
the competent ectoderm cells. These results
suggest that the mE cells acquire the capacity to
contact with the inner surface of competent
ectoderm during their mesodermalization process,
although the competent ectoderm couldn't closely
contact each other on the inner surfaces. SEM
observations showed that there were significant
differences in the inner surface between the
control and mE cell membranes and mE cells showed
migrating features. Both SEM and TEM observations
indicated that there were many interconnections
between the competent and mE ectoderm cells by
filopodia and cellular processes. We examined the
interface between neuro-ectoderm and chorda-
mesoderm of late gastrula and obtained same results
as that of mE.

It seems to be probable that appearance of
contact affinity with inner surface of competent
ectoderm in the mE cells is related to acquisition
of neural- inducing activity.

1264

Developmental Biology

DB 62

STUDIES ON THE NEURAL-INDUCING ACTIVITY OF
NEWLY -MESJDERMALI ZED ECTODERM. 2

A. Suzuki , J. Matsuda and J. Miyagawaz . "Dept. of
^iology, Fac. of Gen. Educ., Univ. of Kumamoto,
Dept, of Biology, Fac. of Science, Univ. of
Kumamoto, Kumamoto.

It has been shown that the competent ectoderm of
early Cynopus gastrula is mesodermal i zed by being
placed in contact with Carassius swimbladder. We
found that the newly-mesodermalized ectoderm(mE)
acquires the neural-inducing activity during the
early phase of its mesodermalization.

SEM and TEM observations on the interface be¬
tween the competent and mesodermalized ectoderms
showed that the mE had an ability to contact close¬
ly with the inner surface of the competent ecto¬
derm. What are the factors to be participated in
the neural-inducing activity being brought about
after mesodermalization ? It was inprobable that
the inducing factors were secretory substances,
because we couldn't observe the neural-inducing
activity in the conditioned medium in which the mEs
were cultured for 18hrs. The homogenate of mEs was
centrifuged at l,000g and the supernatant was cen¬
trifuged at 105,000g for 30 minutes. The precipi¬
tate was intactly wrapped with the competent ecto¬
derm. The neural induction was observed in both
cases of control and mesodermalized ectoderms. The
mE was killed by different means. The mE killed
with cold 90% ethanol showed a strong activity, but
aldehyde-killed mE showed very weak activity.

The present study suggests the possibility that
the neural-inducing activity of mE is due to the
contact activation of competent ectoderm cells by
molecular modification of cell membrane of
mesodermalized ectodermal cells.

DB 63

NEURAL INDUCTION AND PROTEIN SYNTHESIS OF
MESODERMALIZED ECTODERM

J. Matsuda and A. S. Suzuki. Dept, of Biol.,
Fac. of Gen. Edu ., Kumamoto Univ., Kumamoto.

The presumptive ectoderm of Cynops gastrula was
mesodermalized by swimbladder of Carassius sp., and
the mesodermalized ectoderm (mE) acquired very
strong neural-inducing capacity. In the present
study, we used two-dimensional SDS-polyacrylamide
gel electrophoresis method (2D SDS-PAGE) to analyze
the molecular changes in the mE. The 13,000g
supernatants were applied to 2D SDS-PAGE, there
were quantitative differences in a few protein
spots between the control and mesodermalized
ectoderms. In addition, the proteins of the neural
plate and mesoderm of early neurula embryo were
compared with those of the presumptive ectoderm on
the 2D SDS-PAGE. However, we couldn't found the
specific spots which seem to be participated in the
neural-inducing activity of mesoderm cells,
although there were quantitative differences in a
few spots between mesoderm and neural
plate/presumptive ectoderm.

It has been already shown that the neural-inducing
activity of mE was decreased by being treated with
protein-synthesis inhibitors, then newly-
synthesized proteins of mE were analyzed by the
fluorography on 2D SDS-PAGE. We could found
differences in 16 spots out of about 100 spots. 9
spots were heavily labeled in the control ectoderm
and 7 spots were heavily labeled in the mE. But the
specific spots were not observed. Therefore, We
conclude that the neural-inducing activity of the
mE might be evoked by molecular modifications or
transformations of cell surface rather than newly-
synthesized factors.

DB 64

EFFECT OF THE MESENCHYME ON THE DIFFEREN¬
TIATION OF THE DUODENAL AND GIZZARD EPITHE-
LIA OF THE CHICK EMBRYO.

S . Matsushita . Dept. Biol., Tokyo Women's
Medical College, Tokyo. _

Duodenal and gizzard epithelia of the
6-day chick embryo were cultured in recom¬
bination with the digestive-tract mesen¬
chyme, in the coelomic cavity of 3- to 3.5-
day chick embryo. Sucrase expression re¬
vealed by the immunofluorescence method and
mucus production detected by alcian blue-
staining were examined as the markers of
intestine- and gizzard-type differentia¬
tion, respectively.

Duodenal epithelium mostly differentia¬
ted into the intestine-type epithelium with
sucrase in the grafts of the intact duode¬
num and of the combination with the oeso¬
phageal or proventricular mesenchyme, while
the mucus-producing epithelium prevailed in
the grafts of the combination with the
gizzard mesenchyme. On the other hand, the
mucus-producing epithelium predominantly
appeared in the grafts of the intact giz¬
zard and of the combination of the gizzard
epithelium and the oesophageal or proven¬
tricular mesenchyme, while the intestine-
type epithelium with sucrase appeared in
wide area in the combination of the gizzard
epithelium and the duodenal mesenchyme.
Thus, duodenal and gizzard mesenchymes can
elicit mesenchyme-specific differentiation
in the epithelia of their neighboring
regions, that is, gizzard and duodenal epi¬
thelia, respectively.

DB 65

TRANSFILTER ANALYSIS OF THE INDUCTIVE
MESENCHYMAL INFLUENCE ON DIFFERENTIATION
OF STOMACH EPITHELIUM IN CHICK EMBRYOS.

K. Takiguchi-Hayashi1 , S. Yasugi1 and T.
Mizuno2. 'Zool. Inst., Univ. of Tokyo,
Tokyo; 2Fac. of Pharmac. Sci., Teikyo
Univ., Kanagawa.

To study the nature of transmission of
the inducing signal of proventricular
mesenchyme, we have done transfilter
experiments using gland formation and
pepsinogen expression as markers of
proventricular differentiation.

6-day or 4.5-day proventricular or
gizzard epithelium and 6-day

proventricular mesenchyme were recombined
with an intervening Nuclepore filter with
a pore size of 0. 2-3.0 //m and cultured

in vitro for 6 days. Gland formation and
pepsinogen expression were observed in

explants with a Nuclepore filter with a
pore size of more than 0.6 ^ m , but not
with a filter of 0.2 >um. As regards
gland formation there was no difference
between proventricular and gizzard
epithelia, but expression of pepsinogen

was lower in 6-day gizzard epithelium

than 6-day proventricular or 4.5-day
gizzard epithelium, indicating that the
reactivity of 6-day gizzard epithelium

had decreased rapidly before the signal
reached it.

These results suggest that the
transmission of inductive signal to
epithelium coming from the proventricular
mesenchyme is not achieved by a
long-range diffusion of signal substances.

Developmental Biology

1265

DB 66

DERMAL-EPIDERMAL INTERACTIONS IN A RECONSTI¬
TUTED SKIN MODEL OF RAT.

K. S. Nishikawa, A. Nishikawa and K.
Yoshizato. Dept, of Biol. Tokyo Metro¬
politan Univ., Tokyo.

A reconstituted skin model is prepared by
combining epidermal cells with the dermal
model that is composed of dermal
fibroblasts and collagen gels. This skin
model is thought to be useful for studying
the mechanism of regulation of collagenase
activity during wound healing at the
cellular level. We previously showed that
collagenase activity of the skin model is
higher than that of epidermal cells plus
that of dermal models. In this study, we
demonstrated that the increase of
collagenase activity of the dermal model
was induced by a factor! s ) in epidermal
cell conditioned medium ( ECCM ) . Collagenase
activity in ECCM was not affected by
repeated opperations of freezing and
thawing. This suggests that the factor is
not latant collagenase. Secretion of the
factor was highest at 6-7 days in culture
and was not influenced by the presence of
serum. The ultraf i 1 tration experiment
showed that molecular weight of the factor
was more than 10 K. Polarized epidermal
cells on a permeable collagen film secreted
the factor into the apical medium, sugges¬
ting that the polarity in secretion of the
factor might be important for regulation of
eollagenolysis of skin during wound
healing. ( Supported in part by Terumo Life
Science Foundation. )

DB 67

RETINOIC-ACID BINDING PROTEIN EXTRACTED
FROM FETAL RAT LIVER NUCLEUS.

Y. Fujino and T. Fujii. Dept, of
Pharmacol., Teikyo Univ. Sch. Med., Tokyo.

This study was designed to detect and
Isolate retinoic acid-binding protein (RA-
BP) in the nucleus of rat liver and to
clarify the role of retinoic acid on
hepatic cells. Retinoic acid-binding
capacity (RA-BC) was present in the
nuclear fraction obtained from fetal rat
liver (15, 18 and 20 days of gestation, as
far as examined) , using HPLC for the
detection of retinoic acid. The binding-
capacity was undetectable in infants (2,

6, and 12 days after birth) and adult
rats. This capacity was extracted from
the nuclear fraction by sonication in the
presence of 0.4 M NaCl . Nuclei were
completely disrupted in this condition,
when observed under a phase contrast
microscope. DNA, histones and phospho¬
lipids were not solubilized by this
treatment. RA-BC in the extract was eluted
in the void volume fraction from Sephadex
G-25 column. RA-BC was heat labile.

These results suggest that the ligand is
non-histone nuclear protein. The role of
RA-BP was unknown at present, but it may
participate in cell division, cell
differentiation or cell maturation in the
fetal liver.

DB 68

TROPOMYOSIN ISOFORMS IN CHICKEN GIZZARD
SMOOTH MUSCLE DEVELOPMENT.

M. Hosoya and T. Hirabayashi. Inst, of
Biol. Sci., Univ. of Tsukuba, Tsukuba

In the differentiation of chicken
gizzard smooth muscle, accumulation ratio
of tropomyosin (TM) to actin was reported
to be reasonably constant after hatching.
However, during the embryonic development,
it was relatively low.

In embryonic smooth muscle, we have
previously showed the presence of 10 TM
isoforms (E1-E10) in addition to smooth
muscle type isoforms (o^,/2). Relative
amounts of E1-E10 in total TM were
predominant at earlier embryonic stages.
These newly identified isoforms were
supposed to bring the TM/actin ratio up to
the constant ratio observed after hatching.

By an image analysing system, we have
determined the ratio of total TM to actin
in gizzard development. Although the ratio
obtained in embryonic gizzards was lower
than expected, it gradually became higher
along the embryonic development. Besides,
the ratio in thin-f ilaments from embryonic
muscle was lower than in total homogenate
of the embryonic tissue. These results
suggest that the regulation for coodinate
accumulation of actin and TM is only in the
development after hatching and that other
kind of regulation may be at the embryonic
stages .

DB 69

IMMUNOCYTOCHEMICAL STUDY ON MUSCLE
DIFFERENTIATION IN SEA URCHIN EMBRYO BY
USING PARAMYOSIN AS A MARKER PROTEIN.

M. Satoh and T. Obinata. Dept, of Biol.,
Fac. of Sci., Chiba Univ., Chiba.

Paramyosin is a good marker for the
study of myogenesis in sea urchin, since
this protein is a muscle-specific protein
which is widely distributed in invertebrate
animals including echinoderms. We prepared
a monoclonal antibody (SPM-2) to paramyosin
of sea urchin lantern muscle. This anti¬
body recognized paramyosin specifically in
sea urchin but did not react with the same
protein of starfish or molluscans, when
whole lysates of muscles or purified
paramyosin was examined by immunoblotting
combined with SDS-PAGE. With SPM-2,
appearance and assembly of myogenic cells
during embryonic development of sea urchin
were studied. Paramyosin-containing cells
were first detected in coelomic sacs at the
early prism stage and then they were
distributed and clustered around oeso¬
phagus. Later, they encircled oesophagus to
form several layers of fibrous ring-
structures. Paramyosin was detected in
anus but not around either stomach or
intestine. SPM-2 positive cells were also
stained by rhodamine-phalloidin, an anti-
actin agent, although this agent stained
the other regions of embryos as well less
brightly, probably because of the presence
of actin in non-muscle cells.

1266

Developmental Biology

DB 70

APPEARANCE AND NATURE IN CULTURE OF EMBRYONIC
MUSCLE CELLS OF THE STARFISH, ASTER I AS AMURENSIS.
Y. Kawahara, H. Kaneko and M. Dan-Sohkawa.

Dept, of Biol., Osaka City University.

We have reported in the 57th conference of
this Society (1986) that a pure culture of
embryonic mesenchyme cells is obtained by treating
dissociated mid gastrula cells for 24 hours with
0.6 M glycine in Ca7 Mg* free sea water.

We have found, recently, that cells showing a
spontaneous contraction and relaxation make their
appearance after 6-8 days of this culture when the
stage of the material embryo was raised to late
gastrula and the treatment period was shortened to
12 hours. The majority of these cells are slender
in shape 100-200 jjiti long. All of them contracts
within 1-2 seconds when acetylcholin (Hf^M) is
dropped over them from above the culture medium.
These cells contain bundles of thick (10-20 nm)
and thin (5-8 nm) filaments which are not alingned
in a sarcomeric pattern. Staining the cell by
anti nematode-myosin polyclonal antibody and NBD-
phallacidin revealed that they are rich in myosin
and F-actin. These cells were determined to be
larval muscle cells by their resemblance in shape,
the features of the thick filaments and the
reactivity to acetylcholin.

DB 71

MOLECULAR CLONING OF cDNAs SPECIFIC TO
DIFFERENTIATING MUSCLE CELLS OF ASCIDIAN
EMBRYOS .

K.W.Makabe* 1 , J. Suzuki1 , T . Deno2 3 and
N . Satoh' . 1 Dept . of Zool . , Fac . of Sci . ,

Kyoto Univ., Kyoto, 2 Dept . of Biol., Fac.
of Educ . , Osaka Kyoiku Univ., Osaka.

The B4. 1-line muscle cells of 'the
ascidian (Halocynthia roretzi) embryo are
characterized by the ability to
differentiate autonomously. To
investigate the molecular mechanisms
involved in specification of the muscle
cells from the viewpoint of activation of
genes expressed in tissue-specific manner,
we attempted to clone three cDNAs which
are expressed only in the differentiated
muscle cells. Our cDNAs are as follows.

1. cMll: Myosin heavy chain cDNA , which
was isolated using our muscle-specific
monoclonal antibody (Mu-2) from an
expression library. This cDNA is 1.6kb in
length .

2. cA12: Muscle actin cDNA, which gave
the strongest signal in the hybridization
with a muscle actin cDNA of Styela plicata.
This is 1 . 3kb in length.

3. c3117e: Acetylcholine esterase
cDNA, which was obtained by screening -a
cDNA library with two groups of chemically
synthesized oligonucleotides corresponding
to amino acid sequences conserved beyond
species. This is 1.4 kb in length.

DB 72

DEVELOPMENTAL CHANGES OF CYT0KERATIN EXPRESSION IN
EMBRYONIC CHICK ORGANS.

Y. Nishizawa & N. Shiojiri. Biol. Inst., Shizuoka
Univ., Shizuoka.

Cytokeratin is one of cytoskeletal proteins of
epithelial cells and has about twenty subtypes. The
expression of them in adult organs depends on the
type of epithelial tissues. Epithelial tissues in
embryonic organs change their architecture with
development. The present study was undertaken to
analyze how cytokeratin subtypes are expressed with
the development of epithelial tissues.

Chick embryos at 10 to 20 days were used. The
antiserum was made in guinea pig using one of crop
cytokeratins(48kD) and leg skin epidermal
cytokeratins as antigen. The antiserum showed broad
crossreactivity with other subtypes of cytokeratin.
Indirect immunofluorescence and Western blotting
were carried out to examine cytokeratin expression
in embryonic organs.

All epithelia of leg skin, crop, small intestine
and liver were stained immunof 1 uorescetly.
Cytokeratin of 48kD appeared in the crop at 14 days
when the epithelium stratified. Leg skin epidermis
began to express cytokeratin of 48kD at 12 days
when the stratification of the epithelium occured,
and cytokeratin of 64kD at 16 days. Small intestine
and liver, not stratified epithelium also in the
embryo, did not express these cytokeratins. These
results suggest that cytokeratin expression in the
embryonic organs change during development, and

cytokeratin of 48kD seems to be specific for the

stratified epithelium through development.

DB 73

LOCALIZATION OF ELASTIN mRNA IN MOUSE ORAL
REGION BY IN SITU HYBRIDIZATION WITH RIBO-
PROBES .

T. Yamaai, S. Noji^ , E. Koyama^ and S.
Taniguchijg , Depts. of Oral Anat., Biochem.,
and Maxillofacial Surg., Okayama Univ. Dent.
Sch . , Okayama .

We have reported that tbe distribution of

buccal mucosa type cells correlates with
that of the elastic fiber in the proper lay¬
er of the oral cavity. To clarify the
correlation between the cytodifferentiation
and elastic fiber formation in the oral re¬
gion of an immature mouse, the localization
of the elastin mRNA was analyzed by in situ
hybridization using the ribo-probes with aim
to detect cells synthesizing the elastin.

One week old male C3H mice were fixed by
perfusion from the left ventricle with 4%
paraformaldehyde. 35S-labeled anti-sense
and sense RNA probes (= 150 b in size) were
both synthesized from cDNA cloned from a
chick aorta (kindly supplied by Dr. Tajima) .
In situ hybridization was carried out on 5/^m
thick serial sections and subsequently, they
were made into autoradiograms. Exposed
silver grains were observed on the fibro¬
blasts of the proper layer of the alveolar
side of the dental lamina and of the
transverse palatal ridge. The grains were
also observed on the upper half of the loose
connective tissues paralleled the epithelium
of the vomeronasal organs and arterioles.
No signal grains of the sense RNA probe were
found on the same region mentioned above.
These results demonstrate that the produc¬
tion of the elastin by fibroblasts of the
oral proper layer is closely related to the
cytodifferentiation of the oral epithelium.

Developmental Biology

1267

DB 74

MONOCLONAL ANTIBODY STAINING NERVE FIBERS
OF EMBRYONIC RAT BRAIN.

T. Seki and Y. Arai . Dept. of Anat . ,
Juntendo Univ. Sch. of Med., Tokyo

Monoclonal antibodies(MAbs) against
cell suspension, containing broken cells
of 6-day-old rat hypothalami were produced
and screened by immunohistochemically
staining A% paraformaldehyde-fixed frozen
sections of adult and 6-day-old rat
hypothalamus. In a total of about 700
hybridoma lines, 16 reacted with glial
elements and 12 reacted with cell bodies
of neurons or nerve fibers. Four reacted
with cell nucleus of neurons, glia and
ependymal cells. They also stained the
cell nucleus of testis, kidney and adrenal
medulla strongly and liver weakly.
Further immunohistochemical study revealed
that MAb 5F7 , one of these four MAbs
originally screened by the stainability of
cell nucleus, strongly stain the nerve
fibers of embryonic rat brain, whereas
adult brain were weakly stained by this
MAb. In the developing cerebral cortex,
the nerve fibers stained by MAb 5F7
appeared in the intermediate zone, which
becomes white matter in future. Strongly
immuno re active fibers were observed in
the rat brain from days 16 to 20 of
gestation. After birth, the staining
intensity of cortical fibers was
progressively reduced. These results
suggest that MAb 5F7 may have a potential
for identifying the nerve fiber in the
early developmental stages of rat brain.

DB 75

Occurrence of larval leukocytes that are
not derived from blood islands in Xenopus
laevis

H.Ohinata, S.Tochinai and Ch.Katagiri.
Zool. Inst., Fac. of Sci., Hokkaido Univ.

Immunohistochemical observations employing
the monoclonal antibody (XL-1 ) recognizing
all types of leukocytes of X^ laevis
revealed that XL-1 + cells are distributed
in the mesenchyme of st. 34/35 embryos
before the appearance of any lymphocytes.
Experiments were made to determine whether
or not these leukocytes are originated
from the ventral blood islands (VBI), like
lymphocytes and erythrocytes are. When
the head region cut at the level of gill
region at st. 22/23 were explanted and
cultured for about 60hr . (until st.43 in
control embryos), a significant number of
XL-1 + cells were differentiated. The
orthotopic VBI grafts of st. 22/23 X .
laevis x X_i, borealis hybrid (LB) embryos
were made to the stage matched X_^_ laevis
embryos. Examinations of host individuals
after about 70 hr. culture (st. 44/45)
showed that in all experimental larvae the
proportion of XL-1 + cells possessing a LB
graft marker was always significantly
lower than that of erythrocytes having the
same marker. These results strongly
indicate that there is a significant
population of larval leukocytes that are
not derived from the VBI region, but are
widely distributed in the mesenchyme
throughout the embryonic body.

DB 76

PRENATAL DEVELOPMENT OF THYROID C-CELL IN
CULTURE.

I. Nishiyama and T. Fujii. Dept, of Pharmacol.,
Teikyo Univ. Sch. of Med., Tokyo.

Prenatal development of rat thyroid C-cell was
immunocytochemically investigated jn vitro. The
thyroid glands, taken from the embryos at various
gestational ages, were enzymatically dispersed and
cultured in Dulbecco»s modified Eagle medium
supplemented with 5% fetal calf serum and 4% rat
embryo extract. The appearence of C-cell was
examined by using immunoreactivities for
calcitonin (CT), CT gene-related peptide (CGRP)
and somatostatin (SRIF). In the culture of the
dispersed thyroid cells from Day 14* embryos, CT-,
CGRP- and SRIF-immunoreactive cells first appeared
at Day 2 in culture in four cases of seven
independent experiments. Prolonged incubation
period (3-5 days) did not affect the results.
C-cells appeared without exception in the culture
from embryos at Day 15 or later stages. The
appearence of CT- and CGRP-immunoreactivities was
inhibited by cycloheximide (Ipg/ml) or puromycin
(IpM), but not by actinomycin D (lOOng/ml) in the
cultured cells from Day 15 embryos. These results
show that C-cell precursors taken from the embryos
were able to differentiate to produce these
peptides _in vitro even after enzymatically
dispersed.

Supported in part by a research grant to T.F.
from The Mitsubishi Foundation.

* The day on which sperm were present in the
vaginal smear was designated as Day 0 of
gestation.

DB 77

THE REGULATION OF MELANOGENES IS DURING
HAIR CYCLE OF MICE.

Y . NISHITANI 1 , K . I TO 1 , S . H 0 R I U C H I 1 ,
T.MORITA1, Y.KOSHIDA1, H. YAMAMOTO2,
T . TAKEUCHI 2 , 1 Dept . of Biol., Coll, of

Gen. Educ. , Osaka Univ., Toyonaka, 2Biol.
Inst., Tohoku Univ., Sendai.

Melanogenesis during the hair cycle has
been studied with the nude mouse (C3H/He-
nu) , which changes the melanogenic pattern
in the skin with passage of the time.

As the black area of the nude mouse
skin has the typical size and the number
of hair bulb, the number of melanocytes in
the black area might be similar to that of
the wild type. The ultrastructure of
melanocytes was identical to that in the
wild type. These results suggest that the
phase in the hair cycle of black area is
in anagen. On the other hand, the size
and the number of hair follicles in the
white area suggest that the phase of this
area may be in catagen or telogen.
Structures of those melanosomes enclosed
with the membrane which were found in the
white area's melanocytes were comparable
to the autophagic vacuoles reported by
Feeney-Burns and Mixon (1979).

It would be conclusive that the nude
mouse is very useful for the study of
melanogenesis during the hair cycle.

1268

Developmental Biology

DB 78

CELL DEATH IN WING MORPHOGENESIS OF LEPI-
DOPTERAN INSECTS.

A.Yoshida and K.Aoki. Life Sci. Inst., So-
phia Univ. , Tokyo. _ _

Adult wings of lepidopteran insects are

formed by degeneration of some cells along
the margin of the pupal wing disc (Suffert,
1929). We examined morphological difference
between cells in the degeneration region
and in the differentiation one, and studied
development of marginal hairs after barrier
insertion into the disc. A small white
cabbage butterfly, Pieris rapae , was used.

Immediately after pupation, a pupal
wing disc was examined with SEM. Polygonal
epidermal cells are closely packed both in
the degeneration region and in the dif¬
ferentiation one. Each cell is larger in
the degeneration region than in the dif¬
ferentiation one. A whole mount of the disc
was stained with Feulgen and was examined
with a light microscope. A nucleus of each
cell is also larger in the degeneration
region than in the differentiation one. A
small piece of alminum foil was inserted
into the degeneration region immediately
near the margin of the differentiation
region. Marginal hairs of the emerged
adult wing are much shorter in the area
which was adjacent to the inserted alminum
piece than in the intact area. This
results suggests that the degeneration of
the marginal part of the pupal wing disc
may be necessary for the normal develop¬
ment of the marginal wing hairs.

DB 79

REGENERATION CAPACITY AND PATTERN
FORMATION OF WING IMAGINAL DISC IN
SILKWORM, Bombyx mori
Y. Higaki and T. Ohtaki

Department of Biology, Faculty of Science,
Kanazawa University, Kanazawa 920.

In order to examine the regeneration
capacity and process of pattern formation,
we removed surgically a parts of wing
imaginal discs of Bombyx mori in situ on
different developmental stages. The
results showed that wing discs of
penultimate instar larvae possessed a high
regeneration capacity, but those of last
instar larvae have lost the capacity
during day 1 through day 3. Moreover,
duplication of posterior pattern was
observed in animals of which anterior or
apical parts of the discs was removed at
the penultimate instar. The time losing
regeneration capacity well corresponds to
the time of commitment change of the discs
to pupal type.

DB 80

TIBIA TO FEMUR RATIOS OF UNALTERED AND
REGENERATED LEGS OF THE STUMPY MUTANT OF
THE GERMAN COCKROACH.

A. Tanaka1 and M.H.Ross . 1Dept. of Biol.,
pFac. of Sci., Nara Women's Univ., Nara,
Dept, of Entomol . , VPI&SU, Virginia, USA.

In wild type, the ratio of tibia length
to femur length (T/F ratio) is low in the
foreleg, high in the hindleg, and interme¬
diate in the midleg throughout stages.
After autotomy at the trochanterof emoral
articulation, regenerates from foreleg
coxae show intermediate T/F ratios between
the normal foreleg and midleg, and those
from the hindlegs, intermediate between the
normal hindleg and midleg. those from the
midlegs are unaltered in the ratios. The
relative length of stumpy ( sty) legs to
wild- type legs was around 80%, showing some
fluctuation with the stage and the type of
legs. The T/F ratios of unoperated sty
fore-, mid-, and hindleg were, respec¬
tively, similar to those of regenerated
wild-type. This suggests that sty may have
primitive legs that are not fully differen¬
tiated from midleg-like nonspecific forms.
The T/F ratios of regenerated sty legs were
similar to those of the unoperated sty
when the legs were autotomized in early
instars. When the legs were autotomized in
middle and late instars, the ratios of sty
regenerates increased in all legs. In the
adult regenerates, the ratios of the sty
fore-, mid-, and hindleg almost reached
those of the sty midleg, sty hindleg, and
wild-type hindleg, respectively.

DB 81

DISTRIBUTION OF AV-1 ANTIGEN IN THE
MESODERM OF CHICK LIMB BUD

Y. YOKOUCHI , K.OHSUGI , and H.IDE. Biol.
Inst., Tohoku Univ. , Sendai .

AV-1 antigen is a position specific
molecule found in early developmental ly
chick, limb buds. Using immuno histo-
chemical method, we investigated the
distribution pattern of AV-1 antigen with
regard to dorsoventral axis. At stage
19, the antigen was detected only in the
mesoderm beneath ventral ectoderm. At
stage 21 , the AV-1 positive region spread
to core area, and at stage 24, expression
was found in all the anteroventral area of
distal mesoderm. At stage 26, it remained
only in the subdistal core area.

To elucidate the mechanism of position
specific expression of AV-1 antigen along
anteroposterior axis, we examined the
relationship between AV-1 expression and
ZPA factor. When the level of the ZPA
factor decreased in the anterior region by
the insertion of impermeable barrier or by
the removal of ZPA, AV-1 expression
remained for 24 hours. When the level of
the factor increased by grafting ZPA
fragment into the posterior distal margin
of anterior 1/3 stump, which was AV-1
positive, the expression disappeared within
24 hours, but reappeared after 48 hours
with the inclease of stump length in
anteroposterior direction.

Developmental Biology

1269

DB 82

THE DISTRIBUTION OF RETINOIDS IN THE LIMB
BUD: ANALYSIS WITH MONOCLONAL ANTIBODY
K.TAMURA. K.OHSUGI. and H. IDE. Biol. Inst..
Tohoku Urtiv. , Sendai. _

In chick limb bud, locally applied retinoic
acid(RA) causes striking effects on digit
pattern. It has been recently shown that retinol
and retinal are converted into RA in the limb
bud. So the distribution of retinoids other
than RA seems to be involved in limb pattern
formation.

To study the distribution of three retinoids
in the limb bud, we made a monoclonal antibody
against retinoids. Antigen used was RA
conjugated BSA (Bovin Serum Albmin) prepared
with acid anhydride reaction. Screening was
carryed out with ELISA. The antibody obtained
bound to three retinoids. Fluoroescence
microscopic observation with FITC-antibody
revealed a different pattern of retinoid
distribution in stage 20-24 limb buds. At stage
20, retinoids were detected only in AER and
underlying mesoderm. At stage 22, strong
binding to the antibody was observed in dorsal
mesoderm. As development proceeds further,
binding remained ectoderm and non-chondrogenic
mesoderm. This antibody, however, showed no
effects on limb bud mesodermal cells treated
with RA in vitro.

The role of these retinoids in pattern
formation was discussed.

DB 83

PATTERN FORMATION IN THE CULTURE OF
DISTAL MESODERMAL CELLS OF CHICK LIMB BUD
H.Ide. Biol. Inst., Tohoku Univ. , Sendai.

Distal mesoderm of stage 20-23 limb
bud is in an undifferentiated growing
state. When the mesoderm was dissociated
and cultured in FI 2 medium containing 1%
FCS, proliferation and gradual
differentiation into cartilage occurred.
Retinoic acid (RA) at 5-25 ng/ml
drastically promoted the growth and
cartilage nodule formation in the distal
mesodermal cells. At the same time,
especially at higher concentration of RA,
the differentiation into fibroblastic
cells also occurred. These results seem
to be a cellular basis for RA-induced
duplicate formation.

When the distal mesodermal cells were
cocultured with retinal pigmented
epithelium, local inhibition of
chondrogenesis in the culture of limb bud
cells occurred. In the rectangular area
of the mesodermal cell culture which had
been surrounded by the pigment cells, a
row of cartilage nodules was formed in
the center. In the wedgy area of the
mesodermal cell culture, Y-shaped
cartilage pattern was formed. These
pattern formation may be a cellular basis
for normal limb development.

DB 84

THE EFFECT OF HALOTHANE GAS OH THE QUAIL
EMBRYO.

H.Hasegawa, H. Tanaka and K.Nonoyama.
Dept, of Biology, Aichi Univ. of Educ.,
Kar iya .

Previously the authors have examined the
influence of an anesthetic gas , halothane
gas, on the liver of hatched chicken in
terms of changes in serum LDH , GOT, GTP ,
YGTP, liver LDH and electron microscopic
pictures. Then theauthors investigated
that 3.0 % of halothane gas influenced on
an abnormal variation in morphological
differentiation of brain, eyes, legs and
wings , particularly to the amount of mela¬
nin pigmentation in the eyes of quail em¬
bryo , however the gas concentration was low
( 0. 5-1.0 % ) revealed skeletal formation
normally. in this experiment, it was found
that 1.0 % of halothane gas influenced on
the growth of quail embryo as to the anes¬
thetic period of time, and the body weight
of the anesthetized quail embryo for 11
days was corresponded to the weight of 6
days of the control, non-anesthetized
quail embryo. The formation and hardening
of leg-bone of were developed normally as
same as the the control above mentioned.
These facts indicated that halothane gas
did not influenced on the abnormal develop¬
ment of the quail embryo, differ to the
embryo anesthetized with enfluren gas which
induced to the malformed development of the
quail embryo.

DB 85

CELL PROLIFERATION DURING BUD FORMATION
OF THE QUAIL UROPYGIAL GLAND
Y. Fukui Dep. Biol., Tokyo Women's
Medical College, Tokyo

A localized increase of labelling
index (L.I.) was observed in bud-forming
epithelium of the uropygial gland of the
Japanese quail embryo through computer-
aided analysis. Proliferative activity of
the uropygial epithelium was examined by
autoradiography of t r i t i a ted- thymi d i ne
incorporation. Positions of labelled
nuclei and of non-labelled nuclei in the
epithelium were recorded by a digitizer,
reconstituted as two-dimensional com¬
puter-graphics and analyzed.

At day 10 no glandular bud was formed
yet anywhere in the epithelium and the
distribution of proliferative activity in
the epithelium was not differential. On
the following day the peripheral buds
were being formed at the externo- 1 ateral
end of the basal luminal epithelium; L.I.
increased locally at these buds. At day
12 a localized increase of L.I. took
place also in the peripheral buds of the
median end; extremely high labelling
indices were observed at the tip of
peripheral buds and developing central
buds .

1270

Developmental Biology

DB 86

ACCUMULATION OF COLLAGEN III AT THE CLEFT POINTS OF
DEVELOPING M(j)USE SUBMANDIBULAR EPITHELIUM.

Y. Nakanish^ , ^H. Nogawa , Y. Hashimotci and
T. Hayakawa . Dept. Chem. , Nagoya Univ., Nagoya,
^Biol. Lab., Coll. Arts and Sci., Chiba Univ., Chiba
Dept. Biochem., Aichi-Gakuin Univ., Nagoya. _

Collagens are widespread extracellular matrix

components in connective and epithelial tissues. Our
previous studies demonstrate that collagen is essen¬
tial to branching morphogenesis of developing mouse
submandibular epithelium. Moreover, an interstitial
collagenase from bovine dental pulp, which was shown
to degrade collagens I and III but not IV and V, in¬
hibited cleft formation, indicating that interstitial
collagens were required for epithelial branching.
However, Kratochwil et al. proved that collagen I
was not involved in the epithelial branching. We,
therefore, tried to identify the type of collagen at
epithelial-mesenchymal interfaces immunohistochemi-
cally. Anti-collagen I antibody stained the extra¬
cellular matrix of the mesenchyme and the interface
of the 12-day gland uniformly. Staining with anti¬
collagen III antibody was prominent at the interface
of the mid 12-day gland. At late 12-day stage, col¬
lagen III appeared to be concentrated at slightly
indented epithelial site. At early 13-day stage,
very intense staining was seen at every definite
cleft point of the epithelium and at the bottom of
the narrow, deep clefts. In contrast, the distribu¬
tion of collagen I was rather uniform as in the 12-
day gland. Staining with anti-collagen IV and V
antibodies showed no specific accumulation in the
early clefts. These results strongly' suggest that
collagen type essential to epithelial branching of
mouse embryonic submandibular epithelium could not
be type I but III.

DB 87

REGIONAL DIFFERENCES IN EPIDERMAL KERATIN
ISOFORMS OF XENOPUS LAEVIS TADPOLE.

T. Kinoshita, H. Takahama and F. Sasaki.
Dept, of Biol., School of Dent. Med.,
Tsurumi Univ., Yokohama.

Regional differences in developmental
changes of epidermal keratin isoforms were
electrophoretical ly compared between back
and tail skins of Xenopus 1 ae v is tadpole
during metamorphosis. In the back skin,
4 9 kD and 63kD keratins (adult type)
appeared at prometamorphosis, and 44kD,
58kD, 59kD and 60kD keratins (larval type)
disappeared during the climax of meta¬
morphosis. In the tail skin, larval type
keratins were found during the whole
larval stages. A small amount of adult
type keratins was recognized in the tail
skin at the final stage of metamorphosis.
Distribution of keratins was examined by
indirect immunof luorescent stainings using
polyclonal antisera raised against 44kD
and 63kD keratins. The 44kD keratin was
located in the tail epidermis during the
whole larval stages. In the back skin,
however, the 44kD keratin became
undetectable in the upper layers of
epidermis during the climax and completely
disappeared in adult skin. The 63kD
keratin appeared in the back epidermis
during prometamorphosis and continued to
exist until the adulthood. In the tail
skin, the 63kD keratin was detected in
several epidermal cells at the final stage
of metamorphosis.

DB 88

IMMUNOCHEMICAL STUDIES ON CATHEPSIN D-LIKE
ENZYME IN TADPOLE OF RANA CATESBEIANA.
Y.Fujii, M.Taguchi, K.Kobayashi and
S . Horiuchi .

Life Sci. Inst., Sophia Univ., Tokyo.

The activity of cathepsin D-like enzyme
increases in tail and liver of metamor¬
phosing frog. Molecular weight of the
enzyme was about 42,000. Purified enzyme
from liver was injected into rabbit to
obtain anti-serum.

By ELISA method and immunoblotting , IgG
fraction of the anti-serum specif ically
reacts to cathepsin D-like enzyme purified
from both liver and tail. This fraction
also inhibits the enzyme activity .

By immunohistochemical techniques,
localization of cathepsin D-like enzyme
in tail and liver was studied. In liver
from stage 23 tadpole, macrophage-like
cells were intensely stained and
hepatocytes were weakly stained. Degrading
area of muscle, macrophage- like cells in
epidermis and dermis in tail were intensely
stained at metamorphic stage.

DB 89

MYOTOMAL MUSCLES IN THE TRUNK AND TAIL IN
Xenopus laevis EMBRYOS: SENSITIVITY OF
THE CULTURED MUSCLES TO THE THYROID
HORMONE .

H. Takahama, T. Kinoshita and F. Sasaki.
Dept, of Biology, School of Dental
Medicine, Yokohama.

Morphological changes in the trunk
and-tail myotomes, explanted from Xenopus
laevis embryos (stage 25-26) and cultured
for 7, 10 and 14 days respectively in the
culture medium without thyroid hormone,
were examined by phase contrast, light
and electron microscopes. After the
administration of the hormone (T3,10_6M),
degeneration occurred in the muscle
cells of the tail somites within 4-5
days. Z bands disappeared and the
sarcoplasmic reticulum dilated in these
cells. There was no evidence that the
degenerated muscle cells were
phagocytosed by other cells like
macrophages. In the trunk myotomes,
however, some muscle cells remained
intact after the T3 administration.
Myogenesis and my o f i br i 1 logenesi s
occurred in both trunk and tail muscle
cells which were explanted in the culture
medium with T3. Degeneration delayed in
these explants and it took 9 days for the
degenerating cells to be observed.

Developmental Biology

1271

DB 90

HISTOCHEMICAL STUDY ON THE FORMATION OF THE
DERMAL CHROMATOPHORE UNIT IN TREE FROGS.

M. Yasutomi . Lab. of Biol., Aichi Medical
Univ. Aichi.

The formation of the dermal chromato-
phore unit (DCU) , which is formed during
metamorphosis and is a factor of rapid
color change in tree frogs, was studied
histochemically . DCU is composed of three
types of chroma tophores , melanophores ,
xanthophores , and iridophores. In tadpole
there are two distinct types of melano¬
phores, namely the epidermal and dermal
melanophores. Both melanophores and also
xanthophores, by migrating at metamorphic
stage, reached the subdermal region and
participated in the formation of the DCU.
While, most of iridophores constituting
the DCU were not supplied from larval
iridophores but differentiated newly from
iridoblasts at metamorphic stage.

The role of the extracellular matrix
(ECM) in controlling chroma tophores migra¬
tion has received considerable attention.
So, the distribution of gly cos aminogly cans
(GAG) in the dermis was examined. GAG were
identified using alcian blue and PAS
stainings. The ECM of the dermis was filled
with acidic and PAS positive GAG. But the
types and amounts of the GAG in the dermis
do not change during metamorphosis. These
facts suggest that the GAG which were
identified with these staining methods do
not affect the migration of chroma tophores
at metamorphic stage.

DB 91

LOCALIZATION OF 14-NM FILAMENT PROTEIN
(49K PROTEIN) JN CONJUGATING TETRAHYMENA
PAIRS BETWEEN C STRAIN AND NORMAL STRAIN
I. Takagi, O. Numata and Y. Watanabe.
Inst, of Biol. Sci . , Univ. of Tsukuba,
Tsukuba.

We have shown that 14-nm filament pro¬
tein (49K protein) is involved in some
nuclear events of fertilization during
conjugation. By double staining immuno¬
fluorescence, we observed that 49K protein
coexisted with tubulin during the third
prezygotic division, migratory pronucleus
transfer across a cell-cell junction and
zygote formation by pronuclear fusion. On
the other hand, at the post meiotic stage,
49K protein with no tubulin was localized
in network structures around the cell-cell
junctior^. In the conjugating pairs be¬
tween C strain and normal strain, we
could not observed typical immunofluores¬
cence^ patterns of 49K protein and tubulin
in C strain side, suggesting that colo¬
calization of 49K protein and tubulin
depends on the existence of pronuclei at
fertilization. On the other hand, at the
post me ip tic stage, network structures
composed of, 49K protein ^lone were ob¬
served in both normal and C strain sides.
It follows from these that appearance of
the network structures is independent on
the existence of pronuclei and micro¬
tubules. We discuss a possible role of
the network structures.

DB 92

ULTRAS TRUCTURAL CHANGES IN DEGENERATING
MICRONUCEI DURING CONJUGATION IN
PARAMECIUM TETRAURELI A

Y . Yashima . Dept .of Biol.,Sch. of Lib. Arts
and Sci.,Iwate Med. Univ., Morioka.

When two micronuclei complete meiosis
during conjugation, eight haploid micro¬
nuclei are produced. During the micro-
nuclear migration, one of these nuclei
located in the paroral region survives
and the remaining nuclei located outside
the paroral region degenerate. To better
understand the morphological changes asso¬
ciated with the micronuclear degeneration
process, conjugation was induced between
st. 51 O and E, and degenerating nuclei
were compared with the surviving nucleus.
According to TEM observations 3 - 4 hrs

after the beginning of conjugation, no
differences among micronuclei were present
but, 5 hrs after the beginning of con¬
jugation, distinct differences among mi¬
cronuclei were observed. In contrast to
the surviving micronucleus which had an
intact nuclear membrane, chromosomes, and
microtubules, the degenerating nuclei dis¬
played an indistinct nuclear membrane
which was discontinuous in one or two
places, indistinct microtubules, and lyso-
somes surrounding these areas of dis¬
continuity. Furthermore, cytoplasm entered
into one of the degenerating nuclei at the
points of discontinuity, karyoplasm sepa¬
rated from the nuclear membrane, and the
chromatin condensed.

DB 93

IS THERE ANY ANTIGEN WHICH APPEARS SPECIF¬
ICALLY IN THE SEXUAL REPRODUCTION PROCESS
IN PARAMECIUM CAUDATUM?

A.Yanagi and H. Yamamoto, Biol. Inst., Fac.
of Sci., Tohoku Uni v ., Sendai .

To understand molecular mechanism controlling
conjugation in _P. caudatum, identification of the
materials produced only during the conjugation is
useful. We are trying to detect specific materials
for the conjugation process by the monoclonal anti¬
body. We obtained a monoclonal antibody which
recognizes only some material in the nucleus during
conjugation. The nuclear antigen recognized by the
antibody begins to appear in micro- and macro¬
nucleus at the micronuclear stage II or III of
early conjugation process (4 hours after mating
reaction). In the micronucleus, the nuclear anti¬
gen exists until the crescent stage (8h). The
micronuclear antigen disappears after the crescent
stage and reappear after the 3rd postzygotic
division (25h). It seems to appear in the 4 macro-
nuclear anlagen about 30 hours after mating reac¬
tion. In the macronucleus, the antigen persists
for a while after the stage of macronuclear frag¬
mentation (about 35h) and then disappears. The
strong fluorescence seen in the nucleus by the
indirect immunofluorescence stainning suggests
production of a new component in the nucleus or
modification of the nuclear component by acetyla¬
tion, phosphorylation or others during the conju¬
gation process.

1272

Developmental Biology

DB 94

Micronuclear division inducing factor at
meiosis of Paramecium caudatum; an analysis
by micromanipulation.

K.Mikami, Res. Inst, for Sci. Educ., Miyagi
Univ. of Educ., Sendai. _

The result of macronuclear elimination
from both cells of a mating pair revealed
that factor(s) necessary for micronuclear
division of meiosis was given from the
macronucleus (Mac) at about 8 hrs of
conjugation, about 1 h before the first
division. When Mac was eliminated 2-2.5 hrs
of conjugation from a cell of a pair, the
micronucleus (Mic) in the cell skipped over
crescent stage and then divided twice.
However, none of the division products went
into paroral cone and disintegrated, while
in non-operated pair one of the 4 products
went into paroral cone and other 3 degene¬
rated. Then, the pair separated as well as
non-operated pairs. When the stage of Mac.
elimination was shifted to 3-3.5 h, the
crescent formation and all of Mic events
occurred before pair separation. So far,
the Mic division lacking crescent formation
is considered to be mitotic rather than
meiotic. If this is the case, we can
conclude that the f actor (s) for meiosis is
the same as that for mitosis.

When Mac was eliminated from both cells
of a pair at about 9 hrs of conjugation.

Mic repeated division and produced 5-1 6
nuclei though the normal number is 4 in
non-operated. This suggests that the
factor (s) has an ability to induce division
repeatedly until the information is given
from Mac to stop Mic division.

DB 95

ANALI T I CAL STUDIES BY MICROINJECTION OF A
MUTANT AFFECTING SURFACE ANTIGEN EXPRESSION
IN PARAMECIUM TETRAURELI A .

S. Koizumi and S.Kobayashi. Dept, of Biol.,
Miyagi Coll, of Educ., Sendai.

The mutant d48, which cannot express the
A antigen, does not contain the A gene in
its macronucleus, but contains in its
micronucleus, because the cytoplasm affects
the development of new macronuclei. It was
demonstrated that transfer of nucleoplasm
and cytoplasm from 51, wild type, caused
d48 to express A. These results show that
some factor which enables the A gene to be
passed to the newly forming macronuclei is
present in the 51 macronucleus throughout
the vegetative stage and in the cytoplasm
during the sexual stage. To obtain a clue
whether this factor is the product of the A
gene or not, a plasmid pSA14SB containing
the A gene was injected into d48 macro¬
nucleus at the 2nd and the 10th cell cycle.
Over 20% lines among both cell cycles of
recipient cell lines expressed A and main¬
tained A through successive fissions; all
transformants except 1 stably expressed A
even after the next autogamy. Injections of
cytoplasm from 51 serotype B cells during
autogamy into the cytoplasm of d48 showed
no transformation to express A in
d48(0/39) . These results reveal that the A
gene produces, or causes the cell to
produce, the cytoplasmic factor necessary
for proper processing of micronuclear DNA
into macronuclear DNA at auto-gamy.

DB 96

HOST GENES CONTROLLING MAINTENANCE OF A
MACRONUCLEUS-SPECIFIC SYMBIONT HOLOSPORA
OBTUSA OF PARAMECIUM CAUDATUM.

M. Fujishima and Y. Mizobe. Biol. Inst.,
Fac. of Sci., Yamaguchi Univ., Yamaguchi.

The gram-negative bacterium Holospora
obtusa is a macronucleus-specific symbiont
of Paramecium caudatum. This bacterium can
infect the macronucleus of P. caudatum, P.
multimicronucleatum and P. aurelxa species,
but it cannot infect the nucleus of P. jen-
ningsi , P. bursaria , P. trichium, P . dubo-
scqui and P. woodruff r. Furthermore,
maintenance of H. obtusa in the host macro¬
nucleus is achieved only in certain strains
of P. caudatum. This indicates that the
infection and the maintenance of H. obtusa
are separately controlled phenomena, and
that the maintenance of the bacterium is
controlled by host genes. To confirm this
possibility, we crossed various strains of
P. caudatum syngen 13 and tested whether
the offsprings can maintain the bacterium
or not. As a result, we found that the
maintenance of H. obtusa is controlled at
least two different host genes: M and S
loci. Maintenance of the bacterium requires
a dominant gene M, but a dominant gene S
can suppress the effect of M gene.
Therefore, genotypes of cells have to be
either MMss or Mmss for maintenance of the
symbiont in their macronucleus.

DB 97

EXTRACTS OF MASS CULTURES OF THE JUMYO
MUTANT OF PARAMECIUM TERAURELIA PROMOTE
ITS LOW FISSION RATE IN ISOLATION CUL¬
TURES. . »

1 Y . Takagi , H. Kinoshita and K. Kaji .
^Dept. of Biol., Nara Women's Univ., Nara.
^Isotopes Lab., Tokyo Metro. Inst.
Gerontol., Tokyo-.

The jumyo mutant of P. tetraurelia with
short clonal lifespan can grow fast in mass
cultures, while it can grow only slowly in
daily re-isolation cultures. Since a single
cell transfer to fresh culture medium with
a micropipette is repeated in daily re-iso-
lation cultures, 1 ) cells might be injured
through pipetting or 2) their excreting
substances, if relevant, might become in¬
valid through dilution. The possibility 1 )
was excluded by experiments in which the
fission rate in isolation cultures with
daily transfer of 10 cells vs. 1 cell was
compared: the fission rate was significant¬
ly higher in the 10 cells-transf er cul¬
tures. The cell-free medium of mutant mass
cultures was found to include substances
which promoted the mutant's low fission
rate in isolation cultures: crude extracts
concentrated hundredfold by ultrafiltration
were effective at 6pl in 0.4 ml culture
medium. Both the heated extracts (100°C,
5 min. ) and the eluted fraction with molec¬
ular weights of less than 1000 were inef¬
fective. The question how the jumyo gene
is related to the fission promoting fac¬
tors) remains open.

Developmental Biology

1273

DB 98

EFFECT OF INCLINATION ON THE SUBCORTICAL
ROTATION OF NEWT EGGS.

M. Fujisue and M. Sakai. Dept, of Biol.,
Fac. °f Sci. , Kagoshima Univ., Kagoshima.

A 30° rotation of the subcortical

contents relative to its surface occurs
during the first cell cycle of newt eggs.
The movements are directly implicated in
the specification of the embryonic
dorsoventral axis. On the other hand,
90° inclination of the egg can specify
the dorsoventral axis. However, the
effect of 90° inclination on the
subcortical rotation remained unclear.

Newt eggs were embedded in 9% geratin
and inclined at times after fertilization
and the movement of the Nile-blue stained
subcortical cytoplasm were followed
during the inclination period.

At the normalized time 0.2 (0=
insemination, 1.0=first cleavage), no
gravity-driven movement was found.
Injection of colchicine at this time has
no effect on the displacement of the
subcortical cytoplasm. At the time
after 0.4, when the subcortical rotation
initiates, the inclination caused a
slight (about 4°) displacement of the
Nile blue marks. The displacement was
emphasized in the presence of colchicine,
especially at the time 0.4.

These results suggest that the
cytoplasmic firmness at 0.4 depends
predominantly on microtubules and it
depend on other cytoskeletal elements at
before 0.2 and after 0.6.

DB 99

CORRELATION BETWEEN THE PIGMENTATION PAT¬
TERN, THE FIRST CLEAVAGE PLANE, AND THE
BODY AXIS IN Xenopus EMBRYOS

R. Masho, Dept, of Polymer Science, Fac.
of Technology, Gunma University, Kiryu.

One of the blastomeres of 2-cell-
stage Xenopus embryos with three types of
distinguished pigmentation pattern was
labeled with injected fluorescein dextran
amine (FDA). The first cleavage plane was
directly shown as the boundary between
labeled and non-labeled cells in resulting
embryos under an epif luorescence micro¬
scope .

The first cleavage plane fell to
divide an embryo roughly into lateral
halves in majority of the cases where the
plane bisected the pigmentation pattern
symmetrically. In the cases where the
first cleavage plane was perpendicular to
the axis of bilateral symmetry and divided
the egg into dark and pale blastomeres,
the first cleavage plane fell frontally to
divide the resulting embryo into ventral
and dorsal regions.

It seems that the cleavage furrow
fall close to the future dorsal-ventral
axis, then perpendicular to it in the
second turn. In some cases, however, the
furrow takes the perpendicular position
first, giving rise to the frontal pattern.
The pigmentation pattern, too, serves as a
rough indicator of the body axis; the
darkly pigmented area is ventral and the
palely pigmented area is dorsal.

DB 100

LOCALIZATION OF THE DORSAL AXIAL STRUCTURES
INDUCING COMPONENTS OF XENOPUS EGGS.

H. Kageura. Biological Laboratory Faculty
of Science Fukuoka University, Fukuoka.

The experiments here were designed to

reveal the distribution in early embryos of
components which induce the dorsal axial
structures. In the experiments embryos were
selected on the bases of a precisely
regular pattern of cleavage and a symmet¬
rical pattern of color. In such selected
32-cell embryos, the components seem to be
contained densely in the most dorsal blas¬
tomeres of the third tier and the fourth
tier, and be contained thinly in the most
dorsal blastomeres of the second tier from
the results of defect and transplantation
experiments .

Only the difference of density of these
components makes a difference of develop¬
mental capacity among dorsal, latefal and
ventral blastomeres in each tier.

DB 101

DRAMATIC ALTERATION AT THE MIDBLASTULA
TRANSITION OF THE EFFECT OF LITHIUM ON
XENOPUS PATTERN FORMATION.

Y. Yamaguchi and A. Shinagawa, Dep. Biol.
Fac. Sci., Yamagata Univ., Yamagata+990 .

The stage-dependent effect of Li on

pattern formation of Xenopus embryos was
reexamined by pulse exposure t£ Li at
various stages. Effect of Li on the
pattern formation appeared at the 2-cell
stage and disappeared at the early neurula
stage,. The degree of the effect varied
well continuously with one peak at 128- to
256-cell stages. The mode of the effect
altered drastically at 12th cleavage, when
midblastula transition (MBT) occurs; ex¬
posure at pre-MBT stages induced exclsive-
ly anteriorization , while that at post-MBT
stages induced exclusively posterioriza-
tion. Such alteration at MBT of the mode
occured also in embryos developed from
half-egg fragments, in which MBT occurs
at 11th cleavage. Enhanced posterioriza-
tion was induced by double exposure at
pre- and post-MBT stages.

From these findings, the followings are
speculated: The mode and the degree of

effect of Li on Xenopus pattern format¬
ion may be regulated by separate mecha¬
nisms. 2. Alteration of the mode of the
effect must be timed by the same timing
system as of MBT, i.e. a system in which
the "nuclear/cy toplasmic ratio" plays a
key role. 3. Both the mode and the
degree of the effect are progressively
determined during the effective period.

1274

Developmental Biology

DB 102

LOCALIZATION OF A SURFACE ANTIGEN IN
DROSOPHILA EARLY EMBRYOS DERIVED FROM
MATERNAL-EFFECT MUTANTS ALTERING THE
ANTERIOR-POSTERIOR EMBRYONIC PATTERN.

F. Maruo* and M. Okada . Inst. Biol. Sci.,
Univ. of Tsukuba, Ibaraki. _ _

One of the monoclonal antibodies against

ovarian cells from benign gonial cell
neoplasm mutant females, was found to
react strongly with the pole cell surface
in wildtype embryos. The antigen was also
prominent on the posterior somatic region
adjacent to pole cells. Before pole cell
formation, it was mainly present in
surface of the both poles of cleavage
embryos. In this study, the effect of some
maternal polarity genes on localization of
the surface antigen was examined.
Mutations in torso and trunk delete the
anterior- and posterior-most structures of
the embryo. But the antigen showed perfect
wildtype distributions. The mutation in
en deletes the abdomen and pole
cells. In cleavage stage, the antigen was
found in both poles, but no antigen was
seen in the blastoderm embryos with no
pole cells. Embryos from bicoid females
lack head and thorax. No antigen was found
in most of the cleavage embryos, however,
it was faintly seen only in the posterior
pole of some embryos. In blastoderm stage,
it showed the wildtype distributions. The
bicoid'1' activity may be necessary to the
anterior localization of the antigen, and
maintenance of its posterior localization
may require the presence of pole cells.

DB 103

DISTRIBUTION OF CYTOPLASMIC ACTIVITY
DEPENDENT ON THE TORSO GENE IN THE EARLY
DROSOPHILA EMBRYO.

S. Sugiyama and M. Okada

Inst . Biol . Sci . , Univ. of Tsukuba , Tsukuba .

Anterior terminal (labrum) and posterior
terminal (telson) structures are deleted
from embryos of the maternal-effect mutant
torso . Deleted structures such as the
median tooth (labrum) or the posterior
spiracles and filzkorpers (telson) are
restored to mutants by the injection of
cytoplasm from wildtype embryos (2.5% egg
volume) at the early cleavage stage. While
anterior and lateral wildtype cytoplasm
transplanted to the mutant anterior
rescued labrum formation, posterior
cytoplasm did not. Similarly, posterior
and lateral wildtype cytoplasm rescued
telson formation but anterior cytoplasm
did not. Interestingly, bicoid mutant
anterior cytoplasm restored telson
structures and nanos mutant posterior
cytoplasm restored labrum structures when
injected into the posterior and anterior
of torso embryos, respectively. Presuming
the distribution of torso+ activity is
normal in bicoid and nanos mutants, torso+
dependent activity required for both
labrum and telson formation can be assumed
to be distributed over the entire length
of the wildtype embryo.

db m

A MATERNAL-EFFECT MUTATION AFFECTING
NUCLEAR APPROACH TO THE CORTEX IN
DROSOPHILA EARLY EMBRYOS.

K. Hatanaka and M. Okada. Inst. Biol.
Sci., Univ. of Tsukuba, Ibaraki.

Embryos deriving from females
homozygous for l(3)c21RKWbJU (RW630)
showed a delay in nuclear approach to the
entire egg cortex in the late cleavage
stage. Reduced separation of daughter
nuclei was also frequently observed in the
cortex of RW630 embryos at the syncytial
blastoderm stage, resulting in a layer of
"nuclear twins". The former phenotype can
be mimicked by treatment of wild-type
embryos at early cleavage stage with 0.2-
0.5 pg/ml colchicine for 5 minutes, and
the latter phenotype can be mimicked by
treatment of wild-type embryos at
syncytial blastoderm stage with 1 pg/ml
colchicine for 5 minutes. Both treatments
do not inhibit the nuclear division
itself. Treatment with 10 pg/ml
1 urn i c o 1 ch i c i ne for 5 minutes had no
visible effect on nuclear migration, and
we have previously reported that
cytochalasin B has a quite different
effect on nuclear migration, i.e., it
causes a spherical nuclear distribution in
the anterior part of the embryo. even in
the late cleavage stage. From these
results, we conclude that RW630 has a
genetic defect in a function of
microtubules which is required for nuclear
migration toward the cortex.

DB 105

THE FUNCTION OF 1.5KB POLY ( A ) +RNA REQUIRED
FOR POLE CELL FORMATION IN DROSOPHILA
S.Kobayashi and M. Okada. Inst. Biol. Sci.,
Univ. of Tsukuba, Ibaraki

Pole cells, determined for germ line,
are prevented from forming by uv
irradiation tTo~~ the posterior pole of
cleavage embryos. We have reported that
a cDNA clone (pDe20.6), highly homologous
to "mitochondrial large rRNA gene hybrid-
select a 1 . 5kb poly(A)+RNA which is able
to restore pole cell f otming ability to
the uv-irradiated embryos. The 1 . 5kb RNA
selected by pDE20.6 was also able to
induce pole cells in the anterior region
of embryos when it was co-injected with
uv-irradiated polar plasm, although the
RNA or uv-irradiated polar plasm was not
effective. Injection of polar plasm into
the anterior induced amn i opr oct odea 1
invagination and Malpighian tubules as
well as pole cells in the anterior. In
contrast, uv-irradiated polar plasm never
induced these structures when injected
into the anterior. When the irradiated
polar plasm was co-injected with the 1 . 5kb
RNA selected by pDE20.6 into the anterior,
the embryos formed pole cells but neither
amnioproctodeal invagination nor
Malpighian tubules was found in the
anterior. These results suggest that the
1 . 5kb RNA selected by pDE20.6 is involved
in the mechanism underlying pole cell
formation but not specification of
posterior structure.

Developmental Biology

1275

DB 106

ENUMERATION AND ANALYSIS OF MOLECULES
LOCALIZED IN CION A EGG MYOPLASM.

T. Nishikata, K. W. Makabe, and N. Satoh.
Dept, of Zool., Fac. of. Sci., Kyoto Univ.,
Kyo to .

Ascidian myoplasm has been thought to
contain cytoplasmic determinants for the
muscle cell differentiation. In order to
search for the candidates for the
determinants, we compared differences in
the quality of proteins and RNAs between
isolated myoplasm and the other egg
cytoplasm. Moreover, we characterized
antigenic polypeptides recognized by
myop 1 asm-spec i f i c monoclonal antibodies we
produced .

(1) Protein localization: Proteins of
myoplasm and the other cytoplasmic region,
isolated from Ciona eggs, were subjected to
2-D gel electrophoresis and visualized with
silver staining. We found 16 myoplasm-
specific spots and 17 myoplasm rich spots
on these gels.

(2) RNA localization: RNAs were extracted
from both isolated myoplasm and the other
cytoplasm, and translated in vitro using
rabbit reticulocyte lysate. All 40 spots
translated from isolated myoplasmic RNAs
were found among the proteins translated
from the other cytoplasm RNAs. At present,
we could not find myoplasm specific mRNA.

( 3 ) Myop 1 asm-spec i f i c antigens: We have
already produced 12 myop 1 asm-spec i f i c
monoclonal antibodies. One of them, IIG6B2
antibody which has an inhibitory effect on
the muscle cell differentiation, recognized
one single polypeptide corresponding to a
molecular weight of approximately 40kd and
pi of 5.0. After analysing molecular
natures of these myop 1 asm-spec i f i c
antigens, we classified them into 10 types.

DB 107

FORMATION OF PRIMORDIAL GERM CELLS WITH
GRANULE-FRACTIONS DERIVED FROM RANA NIGRO-
MACULATA EMBRYOS AT THE TAIL-BUD STAGE.

T. Shirane, Faculty of Integrated Arts and
Sciences, Hiroshima University, Hiroshima.

It has been well known in Anura that the
formation of primordial germ cells (PGCs)
is correlated with a specific cytoplasm
(germ plasm). In Rana nigromaculata and
a few other species, however, presumptive
PGCs with germ plasm cannot be detected
after the neurula stage. In these species,
PAS positive granules appear to resemble
the localization of presumptive PGCs in
other species.

In order to examine whether these gran¬
ules correlate to the formation of PGCs,
three fractions of granules of R^ nigro¬
maculata embryos at the tail-bud stage were
prepared by centrifugation: heavy granules
(H), crude lysosomes(L) and crude mitochon-
dria(M). One ul of each fraction (3 embryos
/ul in PBS) was injected into the abdomen
of tail-bud embryos, which were raised from
eggs irradiated with UV light at the 2-cell
stage. The number of PGCs at the pre-larval
stage was examined in the experimental
animals. The numbers of PGCs were signifi¬
cantly higher in the H and M fractions than
those in the L fraction. These results in¬
dicate that some of the heavy granules in¬
volve a determining factor of PGCs, al¬
though it remains to be proved whether or
not the available granule is similar to the
PAS positive one.

DB 108

AN IMMUNOSTAINING TO SEA URCHIN OOCYTES AND

EGGS WITH ANTI-RNP ANTIBODY. - RESULTS

AND COMPARISON WITH THE Ag-NOR STAINING -

Reiko M. Amikura. Tokyo Women’s Medical
College, Tokyo. _ _

An anti-RNP antibody was applied to sea

urchin eggs, oocytes and isolated germinal
vesicles (Giudice, 1972). Proteins which
were boiled in the SDS sample buffer were
loaded on polyacrylamide gradient (10 - 20
%) gels and electrophoretical ly transferred
to nitrocellulose membranes. Staining was
made by a blotting detection kit (Amarsham,
U.S.A.). The anti-RNP antibody was
monoclonal, and commercially available
(Synbiotics Co, U.S.A.; Funakoshi Co.
Tokyo). Three to five bands were stained
with anti-RNP antibody in each of the three
samples, indicating that this antibody
detects various sizes of RNP .

In the ovaries fixed in Carnoy’s
solution (methanol, acetic acid) the
paraffin sections were immunos tai ned with
anti-RNP antibody using avidin-biot in¬
peroxidase complex method (Vectastain ABC
kit, Vector Lab. , U.S.A. ) Followings were
observed: 1) In oocytes, nucleoplasm of the
germinal vesicle was stained but not for the
nucleoli. 2) sea urchin egg cytoplasm was
reacted with the antibody whereas the
oocyte cytoplasm was not. The staining
pattern was compared to the results of Ag-
NOR staining (Amikura and Ihnuma, 1984),
suggesting that Ag-NOR protein is different
from the RNP (snRNP) which is detected by
the anti-RNP antibody.

DB 109

MASS ISOLATION OF GERMINAL VESICLES FROM
ASCIDIAN OOCYTES AND PRODUCTION OF GERMINAL
VESICLE SPECIFIC MONOCLONAL ANTIBODIES
S.Fujiwara and N. Satoh. Dept. of Zool., Kyoto
Univ . , Kyo to . _ - _

The germinal vesicle (GV) could be regard-

ed as a maternal stock of nuclear proteins.
In order to investigate the role of GV
contents for the ascidian development, we
attempted mass isolation of GVs from oocytes
of the ascidian, Halocynthia roretzi .

Oocytes were chemically dechorionated to
remove follicle cells and test cells. We
first tried to isolate GVs from unfixed
oocytes. But GV isolation from unfixed oo¬
cytes was difficult, because the cytoplasm of
Halocynthia oocytes was quite sticky. We,
therefore, fixed the dechorionated oocytes
with 70% ethanol so that the stickiness of
the oocyte cytoplasm became reduced. Fixa¬
tion method resulted in higher yield of GVs.
The estimated yields were about 40 - 50% in

number. SDS polyacrylamide gel electro¬
phoresis showed that there was not a severe
loss of GV proteins. Therefore, we assumed
that most of the GV contents, including
soluble proteins, were retained in the iso¬
lated GVs.

Using the GVs isolated from fixed oocytes
as crude antigens, we produced monoclonal
antibodies against the GV materials. At
present, we have obtained a few hybridoma
clones producing GV-specific monoclonal anti¬
bodies. One of the antigens were found in
nuclei of most of the embryonic cells until
tailbud stage, but not in nuclei of follicle
cells and test cells.

1276

Developmental Biology

DB 112

DB 110

CHARACTERIZATION OF GERMINAL VESICLE
FACTORS REQUIRED FOR INDUCING THE CYTO¬
PLASMIC CYCLE OF STARFISIjj OOCYTES. 2

N. Hashimoto1, S. Nemoto1 and k. Yamamotc>
1Tateyama Marine Lab., OChanomizu Univ. ,
Tateyama, and zDept. of Biol., Fac. of
General Edu., Gifu Univ., Gifu. _

“ Germinal vesicles (GV) of starfish

oocytes contain some factors which are
indispensable for inducing the cytoplasm
of oocytes to initiate its own cyclic
activity during meiotic division. To re¬
veal the mechanism of this phenomenon, it
is necessary to obtain GV materials in
ample quantities. We developed a
mass-isolation procedure for the GVs from
Asterias amurensis and Asterina pectini-
f era oocytes. In the present study, some
properties of the GV factors were examined
using extracts prepared from isolated GVs.

Isolation of GVs and the assay of the
GV factors were made as previously
reported (Zool. Sci., 1987). When GV homo¬
genate in a bufferizeg hypotonic medium
was centrifuged at 10 xg, the factors were
found in the supernatant. The factors
did not pass through a ultraf iltratio^
membrane which is available with 2x10° M.
W. cutoff. Since the factors became
inactive by treatments with either heat or
papain, the factors may be proteins with
relatively high molecular weight. The GV
factors of sea cucumber oocytes (Stichopus
japonicus and Holothuria moebi ) were
effective to initiate the cytoplasmic
cycle of Asterina pectinifera oocytes.

DB 111

DIFFERENTIAL DEPENDENCE ON GERMINAL VESICLE
(GV) MATERIALS OF THE EVENTS ASSOCIATED
WITH MEIOTIC AND MITOTIC CELL-CYCLES OF
XENOPUS OOCYTES AND EGGS.

K. Ohsurai and Ch. Katagiri . Zool. Inst.,
Fac. Sci., Hokkaido Univ., Sapporo

We have previously shown that the eggs
from which the GV had been removed (enucle¬
ated) do not exhibit the periodic changes
in rigidity that correspond to the cleavage
cycle in fertilized eggs. Similar determi¬
nation of rigidity revealed that during the
progesterone-induced maturation process,
enucleated oocytes exhibit periodic changes
in synchrony with the first and second
meiotic divisions of nucleated oocytes.
Determination of maturation-promoting
factor (MPF) activity measured by transfer
of cytoplasm to full-grown oocytes showed
that after progesterone treatment enucle¬
ated oocytes generate MPF-cycling coinci¬
dent with the meiotic cycles of nucleated
oocytes. Upon electric stimulation,
enucleated eggs exhibited a normal acti¬
vation response and rapidly lost MPF, but
showed a delayed periodic change in
rigidity at extremely longer intervals. In
these eggs, MPF activity reappeared at the
normal timing of activated nucleated eggs,
but retained a high level until it disap¬
peared in synchrony with the delayed change
in rigidity. These results indicate that
meiotic and mitotic cell-cycle events of
oocyte and egg cytoplasms possess differ¬
ential requirements for the GV materials.

CELL-FREE PREPARATIONS FROM XENOPUS EGGS
ACCELERATE CYTOPLASMIC CYCLE OF NEWT EGGS.
H. Kasuga and M. Sakai. Dept of Biol.,

Fac. of Sci., Kagoshima Univ., Kagoshima.

vie have shown that an injection of

cytoplasm from a fertilized Xenopus egg
into a newt egg accelerates cytoplasmic
cycle of the recipient egg.

To characterize the factor in Xenopus
eggs which accelerates the cytoplasmic
cycle of newt eggs, cell free
preparations of the cytoplasm from
activated or unactivated eggs of Xenopus
were made and their activity to
accelerate the cytoplasmic cycle of newt
eggs were tested.

He jellied activated or unactivated
Xenopus eggs were rinsed twice w ith
ice-cold extraction buffer containing
250mM sucrose, 200mM KC1 , 1 . 5mM MgCl-,

2mM 2-mercaptoethanol , 1 OmM tris-HCl and
1 OmM EGTA (pH 7.5). Eggs were
transferred to a 1.5 ml centrifuge tube
and crashed by centrifugation at 10,000g
for 1 Omin and the cytoplasmic layer above
the packed pigment and yolk was used as a
cell-free preparation.

When the cell free preparation from
activated Xenopus eggs was injected
into a newt egg, the cytoplasmic cycle
(indicated by the surface contraction
waves: SCT7s) acceleraed. Injection of
unactivated egg extract caused precocious
first SCT7 however, thereafter the
cytoplasmic cycle of the recipient egg
ceased completely.

DB 113

ROLE OF Ca2+ ON THE MORPHOLOGICAL CHANGES
AND REINITIATION OF MEIOSIS AT

FERTILIZATION IN ASCIDIAN OOCYTES
N. Sensui . Misaki Mar. Biol. St. Fac.

of Sci . , Univ. Tokyo

It is well known that deformation of
oocyte, segregation of ooplasm and polar
body formation occur at fertilization in
a^scidian oocytes. In the gjesent study,
changes in intracellular Caz in oocytes
of ascidian, Ciona^^avignyi were measured
using aequorin Ca2 reaction. 2+ It was
showed that intracellular Caz rised
transiently just after i nsemi nat i og+
Calcium ionophor|+ also induced the Ca
increase. The Ca+ rise was independent of
extracellular CaZ . Furthermore, injection
of2£GTA caused suppression of intracellular
Ca2 rise , deformation of oocyte and polar
body formation when oocytes were
inseminated with sperm. In cgjtrast, an
increase in intracellular Caz by the
injection of CaZ buffer caused deformation
of oocyte and progression of meiosis.
Oocyte deformation and ooplasmic

segregation were„+ induced by low
concentration of CaZ , however reinitiation
of meiosis and decondensation of nucl|^s
required high consen trat i on of Ca 2 +
These suggest that intracellular Ca
increases at fertilization in the oocyte of
ascideans and morphological changs in
oocytes and reinitiation of meiosis induced
by the increased Ca

Developmental Biology 1277

DB 114

Inhibition of transition to metaphase in
mouse oocytes by an inhibitor of a calcium
activated neutral pro^ase, E64-d. **
N.Hashimoto^, Y. Kasai , S^Jwaghita ,
T.Kishimoto and K.Imahori*^. Biol. Lab.,
Tokyo Inst. Tech., Tokyo, Mitsubishi-
Kasei Inst. Life Sci., Machida. _

E64-d is a membrane permeable derivative
of E64-c, a specific inhibitor of calcium
activated neutral protease (CANP).

Exposure to E64-d prominently increased
mitotic index of HeLa cells as previously
reported in A431 cells. In mouse oocytes,
E64-d (400 pg/ml ) blocked the first polar
body emission but not germinal vesicle
breakdown. Double staining with FITC-
conjugated anti-tubulin antibody and DAPI
revealed that meiotic spindles were not
formed in E6 4-d-treated oocytes though
chromosome condensation occurred. During
the prolonged culture, condensed
chromosomes dispersed in the cytoplasm.
When exposed to E64-d after prometaphase,
about 30% of the oocytes emitted the first
polar body but never formed spindles at
the second metaphase. Thus, E64-d inhibit
transition from prometaphase to metaphase
during both first and second meiotic
division. Further, cytoplasmic transfer
experiments demonstrated that the activity
of maturation-promoting factor in mouse
oocytes remained at high levels in the
presence of E64-d. It is possible that
CANP is related to the regulation of
microtuble assembly both in meiosis and
mitosis .

DB 115

PERIODICAL CHANGES OF SPERM ASTER DURING
MEIOSIS IN STARFISH OOCYTES.

K. Kyozuka and K. Osanai. Mar. Biol. Stn.,
Tohoku Univ., Aomori. _

The distribution of microtubules during
meiosis after sperm incorporation was
investigated with anti-tubulin indirect
immunofluorescence microscopy in starfish,
As terina pectinif era. Aster ina sperm
normally fertilize the oocyte at the meta¬
phase of first meiotic division (metaphase
I). During the first polar body formation,
the sperm aster developed conic, the apex
of which looked toward the animal pole.
Then, the sperm head migrated to the center
of the oocyte. The sperm aster became
smaller when oocytes were at metaphase II.
After the completion of both meiotic
divisions, astral rays developed in all
directions from the egg center to the
periphery of the egg. The female pronucleus
migrated to the egg center where the male
pronucleus had been located. The interphase
aster disappeared from the inner side with
the development of the mitotic aster. When
sperm were added to the immature oocyte of
the germinal vesicle stage, they penetrated
without egg activation. Conical sperm
asters developed at the telophase of first
meiotic division. They became smaller at
the metaphase II. These results indicate
that the organization of microtubules in
the sperm aster changes periodically with
meiotic cell cycles in inversely pro¬
portional to the development of meiotic
spindle.

DB 116

BEHAVIOR OF SPERM NUCLEI INCORPORATED INTO NUCLEATE AND
ANUCLEATE FRAGMENTS OF HAMSTER EGGS AT VARIOUS STAGES OF
MATURATION.

Y. Hirao, Dept, of Biology, Wakayama Medical College,
Wakayama.

Zona-free ovarian eggs (GV stage and Proaet-I) and ovu¬
lated eggs(Met-II) were divided into nucleate and anucleate
fragments by the centrifugal force. Artificial insemination
in vitro was performed with capacitated and acrosome react¬
ed spermatozoa in which were incubated in mTAPL during 3-4
hr at 37 °C under 5 X C02 gas phase.

Between 3-4hr after insaination, egg fragments were exa¬
mined with a phase-contrast microscope for the morphology
of sperm heads. In both nucleate and anucleate fragments
at GV stage, incorporated sperm heads remained unchanged
in egg cytoplasm. On the contrary, the decondensation of
sperm heads occurred in both nucleate and anucleate frag¬
ments at Promet-I. But, they did’nt develope into male pro-
nuclei. In nucleate fragments at Met-II, sperm heads trans¬
formed well into developmental stage of male pronuclei,
when the female pronucleus was completely formed.

These results indicate that (1) the egg cytoplasmic fac¬
tor inducing the decondensation of sperm head appears after
GVBD, (2) the factor controlling the development of the fe¬
male pronucleus is identical to that regulating the de¬
velopment of the male pronucleus.

DB 117

ETHANOL-PHOSPHOTUNGSTIC ACID (E-PTA) AND
BISMUTH STAINING OF MOUSE SPERM NUCLEI AT
FERTILIZATION.

I. Takeuchi and Y. Takeuchi . Inst. Dev.
Res., Aichi Prefect. Colony, Kasugai and
Gifu Coll. Med. Technol., Seki.

The decondensing sperm nuclei and male
pronuclei in the fertilizing mouse eggs
were examined electron-microscopically
afteir E-PTA staining, formaldehyde-bismuth
(FA-Bi) staining and glutar a ldehyde-
bismuth (GA-Bi) staining. The results ob¬
tained reconfirmed the previous observa¬
tion that the nuclei of the ejaculated
sperms contained numerous basic proteins.
The decondensing sperm nuclei in the fer¬
tilized egg cytoplasm rapidly lost the
basic proteins, and a number of
p e r i c hr oma t i n granules and a few
nucleolus-organizing regions were formed
in association with the dispersing
chromatin materials. In the male

pronuclei, numerous basic proteins reap¬
peared in the chromatin materials and con¬
densed spherical nucleoli. Fibrillar cen¬
ters localized at the periphery of spheri¬
cal nucleoli also contained numerous basic
proteins. Per ichromatin granules were
somewhat distributed in the nucleoplasm,
but no interchromatin granules were found.
These results suggest that some
heterogeneous nuclear RNAs may be syn¬
thesized by the transcription of paternal
genome soon after the entrance of the
sperm in the egg cytoplasm.

1278

Developmental Biology

DB 118

AN ULTRASTRUCTURAL STUDY OF THE SECRETORY
CELLS OF THE OVIDUCT EPITHELIUM IN THE
GOLDEN HAMSTER.

H. Abe and T. Oikawa. Developmental and
Reproductive Biology Center, Yamagata.

The secretory cells of the oviduct
epithelium in the golden hamster were
investigated by means of electron
microscopy. In the fimbrial secretory
cells, two types of secretory granules
were observed: 1) with a moderately
electron-dense, homogeneous matrix
containing an e 1 ect ron -dense spot, and 2)
with an e 1 ect ron - 1 ucen t matrix. In the
ampullar and isthmic cells, numerous
secretory granules that were moderately
electron-dense and possessed a homogeneous
matrix were observed in the supranuclear
area. E 1 ect ron - 1 ucen t vesicles partly
filled with a dense substance and/or a
finely granular substance were observed in
the intranuclear cytoplasm in the ampullar
and isthmic cells. Fusion of the vesicles
with secretory was frequently observed in
the isthmic cells. Secretions such as
exocytosis of the secretory granules were
suggested in all segments. Evidence of
apocrine and holocrine secretion
mechanisms were obtained in the fimbrial
and ampullar cells, but not in the isthmic
cells. Moreover, secretory granules and
vesicles were found in the intercellular
space in the ampulla and isthmus. The
findings were different among the three
segments .

DB 119

PARTIAL PURIFICATION AND INITIAL CHARACTERIZATION
OF COELOMIC ENVELOPE GLYCOPROTEIN CONVERTASE.

D.M. Hardy, and J.L. Hedrick. Department of
Biochemistry and Biophysics, University of
California. Davis. CA 95616 _

Eggs recovered from the body cavity (coelom)
of anuran amphibians are not fertilizable .

Previous studies showed that these coelomic eggs
must pass through the pars recta of the oviduct
before they can be fertilized. In Xenopus laevis ,
the coelomic egg envelope (CE) is modified during
transit through the oviduct. One of the observed
changes in the envelope is conversion of the
43,000 Mr glycoprotein component of the CE to
41,000 Mr. We isolated the contents of pars recta
secretory granules, and assayed for proteolytic
activity using a panel of substrates. Activity
was detected only using substrates with Lys or Arg
in the PI postion. The substrate hydrolyzed at
the highest rate was Boc-Phe-Ser-Arg-methyl-
coumarylamine (FSR-MCA) . The pH optimum for
hydrolysis of FSR-MCA was 8.0. In the absence of
Ca++, the FSR-MCA activity decreased by 50% in 25
min at 23°C. We partially purified the FSR-MCA
activity by gel filtration HPLC of pars recta
granule contents on tandem TSK3000 and TSK2000
columns. The FSR-MCA activity eluted with Mr less
than 30,000 during HPLC, and readily converted the
43,000 Mr glycoprotein component of the CE to
41 , 000 Mr . We conclude that the enzyme which
converts the 43,000 Mr component to 41,000 Mr is a
Lys/Arg specific protease with Mr less than
30,000. Supported by grants HD04906 and HD07088.

DB 120

COMPARATIVE STUDIES OF THE COELOMIC- AND
UTERINE EGG ENVELOPE IN THE NEWT, CYN0PS
PYRRHOGASTER.

K. Onitake, T. Adachi and N. Yoshida.

Dept, of Biol., Fac. of Sci., Yamagata
Univ. , Yamagata. _ _ _ _

As a step toward defining the role of

vitelline envelope (VE) , VEs were isolated
from coelomic-, uterine- and fertilized
eggs of Cynops pyrrhogaster. The macro-
molecular composition of the VE was deter¬
mined by sodium dodecyl sulfate polyacryl¬
amide gel electrophoresis (SDS-PAGE) . The
uterine egg envelope (UVE) contained at
least 7-9 proteins with molecular weights
ranging from about 120000 to 18000. SDS-
PAGE analyses of VE demonstrated that in
comparison with CE, UVE lacks 37-41K pro¬
teins concomitant with the increased stain-
ability of 35K protein and posses 74K .pro¬
tein which is not appeared in CE. Immuno-
blotting against the absorbed anti-UVE
antiserum revealed that the CE lacks 35K
and 74K protein while the UVE reacts
strongly to antiserum at the regions of
35K and 74K protein. 35K and 74K proteins
on SDS-PAGE are stained clearly with per¬
iodic acid-Schiff (PAS) -staining.

It is probable that in Cynops pyrrho¬
gaster the coelomic egg envelope changes
into the uterine egg envelope by conversion
of 37-41K to 35K glycoprotein and addition
of 57K glycoprotein by the oviduct.

DB 121

Electrical Block to Polyspermy in the
salamander, Hynobius nebulosus eggs.

Y. Iwao and L. A. Jaf fel . Biol. Instit. ,
Fac. Sci., Yamaguchi Univ., Yamaguchi.
^ept. Physiol., Univ. Connecticut Health
Center, U. S. A.

To determine whether the positive
fertilization potential (FP) elcited by
the Hynobius eggs operates as a fast block
to polyspermy, we investigated the voltage
dependence of the Hynobius fertilization
by voltage-clamp technique.

The fertilization was blocked by the
clamp of egg membrane at +40 mV, but the
egg, then, elicited the FP within 2-6 sec
after the clamp was turn off. All the
eggs were monospermic. The fertilization
was not affected at all by the clamp at
+20 mV, so that the holding current was
shifted in inward direction 2-3 min after
insemination. Several sperm entered the
egg; polyspermy occurred. Since all the
eggs deposited by natural mating received
only one sperm, the positive FP is the
most important mechanism to prevent poly-
sermy in the Hynobius fertilization.

The fertilization of the anuran, Bufo
.iaponicus was blocked by the clamp at
+20 mV, but was not affected at 0 mV. The
cross-fertilization of the Bufo eggs with
the Hynobius sperm was not affected by the
clamp at +20 mV, but was blocked at +40
mV. These results demonstrate that the
voltage level required to block fertiliza¬
tion Is determined by the sperm species.

Developmental Biology

1279

DB 122

LOSS OF SUSCEPTIBILITY OF VITELLINE COAT
AGAINST LYSIN AS A SLOW BLOCK TO POLYSPERMY
IN BUFO JAPONICUS .

H. Yamasaki and Ch . Katagiri. Zool.

Inst., Fac . of Sci . , Hokkaiko Univ . ,

Sapporo .

The vitelline coat (VC) lysin from
Buf o japonicus sperm readily digests the
uterine egg VC (UEVC) but not the activated
egg VC (AEVC). Assays of sperm lysin-
induced hydrolysis of UEVC in. vitro showed
that this increase in resistance to lysin
occurs within 5 min after activation.
Treatment of fragmented UEVC with the
exudates of jelly-less activated eggs (AEX)
resulted in both the decrease in the
sensitivity to lysin and the change in the
melting property of the VC, to a degree
comparable to those exhibited by AEVC.
Removal of UEVC and insemination of the
eggs in reconstituted salt solution (RSS)

30 min after electric stimulation gave rise
to a high percentage of polyspermic
fertilization. Pretreatment with AEX of
dejellied- but not denuded uterine eggs
inhibited fertilization upon insemination
in RSS. Activation exudates collected from
denuded eggs (dAEX) exhibited the
activities of AEX both iri vivo and in
vitro more efficiently than the latter.
These results indicate that the
modification of VC induced by 'the exudates
of activated eggs constitutes a slow and
permanent block to polyspermy.

DB 123

COMPARISON OF TWO LECTINS ISOLATED FROM
CORTICAL GRANULES IN XENOPUS EGGS.

N. Yoshizaki, Dept, of Biol., Fac. of Gen.
Educ . , Gifu Univ., Gifu.

Two cortical granule lectins (CGL-1 and
CGL-2) of Xenopus laevis were compared by
PAGE, immunodiffusion, immunoblot, atomic
absorption and amino acid analysis. The
CGL was released following activation of
coelomic eggs by electric stimuli and
purified by affinity chromatography with
immobilized melibiose . The two lectins
were isolated electrophoretically from
polyacrylamide gel. The CGLs had the
following characteristics in common: the
specific activity of hemagglutination, the
specific binding to galactosides , a cal¬
cium requirement, calcium content of 5.69-
7.24 mole per 10 g protein, and antigen¬
icity to an antiserum raised against CGL-1.
Differing characteristics included: the
molecular weights of the subunits (42 kDa
and 46 kDa for CGL-1, and 39-46 kDa for
CGL-2), the pi value of the native lectins
(4.55 for CGL-1 and 4.70 for CGL-2), the
isofocusing pattern of the subunits, and
the amino acid composition. So CGL-1 and
CGL-2 differ in their particular combina¬
tion of subunits, each of which is hetero¬
geneous in size, charge and amino acid
sequence. That the CGLs are confined to
cortical granules in the eggs was confirm¬
ed by an immunoelectron microscopic
protein A-gold technique, using the anti¬
serum against CGL-1.

DB 124

FACTORS AFFECTING GEMMULE HATCHING OF
THE FRESHWATER SPONGES, EPHYDATIA
FLUVIATILIS (PORIFERA) .

Y. Masuda and K. Satoh. Dept, of Biol.,
Kawasaki Medical School, Kurashiki. _

When these gemmules were incubated in
the M-medium (Rasraont, 1961) at 25°C,
they reached final hatching rate of 90-100%
in 5 days (Control) . Gemmules were not
hatched in distilled water (DW) even at
25 °C in 16 days. When placed back in
the M-medium, they were hatched at the
same rate as control. However, the DW
pre-treatment gemmules took shorter time
to initiate hatching than control. The
gemmules seems to germinate in DW. These
results suggest that temperature and water
are not enough to cause gemmule hatching.

In CaCl2 solution (2 ”2-22mM) , gemmules
were hatched at the same rate as control.
However, in other solution such as NaCl
(2-8-22M) and MgCl2 (0.5-3.5mM), they
could not be hatched. The Ca-ion seems
to be essential for gemmule hatching.

Gemmules dried previously in the air
(RH 60-80%) for the various period (1-34
days) took longer time to initiate hatching
in the M-medium than control. As period
of the desiccation was extended, the time
to initiate hatching became longer. More¬
over, gemmules dehydrated in concentrated
NaCl solution (2~1-21M) took longer time
to initiate hatching in the M-medium than
control. Amount of water contained in
the gemmules seems to affect to initiate
hatching .

DB 125

TWO KINDS OF FERTILIZATION MEMBRANE DIS¬
SOLVING ACTIVITIES IN THE SUPERNATANT OF
HATCHED BLASTULAE OF THE SEA URCHIN
HEMICENTROTUS PULCHERRIMUS

H. Tanaka1, M. Yamaguchi 1 , N. Suzuki1 and
K. Nomura2. 1Noto Marine Lab., Kanazawa
Univ., Uchiura, 2Tokyo Metro. Inst, of
Gerontol., Tokyo. _ _

Two fertilization membrane dissolving
activities (FMDA) were isolated from the
supernatant of hatched blastulae of H. pul-
cherrimus by gel filtration on a Sepharose
4B column (4B-I and 4B-II ). When the ac¬
tivity was assayed using ethanol fixed
embryos and medium described by Barrett
which is 43% ionic strength of artificial
sea water (ASW), FMDA in 4B-I thinned and
then dissolved the entire fertilization
membrane (FM), while FMDA in 4 B — II dis¬
solved the FM like normal hatching. Both of
FMDAs were purified by chromatography on a
4-phenylbutylamine agarose column. Gluca-
nase activity in 4B-II was removed by this
chromatography while 4B-I still contained
the activity after the chromatography. FMDA
from 4B-II was further purified by HPLC
using TSK gel-G 3000 SW column. SDS-PAGE of
the purified FMDA from 4 B — II indicated that
the activity was a single protein with
molecular weight of 21-22K. This protein
retained FMDA in 83% ionic strength of ASW.
Although it took much longer time than the
case of H. pulcherrimus FM , FMDA from 4B-II
dissolved FM of G. crenularis , C. japoni¬
cus , A. manni and P. japonica but not the
membrane of A', crassispina .

1280

Developmental Biology

DB 126

SPECIFICITY OF THE PROTEASES INVOLVED IN THE
HATCHING1 PROCESS OF ^EA URCHIN EMBRYOS.

K. Nomura , H. Tanaka and N. Suzuki . Dept, of
Biocjjem., Tokyo Metropol . Inst, of Gerontol., Tokyo
and Noto Marine Lab. , Kanazawa Univ. , Uchiura.

Hatching enzyme(HE) and a caseinolytic protease,
tentatively named caseinase(CA) , were partially
purified from the supernatant over the hatching
embryos of the sea urchin, Hemicentrotus pulcher-
rimus , by (NH4)2S04 precipitation and Sephadex
G-100 chromatography. Sixteen bioactive peptides
(oxidized insulin B chain, angiotensin I, masto-
paran, substance P, etc.) were hydrolyzed by these
enzymes as well as by chymotrypsin(CH) and Subtili-
sin(SU) to compare their specificity and mode of
action. The fragment peptides in the digest were
isolated by HPLC on a C8 column, and identified by
their amino acid compositions.

The four enzymes acted differently from one
another on most of the peptides with respect to the
sites of cleavage and the fragments they produced,
indicating that their specificity and mode of
action are different. Among the peptides, serum
thymic factor (Pyr-Ala-Lys-Ser-Gln-Gly-Gly-Ser-Asn)
was the most prominent ^in discriminating the four
enzymes; HE cleaved Ser - and Gin-, CA, only Gin-,
CH , no cleavage and SU, Lys- only. Totaly HE and CA
cleaved Tyr-, Phe-, Trp-, Leu-, Met-, Glu-, Gin-,
Asn-, Thr-, Ser-, Pro-, Ala-, Gly-, Arg- and Lys-.

Leupeptin (ImM) inhibited HE and CA slightly,
while chymostatin (ImM) and EDTA (lOmM) inhibited
them almost completely. Chymostatin inhibited CH
and SU completely, but leupeptin and EDTA did not.

The results indicate that HE and CA are neither
Glu/ Asp-specific , nor chymotrypsin-like as previ¬
ously reported on the enzymes of other species.

DB 127

DEVELOPMENT OF THE HATCHING GLANDS OF THE
TELEOST, PARALICHTYS OLIVACEUS.

S.Yokoya and Y. Ebina. Division of cell
science, Fukushima Medical college,
Fukushima . _ _ _ _ _ _

For the further studies on the differ¬
entiation of the hatching glands (HG), we
examined some preliminary problems: (1)
The normal developmental stages with
special reference to development and
distribution of the HG. (2) Effects of
actinomycin C and cycloheximide on the
hatching out. (3) Properties of the hatch¬
ing enzymes to have knowledge for purifi¬
cation of the hatching enzymes from the
hatching medium. At 150, the HG appeared
first on the caudal part of the embryo of
14 hrs before hatching out They came to
distribute posteror half of the embryonic
body and yolk sac epidermis at the just
before hatching stage. By treatment with
lOug/ml of actinomycin C at 24 hrs before
hatching out, 20-50% of the embryos were
arrested the hatching out. Although
cyclohexymide of 0.5ug/ml showed toxicity
for the development of embryos, it had no
effect on the hatching out of the embryos
treated at 14 hrs before hatching out.

The hatching medium contained more than 22
different kinds of protein in SDS-PAGE and
isoelectric focusing. Six of them had
proteolytic activity in zymography on a
gelatin plate. Partially purified samples
by DEAE-Sephadex and Z-L-Phe-T-Sepharose
still contained 4 and 2 different proteo¬
lytic substances, respectively.

DB 128

Mode of Choriolytic Action of The Medaka
Hatching Enzyme (HCE)

S. Yasumasu, S. Kato , Y, Umino1 , I. luchi
and K. Yamagami, Life Sci . Inst., Sophia
Univ., and 1 Natjl| Inst. Health Japan.,
Tokyo .

It has been suggested that high chorioly¬
tic enzyme (HCE), a component of the hatch¬
ing enzyme, is markedly of high affinity
to the chorion. Semiquantitative examina¬
tion of its binding to the chorion revealed
that 2 mg of the coarse fragments of chori¬
on was saturated by about 1.5 ug of HCE.

Anti-HCE-monoclonal antibody (E-72 Fab'),
which inhibits the choriolytic activity but
not the caseinolytic (proteloytic ) activity
of HCE, was found to inhibit also the bind¬
ing of HCE to the chorion. This fact sug¬
gested that the binding of the enzyme to
the chorion was prerequisite to its
choriolytic action. EDTA inhibited both of
the proteolytic and the choriolytic ac¬
tivities, but did not inhibit the binding
of HCE to the chorion, probably because the
reagent affected the catalytic site. The
pH optimum of HCE for binding to the
chorion was almost identical with that for
choriolytic activity but different from
that for caseinolytic activity. This result
also suggested that the choriolytic ac¬
tivity of HCE was closely related to its
binding to the chorion.

The binding site and the catalytic site
may be present separately in the HCE
molecule .

DB 129

YOLK GRANULES IN SEA URCHIN EGGS AND
EMBRYOS .

Y. Yokota1 and K. H. Kato2. J-Biol. Lab.,
Aichi Pref. Univ., and 2Inst. Biol., Coll.
Gen. Edu., Nagoya City Univ.

Yolk granules in eggs and embryos of the
sea urchin, Hemicentrotus pulcherrimus ,
were bbserved with an electron microscope.
Yolk granules of the eggs and the embryos
were divided into 4 classes, dense granules
moderate granules, coarse granules and
lysosomal granules. The coarse and lyso¬
somal granules were not observed in unfer¬
tilized eggs. In gastrulae, on the other
hand, the dense granule was not found.

Dense granules decreased and, moderate and
coarse granules increased during embryo-
genesis. More than 90% of yolk granules in
unfertilized eggs were dense granules. In
blastulae and gastrulae, more than 90% of
yolk granules were moderate and coarse
granules . The ratio of area of yolk gran¬
ules to eggs and embryos was estimated on
the basis of the electron microgram. The
ratio of the unfertilized eggs, blastulae
and gastrulae were 22.3%, 22.5% and 13.9%,
respectively. In plutei, oil droplets,
which seemed to be derived from yolk
granules, were observed.

Morphological changes in yolk granules
of developing sea urchin embryos were dis¬
cussed in reference to the changes in the
protein components of yolk granules during
the development.

Developmental Biology

1281

DB 132

DB 130

YOLK PHOSPHOPROTEIN METABOLISM DURING EARLY
DEVELOPMENT OF MEDAKA.

M. Murakami, I. Iuchi , K. Yamagami.
Life Sci. Inst., Sophia Univ. , Tokyo.

The contents of phosphorus of protein
fraction measured by the Schneider's method
is known to decrease during early develop¬
ment of the medaka (Yamagami, 1960). We
identified four phosphoproteins in the yolk
of blastula stage by polyacrylamide gel
electrophoresis. Two types of phos¬
phoproteins were fractionated by DEAE-
cellulose chromatography. One of them con¬
sisted of two kinds of high molecular
weight phosphoproteins containing 0.7 %
(w/w) phosphorus and their amino acid com¬
position was similar to that of vitel¬
logenin of medaka. The other was composed
of two kinds of low molecular weight phos¬
phoproteins, characterized by containing
the large amount of phosphorus [31.5 %
(w/w)] and serine (34.2 mole%) and few
aromatic amino acid residues. These
characteristics markedly resembled those of
phosvitins of various fish eggs. These
phosphoproteins were degraded and decreased
through the development. Besides par¬
ticipation of acid phosphatase (Yamagami,
1961 ), the present _in vitro experiment
strongly suggested that some protease in
the fraction of the pellet of the egg
homogenate was associated with the
phosphoprotein degradation.

TRANSIENT INCREASE IN MITOCHONDRIAL
PROTEIN SYNTHESIS OF CRICKET EMBRYOS
DURING DEVELOPMENT.

S. Asada and T. Nakazawa. Dept, of Biol.,
Fac. of Sic., Toho Univ., Funabashi.

We previously reported that a transient
increase in mitochondrial protein
synthesis, which was inhibited by
chloramphenicol, was observed in embryos
of cricket, Gryllus bimaculatus , 25hr
after oviposition.

Decrease in mitochondrial protein
synthetic activity after the transient
increase depended on contents of free
amino acids in the embryo. Therefore,
increase in free amino acids by activated
protease during development probably
induced to inhibit [JH] -amino acids
incorporation in mitochondria.

Increase in the activity was found to
be related to the nucleic acids in
mitochondria. The enzyme activity of
mitochondrial protein synthesis was not so
different between Ohr and 25hr embryos.
However the addition of extracted nucleic
acids from 25hr embryos alone induced to
increase in protein synthesis of
mitochondrial systems. The activity of
[ H]-UTP incorporation in mitochondria of
25hr embryos was also considerably higher
than in those of Ohr embryos.

From these results, transient increase
in mitochondrial protein synthesis might
be due to the increase in transcripting
activity in 25hr embryos.

DB 131

DB 133

STIMULATION OF PROTEIN SYNTHESIS IN THE
MITOCHONDRIA AT THE MESENCHYME BLASTULA
STAGE OF SEA URCHIN EMBRYO.

T.Kawashima and T. Nakazawa. Dept. of
Biol., Fac. of Sci., Toho Univ Funabashi .

A transient increase in protein

synthesis _in vitro was observed at the
mesenchyme blastula stage of sea urchin
embryos. This stimulated activity was
localized in the mitochondrial fraction
and reconstituting experiments in which
poly U-dependent protein synthesis showed
the mitochondrial peptide elongation
factor to be essential for increasing
the protein synthetic activity in
mesenchyme blastula. The peptide

elongation factor in the mitochondrial
fraction was prepared from swimming
blastula, mesenchyme blastula and late
gastrula. In the case of mesenchyme
blastula, DEAE-Sephadex A 50 column
chromatography revealed two fractions
which were active for increasing the
protein synthesis in the reconstituting
system. The two fractions were separated
by second DEAE-Sephadex A 50 into three
active peak. But in the swimming blastula,
it showed weak stimulation. This result
suggested that stimulation in protein
synthesis of mitochondria was due to the
peptide elongation factor. To clarify the
nature of protein synthesized in the
mitochondrial fraction of sea urchin
embryos, the protein was purified with
column chromatograhy . Properties of the
syntheticprotein will be examined in future.

CELL-SUBSTRATUM INTERACTION IN DIFFERENTI¬
ATION OF MOUSE UTERINE EPITHELIAL CELLS IN
SERUM-FREE PRIMARY CULTURE.

H. Fukamachi1, and J. A. McLachlan2 .

Izool. Inst., Fac. of Sci., Univ. of Tokyo,
Tokyo, and 2Lab. of Reprod. Develop.
Toxicol., Natl. Inst, of Env. Health Sci.,
Res. Triangle Park, NC, USA. _

Uterine epithelial cells were obtained
from 17- to 21-day old CD-I mice, and were
cultured in a 1:1 mixture of Ham's F-12 and
Dulbecco's modified Eagle's medium (0.1 mM
Ca) supplemented with bovine serum albumin,
insulin, transferrin, and epidermal growth
factor.

When seeded on type I collagen gels,
uterine epithelial cells dedifferentiated
to form simple squamous monolayer, and pro¬
liferated rapidly. When seeded on Matrigel
(a reconstituted basement membrane-like
substratum) , they proliferated slowly, but
differentiated to form polarized columnar
monolayer with junctional complexes, micro¬
villi, and lipid granules, comparable to
normal uterine epithelial cells in vivo. No
significant effect of estradiol-17- (2> could
be found on the proliferation and differen¬
tiation of uterine epithelial cells.

These results indicate that the prolifer¬
ation and differentiation of uterine epi¬
thelial cells are more easily affected by
substratum than by estrogens in our culture
system.

1282

Developmental Biology

DB 134

A VOLATILE GROWTH-PROMOTING ACTIVITY
M.Kariya and H.Namiki

Dept, of Biol., Sch. of Educ., Waseda
Univ. , Tokyo.

Effects on the cell proliferation of
the cover lid of the culture plate were ex¬
amined. Human dermal fibroblast cells (HDF)
were cultured in a multiplate either cover¬
ed with lid (+lid) or without lid (-lid).
The cells in the (-lid) plate significantly
less proliferated than those in the (+lid)
plate. CO2 concentration above the culture
media was always as constant as of the at¬
mosphere in the incubator ( 5% ) . The phenome¬
non therefore may be due to the volatil¬
ization of a f actor (s) either secreted from
the cells in culture or contained in the
supplemented serum. The latter possibility
was eliminated by an experiment using thor¬
oughly dialyzed serum, showing practically
the same result as that of the initial ex¬
periment. We thus concluded that HDF se¬
creted a volatile growth-promoting factor (s
) . We also carried out an experiment uti¬
lizing HeLa cells where no lid-effect was
observed. Explanations of no effect are as
follows: (1)HeLa cells may be little depen¬
dent on the factor, (2)they might secrete a
considerable amount of the factor, or(3)
they might secrete little or no such fac¬
tors .

DB 135

SERUM-FREE CULTURE OF MARINE FISH CELLS.
N.Ebitani1, T.Kubo2 and S .Nadamit su 3 .
1Biol. Lab., Shohoku Coll., Atsugi, 2Biol.
Lab., Sophia Univ., Tokyo, and 3Biol. Lab.
Hirosima Women's Univ., Hirosima.

Cells were obtained from the fins of the
scorpion fish, Sebasti scus marmoratus , and
cultured in a serum-free medium (GIT).

After 4 months culture, these cells were
cultured in four kinds of culture medium,
serum-free medium (GIT), GIT containing 5%
FBS (G-5FBS), GIT containing 15% FBS (G-15
FBS) and 199 containing 15% FBS (199-15FBS).
The cells cultured in G-15FBS showed the
greatest growth. The growth of the cells
was higher in GIT than in 199-15FBS. The
population doubling time was 37.4 (G-15FBS),
42.7 (G-5FBS), 49.3 (199-15FBS) and 51.2 hr.
(GIT). The extent of cellular attachment
was 83.3 (G-5FBS), 71.3 (199-15FBS), 70.2
(GIT) and 68 . 1% ( G-15FBS) . The morphologi¬
cal difference was not observed among the
cells cultured in four kinds of medium.

These results showed that primary cul¬
ture of fish diploid cells and its long¬
term serial transfer could succeed in a
serum-free medium, as well as using serum
containing medium. Addition of serum to
the culture medium did not lead to inhibi¬
tion of growth but led to promotive effect
of growth in this cells. Cultures derived
in serum-free medium contained little num¬
ber of cells with karyotypes in the tetra-
ploid range.

DB 136

REDIFFERENTIATION OF ONCOGENE-TRANSFECTED
FISH PIGMENT CELLS.

T. Akiyama and J. Matsumoto. Dept of Bio-
logy, Keio Univ. Yokohama-Hiyoshi , Japan

It is shown that (1) enhanced expression
of ras gene is observed in goldfish ery-
throphoroma cells, when examined by Nor¬
thern blotting (Nemoto et al. 1984) and (2)
pp 60c-src protein kinase activity is mar¬
kedly increased in melanomas appearing in
swordtail-platyf ish hybrids (Anders et al.
1984). Based on these findings, attempts
have been made to produce tumorous pigment
cells by introducing such oncogenes into
normal pigment cells. For this purpose,
xanthophores, melanophores and chromato¬
blasts isolated from goldfish skins are
subjected to transfection of v-H-ras , v-
myc, and v-src genes, and subsequent selec¬
tion by geneticin and a lowered concentra¬
tion of ACTH or MSH. Transf ectants show a
varying degree of growth, crisscross-like
arrangement in a monolayer culture and
colony formation in semi-solid soft agar.
The vast majority of less to non-pigmented
transf ectants spontaneously redifferen¬
tiated into heavily pigmented xanthophores
or melanophores within three months. One
cell clone transfected with v-myc and v-src
is successfully cultivated over one year
under constant growth. Immunoblotting dis¬
closed that the leve'ls of oncogene products
in such transf ectants are comparable to or
lower than those in ery throphoroma cells,
whilst higher than those in normal pigment
cells .

DB 137

THE DIFFERENTIATION IN VITRO OF MOUSE NEURAL
CREST CELLS

K. Ito and T. Morita. College of General
Education, Dept, of Biology, Osaka Univ.,
Osaka

In the present study, we established the
culture system" in which three cell types
derived from neural crest cells( sensory
neurons, adrenergic neurons, melanocytes)
differentiated. When 'mouse trunk neural
crest cells were cultured in the medium only
with serum at low concentration (5% calf
serum(cs)), substance P-immunor eacti ve
sensory neurons appeared. These neurons
also displayed the immunor eactivity against
6 8 K D - n e u r o f i 1 a m e n t . In the medium
containing 2-4% chick embryo extract along
with 5% CS, on the other hand, adrenergic
phenotype exclusively expressed. Judging
from the difference between sensory and
adrenergic neurons of the nutrient
requirement ( medium composition) and cell
proliferation examined with BrdU method, we
conclude that sensory and adrenergic neurons
develop from distinct populations of mouse
neural crest cells and respective cell
lineages are probably separated early in the
neural crest development.

The differentiation of mouse neural crest
cells into melanocytes was remarkably
promoted by the addition of a phorbol
ester(TPA) to the medium. This result
suggests that protein kinase C activity
influences the melanocyte differentiation in

mouse .

Developmental Biology

1283

DB 138

CELL CULTURE CONDITIONS FAVORABLE FOR
CHICK EARLY EMBRYONIC NEURAL CELLS TO
DIFFERENTIATE IN VITRO.

T. Kobayashi. Department of Biochemistry,
Jikei University School of Medicine, Tokyo

An in vitro culture of embryonic neural
cells that supports their entire differ¬
entiation including neuroblast migration,
synaptogenesis and myelination would pro¬
vide an excellent model system to investi¬
gate brain development. Cells isolated
from cerebrum cortex of 6 day old chick
embryo remained bipolar in their shape and
could not differentiate morphologically
any more when they were cultured in a me¬
dium containing 1 0% of non-chicken serum.
On the other hand, cells isolated from 18
day old embryo and cultured in the same
medium showed semi-differentiated multi¬
polar shape with short neurites. Neuron¬
like cells perished within 3 weeks in both
culture. Addition of 5% chicken serum and
10% chick embryonic brain extract to the
culture medium sustained their survival
and gave them enough time to achieve mul¬
tipolar cell shape with long neurites.
Cell aggregate formation by gyratory shak¬
ing just after cell isolation remarkably
improved culture conditions for 6 day old
embryonic cells. They appeared as differ¬
entiated as 18 day old embryonic cells.
Addition of 25 mM KC1 or 0.1 mM cAMP ana¬
logue was effective for the differentia¬
tion of 18 day old embryonic cells but not
of 6 day old embryonic cells.

DB 139

ESTABLISHED ASTROCYTE-LIKE CELL LINE
DERIVED FROM XENOPUS LAEVIS.

M. Asashima, M. Aikawa , T. Sasaki, M. Mura¬
kami and K. Shimada. Dept, of Biology,
Yokohama City University, 22-2 Seto, Kana-
zawa-ku, Yokohama 236.

We previously reported long-term culti-

vation of four cell types derived from a
Xenopus laevis tumor. In this time, we
characterize a cell line established from
long-term cultivation of one of these cell
types, the neuroid cells. This new cell
line has proliferated in standard amphibian
culture medium (ACM) for over 4 years (800
generations) . We have renamed these cells
" astrocyte-like cells (XAY ) " because they

resemble central nervous system (CNS)
astrocytes morphologically and also react
with antibodies specific for glial fibrill¬
ary acidic protein (GFAP ) and S-100, the
molecular marker for mammalian and chicken
astrocytes. This is the first report of an
established amphibian neural cell line,
and the first report from any animal of
cultured astrocytes originating outside the
CNS. These cells grow with a doubling time
of 45 hr. These cells vary in size and
shape ( cell diameter, 20 to 200 pm ) , but
fall into two basic morphologic types: the
process-bearing type and the flat membran¬
ous type. In the process-bearing type a
number of branching dendritic processes,
some of which carry surface enlargements,
emerge from a large perikaryon. The flat
cell membranous type cells bear no evident
processes .

DB 140

A GOLGI STUDY ON THE NEURONS AND NEUROGLIA
IN EARLY DEVELOPMENTAL STAGES OF THE DOMES¬
TIC HEN

N. IWAHORI , Dept, of Anat., Fac . of Med.,
Nagasaki Univ., Nagasaki.

Morphological features of the neurons
(NR) and neuroglia (NG) in early develop¬
mental stages were studied in the spinal
cord of the domestic hen using the rapid
Golgi method. The NR in early developmental
stages were unipolar spindle cells located
near the internal limiting membrane. From
the somata extended a process which was
smooth in contour and travelled in an ex¬
ternal direction branching out a few long
collaterals. Upon reaching the vicinity of
the external limiting membrane (ELM) , the
processes swung rostrally, caudally or
ventrally to run further distally. The NG
appeared later than the NR. The NG in early
developmental stages were also unipolar
cells observed near the internal limiting
membrane. The somata of the NG were spindle-
shaped and generated a process which was
thicker and coarser than that of the NR.
These processes ran externally, branching
out numerous fine short collaterals and
ended at the ELM, forming terminal swell¬
ings. Thus, while the NR and NG in early
developmental stages were unipolar spindle
cells, the morphology of the processes of
these two cells differed: the processes of
the NR were smooth and extended for a long
distance bypassing the ELM, whereas those
of the NG were thick and terminated at the
ELM.

DB 141

IN VITRO CELLULAR AGING AND CELL MIGRATION:
ON THE DIFFERENCE BETWEEN HUMAN LUNG AND
SKIN FIBROBLASTS.

H. Kondo, Y. Yonezawa and T.A. Nomaguchi.
Dept, of Biol., Tokyo Metropol. Inst, of
Gerontol., Tokyo.

Since human fibroblasts have the
intrinsic limit of cell division potential,
they are often used as a model system of
cellular aging in vitro. Human fibroblasts
derived from different tissues have similar
features each other. But we have reported
that human fibroblasts have the hetero¬
geneity and tissue specificity as shown
from the hydrocortisone responsiveness to
cells and its specific binding. In the
present study, we also observed using
Stenn's method (Stenn, 1980) that human
lung fibroblasts had different migratory
ability from human skin fibroblasts. This
was obtained from the following results:
(1) Human lung fibroblasts (TIG-3) migrated
more slowly than human skin fibroblasts
(TIG-3S) derived from the same fetus, at
all passage levels. (2) TIG-3 cells
migrated in medium which was not
supplemented with 10% FBS as effectively as
in medium which was so, in contrast, TIG-3S
cells migrated in medium which was not
supplemented with 10% FBS much more slowly
than in medium which was so. The same
result was obtained when young adult and
elderly donor skin fibroblasts were used.
(3) TIG-3 cells were less sensitive to pH
of culture medium than TIG-3S cells.

1284 Developmental Biology

DB 142

THE EFFECT OF ALPHA-TOCOPHEROL ON THE AC¬
CUMULATION OF LIPOFUSCIN IN MOUSE CEREB¬
RAL CORTICAL CELLS IN PRIMARY CULTURE

5. Kan, S. Asha Devi, S. Kawashima. Zool.

Inst., Fac. Sci., Hiroshima Univ., Hiro¬
shima. _ _ _ ^ _

Lipofuscin is one of the biochemical

markers of aging in postmitotic cells. We
have established a method for the extrac¬
tion and quantification of lipofuscin in
cultured cerebral cortical cells, and the
effect of alpha-tocopherol on the lipo¬
fuscin accumulation was studied. Lipofus¬
cin was extracted by the method of Schmidt
-Thannhauser-Schneider with some modifi¬
cation. Autofluorescence of lipofuscin
was measured at the excitation wavelength
of 360 nm and intensity of fluorescence
of the emission wavelength between 400
and 470 nm. The auto fluorescence peak was
observed approximately at 435 nm. After 6
days of primary culture mouse cerebral
cortical cells were treated with alpha-
tocopherol acid succinate (0.01 IU/ml) for

6, 12 or 18 days. A less amount of lipo¬
fuscin was accumulated in the alpha-toco¬
pherol-treated cultures than in the un¬
treated ones. The extent of decrease was
directly proportional to the length of
treatment period and it ranged from 77 • 5
to 86.8 % of the control level.

DB 144

REDUCTION OF THE LIFESPAN OF HUMAN ENDO¬
THELIAL CELLS BY TUMOR NECROSIS FACTOR

Y.Shimada, M.Fukuda, and K.Kaji. Tokyo
Metropolitan Inst, of Gerontol., Tokyo-173.

Lacking the knowledge of cell growth
regulation by endogenous substances such as
hormones, growth factors, and cytokines,
the mechanism of cell growth and cellular
aging cannot be elucidated. We have exa¬
mined the effect of tumor necrosis factor
( TNF ) on the lifespan of human endothelial
cells (ECs) from umbilical veins, umbilical
arteries and a pulmonary artery. The fol¬
lowing results were obtained. (1) In a
short-term assay (6 days), TNF inhibited
the growth of ECs in a dose -dependent
manner. (2) In a long-term assay, TNF re¬
duced the growth ability of ECs and ceased
growth at much less cumulative population
doublings (CPDs); The CPDs of the TNF-
treated cells were 30% of those of the
untreated control cells, regardless of the
types of ECs. (3) The growth-inhibiting
effect of TNF seemed not to be correlated
with its lifespan-shortening effect. (4)
The division-arrested cells resumed cell
division after they were returned to normal
culture conditions. Their lifespan, howe¬
ver, was shortened in dependence on the
length of TNF treatment.

In conclusion, iji vitro aging of endo¬
thelial cells is modulated by TNF.

DB 143

STUDIES ON CREATINE PH0SPH0KINA3E ACTIVITY
IN PRIMARY CULTURE OF DISSOCIATED MOUSE
CEREBRAL CORTICAL CELLS.

S. Asha Devi, S. Kan and S. Kawashima.
Zool. Inst; Fac. of Sci; Hiroshima Univ;
Hiroshima _ _

Creatine phosphokinase ( CPK. L. C. 2. 7. 3-
2), a key enzyme in the interconversi¬
on of ADP and CP from ATP and creatine is
universally distributed in the CNS. Since
the cerebral cortex derives its energy mo¬
stly from glucose, the supply of ATP from
ADP and CP is one of the prerequisite
reactions involving CPK.

We have assayed CPK activity at 1, 3>

8, 16 and 21 days of culture by the coup¬
led enzyme assay scheme of Rosalki (1967).
DNA and protein levels were measured by
the methods of Gold and Shochat( 1980) and
Lowry et al.(195D*

Specific enzyme activity increased pr¬
ogressively from 1 day (6.5 mU) to 16 days
(13.4 mU) and declined at 21 days(6.8 mU).
Absolute activities were 373» 905 and 1076
mU/dish at 1,16 and 21 days respectively.
Protein accumulation was linear from day 1
(36.5 ug/dish) to 21 day (158 ug/dish) which
was four-fold greater in relation to the
former. These changes were accompanied by
a thirteen- fold increase in DNA at 21 days
in culture.

Our results indicate an age-related de¬
cline in specific enzyme activity of cere¬
bral cortical cells and may suggest a pos¬
sible decrease in turnover of ATP molecu¬
les from creatine phosphate.

DB 145

LOCALIZATION OF POLLEN PARTICLES INJECTED
INTO POST-IMPLANTATION MOUSE EMBRYOS
IN UTERO. *

Y. Kajiwara, T. Kuwana and M. Inouye.
Pathol. Sec., Natl.#Inst. for Minamata
Dis., Minamata and 3rd Dept, of
Anat. , Kumamoto Univ., Med. Sch. , Kumamoto.

Pollens (paper mulberry, 12-13 ^im in
diameter) were fixed with 0s04j rinsed in
Hanks' solution and injected into mouse
embryos through the uterine wall on day
8 . & of gestation. The embryos were
removed after 48hr and checked the
existence of the pollens in their bodies
under a dissecting microscope.

One hundred and twelve embryos survived

among 1

67 em

bryos injecte

d .

The

pollen

contr ibu

ti on

was observed

in 26 em

bryos .

Pollens

wer

e localized

in

fol

lowing

regions ;

do

rsal midlin

e in the

head

(21/26) ,

pos

terior trunk

anc

1 the

tail

(13/26) ,

and

the breast (

12/26

5) .

These

results

sugge

st that polle

ns ar

'e pas

s i vely

taken in

to em

bryos on the

mi d-

-dorsa

, 1 l-ine

of the

head

and posterior

trur

ik , a

nd on

the bre

as t d

uring the pro

cess

of

neural

tube an

d so

matopleure e

nclos

;ure .

In 3

embryos

pol

lens distri

butec

1 in

the

embryoni

c m

esenchyme al

ong

the

dorso-

ventral

rout

e resembling

to

the

neur al

crest pa

.thway

. This sugg

ests

that

neural

crest ce

11s a

re , in part ,

tr ar

is 1 oc a

ted by

the emb

ryoni

c environmen

t in mam

mal i an

embryos ,

s i

milarly to

those

; of

avian

embryos .

Developmental Biology

1285

DB 146

DEVELOPMENT OF MOUSE EMBRYOS FROM THE
PREIMPLANTAION STAGE IN VITRO. (MIGRATION
OF INNER CELL MASS THROUGH THE
TROPHECTODERM )

I.Naruse and R.Shoji. Institute for
Developmental Research, Aichi Prefectural
Colony, Kasugai, Aichi. _

It is possible to culture the mouse
embryos from the preimplantaion stage to
the forelimb bud stage for 10 days. But,
the rate of development of embryos to the
forelimb bud stage is extremely low.
Inner cell mass (ICM) makes an upward
migration through the trophec toderm on
day 3 of cultivation in the statinary
culture system. This process has been
considered to be the main barrier for the
further development. In the present
experiment, the hanging drop culture
technique was employed to culture the
mouse blastocysts in 3 - d i m e n s i o na 1
circumstance. Even by the employment of
hanging drop culture method, ICM migrated
through the trophec toderm on day 3 of
cultivation. In other words,

trophectoderm retreated from the covering
of ICM to form the ectoplacen t a 1 cone,
and transformed to the trophoblastic
giant cell. From this fact, it was
suggested that the migration of ICM
through the trophectoderm was not due to
the 2-dimensional condition of stationary
culture method, and that trophoblastic
cells transformed to the giant cells
without any places to attach.

DB 147

EXAMINATION OF CULTURE METHODS FOR PERI-
IMPLANTATION MOUSE EMBRYOS.

M.Taguchi1, H.Ohyama2 and T.Yamada1. Divs.

1 Biol . and 2Radiat. Health, Natl. Inst.
Radiol. Sci., Chiba. _

Interest in embryo culture as a screen
for environmental teratogenesis arose from
observations that mammalian embryos, both
in vivo and in vitro, are highly
susceptible to the damaging effects of
ionizing radiation. The literature on
peri-implantation embryo culture was
sparse, because the culture procedure used
was complex and the success rate was low.
In this study, we examined the procedure
developed mainly by Spindle, with
particular attention to the development of
those cultures as systems for evaluation of
embryotoxicants . Female mice ( BC3F x or
ICR) were superovulated , and blastocysts
were obtained by culturing 2-cell embryos
flushed from oviducts or zygotes fertilized
in vitro. Then blastocysts were cultured
in Spindle's medium containing amino acid
mixture, or in M280 medium we used for the
zygote culture, to the stage of inner cell
mass (ICM) outgrowth through hatching and
implantation.

About 80 % of embryos could developed
to the ICM growth stage in Spindle's medium,
but no ICM growth observed in M280 medium.
No significant difference in the embryo
development was observed between BC3F! and
ICR strains. These results suggest that the
major variations affecting the success rate
of the peri-implantation culture are the
choice of medium.

DB 148

OBSERVATION AND MEASUREMENT OF SAND DOLLAR
EGGS WITH A CENTRIFUGE MICROSCOPE
Y. Hiramoto, A. Izumi and S. Sekiguchi.
Biol. Lab. Univ. Air, Wakaba, Chiba. _

We observed eggs of the sand dollar,

Clypeaster j aponicus with a novel cen¬
trifuge microscope devised by Kamitsubo
(1 986). The nucleus and the drops of
paraffin oil injected into the egg moved
in centripetal direction in the cytoplasm
when a centrifugal acceleration was
applied. The apparent viscosity of the
cytoplasm was calculated from the speed of
the nucleus or the oil drop. In unferti¬
lized eggs, the larger apparent viscosity
was obtained when the speed of the larger
oil drop was used in the calculation.

This fact suggests the presence of a
network structure in the cytoplasm. The
apparent viscosity of the cytoplasm calcu¬
lated from the speed of the nucleus, which
was about 200 poises before fertilization,
decreased shortly after insemination and
increased thereafter. The mitotic appara¬
tus moved in centrifugal direction when a
centrifugal acceleration was applied. The
cleavage furrow of the egg in which the
mitotic apparatus had been displaced to
the centrifugal region of the cell was
formed at cell surface close to the equa¬
tor of the spindle, irrespective of the
direction of the spindle axis in reference
to the direction of the centrifugal acce¬
leration. The furrowing started earlier
in the region closer to the spindle than
the opposite region farther to the
spindle .

DB 149

BEHAVIOR OF CHROMOSOMES IN ASTERINA
PECTINIFERA EMBRYOS AT EARLY STAGES IN
THE PRESENCE OF APHIDICOLIN.

T. Saiki, Y. Hamaguchi. Biol. Lab., Tokyo
Inst, of Technol., Tokyo.

Fertilized eggs of A. pectinif era
develop to blastulae in the presence of
aphidicolin, which is a specific inhibitor
of DNA polymerase-a (see ref. NAGANO et
al. 1981 Dev. Biol. 85, 409-415). We
reinvestigated the behavior of chromosomes
in these embryos. At first, we
microinj ected f luorescently labeled
histones extracted from calf thymus into
the oocyte to make chromosomes visible and
observed the behavior of chromosomes. The
first cleavage completed although
chromosomes did not separate after the
first metaphase. The chromosomes did not
accompany the aster of the mitotic
apparatus or did not become nucleus,
although they entered one of blastomeres
in a lump. Secondly we fixed these
embryos at various stages during cleavage
and stained microtubules indirectly with
anti-tubulin antibody and chromosomes with
DAPI, which binds to DNA. The behavior of
chromosomes mentioned in the first half
was confirmed by the experiments in the
second half. Furthermore, we found that
after the second cleavage blastomeres
continued to divide without chromosomes or
the spindle, but only with asters.

1286

Developmental Biology

DB 150

EFFECTS OF ATP ANALOGS ON CELL MOTILITY IN
SAND DOLLAR EGGS

Y. Hamaguchi and R. Kuriyama . Biol.
Lab., Tokyo Inst, of Technol., Tokyo and
2Dept. Cell Biol. Neuroana., Univ. of
Minnesota, Minaapolis, USA.

When adenosine 5'-0-( 3-thiotriphosphate)
(ATPyS ) was injected into fertilized eggs
of the sand dollar, Clypeaster j aponicus
at the concentration of more than 0.1 mM
in the egg cytoplasm, cleavage was
inhibited. At about 1 mM ATPyS in the egg
cytoplasm, furrow stopped proceeding and
relaxed. Furthermore, the hyaline
cytoplasm protruded from the egg cortex a
few min after injection, the egg shape
became irregular, and the injected egg
began to move around like amoeba.

When AMP-PNP (8,y imido-ATP) was
injected into sand dollar eggs shortly
after fertilization, the migrations of egg
and sperm pronuclei became slowly and then
stopped. When AMP-PNP was injected at
metaphase at the concentration of about 2
mM, birefringence of the spindle did not
changed, but chromosomes moved slowly
during anaphase. However, the spindle did
not elongate but shortened, and a single
nucleus was formed in the center of the
egg during anaphase and telophase.

Division did not complete though furrow
appeared usually. No effect- was observed
when AMP-PCP (6,Y methylene-ATP) was
inj ected .

DB 151

THE CORTICAL BEHAVIOR OF DIVIDING
NEUROBLASTS OF THE GRASSHOPPER.

K.Kawamura and K.Kamata. Biology, Rakuno
Gakuen Univ., Ebetsu, Hokkaido. _ _

Unequal division to produce a small
ganglion cell is determined by the
eccentric location of the spindle at late
anaphase. Cortical movements seems so far
to play a main role to shift the spindle
toward the ganglion cell (GC)-side. In
the present study, cortices of normal and
Cytochalasin B (CB) -treated neuroblasts
were marked with Latex beads, and cortical
behavior during mitosis was analyzed by
means of time-lapse cinematography.
Expansion of cap cell (CC)-side cortex and
shrinkage of GC-side cortex occurred
simultaneously during anaphase. As a
result of such antipodal movements of the
cortex, the spindle was shifted toward GC-
side, until GC-side pole of the spindle
was caught by the cortex. Initiation of
cortical expansion and shrinkage was well
coincided with beginning of anaphase
movement of chromosomes. The movements of
the beads on the cortex showed that the
equatorial cortex of the metaphase cell
developed into the cleavage furrow which
initiated at late anaphase in the cortex
close to the GC-pole of the cell. The
fact that CB-treatment interfered the
formation of cleavage furrow, the cortical
movements, and the anchoring of spindle
pole on GC-side cortex suggests the
important role of microfilaments in the
action of cell cortex during mitosis.

DB 152

CENTRIOLES IN THE CELL CYCLE OF
GRASSHOPPER NEUROBLASTS.

N . Yamashiki . Biology, Rakuno Gakuen Univ.,
Ebetsu, Hokkaido. _ _

Grasshopper neuroblasts (NB) repeat

unequal division to produce ganglion cells
(GC) in a definite direction. The behav¬
ior of centrioles in the cell cycle of NBs
was studied by electron microscopy. After
the preceding division was completed, two
centrioles from a centrosome took a random
position within a nuclear core of the NB.
A procentriole appeared beside the mother
centriole to make a duplex either at
interphase or early prophase. At late
prophase a spindle began to form between
two duplexes and oriented along the
nuclear core. When the spindle began
shifting toward the GC pole of the cell at
middle anaphase, two centrioles of the NB-
side pole duplex separated in the flatly
stretched centrosome. The duplex of the
GC-side pole of the spindle, which had
anchored the cell cortex at late anaphase,
remained unchanged in the spherical
centrosome. Two centrioles in the
daughter GC separated at early telophase,
and the pericentriolar material disappear¬
ed during interphase. No procentrioles
were formed in the GC . When anchoring of
the GC-side spindle pole to the cortex was
forcedly prevented by micromanipulation,
separation of the centrioles occurred in
the GC-side duplex at late anaphase. Some
actions of the cortex were suggested in
preventing the separation of duplex.

DB 153

ARRANGEMENTS OF MICROTUBULES DURING
BLASTULA FORMATION OF SEA URCHIN EMBRYOS.
I. Uemura. Dept, of Biol., Tokyo Metro¬
politan Univ. , Tokyo.

From the 7th cleavage to hatch stage,
distribution of microtubules (mts) was
examined of the detergent-treated embryos
of the sea urchin, Hemicentrotus pulcher-
rimus , by immuno-electron microscopy. Few
mts can be found in blastomeres upto the
8th" cleavage stage, though blastula for¬
mation began after this stage. The 9 th
cleavage stage embryos showed an arrange¬
ment of mts indistinguishable from that
in the mesenchyme blastulae, except the
less density of mts, and the columnar
shape of each blastomere, indicating the
completion of blastula formation. Height
of blastomeres changed little from the
7th to the 9th cleavage stage. From 9hr-
blastula stage with the emergence of
regional differences in cell height, the
vegetal cells elongated with a peak at
the mesenchyme blastula stage; the equa¬
torial cells continued to shorten, and
the apical cells reached bottom in cell
height around hatch stage, then began to
increase. Arragement of mts correlated
well with change of height of ectoderm
cells. Treatment of mesenchyme blastulae
with microtubule inhibitor, nocodazole,
confirmed the above correlation between
mts arrangement and change in cell
height.

Developmental Biology 1287

DE 154

A NOVEL 70KDA M I C ROT U BU L E - M 0 D U L A T I N G
PROTEIN ( 70KDA PROTEIN ) FROM STARFISH
OOCYTES

N. Hosoyal, H. Hosoya2 and H. Mohril.
iDept. Biol., Univ. Tokyo, Tokyo and
2Dept. Ultrastruct. Res., Tokyo Metro.
Inst. Med. Sci., Tokyo.

A novel microtubule-modulating protein
( 70kDa protein ) was purified from ma¬
tured oocytes of starfish, A s t e r i a s
amurens i s , with taxol by means of
hydroxyl apat i te column and gel filtration
column chromatography ( Sephacryl S-300 ).
This protein consisted of a single
polypeptide chain having an apparent
molecular mass of 70,000 on sodium dodecyl
sulfate polyacrylamide gel electro¬
phoresis. Three isoforms were identified
by isoelectric focusing. The protein bound
to porcine brain microtubules in the
presence of taxol in a molar ratio of
4-5:1 ( tubulin dimer to 70kDa protein ).

In the presence of brain microtubule-
associated proteins ( MAPs ), the 70kDa
protein also bound to brain microtubules
in the similar ratio to that in the ab¬
sence of MAPs. In the absence of taxol,
the 70kDa protein inhibited the rate and
the extent of microtubule assembly in a
concentration-dependent manner.

These results show that the 70kDa
protein is a novel micro tubule- modulating
pro tei n .

DB 155

INHIBITION OF CELL DIVISION IN SEA URCHIN
EGGS MICROINJECTED WITH ANTI-51K MONOCLONAL
ANTIBODY.

M.Toriyama1, K.Ohta1, Y . Hamaguchi2 , S.Endo1
and H. Sakail. iDept. of Biophys. and Bio-
chem. , Fac. of Sci., Univ. of Tokyo, Tokyo,
2Biol. Lab., Tokyo Inst, of Technol., Tokyo.

We have previously presented evidences
that the 51-kD protein is a component of sea
urchin mitotic centrosomes, possibly involv¬
ed in the microtubule initiation in the for¬
mation of the mitotic apparatus. We pre¬
pared monoclonal antibody, HP-1, against the
51-kD protein. Although HP-1 did not
inhibit reconstruction of asters in vitro as
did the polyclonal antibody, it inhibited
cell division depending on the stage of
microinjection during cell cycle in four
species of sea urchins. When injected at
the stages between monaster and prophase,
most of the eggs failed to divide. When
injected around prometaphase, some eggs did
not cleave, but some showed abortive furrow.
Spindle formation was largely suppressed
showing only a spherical mass of birefrin-
gent materials. The HP-1 injection became
ineffective after metaphase and normal clea¬
vage ensued. However, the second cleavage
never occurred. Using Clypeaster eggs with
which the mitotic stage can easily be as¬
signed, the transitional stage for the in¬
hibition of cleavage was determined to be
prometaphase. It was suggested that the 51-
kD protein concerns the formation of the
centrosome and the function of the spindle
in mitotic sea urchin egg.

DB 156

ROLE OF 51-KD PROTEIN IN MITOTIC APPARATUS
(MA) FORMATION AS A COMPONENT OF MA MATRIX.
K.Ohta, M.Toriyama, S.Endo, Y. Hamaguchi1 ,
and H. Sakai. Dept, of Biophys. and Biochem.,
Fac. of Sci., Univ. of Tokyo, Tokyo, 1Dept .
of Sci., Tokyo Inst, of Tech., Tokyo. _

The 51-kD protein, a major MA-associated
protein in sea urchin eggs, localizes in the
centrosome, the basal region of asters, and
the spindle which led us to study possible
function of the 51-kD protein in MA besides
the role in microtubule( MT ) nucleation in
the centrosome previously reported. Indi¬
rect immunofluorescence staining revealed
that this protein becomes to be localized in
the nuclear region just prior to formation
of spindle MTs. Using colloidal gold-anti¬
body, it was further shown that the 51-kD
protein is associated with amorphous struct¬
ures surrounding rather than in direct con¬
tact with the MTs. These structures remain¬
ed in the insoluble component of isolated
MAs after depolymerization of MTs at 0°C.
Colcemid treatment caused almost total depo¬
lymerization of MTs in MA. However, the re¬
gion where MA had remained as a clear cyto¬
plasmic zone from which granules were ex¬
cluded. It was demonstrated that the 51-kD
protein and dimeric or oligomeric tubulin
are localized in this region, named MA matrix
(MAM). It was suggested that the 51-kD pro¬
tein in MAM may have important contributions
to the formation of MA.

DB 157

FORMATION OF MINIASTERS IN THE CYTOPLASM OF
HEXYLENEGLYCOL TREATED SEA URCHIN EGGS.
S.Endo, M.Toriyama, K.Ohta and H. Sakai.

Dept, of Biophys. and Biochem., Fac. of Sci.,
Univ. of Tokyo, Tokyo

We previously reported that many mini¬
asters were formed throughout the cytoplasm
of 5% hexyleneglycol treated sea urchin eggs
at metaphase and MTOG-like granules were
found in their centers. We investigated
the^dependence of the miniaster formation on
cell cycle by immunofluorescence microscopy
using whole mount cells. In the eggs treat¬
ed at prophase, a small number of MT frag¬
ments was observed and in those treated at
prometaphase, a few miniasters and many MT-
fragments were seen. When treated at mata-
phase, we found a great number of miniasters,
250-350 in one egg. In contrast, none of
miniaster was seen in the eggs treated at
anaphase, though many long MTs were spread¬
ing among the cytoplasm. In the eggs treat¬
ed at telophase, we scarecely noticed MT
structures in the cytoplasm. We also ob¬
served that the 51-kD protein was localized
in the region of the miniasters. It was
suggested that the aster forming activity
of MTOG is cell cycle dependent with the
highest activity at metaphase and that acti¬
vated cytoplasmic MTOGs that contain the 51-
kD protein form many aggregates thereby
causing the formation of numerous miniasters.

1288

Developmental Biology

DB 158

DB 160

THE INTERACTION OF A PAIR OF ASTER MODELS
POLYMERIZED ENDOTHERMICALLY
M.Yamanaka1 , F . Tamaki 2 and S.Xshizaka .

^ Dept . of Physics, Nihon Uni v . , Tokyo,

^ I n s t , of Biol. S c i . , Univ. of Tsukuba ,
Tsukuba .

We observed the interaction of a pair

of aster models polymerized endo¬
thermically, under a polarizing microscope
to clarify the role of asters in cell
division .

A pair of m i c r o e 1 e c t r o d e s , (tip
resistance : 1-10M fl) was inserted in a cold
solution of tubulin. Joule's heat was
generated at the tips of the micro¬
electrodes. A temperature gradient was
produced radially from the tips. Tubulin
polymerized endothermically along the
thermal gradient, forming a pair of aster
models, which competed with each other.

A spindle like structure was formed
between the tips when the microelectrodes
were fixed. Astral rays growing from the
tips oriented in anti-parallel directions.

On the other hand, when we mechanically
moved one of the m i c r o e 1 e c t r o d e s , the
astral rays growing from the tips took two
different states according to the velocity
of the mechanical movement.

(1) When the velocity was very low (about
lpm/min), the astral rays oriented in
an t i - p a r a 1 1 e 1 directions, and formed a
spindle like structure.

(2) When the velocity was higher, the
astral rays temporarily oriented in
parallel directions along the equatorial
plane. Then the rays changed gradually
from parallel to anti-parallel, when the
aster models were pulled apart slowly and
even when left untouched for 10-30 mins.

ACTIN ASSEMBLY AND INHIBITION OF NEMATODE
MOTILITY INDUCED BY THE METABOLIC IN¬
HIBITOR , NaN3

H.Hosoya and H . Takano-Ohmuro .

The Tokyo Metropolitan Institute of Medi¬
cal Science, Tokyo.

We found that the inhibitor of energy
metabolism, sodium, azide ( NaN3 ) , can in¬
duce polymerization of actin. This
effect was studied by viscometry,
spectrophotometry and electron microscopy:
(a) NaN3-induced assembly of actin
produced normal microfilaments. (b)
The initial rate of actin polymerization
was lower than that of KCl-Mg induced as¬
sembly. (c) Depolymerization of F-
actin occurred at a relatively high con¬
centration of NaN3.

It is known that N aN 3 inhibits
nematode ( C . elegance ) motility. Our
fluorescence and electron microscopic ob¬
servations showed that many actin fila¬
ments were produced in the NaN3-treated
nematode. These results suggest that
there is an apparent linkage between in¬
duction of actin polymerization and in¬
hibition of nematode motility by NaN3 .

DB 159

ISOLATION OF CLEAVAGE FURROWS FROM DIVIDING
SEA URCHIN EGGS AND I M M U N 0 F L U 0 R E S C E N T
ANALYSIS OF MYOSIN IN THE CONTRACTILE RING.
S.Yonemura , S.Tsukita^ and I.Mabuchi .

' D e p t. of Biol., College of Arts and Sci.,
Univ. of Tokyo, Tokyo, 1 U e p t . of
Ultrastruct. Res., The Tokyo Metropolitan
Inst, of Medical Science, Tokyo. _ _

A s a direct approach to clarify the
mechanism of cytokinesis at a molecular
level, we attempted to develop a new
procedure to isolate cleavage furrows from
dividing sea urchin eggs. We employed a
detergent treatment of C 1 ypeaster ,j a p o n i c u s
eggs and subsequent sucrose gradient
centrifugation. Using light and electron
microscopies, we confirmed that cleavage
furrows were isolated with high purity,
maintaining the structural integrity of the
contractile ring. Two dimensional gel
electrophoresis and immunoblotting using
antibodies against starfish oocyte myosin
revealed that actin, myosin, ot-actinin and
spectrin were contained in the isolated
cleavage furrow fraction.

Egg cortices containing cleavage furrows
were isolated on protamine-coated
coverslips and myosin was localized by
immunofluorescence microscopy after various
treatments on the cortices in order to
determine how myosin and the contractile
ring were associated. Both myosin and actin
filaments disappeared from the furrow
region by extraction with 0.6M KC1 plus 5mM
ATP or with 45K protein which severes actin
f i laments.

DB 161

ENVIRONMENTAL CONTROL OF GONAD MATURATION
IN THREE SPECIES OF SEA URCHINS
M. Yamamoto,1 K. Sakairi,2 M. Ishine 2
and M. Yoshida.2

1 Dept. Biol., Okayama Univ., Okayama.

2 Ushimado Marine Lab., Okayama Univ.,
Ushimado, Okayama.

Juveniles of Pseudoeen tro tus depressus ,
Anthocidar i s crass i spina and Heroi cent-
ro fus pul cherr i mus . each derived from
zygotes of the same batches, were reared
under various light and temperature
regimes. The light regimes tested were
constant light, constant darkness, and
out of phase and in phase with ambient
daily photoperiod. Stages in game to-
genesis were determined according to a
histological standard.

In all the three species, gonadal matu¬
ration proceed irrespective of photic
conditions tested. In P. depressus kept
at constant 20°C , the gonads matured
within a year. In A. crassispina, the
gonads matured within a year at constant
20 or 2S°C but did not at constant 15 °C .
In H_s_ pul eherr i mus . the gonads began to
mature after animals experienced tempe¬
rature drop (e.g. , 2 5°C to 1 5°C ) but
remained immature as long as animals were
kept at constant temperatures (15, 20 or
25 °C ) . These results indicate that not
the light but the temperature is an im¬
portant factor to determine the breeding
season in these sea urchins.

Developmental Biology

1289

DB 162

IN VITRO TESTICULAR DIFFERENTIATION IN
THE SEXUALLY SEMI DIFFERENTIATED SPECIES
OF ANURAN AMPHIBIANS.

Akihiko Tanimura and Hisaaki Iwasawa .
Biol. Inst., Niigata Univ . , Niigata.

DB 164

Protein secretory patterns during the
development of rat testis and epididymis
H. Ueda and S. Fuj imoto

Dept, of Anat. , Sch. of Med., Univ. of
Occup. and Environ. Health, Kitakyushu
807, Japan

Differentiation of primordial gonads
into the testicular structure in vivo and
in vitro was examined in genetic male
tadpoles of Rana catesbeiana and
Rhacophorus arboreus . In R. catesbeiana
tadpoles in the toe development stage,
gonadal development was resumed when the
temperature was raised from 1 0 °C to 22°C
in vivo , and the primordial gonads were
differentiated into testicular structure.
On the other hand, in vitro (70% MEM
Hanks solution containing 7% fetal bovine
serum, at 22°C), the differentiation was
seen on the 14th day. But the differen¬
tiation was not recognized in the
primordium containing many auxocytes. In
Rh. arboreus tadpoles, the sexual
difference in primordial gonads was
recognized histologically from the
forelimbs stage onward. The medullary
cell mass was reformed on the 5th day
later from this stage, and cell cords
were formed on the 7-9th day in vivo . On
the other hand, in vitro , the differen¬
tiation into the testicular structure was
seen on the 4th day in the primordial
gonad removed in the forelimbs stage.
This differentiation, however, was not
recognized in the gonadal primordium
removed in the toe development stage.

During the development, testis and
epididymis synthesize the proteins with
possible roles in spermatogenesis and
sperm maturation. We have used two dimen¬
sional poly-acrylamide gel electrophoresis
and autoradiography to study the time
course secretion of [ Jmethionine
labeled protein accumulated in the medium
of incubated testis and epididymis tissues
from fetal to mature rats. In the fetal
stage, testis and epididymis displayed al¬
most similar protein secretory patterns;
F 3 7 ( Mr /pi 3 7 , 5 0 0/ 4 . 63) and T35

(35,000/4.67) were prominent. During the
testicular development, T35 disappeared at
the onset of spermatogenesis, and Sertoli
cell-specific secretory proteins S70, S45

and S35 became progressively prominent.
In the epididymis, F37 and T35 were
present until sperm was detected in the
lumen, and disappeared at the time of
sperm maturation. There was no sig¬
nificant changes in protein secretory pat¬
terns between the three portions of
epididymis (head, body and tail). The
significance of F37 and T35 in sper¬
matogenesis and sperm maturation is under
investigation.

DB 163

EFFECTS OF TAMOXIFEN ON THE POPULATION OF
GERM CELLS IN THE GONAD OF DEVELOPING FEMALE
QUAIL.

Y. Araki and N. Yamamoto. Dept, of Physiol..
Gifu Univ.. Sch. of Med.. Gifu.

Tamoxifen (TAM) (0. 2m gj was injected into
yolk sac of fertilized egg of genetic female

quails on the zero day stage. The ovaries of
females from the sixth to the thirteenth day
of incubation were removed. Various measure¬
ments were made on histological sections with
a D i g i gr amme r . The maximal dimension of
gonads (cortexes and medullae) was determined
from the surface area of cross sections and
the number of germ cells of them was counted
out. Moreover the proportion of germ cell
numbers to the area of the go n a d ( 1 0~ 2mm2) was
estimated and compared with that of treated
and control females. The left cortex of
treated females showed 13 cells on the sixth
day. 35 cells on the ninth day and 61 cells
on the thirteenth day, and exhibited an in¬
crease over that of the control (18 cells on
the sixth day. 31 cells on the ninth day and
42 cells on the thirteenth day). On the other
hand, the left or right medulla in treated
females contained 4 cells on the sixth day.

3 cells on the ninth day and 1 cell on the
thirteenth day. and was similar to that in the
control (5 cells on the sixth day. about 3
cells on the ninth day and 1 cell on the
thirteenth day) .

These results indicate that TAM retarded
cortical proliferation in the left gonad and
increased the left or right medulla: the fe¬
male gonad develops into a testis-like size.

DB 165

SEMINAL PLASMA LECTIN OF THE SEA URCHIN,
HEMICENTROTUS PULCHERRIMUS .

H.Watanabe and T. Suyemitsu. Dept, of Regulation
Biology, Fac. of Science, Saitama University,

Urawa. _ _

Hemagglutinating activity on trypsin-treated,
formaldehyde-fixed, human erythrocytes of A blood
type was detected in the seminal plasma of the sea
urchin, Hemicentrotus pulcherrimus, whereas no
hemagglutinating activity on the human erythrocytes
was detected in the coelomic fluid of the sea
urchin. The hemagglutinating activity requirred
calcium ions. When calcium ions were removed, no
activity was observed. The following sugars failed
to give a significant inhibition against hem¬
agglutination of human erythrocytes when tested up
to a concentration of 3.3 mM: D-glucose, D-galac-
tose, D-mannose, L-fucose, methyl -d-D-mannoside,
N-acetyl neuramic acid, lactose, D-glucosamine, D-
galactosamine, D-mannosamine, D-GlcNAc, and D-
ManNAc. Only D-GalNAc showed an inhibitory effect
against two units of the hemagglutinating activity
at the concentation of 8 jjM. Seventy-five percent
of the activity was lost when the hemagglutinin was
boiled for 30 sec.

This hemagglutinin was purified by affinity
chromatography on GalNAc-Sepharose column and high
performance liquid chromatography on Tskgel G3000SW
column. The purification procedure of the hem¬
agglutinin from the seminal plasma resulted in
about a 24-fold increase in the specific activity.
On sodium dodecyl sulfate-polyacrylamide gel
electrophoresis, it showed a single protein band
with a molecular weight of 12,000 daltons. The
amino acid composition of the hemagglutinin was
analyzed.

1290

Developmental Biology

DB 166

LECTIN BINDING IN TESTIS OF FISH ORYZIAS LATIPES .

N. Shibata1, S. Haraaguchi2, and H. Mitsui1 ‘Dept.
Biol., Fac. Sci., 2Dept. Biol., Coll. Gen. Educ.,
Niigata Univ. , Niigata _ _

Thirteen HRP-conjugated lectins were used to
study the changes of sugar-terminals on cell
surfaces in adult testis of Oryzias latipes. In
germ-line cells, BPA strongly stained type A
spermatogonia and weakly type B spermatogonia,
while LPA stained nuclei of spermatocytes and of
spermatids. SBA stained type A spermatogonia
irregularly; some were stained but others were
weakly or not stained. This observation possibly
reflects the difference of type A spermatogonia
between stem-type cells and those destined to
differentiate into type B spermatogonia.

LCA and RCA-I strongly stained all types of
Sertoli cells. BPA stained strongly Sertoli cells
around type A spermatogonia (type I), but weakly
the cells of germinal cysts (typell). Also MPA and
PNA stained type A Sertoli cells but not type II.
Sertoli cells which have phagocytoic activity of
residual bodies were stained by BPA and weakly by
MPA, PNA, and WGA. Sertoli cells which structure
walls of efferent duct was stained by BPA, Con-A,
MPA, PNA, SBA, and WGA. Thus the surface of
Sertoli cells as well as of spermatogenic cells do
differentiate according to the progress of
spermatogenesis and these differentiation might be
involved in the cooperative mechanism between germ
cells and Sertoli cells.

DB 167

THE MUTANT GENE ter CAUSES GERM CELL
DEFICIENCY, NOT ACCOMPANIED BY TESTICULAR
TERATOCARCINOGENESIS, IN ter CONGENIC MICE.
M. Noguchi, 1, C. Watanabe, 1, H. Kato, 2 and K.
Moriwaki, 3. 1. Biol. Inst. , Fac. of Sci. , Shi¬
zuoka Univ. , Shizuoka, 2. Central Inst, for
Exp. Animals, Kawasaki, 3. Natl. Inst, of
Genetics, Mishima. _

A recessive mutant gene ter causes
germ cell deficiency, accompanied by a
high incidence of testicular teratomas in
129/Sv-ter strain mice (T. Noguchi & M.
Noguchi, 1985) . In order to analyse the
influence of a genetic background on the
ter action in the ter/ter mice, we have
backcrossed animals carrying the mutation
to LTXBJ and C3H/HeJ strains for N1-N4
generations. Both testes and ovaries were
examined histologically in offspring from
matings of N1-N4 (+/ter) x 129/Sv-ter
(+/ter) or N1-N4 (+/ter) x N1-N4 (+/ter) .
About 26% of examined gonads reduced in
mass and appeared deficient in the number
of germ cells. These would be ter/ter
genotypes. The incidence of such cases
was close to the theoretical frequency of
25%. The incidence of testicular terat¬
omas in ter/ter males descended drastical¬
ly. Thus, it was shown that the gene ter
causes germ cell deficiency, not accom¬
panied by testicular teratocarcinogenesis,
in ter homozygotes in the ter congenic
mice, LTXBJ-ter, C3H/HeJ-ter and C57BL/6J-
ter (M. Noguchi et al. , 1987) and also that
ovarian teratocarcinogenesis is not caused
by the gene ter in a LTXBJ background.

DB 168

ABNORMALITIES AT THE MOLECULAR LEVEL IN
THE TESTES HOMOZYGOUS FOR THE ter MUTATION
IN ter CONGENIC MICE.

T . Hasegawa and M. Noguchi. Biol . Inst . , Fac.
of Sci., Shizuoka Univ., Shizuoka. _

A recessive mutant gene ter causes
germ cell deficiency, accompanied by a
high incidence of testicular teratomas in
ter/ter males in 129/Sv-ter mice. The
gene ter introduced onto a C57BL/6J
genetic background, however, causes the
reduction in testis size and germ cell
deficiency, not accompanied by testicular
teratocarcinogenesis in ter /ter males (M.
Noguchi et al.,1987). In order to analyze
the gene ter action at the molecular level,
we examined the component proteins in
testis, liver, kidney and pancreas in ter/
ter , +/ter and +/+ males in ter congenic
strain C57BL/6 J-ter by two-dimensional
polyacrylamide gel electrophoresis. The
results showed that 2 specific spots and
10 spots in more amount were detected in 1,
2 and 3 months old testes in ter/ter males.
Further, we compared ter /ter testes with
cryptorchid testes of C57BL/6J, in which
spermatogenesis was inhibited. The 2
specific spots above mentioned were never
observed in the latter testes.

Thus, it is concluded that the
abnormality at the molecular level in germ
cell deficient testes in ter/ter male
could be derived from the ter gene action
in ter congenic mice.

DB 169

ANALYSIS OF STAGE-SPECIFIC PROTEIN SYNTHE¬
SIS OF THE RAT DURING SPERMATOGENESIS BY
ID- and 2D-PAGE .

M. Nakamura, S.Okinaga and K.Arai. Dept, of
OB/GYN, Sch. of Med., Teikyo Univ., Tokyo.

Round spermatids and pachytene sperma¬
tocytes isolated from rat testes by unit
gravity sedimentation have been used to
analyze the newly synthesized prc^eins in a
chemically defined medium ^ith [ J:3S Jmethio-
nine. Lactate stimulated [ JDS ]methionine
incorporation into protein in both sperma¬
tids and spermatocytes. The rate of protein
synthesis was closely related to the intra¬
cellular level of ATP. For both types of
cells, labeled proteins in the water-solu¬
ble and insoluble fractions were compared
by ID- and 2D-PAGE and autoradiography. The
synthesis of several stage-specific prote¬
ins was observed. Interestingly, in the
presence of lactate, the synthesis of six
proteins (Mr=1 4K, 1 6K, 43K, 55K, 84K and 135K)
in the water-soluble fraction of spermatids
and spermatocytes were markedly increased.
After ID- and 2D-PAGE and Western transfer,
two proteins ( 43K and 55K) were identified
as actin and tubulin. Tubulin was also pro¬
duced by the in vitro translation in the
spermatid lysate. These results indicate
that lactate may play an important role in
changing cell structure and shape in sper¬
matogenesis by regulating the synthesis of
actin and tubulin. The in vitro translation
system for tubulin production will be usef¬
ul tool to examine the mechanism for regu¬
lation of tubulin synthesis in germ cells.

Developmental Biology 1291

DB 170

PREPARATION OF MONOCLONAL ANTIBODIES
AGAINST GAMETOCYTE- SPECIFIC COMPONENTS IN
THE STARFISH, ASTERNIA PECTINIFERA.

C. Inoue and H. Shirai. Ushimado Marine
Laboratory, Okayama University,

Ushimadocho, Okugun. 107-43. _

Monoclonal antibodies against gametocyte-
specific components were prepared as
follows. For the detection of both early
oocytes and spermatocytes, components of
nuclear structures, such as synaptonemal
complex, were preferable to those of
cytoplasm. Germinal vesicles (GV) were
collected from small oocytes. One volume of
packed oocytes, free of follicle cells,
jelly and vitelline coats, were ruptured
by shaking them with about 100 volumes of
isolation medium containing detergent
(Toriyama et. al . , Cell Structure and
Function, 9, 213, 1984), and the GV were
spun down and washed. GV were first
extracted with cold 0.35 M NaCl (for non¬
histones), second with cold 0.4 M HaS04(for
histones) , and the residue was digested
with a mixture of DNase and RNase (for the
nuclear matrix) . Nuclei from other tissues
such as follicle cells, intestine, stomach,
tubefeet, coelomocytes , and embryonic
gastrula were treated similarly. SDS-PAGE
was carried out on the three fractions,
and staining patterns were compared.
Candidate gamete-specific proteins from
slab gels were used as antigens. Culture
medium containing antibodies which
immunoreacted only with GV will be used to
detect gametocytes differentiated from
gametogonia .

DB 171

INHIBITION OF IN VITRO SPERMATOGENESIS OF
XENOPUS LAEVIS BY THEOPHYLLINE AND IBMX.
S.Asakura and S.Abe, Dept, of Biol., Fac .
of Science, Kumamoto Univ. , Kumamoto. _

Dissociated primary spermatocytes from

Xenopus undergo two meiotic divisions and
early, mid-spermiogenesis in vitro with
high efficiency (Zool. Sci., 1987, Abe and
Asakura) . In this study, in order to
examine the effects of the elevation of
intracellular cAMP level on in vitro sper¬
matogenesis of Xenopus , theophylline and
IBMX, phosphodiesterase inhibitors, were
applied on spermatocytes and spermatids.

Dissociated spermatocytes were inocula¬
ted in pLL-coated dishes or in cell suspen¬
sion. The cells were cultured in L-15
medium + 10% fbs at 22°C.

In pLL-coated dishes, 20mM theophylline
inhibited progression of in vitro spermato¬
genesis at stages of diplotene and inter¬
phase II by about 90 and 100%, respective¬
ly, when incubated for 4 days. Theophylline
also inhibited acrosomal and flagellar
formation in round spermatids almost
completely. When theophylline was washed
off after 2 days' incubation, about 50-90%
recovered from the inhibition. 6mM IBMX
had similar inhibitory effect on in vitro
spermatogenesis. These results indicate
that the elevation of intracellular cAMP
level inhibits in vitro spermatogenesis of
Xenopus at several stages.

DB 172

INHIBITION OF FLAGELLAR GROWTH IN ROUND
SPERMATIDS FROM XENOPUS AND NEWT BY THEO¬
PHYLLINE (TH) , IBMX AND DIBUTYRYL CYCLIC AMP
Is . Uno , 1 H . Hiyoshi , ]k . Imakado , 2K . Kubokawa ,
2S.Ishii and ^ S.Abe, ^Dept. of Biol., Fac.
of Sci. , Kumamoto Univ. , Kumamoto and
2Pept. of Biol., Waseda Univ., Tokyo. _

The effects of phosphodiesterase
inhibitors and dibutyryl cAMP on flagellar
growth in round spermatids from Xenopus and
newt were examined in vitro .

The dissociated primary spermatocytes
were inoculated in pLL-coated dishes
(Xenopus) or within collagen matrix (newt) .

In Xenopus , 10 and 20mM TH, when incuba¬
ted from 0 to 2-day post second meiotic
division, suppressed the flagellar growth
by about 60 and 90%, respectively, whereas
5mM did not. 4mM IBMX and 1 OmM dbcAMP also
inhibited the growth by about 70 and 100%,
respectively. Washing off TH and IBMX,
flagellar growth recovered about 94%. Also
in the newt, 20mM TH or 4mM IBMX inhibited
flagellar growth by more than 80%. Radio¬
immunoassay showed 3-fold increase of cAMP
level in the presence of 20mM TH than
control. Fluorogram of 2-dimensional gel
electrophoresis following incorporation of
35s-methionine showed remarkable decrease
in the intensity of almost all spots in the
presence of TH or IBMX. These results
indicate that increase of intracellular
cAMP level suppresses flagellar growth of
round spermatids from Xenopus and newt,
probably by inhibition of protein
synthesis .

DB 173

BEHAVIOR OF MITOCHONDRIA DURING IN VITRO
SPERMATOGENESIS OF XENOPUS LAEVIS.

M. Noyori and S. Abel Dept . of Biol.,

Fac. of Sci., Kumamoto Univ., Kumamoto.

We previously reported by electron
microscopic observation that mitochondria
assemble and attach to nuclear envelope of
Xenopus round spermatids cultured in vitro
(Zool. Sci., 1987, Abe & Asakura). In this
study, the behavior of mitochondria stained
by rhodamine 123 was observed by fluores¬
cent microscopy during in vitro spermato¬
genesis of Xenopus .

The mitochondria in primary spermato¬
cytes in prophase were localized in a
restricted region as a cluster. In meta¬
phase I and II, mitochondria were distri¬
buted evenly in the region excluded by a
cluster of chromosomes. In telophase I and
II, mitochondria were localized in the
middle region between two sets of separated
chromosomes. Just after the second division
, mitochondria were distributed evenly
throughout the cytoplasm of round sperma¬
tids. After 2 hrs following telophase II,
mitochondria began to assemble around
nuclear envelope and after 4-5 hrs most of
the mitochondria attached to the nuclear
envelope. From the experiments in which
various reagents were applied, it was
indicated that cytoskeletons , RNA and
protein synthesis in spermatids, protein
synthesis in mitochondria, ATP, Ca2+,
calmodulin, cAMP in spermatids played no
role in mitochondrial assembly around the
nuclear envelope.

1292

Developmental Biology

DB 179

SPERMATOGENESIS OF THE LEPIDOPTEROUS
INSECTS , THE CABBAGE- ARMYWORM (MAMESTRA
BRASSICAE) AND THE COMMON ARMYWOM (LEUCANIA
SEPARATA)

T . Shimizu'*' , S.Yagi^and N.Agui^

''Department of Medical Entomol- The Natl.
Inst. of Ho 1th , Shinagawa , Tokyo , Div.of Ent. ,
Natl. Inst. of Agro-Environ . Sci . , Tsukuba

A spermiogenesis proceeds at "the final

larval stage in L. separata , but at the
pupal stage in M. brassicae . According to
comparative investigation of spermiogenesis
in both insects , we attempted to make a
clear hormonal regulation of spermiogenesis
in these two species. In M. brassicae , ecdy-
steroids in the heme lymph of the nondia-
pausing pupae gradually increased after
pupation., and the c-AMP level in the testis
was higher than that in the diapausing
pupal testis.

In L . separata , sperxnatocysts from larvae
that have been ligatured between the thorax
and abdomen on day 0 showed elongated
spermatocysts 4 days after the ligature.

The initiation of spermiogenesis is well
correlated with the period ( 3rd days after
the last larval eedysis ) of PTTH secretion
for pupal eedysis, but not with hemolymph
eedysteroid surge.

We concluded that spermiogenesis is
induced by eedysteroids containing in the
testis itself, if hemolymph eedysteroid
titer is absent. However, the clear roles
of testicular eedysteroids in spermio¬
genesis are not elucidated yet .

DB 175

STRUCTURAL CHANGES IN SPERM FLAGELLAR
MEMBRANE OF APYRENE SPERM OF THE SILKMOTH
RELATED WITH THE MOTILITY
H. KASUGA, M. OSANAI AND T. AIGAKI.

Dept. Biol., Tokyo Metrop. Inst. Geront.,
Itabashiku, Tokyo.

The silkmoth, Bgmbvx morjL, has

nucleated eupyrenes and anucleated
apyrenes. The eupyrenes fertilize eggs,
while the function of the apyrenes is to
stir the viscous and heterogeneous
contents of the spermatophore, which acts
as a reactor for promoting, various
reactions. On ejaculation, immotile
apyrene sperm stored in v. seminalis
acquire motility by initiatorin contained
in the prostatic secretion. However,
they move already at passing through the
basement membrane of testis and then stop
moving in the proximal part of the d.
deferens. The flagellar membrane of the
motile apyrene sperm present in the d.
deferens and spermatophore was only half
the thickness of that of the immotile
sperm in the distal part of d. deferens
and the v. seminalis. Moreover, the
flagellar membrane of apyrenes present in
the latter organ was rich in the non -
elecctron dense spaces which formed the
regularly arranging slit-like structures
perpendicular to the membrane-surface.
The flagellar membrane of apyrenes in the
v. seminalis activated by trypsin or
initiatorin reduced the thickness to half
and the structure was very similar to that
of the in vivo motile apyrene sperm.

DB 176

ULTRASTRUCTURAL OBSERVATIONS OF SPERMATOZOA
IN THE JAPANESE WATER SHREW, CHIMARROGALE
HIMALAYCA AND ITS TAXONOMIC IMPLICATION.

T . Mori1, T.A.Uchida1 and S . Aral2, ‘Zool . Lab.,
Fac . of Agr. , Kyushu Univ. , Fukuoka , 2Gen .
Lab. Center, Kyushu Dent. College, Kita-
Kyushu .

C. himarayca has been regarded as belong¬
ing to the subfamily Crocidurinae based on
the teeth with white-colored tips. Spermato¬
zoa of the family Soricidae are character¬
ized by the presence of a much longer middle
piece, outer dense fibers arranged in a
horseshoe fashion and well-developed satel¬
lite fibrils. However, spermatozoa of C.
himalayca differ from those of crocidurines
in size and shape of the sperm head ; i.e.
the former head displays a small, spatula te
shape, but the latter a large, fan-like
shape. In addition, in the water shrew sper¬
matozoa possess an unusually long, slender
and wavy apical body and a proximal, centri-
ole filled with electron-dense materials in
its lumen, but in crocidurines they have a
serrated apical body and a proximal centri-
ole with a fistulous lumen. Spermatozoa of
the water shrew are much the same as those
of the subfamily Soricinae in the ultra-
structural feature.

The evidence presented here strongly sug¬
gests that C. himalayca belongs to the sub¬
family Soricinae rather than Crocidurinae,
together with the detection of red-colored
tips of the teeth under ultraviolet rays
(Repenning, 1967) and the karyotypic analy¬
sis (Obara , 1 985 ) .

DB 177

WGA-RECEPTOR IS INVOLVED IN SPECIES- SPECIFIC
CELL ADHESION OF SEA URCHIN EMBRYO.

J.Kamimura, K.Abe and Y.Tonegawa.

Dept. Regulation Biology, Saitama Uni vers ity,Urawa.

Former comparative studies on effects of lectins
on sea urchin development have shown that WGA
dissociates pulcherrimus embryos species- and
sugar-specifica'l ly.The current study was performed
to isolate WGA-receptor glycoprotein and to
characterize the way of its involvment in cell
adhesion.

Dissociated blastula cells of pulcherrimus
was extracted with KC1 -Tris-HCl buf feT(TMs D0mMT~
containing l%TritonX-l 00 and WGA-receptor activity
was monitored with the inhibition of hemaggluti¬
nation by WGA. The Triton extract was fractionated
with ammonium sulfate precipitation and Phenyl -
Toyopearl ,WGA-Sepharose chromatography. The glyco¬
proteins eluted from WGA-Sepharose with GlcNAc
presented three bands around molecular size 5Q-70K,
when el ectorophoresed on SDS-polyacrylamid gel and
stained with HRPO-labeled WGA.

To examine the biological activities on live
sea urchin cells and embryos, purified glycoproteins
were treated with trypsin. Resulting glycopeptides
inhibited reaggregation of dissociated cells
mediated by the cell aggregation factor. In addi-
tion,normal cell association was disturbed signifi¬
cantly when they were added to the culture medium
of demembranated embryos. From these results, it is
highly probable that WGA-receptor on the cell
surface constitutes the receptor for the cell
aggregation factor of H._ pulcherrimus embryos.

Developmental Biology

1293

DB 180

DB 178

SPICULE FORMATION IN THE AGGREGATES DERIVED
FROM MESOMERES ISOLATED FROM 16-CELL-STAGE
EMBRYOS OF A SAND DOLLAR, PERONELLA
JAPONICA

S. AMEMIYA. Misaki Marine Biological
Station, University of Tokyo, Kanagawa.

The developmental potency of the animal
half in embryos of a sand dollar, Perone 11a
j aponica, was examined. Embryos were
dissociated into single cells at the 4-cell
stage in Ca^+-free sea water. When the
quarter-embryos had each developed into 4
cells (corresponding to control embryos at
the 16-cell stage), embryos consisting of 2
mesomeres, 1 macromere and 1 micromere were
isolated. Then, each quarter embryo thus
isolated was gently shaken to dissociate
each mesomere pair from each macromere-
micromere pair. Each pair of mesomeres was
isolated using a pipette and four to eight
mesomere pairs were placed into 1 0-ml
conical glass tubes containing 0.5 ml of
culture medium (100 pg/ml streptomycin and
100 u penicillin/ml in Jamarin sea water),
followed by gentle centrifugation in order
to group them together on the tube bottom.
The mesomeres thus developing on the bottom
of each tube formed an aggregate and
started to swim in the medium within one
day. Two to three days later, the mesomere
embryoids were found to have formed
spicules. These findings show that the
mesomeres isolated by those procedures has
a greater potency to differentiate than
previously believed.

DB 179

EFFECTS OF ACTINOMYCIN D ON THE GROWTH OF
PSEUDOPODIAL CABLE AND FORMATION OF
SPICULES IN CULTURED CELLS DERIVED FROM
MICROMERES OF SEA URCHIN EGGS.

S.Kuno, K.Mitsunaga and I.Yasumasu

Dept. of Biol., Sch. of Educ., Waseda

Univ. , Tokyo.

Growth of pseudopodial cables in the
cells derived from micromeres of sea urchin
eggs was blocked by actinomycin D added to
the culture at 1 Ohr of culture, at which
the cells were going to undergo the cable
growth but normally occured when
actinomycin D was added at 1 5hr of culture.
Spicule formation in pseudopodial cable
was blocked by actinomycin D when it was
added to the culture at 1 5hr of culture but
was hardly blocked by this compound added
at 20hr of culture. The activity of protein
kinase Cf which was only detectable in
micromere derived cells after the
initiation of cable growth, was hardly
found in the cells kept with actinomycin D
from 1 Ohr of culture but was as high as in
control cells, when it was added at 1 5hr of
culture. The H+,K+,ATPase activity, which
was found at the start of spicule formation
was not found in the cells exposed to
actinomycin D from 1 5hr of culture but was
found in the cells kept with the compounds
from 20hr of culture. Transcription at
these periods probably initiates the growth
of the cable and spicule formation.

IN VITRO FORMATION OF SEA URCHIN SPICULE;
EFFECT OF BLASTCOELIC FLUID(2)

M.Kiyomoto and J.Tukahara. Dep. of Biol.,
Fac. of Sci., Kagoshima Univ., Kagoshima.

OKAZAKI (’75) demonstrated that

descendants of isolated micromeres ( Dills )
of sea urchin embryos produced spicule
when they cultured in sea water
containing horse serum. From the result
it is expected that such effective
substances are also existed in the
blastocoel of embryo. In the previous
report we have shown that DIM could made
spicule in the blastocoelic fluid(BCF)
containing sea water ('87).

In the present study, DIMS of
Anthocidaris crassispina were cultured in
sea water containing BCF obtained from
Toxopneustes pileolus blastula or in the
horse serum. In the former culture
condition, DIMs began to migrate at the
primary mesenchym stage. DIMs had many
filopodia and connected each other by
them. DIMs made spicule at the gastrula
stage. In the latter culture condition,
DIMs had remained the aggregate, had
scarcely filopodia, and made spicule
rarely. When the additional timing of
BCF delayed to 1 8hr after
insemination(early gastrula stage), the
begining of spicule formation also
delayed, although the rate of the
formation was not change. If the timing
delayed more than 1 8hr , the rate of
formation was lower.

DB 181

SYNTHESIS OF PROTEINS ENRICHED IN
VEGETALIZED SEA URCHIN EMBRYOS DURING EARLY
DEVELOPMENT.

M.Komukai, Y.Iizuka and I.Yasumasu

Dept, of Biol., Sch. of Educ., Waseda

Univ., Tokyo.

We exam ined the conditions of Li + -
treatment to obtain the maximum population
of vegetalized embryos. The most of the
embryos were completely vegetalized when
they were treated by 60 mM LiCl for 3.5 hr
starting from 2.5 hr after fertilization.
Under these conditions, we analyzed the
differences of newly synthesized proteins
between normal and vegetalized embryos by
two-dimensional polyacrylamide gel electro¬
phoresis (2D-PAGE). No significant
difference was observed until the swimming
blastula stage. However, several proteins
enriched in vegetalized embryos began to be
prominently synthesized from the mesenchyme
blastula stage, their relative rates of
synthesis reached the maximum value at the
late gastrula stage and again their
synthesis declined substantially by the
prism stage. Our results suggest that
several proteins involved in endodermal
differentiation had been synthesized before
the morphological changes occurred.

1294

Developmental Biology

DB 182

EXOGASTRULAE OF SEA URCHIN INDUCED BY THE
TREATMENT WITH A23187, RYANODINE AND
PROCAINE .

K.Kataoka, A.Fujiwara and I.Yasumasu
Dept, of Biol., Sch. of Educ., Waseda
Univ. , Tokyo.

Sea urchin embryos became exogastrulae
with undeveloped bodies at 45 hr of culture
at 20 °C, when they were kept for 1.5 hr
from the times between 2 hr and 4 hr after
insemination in the presence of A23187,
ryanodine, miconazole and several com¬
pounds, which induced Ca2+ release from
sedimentable fraction. This fraction under¬
went Ca2+ uptake at the expense of ATP.
Procaine at low concentrations, which
inhibited Ca2+ release induced by these
compounds in this fraction, blocked exogas-
trula formation but produced exogastrulae
at high concentrations, at which Ca2+
release from this fraction was activated.
Embryos, treated by Li+ for 1.5 hr at these
early stages, became quasi-normal plutei,
when the treatment was performed in the
presence of procaine and ryanodine, at
concentrations which produced only small
number of exogastrulae in the absence of
Li+. Abnormal cytosolic Ca2+ levels at
early stages probably result in exogastrula
formation.

DB 183

ONTOGENY AND ROLE OF THE BLASTOCOELIC
NETWORK IN EARLY STARFISH DEVELOPMENT.
T.Kominami1, T.Nishikata^ and N.Satoh^.
^Dept. of Biol., Fac. of Sci., Ehime
Univ., Matsuyama and ^Dept. of Zool . , Fac.
of Sci., Kyoto Univ., Kyoto. _

Monoclonal antibodies were prepared
against Asterina pectinif era blastulae,
and three of them (BN-1, BN-2 , BN-3) were
used to study the ontogeny and role of the
blastocoelic network in early strfish de¬
velopment. Although distribution of the
antigens recognized by these antibodies
were same as in the ovary during early
cleavage stages, a dramatic change occur¬
red at the 8th cleavage. Fine filaments
recognized by two antibodies (BN-1, BN-3)
were first observed in the blastocoel at
the 256-cell stage. Another type of fila¬
ments containig BN-2 biding sites appeared
at the same stage. This type of filaments
rapidly increased in number and formed
network before the initiation of gastrula-
tion. Well-developed network containing
epitopes of BN-1 and BN-3 was formed at
the mid-gastrula stage and persisted up to
later stages. However, the former network
gradually disappeared from the blatocoel
after the mesenchyme cell migration. The
results of collagenase injection into the
blastocoel, which was shown to digest the
blastocoelic network, suggested an impor¬
tant rcple of this structure on the forma¬
tion of coelomic sacs and oral opening, or
the expression of DV-axis during starfish
larval development.

DB 189

ALKALINE SUBSTANCE AND OTHER CONSTITUENTS
IN THE BLASTOCOEL FLUID OF NEWT EMBRYO.

T. Asao . Biol. Lab., Sch. of Med., St. Mar-
ianna Univ., Kawasaki. _ _

The authors studied the main cause making
blastocoel fluid of the newt gastrula alka¬
line, and the biological roles of this phe¬
nomenon in the newt’s development. The flu¬
id constituents were initially separated by
gel permeation chromatography using Cellu-
lofine GCL 25 sf . The last of four separate
peaks was fairly alkaline. Infrared spec¬
trum of the alkaline specimen was similar
to that of sodium bicarbonate but without
absorption at near 1400 cm (a characteris¬
tic feature of C03 . It was accordingly sup¬
posed that blastocoel fluid contained sodi¬
um bicarbonate and sodium carbonate. This
was ascertained by chromatography of both
sodium carbonate mixtures carried out under
the same conditions. SDS-polyacrylamide gel
electrophoresis of the first peak fraction
showed it to contain at least twenty kinds
of protein. The infrared spectrum of acidic
third peak suggested that this was possibly
a low molecular carbonate. Eventually, we
came to the conclusion that blastocoel flu¬
id contained the following: ca.40 mM of so¬
dium bicarbonate and sodium carbonate mix¬
ture causing alkalinity; a minute amount of
organic carbonate; twenty kinds of protein
from 17 to 80 kd; and several familiar in¬
organic ions including ca . 70 mEq/l Na . The
role of the alkalinity of blastocoel fluid
has remained unknown until now.

DB 185

METHYLATION OF DNA IN SEA URCHIN EMBRYOS.
K.Murofushi, M.Fujii and I.Yasumasu
Dept, of Biol., Sch. of Educ., Waseda
Univ., Tokyo.

During early development of sea urchin,
the [ 3H-methyl ] radioactivity in DNA
fraction was high in isolated nuclei
exposed to [^H-methyl] S-adenosylmethionine
at the morula stage and at the stages in
the gastrulation period. The same was found
in embryos exposed to [ ’H- methyl ] methio¬
nine. In embryos the incorpolation of [2H-
methyljinto DNA was strongly blocked by 5'-
deoxy-5 ' -S-i sobutyladenosine (SIBA). The
embryo kept with SIBA for 12 hr period
including the morula stage normally
developed when those treated embryos
were successively kept in normal sea water.
Those kept with SIBA for the same duration
including the gastrulation period, however,
because undeveloped embryos having poor
archenterons or exoguts. These suggest that
DNA methylation in the period of
gastrulation supports archenteron formation
in sea urchin embryos.

Developmental Biology

1295

DB 186

CHANGE IN ADP-RIBOSYL TRANSFERASE ACTIVITY
IN SEA URCHIN EMBRYOS DURING EARLY
DEVELOPMENT.

Y.Kamata and I.Yasumasu

Dept, of Biol., Sch. of Educ., Waseda
Univ. , Tokyo.

The ADP-ribosyltransf erase activity in
sea urchin embryos was relatively high in
nucleus fraction at the morula (6 hr after
fertilization) and early gastrula stage.
The activity was also found in plasma
membrane fraction and hardly altered during
early development. The activity was
inhibited by 3-aminobenzamide (3-ABA). The
embryos kept with 3 -ABA for 10 hr from the
time of fertilization normally developed to
plutei. Those kept with 3-ABA in a period
between 20 and 30 hr became undeveloped
embryos having small guts or exoguts, which
were very alike in their shape to abnormal
ones in which DNA methylation was blocked
by 5 1 -deoxy-5 '-S- isobutyl adenosine. [^H-
methyl] radioactivity in DNA fraction in
isolated nuclei, observed after an exposure
to [^H-methyl] S -adenosy lmethionine , was
enhanced by NAD+ and was markedly reduced
by 3-ABA. These suggest that DNA methyla¬
tion is stimulated by [ ADP-ribosy 1 ] ation of
nuclear proteins and supports gut formation
in the period of gastrulation.

DB 187

CHANGES OF CHROMATIN PROTEINS DURING THE
BLASTURA STAGE OF THE EMBRYO OF THE STAR¬
FISH, ASTERINA PECT1NIFERA.

S.Ikegami, T.Mitsuno, N.Itoh and N.Tsuchi-
mori. Dept, of Applied Biochem., Hiroshima
Univ. , H i gash i h i rosh ima .

In the embryo of the starfish, Asterina
pectinifera, the rate of RNA synthesis be¬
comes high and G1 phase of the cell cycle
becomes measurable during the blastula
stage. We have analyzed histones and non¬
histone proteins of embryonic cells of the
blastula. We have found that a molecular
species of histone HI with an apparent
molecular weight of 29 kilodaltons is
produced specifically at the midblastula
stage. We also found that a non-histone
protein with an apparent molecular weight
of 29 kilodaltons is produced specifically
during the blastula stage. These proteins
are considered to be involved in the
transtion of the chromatin structure and

DB 188

ISOLATION AND CHARACTERIZATION OF HOMEO BOX
CONTAINING GENE FROM THE ASCIDIAN, HALO-
CYNTHIA RORETZI

H. Saiga1, K.Makabe2, A.Mizokami1, N.Satoh2,
and T.Mita1, 1. Dept .Mol .Biol . , Univ.Occup.
Env. Health, Kitakyushu. 2. Dept.Zool.,
Kyoto Univ., Kyoto.

To obtain a clue to understand the func¬
tion of homeo box containing gene, we are
interested in examining spatial and tempo¬
ral expression pattern of the homeo box
containing gene in the ascidian whose cell
lineage in early development is well known.
For this purpose we have started isolation
of homeo box containing gene from the as¬
cidian, Halocynthia roretzi .

So far we have isolated 2.7 kb DNA frag¬
ment which cross-hybridized with Bombyx
homeo box. Nucleotide sequence analysis
reveals that this DNA fragment contains the
nucleotide sequence corresponding to the
4 5-6 0th amino acid residues of the homeo
box. In the immediate upstream of the 45th
codon, the consensus nucleotide sequence
for exon-intron junction is found, suggest¬
ing that putative ascidian homeo box is
interrupted by at least one intron . North¬
ern blot analysis with gonad RNA shows that
putative ascidian homeo box containing gene
is expressed, producing about 2.8 kb tran¬
script. Southern genomic blot analysis
with probes from the isolated DNA fragment
suggests that it is unlikely that a gene
which has homologous homeo box is present
in the genome .

DB 189

CLONING OF THE EARLY HISTONE GENE IN
EMBRYOS OF THE SEA URCHIN, B . PULCHERRIMUS

E. MAZDA1, I.YASUMASU1, K.YAMADA2 ,
K. AKASAKA2 AND H . SHIMADA3

^ept. of Biol., Sch. of Educ., Waseda
Univ., Tokyo. 2Zool. Inst., Fac. Sci.,
Univ. of Tokyo, Tokyo. 3Zool. Inst., Fac.
of Sci., Hiroshima Univ., Hiroshima.

The early histone gene of the sea urchin
H . -.pulcherrimus was cloned using cloned
early histone gene of the A. crassisspina
(Watanabe et.al.) as prove. Two clones,
about 3.2kb and 6 . 5kb in length, were
identified by hybridization. It is
expected that both clones construct
"histone gene cluster" and contain all
five histone genes. The nucleotide
sequence of latter clone was almost
determined and revealed its high degree of
homology of coding and flanking region
with S. purpuratus. This clone contains
H2B , H2A, HI and a part of H4 genes. H3
gene was not detected. This gene is
probably contained in not yet analyzed
region of this clone. The rest of H4 gene
will be found in another clone.

The present analysis of this clone
indicates that the early H. pulcherrimus
histone gene has the relative 3 ' —5 ' order
of the five genes , Hi ,H4 ,H2B,H3 and H2A, on
the transcribed strand in the same order
to the other sea urchin species.

functions at the midblastula stage.

1296

Developmental Biology, Endocrinology

DB 190

DB 192

MOLECULAR CLONING OF ARYLSULFATASE (ARS)
GENE FROM HEMICENTROTUS PULCHERRIMUS
K.YAMADAl, K.AKASAKAl, H.ISHIKAWAl and H.
SHIMADA2.

1. Zool . Inst . , Fac. Sci . , Univ. of Tokyo,
Tokyo .

2. Zool . Inst Fac . Sci Univ . of Hiroshima,
Hiroshima .

In the sea urchin embryos, Ars mRNA
begins to accumulate in the mesenchymal
blastula stage and is most abundant in the
prism stage. It's amount reaches about
1000-fold higher than that in the blastula
stage .

Recently we have coined the Ars cDNA
from the cDNA library of sea urchin gast-
rula ( Sasaki et al . , Europ. J. Biochem. ,
in press ) .

In this paper, we describe isolation and
characterization of the sea urchin Ars gene
with this Ars cDNA as the probe. The frag¬
ment of the Ars DNA cloned here was about
2x10^ base-pairs long, and it was interrup¬
ted by six introns. The 5' flanking region
of this Ars gene was found to contain TATAA
box as well as the CCAAT consensus sequence.

CDNA CLONING OF SEVERAL ATPASES FROM
EMBRYOS OF THE SEA URCHIN, HEMICENTROTUS

, M. Hatch1 , N.Takahashi1 ,
Fujino , K.Yamada3,

H.Shimada4 and I .Yasumasu .
'Dept, of Biol., Sch. of Educ., Waseda
Univ., Tokyo, 2 Dept . of Pharmacol., Teikyo
Univ. Sch. of Med., Tokyo, 3Zool. Inst.,
Fac. of Sci., Univ. of Tokyo, Tokyo, 4Mol.
Genetics, Fac. of Sci., Hiroshima Univ.,
Hiroshima _ _ _ _

During early development of sea urchin

embryos, the activities of H+ , K+-ATPase ,
Cl- , HCQg-ATPase and Na+ , K+-ATPase change
in a stage specific manner. We tried to
isolate cDNA clones of these ATPases to
study the developmental regulatory
mechanism of these genes' expression.

The cDNAs were identified using an
oligonucleotide probe derived from a
conserved amino acid sequence of the ATP
binding site of the aspartylphosphate
family of transport ATPases. With this
probe, we screened about 100,000 clones of
the cDNA library from prism poly(A)+RNA
which had been built into XgtIO.

We got cDNA clones coding a sequence of
ATP binding site. Northern blot analysis
revealed that these mRNAs , about 4.5 Kb
long, are developmentally regulated with
maximum expression at the mesenchyme
blastula or the gastrula stage.

PULCHERRIMUS

K.Mitsunaga
E. Mazda1 , Y.
K . Akasaka3 ,

DB 191

DETECTION BY IN SITU HYBRIDIZATION OF ARYL¬
SULFATASE mRNA- PRODUCING CELLS IN SEA
URCHIN EMBRYO.

K. Akasaka1-, T. Ueda2, T. Higashinakagawa2 ,
K. Yamadal and H. Shimada2.

1. Zool. Inst. Fac. of Sci., Univ. of
Tokyo. 2. Mitsubishi-Kasei Inst, of Life
Sci., Machida. 3. Zool. Inst., Fac. of
Sci., Hiroshima Univ., Hiroshima. _

In the sea urchin embryo, the aryl-
sulfatase (Ars) activity is known to
increase dramatically after the mesenchyme
blastula stage. We have demonstrated that
Ars mRNA exists in the unfertilized egg in
small amounts, while it markedly increases
after the blastula stage preceding the
increase of the Ars activity. In order to
detect cells responsible for the mRNA
production in the embryos, we carried out
in situ hybrydization with the use of RNA
probes. The RNA probes were synthesized
in vitro from Ars sequences in a trans¬
cription vector-pT7T3 . The Ars seauence
(EcoRI-Xbal: coding region) was inserted
into the EcoRI and Xbal sites of pT7T3.

This template was truncated with EcoRI for
synthesis of an antisense transcript
with T7 RNA polymerase, or with Xbal for
synthesis of a sense transcript with T3
RNA polymerase. Ars probes were labeled
with 33S-UTP. Hybridization to the
paraffin sections of gastrulae revealed
that Ars transcripts are restricted to
ectoderm cells.

EH 1

ISOLATION AND CHARACTERIZATION OF

ALDOSTERONE-RELEASING SUBSTANCES IN

BULLFROG ANTERIOR PITUITARY GLAND
S.Iwamuro, S.Kikuyama, and H.Hayashi1.
Dept, of Biol., Sch. of Educ., Waseda
Univ., Tokyo. ’Inst. of Endocrinol.,
Gunma Univ., Maebashi . _

Two active substances which
stimulate aldosterone release from
bullfrog or Xenopus laevis interrenals
both in vivo and in vitro were isolated
from bullfrog anterior pituitary
glands by extraction with acid-acetone
and separation by a C18 SEP-PAK
cartridge, Sephadex G-50 column and
reverse-phase HPLC . At the final step
of purification by HPLC, one active
principle (A) emerged at a

concentration of 45% MeCN and the other
(B) at a concentration of 50% MeCN. A
and B have molecular weight of 4,000
and 5,000 respectively. Amino acid
sequences of A and B are as follows :
A=EDIANYPILNLFPGSDNQNTQEGIMEDEALDRQDS,
B=XCEESN'KCTDLSSEDGILECIKACKMDLSAESPVFP-
GNGHMQPLSENI . A shares 54% sequence
homology with _X* laevis joining peptide
of POMC and B shares 63% sequence
homology with bovine N-terminal peptide
of POMC.

Endocrinology

1297

EN 2

ISOLATION OF EEL PROLACTIN (PRL) .

R. Suzuki1 , A. Yasuda2 , H. Kawauchi2 , T.
Hirano1 . 'Ocean Research Inst., Univ. of
Tokyo, Tokyo. 2School of Fish. Sci.,
Kitasato Univ., Iwate.

A highly purified PRL was isolated
from the pituitary of the Japanese eel
(Anguilla j aponica ) by extraction with
acid acetone, gel filtration on a Sephadex
G-75 column equilibrated with 0.1 N acetic
acid, reversed-phase HPLC on TSK-gel ODS
120T and on TSK-gel TMS 250, with a yield
of 0.8 mg/g of wet tissue. At each stage
of purification, fractions were monitored
by the Western blot method using antiserum
against chum salmon PRL. Eel PRL
comprised two variants, which were
separated by HPLC on an ODS column. The
two PRLs were also secreted by organ-
cultured pituitaries kept in a defined
medium for several weeks. The secreted
PRLs were isolated by subjecting the
insoluble fraction of the dialysed medium
in 0.01 M ammonium acetate (pH 9.0) to HPLC
on an ODS column. Both PRLs had a
molecular weight of 22 kDa in SDS-PAGE, an
isoelectric point above 10 by gel
electrof ocusing, and N-terminal residue of
valine. Amino acid composition of the eel
PRLs was similar to that of salmon and
carp PRLs rather than eel GH. Comparison
of the fragments after digestion with
lysyl endopeptidase revealed two different
sites in amino acid sequence between the
two PRLs.

EN 3

PURIFICATION AND PROPERTIES OF NEWT PROLAC¬
TIN AND DEVELOPMENT OF HOMOLOGOUS
RADIOIMMUNOASSAY

K. Matsuda, K. Yamamoto and S. Kikuyama.
Dept, of Biol., Sch. of Educ., Waseda
Univ., Tokyo.

A highly purified prolactin( PRL ) was ob¬
tained from the acid acetone extract of
pituitary glands of the newt, Cynops pyr-
rhogaster , by HPLC on anion exchange, gel
filtration and reverse-phase columns.
Purification was monitored by SDS-PAGE and
Western blotting analysis employing an¬
tiserum to bullfrog PRL. The newt PRL thus
obtained has a molecular weight of 23,000
as determined by SDS-PAGE. The isoelectric
point is 4.7 when determined by isoelectric
focusing. The amino acid composition
resembles that of anuran PRLs suggesting
that newt PRL possesses three disulfide
bonds. Antiserum was prepared by immunizing
a rabbit with the newt PRL. The radioim¬
munoassay for newt PRL was developed
employing the antiserum, standard newt PRL
and ~223I-newt PRL. Several dilution of
plasma and anterior pituitary homogenate of
newts yielded dose-response curves which
were parallel to the standard curve. Bovine
and ovine PRLs as well as plasma from
hypophysectomized newts did not react in
this assay. The interassay coefficient of
variation and intraassay coefficient of
variation were 7.66% and 2.53%, respec¬
tively. The sensitivity of the assay was
0 . 03±0 . 004ng/ml .

EN 9

EXAMINATIONS OF MEASUREMENT OF CALCITONIN
BY ENZYME-LINKED IMMUNOSORBENT ASSAY (EL¬
ISA) IN LOWER VERTEBRATES

Y. Sasayamai N. Suzuki* C. Oguro* and A.
Kambegawa? *Dept. of Biol., Fac. of Sci.,
Toyama Univ., Toyama, 2Dept. of Obst. Gyne-
col., Fac. of Med., Teikyo Univ., Tokyo.

ELISA for calcitonin has been developed
only in frog (Rana pipiens ) throughout
vertebrates (Robertson, 1981 in Gen. Comp.
Endocrinol.) However, there is no paper
in which this method was adopted. There¬
fore, we tried to rebuild the assay system
and examined whether it is practical in
other frogs or also in other animals.
An antiserum was raised to synthetic salmon
calcitonin in rabbits. The standard curve
was calibrated by the synthetic salmon
calcitonin. Serial dilutions of extract
of ultimobranchial glands of frog (Rana
n igromaculata) exhibited parallel curves
to the standard. On the other hand, extract
of parathyroid glands of frog or bovine
PTH (1-34) and rat CGRP showed no cross-
reactivity. However, in ultimobranchial-
ectomized frogs (even 8 days after the
operation), calcitonin was detected as
large as in sham-operated frogs. This
fact implies two possibilities. The first
one is that there is a problem in the
specificity of the antiserum. The second
one is that ultimobranchial gland may
not be the only source of calcitonin in
frogs. Therefore, it was suggested that
this method may be useful if this dis¬
crepancy can be clarified.

EN 5

EFFECT OF TRH AND HYPOTHALAMIC FACTOR ON
PROLACTIN LEVELS IN PROMETAMORPHIC
BULLFROG TADPOLES

D. Uchida, S. Kikuyama.

Dept, of Biol., Sch. of Educ., Waseda
Univ., Tokyo.

We have isolated a substance which has a
prominent prolactin-releasing activity
from acid extract of bullfrog hypothalami
by Sephadex G-1 5 column chromatography and
reverse phase HPLC ( TSK ODS-120T, 0-50%
MeCN gradient in 0.1% TFA) . Prolactin¬
releasing activity was monitored by
measuring prolactin released from the
adult bullfrog pituitaries in the presence
of test substances iri vitro. The isolated
substance showed the same retention time
as synthetic TRH on the column. In order
to see whether the hypothalamic factor as
well as synthetic TRH elevates prolactin
levels in the tadpoles, each substance was
administered intraperitoneally to the
stage 18 tadpoles. One-two hours after the
injection both hypothalamic factor (1
hypothalamic equivalent) and TRH (0.1 /ug)
elevated the prolactin levels upto 200% of
the control value.

1298

Endocrinology

EN 6

IN VIVO EFFECT OF THYROTROPIN-RELEASING
HORMONE ON PLASMA TSH AND PROLACTIN
LEVELS IN BULLFROGS , RANA CATES BE I ANA
K. Niinuma1, S. Sakai1, S. Kikuyama1,

S . Tanaka2, H . Hayashi2 and Y . Hanaoka2.
’Dept of Biol., Sch. of Educ . , Waseda
Univ. , Tokyo. 2Inst. of Endocrinol.,
Gunma Univ., Maebashi .

Experiments were carried out to see
whether the release of TSH and prolactin
from the frog hypophysis is elicited
simultaneously by synthetic TRH . Such
experiments have not been done before
since specific radioimmunoassay system
for amphibians was lacking. Recently, we
have developed radioimmunoassay for
amphibian TSH using highly purified TSH-/3
subunit of bullfrog origin and its
antibody. TRH injections (5-20 Mg/ 100 g
B.W.) were made into the cannulated vena
abdominalis of adult bullfrogs. Blood
samples were collected upto 8 hr through
the cannula. Plasma prolactin and TSH
concentrations were determined by the
homologous radioimmunoassay. TRH elevated
prolactin levels 1 hr after injection.

The prolactin levels in the TRH-treated
animals declined to the control levels 8
hr after injection. On the other hand, no
significant elevation of plasma TSH by
the tripeptide was observed during the
experimental period.

EN 8

CLONING OF cDNA FOR THE CHICKEN LUTEINIZING
HORMONEyS- SUBUNIT

H. Ando, T. Noce, 2T. Higashinakagawa and
S. Ishii 'Dept, of Jiol., Sch. of Edu. ,
Wasada Univ., Tokyo, 2Dept. of Dev. Biol.,
Mitsubishi-Kasei Inst. of Life Sciences,
Tokyo.

An expression cDNA library was
constructed from poly(A) RNA of chicken
adenohypophyses with Agtl 1 . After the
screening with an anti-chicken LH
antiserum, a cDNA clone ( Li 2 ) encoding the
most part of the chicken LH subunit was
obtained. Another cDNA clone (LF127) was
isolated by rescreening with a subclone of
LI 2 ( pLl 2 ) as a probe. The LF1 27 was the

full length cDNA clone encoding the chicken
LH precursor. The leader sequence of the
subunit was longer (117bp) than that of
mammals. The Li 2 and LF1 27 differed in the
length of the 3 ' -untranslated region. The
deduced amino acid sequence of chicken LH/?
apoprotein revealed that the position of
all the twelve Cys residues were conserved
throughout gonadotropins of vertebrate
species studied, and the number of proline
residues was larger in phy logenet ical ly
higher vertebrate classes. Hybridization of
the pLl 2 cDNA insert to a RNA fraction was
detected only in the adenohypophysis among
various organs of chicken.

EN 7

CLONING OF BULLFROG PROLACTIN PRECURSOR
cDNA

K . Yoshihama 1 , S . K i kuayama 1 , K. Yamamoto1,
K . Wakabayash i 2 , and Y.Kato2 ‘Dept . of
Biol., Sch. of Educ., Waseda Univ., Tokyo,
2Inst. of Endocrinol., Gunma Univ.,
Maebashi .

In amphibians, prolactin ( PRL ) is known
to be involved in growth, metamorphosis,
osmoregu Lation and reproduction. We report
the molecular cloning of bullfrog PRL
precursor cDNAs as a basis for understand¬
ing of control mechanism of PRL gene ex¬
pression during metamorphosis and develop¬
ment .

From 100 bullfrog anterior pituitaries,
total RNAs were obtained by guanidine
i soth iocyanate-CsCl method. A cDNA library
was constructed using a cDNA synthesis
kit, the expression phage vector lambda
ZAP and a iji vitro packaging kit. A PRL
cDNA clone was detected from the library
by antiserum to bullfrog PRL. This clone
contained 235 bp insert. The result of the
chainterminat ion sequencing revealed that
it codes the carboxyl amino terminal
region corresponding to the known mam¬
malian PRLs . The longer insert clones are
under investigation. Northern analysis of
bullfrog anterior pituitary RNAs showed
that the bullfrog PRL mRNA is composed of
about 1300 nucleotides.

EN 9

BINDINGS OF BULLFROG FSH AND LH TO THE
LIVER OF BULLFROG TADPOLES.

K.Kubokawa and S. Ishii, Dept, of Biol.,
Sch. of Edu. , Waseda Univ. , Tokyo.

FSH and LH of Rana catesbeiana bound
specifically not only to the gonad but also
to the liver in amphibians. They also in¬
crease the cAMP concentration in bullfrog
liver slices. These phenomena were ob¬
served in sexually mature bullfrogs. When
do the hepatic gonadotropin receptors ap¬
pear in the course of the metamorphosis?
The specific bindings of LH to the liver of
tadpoles were very low and less than 40 %
of nonspecific binding from stage 5 to 20
at which the frontlimb erupt. At stage 21 ,
the LH bindings increased significantly,
and this level was maintained to the end of
metamorphosis. The specific bindings of
FSH were also very low from stage 5 to 18.
It started to increase at stage 19, and
continued to increase gradually through the
period from stage 20 to 25, when the tail
length shortens. The FSH bindings reached
to twice of nonspecific binding at the end
of metamorphosis. Both FSH and LH did not
change the cAMP concentrations in liver
slices at stage 18 to 20 when their sig¬
nificant specific bindings were not
detected. They increased the concentration
at stage 24 and 25, when high levels of
specific bindings were observed. These
findings have revealed that the hepatic
gonadotropin receptors appear after
frontlimb eruption and may regulate some
physiological function of the liver.

Endocrinology

1299

EN 10

SOLUBLE GONADOTROPIN RECEPTORS OF THE
POSTOVULATORY OVARY OF AMAGO SALMON.

A. Kanamori and Y. Nagahama. Lab of Reprod.
Biol., Natl. Inst, for Basic Biol., Okazaki.

Gonadotropin (GTH) receptors were efficiently
solubilized by sonicating particulate membrane
preparations of amago salmon postovulatory ovaries
for 10 min at 4 °C. Preparations were suspended in
Tris-buffer containing 20% glycerol, 1% Triton-X 100
and protease inhibitors. General properties of the
soluble GTH receptors were examined by binding
of 125l-labeled chum salmon GTH (CSG-SII).
Optimum specific binding occurred at pH 7.5 after
24 hr of incubation at 4 °C in the presence of 10
mM Ca2+. Over the range of 0-1 mg, there was a
linear increase in specific binding as protein in
soluble extracts increased. As the concentration of
GTH (labeled or unlabeled) increased, specific binding
of GTH approached saturation. Scatchard analysis
of GTH binding to the soluble receptors showed no
significant differences between the properties of the
soluble receptors and those of the particulate
membrane preparations (Kd= 1 nM, Bmax= 0.3x10" 13
mole/mg protein). Also binding specificity of the
soluble receptors seemed almost unchanged from that
of particulate membrane preparations as revealed by
competition experiments. These results show that
GTH receptors of the amago salmon postovulatory
ovary were efficiently solubilized as a functional
form and provide an approach to future studies on
the structure and properties of GTH receptors.

EN 11

GROWTH HORMONE RECEPTORS IN EEL LIVER.

M. Kishida and T. Hirano.

Ocean Res. Inst., Univ. of Tokyo, Tokyo.

Highly purified eel growth hormone (GH)
binds specifically and significantly to a
crude plasma membrane fraction of eel
liver. The specific binding to the lOOOOg
pellet derived from 30 mg fresh tissue was
10-15% in 20 h at 15 C, with nonspecific
binding less than 5%. The specific binding
to the 600g pellet was about 4%. The other
tissues such as gills, kidney, muscle,
brain and intestine of the eel showed
little or no specific binding. Specific
binding of 1251-eel GH to plasma membranes
ia a saturable process, and time-, pH-, and
temperature-dependent, with the optimum pH
occurring at pH 7.3, and with greater
specific binding occurring at 15 and 25 C
than at 35 C. Competition experiments
revealed that natural and recombinant eel
GHs as well as recombinant chum salmon GH
competed equally with 1251-eel GH . No
significant binding was detected with eel
prolactin ( PRL ) . Cvine GH competed with
1251-eel GH more effectively than eel GH.
Ovine and salmon PRLs were needed in
greater amounts (40 and 500 times,
respectively) than eel GH to displace the
radiolabeled hormone. Starvation seems to
cause a decrease in the specific binding in
the liver membrane, whereas transfer of the
eel from fresh water to seawater did not
affect the specific binding.

EN 12

A COMPARATIVE STUDY OF THYROID STIMULATING
HORMONE RECEPTORS IN MAMMALS AND AVES

M. Sato, H. Sakai, S. Wakabayashi and T.
Yoshida* . Dept. of Biol., Nihon Univ.
School of Dentistry, Tokyo, and *Tsukuba
Primate Center for Medical Science,

N. I .H. , Tsukuba.

Thyroid stimulating hormone (TSH)
receptors were studied to ascertain the
difference between mammals and aves. The
homogenates of thyroid glands in Macaca
fascicularis , Mus musculus and Gal lus
domesticus were used as the receptor
preparations and 125I-bovine TSH was used
as a labeled hormone. The specific binding
of TSH and receptor could be detected with

0.06 mg/ tube of receptor preparations.
The maximum amount of specific binding was
observed in 15-min incubation at pH 5.0 in
mammals and aves. Also observed was the
greater increase in the specific binding
of receptor preparations of mammals than
aves when pretreated with 50 mM of NaHCC>3 .
In mammals with pretreated receptors and
excessive TSH, the specific binding
decreased more than untreated ones, but,
in aves, excessive TSH, regardless of
treatment, did not decrease.

We therefore concluded that respective
TSH receptors of mammals and aves are
identifiable through the application of
bovine TSH.

EN 13

A QUANTITATIVE I MMUNOH ISTOCHEMI CAL STUDY ON ONTOG¬
ENY OF THE VASOPRESSINERGIC NEUROSECRETORY SYSTEM
IN THE PERINATAL MOUSE.

C. Yamada, T. Takezawa, S. Hyodo1 and A. Urano.
Dept, of Regul. Biol., Saitama Univ., Urawa and
^Ocean Res. Inst., Univ. of Tokyo.

Vasopressin (AVP) is a major mammalian neuro¬
hypophysial hormone involved in regulation of
water and mineral metabolism. Since mammalian
neonates are exposed to drastic environmental
osmotic changes, AVP is considered to play impor¬
tant osmoregulatory roles in the perinatal period,
although it is not yet clear what function AVP
has. We thus studied morphological and stereomet¬
ric changes in AVP immunoreactive neurons in the
brains of perinatal mice, because such information
would help better understanding of their physio¬
logical roles. AVP immunoreactivity was first
found in the median eminence on day 15 of embryon¬
ic age (E15), and then in the pars nervosa on E16.
Both of them are the terminal region of neurose¬
cretory fibers. AVP immunoreactivity in neuronal
cell bodies appeared on E16 in the supraoptic
nucleus (SON), and then in the paraventricular
nucleus (PVN) on E17. Thereafter, the number of
immnostained neurons gradually increased by day 0
of postnatal age (PN0), dropped somewhat on PNl to
PN2, and again increased to the adult level. The
sizes of AVP neurons also increased gradually in
the embryonic period. However, in concordance
with the change in the cell number, the magnitudes
of cross sectional cell areas of AVP neurons were
significantly decreased on PNl and PN2 . These
results suggest that secretory activity of AVP
neurons is elevated in neonatal mice.

1300

Endocrinology

EN 14

ONTOGENY OF VASOPRESSIN GENE EXPRESSION IN
THE PERINATAL MOUSE HYPOTHALAMI
S. Hyodo', C. Yamada2, T. Takezawa2 and A.
Urano2. 'Ocean Res. Inst., Univ. of Tokyo,
Tokyo and 2 Dept, of Regulat. Biol., Fac.
of Sci., Univ. of Saitama, Urawa.

Ontogeny of vasopressin (AVP) gene
expression in magnocellular neurons of the
perinatal mouse hypothalami were
determined by in situ hybridization and
immunohistochemical methods. The AVP mRNA
was first detected on day 16 of embryonic
age (El 6) in the supraoptic nucleus (SON)
and on El 9 in the paraventricular nucleus
(PVN). AVP mRNA levels were conspicuously
increased on day 0 of postnatal age (PNO)
and PN1 in the SON, on PNl and PN2 in the
PVN. AVP mRNA levels were once decreased
on PN2 in the SON and on PN5 in the PVN,
and were increased again to the adult
level after PNl 0 in the SON, and PN20 in
the PVN.. Immunostainability of AVP in
both nuclei was markedly decreased on PNl
and PN2. Sizes of AVP immunoreactive
neurons and their nuclei were increased on
El 9 and PNO in both the SON and the PVN.
Immunostainability in the neurohypophysis
did not noticeably change in the perinatal
period. These results suggest that
synthesis, transport and release of AVP
were stimulated in the perinatal period in
response to the drastic changes in water
metabolism after birth.

EN 15

REGIONAL DIFFERENCES OF THE EFFERENT PRO¬
JECTION AND THE DISTRIBUTIONS OF VASO¬
PRESSIN NEURONS OF THE SUPRACHIASMATIC
NUCLEUS IN RATS.

K. Yamase, K. Nomura, S. Takahashi and
S. Kawashima

Zoological Institute, Faculty of Science,
Hiroshima University, Hiroshima. _

The suprachiasmatic nucleus (SCN) is
related to the regulation of discharge of
gonadotropin-releasing hormone via the
medial preoptic area, the ventromedial
and arcuate nuclei and the median
eminence. Recent studies have shown a
regional differentiation in this nucleus.
Based on the finding that the lesion of
dorsomedial SCN(dm-SCN) but not ventro¬
lateral SCN ( vl-SCN ) eliminated the cir¬
cadian rhythms in feeding and drinking,
it was concluded that the dm-SCN was a
crucial site for the maintenance of cir¬
cadian rhythmicity. The present studies
were aimed to clarify the efferent pro¬
jections of the SCN by means of retro¬
grade transport of horseradish peroxidase
( HRP ) and the distribution of AVP-immuno-
reactive cells of the SCN. The distribu¬
tion of HRP-labeled somata in the SCN was
preferentially observed in the dm-SCN.
Moreover, the number of HRP labeled
somata with AVP-immunoreactivity in the
dm-SCN was significantly greater than the
other divisions. The present results show
that the SCN has a regional difference
within the nucleus, and the structural
differences may be related to several
diverse functions of the SCN.

EN 16

EFFECTS OF SUPRACHIASMATIC DEAFFERENTATION
ON LHRH CONTENT AND CATECHOLAMINE TURNOVER
IN HYPOTHALAMIC NUCLEI IN FEMALE RATS.

K. Nomura, S. Takahashi and S. Kawashima
Zool. Inst., Fac. of Sci., Hiroshima Univ.,
Hiroshima . _

Estradiol benzoate ( EB ) -induced LH re¬
lease was blocked by complete suprachiasmat¬
ic deaf f erentation ( CSD ) . In this study,
effects of CSD on EB-induced changes in
hypothalamic LHRH content and catecholamine
turnover were examined. CSD and control
rats were ovariectomized and used for ex¬
periments. Two days after EB administration,
rats were decapitated at 1000, 1300 and
1600h,and LHRH contents in the medial basal
hypothalamus were assayed. In control rats,
LHRH content increased before LH surge (1300
h) and decreased during the surge ( 1 600h ) .
In CSD rats LHRH content increased at 1 300h
and remained the same at 1600h. In the next
study, rats were injected with oc-methyl-p-
tyrosine at 1000 or 1530h, and killed 90min
later. Catecholamine contents in punched -
out tissues of some hypothalamic nuclei
were measured with HPLC, and turnover rate
of norepinephrine(NE) was calculated. EB-
induced elevation of NE turnover rates was
observed in the afternoon, during LH surge,
in the medial preoptic area, medial preop¬
tic nucleus, suprachiasmatic nucleus, arcu¬
ate nucleus and median eminence. Enhanced
NE turnover rates were also observed in CSD
rats, but this was not accompanied by LH
surge, indicating that the suprachiasmatic
nucleus is more closely related to EB-in-
duced LHRH release than NE neurons.

EN 17

INCREASE OF LHRH \mMUNOREACTIVE FIBERS IN
MEDIAL AMYGDALOID \NUCLEUS AFTER REMOVAL OF
ACCESSORY OLFACTORY BULB.

M. Ichikawa^ and Y. Oka^. ^Dept. Anat &
Embryo., Tokyo Metro. Inst. for Neurosci.,
Tokyo, and ^Zool. Inst., Fac. of Sci.,
Univ, of Tokyo .Tokyo. _ _

We have already: reported that the afferent
terminals in the medial amygdaloid nucleus
(MAN) are rearranged after the denervation
of fibers from accessory olfactory bulb
( AOB ) . On the other hand, it has been
reported that the LHRH immunoreactive
( LHRH-IR ) fibers originating from the basal
forebrain and/or medial preoptic area
distribute in MAN. In the present study,
we examined, using the immunohistochemical
technique, if the LHRH-IR fibers in the MAN
undergo the plastic changes after AOB
removal. The rat AOB was removed
unilaterally and the rats were perfused at
4 days or 2 months survival time. The
tissue including the MAN was processed by
the immunohistochemical method using anti-
L HR H monoclonal antibody. The LHRH-IR
fibers were clearly recognized in MAN.
The total length of LHRH-IR fibers on the
side of AOB removal in MAN was compared
with that on the intact side. Total length
of LHRH-IR fibers inceased significantly on
the side of AOB removal at 4 days survival
time. Similar tendency was also recognized
at 2 months survival time. These results
indicate that the rearrangement of LHRH-IR
fibers take place in MAN after the
dennervation of AOB fibers.

Endocrinology

1301

EN 18

AGE-RELATED CHANGES IN GH mRNA CONTENTS
IN THE RAT.

S. Takahash i^, S. Kawa sh i mal , H. Seo2. and
N.Matsui2.. ^Zool. Inst., Fac. Sci.,
Hiroshima Univ. , Hiroshima and 2pes> Inst.
Environ. Med. , Nagoya Univ. , Nagoya.

Growth hormone (GH) secretion declines
and prolactin (PRL) secretion increases
with aging in rats. We measured pituitary
GH mRNA and PRL mRNA contents using cDNA
probes of each hormone in the rat of
Wistar/Tw strain at G, 12 and 18 months
of ages. Estrous females (6 months of
age) and persistent estrous females (12
and 18 months of age) were used. The mRNA
levels were measured with the method of
cytoplasmic dot hybridization. The GH
mRNA content in pituitary glands did not
change in both sexes with aging, although
pituitary GH content decreased at 18
months of age. The GH mRNA level per pg
GH cell DNA decreased at 18 month of age
in both sexes. The PRL mRNA content
decreased in male rats at 18 months of
age, but increased in female rats at 12
and 18 months of age. The pituitary PRL
content decreased in male rats, and
increased in female rats with aging. The
PRL mRNA level per ^ig PRL cell DNA
decreased at 18 months of age in both
sexes. The increase in total PRL mRNA in
12_ and 18— month old females was due to
the increase in the number of PRL cells.
These results suggest that the syntheses
of GH mRNA and PRL mRNA might decrease
in each GH and PRL cell.

EN 19

EFFECTS OF TESTOSTERONE ON THE HISTOPATHO-
LOGICAL CHANGES OF THE KIDNEY OF MALE AND
FEMALE WISTAR/TW RATS.

Win Win Yee, S. Takahashi and S. Kawashima.
Zoological Institute, Faculty of Science,
Hiroshima University, Hiroshima.

Kobayashi and Kawashima (1982) and our
earlier reports have indicated sex dif¬
ference in physiological, histopathological
and ultrastructural changes of the kidney
of Wistar/Tw rats during aging. To further
elucidate the basis for the sexual dif¬
ference testosterone propionate (TP) re¬
placement studies were undertaken in cas¬
trated animals. Rats were castrated at 1 5
days of age and immediately implanted with
silastic tube filled with TP in both sexes.
At 1 , 3 and 6 months of age kidneys were
examined with light and electron micro¬
scopes after water intake (WI) and urinary
volume (UV) measurements. All the treated
animals at 3 months of age showed greater
WI , UV and kidney weight than the controls.
Histopathological changes of the control
rats kidneys were only weak at 6 months of
age while treated rats showed initial le¬
sions at 3 months. All the treated 6-month-
old male rats showed profound lesions than
similarly treated female rats. After 3
months ultrastructural changes of the glo¬
merulus in treated included the presence
of huge vacuole in the epithelial cells,
epithelial attachment to the glomerular
basement membrane and enlargement of the
mesangium .

EN 20

EFFECTS OF AGING ON URINARY AND SERUM
PARAMETERS FOR RENAL FUNCTION IN THE
WISTAR/TW MALE RATS: II. OVER 13-MONTHS-OLD .
Y. Kobayashi, M. Senoo* and K. Hattori.
Dept, of Pharmacol, and *Cent. Res. Lab.,
Shimane Med. Univ., Izumo.

The Wistar/Tw strain male rats show
polyuria after 16 months of age (Mo) .
Morphological and functional changes in re¬
nal glomeruli in young period have been re¬
ported, however, the mechanism of enhance¬
ment of urinary volume at aged rats has
been unclear. In the present study, uri¬
nary and serum parameters for renal func¬
tion were measured in rats at 13, 15 and 17
Mo and they were compared with those of
younger rats. Urinary protein concentra¬
tions were increased at 13 Mo and showed a
peak at 15 Mo. Daily excretion of urinary
protein increased with aging and showed the
maximum at 17 Mo. Urinary creatinine
concentrations were decreased at 17 Mo.
Serum creatinine concentrations and blood
urea nitrogen values were higher at 17 Mo,
glomerular filtration rate was much lower,
suggesting the further dysfunction of
glomeruli at 17 Mo. On the other hand,
urine concentrating rate, calculated from
urine and serum creatinine concentrations,
reduced with aging. In conclusion,

dysfunction of renal glomeruli occurred in
younger period and it developed during
aging, however, dysfunction of tubule
reabsorption was occurred at aged rats and
it masked the reduction of glomerular
filtration rate and resulted in polyuria.

EN 21

SUBCELLULAR LOCALIZATION OF ANTERIOR
PITUITARY HORMONES ON ULTRATHIN
CRYOSECTIONS

S. TANAKA AND K.KUROSUMI. Dept, of Morphol.,
Inst, of Endocrinol., Gunma Univ. Maebashi ■

There are three techniques for
immunoelectron microscopy; the pre¬
embedding, post-embedding and nonembedding
methods. Among them, the nonembedding
method using cryomicrotomy is the most
advantageous for study by immunoelectron
microscopy, because there is little loss of
antigenicity during dehydration and
polymerization of the embedding media, and
the access of antigens and antibodies may
easily occur in the unembedded specimen.

When ultrathin cryosections of rat
anterior pituitary were immunostained with
anti-rat PRL labeled with gold particles,
we observed on round and polymorphic
secretory granules in PRL cells. Many gold
particles were also localized in the Golgi
saccules and in the cisternae of rough ER.
The gonadotrophs of adult male bullfrog
were characterized by two types of
secretory granules; round or irregular
secretory granules, and large globules.
When ultrathin cryosections were dobule-
immunolabeled with mouse monoclonal
antibody to bullfrog LH-beta (5nm gold
particles) and rabbit anti-bullfrog FSH-
beta serum (15 nm) , subcellular localiza¬
tion of different sizes of immunogold
particles revealed that individual
secretory granules in gonadotrophs contain
only one or both gonadotropins.

1302

Endocrinology

EN 22

I MMUNO CYTOCHEMISTRY ON MELANOTROPHS OF PARS
INTERMEDIA IN AMPHIBIAN PITUITARIES.

S.Kurabuchi1 and S. Tanaka2. 1 Depart. of
Histol . , Sch.of Dentist., The Nippon Dental
Univ. , Tokyo and 2Dept. of Morphol., Inst,
of Endocrynol., Gunma Univ., Maebashi.

Melanotrophs in pars intermedia of 1 1
amphibian species were immunostained by PAP
method using antisera to a-MSH, porcine
ACTH , human 3-LPH and camel 3-Endorphin.
Almost MSH cells showed a positive reaction
to anti-a-MSH, 3-LPH and 3-End. sera in all
species examined. However, the immunoreac-
tivity of ACTH showed a species-specf icity ;
positive reaction in Hinobius retardatus ,
H . nebulosus tokyoensis , Ran a j aponica , R .
brevipoda porosa, R. catesbeiana and
Xenopus laevis’, negative in Onychoductylus
japonicus , Ambystoma mexicanum and Cynopus
pyrrhogaster pyrrhogaster , Hyla j aponica
and Bufo bufo formosus.

When the white-background- adapted frogs
R. b. porosa) were adapted in black one,
the skin colour changed to be dark within
one hr. In the MSH cells after two hrs.,
an immunoreactive ACTH became remarkably
faint, and was not observed at all after 6
hrs. On the other hand, the immunoreacti-
vity of a-MSH, 3-LPH and 3-End. was also
weak even in the frogs adapted for a long
period in black background. This remarkable
decrease and disappearance of immunoreac¬
tive ACTH suggests that the synthesis and
release of a-MSH increased as the result of
proteolitic cleavage of ACTH into a-MSH.

EN 23

RELATIONSHIPS OF THE HYPOPHYSIAL PORTAL
SYSTEM AND DISTRIBUTION OF THE ADENO¬
HYPOPHYSIAL CELLS IN THE SNAKES
I.Koshimizu and Y.Oota. Biol. Inst.,
Fac . of Sci., Shizuoka Univ., Shizuoka.

The vascularization of the hypophy¬
sis and distribution of the adenohypo¬
physial cells were studied in the
snakes, Elaphe quadrivirgata, Elaphe
climacophora, Elaphe conspicillate and
Rhabdophis tigrinus . Blood vascular
casts were observed by scanning elec¬
tron microscope. The primary capillary
plexus (PCP) of median eminence re¬
ceives an arterial supply from the
infundibular artery. The PCP converges
into several large portal vessels that
supply the blood to the pars distalis
(PD). In the PD, the portal vessels
give off the secondary capillary
plexus. The regional distribution of
the portal vassels into the PD is
demonstrated . Immunocytochemically ,

corticotropin-immunoreactive cells are
found in the cephalic lobe, whereas
thyrotropin-immuno reactive cells are
located in the ventro-caudal lobe. The
gonadotropin-immuno reactive cells are
distributed throughout the PD. The
results suggest the close correlation
between the vascular flow and the
regional distribution of the cells in
the PD.

EN 24

IMMUNO-ELECTRON MICROSCOPY OF TWO DISTINCT
GONADOTROPHS IN TELEOST PITUITARIES.
Naito,N.l, Nakai,YjL, Nagahama,Y?, Suzuki, K?
and Kawauchi,H^ ^Showa Univ. Sch. Med . Tokyo ,
?Nat . Inst . for Basic Biol . Okazaki , 3Kitasato
Univ. Iwate.

Two chemically distinct gonadotropins,
designated GTH I and GTH II, were purified
from chum salmon pituitaries. GTH II
corresponds to the classical GTH and GTH I
is a new hormone. In our light microscopic
immunocytochemistry , specific antisera
against 3-subunits of chum salmon GTH I and
GTH II cross-reacted with different cells
in the amago salmon pituitary. Here, we
examined these two distinct gonadotrophs,
GTH I and GTH II cells, with immuno-
electron microscopy using the same
antisera. In ovulated pituitaries, the
gold-labeled anti-GTH II were concentrated
both on small granules (200-400 nm ) and
large globules (500-2000 nm ) of GTH II
cells. In the GTH I stained cells, large
globules were scarce and gold-particles
were located on small granules (100-400
nm ) . In vitellogenic pituitaries, as
compared with ovulated ones, GTH I and
GTH II cells were much smaller in number,
and immuno-positive granules in both cells
were smaller in size (1 00-300 nm). These
results suggested that ultrastructure of
GTH I and GTH II cells were quite similar
to each other, and thus it was difficult to
identify them exactly without immuno¬
cytochemistry at any reproductive stages.

EN 25

EFFECT OF TRH AND DOPAMINE (DA) ON CULTURED
PRL CELLS FROM YOUNG AND OLD RATS .

T. Shinkai1 , M. M. Takahashi2 and H. Ooka1 .

' Dep t . of Biol., and 2Cell Cult. Lab.,
Tokyo Metropol. Inst, of Gerontol., Tokyo.

Anterior pituitary cells of young(6mo)
and old(22mo) female rats were cultured
with TRH(10"7M) or DA(1CT4M) for 3.3, 6.7
and lOh. PRL levels in the medium were
measured by RIA, and the ultrastructure of
young PRL cells (YPC) and old PRL cells (OPC)
was observed. PRL secretion by YPC did not
differ from OPC in the absence of TRH or
DA. Most YPC had small granules ( SG ) . Large
granules(LG) were contained in 3/4 of OPC.
The elevation of PRL levels by TRH was
observed only in the first 3.3h. YPC
cultured with TRH contained SG, but some LG
cells were observed at lOh. Three types of
the cells which contained SG, median
granules (MG) and LG were observed in OPC.
PRL secretion by both of YPC and OPC was
suppressed by DA in the first 3.3h, but
that of OPC gradually recovered in spite of
the presence of DA. Most of YPS in this
conditions were SG type, whereas OPC
contained of SG(1/3), MG(1/2) and LG(1/6).
These results indicate that functional and
morphological responses of PRL cells to TRH
and DA are affected by the age of the
cells .

Endocrinology

1303

EN 26

AGE-RELATED CHANGES OF LACTOTROPIC CELLS
OF FEMALE RATS IN THEIR RESPONSIVENESS TO
POSTERIOR PITUITARY EXTRACT
M.M.Takahashi , T.Shinkai2 and H.Ooka ,
'Cell Cult.Lab. , Dept.of Biol., Tokyo
Metropol.Inst.of Geronto 1 Tokyo

To study the possible mechanisms
involved in hyperprolactinemia of aged
rats, responsiveness of lactotropic cells
obtained from young and aged rats to VIP
and posterior pituitary extract was
examined in serum-free medium. In control
cultures, PRL secretion by lactotropic
cells of aged rats was comparable level to
that of young rats. Addition of VIP at
concentrations in the range from 10-13 to
10_'M had no significant effect on PRL
secretion of the cells from both young and
aged rats. However, posterior pituitary
extract(PPE) of rats significantly
stimulated the cells from either young or
aged rats to secrete PRL. Responsiveness
to PPE of the cells from aged rats was far
greater than that of the cells from young
rats. The present results suggest that an
elevated PRL level of aged female rats may
be due to at least in part an increased
response of lactotropic cells in aged rats
to some factors in posterior pituitary.

EN 27

FUNCTIONAL HETEROGENEITY OF PRL AND LH
CELLS IN THE RAT ANTERIOR PITUITARY GLAND.
Y. Sakai1. A. Hattori2, T. Yashiro2, and
T. Suzuki2

1 Coll, of Agric. arud Vent. Med., Nihon
Univ., Fujisawa and 2 Dept, of Anat., St.
Marianna Univ. Sch. of Med., Kawasaki.

The functional heterogeneity of hormone
producing cells of the anterior pituitary
gland was studied especially focusing on
PRL cells, using reverse hemolytic plaque
assay (Neill et al. 1983). In this study,
we tried to obtain further information
concerning functional heterogeneity of LH
cells in the rat.

Dispersed cells of the gland from 20-day
and 60-day-old Wistar-Imamichi male rats
were incubated in Cunningham chambers with
protein-A coated ovine erythrocytes in the
presence of oLHbeta antiserum. The LH
cells showed complement-mediated plaque
formations. The number and size of
plaques increased significantly with the
prolongation of incubation time for up to
3hr with the addition of LHRH-A, (D-
Ala6 ,des-Gly 1 0 ) -LHRH ethylamide. Increment
of LHRH-A concentrations from 1 0_1 1 to
10-9M increased the plaque number and to¬
tal plaque area although the mean plaque
area showed no remarkable change. These
results suggest the existence of
heterogeneity as to hormone release among
the LH cells.

EN 28

AN IN VITRO STUDY OF CELL TO CELL AFFINITY
IN CULTURED FISH ANTERIOR PITUITARY CELLS.
A. Hattori1, T. Yashiro1, T. Suzukir, J.
Takimori , M. Kobayashi3 and K. Aida3
Dept, of Anat., St. Marianna Univ. Sch.
of Med., Kawasaki, 2Coll. of Agric. and
Vent. Med., Nihon Univ., Fujisawa and
3Pept. of Fish., Univ. of Tokyo, Tokyo.

In order to examine the cell to cell af¬

finity of each hormone producing cell in
teleost pituitary gland, we employed a
suspension culture rotated by a magnetic
stirrer. The glands from immature rainbow
trout and mature goldfish were dispersed
with 0.2% collagenase for 30 min, and the
cells were then cultured in suspension by
rotation (80rpm) under 99% air 1% C02 at
15 or 25 C. After the culture for 3, 6,
12, 24, 48, and 72hr, the cell aggregates
were fixed by sublimate-formalin, and im-
munostained using several antibodies gen¬
erated against each hormone. Within 24hrs
in culture, the aggregated pituitary cells
became organized in a tissue-like con¬
figuration. In trout, most PRL cells were
found within well-defined clusters, but
other hormone producing cells appeared
scattered. PRL cells in goldfish also
gathered with well-defined cell colonies,
and GTH cells sometimes formed two or
three cell colonies. These results indi¬
cate that isolated PRL cells allowed
reaggregation in vitro in homologous cell
colonies which maintained in vivo charac¬
teristics .

EN 29

THE APPEARANCE OF STRIATED MUSCLE FIBERS IN
MONOLAYER CULTURES OF ANTERIOR PITUITARY CELLS.
Y. G. Watanabe, Biol. Inst., Fac. of Sci., Univ.
of Niigata, Niigata

Striated muscle fibers were found to appear in
monolayer cultures of rat anterior pituitary
cells maintained in a MEM. Since this medium has
not commonly used, a quantitative study was
performed to test the effects of culture media
on the differentiation of muscle in vitro. Three
different media, Medium 199, DMEM and a MEM were
used. Of these media, a MEM was most effective in
inducing muscle differentiation in vitro. Both
the concentrations (1-10 $) and nature of serum
(fetal calf, calf and horse) were not essential
for the myogenesis.

In a MEM-maintained cultures, striated muscles
differentiated as follows : 1) the appearance of
elongated cells resembling myoblasts, 2) their
fusion, and finally , 3) the formation of the

cross striations. Rhythmical contraction was
observed before the appearance of cross striations
or even before the fusion of myogenic cells.

Cultivation of leptomeninges has also resulted
in the appearance of muscles, though their inci¬
dence was far lower than in pituitary monolayers.
The possible origin of striated muscles in these
cultures is discussed.

1304

Endocrinology

EN 30

QUANTITATIVE ELECTRON MICROSCOPIC STUDY ON
THE PARS INTERMEDIA OF THE PITUITARY GLAND
IN MICE UNDER EXPERIMENTAL POLYDIPSIA:
EFFECTS OF NACL SUPPLEMENTATION.

M.Okada and Y.Kobayashi, Dept, of Biol.,
Fac.Sci., Okayama Univ. , Okayama.

Male mice of the ICR/JC1 strain showed
polydipsia and polyurea when animals were
deprived of lab chow and, instead, given
5% glucose solution for drinking. This
copious drinking produced marked hyper¬
secretion in cells of the pars inter-
media(PI), which was interpreted as a
result of hyponatremia induced by poly¬
dipsia and polyurea. In the present study,
morphometric study on PI cells of the pitu¬
itary was undertaken in mice after inges¬
tion of (.1)5% glucose, (2)5% glucose added
with final 0.225% NaCl, (3)5% glucose with
0.45% NaCl and (4)5% glucose with 0.9%
NaCl, for 3 days. Mice fed a lab chow and
given DW served as the control. Drinking
volume relative to body weight (v/w) was
(1)200%, (2)230%, (3)220% and (4)200%,
respectively, whereas 20% in the DW con¬
trol. The % area of r-ER in PI cells
increased from 9.5% in the control to
(1)27.6%, (2)27.6%, (3)26.6% and (4)25.2%.
On the contrary, the numerical density of
secretory granules decreased from 2.2 in
the control to (1)1.7, (2)1.7, (3)1-4 and
(4)1.7, respectively. These results indi¬
cate that the mouse PI cells show hyper¬
secretion in response to copious drinking
despite simultanous intake of excess NaCl.

EN 31

ULTRASTRUCTURAL STUDIES ON THE EMBRY¬
ONIC DEVELOPMENT OF THE PARS TUBERALIS
IN THE SALAMANDER, HYNOBIUS NEBULOSUS
NEBULOSUS

Y.Oota, T.Saga and I.Koshimizu, Biol.
Inst., Fac. of Sci., Shizuoka Univ.,
Shizuoka.

The embryonic development of the
pars tuberalis (PT) in the salamander
was studied. The salamander PT forms
two distinct rostral, bilateral and
disk-shaped plaques. In the premeta-
morphic stages of the larvae, the
plaque elongates. They are composed
only a few free epithelial cell mass.
Considerable increase in number and
volume of PT cells is demonstrated
during prometamorphic and climax
stages. Most of the PT cells are chro¬
mophobic. Immunocytochemically , some
cells reacting with TSH- and FSH-anti-
sera are observed in the climax stage
of larvae. Electron microscopically,
most part of the PT is composed of the
non-granul ated cells. At the climax
stage, two types of glanulated cells
are identified. Type 1 cells are
characterized by the presence of a few
dense granules (ca. 150-200 nm in
diameter). Type 2 cells contain large
electron-dense granules (ca. 200-300
nm). They may correspond to the TSH-
and FSH-immuno reactive cells, respec-
t ively .

EN 32

METABOLISM OF ECDYSTEROID CONJUGATES IN
THE OVARIES OF BOMBYX MORI .

E.Ohnishi1 and S . Y . Takahashi2 . 1Dept. of
Biol. ,Facult. of Sci., Nagoya Univ. , Nagoya
andzDept. of Biol. ,Facult. of General Educ .
, Yamaguchi Univ., Yamaguchi.

Six major ecdysteroid conjugates, which
had been identified as phosphate esters,
accumulate in mature ovaries of Bombyx mori
. Relative contents of these conjugates
roughly parallel to those of the correspon¬
ding free forms, suggesting a rapid inter¬
conversion between the two forms. The
synthesis of the conjugates is considered
to be catalyzed by phosphotransferases,
and the metabolism by phosphatases. Two
kinds of acid phosphatases, having differ¬
ent pH optima, could be detected in the
extract of ovaries. The phosphatase having
higher pH optimum has been characterized.
The enzyme was found in the extract as
partly bound to vitellin, one of the major
proteins found in the ovary. Ecdysteroid
conjugates were hydrolyzed by this enzyme
only slowly: 2 , 22-dideoxy-20-hydroxyecdy-
sone 3-phosphate was split most efficient¬
ly, but bombycosterol 3-phosphate was not
hydrolyzed appreciably. Other conjugates
were hydrolyzed only slightly.

EN 33

DIFFERENCES OF MAJOR ECDYSTEROIDS PRODUCED
BY THE PROTHORACIC GLANDS AMONG THE INSECT
SPECIES

S.Kiriishi, S.Sakurai. Dept of Biol, Fac.
of Sci., Univ. of Kanazawa, Kanazawa.

The prothoracic glands of insects is
believed to stcrete ecdysone which is
converted to the major molting hormone, 20-
hydroxyecdysone . Recently, the prothoracic
glands of Manduca sexta was shown to
release a mixture of 2-dehydroecdysone and
3-dehydroecdysone (1:2), which is rapidly
converted to ecdysone in the presence of
hemolymph protein fraction.

We analyzed ecdysteroids produced by
prothoracic glands of lepidopteran,
dipteran and orthopteran insects by HPLC
and RIA using 2 kinds of antisera .which can
recognize the difference of side chain
modification and AB ring modification of
ecdysteroids. The results showed that
insects can be classified to following
three groups, (1) like as Manduca, the
major product is dehydroecdysone (Papilio
xu thus , Pier is rapae ), (2) the major

product is ecdysone (Bombyx mori,
Sarcophaga peregrina) , (3) nearly same

amount of ecdysone and dehydroecdysone was
produced (Leucania separata , Periplaneta
americana) .

Endocrinology

1305

EN 34

EN 36

DIFFERNT ACTION OF ECDYSONE AND
DEVELOPMENT OF IMAGINAL DISCS OF THE
FLESH FLY BETWEEN DIAPAUSE- AND NON¬
DIAPAUSE-DESTINED INDIVIDUALS.

1 2 3

A. Moriba^ashi , H . Kurahashi^ , K.Ogino and

T.Ohtaki . Dept . Technology , Dept . Entomology
N . I . H . , Tokyo , ^Taisho Pharmaceutical , Co . , and
Dept . Biology , Faculty of Science , Kanazawa
Univ. , Kanazawa.

Larvae of the flesh fly (Boettcherisca
peregrina) develop to adult continuously
when they grow under a long day condition
(15L:9D)at 20°C, while larvae reared under
a short day condition ( 9L: 15D) at 20°C enter
pupal diapause.

Diapause- and nondiapause-destined larvae
pupariate in 30hr and 40hr after the
transfer to dry conditions, respectively.

Pupation in diapause-destined individuals
takes place 50hr after pupariation,but that
in nondiapause ones does 60hr after.
Therefore, the larvae of both populations
pupated 90hr after the dry treatment.

Ecdysteroid contents in individuals of
both cultures gradually increased and made
a peak at about 15hr after the pupariation.

The second increase of the ecdysteroid
level was observed in nondiapause-destined
animals several hours before pupation, but
was not in diapause-destined ones. The
second increase of ecdysone contents is
indispensable for adult development, and
also may cause a delay of pupal development
of wing discs in nondiapausing pupae.

INCUBATION TEMPERATURE ON THE TERMINATION
OF THE DIAPAUSING CYNTHIA PUPA
A. Koenuma. Dept. of Biol.,Fac. of Sci.,
Univ. of Shinshu, Matsumoto.

The termination of diapause of the pupa
of Sami a cyn th i a orveri is induced by Act¬
ivation of the brai n-pro thoraci c gland
system. When pupae kept in 5°C-storage
were incubated with 26°C, they developed
to imagineg. When the incubation temperat¬
ure was 20°C,they did not develop. But
when the pupae were incubated with 20°C
for 15 days then with 26°C, they were
able toQdevelop.The failure of development
with 20 C-incubation was able to consider
as the failure of secretion of ecdysones
from the prothoracic glands, because the
pupa devel^d with ecdysone injection. Some
experiments were carried out to invest¬
igate whether or not the failure of deve¬
lopment with 20°C- i ncubat i on ascribe the
failure of the activation of prothoracic
glands. When brain was removed from the
pupa after the 15-day incubation with 20°C
then the pupa was incubated with 26°C, no
pupa developed. When the pupae with 20°C-
incubation were stimulated by electric
pul ses ( 1 V/mm , 1H1 z ) f or 3 days, all survived
pupae developed. These results show that
the failure of development with 20°C is
not ascribe the failure of activation of
the prothoracic glands. It seems to as¬
cribe the failure of PTTH secretion from
the pupal brain.

EN 35

DIFFERENCE IN SEPHADEX G-50 GEL-FILTRATION
PATTERNS OF SMPH- AND PTTH-ACTIVITIES IN
POLYGONIA C-AUREUM L.

K. Endo , Y. Fujimoyo , T. Masaki and K.
Kumagai-*- . Biol. Inst., Fac. of Sci., Yama-
guchi Univ., Yamaguchi and ^Biol. Inst.,
Fac. of Liberal Art, Yamaguchi Univ.,
Yamaguchi. _

In Polygonia c-aureum L. , a summer -
morph-producing hormone (SMPH) was a pep¬
tide of M.W. about 4,500, which was pro¬
duced in medial neurosecretory cells of
the brain and secreted into hemolymph in
the early pupal stage. In addition, 20-
hydroxy-ecdysone in pupae was thought to
play a significant role in the development
of seasonal morphs.

Brain-extracts were made from brains of
P . c-aureum and B^ mor i and were applied
to a column of Sephadex G-50. Activities
of SMPH and PTTH in each fraction were
assayed by Polygonia pupal assay for SMPH
and Polygonia in vitro assay for PTTHs ,
respectively .

From gel-filtration patterns of SMPH-
and PTTH-activities through Sephadex G-50,
PTTHs of two different molecular sizes
(small and big) may exist in P^ c-aureum.
Molecules of Polygonia SMPH were judged as
being much more smaller than those of the
small PTTH being smaller than those of the
other. Furthermore, one of Bombyx PTTHs
(bombixin) extracted from adult brains of
B. mori activated larval prothoracic
glands (PGs) in P_^ xuthus , but it showed
no effects on larval PGs in P. c-aureum.

EN 37

REVERSIBILITY OF ADULT DIAPAUSE IN THE
BEAN BUG, RIPTORTUS CLAVATUS
H. Numata , Dept, of Biol., Fac. of Sci.,
Osaka City Univ., Osaka.

Riptortus clavatus Thunberg (Insecta,
Heteroptera, Alydidae) shows a facultative
adult diapause, which is induced and
maintained by short-day photoperiod (SD),
and prevented or terminated by long-day
photoperiod (LD). The insects maintained
the photoperiodic sensitivity during
nymphal and adult stages. Diapausing
adults began to lay eggs responding to LD,
and non-diapausing and post-diapausing
adults entered diapause responding to SD.

Diapause in insects is not a rhythmic
event but occurs once in the life of an
individual in general. However, adult
diapause in R. clavatus was induced and
terminated reversibly several times
(max. 4 times) by photoperiodic changes.
Taking advantage of this reversibility, we
can study new aspects of insect diapause.

At the beginning, I examined the
individual variation in the period (the
number of photoperiod) required for
diapause induction and termination.
Diapausing adults were transferred from SD
to LD, then to SD, and finally to LD
again. The sooner an individual
terminated the first diapause after
transferred to LD, the later the
individual entered the second diapause
after returned to SD and terminated it the
sooner after re-transferred to LD.

1306

Endocrinology

EN 38

REGULATION OF DIAPAUSE-HORMONE SECRETION
IN THE SILKWORM, BOMBYX MORI.

K. Hasegawa, I. Shimzu, and T. Matsui.

Res. Sec. of Environ. Biol., Lab. for Plant
Ecol. Studies, Fac. of Sci . , Kyoto Univ. ,
Kyoto . _ _

We studied the neural regulation of dia¬
pause-hormone (DH) secretion from a subeso-
phageal ganglion (SEG) in the silkworm. In¬
jection of y-aminobutylic acid (GABA) , an
inhibitory neurotransmitter , into female
pupae of diapause-egg producers induced
non-diapause eggs. On the other hand, in¬
jection of picrotoxin, a specific potent
blocker to GABAergic neuron, into pupae of
non-diapause-egg producers caused diapause
eggs. Then to see the effect of GABA and
picrotoxin on the DH secretion directly, we
incubated brain-subesophageal ganglion com¬
plexes isolated from female pupae in a cul¬
ture medium with or without these reagents.
DH released in the medium was estimated by
the amount of 3-hydroxykynurenin , a precur¬
sor of ommocrome pigments, in the ovaries
of the female pupae into which the medium
was injected. The results showed that GABA
inhibited the secretion of the DH from the
SEG of diapause-egg producers and that pic¬
rotoxin induced the release of the DH from
the SEG of non-diapause-egg producers .

These pieces of evidence suggested a possi¬
ble involvement of a GABAergic neuron in
the regulation of the DH secretion in the
silkworm.

EN 39

GENE ORGANIZATION AND EXPRESSION OF
BOMBYX INS, THE I N S U LI N- F AM I L Y PEPTIDES
INVOLVED IN INSECT METAMORPHOSIS.

A. Kawakami , M. Iwami* 1 . T. Adachi1’ , H.
Ishizaki , H. Nagasawa^ and A. Suzuki .
^Biol. Inst., Fac. Sci., Nagoya Univ.,
Nagoya, .Do p . of Agr. Chem.. Fac. of Agr . ,
Univ. of Tokyo, Tokyo, JDiv . of Cell
Differ., Natl. Inst. for Basic Biol.,
Okazaki .

Bombyxins are insulin-family peptides
that are derived from Bombyx mori and have
the activity of inducing metamorphosis in
another species of moth Sarnia cynthia
_r _i c__i n__i when tested by debrained-pupa
assay. Recently, we isolated a bombyxin
gene and clarified that bombyxin was
first synthesized in a precursor form. We
characterized here that many gene copies
exist in the Bombyx genome, and these are
classified into two families A and B
according to the nucleotide sequence
homologies between them. The numbers of
the gene copies in respsctive families are
estimated over ten by Southern blotting.
The two-family genes are organized such
that each of the A and B family genes are
paired and divergently transcribed. The
two-family genes share little sequence
homologies in the flanking regions except
the cap site, TATA-box and poly(A) signal,
although both family genes are expressed
t i s s u e - s p e c i f i c a 1 1 y in the brain as
assessed by Northern blotting.

EN 40

EXPRESSION OF BOMBYXIN GENES OF Bombyx mori
DURING L^R^AL-PUPAL DEVELOPMENT 1
T. Adachi ' 2 , S.Takiyg , Y. Suzuki ^
A.Kawakami^, M.Iwami , g . I jhizaki ,
H.Nagasawa and A. Suzuki . Div. of Cell
Differ., fcjatl . Inst, for Basic Biol.,
Okazaki, Biol . ^ In st . , Fac. 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 been shown to
consist of highly heterogeneous molecular
species. Bombyxin genes and cDNAs we have
already cloned can be classified into at
least three families. A, B and C. The base
sequence homologies among them are less
than 60% in all cases. Thus we can
distinguish each of the- specific genes or
mRNAs under a syringent condition of
hybridization experiment..

Quantitative analysis of each mRNA
contained in the 5th- in star 0-day. brain was
carried out by the use of standard bombyxin
A, B and C RNAs produced from the
respective genes or c DN As with T7
polymerase. Their concentration in
bombyxin-producing neurosecretory cells was
estimated as 3 - 5 x 10 molecules per ;ig
of total RNA, which is evaluated as
remarkably high. Northern hybridization
analyses revealed that the titre remained
essentially unchanged during larval-pupal
development .

EN 41

GENES ENCODING THE BOMBYX ! X- LIKE PEPTIDE
OF SAMI A CYNTHIA RICINI

M.Kimura , A . Kawakami M . Iwami ^ i
H . Ishizaki1, H.Nagasawa, A. Suzuki.

D. e p t . of Biol., .Fac. of Sci., Nagoya
Univ., Nagoya, ^Dept.of Agr. Chem., Fac. of
Agr., Univ. of Tokyo ,' Tokyo .

Bombyxin derived from Bombyx mori is a
brain hormon which^^induces metamorphosis
in Sarnia . The genes encoding bombyxin had
bee’ll isolated from Bombyx . Here we report
the isolation of homologous genes from the
Sarnia genome. A 2.7kb part of a clone
hybridized with a Bombyx bombyxin gene has
been sequenced to reveal two bomby xin-like
genes: the amino acid sequences predicted
from these genes show homology with the
precursor proteins for bombyxins and other

i n s u 1 i n - f am i 1 y peptides and the same
domain organization as in these precursors
is revealed. These genes lie close to each
other and are paired to be transcribed
divergently like bombyxin genes in the
Bomby x genome. Genomic Southern blotting
shows that Sarnia genome does not contain
multiple gene copies closely related to
either of these genes, contrasting to the
situation in Bombyx . Northern hybridi¬
zation analyses using these genes as
probes indicate that only one of the two
b o mb y x i n - 1 i ke genes is expressed in the
brain, suggesting that the other gene
might be non-functional.

Endocrinology

1307

EN 42

WHAT IS THE PHYSIOLOGICAL ROLE OF BOMBYXIN
IN BOMBYX MjORI ?

A.Mizoguchi , H.Ishizaki , H.Nagasawa and
A. Suzuki2. iBiol. Inst., Fac. of Sci.,
Nagoya Univ. , Nagoya and 2Dept. of Agr.
chem. , The Univ. of Tokyo, Tokyo.

Bombyxin which is produced in the brain
of Bombyx mori has a "PTTH" activity on the
prothoracic glands of Sarnia cynthia ricini .
Its role in Bombyx itself, however, is not
known. The bombyxin content in the Bombyx
brain was high in the first half but low in
the second half of the 4th larval instar.

It was high again at the beginning of the
5th larval instar, but decreased gradually
until larvae entered the wandering stage,
then increased and returned to its initial
level at the time of pupation. The gradual
decrease of bombyxin in the brain indicates
the continual release of bombyxin into the
blood. On the other hand, the increase of
that suggests the stop of bombyxin release.
The agreement of the time of bombyxin re¬
lease with the growing stage of larvae
suggests that the bombyxin is involved in
the regulation of larval growth.

The immunoblotting analyses using bom¬
byxin antibody demonstrated that some of
the bombyxin immunoreactive molecules
occured in the specific developmental stage
and that one of the bombyxin immunoreactive
molecules occured much more in female than
male larvae, suggesting that some of the
bombyxin family peptides are responsible
to the developmental stage- or sex-specific
functions .

EN 44

ON THE ANDROGENIC GLAND OF THE CRAYFISH,
PROCAMBARUS CLARKII.

Y. Taketomi and M. Murata, Dept, of Biol.,
Fac. of Sci., Kumamoto Univ., Kumamoto.

One of the sexual characters of male
crayfish, Procambarus clarkii , is that
there are pointed spines on the iscia of
the third and fourth walking legs. In the
present study, we found males having these
spines were able to mate with the females.
All the males more than 30 mm in carapace
length had these spines. These spines seem
to be essential for mating behavior. In
fact, the crayfish less than 25 mm in
carapace length had no spine, and were not
active sexually. The androgenic gland of
adult male was homogenized in Van
Harreveld's solution by use of a sonicator,
and the homogenate was injected into young
males 17-20 mm in carapace length. At the
next molt all of them developed these
spines. This suggests that the androgenic
gland hormone stimulates development of
these spines directly. The EM observations
revealed that the androgenic gland cells
are different in the crayfish with spines
from the ones without them. The cells in
the former show a highly developed rER and
Golgi complex, whereas the cells of the
latter have few such organelles but instead
have many vacuoles, suggestive of an
inactive gland.

EN 43

ON THE MANDIBULAR GLAND AND THE Y-ORGAN OF
THE CRAYFISH, PROCAMBARUS CLARKI .

Y. Taketomi, M. Motono and M. Miyawaki.
Dept, of Biol., Fac. of Sci., Kumamoto
Univ., Kumamoto

The function of the mandibular gland
of decapod crustaceans has not yet been
clarified. We have performed a series of
experiments designed to elucidate the role
of this enigmatic structure.

Mandibular glands of the crayfish,
Procambarus clarki (Reptantia) , were
homogenized in Locke's solution with the
use of a sonicator, and homogenate was
injected into shrimp, Caridina denticulata
(Natantia) which had just undergone
ecdysis. This species undergoes ecdysis
every 11-13 days at 26°C, and 14-16 days at
21 C. The control shrimp reached the
premolt stage of the next molt cycle (stage
D ) after 5-7 days at 26°C, but for the
shrimps that received the homogenate it was
reached after only 1-3 days .. Likewise , at
21°C, the shrimp attained D after 2-5
days, but for the control sRrimps not until
after 7-8 days.

These results suggest that the
mandibular gland of decapod crustaceans
either itself produces molting hormone(s)
and thereby accelerates the molting process
directly or produces a factor; that
activates the Y-organs.

EN 45

VITELLOGENIN SYNTHESIS IN ANDRECTOMI ZED
MALES OF ARMADILLIDI UM VULGARE .

S. Suzuki*, K. Yamasaki** & Y. Katakura***.
*Biol. Lab. Kanagawa Pref . Col., Yokohama.
**Dept. Biol. Tokyo Metropol. Univ., Tokyo.
***Dept. Biol. Keio Univ., Yokohama.

When adult females were ovariectomized ,
synthesis of vitellogenin (Vg) lasted. On
the other hand, females transplanted andro¬
genic glands from males decreased the syn¬
thetic activity for Vg in their fat body.

In order to elucidate the regulatory role
of the androgenic hormone for Vg synthesis,
the occurrence of Vg was studied with andr-
ectomized males. Three pairs of androgenic
glands were ablated from males at the 5th
stage of post embryonic development (3 mm)
in carapace length) . Vg was estimated using
rocket immunoelectrophoresis . Vg synthesis
in the fat body was analyzed by SDS-PAGE.

Andrectomized males ceased the elon¬
gation of the endopods (copulatory organ ) .
Several months later following operation
they gave to rise to adult size (7.5 mm).
These animals contained Vg in their hemo-
lymph. However, Vg synthesis in their fat
body was low level and the site of it could
not be defined.

These results indicate that andrectomized
males seem capable of Vg synthesis, so one
of the role of the androgenic hormone is to
inhibit Vg synthesis. In A. vulgare ,
ovarian factors may not be involved in the
regulatory mechanism for Vg synthesis.

1308

Endocrinology

EN 46

FMRF- AM I DE-LI KE I MMUNOREACT I V I TY IN THE
NERVOUS SYSTEM OF THE EARTHWORM, El SEN I A
FOETIDA

K. Fuj i i , N.Ohta, T. Sasaki, Y.Sekizawa and
H.Kobayashi. Zenyaku Kogyo Co., Ltd.,
Research Laboratory, Tokyo

A number of large and small FMRFamide-
like i mmunoreac t i ve cell bodies were found
in the cerebral and subesophageal ganglia,
in the segmental ganglia of the ventral
nerve cord, and in the ganglia in the
gangliated chain in the pharyngeal plexus
of the earthworm, Ei sen i a f oet ida. From the
cerebral and subesophageal ganglia, immuno-
reactive fibers proceed to the prostomium
and the 1st segment, respectively. From
the ganglia in the gangliated chain, the
i mmunoreac t i ve fibers proceed towards the
pharynx. From each segmental ganglion,
three i mmunoreac t i ve segmental nerve trunks
emerge bilaterally and extend to the body
wall. Then they run dorsal ly between the
circular and longitudinal muscle layers of
the body wall, branching off into the skin.
Furthermore, i mmunoreac t i ve fibers were
observed between the epithelial cell layer
and the muscle layer of the alimentary
canal. I mmunoreac t i ve products are
occasionally present in contact with the
blood vessels in the cerebral and
subesophageal ganglia. FMRFamide seems to
act both as a neuro transmi t ter or neuro¬
modulator and as a neurohormone in the
ear thworm .

EN 47

DEVEROPMENTAL CHANGES IN ATRIAL NATRIURE¬
TIC PEPTIDE-IMMUNORE ACTIVE ( ANP-ir )
GRANULES OF TOAD CARDIOCYTES .

T. Hirohama, S. Nakamura, H. Uemura and
T. Aoto. Biol. Lab., Kanagawa Dental
Coll., Yokosuka.

The ontogeny of ANP-ir granules in
cardiocytes was studied in developing and
adult toads ( Buf o j aponicus f ormosus ) .
ANP immunoreactivi ty was determined by
the postembedding immunogold technique.
ANP-ir granules ( 120 nm ) were first

noted in both the atrial and ventricular
cardiocytes of limb-bud stage larva
( St. 28 ). Their number was somewhat

greater in atrial than ventricular
cardiocytes and, as development proceeded,
showed rapid increase only in the atrial
cardiocytes. Two kinds of ANP-ir
granules were discernible in the same
cardiocytes of adult atrium and
ventricle. Their median size was 120 nm
and 200 nm, respectively.

Some specimens were hypophysectomized
at the late neurula stage, and their
cardiocytes compared with those of intact
larva which were in the metamorphic stage
( St. 44 ). In hypophysectomized larva,

atrial cardiocytes possessed more
abundant and larger (the median, 130 nm)
granules than those of intact controls.

EN 48

SIGNIFICANCE OF ANP EEL CARDIOCYTES.

S. Nakamura1, T. Hirohama1, H. Uemura1, T.
Aoto1 and M. Naruse . 1Biol.~ Lab., Kana¬
gawa Dent. Coll., Yokosuka, ZDept. of Med.,
Tokyo Women's Medical Coll., Tokyo.

Secretory granules in the atrial cardio¬
cytes of the eel Anguilla j aponica were
found immunocy tochemical ly (protein A-gold
technique) to contain an atrial natriuretic
peptide (ANP)-like substance. The number of
these "ANP granules" underwent no change in
15 min or 1 h after transferring the fish
from freshwater (FW) to 100% artificial
seawater (SW). However, after 3 h in SW,
secretory granules were noted to start de¬
creasing and after 24 h, both they and the
Golgi cisternae were present in only slight
number. In the ventricular cardiocytes of
the fish after 24 h in SW, however, well-
developed Golgi cisternae and many granules
could be observed. In eels kept for a week
in SW, atrial cardiocytes contained ANP
granules in abundance and well-developed
Golgi cisternae, but Golgi cisternae in the
ventricular cardiocytes were developed to a
lesser degree and the number of granules
was less than that of the control (SW) fish.
After one month in SW, the Golgi body and
ANP granules in the atrial and ventricular
cardiocytes were essentially the same in
development and number as those of the con¬
trol. With respect to ANP production, ven¬
tricular cardiocytes may augment the inade¬
quacy of atrial cardiocytes.

EN 49

EFFECTS OF HYPOPHYSECTOMY AND THYROID¬
ECTOMY ON ATRIAL NATRIURETIC PEPTIDE-LIKE
IMMUNOREACTIVITY (ANP-LI) IN LARVAL TOAD
CARDIOCYTES.

H. UEMURA1, T. HIROHAMA1, S. NAKAMURA1, S.
KIKUYAMA2 and T.AOTO1. 1Biol.Lab., Kana¬
gawa Dent. Coll., Yokosuka, 2Dept. of
Biol., Waseda Univ. , Tokyo.

Ontogeny of ANP-LI in toad heart
tissue was studied, applying the PAP
method to intact, hypophysectomized ( HX )
and thyroidectomized ( TX ) larvae of the
toad Buf o j aponicus f ormosus . Also, some
TX animals treated with 1 0“^M T4 ( TXT4 )
were used. In intact animals, ANP-LI
was first detected in atrial and
ventricular cardiocytes of the larva with
completed operculum. During development,
ANP-LI decreased but was consistently
positive in the atria. In the
ventricles, it disappeared soon but was
again slightly evident in toadlets.
In all experimental groups, atrial
cardiocytes consistently showed
stronger ANP-LI than those of intact
controls. The ANP-LI was evident in the
ventricular cardiocytes of TX animals,
while weakly positive in TXT4 and HX
animals. From these findings, it thus
follows that neither hypophysis nor
thyroid may be essential for ANP
synthesis in larval hearts but thyroid
hormones may possibly modulate synthesis
and/or release of ANP from the
cardiocytes, at least of the ventricle.

Endocrinology

1309

EN 50

DEPRESSOR EFFECT OF ATRIAL NATRIURETIC
PEPTIDES (ANP) IN THE QUAIL: A STRUCTURE-
ACTIVITY RELATIONSHIP.

Y. Takei1 and T. X. Watanabe . Dept.
Physiol., Kitasato Univ. Sch. Med.,
Sagamihara, and 2Protein Res. Inst., Minoh.

An intravenous injection of (1-28) human
ANP ( hANP ) decreased arterial pressure and
heart rate in the quail. The minimum
effective dose was as low as 16 pmol/bird.
The degree of hypotension was almost 50% of
the normal level at high doses, and the
effect was quite reproducible even after
repetitive injections. Thus, the depressor
activity in the quail appears to be a
useful bioassay system for ANP _in vivo .
The hANP peptides with amino acids removed
from the NH2-terminus were as potent as
hANP, but the removal of amino acids from
the COOH-terminus decreased the potency.
Modifications of amino acids within the
ring structure of hANP had variable effects
on the potency. Thus, amino acid residues
at the COOH-terminus and some amino acids
within the ring structure of hANP are
important for the expression of its
depressor activity in the quail. Immuno-
histochemically , quail hearts contain
substances that cross-react with
antibodies to hANP, but the content in the
heart measured by radioimmunoassay of hANP
was 1/250,000 of that measured by depressor
bioassay in the quail. Thus, quail appear
to have ANP-like substance in the heart,
but its amino acid sequence may differ from
hANP.

EN 51

IMMUNOHI STOCHEMICAL STUDIES OF JUXTA¬
GLOMERULAR CELLS AND THE CORPUSCLES OF
STANNIUS IN THE JAPANESE EEL.

12 1
H.Tazawa , M.Mukai and M.Ogawa

^Dept.Biol. ,Saitama Univ. ,Urawa , Saitama ,
^Dept . Pathol . ,Keio Univ . Sch . Med . , Tokyo.

The localization of immunoreactive renin
in the juxtaglomerular cells (JGC) and the
corpuscles of Stannius(CS) was studied in
the Japanese eel , Anguilla japonica. Anti¬
serum against mouse submandibular gland
renin raised in rabbits was used for the
immunohistochemical study.

In the teleostean kidney, JGC have been
observed in the walls of the afferent
arterioles and/or the small arterial
branches. The CS have been implicated in
calcium homeostasis in fishes. Renin-like
activity was found in the CS of the carp
(Sokabe et. al.,1970).

The cells resembling JGC in the walls
of the sfferent arterioles and the small
arterial branches were intensely immuno¬
reactive to anti-renin serum. The immuno¬
reactive JGC were located extensively and
dispersedly. They were detectable along the
tunica media of arterial vessels . However ,
no renin immunoreact ivity was detected in
the eel CS even though immunoreact ivity
occurs in the kidney in the same section.
This result suggests that the eel CS do not
contain any substances which react with
anti-mouse submandibular gland renin serum.

EN 52

CALCITONIN-LIKE IMMUNOREACTIVE SUBSTANCE
FOUND IN PANCREATIC ISLETS OF MEDAKA (TELE-
OST), ORYZIAS LATIPES

Y.Sasayamai T. Koizumi, C.Oguro’and A.Kambe-
gawa2 ^ept. of Biol., Fac. of Sci., Toyama
Univ., Toyama, 2Dept. of Obst. Gynecol.,
Fac. of Med., Teikyo Univ., Tokyo. _

In Medaka, calcitonin (CT)-like immuno-

reactivity was found in pancreatic islets
by PAP method using anti-salmon CT anti¬
serum. Number of the cells which show
the reactivity was by far many than in
ultimobranchial gland. These cells were
also stained with anti-eel CT antiserum.
However, anti -human CT antiserum did not
bring about positive reaction. Therefore,
it was suggested that the immunoreactive
substance has some common features with
fish CT. Anti -somatostatin antiserum caused
also positive reaction in pancreatic is¬
lets, but it was in different cells. CT-
like immunoreactive cells were also stained
with aldehyde fuchsin. These facts imply
that these cells may be identical with
pancreatic B cells. Furthermore, administ¬
ration of extract of the islets to rats
brought about significant decreases in
the serum Ca and Pi levels at 30 min after.
However, the hypocalcemic potency was
by far weak when compared with that of
extract of ultimobranchial gland. There¬
fore, it was suggested that this CT-like
immunoreactive substance in pancreatic
islets may have not hypocalcemic function
but some unknown functions in Medaka.

EN 53

HISTOLOGY OF ULTIMOBRANCHIAL GLAND (UB)
AND PARATHYROID GLAND (PT) OF CRAB-EATING
FROG, RAN A CANCRIVORA

Y.Sasayamal C.Oguro^nd T.Ogasawara2 ^ept.
of Biol., Fac. of Sci., Toyama Univ., Toyama
2Ocean Res. Inst., Univ. of Tokyo, Tokyo _

In crab-eating frog, the control mecha¬
nisms of divalent salts has not been inves¬
tigated so far. In the present study,
histology of UB and PT was examined.

UB was expected to be activated. The
diameter of UB was 200-400 pm in frogs
(7-35g) examined. Most of histological
features were similar to those in other
common frogs. However, hypertrophy was
noted in the parenchymal cells. Further¬
more, those cells exhibited intense immuno-
reactivity to anti -calcitonin antiserum.
It was known also by EM that secretory
granules are distributed throughout the
gland. These features show that UB is
activated chronically in this species.
On the other hand, the diameter of PT
was about 200-400 pm. PT was expected
to be inactivated. However, the histo¬
logical features does not differ from
those in other anurans. Also in observa¬
tions by EM, there was no symptom of dege¬
neration. This implys that PT is functional
at least as well as in the same degrees
in other common frogs. It seems that this
phenomenon is interesting, when taking
high Ca environment into considerations.

The present study was supported by
Grant-in-Aid for Overseas Scientific Re¬
search (No. 62041035).

1310

Endocrinology

EN 54

CALCITONIN ACTIVITY OF THE ULTIMOBRANCHIAL
GLAND OF THE CRAB-EATING FROG.

C. Oguro, T. Ishijima and Y. Sasayama. Dept,
of Biol., Fac. of Sci., Toyama Univ., Toyama.

The crab-eating frog, Rana cancrivora is
known only amphibian species which can tole¬
rate to estuarine water. Serum monovalent
ion and urea concentrations were reported
in this species. However, nothing has been
known on serum divalent ion concentrations.
In the present study, the crab-eating frogs
were collected in the coast of Thailand.
Serum Ca concentration was higher than that
of the other frogs. It is of interest to
know calcitonin (CT) activity of the ultimo-
branchial gland (UBG) in this species in
relation to high serum Ca concentration.

Some adults were sent to Japan and their
UBG were removed for the measurement of CT
activity. Hypocalcemic activity was deter¬
mined by rat bioassay and compared with
salmon CT activity. Average CT activity of
this species was 25A9 mU/lOOg bw. The
activities in the frogs, Rana nigromaculata
and R. ornativentris and the newt, Cvnoos
pvrrhogaster were 850, 1550 and 13 mU/lOOg
bw, respectively. These results indicate
that CT activity of UBG in this species is
fairly high among amphibians.

Supported in part by Grant-in-Aid for Over¬
seas Scientific Research from the Ministry
of Education, Science and Culture of Japan
(no. 62091035) .

EN 56

EFFECTS OF ADRENALECTOMY ON SERUM MINERAL
AND GLUCOSE CONCENTRATIONS IN THE SNAKE,
RHABDOPHIS TIGRINUS TIGRINUS.

C. Oguro, E. Ohba and Y. Sasayama. Dept, of
Biol., Fac. of Sci., Toyama Univ., Toyama.

Function of the adrenal gland ( ADG ) of
reptiles has not been thoroughly under¬
stood because of the difficulty of surgical
removal of ADG. Recently, a technique for
complete adrenalectomy (ADX) in snakes
without defect in kidney function has been
developed in this laboratory.

In the present study, effects of ADX on
serum mineral and glucose concentrations
were studied in the water snake. Further¬
more, effects of corticosterone administ¬
ration were examined after ADX.

Ten days after ADX, serum K concentration
increased by 36%, while serum Na, Ca and Mg
levels were not significantly changed.

Serum glucose concentration decreased by
68%. Replacement therapy resulted in a
shift of serum K and glucose values to the
normal level, although the values were not
completely corrected in the dose used in
the present study (5ug/10g bw/day).

The present results confirmed our former
conclusion that in snakes ADG is involving
in the control of serum K and glucose con¬
centrations. However, ADG has a minor role
in the control of serum Na concentration.

Supported in part by Grant-in-Aid for
Scientific Research from the Ministry of
Education, Science and Culture of Japan
(no. 63540598) .

EN 55

STRUCTURE OF KIDNEY AND INFLUENCE OF AC¬
CLIMATION TO VARIOUS SALINITIES IN THE
CRAB-EATING FROGS, RANA CANCRIVORA.

M. Uchiyama1 , T. Murakami1 , hT Yoshizawa2
and C. Wakasugi3. ^ept. of Oral Physiol.,
The Nippon Dental Univ., Niigata, 2Dept. of
Histol., Matsumoto Dental Coll., Shioj iri
and 3Dept.of Oral Histol., The Nippon Den¬
tal Univ., Niigata.

The structure of nephrons of the eu-
ryhaline frog was investigated by means of
light and electron microscopy. The nephron
has the following parts; renal corpuscle,
ciliated neck segment, proximal tubule,
ciliated intermediated segment, distal
tubule, connecting tubule and collecting
tubule. The proximal tubules are composed
of cylindrical cells with a dense luminal
brush border of long microvilli and
numerous apical vesicles and vacuoles. The
distal tubule consists of heavily inter-
digitated cells. Mitochondria of this por¬
tion are prominent with a very regular
palisade arrangement. The connecting tubule
consists of the clear cells and the
canaliculi cells. These findings are simi¬
lar to the structure of the other amphib¬
ian kidneys. In the nephrons of crab-eating
frog adapted to high salinities, the size
of glomeruli was reduced and the apical
cytoplasm of the epithelial cells showed
hyperplasia in distal portions of distal
tubules. These structures may be related to
a little production of hypotonic urine in
concentrated media.

EN 57

HYPERCALCEMIA IN THE OVIPAROUS SNAKE, Elaphe quad-
rivirgata , AT THE END OF OPPOSITION.
M.Yoshihara , M.Uchiyama , T. Murakami1.
H.Yoshizawa , N. Suzuki2, Y. Sasayama2, M. Fujimori2
and C. Oguro3. 1Dept. of Oral Physiol., Sch. of
Dentistry at Niigata, The Nippon Dental Univ.,
Niigata, 2Dept. of Oral Histol., Matsumoto Dental
Coll., Shioj iri, 2Dept. of Biol., Fac. of Sci.,
Toyama Univ. , Toyama. _

In the oviparous snake, Elaphe guadrivirqata ,
plasma Ca, Mg and inorganic phosphorus (Pi) levels
at circa-oviposition were studied. Plasma Ca level
of two weeks before oviposition was 12.5 mg/ 100
ml. This level was lower than that in the male
snake (14.2 mg/100 ml). At the end of oviposition,
plasma Ca level increased (15.6 mg/100 ml) and it
decreased two weeks after oviposition (14.6 mg/100
ml). On the other hand, plasma Mg and Pi levels at
the end of oviposition were lower than those of
two weeks before oviposition and two weeks after
oviposition. Plasma Mg levels of two weeks before
oviposition, at the end of oviposition, two weeks
after oviposition and the male snake were 3.4,
2.9, 3.3 and 3.1 mg/100 ml, respectively. Plasma
Pi levels of those specimens were 4.5, 3.3, 4.4
and 3.8 mg/100 ml, respectively. The hypercalcemia
at the end of oviposition may be a result of over¬
shooting of Ca mobilization from the bone after
the end of eggshell calcification in the oviduct.
Similar results have been reported in laying hen.

Endocrinology

1311

EM 58

GROWTH HORMONE KINETICS DURING SEAWATER
ADAPTATION OF RAINBOW TROUT.

T. Sakamoto, T. Ogasawara and T. Hirano.
Ocean Res. Inst., Univ. of Tokyo, Tokyo.

To clarify the mode of growth hormone
(GH) action in salmonids during seawater
(SW) adaptation, we examined kinetics of GH
in chronically cannulated rainbow trout,
weighing about 1 kg. After direct transfer
to SW the fish failed to survive for more
than 2 days. When they were transferred
from fresh water (FW) to 75% SW, they
adjusted the plasma Cl concentration within
1 week. Following transfer to 75% SW, the
plasma GH level increased during days 1-3,
then decreased to the initial level at day
4 and remained constant until day 7. In
contrast, plasma cortisol levels increased
dramatically within 1-2 hours after
transfer. These results suggest that both
hormones have important roles in
adaptation to an hyper-osmotic environment.

Metabolic clearance rate (MCR) and
secretion rate (SR) of recombinant chum
salmon GH were calculated after intravenous
injection of GH. The MCR and SR were not
affected by injected doses of GH between
8.4 and 42 ng/g. Four days after transfer
to 75% SW, both the MCR and SR were five
times higher than in FW. However after 3-4
weeks the MCR and SR returned to the FW
levels. Thus, GH seems to be involved in
the development of hypoosmoregulatory
mechanisms in rainbow trout, especially
during the early phase of adaptation.

EN 60

SEASONAL CHANGES IN SEAWATER ADAPTABILITY
OF LAND-LOCKED SOCKEYE SALMON ( ONCORHN -
CHUS NERKA) .

T. Yada1 , K. Takahashi2 and T. Hirano .
^Ocean Res. Ins., Univ. of Tokyo, Tokyo and
2Yamanashi Pref. Fishry Center, Yamanashi.

Development of seawater (SW) adaptability
was assessed by SW-charenge tests in hime
salmon, land-locked sockeye salmon ( 0.
nerka ) . Compared with amago salmon ( 0.
rhodurus ) , hime salmon showed greater SW
adaptability at all stages until spring of
the 2nd year. There was a tendency toward
an increased adaptability in spring. In
most Salmonids, SW adaptability is
accompanied with parr-smolt transformation,
and some hormones are suggested to be
involved in the process. In hime salmon,
plasma thyroid hormones showed seasonal
changes coinciding with morphorogical
changes associated with smoltif ication.
Plasma levels of growth hormone (GH)
decreased while prolactin (PRL) levels
increased significantly in spring , when
the SW adaptability increased maximally.
After SW transfer, PRL levels decreased in
both summer and spring, and GH levels
increased only in spring, suggesting
involvement of PRL in freshwater adaptation
and GH in SW adaptation.

EN 59

PROLACTIN KINETICS IN THE RAINBOW TROUT.

T. Ogasawara, S. Hasegawa and T. Hirano.
Ocean Res, Inst., Univ. of Tokyo, Tokyo.

It has been well estabilished that

prolactin (PRL) plays an important role in
maintaining hydromineral balance of
eur yhaline teleosts in fresh water (FW).
When measured by a homologous RIA, the
plasma PRL concentrations of the rainbow
trout were kept at low levels either in FW
or in seawater (SW). When transferred
from SW to FW , an increase in plasma PRL
was seen during the first 1-10 days.

In the mature chum salmon, on the other
hand, plasma PRL in the male showed a
little increase on transfer to FW , whereas
a marked increase was seen in the female.
Since plasma level of a hormone reflects
equilibrium between the secretion rate and
the matabolic clearance rate, kinetic
study was undertaken to clarify the mode
of action of PRL. Salmon PRL was injected
through a cannula implanted into the
dorsal aorta of the rainbow trout (1 kg),
and changes in plasma PRL were followed
until 7 hrs after the injection. Injected
PRL was cleared from circulation earlier
in the fish in FW than in 75% SW .

Metabolic clearance rates and secretion
rates of PRL were 69.6 ml/hr. kg and 13.8
ng/hr.kg in FW , and 50.3 ml/hr. kg and 6.9
ng /hr . kg in 75% SW fish, respectively.
Turnover rate of PRL in the rainbow trout
seems to be greater in FW than in SW .

EN 61

EFFECTS OF THYROID HORMONES AND CORTISOL ON
FIN RESORPTION IN OLIVACEUS: AN IN VITRO
STUDY.

E. de Jesus1, Y. Inui^, S. Miwaz and T.
Hirano1. 1Ocean Research Institute,
University of Tokyo ^National Research
Institute for Aquaculture _ _

' Thyroid hormone leve 1 s in t founder

show remarkable changes during early
development. T3 concentration is high in
fertilized flounder eggs, declines
gradually until hatching then drops sharply
to below 1 ng/g one day after hatching. In
contrast, T4 level is below 1 ng/g in eggs
and larvae from the time of hatching
through early stages of metamorphosis. At
the onset of climax, T4 increases to a peak
value of 1 4 ng/g at climax and decreases by
about half at post climax. To better
understand the coincident increase in whole
body concentrations of T4 and T3 with
metamorphic events, in vitro studies on fin
resorption were done using elongated dorsal
fin rays from prometamorphic fish. Results
show that addition of thyroid hormones into
the culture medium accelerated fin
resorption in a dose dependent manner with
T3 being more potent than T4. When
cortisol was added with T4, an enhancing
effect was seen in combination with the
lower T 4 dose. Cortisol alone had no
effect. In a preliminary study,
significant amounts of cortisol were
detected in flounder embryos at an early

developmental stage and in metamorphosing
larvae; the concentration in the latter
seemed greater than in the former.

1312

Endocrinology

EN 62

COMPARATIVE PROFILES OF THYROID HORMONES
DURING EARLY DEVELOPMENT OF SEVERAL TELEOST
FISHES.

M.Tagawa^ , T.Hirano^ and M . Tanaka ^ . ^Ocean
Res. Inst., Univ. of Tokyo, Tokyo and
^Dept. of Fish., Fac. of Agr. , Kyoto Univ.,
Kyoto.

In order to examine the role of thyroid

hormones during early developmental stages
of teleost fishes, thyroxine (T4) and
triiodothyronine (T3) in eggs and in whole
larvae were extracted and measured in RIAs.
T4 and T3 concentrations in eggs varied
among different species. Significant
elevation in T4 concentrations was observed
during metamorphosis in flounders and
during transformation from larvae to
juveniles in ayu and greenling. A tran¬
sient increase in T4 concentration was also
observed at the stage when larvae change
their habitat in nature in chum salmon,
black sea bream and some other species. No
significant changes in T4 and T3 concentra¬
tions were seen when there were no remark¬
able changes in larval forms and habitat as
in moroko, Gnathopogon elongatus caeru-
lescens . In sea perch, Pacific herring and
striped jack mackerel, although there was
no close correlation between developmental
events and elevation in T4 concentration, a
gradual increase in T4 and T3 concentra¬
tions was observed during early develop¬
ment. These observations suggest the in¬
volvement of thyroid hormones in larval
metamorphosis and changes in habitat in
fishes .

EN 63

A HOMOLOGOUS RADIOIMMUNOASSAY FOR BULLFROG
THYROTROPIN

M. Sakai, K.Niinuma, S.Kikuyama, S. Tanaka1,
Y.Hanaoka1 and H.Hayashi1

Dept, of Biol., Sch. of Educ. , Waseda
Univ., Tokyo. 1Inst. of Endocrinol., Gunma
Univ., Maebashi.

A double antibody radioimmunoassay (RIA)
for bullfrog thyrotropin ( TSH ) was
established. The antiserum was raised
against 3-subunit of highly purified
bullfrog TSH. Bullfrog TSHB was used for
r ad i o i od i na t i on and as a reference
standard. The standard curve was obtained
with 0.1-51.2 ng/ml of bullfrog TSHB •
Bullfrog intact LH, FSH and their subunits
did not react in this assay. When bullfrog
sera were assayed by this RIA system, the
intra-assay coefficient of variation and
inter-assay coefficient of variation were
3.1% and 3.6%, respectively. The accuracy
of the assay was shown as the recovery of
bullfrog TSHB added to the pooled bullfrog
sera. With a range of 0.8-25.6 ng/ml, the
recovery was 103.3 + 13.3% (Mean+S.D.).
Several dilutions of serum of both adult
and larval bullfrogs yielded dose-response
curves which were parallel to the standard
curve. RIA of serum TSHB in larval
bullfrogs of various developmental stages
was performed. Serum TSHB levels were
relatively low during prometamorphosis.
The levels rose markedly at the onset of
climax and declined at the end of climax.

EN 64

H0M0L0GUS RADIOIMMUNOASSAY OF BULLFROG
GROWTH HORMONE

T. Kobayashi1, S. Kikuyama2, T. Ogasawara3,
T. Hirano3.

1) Dept. of Basic Human Sciences, Sch. of
Human Sciences, Waseda Univ., Saitama.

2) Dept. of Biol., Sch. of Educ., Waseda
Univ., Tokyo. 3)0cean Res. Inst., Tokyo
Univ . , Tokyo .

Antiserum to bullfrog(f) growth
hormone(GH) was produced by immunizing
rabbits with highly purified preparation.
The antiserum together with fGH and 1251-
f GH were used to develop a
r ad i o i mmuno a s s ay ( R I A ) for frog GH.
Several dilution of plasma as well as
pituitary homogenate yielded dose
response curves which were parallel to
the standard curve. Plasma from
hypophysectomized bullfrogs contained no
measurable GH. Ovine PRL and GH, eel and
salmon GHs , f LH , f FSH , f TSH , f PRL and
bullfrog neurointermediate homogenate did
not react in this assay. Plasma and
pituitary homogenate of Xenopus laevi s
gave inhibition curves which did not
parallel to the standard. The interassay
coefficient of variation, intraassay
coefficient of variation and sensitivity
of the assay were 4.3%, 4.1% and 0.97 ±
0.44 ng/ml, respectively. RIA of plasma
GH in larval bullfrogs revealed that GH
levels rise gradually as metamorphosis
progresses, reaching maximum (78 ng/ml)
at the end of metamorphosis.

EN 65

THE ROLE OF THYROID HORMONE AND
ALDOSTERONE ON THE DIFFERENTIATION OF
THE ACTIVE Na TRANSPORT SYSTEM ACROSS
TADPOLE SKIN DURING METAMORPHOSIS.

M. TAKADA, Dept, of Physiol., Saitama
Med. Sch., Moroyama, Saitama.

Dif f erentiatrion of the active Na
tran'sport system across R. catesbeiana
skin may be controlled by hormones since
the system appears during metamorphosis
and the serum concentrations of thyroid
hormone and aldosterone increased during
that period. An injection of
aldosterone ( 200 nM) into a tadpole
every other day for 2 weeks had no
effect on the differentiation. On the
other hand, an injection of T3 (10 nM )
every other day for 2 weeks induced
differentiation. The differentiation
was still induced by T3, even when
spironolactone (20 uM ) was injected.
The electromotive force of the active Na
transport induced by the injection of a
mixture of T3 and aldosterone was larger
than that induced by an injection of
only T3. The differentiation of the
active Na transport system during
metamorphosis is induced by thyroid
hormone alone. Aldosterone is able to
stimulate the active Na transport system
which is already differentiated, but it
is not able to induce differentiation.

Endocrinology

1313

EN 66

LUNG MATURATION OF BULLFROG TADPOLES DURING
METAMORPHOSIS

M.Ohkawa' , M.Mita2, N.Ueta2 and S.Kikuyama'
^ept. of Biol., Sch. of Educ . , Waseda
Univ. , Tokyo 169 and 2Dept. of Biochem. ,
Teikyo Univ., Tokyo 173.

In amphibians, the lung starts to function
as a respiratory organ at metamorphosis. It
is well known that in the mamalian fetus,
the lung surfactant which is mainly com¬
posed of phosphatidylcoline (PC) appears as
lung maturation progresses. As a prelimi¬
nary step for the investigation of the
mechanisms involved in lung maturation in
bullfrog tadpoles ( Rana catesbeiana ) , lipid
analysis in the whole lung and isolated
surfactant from tadpoles of various
developmental stages was conducted. PC was
a principal lipid both in the lung and in
the surfactant. The main fatty acid com¬
ponent of PC was palmitic acid. The amount
of whole lung lipid remained rather con¬
stant throughout TK stage 18-24. On the
other hand, the amount of surfactant lipid,
especially PC increased markedly around
stage 22, suggesting that lung maturation
in bullfrog tadpoles takes place at late
climax.

EN 67

CELL DEATH IN THE ANURAN TADPOLE TAIL:
EFFECTS OF TRIIODOTHYRONINE ON FIBROBLASTS.
A. Nishikawa and K. Yoshizato. Dept, of
Biology, Tokyo Metropolitan Univ., Tokyo. _

To know the cell death mechanism during

metamorphosis of anuran amphibia, the
effects of triiodothyronine (T^) on DNA
synthesis of fibroblasts and collagen
synthesis in dermis were examined and those
of tail skin and body skin were compared.

Fibroblasts from tail fin and dorsal body
skin of Rana catesbeiana tadpole were
cultured for one week. The DNA synthetic
actibity (DSA) of fibroblasts was measured
by labeling with 3H-thymidine for 12 hr.

The DSAs of tail and body fibroblasts were
3.5x102 dpm/pg of DNA and 1.3x103 dpm/pg of
DNA, respectively. To (10-°M) decreased
the activity to 1 /9 of control value in
tail cells. In the case of body cells,
however, the activity increased 1.5-fold in
the presence of T^ (10~°M) and cortisol
(5x10-'M) during culture. These results
suggest that T3 has opposite role for
growth of tail and body fibroblasts.

Collagen synthesis of body skin and tail
skin (fin) were measured by labeling in
vivo with ' 4C-proline and analysis for
proline-hydroxylation with HPLC. The
activity of proline-hydroxylation was
decreased by T3 (1 0"°M) to 65% of control
value in tail skin. On the other hand in
body skin, T3 increased the activity 2.3-
fold. These results suggest that T3
differently regulate the collagen synthesis
of tail and body skin.

EN 68

IMMUNOREACTIVE OXYTOCIN IN DERMAL GLAND OF
BUFO JAPONICUS.

K.Kawamura and S.Kikuyama. Department of
Biology, School of Education, Waseda Univ.,
Tokyo 169.

Oxytocin-like substance was immunohisto-
chemically detected in the dermal granular
gland of the toad, Bufo japonicus. This
immunoreactive oxytocin (irOT) coexisted
with CRF- or sauvagine-like substance in
the gland. IrOT appeared at metamorphic
climax and disappeared after the completion
of metamorphosis. In the hypophysectomized
tadpoles, neither irOT nor glandular struc¬
ture developed in the skin. If hypophys¬
ectomized tadpoles were artificially meta¬
morphosed with T4, the skin glands
developed, but they were irOT-negative . If
these animals were supplemented with ACTH
and prolactin ( PRL ) in addition to T4 , irOT
became detectable. However, treatment with
ACTH and PRL alone did not induce the
formation of the dermal glands. According¬
ly, T4 , ACTH and PRL are necessary for the
irOT-containing dermal glands to develop.

If the neural crest was removed from early
neurulae, the glands failed to develop,
suggesting that the irOT-positive cells are
derived from the embryonic neural tissue.

Acid extract of the toadlet skin was
analyzed with reversed phase HPLC, and the
fractions were tested for the immuno-
reactivity. It was found that the reten¬
tion time of the irOT is different from
that of authentic oxytocin nor mesotocin.

EN 69

INDUCTION OF MEIOSIS IN CULTIVATED MALE
EEL ( Anguilla j aponica)

' 2T.Miura 2A.Kanamori 2H.Ueda ^K.Yamauchi
and 2Y.Nagahama Fac. of Fisheries, Hok¬
kaido Univ., Hakodate, 2Labo. of Reprod.
Biol., Natl . Inst. For Basic Biol.,
Okazaki, 3Univ. of Occupational and En-
viromental Health, Sch. of Med., Kitakyushu

In cultivated male eel, germ cells
are all spermatogonia, and meiotic cells
are not observed in the testis. In this
study, male eels were given a single in¬
jection of human chorionic gonadotropin
(?ICG) (5 IU/g BW) , and histrogical changes
in the testis were observed. Serum 11-
ketotestosterone (11 -KT) profiles and 11-
KT production by the testis in vitro were
mesured. 1) Histology of testis; One day
after HCG injection, Sertoli cells and
Leydig cells were activated. After 6 days,
spermatogonia multiplied mitotically, and
after 12 days, some germ cells started
meiosis. 2) Serum 1 1 -KT profile; Initial
concentration of 1 1 -KT were relatively low
(0.48 ng/ml). Six days after HCG injec¬
tion, the concentration increased
remarkably, to 7.39 ng/ml. After 12 days
it decreased to 4.62 ng/ml 3) In vitro
production of 1 1 -KT by the testis; The
testis of uninjected eels produced 1 1 -KT
when cultured in vitro with HCG. Produc¬
tion was proportional to the concentration
of HCG. These results indicated that a
single injection can induce the d the in¬
itiation of meiosis, coincident with
remarkable endocrinological changes.

1314

Endocrinology

EN 70

MECHANISM OF 1 7a, 203-DIHYDRQXY-4-PREGNEN-3-
ONE PRODUCTION BY THE RAINBOW TROUT TESTIS
DURING SPERMIATION

N. Sakai, *A H. Ueda,* A. Kanamori* N. Suzuki A
and Y. Nagahama* *Lab. of Reprod. Biol.,
Natl. Inst, for Basic Biol., Okazaki , ANoto
Mar. Biol. Lab., Kanazawa Univ., Uchiura,
#Univ. of Occupational and Environmental
Health, Sch. of Med., Kitakyushu _

The sites and mechanism of production of

1 7a, 20$-dihydroxy-4-pregnen-3-one ( 1 7a, 208-
diol) were examined using preparations of
sperm-free testicular fragments and sperm
of the rainbow trout ( Salmo gairdneri )
testis during spermiation. _I n vitro
production of 17a,203-diol by intact
testicular fragments and by co-cultures of
sperm-free testicular fragments and sperm
preparations was markedly enhanced by
gonadotropin ( GTH ) , but neither of the
latter alone were capable of producing
17a,203-diol in response to GTH. When the
preparations were incubated with [14C] 17a-
hydroxyprogesterone, sperm-free testicular
fragments metabolized the precursor to
various androgens. 17a,208-Diol was the
only metabolite identified from incubations
with sperm preparations. Specific binding
of [ 5I] chum salmon GTH was highest in
sperm-free testicular fragment membrane
preparations. These results indicate that
the interaction of testicular somatic cells
and sperm is necessary for the production
of 17a,208-diol in response to GTH, and
further suggest that GTH acts by enhancing
the production of 1 7a-hydroxyprogesterone
by testicular somatic cells.

EN 71

REGULATION SYSTEM OF AROMATASE ACTIVATION
IN MEDAKA OVARIAN FOLLICLES.

A. Matsuhisa1 ' 2 , T. Iwamatsu ,

Y. Nagahama2, and N. Sakai2. ^Aichi Univ.
of Education, 2Natl. Inst. for Basic
Biol . , Okazaki . _ _

We investigated the aromatase activation

system that is regulated by gonadotropin
in the medaka ( Oryzias latipes ) .

Ovarian follicles, isolated 32 hr before
spawning, were incubated with pregnant
mare serum gonadotropin ( PMS , 1 OOIU/ml ) ,
forskolin (FK,10]jM) or dibutyryl cyclic
AMP (dbcAMP,5mM) in the presence or
absence of exogenous testosterone
(lOOng/ml). PMS, FK, and dbcAMP alone each
induced significantly increased estradiol-
1 78 production, and they enhanced
conversion of exogenous testosterone to
es tradiol - 1 7 8 . Cyanoketone (lug/ml), a
potent inhibitor of 3 8 -hydroxy steroid
dehydrogenase, completely suppressed PMS-
and FK-induced estradiol-178 production,
but did not influence their enhancing
effect on testosterone conversion.
Actinomysin-D (10ug/ml) and cycloheximide
(10ug/ml), inhibitors of mRNA and protein
synthesis respectively were able to
suppress this enhancing effect.

These results indicate that gonadotropin
induces aromatase activity in vitellogenic
follicles of the medaka, and suggest that
this action is regulated by the adenylate
cyclase-cyclic AMP system, mRNA synthesis
and new protein synthesis.

EN 72

GONADOTROPIN-INDUCED SPERMIATION IN VITRO
IN RANA NIGROMACULATA.

Tohru Kobayashi, Akiko Oshimi and Hisaaki
Iwasawa .

Biol. Inst., Niigata Univ., Niigata.

To clarify the mechanism of spermia¬
tion, an in_ vitro system was developed
with testes of Rana nigromaculata .
Small pieces of the testes were incubated
in modified Hank's-Hepes or Medium 199.
The induction of release of spermatozoa
from the Sertoli cells due to incubation
was not recognized in these media. But
the addition of hCG or glycoprotein (GP)
isolated from bullfrog pituitary to the
medium induced the release of sper¬
matozoa. This reaction was caused even
in the incubation, of hCG: 10 IU/ml or GP:
5 jug/ml for 1 . 5hr at 18°C. Furthermore,
effects of temperature at 18 and 25 °C
were compared. Dose-response charac¬
teristics of these reaction depended upon
incubation time and temperature regard¬
less of the kind of medium used. These
results suggest that in the presence of
gonadotropin (GTH), the release of sper¬
matozoa from the Sertoli cells is induced
by incubation at least for ,1.5 hr at 18
or 25 °C. The action of GTH on spermia¬
tion was examined by means of this sys¬
tem. Gonadotropin-induced spermiation
was inhibited by the addition of
cyanoketone as a specific inhibitor of £$
-3g-HSD. GTH-induced spermiation is in¬
volved in steroidogenesis.

EN 73

POLYOVULAR FOLLICLES IN MOUSE OVARY
EXPOSED NEONATALLY TO DIETHYLSTILBESTROL
IN VIVO AND IN VITRO.

Y.Fukazawa, T.Iguchi, and N.Takasugi
Dept. Biol., Yokohama City Univ., Yokohama
Neonatal treatment of C57BL/Tw female

mice with 1 pg DES for 5 days (DES mice)
induced a signif icantly higher incidence
of polyovular follicles (PF) in the ovary
of mice at 35 days than that in the
controls. Ovaries of prepubertal control
and 1)ES mice showed an enhancement of
ovulation and luteinization following a
combined treatment with hCG and PMSG. PF
incidence in control mice was not changed;
however, in DES mice, the incidence was
reduced by injections of hCG and PMSG. A
high incidence of PF was also found in
newborn ICR mouse ovaries transplanted for
30 days into intact hosts or
ovariectomized (OVX) hosts given DES
injections, but not in the OVX controls.
When neonatal ovaries were cultivated in
medium contaning DES and then transplanted
into OVX hosts, PF were formed in the -
ovarian grafts, but not in the DES-
unexposed grafts. The present studies
suggest that neonatal ovaries exposed to
estrogen in vivo and in vitro, which
produce PF in prepubertal hosts, are
capable of responding to hCG plus PMSG
given later, resulting in a reduction of
PF incidence, and that endogenous estrogen
in prepubertal mice can also act on
neonatal ovaries to induce PF.

Endocrinology

1315

EN 74

REPRODUCTIVE TRACT ABNORMALITIES IN
OVARIECTOMI ZED ADULT MICE EXPOSED
PRENATALLY TO DIETHYLSTILBESTROL (DES).

S. Ozawa, T.Iguchi and N.Takasugi. Dept.
Biol., Yokohama City Univ., Yokohama. _

Pregnant ICR mice were given 4 daily sc
injections of 2-2000 ug DES from Day 15 of
gestation. The prenatally DES exposed mice
(DES mice) were ovariectomized at 30 days
and killed at 120 days of age. In
addition, offspring of mice given 2000 ng
DES were killed at 3 days of age to
examine the distribution of the nodules
composed of enlarged polygonal cells. The
nodules were encountered under the
columnar epithelium of the Mullerian
vagina without spatial continuity to the
original cuboidal epithelium of sinus
vagina. Ovary-independent epithelial
stratification of the vagina was induced
in all DES mice. Thus, the appearance of
nodules found in the vaginal epithelium of
3-day-old DES mice seems to be a predromic
phenomenon of the later ovary-independent
vaginal stratification. The incidence of
ovary-independent vaginal cornif ication
was significantly higher in 20-2000 yg DES
mice than in the controls. Epithelial
downgrowths and/or pegs were observed in
vaginae of 20-2000 pg DES mice. Wolffian
remnants (77-100%) and hypospadia (100%)
were encountered in DES mice.
Ovary-independent stratif icatin of uterine
epithelium (8-42%) and disorganization of
the circular musculature of the uterus
(38-100%) were also induced in DES mice.

EN 75

5ct-DIHYDROTESTOSTERONE IS PREREQUISITE TO
DEVELOPMENT OF UROGENITAL SINUS
DERIVATIVES IN MICE.

Y. Uesugi* T. Iguchi, N. Takasugi and
V. Petrow . Dept. Bigl . , Yokohama City
Univ., Yokohama, and Dept. Physiol., Duke
Univ. Med. Center, North Carolina _

Development of testis, prostate,
coagulating gland, bulbourethral gland and
seminal vesicle was morphometrically
examined in male ICR mice given 7 daily sc
injections of 5a-reductase inhibitor, 400
mg/kg BW of 6-methyleneprogesterone ( 6MP ) ,
starting on day 12 of gestation. Female
mice were also exposed prenatally to 6MP .
On day 19 of gestation, fetuses were
removed by caesarian section for
examination .

In 6MP exposed male mice, anogenital
distance was significantly shorter than
the controls. Feminization of the nipple
and translocation of the urethra to the
base of the phallus were induced by
treatment with 6MP in utero. Development
of prostate, coagulating gland and
bulbourethral gland was significantly
suppressed by the inhibitor but not of
seminal vesicle and testis.

These results suggest that 5a-
dihydrotestosterone is indispensable for
normal development of urogenital sinus
derivatives (prostate, coagulating gland,
bulbourethral gland) and phallic urethra,
and for the regression of nipple in male
mice .

EN 76

NEONATAL EXPOSURE TO TAMOXIFEN INDUCES
ABNORMAL DEVELOPMENT OF OS PENIS IN MALE
MICE.

S.Irisawa*, T.Iguchi and N.Takasugi. *Lab.
Biol., Tokyo Kaseigakuin Univ., Machida,
Tokyo, and Dept. Biol., Yokohama City
Univ. , Yokohama _

In neonatal C57BL male mice, the

proximal segment (P) of os penis was
composed of condensed cell mass; however.

P developed into the membrane bone (MB)
with trabecula (T) and hyaline cartilage
(HC) at 5 days of age. The distal segment
(D) of os penis which had been composed of
mesenchymatous tissue until 10 days
developed into f ibrocartilage (FC) at 20
days. By contrast, in mice given neonatal
injection of 100 ng tamoxifen (Tx) for 5
days, HC and T of P disappeared at 10 and
30 days, respectively. The maximal area of
both P and D gradually increased with age
in control mice, whereas P area decreased
with age in Tx mice. In 60-day-old mice
exposed neonatally to clomiphene and
nafoxidine, P was composed of HC; however,
D lacked FC . P failed to from T in all
60-day-old mice given Tx starting within 3
days. HC of P and FC of D disappeared in
all mice given Tx starting within 5 days.
Neonatal castration did not suppress the
formation of T and FC in os penis, though
the size was smaller. These results
suggest that Tx has an inhibiting effect
on the development of mouse os penis,
resulting in suppression of formation of
both D and T of P .

EN 77

SEASONAL CHANGES OF PITUITARY-THYROIDAL
FUNCTIONS IN THE SQUIRREL MONKEY
T.Yoshida} S.Hattori^ M.Sato^and K.Ohotoh2*
1Tsukuba Primate Center, N.I.H., Tsukuba,
2Amami Branch Lab., Inst. of Med.Sci., Univ.
of Tokyo, Oshima-gun, 3Dept . of Biol., Nihon
Univ., School of Dentistry, Tokyo, and
^Corporation for Production and Research of
Laboratory Primates, Tsukuba. _

The squirrel monkey ( Saimivi soiurea)

has a well-defined annual breeding season
during which adult males undergo morpho¬
logical and behavioral changes. Conceptions
occur from December to March under captive
conditions in Japan. We investigated sea¬
sonal changes in pituitary-thyroidal func¬
tion in the male squirrel monkey. TRH (40
jig/Kg body weight) was injected intrave¬
nously under anesthesia with ketamine-HCl .
Blood samples were taken at 0, 20, 40 and
60 minute intervals after TRH-injection .
This procedure were repeated five times
from April 1987 to April 1988. Serum con¬
centrations of TSH and thyroid hormones
(T3 ,Ti, ) were determined by an IRMA and
RIAs , respectively. Similar results were
obtained among the experiments. Although
the significant increase in serum TSH con¬
centrations was detected, no significant
changes in serum T3 and Ti» concentrations
were observed after TRH-injection. However,
serum basal values of T3 and Tij were lower
in July than those in January. These re¬
sults suggest that the seasonality of the
squirrel monkey is mediated by the hypo-
tharamic pituitary-thyroidal function.

1316

Endocrinology

EN 78

ANNUAL REPRODUCTIVE CYCLE IN FEMALE
SALAMANDERS OF HYNOBIUS NIGRESCENS.

Masato Hasumi , Kazuya Nagai , Hisaaki
Iwasawa and Yoshitaka Nagahama* . Biol.
Inst., Niigata Univ. , Niigata and *Lab.
Reprod. Biol., Natl. Inst, for Basic Biol.,
Okazaki .

Adult female salamanders collected from
a breeding pond in March were raised in
terraria located outdoors throughout the
year. Just before submergence into water
all the trapped females were in the pre¬
ovulatory condition. Ovarian eggs were
translucent from the time of ovulation
until July, and the eggs assumed a mint
green color in August when the accumulation
of yolk granules began. Melanin began to
be deposited in the ovarian eggs in Septem¬
ber, and the entire egg was dark brown in
October and the succeeding winter months.
The sizes of the ovarian eggs and cloacal
glands increased markedly from July to No¬
vember, and further increased gradually
during hibernation. Ovary weight began to
increase in August, and reached the same as
that of females just before submergence
into water in January. The maximum mean
weight of the fat body was recorded in Au¬
gust. The weight of the oviduct plus ovi¬
sac increased from August toward the breed¬
ing season. The peak in the circulating
estradiol-17B levels was seen in October;
in the circulating progesterone levels, in
females just after ovulation.

EN 79

Response of the Mauthner cell in hypophy-
sectomized and castrated newts to androgen
and prolactin.

Y. Suzuki and S. Kikuyama*

Lab. Biol., Dept. Lib. Arts, Asia Univ.,
Musashino-shi , Tokyo and *Dept. Biol., Sch.
Edu. , Waseda Unib., Shinjiku-ku, Tokyo.

We have previously reported that the
Mauthner (M) cell of newts shows sexual
dimorphism during breeding season. During
this period, the nuclear volume of the M
cell is much greater than in the female.

In the present study, we attempted to'
examine whether the treatment of hypophy-
sectomized and castrated animals with T
and/or PRL could induce the enlargement of
the nuclear volume of the M cell. In the
hypophysectomized and castrated male newts,
PRL did not affect the nuclear volume of
the M cell. In the male newts bearing T
pellet, significant enlargement of the
nuclear volume of the M cell was observed.
Further increase in the M-cell volume was
recognized in the T pellet-bearing animals
treated with PRL. On the other hand, in
the hypophysectomized and ovaryectomized
female newts, implantation of T pellet and/
or treatment with PRL did not induce any
changes in the nuclear volume of the M cell.

EN 80

THE RELATIONSHIP BETWEEN AMPLEXUS BEHAVIOR
AND PLASMA GONADOTROPIN LEVELS IN MALE
TOADS, RUED J A PON TOIIS .

M. Itoh1 , S. Ishii1 and Y. Oshima2 ^Dept.
of Biol., Sch. of Edu., Waseda Univ., Tokyo
and ^Dept . of Basic Human Sci., Sch. of
Human Sci., Waseda Univ., Tokorozawa.

A gonadotropin surge takes place not
only in female toads but also in male
toads, when they reach the pond for breed¬
ing in the spring. Plasma LH and FSH levels
of males in amplexus with females in the
field are significantly higher than those
of solitary males. Sexually active male
toads were divided into two groups: 6 and 8
males caged with females in Exp.1 and 2,
respectively, (Group A) and 5 and. 7 males
alone, respectively, (Group B). In Exp.1
started on March 8, the mean plasma LH and
FSH levels with ± S.E.M. of Group A in¬
creased from 0.40 ± 0.09 to 6.92 ± 2.20
ng/ml and from 2.33 ± 0.36 to 5.05 ± 1.13
ng/ml, respectively, over 33 hours during
amplexus. In Exp. 2 started on March 1 8 and
22, the LH and FSH levels changed from 1.94
± 0.87 to 16.83 ± 2.42 ng/ml, and from 3.01
± 0.40 to 5.61± 0.83 ng/ml, respectively,
over 12 hours during amplexus. The levels
declined thereafter. There was no sig¬
nificant increase of LH and FSH in Group B
in either experiment. These results suggest
that amplexus with the female results in
the male gonadotropin surge which
presumably induces spermiation.

EN 81

PHYSICAL AND CHEMICAL PROPERTIES OF THE
MALE ACCESSORY GLAND SUBSTANCE MODIFYING
VIRGIN FEMALE CIRCADIAN PATTERN
M. Yoshii1 ,Y. Yamamoto1 , K.Tomioka1 , Y. Chiba1 ,

S . Takahashi2 . lEnviron . Biol . Lab . , Biol . Inst . ,
Fac .of Sci. , Yamaguchi Univ ., Yamaguchi, 2Biol .
Inst.,Fac.of Lib . -Arts .Yamaguchi Univ.,
Yamaguchi. -

The virgin female mosquito ( Culex pipi-
ens pal lens ) shows a diphasic circadian
activity pattern, peaking at light-off and
-on. This activity pattern is modified, by
male accessory gland substance received on
copulation, so that the light-off peak is
followed by an intense activity continuing
nearly entire night. This modification is
directly induced by injecting the male ac¬
cessory gland homogenate (MAGH) into abdo¬
men of virgin female. In the present study,
we injected the MAGH, treated variously,
into virgin female mosquito to investigate
the physical and chemical properties of the
rhythm modifying substance and obtained the
following results. (1) The rhythm modifying
substance was extracted from MAG with 0.9%
NaCl , but not with 80% acetone or 80% etha¬
nol. (2) It was inactivated with trypsin
-hydrolysis or boiling at 100°C for 5 min.
(3) The substance was nondialyzable . (4)
When ultrafiltration with UF3LGC00 was ap¬
plied, the activity was found only in the
retentate. These results suggest that the
rhythm modifying substance may be a peptide
with a molecular weight more than 10000.

Endocrinology

1317

EN 84

EN 82

CROWING BEHAVIOR IS MODIFIED BY VISUAL
CUES.

J. Igarashi and M. Wada. College of
Agr.& Vet. Medicine., Nihon Univ. ,
Fujisawa, Kanagawa. and Dept. Gen.
Educat. , Tokyo Med. Dent. Univ., Ichikawa,
Chiba _

Crowing behavior in Japanese quail is
dependent on testosterone, and has a clear
circadian rhythm due to a circadian clock
mechanism. However when the other factors
are present, for example social cues,
crowing behavior may change. We studied
modification of crowing by social interac¬
tions, using an automated recording system
of crowing through days. Sexually imma¬
ture birds were purchased and kept in an
individual recording cage. After normal
sexual maturation by photostimulation, the
birds were castrated and Silastic tubes
filled with testosterone were sub¬
cutaneously implanted, and experiments
started. When mirror was placed by the
cage, crowing was suppressed at during
the light period. Presence of a male and
a female, respectively, in the next cage
separated by a clear screen also
suppressed crowing at daytime. However,
when the cage were separated by a blind
instead of the clear screen, a pattern of
crowing rhythm was similar to a typical
one even though the total number of crows
decreased. These observations indicate
that crowing behavior in male is not a
threating behavior, but it serves as to
keep pairing and to maintain a territory.

EN 83

SEASONAL CHANGES IN CIRCULATING LH OF
JAPANESE QUAIL KEPT IN OUTDOOR CAGES.
H.Tsuyoshi and M. Wada

Dept. Biol., Sch. Educat., Waseda Univ.,
Tokyo and Dept. Gen. Educat., Tokyo Med.
Dent. Univ., Ichikawa, Chiba _

It is widely accepted that an annual
breeding cycle of birds in the temperate
zone depends on changes in photoperiod.
In Japanese quail, however, the
photoperiod is not omnipotent, especially
in LH decrease; a photoper iodic change
from long days to short days do not show a
steady testicular regression in a
laboratory condition. To clarify a
mechanism of photoperiodic regulation of
in circulating LH release, seasonal
changes of LH and thyroxine and in the
size of cloacal protrusion were measured
in the birds kept in outdoor cages. Daily
ambient temperature was also recorded.
Circulating LH began to increase in late
February when the daylength exceeded 12hr
and temperature fluctuated at the lowest
ranges. Higher LH and the developed cloa¬
cal protrusion were maintained from April
to August. In early September LH began to
decrease gradually, when the daylength was
still 13 to 14hr but temperature began to
decrease. These results indicate that in¬
crease in Circulating LH is induced by the
daylength but its decrease is induced by
temperature decrease together with the
daylength. Temperature may also modify
the induction of LH release in early
spring.

EFFECT OF PHENOBARBITAL ON LH HYPERSECRE¬
TION AFTER REMOVAL OF TESTOSTERONE NEGA¬
TIVE FEEDBACK IN JAPANESE QUAIL
T. Tsunenari and M. Wada

Dept, of Biol., Sch. of Educat., Waseda
Univ. , Tokyo. and Dept. Gen. Educat. ,
Tokyo Med. Dent. Univ., Ichikawa, Chiba.

It is generally considered that LH
secretion in avian species are regulated
by the catecholaminergic system in posi¬
tive fashion through the LHRH neurosecre¬
tion. In a negative feed back inhibition
of LH release, testosterone is postulated
to act via the catecholaminergic system.

In this study, phenobarbital , a blocker of
catecholaminergic neurons, was employed
to reveal an involvement of the
catecholaminergic system in testosterone
negative feedback on LH secretion.
Sexually matured male quail were castrated
and implanted testosterone with Silastic
tubes. Removal of Silastic implants in¬
duced a rapid increase of plasma LH
levels, and phenobarbital injection failed
to suppress this increase even though a
transient delay was observed. In intact
birds, phenobarbital injection suppressed
plasma LH levels for long term. These
results indicate that there is at least
another pathway regulating LH secretion
positively and sensitive to circulating
testosterone other than the positive
catecholaminergic system.

EN 85

ANDROGENIC REGULATION OF SYNAPTIC INPUT TO
SPINAL MOTONEURONS IN MALE RATS.

A. Matsumoto1, A.P. Arnold*and P.E.
Micevych* ‘Dept, of Anat . , Juntendo Univ.
Sch. Med., Tokyo, aBrain Res. Inst., Univ.
California at Los Angeles, Los Angeles.

Adult male rats ( Sprague-Dawley ) were
castrated and implanted with Silastic
tubes containing testosterone (T) or
nothing. Four weeks after castration,
horseradish peroxidase (HRP) was injected
into the'’ bulbocavernosus muscles and
animals were killed 2 days later. The
somatic and proximal dendritic membranes
of HRP-labeled motoneurons in the spinal
nucleus of the bulbocavernosus (SNB) were
examined ultrastructurally . Neuronal
structures apposing the membranes of 150
HRP-labeled SNB cells were analyzed by
measuring the percentage of somatic and
proximal dendritic membranes covered by
synaptic, synaptoid and neuron-neuron
contacts. Most of the neuronal contacts
consisted of synaptic contacts. The
percentage of somatic and proximal den¬
dritic membranes covered by synapses 4
weeks after castration was reduced to 30%
of those in controls. However, treatment
of T for 4 weeks prevented this decline.
Castration and T treatment also influenced
the size and length of synapses. These
results indicate that androgen is critical
for maintaining the organization of syn¬
aptic inputs to SNB motoneurons in adult
male rats.

1318

Endocrinology

EN 86

NEURAL MECHANISMS OF SEXUAL BEHAVIOR IN
MALE RATS: INTERACTION OF PREOPTIC AREA AND
AMYGDALA .

Y. Kondo , M. Nishizuka, Y. Arai , and K.
Yamanouchi*. Dept, of Anat . , Juntendo Univ.
Sch. Med., and Dept. Human Sci . , Fac. Human
Sci., Waseda Univ. , Tokyo.

There is a considerable body of evi¬

dence that the medial preoptic area (MPOA)
and the amygdala (AMG) have a facilitatory
role in masculine sexual behavior of rats.
In this study, we examined the relation
between these two limbic structures in
regulating sexual behavior. Male rats were
orchidectomized and simultaneously received
any of 6 types of the lesions as follow :
contralateral lesions in the MPOA and the
AMG, ipsilateral lesions in the MPOA and
the AMG, unilateral lesion in the MPOA or
the AMG, bilateral lesions in the MPOA or
the AMG. Three weeks after lesioning, all
animals were received the implantation of
Silastic tubings containing testosterone
(T) . Behavioral observations of masculine
sexual behavior were carried out 4 times
every 5 days after the implantation of T.
The contralateral lesions in the two struc¬
tures significantly suppressed masculine
sexual behavior, while the ipsilateral
lesions did not have a suppressive effect
on copulatory behavior. However, the copu-
latory activity of males with the unilat¬
eral lesions in the MPOA or the AMG was
also suppressed as the male with the bilat¬
eral lesions. Further studies are required
to elucidate this contradictory phenomenon.

EN 87

THE ROLE OF THE MEDULLARY RAPHE NUCLEUS IN
REGULATING FEMALE SEXUAL BEHAVIORS IN
FEMALE RATS
K. Yamanouchi

Neuroendocrinology, Dept, of Basic Human
Sciences, Sch. of Human Sciences, Waseda
University, Tokorozawa

The role of the medullary raphe
nucleus, especially raphe obsculus, in
regulating female sexual behaviors was
examined in female rats. In 9 females,
radiofrequency lesions in the midline of
the caudal medulla were made (MRL) . Six
females received a sham operation (Sham)
and 10 females received no brain surgery
(Control). All females were ovariectomized .
Four to 5 weeks after the surgery, all
animals were implanted with a 3 cm
Silastic tubing (1.57 X 3.18) containing
estradiol. Behavioral tests were started
2 days after implantation of estrogen and
carried out daily for 5 days. Increases
of the mean lordosis quotients (LQ) were
observed through the tests in all groups.

In the first and second tests, however,
the mean LQ in the MRL group was higher
than those in the Sham and Control groups.

In addition, the median day of onset of
ear-wiggling (one of the soliciting
behaviors) in the MRL group was advanced
significantly, when compared with those in
other groups. These results suggest that
the medullary raphe nucleus may play an
inhibitory role in regulating female
sexual behaviors in female rats.

EN 88

EFFECTS OF PRENATAL STRESS AND PCS '.''WEANING
ISOLATION ON ADRENOCORTICAL FUNCTION IN
MALE AND FEMALE MICE.

T. Noumura and M. Kaneko. Dept, of Regulat.
Biol., Fac. of Sci., Saitama Univ., Urawa.

The effects of prenatal stress and post-
weaning housing conditions on adrenocorti¬
cal function were studied in growing male
and female mice.

ICR pregnant females were stressed by
being placed with groups of unfamiliar,
daily-renewed males from Day 13 through
Day 18 of gestation. The offspring of both
sexes from stressed and unstressed mothers
were individually-housed or maintained in
groups of five from weaning at 20 days of
age. Mice were killed every 10 days from
weaning to 200 days to assay plasma levels
of corticosterone.

Female mice had heavier adrenals and
higher levels of plasma corticosterone
than males had. Prenatal stress resulted
in a decrease of adrenal weight but no
significant difference in plasma level of
corticosterone in both sexes. Post-weaning
isolation induced higher level of plasm
corticosterone in mice of both sexes. Male
mice were more affected by both conditions.
The results indicated that prenatal stress
and post-weaning isolation seemed to
influence synergistically on adrenocortical
function in mice.

EN 89

NEONATAL ANDROGEN EXPOSURE AND CELL DEATH
IN EARLY DEVELOPMENT OF THE ANTERQVENTRAL
PERIVENTRICULAR NUCLEUS (AVPv-POA) IN THE
FEMALE RAT.

S. Murakami and Y. Arai. Dept of Anat.,
Juntendo Univ. Sch. of Med., Tokyo.

The anteroventral periventricular
nucleus (AVPv-POA-) is an intensely stained
neuron group in the anteroventral
periventricular gray of the preoptic area.
Its volume is significantly greater in
females than in males. Neonatal treatment
of female rats with testosterone
propionate (TP) for 5 days from the day of
birth induced a significant reductiion of
the volume in the AVPv-POA to male level.
In order to examine the action of androgen
on the neural substrate of the AVPv-POA to
decrease Its volume, the rate of degenera¬
ting cells to normal cells in the AVPv-POA
during postnatal period was examined in
females and TP-treated females. Pycnotic
cells were seen in the AVPv-POA during the
first 13 days of life, peaking on day .3.
There was no significant difference in the
pycnotic rates between females and TP-
females at days 3 and 5. At postnatal days
7 and 10, however, the pycnotic rate in
TP-females was significantly higher than
that of normal females. These results may
suggest that neonatal androgen decreases
neuron number of the AVPv-POA by
regulating neuronal death, and this may
account for the reduction of the volume of
the AVPv-POA following neonatal treatment
with TP.

Endocrinology, Morphology

1319

EN 90

DISTRIBUTION PATTERN OF NEURONS AND
GLIAL CELLS IN THE ANTEROVENTRAL PERI¬
VENTRICULAR NUCLEUS (AVPv) IN RATS.

M. Miyakawa and Y. Arai . Dept. Anat . ,
Juntendo Univ. Sch. Med., Tokyo.

Sexual dimorphism in nuclear volume
has been shown in the AVPv of the preoptic
area. In this study, the distribution of
neurons was examined on Epon-embedded lum
thick sections in males(M) , females(F) and
females treated with testosterone propio¬
nate for the first 5 days of postnatal
life (50 ug/day ) (TP-F) . The mean density
of neurons in 5 sections of 40 urn interval
/ each brain was F>M>TP-F (p< 0.05). The
distribution pattern of neurons in F dif¬
fered from that in M and TP-F, showing the
presence of more grids of higher density.

Glial cells in the AVPv was examined
by immunocytochemical staining with mono¬
clonal antibody of GFAP and vimentin
(Boehringer-Mannheim) in adult and neo¬
natal rats. In adult AVPv, GFAP-posi tive
astrocytes were evenly distributed and
vimentin positive long processes of tany-
cytes was found to extend horizontally in
the AVPv from ependyma to the anterior
direction. In the neonates, GFAP-positive
astrocytes appeared in dorsolateral peri¬
phery of AVPv around day 4. The pycnotic
cells enclosed with GFAP-positive sub¬
stance were found occasionally. Vimentin-
positive long processes in close vicinity
to neurons were observed to run to the
ventrolateral direction from ependyma.

EN 91

BR0M0DE0XYURIDINE (BrdU) -ANTI BRDU ANALYSIS
OF NEUROGENESIS IN SEXUALLY DIMORPHIC
ANTEROVENTRAL PERIVENTRICULAR NUCLEUS
(AVPv) OF THE PREOPTIC AREA.

M. Nishizuka, H. Sumida* , Y. Kano* and Y.
Arai. Dept. Anat., Juntendo Univ. Sch. Med. ,
Tokyo and *Lab. Anim. Physiol., Fac. Agri.,
Meiji Univ., Kawasaki. _ _

To elucidate possible effects of androgen

in the process of neurogenesis in AVPv,
fetal rat brains were exposed to testoste¬
rone propionate (TP) and BrdU, a thymidine
analogue , to label final cell division. Preg¬
nant Wistar rats received daily s.c. injec¬
tions of 2 mg TP from day 14-16 or 14-18 of
gestation. The rest of pregnant rats were
given oil as for controls . All rats were i.p.
injected with 50 mg BrdU on day 15 and killed
on day 17 or 21 of gestation. Female fetal
brains were immunostained using anti BrdU.

A number of BrdU immunoreactive cells were
obtained in presumptive AVPv area of fetal
brains. The number of these cells in TP-
treated females seemed to be greater than
that of control females on day 17. The
number on day 21, however, became a similar
level to controls. Total neuron population
on day 17 was not significantly different
between two groups. On day 21, in contrast,
AVPv area of TP-treated rats consisted of
statistically smaller number of neurons com¬
pared to controls. These results indicate
that androgen modifies neurogenesis, migra¬
tion and/or survival of AVPv neurOns. Such
prenatal effects of androgen may probably
underlie formation of sexual dimorphisms
which appear in AVPv during postnatal period .

MO 1

ULTRASTRUCTURAL CHANGES OF THE MYONEME
BETWEEN THE CONTRACTED AND ELONGATED
STATES OF THE CILIATE, SPIROSTOMUM.
H.Ishida, A.Matsuno and M.Kuroda
Department of Biology, Faculty of Science,
Shimane University, Matsue.

A large heterotrichous ciliate,
Spirostomum ambiquum shows the
characteristic twisting contraction when
stimulated by mechanical or chemical
factors. This contraction is attributed to
a special contractile fibrillar system
which is termed myoneme. The myoneme
features by being surrounded by various
sizes of vacuoles, and located at the
transitional plate between ectoplasm and
endoplasm . Every myoneme is composed of a
great number of fine filaments which are
heavily packed and lie in parallel with
each other inside the myoneme. The
diameter of individual filament varies
from 3 to 5 nm at elongated state of
organism. But it seems to rise up about 10
nm at contracted state. Observation in
stereo pair, myonemal filaments has many
branches and makes a mesh-like network.

MO 2

MORPHOGENESIS OF VENTRAL CILIATURES
DURING BINARY FISSION OF A HYPOTRICH
CILIATE, HISTRICULUS CAVICOLA.

T. Nakamura, O. Noguchi, T. Matsusaka.
Dept, of Biol., Fac. of Sci., Kumamoto
Univ., Kumamoto.

Ciliary renewal during binary
fission of H. cavicola was first detected
by the formation of new kinetosomes at 2
positions, anterior-left margins of
ventral cirri 1 (VI) and transverse
cirri 5 (T5) . New kinetosomes appeared

at close proximity to the outside wall of
its rampart in VI and to the inside and
outside-left wall in T5 . They were first
formed in parallel to the cell membrane,
turned perpendicular to it and then
ciliogenesis began. The kinetosomes
proliferated in this area to form oral
primordium (OP). 5 cirral streaks,

future cirri, were formed from
disaggregated frontal cirri 6, 7, and 8
in a proter and from disaggregated VI,

V2 , and anterior-most part of OP in an
opisthe. The disaggregation of cirri
occurred by disappearance of their
networks and ramparts. Ciliary
proliferation occurred in the cirral
streaks. The cirral streaks segragated
into 3 or 4 segments to form new cirri.
Neodesmal microtubules were found on all
posterior-most segments of cirral
streaks .

1320

Morphology

MO 3

Physiological Analysis of Macronuclear Inclusion
Body in Paramecium caudatum.

Yuuji Tsukh, Laboratory of Biology, Hosei University, Chiyo-
da-ku, Tokyo 102

In some strains of P. caudatum, macronucleus contains inclu¬
sion body (designated AFG body), which is specifically stained
with fast green FCF at low pH (45% acetic acid). AFG body
within a macronucleus is variable both in numbers and in sizes,
and morphologically distinct from bacterial endosymbionts. Nu¬
clear transplantation experiments showed that the AFG body mul¬
tiplies independently from 'host' macronuclear genome and
moves among macronuclei via cytoplasm during both vegetative
and sexual reproduction, suggesting that the AFG body is a non-
bacterial macronuclear endosymbiont. SDS-PAGE analysis
showed that AFG body consists of 4 or 5 kinds of proteins.

In this report, changes of numbers and sizes of the AFG body
at various physiological conditions were examined. Cells in cul¬
ture medium have mostly one or two large AFG body and some
smaller ones within their macronuclei. However, when the cells
were transferred to a solution of 1 mM CaCh and Tris-HCl (pH
7.0), number of cells with the large AFG bodies gradually de¬
creased and, alternatively, that with numerous small AFG bodies
increased. AFG body then became smaller and smaller, and at
last undetectable under microscope at about 1 hour after the trans¬
fer. This phenomenum occurred in cells at various cultural stages
and temperature, but was inhibited by the addition of NaCl (>2.5
mM). On the contrary, KC1 promoted the disappearance of AFG
body. Dinitrophenol had also an inhibitory effect at very low
concentration (<1 p.M). Other cations (Ba,La,Li,Sr) as well as
other chemicals (antibiotics [G418], Caffeine, Theophiline, iono-
phores, Colchicine, Colcemide, Vinblastine, ConA, Methylurea,
Pronase, Nonidet P-40) did not affect the disappearance of AFG
body. SDS-PAGE of the macronuclei in which AFG body had
disappeared, however, did not show the marked decrease of the
protein components of AFG body, suggesting that AFG body
only dispersed within a macronucleus.

MO 4

QUANTITATIVE ANALYSIS OF RESPONSE TO FATTY
ACID IN AMOEBA PROTEUS.

A. Kihara, K. Ishii. Lab. Biol., Hosei Univ., Tokyo. _

A. proteus is transforming its shape spontaneously by
means of the amoeboid movement. We observed that
global application of caprylic acid caused two typical
transformations depending on the concentration. One is
the flattening at about 10‘4 M, and another is shrinking at
over 5x1 O'4 M . We tried the quantitative assay for these
transformation.

The images of amoeba through microscope were re¬
corded in 35mm high contrast negative film and were
printed on paper. Then they were put in microcomputer
with image scanner. The bit image of the amoeba on the
screen was constituted by the outline black pixels corre¬
sponding to the outer edge of the ectoplasma and by the
inner clusters of black and white pixels corresponding to
the endplasmic granules. We developed the program to
trace the outline edge of the image and to count all pixels
in the inner region of amoeba. By means of these meth¬
ods, we can get the area of amoeba without influence of
inner granule density.

When amoeba changed its shape to flattening at 2x1 0‘4
M, the area increased about 10% greater than control
value. At the concentration beyond 10'3 M in which
amoeba became shrinking, the area become less than
half. So it is suggested that the flattening and the
shrinking are attended with respectively, increase and
decrease of the area. These methods of area calcula-
tionmay take advantage to examine the intermediate re¬
sponses of these shape alteration.

MO 5

AN EXPERIMENTAL STUDY ON THE AMINERGIC
NERVE (AMN) OF THE CEREBRAL BLOOD VESSELS
OF RODENTIA

T.Tagawa, K.Ogawa and Y.Enoraoto, Dept, of
Anat. , Sch. of Med. , Fukuoka Univ. , Fukuoka

After unilateral and bilateral superior
cervical ganglionectomy of rodent ia, an ob¬
servation was made of the varying aspects
of Aminergic Innervation in the cerebral
blood vessels to recognize Amn which is one
of the autonomic nerves controling the cer¬
ebral blood vessels. In the case of uni¬
lateral superior cervical ganglionectomy,
considerable differences in the amount of
Amn were seen on the resected sides both in
the rats and the golden hamsters which have
a slender posterior communicating artery
and a developed anterior communicating ar¬
tery and in the guinea pigs which have a
slender anterior communicating artery and
a deveroped posterior communicating artery,
while at the same time, slight variations
in Aminergic Innervation were found on the
opposite sides. In the case of bilateral
same ganglionectomy, however, Amn complete¬
ly disappeared in all of the cerebral blood
vessels with the exception of the vertebral
artery.

Furthermore, after bilateral superior
cervical ganglionectomy, no variation of
Amn was recognizable in the intra-cerebral
small arteries or capillary.

MO 6

AN ULTRASTRUCTURAL MORPHOMETRIC STUDY OF
THE MOLE CARDIOVASCULAR CHEMORECEPTORS

S . Kikuchi

Dept. of Biol., Sch. of Liberal Art & Sci.,
Iwate Medical University, Morioka.

The carotid and aortic bodies of the
normal adults of Japanese mole, Mogera ko-
beae, were examined by a transmission elec¬
tron microscope. The results of the ultra-
structural morphometric study demonstrate
marked similarity between the carotid and
aortic body. No significant difference in
the glomus cell size and shape was observ¬
ed. Two subtypes of glomus cells, based on
dense-cored vesicle size, are present in
both carotid and aortic body. The majority
is the smaller vesicle-containing cells.

The most characteristic feature of the
mole carotid and aortic bodies is an abun¬
dance of annulate lamellae. The morpho¬
metric analysis revealed that each glomus
cell contains at least one annulate lamel¬
la in the cytoplasm usually near the nu¬
cleus. A direct continuity between the
rough-surfaced endoplasmic reticulum and
the annulate lamellae was frequently ob¬
served. As these annulate lamellae were
fond in all animals of both sexes, the or¬
ganelle seems to be a primary component
of the mole glomus cell.

These findings suggest that the mole
aortic body has a function similar to the
carotid body.

Morphology

1321

MO 7

A HISTOLOGICAL STUDY ON THE NERVE INNERVA¬
TION IN THE URINARY BLADDER OF CLEMMYS JA-
PONICA .

H. Ishihara. Biolog. Lab., Fac . of Sci. &
Engineer., Aoyama Gakuin Univ. , Tokyo. _

The morphology of nerve innervation and
nerve endings in the urinary bladder of
Clemmys japonica was investigated accord¬
ing to the silver impregnation method de¬
vised by the present author. Urinary blad¬
der walls were constituted of four layers,
serosa, muscularis , lamina propria and
epithelium. In serosa, many single nerve
fibers coming from different nerve trunks
were found to form loose plexuses. Uni¬
polar, bipolar and multipolar nerve cells,
which seem to be sensory in nature, were
seen here and there in serosa. Sometimes,
the terminal reticulum was seen widely ex¬
panding outside muscle layer. The nerve
bundles emerging from serosa entered muscle
layer, together with capillaries or by
themselves. After repeated ramifications
and anastomoses , it was observed that these
nerve bundles became gradually thinner in
structure and form rich plexuses in muscu¬
laris. Main nerve bundles containing
large ganglia were observed. Fine nerve
plexuses constituted of neurofibrils were
observed in lamina propria. Generally
nerve bundles were covered with neuroplasm-
mass showing nuclei which scattered here
and there. Various nerve endings were
found in each individual layer. Very often
coil-like or waving nerve fiber and bundles
were seen here and there in each individual
layer.

MO 8

The mental and mylohyoid nerves distribute
to the mandibular sinus hair in crab-eating
monkeys (Macaca fasc i cu 1 a r i s) .

Y. Takahashi and K. Kimura. Dept, of Anat.
Natl. Def. Med. Coll., Tokorozawa.

The distribution of n. trigeminus to the
mandibular sinus hair was examined under
dissecting microscope in fifteen crab-eat¬
ing monkeys (Macaca f asc i cu lar i s) in both
sexes. The simple silver impregnation
(Kimura & Takahashi, 1985) was applied for
defining the nerve fiber and follicle of
sinus hair. The n. mentalis emerged through
the mental foramen. Specimens had one to
five foramens (on the average 2.0) on ei¬
ther side. The marginal mandibular branch
of the n. facialis crossed over the n. men¬
talis on the deep surface of m. depressor
labii inferioris and m. mentalis. Both

nerves did not exchange their nerve fibers
each other. The n. mentalis supplied almost
all the mandibular sinus hairs. The sinus
hair lined up paraboric rows according to
the ramification of n. mentalis. The n.
mylohyoideus supplied a few sinus hairs
succeeding to the lower border of former at
the median line. The cutaneous branch of
n. mylohyoideus passed through between the
right and left anterior bellies of m. di-
gastricus. It was observed in 22 cases (73.
3%) in both sides and in 8 (20.0%) cases
at one sides. Any fiber was not detected
in 2 ( 6 . ?%) cases .

MO 9

HEMATOXYLIN-STAINABLE PROTEINS: HOW THEY
INFLUENCE HEMATOXYLIN STAINING OF NUCLEI.
M. Hara, Y. Sumi and T. Suzuki.

St. Marianna University School of Medicine,
Kawasaki, Japan. _

Chemical properties of hematoxylin are
well known, however the relationships be¬
tween hematoxylin and respective nuclear
components in biological staining still
remain obscure. Using biochemical tech¬
niques, we have succeeded in detecting, at
least two unique proteins that were stained
by hematoxylin in rat liver nuclei. We in¬
vestigated how these proteins influenced
the hematoxylin staining of nuclei, in
relation to how DNA influenced that. DNA in
the purified nuclei was digested by DNasel.
After incubation with 2mg/ml DNasel,
nuclei were hardly stained by hematoxylin.
But with 20ug/ml DNasel, nuclei were
stained with a similar intensity to those
in controls, and the DNA content in nuclei
decreased to about 20% of that of the con¬
trol after 1 20min incubation. Digestion of
DNA was accompanied by release of nuclear
proteins, in which hematoxylin-stainable
proteins were found in the incubation
buffer. Moreover, the molecular weights of
one of the hematoxylin-stainable proteins
and the unique protein we detected were
found to be the same. These data indicate
that hematoxylin mainly stains nuclear
proteins, and decrease of nuclear staining
intensity by treatment with DNasel is due
to releasing of hematoxylin-stainable
proteins than disappearance of DNA.

MO 10

MORPHOLOGICAL CHANGES IN VASCULAR VESSELS
DURING POSTPARTUM ENDOMETRIAL REGENERATION
IN THE CHINESE HAMSTER [CRICETULUS GRISEUS)
H. Ninomiya1 , K. Yamazaki2 , S. Kondo2 and
R. Kubokawa2

1 Dept . of Lab. Anim. Sci., Azabu Univ.,
Sagamihara, technical Sect., Mitsubishi
Kasei Inst, of Life Sci., Machida, Tokyo.

Morphological characteristics of the
vasculature at various stages of postpartum
endometrial regeneration in Chinese hamster
were studied by light and electron micro¬
scopy. Arteriole endothelial cells became
swollen and subsequent cell fusion was
noted 0-2 days postpartum. Lymphocytes
and leukocytes were associated with these
changes in the endothelial cells. The
swollen endothelial cells appeared to
be transformed into giant cells which
plugged the arterioles 3-4 days postpartum.
The giant cells were from 80-150 ym in
diameter and were rich in tubular
endoplasmic reticulum and microvilli.
The plugged regions were replaced by
epithelioid cells and subsequent fibrosis
occurred 5-9 days. Also neovascularization
was evident about the fibrotic tissue
at this period. The origin of the giant
cells could not be determined exactly.
It was concluded that they as well as
epithelioid cells may possibly function
to block the arteioles in order to prevent
bleeding, so that endometrial regeneration
may occur in preparation for future
pregnancy.

1322

Morphology

MO 11

COLLAGENOLYTIC AREA IN MURINE UTERUS DURING
EARLY POSTPARTUM PERIOD.

K . Shimizu , K. Miyoshi and M.Hokano. Dept. of
Anat. Tokyo Med. Coll., Tokyo.

A conspicous features in postpartum
involuting uterus is collagenolysis , since
this represents one of the fastest known
rates in any connective tissue. In the
mouse, postpartum collagen degradation oc_
curred more in the endometrium than in
the myometrium. Although postpartum
collagen degradation is progressing on the
postpartum day 1 , histological observation
could not reveal a proportionate removal
of endometrial collagen ( Anat. Rec . , 1988).
An aim was to determine collagenolytic area
during early postpartum period. Animals
used were female mice of the IVCS strain.
They were mated at 8 weeks of age. The
animals were killed on the day of parturi¬
tion, or 1 , 2 , or 3 postpartum days. Several
portions of uterine horn were fixed with
buffered formaline, embedded in paraffin,
and stained with picro-sirius red. On the
day of parturition, sevral parts in the
endometrium showed edema in which collagen
birefringence was not observed. On the
postpartum day 1 when edema disappeared,
collagen fibers distributed wholly in the
endometrium. It is probable that collagen
fibers are degraded in the edematous area
immediately after the parturition and that
degraded collagen fiber is expelled with
edematous fluid.

MO 12

INDUCTION OF TRANSPLANTABLE HEPATOMA BY
TREATMENT WITH MAM-ACETATE IN AN INBRED
STRAIN OF THE TELEOST ORYZIAS LATIPES.

Y. Hyodo-Taguchil , T. Sasaki2 and K.
Yamagami2 . lNatl . Inst. Radiol. Sci., Chiba,
2Life Science Inst., Sophia Univ., Tokyo.

Adult fish of an inbred strain (H04C)
of Oryzias latipes were exposed for 24 hrs
to an aqueous solution of the carcinogen
methylazoxymethanol (MAM) acetate at con¬
centrations from 0.25-0.75 ppm. Liver
tumors were observable after 2-3 months.
Adult fish of H04C strain was found to be
more sensitive to MAM-acetate than fish of
orange-red variety. Eight fish bearing
liver tumors were used for the transplan¬
tation experiments 2 to 5.5 months after
treatment. By the technique of transplant¬
ation into the eye chamber and peritoneal
cavity of the syngeneic strain, serial
transplantation of the tumor was success¬
ful in the tumors from two of eight fish
examined. The result showed that MAM-
acetate induced-hepatic tumors survived
and grew in the anterior eye chamber and
peritoneal cavity of syngeneic hosts and
that the tumors were malignant.

The changes in isozyme pattern of
esterase were examined during the develop¬
ment of liver tumors. Polyacrylamid slab
gel electrophoretic analyses showed that
the activity of liver-specific isozymes
decreased and anodally migrating bands
were present in transplantable tumor
tissue. The pattern was similar to that
observed in muscle extract and embryo.

MO 13

ANATOMICAL STUDIES ON THE LUNG OF THE
POLYPTERUS SENEGALS.

A.Kimura, T.Gomi, T.Hashimoto and K.Kishi.
Dept, of Anatomy, Sch. of Med., Toho Univ.,
Tokyo .

The lungs were seen to be long, and
narrow sacs; the right lung was a mean
130mm, and the left 45mm in length. The
lungs branched out from the esophagus,
connected with the esophagus with a 2-2. 5mm
slit. Around the slit, ciliated cells
were seen. The pulmonary cavity ran
through the center along the long axis of
the lung, and many grooves were present in
the inner wall. The upheaval area
bordering the grooves were seen to be made
up of the capillary network; on the surface
of these areas could be seen well-developed
alveolar cells having Type I properties.
Thin cytoplasmic projections formed a
blood-air barrier (SD 0.429>im) as a field
for gas exchange. In the cytoplasm, many
pinocytotic vesicles and a few osmiophilic
lamellated bodies (OLBs) were observed.
Inside of the grooves, ciliated cells and
mucous cells were observed. Moreover, the
epithelial cells, including many OLBs
having Type II properties, - were observed.
These morphological characteristics suggest
that, contrary to popular opinion,
Polypterus lungs have respiratory function
in addition to their buoyancy function.

MO 14

GROSS ANATOMICAL AND LIGHT MICROSCOPIC
OBSERVATIONS ON THE TRACHEA AND LUNGS OF
HYNOBIUS NEBULOSUS T0KY0ENSIS.

Y.Kikuchi, T.Gomi, A.Kimura, T.Hashimoto
and K.Kishi. Dept, of Anatomy, Sch. of
Med., Toho Univ., Tokyo.

We observed- the trachea and lungs of
Hynobius nebulosus tokyoensis mainly by
gross anatomically and light microscopy,
and partly by electron microscopy (SEM,
TEM/. The trachea was seen to branch out
from ventral side of the esophagus in the
shape of a slit, and possessed a pair of
cartilage formations. Both the right and
left lungs were single sacs, and occupied
2/3 of the abdominal cavity. The lungs
were mainly divided into two parts, an
airway portion and a respiratory portion.
In the airway portion, smooth muscles were
well developed, and ciliated cells and
goblet cells were seen to exist. In the
respiratory portion, elastic fibers and
capillary networks were well developed, and
epithelial cells ranged from cuboidal to
columnar in shape. These cells possessed
the features of both Type I and Type II
alveolar cells. Granules of neutral and
acid mucopolysaccharides were seen to exist
in the apical cytoplasm of the alveolar
epithelial cells as well as in the goblet
cells. These granules had a moderate
electron density.

Morphology

1323

mo is

ELECTRON MICROSCOPIC OBSERVATIONS OF THE
ALVEOLAR BRUSH CELL OF THE RAT.

T.Gomi, A.Kimura, T.Hashimoto and K.Kishi.
Dept, of Anatomy, Sch. of Med., Toho Univ.,
Tokyo .

The alveolar brush cell of rats were
observed by transmission electron
microscope. They were characterized by
long, thick, peculiar microvilli protruding
from the apical surface of the cell. The
typical cell was ellipsoid in shape with a
constricted neck. On both sides of the
cell, there could be found a very thin
cytoplasmic processes typical of Type I
alveolar epithelial cells, which covered
the capillaries. The microvilli were
about 0.7-0. 9 pm in length and about 0.25-
0.27 pm in diameter. Inside these unique
microvilli, numerous fine filaments
extended from the top to the supranuclear
region of the cytoplasm. In addition to
these filaments, there were vesicles,
vacuoles, mitochondria, Golgi-apparatus ,
endoplasmic reticula, glycogen granules,
free ribosomes and microtubules in the
cytoplasm. The most characteristic
feature was the presence of alveolar brush
cells with long and wide microvilli
protruding from the lateral side of the
cell into the alveolar lumen, instead of
protruding from the top. Moreover, the
alveolar brush cell resembles the type I
taste cell of the taste buds and the fifth
cell type of the olfactory epithelium.

MO 16

ELECTRON MICROSCOPIC STUDIES OF EMBRYONIC
DEVELOPMENT OF THE FRESHWATER PLANARIAN ,
Bdellocephala brunnea . II. THE ROLE OF
ER-AREA.

T. Sakurai. Div. Cell Sci., Cent. Res.
Lab., Fukushima Med. Col., Fukushima.

In the early development of freshwater
planarians , a remarkable phenomenon is syn¬
cytium formation by yolk cells surrounding
the blastomeres. This is preceded by an
aggregation of yolk cells, called "group
formation". It has generally been assumed
that the egg cell participates in the group
formation. Our previous study reported
that a large number of ER-areas derived
from annulate lamellae were found in a fer¬
tilized egg of B_;_ brunnea . The present
study reports the ultrastructural changes
of ER-areas before and after egg laying.
From several hours before egg laying, ER of
tubular profiles in the areas begins to
form a large number of vacuoles of various
sizes by swelling or fusion with one an¬
other. Enlarged ER or vacuoles contain a
fibrillar material. Within a few hours
after egg laying, most of the vacuoles are
released outside the egg cell. In particu¬
lar, those of larger sizes extremely de¬
crease in number. At the same time, yolk
cells surrounding the egg cell comes into a
more close contact with one another than
before the egg laying. It seems likely
that the ER-area produces an inductive
factor which stimulates the aggregation of
yolk cells around the egg cell.

MO 17

ULTRASTRUCTURAL AND HISTOCHEMICAL STUDIES
OF NERVOUS SYSTEM OF OPHIOPLUTEUS .

Y. Nakajima. Dept, of Biology, Keio Univ.,
Yokohama . _ _ _

Localization of nerve cells in 6-armed
ophiopluteus of Amphipholis kochii was
studied by fluorescence and electron micro¬
scopy.

Indirect immunofluorescence microscopy
revealed the presence of 2 to 3 serotoner¬
gic nerve cells on the dorsal side of the
proximal portion of anterolateral arms.
Bundles of axons that emanate from these
neurons run along the ciliary band of the
arms. A pair of serotonergic cells were
also found in the basal part of postero¬
lateral arms. Catecholaminer gic nerves
located by glyoxylic acid fluorescence
method were observed in the lower lip and,
along the ciliary bands of the arms. Both
serotonergic and catecholaminergic neurons
are ectoneural in their origin, involving
three morphologically distinct types. Among
them, one with projections associated with
ciliary band and one with a cilium and fine
radial processes are probably catechol¬
aminergic, while another with coiled cilia
is serotonergic. The last type appears
to be common to the pluteus larvae of
echinoids and ophiuloids.

An effector organ, the circum-esophageal
muscle cells form a thin continous layer,
which is lined by basal lamina. The muscle
layer adjoins esophageal epithelium at the
lower lip, where bundles of axons with
synaptic vesicles are observed.

MO 18

STRUCTURES IN CEREBRAL VESICLE OF THE
AS C I D I AN LARVAE, STYELA PLICATA.

H. Ohtsuki, Biol. Ins., Fac. of Educ. ,

■Uni v, _ of Oita, Oita. _

Structures in cerebral vesicle of the
ascidian larvae. Stye 1 a p 1 i cat a, were
examined with a light microscope (LM) and
a scanning electron microscope (SEM). The
larvae were fixed with 2% g 1 u t ar a 1 dehy de
in 77% seawater for 20-24 hr. A portion
of the fixed larvae was used for serial
paraffine sections (2m m thick) by the
conventional method for LM. Another por¬
tion was dehydrated and ethano 1 - inf i 1 ter ed
materials were cryof ractured at liquid
nitrogen temperature. The larvae, cut
through the cerebral vesicle, were col¬
lected, rinsed with M/30 phosphate buffer
(pH 7.4), dehydrated again and dried by
the method of critical point technique,
with C02. The dried materials were coated
with Pt-Pd utilizing a Hitachi ion sputter
(E102) and examined with a Hitachi S-800
SEM. Ocellus of the larva was dgeneratve
as that of the other species of Styela and
it consisted of only a cell, which con¬
tained a small mass of pigment granules,
in the postero-dorsal wall of the cerebral
vesicle. Statocyte of the larva was hol¬
lowed out on the ventral surface of its
cell body, forming a pigment cup. On the
inner wall of the cerebral vesicle, a
cluster of about 10 small protuberances
(0.7-2. 5m m diameter), surrounded by many
fine processes (0.2m m), was found under
the cell body of the statocyte; the other
protuberances (2m m), projecting several
processes, were shown always in pairs at
the postero-dorsal part of the left wall.

1324

Morphology

MO 19

ON THE ABILITY OF THE HEAD- FORMATION IN A
COMMON BLACK-COLORED SPECIES OF BIPALIUM
FUSCATUM.

Y.Shirasawa and N.Makino, Dept, of Biol.,
Tokyo Med. Coll., Tokyo.

Morphological and histological studies
on the head-formation of the transected
pieces of Bipalium fuscatum have been made
in the various sizes and ages. This mate¬
rial reproducts sexually in May to June,
and the fission was never observed. Small
larvae were transected into five pieces,
and larger ones into seven. Small larvae
always regenerated successfully without
reference to the body level, and each piece
accomplished the head-formation in about
10-14 days after the operation. But larger
groups especially that had oviposit ioned ,
showed remarkable decrease in the ability
of the head-formation, and if they barely
regenerated, the heads collapsed soon. The
second-largest group, which are common
size, also regenerated as well as larvae.
The largest group took much more days than
larvae and revealed that the more posterior
pieces regenerated, the smaller heads were
formed, and the most posterior piece regen¬
erated only a small protuberance. There
were notable difference in the rate of the1
head-formation between the larvae and the
matured group. The histological process of
the head-formation was similar to several
kinds of Bipalium arleady reported and es¬
pecially in larval stage, the typical baso¬
philic granules in parenchyma were observed .

MO 20

FINE STRUCTURAL CHANGES AND ACID PHOSPHA¬
TASE LOCALIZATION IN THE NERVE CELLS OF
THE REGENERATING FRESH- WATER EARTHWORK,
BRAN CHIURA SOWERBYI II.

K. Shirasawa and N.Makino. Dept, of Biol.,
Tokyo Med. Coll., Tokyo.

Cytological changes and fine structural
localization of acid phosphatase activity
in the nerve cells of the head-regenera¬
ting fresh-water earthworm, Branchiura so-
werbyi were investigated light and elec¬
tron microscopically. The ventral ganglion
cells begin rapid propagation and libera¬
tion from the nerve cord in 2-3 days after
the operation. These free nerve cells move
along the new body wall to the dorsal por¬
tion where the cerebral ganglion should be
reconstructed. In comparison with the
original ganglion cells, the free nerve
cells have larger nucleoli, and their
cytoplasm contain more free ribosomes and
dense bodies. Ac Pase activities of Golgi
cisternae and the vacuoles are remarkably
stronger than the former. The dense bodies
made by accumulation of the Golgi vacuoles
are also positive but RER is always nega¬
tive in these cells. Only in the anlage of
the cerebral ganglion, positive reticular
parts of the Golgi apparatus were observed
It is suggested that main function of the
dense bodies relate with the liberation of
nerve cells from the nerve cord, for exam¬
ple, as lyso soir.es in auto lysis of the use¬
less cytoplasm, in generating worms .

M2 21

1MMUNQH! BIOCHEMICAL LOCALIZATION AND THE
ULTRASTRUCTURE OF NEUROPEPTIDE-CONTAINING
CELLS IN A FRESHWATER 01IG0CHAETE,
TUBIFEX HATTA1 .

H. Jaana. Bio!. Lab., Asahikawa College,
Hokkaido Un i v . of Education, Asahikawa.

Cells containing substance P ( S P ) — ,
oxytocin (OXT) - , and arginine vasopressin
C A VP ) — I ike substances were immunohisto-
chemicaliy detected in the cerebral gan¬
glion; in the subesophageai and the ven¬
tral nerve cord ganglia, ACTH- and SP™
immunoreactive cells were found. These
cells showed a different spatial distri¬
bution in the nervous system. Therefore,
these neuropeptides do not appear to
coexist with each other in a single ceil.
No immunoreast i on was, however, observed
with jS-endorphin- and leu •- enkephalin-
antisera. Electron microscopy revealed
clear differences in the structure of
secretory granules of these cells: SP-
immunoreact i ve ceils had e I ectron-dense
and more or less oblong granules (ca. 130
nm in diameter); OXT-immunoreaetive cells
had medium electron-dense and round gran¬
ules (ca. 180 nm in diameter); ACTH-
immunoreaot ive cells had m etj i u m electron-
dense and large granules (ca. 380 nm or
more in diameter). Unfortunately, the
u ! t r a s t r u c t u r e of the AYR- i mmu n o r e a c t i ve
cells have not been identified.

MO 22

STARVATION OF BOTRYOIDAL TISSUE IN THE
LEECH, HIRUDO MEDICINALIS .

H . Inamura . Dept .of Biol., Tokyo Med. Coll.,
Tokyo _ _ _

Cells of botryoidal tissue of hungry leec¬
hes were examined electron microscopically,
The tissue consisted of endothelia , and gl-
ardularcej' 1 es which contained granules of
three a if re,, en t types : ( 1 )Ll-granules which
were droplet-like ,( 2 ) L2 -granules had many
particles upon which ACPase activity had a
positive reaction, and ( 3 ) P-granules which
were pigmented-like and had a limiting me¬
mbrane . Endothelia had pores and uiapiiragma .
The tissue of a hungry individual during
12 months appeared dark— brown to the naked
-eye. It was observed that the majority of
cells had many P-granules , large L2 -granules
and only a few Ll-granules, and ER and gly¬
cogen granules were degenerative. On the
basal parts of the cells , vesicles of low
density were formed by invagination of cell
membrane, and the cell membrane was high de¬
nsity. On blood luminal parts ,microvillies
which were formed by the cell membrane sho¬
wn simple forms. In the cytoplasm, many sphe¬
rical vacuoles of high density fused with
the cell membrane. On lateral parts of the
cells , interdigitation was a simple form, and
the cell membrane was high density. On mito-
chodria , base and lateral of the cells were
lots, and luminal parts of ones were few.

If is suggested that these cells have a po¬
larity which has a relation to endocytosis
on the basal parts and exocytosis on the
blood luminal parts, and are related to met¬
abolism .

Morphology

1325

MO 23

THE ENDOCRINE CONTROL SYSTEM FOR THE
HERMAPHRODITIC GLAND OF THE SLUG ( LIMAX
MARGINATUS MULLER) BY IN VITRO CULTURE.
N.Seo and N.Makino. Dept, of Biol., Tokyo
Med. Coll., Tokyo.

We are making a study of the functional
hermaphroditism in the slug ( Limax margina-
tus ) . The hermaphroditic gland was cultured
for 2-23 days in the medium added the extract
of hearts or brains or suboesophageal ganglia
or optic tentacles. These organs were taken
out from the slug of similar growth stage.

When the extract of hearts was added into
the medium, all explants had abnormal phases
of acini during 2-7 days after the cultiva¬
tion. The shape of germinal epithelial cells
was transformed into a cube or cylinder, and
male germ cells decreased in number.

When the extract of brains or suboesoph¬
ageal ganglia was added into the medium, it
inhibited the transformation of germinal
epithelial cells and the decrease in male
cells during 1 1-23 days after the cultivation.
However, spermiogenesis was dully.

When the extract of optic tentacles was
added into the medium, the transformation of
germinal epithelial cells was remarkable , but
spermiogenesis befor the cultivation continu.

On the other hand, female germ cells of
all explants were maintained the fine condi¬
tion and the growth of oocytes in all media.
7 days after the cultivation in the medium
added the extract of brains-, we observed
mataphase of 1st maturation cell division
on a few oocytes. This phenomenon is very
interesting .

MO 24

ELECTRON DENSE GRANULES FOUND IN THE
VENTRAL RECEPTACLE CELLS OF MATED FEMALES
IN DROSOPHILA M EL ANOG ASTER.

F.Hihara, H. Shimizu and T.Ohno. Ehime
Univ., Matsuyama.

A peptide which is synthesized in the
male accessory glands of D.melanogaster
represses female sexual receptivity and
stimulates oviposition (Chen et.al. ,1 988 ).
This peptide is transferred along with the
sperm to the internal reproductive organs
of female during copulation. The fate of
this peptide after copulation, however, has
not been clarified so far. We searched the
internal reproductive organs of mated
females for morphological changes after
copulation, indicating the receptor cells
of the peptide. Many electron dense
granules having the lamella structure were
observed between the cuticle layer and the

inner surface with many protrusions of the
absorptive epithelium of the ventral
receptacles. These granules appeared about

3 hrs after copulation and remained
constantly for further 10 days. That the
granules might be formed by incorporating
the male accessory gland peptide was
suggested by the fact: simular granules are
found in ventral receptacle cells of mated
females of D.virilis which males entirely
lack the secretes from the ejaculatory
duct.

MO 25

MURINE INTERDIGITATING CELLS IN LYMPH
NODES INTERNALIZED BY ENDOCYTOSIS la
SURFACE ANTIGEN.

S. Tanaka and H. Uda, Dept, of Pathology
Kagawa Med. School, Kagawa. _

Interdigitat ing cells (IDC) in lymph
nodes belong to the group of antigen-
presenting dendritic cells, and have la
antigen on the cell surface. Ia molecule
is the cell surface glycoprotein coded as
the I region of H-2 complex. It has not
yet been clearly characterized. In the
present study, murine IDC were incubated
in antimouse Ia antigen antibody in
conjunction with colloidal gold particles
(1 5nm) solution at 23°C. The morpho¬
logical pathway of endocytosis were
observed by electron microscopy. Thirty
minutes after incubation Ia complexes
distributed on interdigi tating processes
and accumulated at the root of processes.
One hour after incubation Ia complexes
clustered into coated pits on plasma
membrane, these vesicles have been termed
endosomes. Two hours after, they came in
contact with the Golgi system. Some Ia
complexes were found in the Golgi. Four
hours after, some gold particles were
found to be directed to lysosomes. This
results indicate that the endocytic path¬
way of Ia molecules in IDC is similar to
those of other more classical cell-surface
receptors involved in receptor-mect iated
endocytosis. It is possible that Ia anti¬
gen was prosessed as a receptor in order
to intracellular antigen recognition.

MO 26

IMMUNOCYTOCHEMICAL LOCALIZATION OF LAMININ
IN THE ANTERIOR PITUITARY OF FEMALE AND
OVARIECTOMIZED RATS

T . Kikuta1, Y . Kimura2, S . Yamagucnl', N . Shirasawa3,

S .Kusunoki2, H.Namiki1, 1Dept.of Biol. ,Sch.
Educ.,Waseda Univ. , 2Life Sci .Lab. , Advance,
Dept. of Anat.,Jikei Univ. Sch. Med. , Tokyo
Laminin , one of the major components of ’

the basement membrane ( BM) , was detected in
the gonadotrophs of the male rat anterior
pituitary ( AP ) by Tougard et al in 1985.

Through immunocytochemistry , we have
demonstrated the localization of laminin in
the AP of 10-90 day-old female rats, as well
as ovariectomized rats, and at each stage in
the estrous cycle of 60 day-old rats.

Laminin was localized in the BM, capillar¬
ies, and some LH cells in female and ovari¬
ectomized rats, similar to male rats. In
female rats, laminin positive LH cells were
found to already exist in the AP of 10 day-
old rats. The number of these cells increas¬
ed up until 30 day-old rats but decreased
dramatically thereafter. But in 45 day-old
rats ovariectomized at 30 days, laminin
positive LH cells were still remained in its
number as well as 30 day-old, pre-vaginal
open stage in intact rats. There was no dif¬
ference in stainability at any stage of
estrous cycle. Laminin was also detected on
louble like BM in over 30 day-old rats.

These results suggest that laminin is
produced by the LH cells in female rats
until the first LH surge. In other words,
laminin and/or BM may play a part in organ
growth and particularly sexual maturation
and the appearance of the sexual cycle.

1326

Morphology, Behavior Biology

MO 27

CHARACTERIZATION OF MONOCLONAL ANTIBODIES
AGAINST BOVINE PLASMA FIBRONECTIN
H . Ka j ikawa1, 1 . Kido2, J . Arai^H . Miyazaki1, S .
Kusunoki^H.Namiki2, ^ife Sci .Lab. , Advance ,
2Dept.of Biol.,Sch. of Educ. ,Waseda Univ. ,
3Pept.of Anat.,Jikei Univ.Sch. of Med., Tokyo

Fibronectin ( FN ) appears to play a role in

the organization of other matrix components.
We have produced 2-monoclonal antibodies
( MoAb ) against bovine plasma FN.

MoAb(Fib 6-1,44-1) reacted specifically
with bovine plasma FN by ELISA. Epitop was
characterized by proteolytic and sialic acid
digestic techniques. 2-MoAb reacted with
some proteolytic fragments which include
domain 2. And the epitop did not exist on
the carbohydrate chains. Gelatin ( collagen )-
binding site exists on domain 2. The binding
of FN to gelatin was inhibited by Fib 6-1 ,
but not by Fib 44-1 . This suggested that the
epitop of Fib 6-1 exists on the gelatin¬
binding site.

We examined the immunostainability of 2-
MoAb and polyclonal antiserum in some rat
organs. Polyclonal antiserum reacted
strongly to the basement membrane and
especially to the mesangium area in rat
kidney as well as to capillary and sinusoid
in rat liver. On the other hand, Fib 6-1
reacted to the basemant membrane but not to
the mesangium area and sinusoid. Fib 44-1
did not react to any rat organs.

In our study, heterogeneity of FN was
observed by immunohistochemical study. The
heterogeneity may be caused by higher-order
structure FN or by interaction with other
elements in the basal membrane.

MO 28

ULTRACYTOCHEMISTRY OF ACETYLCHOLINESTERASE
IN THE TASTE BUD AND THE GUSTATORY EPITHE¬
LIUM DURING EARLY POSTNATAL PERIOD IN THE
CIRCUMVALLATE PAPILLA OF THE MOUSE.

M. Kudoh. Dept, of Biol., Fukushima Med.
College, Fukushima. _ _ _

The localization of acetylcholinesterase

(AChE) in the taste bud of mouse vallate
papilla has been histochemically investi¬
gated with the use of a highly sensitive
method introduced by Tago at light and
electron microscopic levels.

The intense AChE activity was uniformly
found around all the intragemmal nerves,
especially at the afferent synaptic region
in type III cell. The extensive reaction
product was clearly observed in rER and
perinuclear space only in type III cells,
not in type I and II cells. These findings
may suggest that type III cell is the pri¬
mary gustatory receptor which transmits
the taste stimuli to the related afferent
nerves. AChE-positive granules, resembling
exocytotic vesicles, were often found to
be attached to the nerves, and several
AChE-positive transport vesicles were
scattered in the cytoplasm of basal cell.
These observations would lead us the spec¬
ulation that AChE stored in rER might be
released by exocytosis to the interspace
between cell and nerve through mediation
of transport vesicles. The cells showing
AChE activity in rER were found in the
gustatory epithelium during early postna¬
tal period. These cells may be undifferen¬
tiated progenitors of type III cells.

MO 29

ACTIN FILAMENTS IN THE TASTE BUDS.

Y. Suzuki (Dept. Anat. , Sch. Dent. , Higashi
-Nippon-Gakuen Univ. , Ishikari-Tobetsu)

Filaments in the apical region of the taste
buds of parrot and goldfish, and in the
taste organ of frog were investigated by
electron microscopy and fluorescent micro¬
scopy. After the application of FITC-label-
ed phalloidin, fluorescence was observed in
the pore region of the taste buds of parrot
and goldfish, and appeared in the apex of
the taste organ of frog as a dot-like patt¬
ern. The apical cytoplasm of the taste cell
of frog and the receptor cells of goldfish
terminate in rod-shaped processes. Thin and
straight filaments (6 nm in diameter) fill
the rod-shaped processes, forming a thick
bundle that extends downward about 5-7 /am
in goldfish and 2-3 Mm in frog. Transverse
sections of the rod-shaped processes showed
over 200 hexagonally packed filaments. De¬
coration of these processes with myosin
subfragment SI showed that the arrowheads
point toward the cell center. The support¬
ing cells of frog and goldfish, and the
taste bud cells of parrot possess micro¬
villi, which contain 6 nm filaments in
their core. It is concluded that the 6 nm
filaments in the apical region of the taste
buds of parrot and goldfish, and in the
taste organ of frog are made of f-actin.
Moreover, it is concluded that the well-
developed actin filaments in the rod-shaped
processes are related to the support of the
membrane of taste and the receptor cells
where the taste stimulus may occur.

BB 1

STENTQR COERULEUS SHOWED POSITIVE
PHOTOKINESIS.

K. Iwatsuki. Nippon Petrochemicals CO.
LTD., Tsukuba Life Science Laboratory,
Tsukuba. _ _ _ _ _ _

Stentor coeruleus prefers darker region.
Since if had been reported that S. coeruleus
did not show photokinesis , it was thought
that the organisms accumulated dark by
means of step-up photophobic response and
negative phototaxis. However, we found that
S. coeruleus showed positive photokinesis.
This is the first report about photokinesis
in " Stentor species. In accordance with
increasing light intensity, the swimming
velocity of the organisms increased. We
conclude that S. coeruleus accumulates dark
region by means of these three types of
phot ©responses .

We got a relatively broad action spectrum
for the photokinesi s ( the peak = 68 Onm ) .The
action spectra for the step-up photophobic
response and for the phototaxis were
different from that for the photokinesis in
S. coeruleus. Therefore, we conclude that
the phototransduction system for the
photokinesis is different from those for
photophobic response and phototaxis.

S. coeruleus swims straight, sometimes
swims round, and stops swimming frequently.
Therefore, it is very difficult to measure
swimming velocity of the organisms. To
solve the problem, we used the "Bug-
Tracker" system (sold by Nippon
Petrochemicals CO LTD) which enable to
record the swimming pattern of any moving
material .

Behavior Biology

1327

BB 2

SOCIAL BEHAVIOUR OF AMOEBAE (VI) .

H. Horikami, A. Kihara and K. Ishii.
Lab. Biol, Hosei Univ., Tokyo

Tracks of 3 species in Amoebae, Amoeba
proteus, strain G (A.p.G.), Mayorella sp.
(M.sp.) and Trichamoeba sp. (T.sp.), in sa¬
line in the dark, were compared with each
other to investigate correlation with their
behaviour and feeding habit in low cell den¬
sity (ca.0.3 cells/mm?) under hungry condi¬
tion .

(1) Most of isolated carnivorous A.p.G. ad¬
vance toward right-side as compared with the
start position at high value (79.5 %) . And
in a group the clination is almost dissolved
( Zool . Sci. , 3:992. 1986) .

(2) In ominivorous M.sp., tracks of iso¬
lated cells for 15 min show mostly winding-
type in contrast to a gentle arc-type of
A.p.G.. Besides, their goals lean to left
area with statistical significance, but no
significance in a group.

(3) Tracks of harbivorous T.sp. are very
similar to one of A.p.G.. However, under
both conditions their goals distribute at
random in right and left area.

(4) Differences of directional angle of lo¬

comotion at every 90 sec were measured with
a computer. Mean angle in A.p.G., M.sp. and
T.sp. were as follows: 7.7° in isolated,
2.1° in a group; 7.0°, 3.6° and 4.0°, -5°

respectively when supposed that positive
value was clockwise .

(5) From these results, a possibility of a
relationship between cell behaviour and
their feeding habit was suggested.

BB 3

EXPERIMENTAL ANALYSIS OF DIURNAL SPAWNING
IN THE CHITON ACANTHOPLEURA JAPONIC A BY
ARTIFICIAL LIGHT AND TIDE CONIDITIONS.

E. Yoshioka. Dept, of Biol. , Far. of
Science, Osaka City University, Osaka.

Environmental time cues entraining the
diurnal and tidal spawning of the chiton A .
iaponica were analyzed by means of
experiments in keeping tanks providing
artificial tide and light conditions.
Artificial tide cycle (12.42 hours) was
established which simulated the semi¬
diurnal alternation of submersion and
emersion. Daily light cycle was generated
by alternate switching on and off of
fluorescent lamps on the tanks. Eggs were
commonly released around light-on time,
i.e. experimental morning, before the
middle of the immersion periods. In three
conditions of immersion term (4.21, 6.21,

8.21 hours), no lag of egg release was
observed between the chitons from different
shore level. The timing mechanism of
synchronous spawning is entrained by a
semidiurnal tidal submerging cycle, sensed
at both the beginning and the end of
immersion, and by a diurnal light-dark
cycle. This leads to a release of gametes
principally before the peak of morning high
tide during spring tide. The timing
mechanism guarantees that chiton gametes
are released at the time of high water,
when they are submerged, and that release
occurs within a short period, independent
of the height inhabited by the chitons.

BB 4

CONTROL OF LARVAL RELEASE AND LOCOMOTOR
ACTIVITIES IN THE CRAB SESARMA PICTUM
M. Saigusa. Coll, of Liberal Arts & Sci.,
Okayama Univ., Okayama.

Larval release and locomotor activities
of the terrestrial crab Sesarma pic turn were
monitored in the laboratory in which 1 ight
and temperature were controlled. When ovi-
gerous females were collected from the fie¬
ld and were kept in constant light conditi¬
ons (LL) , they showed a free-running tidal
rhythm, the pattern of which was clearly
' bimodal' (i.e., 12.5-h period). Light-dark
cycles caused a phase-shift of the tidal
rhythm, and the degree of the observed shift
was roughly corresponded to the time diffe¬
rence between sunset and lights-out.

On the other hand, in constant dark (DD),
or even in light-dark (LD) conditions, not
many crabs showed a clear rhythmicity in the
locomotor activity pattern. For the speci¬
mens having a distinct periodicity, the
circadian component (ca. 24-h period) was
observed; and only one individual exhibited
an activity rhythm with about 12-h period.

As noted by previous experiments, the
daily timing of larval release is controlled
by 0C~ fi system, but the presents experiments
suggest that this system is not involved in
the control of locomotor activity, or that
only (X-oscillator (neurons generating a 24-
h period) is weakly related to its expre¬
ssion .

BB 5

THE LOCOMOTOR ACTIVITY RHYTHM AND THE FEEDING
ACTIVITY IN THE HOUSE FLY„,MUSCA DOMESTICA.

M. Aizawa1 and H. Yoshino . Furukawa Girl 's High
School 1 , Furukawa High School 2, Furukawa, Miyagi.

Locomotor activity rhythm of Musca domestica
synchronizes with LD cycle and shows diurnal type
of activity under the LD 12:12. Under the LL,
activity increases and the cycle becomes shorter.
Under the DD, freerunning was observed and the
cycles were 26.3 h in wild type P-9, 23.9 h in the
mutant- wbp and 25.0 h in FI (Aizawa, 1986).

In this study, relation between the freerunning
rhythm and the feeding activity under the DD was
observed. An actograph was prepared with a poly-
acryl pipe of 8 mm in diameter and 100 mm in length.
In the one end, water was supplied with moistened
gauze and in the other end with solidified milk.

The fly was sealed in this pipe. When the fly
exhibits its locomotive activity, the pipe suspend¬
ed on an axis moves up and down and thus leads to
electrical contacts at the relay under the pipe.
This electrical on and off were recorded on a
chart. The amount of feeding activity was measur¬
ed by summing up the times when the fly stays at
the milk. The experiment revealed that feeding was
done during the period of locomotive activity and
its cycle was nearly the same as the cycle of
locomotive activity. However, since the beginning
time of the feeding activity and the time of
maximum feeding was different from the locomotive
activity, the locomotive activity rhythm of the
fly would not be induced by hunger but be the
result of a physiological oscillation endogenous
to the fly.

1328

Behavior Biology

BB 6

FRACTAL PROPERTIES OF THE FEEDING BEHAVIOR
IN DROSOPHILA.

I.Shimada^ and H.Hara2. ^Dept. of Biol.
Sci., Tohoku Univ. and 2Dept . of Eng. Sci.,
Fac. of Eng., Tohoku Univ., Sendai. _

One day old male flies of Drosophila

starved 20 hr were put into micro test plate
and fed 30 min. Foods such as 10 mM sucrose
CIOS) or 5 mM sucrose C 5S ) or pure water
(OS) mixed with II agar were placed in 60
wells on the bottom of the plate. In order
to analyze the feeding movement of the fly
quantitatively we measured the number of vi¬
sits to the foods and the staying time on
the foods by using the video recorder and
personal computer.

When the distribution of the staying time
on 1 0S and 5S alone was described by the
log-log plot, it was characterized by linear
curves with the slopes of -1.1 and -1.5, re¬
spectively. On the other hand, the distri¬
bution on OS was not linear and showed a
breakpoint at 2.5 sec. The fly did not in¬
gest OS since it was given fully with water
under 24 hr starvation. When 60 wells were
alternately filled with 1 0S and 5S (choice
situation) , the fly almost ignored 5S and
ingested 1 0S exclusively. Under such con¬
dition the distribution on 5S was no longer
linear but showed a breakpoint at 4.0 sec,
while the distribution on 1 0S did not show
any breakpoint. The breakpoint, therefore,
indicates a lack of ingestion. The break¬
point at 4.0 sec may suggest a close rela¬
tion between the staying time of 4.0 sec and
the ingestion of 5S. The results were dis¬
cussed from a fractal point of view.

BB 7

COMPARISON OF CIRCADIAN ACTIVITY AMONG
MOSQUITOES IN THE CULEX PIPIENS COMPLEX.
Y. Chiba, Y.Shinkawa, and K.Tomioka. Biol.
Inst., Yamaguchi Univ., Yamaguchi, 753
Japan.

The Culex pipiens complex includes two
well-characterized forms, quinquefas-
ciatus and pipiens , with their inter¬
mediates such as pallens in Japan. In
addition, there is molestus , which has
developed in underground construction of
almost total darkness as an autogenous
strain of pipiens (Barr 1975). We
compared circadian activity among pallens
and four strains of molestus collected in
Egypt, Iran and two domestic parts (Ehira
and Yanagawa) of Nagasaki Prefecture.
Under LD 16:8, the five forms of mosquito
all showed two activity peaks
synchronously with the steps of the light
change. The Iranian and the Japanese
molestus were temporally photophilious ,
being active also in the light fraction,
but the Egyptian tended to be
scotophilious as pallens . Another
remarkable difference was found in the
free-running period (tau) under DD, which
was 21 h in the Iranian and around 24h in
the other forms. Among overt circadian
parameters, tau is the one reflecting
directly the function of clock. In F i
between the Iranian and pallens and in
the F2 , the average length of tau under
DD was intermediate between those in
parents, ranging from 21 to 25h.

BB 8

APPEARANCE OF SPERMATOPHORE PROTRUSION ( SPP)
IN ADULT MALE CRICKET , GRYLLUS BIMACULATUS .

J.Okada and N.Ai. Dept, of Biol., Tokyo Gaku-
gei Univ. ,Koganei, Tokyo. _ _ _

In adult male cricket ( IM7-14 ) projection

of spermatophore from the end of abdomen is
pre-performance for copulation and sexual
attraction for adult female. This behavior
is called spermatophore protrusion(SPP) as
the signal of sexual promotion. In this ex¬
periment, needed time for SPP (SPP duration)
is measured from the end of copulation to
the next SPP. Periods of observation SPP
duration is decided as follows ; 8 : 00-10 : 00 ,
15:00-17:00,20:00-22:00 and 3:00-5:00.
Typical duration time of SPP was normally
about 10 minutes or little more. In this
observation, SPP duration from mid-night to
evening was almost same one. The most short
period as 10 minutes appeared with high fre¬
quency. However, in observation period of
20:00-22:00 series, SPP duration was prol--
onged and duration times were scattered
with wide range. On this period, 20 : 00-22 : 00 ,
sexual promotion in male may be decreased
clearly. Average of SPP duration was two
times longer than that of other period.

When preparation with short SPP duration
selectively was moved to 20:00-22:00 series
it showed change of SPP duration with long
interval. Is seems to be important the time
ing of copulation. This period is the impo¬
rtant for female on oviposition. Isolation
of male from female after copulation makes
clearly plolongation of SPP duration with¬
out 20:00-22:00 series.

BB 9

PHOTOPERIOD DURING POST-EMBRYONIC DEVELOP¬
MENT AFFECTS SOME PARAMETERS OF CRICKET
CIRCADIAN LOCOMOTOR RHYTHM.

K. Tomioka and Y. Chiba Biol. Inst.,
Yamaguchi Univ., Yamaguchi. _

Adult male cricket, Gry 1 lus bimacul atus

exhibits a nocturnal circadian locomotor
rhythm. Effect; on the adult locomotor rhythm,
of the ratio of light to dark phase (L/D-
ratio) of the 24h LD was investigated using
animals reared under either of the three
LDs", i.e. LD16:8, LD12:12 and LD8 : 1 6 . Activ¬
ity pattern differed from each other depend¬
ing on the LD conditions in which the animal
was reared. The length of the nocturnal
activity phase which was about 12h inLD12:
12, shortened to about 8h in LD16:8 but
never lengthened more than 12h even in LD8 :
16 with the longest dark phase. This fact
suggests that the maximal length of the noc¬
turnal activity may be homeostatically regu¬
lated to be 12h. In the constant darkness,
freerunning periods and the ratio of the
activity phase to the rest phase ( <X/fi> -ratio )
were affected greatly by the preceding LD-
ratio. Average freerunning period of the
animal reared in LD16:8 was 24 . 49h signifi¬
cantly longer than about 23. 8h of animals
reared in other LDs . The o<//o -ratio changed
in inversely proportional to the L/D-ratio:
0.88, 0.94 and 1.29 in LD16:8, LD12:12 and
LD8 : 1 6 , respectively. The photoperiod depen¬
dent change in the circadian parameters
suggests that the organization of the crick¬
et circadian system is regulated, at least
partly, by environmental factors.

Behavior Biology

1329

BB in

FACTORS FOR TRIGGERING THE STR I DULATORY
ACTIVITY IN THE LONG -HORNED GRASSHOPPER.

I. Nakatani. Dept, of Biol., Fac. of Sci.,
Yamagata Univ., Yamagata.

The stridulatory activity of the long¬
horned grasshopper, Hexacent rus j apon i cus ,
was restricted to the first half of the
dark period under the photoperiod of
LD12:12 at 24°C. The activity exhibited a
free running rhythm while in constant
darkness, and no activity occurred under
constant light. In the open field,
however, one can observe the singing long-
horned grasshopper under the direct rays
of the sun in late aut umn . To examine the
stimulation inducing the stridulation, the
temperature cycles were studied. The
stridulatory activity occurred in the
first few hours of the cold period while
in constant light and in the temperature
cycle of 10-hours 24°C and 14-hours 14 °C.
When the long-horned grasshopper was
exposed to the cycle of 24°C in the light
and 18 °C in the dark, the activity
occurred in each period. But the activity
did not occur when the photo cycle was
changed to constant light under the same
temperature cycle. These results may show
that the falling temperature can act as a
Zeitgeber and this factor was reinforced
by the light-dark cycle.

BB 11

LOCOMOTOR ACTIVITY RHYTHM OF LOACHES SYN¬
CHRONIZES TO LIGHT-DARK CYCLE AND SCHEDUL¬
ED FEEDING AND FREERUNS UNDER A CONSTANT
CONDITION.

M. Naruse and T. Oishi, Dept. Biol., Fac.
Sci., Nara Women's Univ., Nara.

We investigated (1) an interaction bet¬

ween light-dark cycles and scheduled feed¬
ing on the locomotor activity rhythm of
loaches ( Misgurnus anguillicaudatus ) and
(2) freerunning rhythm under a constant
condition (continuous darkness without
food). The results are as follows. (1)
all the fish except one (93%) exhibited
greater amount of locomotor activity in
dark when food was given at the middle of
dark-phase. Percentage of the fish which
were more active in dark decreased to 36%
when food was given at ~the mi-ddle of
light phase (p < 0.01 by XT- test). (2)
freerunning rhythm under the constant
condition was observed in 29% of the fish
that were previously given a meal at
mid-night and in 7% of the fish that were
previously given a meal at mid-day. But
the freerunning rhythm of these fish
disappeared after about 10 days in the
constant condition and the activity became
arrhythmic. Thus, we conclude that the
locomotor activity rhythm of the loach is
circadian Yn nature (but it__.peems to be
weakly coupled to oscillator ( s ) ) and it is
affected by both LD cycles and scheduled
feeding .

BB 12

EFFECTS OF THE ABLATION OF SENSE ORGANS
AND PARTS OF THE BRAIN ON SWIMMING
BEHAVIOR OF THE HAGFISH.

S. Ooka and H. Kabasawa. Atomi Jr.
Coll., Tokyo, Aburatsubo Marine Park
Aq . , Miura .

The swimming behavior in the hagfish,
Eptatretus burgeri, were examined by
observation and record with an infra¬
red light photocell system. Surgical
ablations or cut of the eyes, ears,
lateral line nerve, telencephalon,
diencephalon, optic tectum, medulla
oblongata and spinal medulla were
performed. Operations were done with
a pair of small scissors while the
animals were lightly anesthetized with
MS-222. Almost all animals survived
after the operations except for the
ablation of medulla oblongata. A rotatory
motion during swimming after the ablation
of ears indicated the function of them
in spite of the morphological devolution.
The swimming pattern after removing
of whole brain except medulla oblongata,
which includes cerebellar rudiment,
did not differ from that of intact
animals. On the other hand, the animals
could not swim up after the cut between
medulla oblongata and spinal medulla,
and they only crept on the bottom of
aquarium. It is concluded that connection
of medulla oblongata and spinal medulla
is important for the coodination of
the movements in swimming.

BB 13

CATADROMOUS MIGRATION OF SALMON SMOLT OBSERVED
IN AN AQUARIUM WITH A CURRENT 2

'J'.Moriya , Y.Miyashita and E.Komiyama .
2Dept . of Biol., Sapporo Med. College, Sapporo,
Living Museum of the Salmon, Sapporo. _

Fish generally swim against the current to
maintain their position in a river; they always
turn their heads toward the upper part of a
river. However, in the case of the salmon
smolts, they have to go to the sea early in
life. If they maintain their general behavior
vis-a.-vis the water current, they can’t reach
the river mouth. Thus, we can expect that some
changes will occur on their swimming behavior.
In order to observe these changes , we made a
circular experimental tank which had a current.
If the young salmon were cultivated in water
with a current continually after hatching, they
always maintained their position in the stream
by swimming against the current. We did not
observe that they swam with the current in the
experimental tank throughout the observation
time (until 130 days after hatching). On the
other hand, if the young salmon were kept in
still water for some periods, and then moved to
the experimental tank, they soon started to
swim with, and faster than, the current. This
catadromous migration continued for several
days. Then, they reversed the swimming
direction from downstream to upstream and kept
their position in the stream. Staying in still
water for some periods may trigger of the
downstream migration. The swimming speed during
catadromous migration was 17 cm/sec. This speed
was faster than the water current (5-7 cm/sec).
The distance traveled during this period, which
was calculated from the swimming speed and
time, seems to be enough to reach the river
mouth .

1330

Behavior Biology

BB 14

THE ROLE OF OLFACTORY MEMORY FOR ORIENTA¬
TION TO THE BREEDING POND IN BUFO JAPONICUS
H. Nishio, K. Kubokawa , H. Yamanouchi, M.
Itoh, M. Kikuchi , H. Funahashi, M. Takizawa
and S. Ishii. Dept, of Biol., Sch. of
Edu. , Waseda Univ. , Tokyo 169.

We have reported that the geographic
memory is used by toads for orientation to
the pond during the breeding migration at
least in a area near the pond. We studied
effects of the surgical disturbance of the
visual and olfactory sense on the orienta¬
tion of toads. Toads were captured at sites
less than 150m from the pond. Some of them
were deprived of vision by sewing both
eyelids together. Control toads were
operated with the needle without threads.
Some other toads were deprived of olfaction
by damaging the nasal mucosa with a 5% sil¬
ver nitrate solution injected through nost¬
rils. Control toads were treated similarly
with saline. Directional choices of blind
toads within a day or two after operation
were random. Nevertheless, 35% of blind
toads were found in the pond finally. This
level was almost identical to control (40%)
and intact (35%) levels. In contrast, only
9% of anosmic toads were found in the pond.
This proportion was significantly Lower
than the corresponding control value (41%).
These results suggest that toads use the
olfactory memory, not visual memory, of the
route to migrate to the breeding pond.

BB 15

URINE-MARKING RESPONSES OF JUVENILE AND
ADULT HOUSE MICE.

M.Daumae and T.Kimura. Dept. Biol., Coll,
of Arts and Sci., Univ. of Tokyo, Tokyo.

We studied urine marking patterns of
male and female mice of various ages, in
response to urine marks of their own, or
of other individuals of the same sex and
age. At 4 weeks of age, both males and
females deposited larger number of urine
spots and larger amount of urine (total
area of spots) in the area previously
marked by other individuals ( 1 other 1 ) than
in the clean area (' clean ') , but they
showed no difference in marking between
'clean' and the area previously marked by
themselves ('own'). At 9-10 weeks of age,
both males and females deposited larger
amount of urine in previously marked area,
whether 'other' or 'own', than 'clean',
but increase of urine spot number was
observed only in males in the presence of
'other', and only in this case, mice spent
more time in the previously soiled area.

These results show that, in mice, urine
marking response to already marked area
changes with age. In adult males, frequent
urination with tiny spots was enhanced by
'other'. This urination pattern suggests
the communicative function. Responses of
juveniles seem to indicate that they
utilize urine marking for communication in
different ways from adults, though further
experiments are needed to elucidate the
function of marking in juvenile mice.

BB 16

EFFECTS OF MALE MOUSE URINARY FACTORS ON
URINE-MARKING ACTIVITY OF FEMALES.

K. Ninomiya and T.Kimura. Dept, of Biol.,
Coll, of Arts and Sci., Univ. of Tokyo,
Tokyo.

We have reported that both urinary and
preputial factors of males are involved in
female attraction. In this study, we
examined female urine-marking activity in
response to the odor of male exereted
urine, in expectation that this can be
used as a measure of soliciting behavior.

Female mice deposited more number of
urine spots on the floor with urine of
normal males than that of castrated ones.
Two androgen-dependent compounds ( dehydro-
exo-brevicomin [ I ] and 2-sec-butyldihydro-
thiazole [ II ] ) , known to be present in
normal male urine, enhanced female marking
when mixed with castrated male urine.

These results are consistent with the
previous results concerning odor prefer¬
ence of females. However, females showed
no marking preference between normal male
urine and preputialectomized male urine,
while females prefer to investigate the
former. Further experiments with various
combinations of urine samples from
differently treated males, with or without
[I] and [II], revealed that females showed
no marking preference when one of two
urine samples was from preputialectomized
males, whether urinary factors were
present or not. It seems likely that the
male odor without preputial factor causes
females to mark by unknown motivation.

BB 17

MALE MICE : SOCIAL DOMINANCE INFLUENCED
BY STRANGE MALE ODORS.

S. Hayashi. Dept. Biol., Fac.Educ.,
Kagoshima Univ., Kagoshima.

Male mice were either individually
raised with a castrated male (C) or raised
with 5 normal males (N) from 5 to 1 5 weeks
of age. Half of C and N groups were
subsequently isolated in clean cages for 2
w^eks (X) while the other half were housed
in isolation for 2 weeks in cages that had
previously housed a strange male (E).
After 1 4 days of isolation-housing each
male of 4 groups (CE, CX, NE & NX) was
placed into pairs with a male from another
groups and observed for 30 min in a
neutral arena.

CE males were more likely to be domi¬
nant than both CX and NE males. When each
male of 4 groups was paired with a
castrated male, CE males behaved more
aggressively than any other groups. On
the other hand, CE and NE males had
heavier preputial glands than CX and NX
males although there was no significant
difference between preputial glands of CE
and NE males.

These results suggest that a hormonal
change was caused by strange male odors.
However, it does not necesarily affect
aggression. The strange male odors en¬
hance aggressiveness in males which have
had little experience of aggression.

Behavior Biology, Ecology

1331

BB 13

ONTOGENY OF LEARNING ABILITY OF ddY MICE:
EFFECT OF VISUAL EXPERIENCE.

N.Mishima, H.Nakano and R.Iguchi. Dep. of
Biol., Fac. of Pharma. Univ. of Tokushima
Bunri, Tokushima.

Passive avoidance task was established
by ddY mice at the age of 1 5 days when
their eye-lids just opened. In the more
complicated task such as LD-discrimina-
tion T-maze avoidance task, An correct
response of the task did not appear till
17 days of age, but acquisition of 23-day
old mice reached almost the same level as
adult mice. Also, learning for food
reward such as Skinner box task or 8-arm
radial maze task could be established by
the weaning time which is about 1 week
after eye-opening. These results show
that learning ability of various types of
tasks rapidly appear during 1 week after
eye opening. To study the effect of
visual experience on appearance of the
learning ability, eye-closed mice whose
eye lids were sewed up from 1 4 days till
16, 19 and 22 days of age was trained in
T-maze at 17, 20 and 23 days age,
respectively. Learning abilities of naive
mice and the eye-closed mice were not
significantly different in acquisition
rates and performance levels at all the
three ages. These results suggest that
the ontogeny of learning ability in mice
is not depend upon the amount of visual
experience, but some genetically deter¬
mined program of neural development.

BB 19

DECREASE IN STRIATAL ACETYLCHOLINE OF
"DULL" LINE GENETICALLY SELECTED FOR
MEMORY RETENTION OF PASSIVE AVOIDANCE.
Y.Maeda, T.Mizuta, R.Iguchi and N.Mishima
Dep. of Biol., Fac. of Pharm. , Univ. of
Tokushima Bunri. Tokushima.

The ninth progeny of "dull" mice were
obtained by selective breedings of poor-
memory animals in 1 -trial passive avoid¬
ance task. In learning of an active
avoidance T-maze task or a food rewarded
Skinner box task, acquisition rates were
significantly lower in "dull" mice than
random-bred ddY mice. To find what caused
such deficits in memory retention and
learning acquisition in "dull" mice, some
biochemical components of their whole
brains were compared with the controls.
There was no significant difference in
contents of water and total lipids,
however, protein concentration of "dull"
mice was about 90% of the control (p<
.001). In addition, ACh levels of cortex,
striatum and hippocampus were measured by
HPLC . Only in striatum, ACh levels of
"dull" mice were significantly lower than
the controls (p< .01). Further, age-
dependent decline of ACh was observed in
striatum of either "dull" and the control
ddY mice. These results suggest that
deficits of memory retention and learning
acquisition in "dull" mice were mostly
related to lower levels of ACh and
related to progress of aging.

EC 1

MORPHOLOGICAL STUDY ON SKELETONS OF
DIASERIS FRAGILIS , A FREE-LIVING CORAL
WHICH REPRODUCES ASEXUALLY BY NATURAL
AUTOTOMY .

H.Yamashiro1 , M.Hidaka2. M.Nishihira2 and
S.Poung-In . ' RI Lab., 2Dept. of Biol.

Univ. of the Ryukyus, and ^Chulalongkorn
Univ.

A disc-shaped individual of Diaseris
f ragilis has radial slits in its skeleton
and finally breaks up into several wedge-
shaped fragment. New skeletons (daughter
segments) are formed around the point of
the wedge-shaped fragment (mother
segment). Specimens of D^ f ragilis at
various stages of regeneration were
collected from the Gulf of Thailand.
Alizarin-stain method and SEM observations
suggest that daughter fragments grow
faster than mother fragments. Slits were
detectable in the daughter fragments whose
radii were greater than 2.2 mm. There was
a fine slit less than 1 0 urn wide at the
initial point of the slit. Skeletons
facing the slits had a smooth surface
suggesting that the calicoblastic tissue
was senescent, though needle-shaped
crystals were observed on the skeleton
facing the fine slit. The present study
suggests that skeletal growth stops at
several points of the growing edge of the
skeleton. Since other areas of the
skeleton continue growing, this, rather
than active dissolution of the skeleton,
results in the formation of the slits.

EC 2

POSSIBLE ROLE OF THE DIVISION OF THE COLONY
IN A SURVIVAL STRATEGY OF THE COMPOUND
ASCIDIAN , POLYCITOR PROLIFERUS .

Y. Taneda and T. Yamaguchi. Dep. of Biol.,.
Fac. of Educat., Yokohama Natl. Univ.,
Yokohama .

Ascidian colony is developed from a
single zooid which derived from a tadpole
larva. The growth of the colony is brought
about by the proliferation of the zooids.
When a colony of Polycitor prolif erus
retains a certain size, it divides into
two or three small colonies. How does
the colony know its size? To answer this
question we measured the size of the colo¬
nies selecting three parameters, length
of the colony, number of the zooids and
the area of the colony. The size of the
colonies is variable with localities,
seasons and micro-topography of the coast.
Simulation experiments by computer revealed
that experiments agreed very closely with
the observations if the probability of
colony division rised along the sigmoid
curve. Environmental factors such as
temperature, pH and dissolved oxygen may
affect the probability of colony division.

In summer, aged zooids are sexually
mature. Mature zooids were abundant in
the large colonies. While, they were
scarecely observed in the small colonies.
Therefore, the possible role of rejuvena¬
tion from the large old colonies is indi¬
cated in the division of the colonies.

1332

Ecology

EC 3

DIFFERENCES IN TEMPERATURE DEPENDENCE OF
EARLY DEVELOPMENT AMONG SEA URCHINS WITH
DIFFERENT DISTRIBUTIONS OR SEASONAL GROWTH
PATTERNS.

H. Fujisawa, Fac. of Educ., Saitama Univ. ,
Urawa.

Sea urchins have species-specific tem¬
perature sensitivities especially in the
embryonic stage, and each sensitivities
evidently determine their distributions as
well as their spawning seasons.

In tidal areas around Japan, several
species of sea urchin are distributed from
the Tropical to Subarctic Zones : Echino-
metra mathaei and Diadema setosum are trop¬
ical species, Anthocidaris crassispina and
Clypeaster japonicus are warm- temperate
ones, Pseudocentrotus depressus and Hemi-
centrotus pulcherrimus are cold- temperate
ones, and Strongylocentrotus intermedius is
subarctic one.

Using these species of sea urchin as
materials, I investigated species-specific
temperature tolerance and dependence during
early development, and clarified quanti¬
tatively how the temperature sensitivities
of embryos of these sea urchins are related
to their spawning seasons and distributions.

EC H

ON A HYDROZOA, ASYNCORYNE SP. SYMBIOTIC WITH THE
ALCYONACEAN SOFT CORAL, CLAD I ELLA PACHYCLADOS.

Y. KAMIZATO AND K. YAMAZATO. BIOL. DEPT. COLL.
SCI., UNIV. RYUKYUS.

Morphological study on a symbiotic hydrozoa, Asyn-
coryne sp. living on an alcynoacean soft coral, Cladiel la
pachyclados was carried out. Symbiotic hydrozoa
were found on 40X of the soft coral colonies in
the study sites at Sesoko Island, Okinawa. Population
of hydropolyps ranged 10 to 12 per every 0,25 cm2 of
the host surface. The hydranth is unbranched, at¬
taining about 2. 0-2. 5 mm in height, with 4-5 short
capitate oral tentacles and with 14-20 scattered
moni I iform aboral tentacles.

Ectodermal cells in the middle part of the hydro¬
caul us are flat and are provided with microvilli,
while those in the basal part are tall and contain
various organelles and inclusions such as rER, Golgi,
free rebosomes, vesicles, electron dens bodies and
vacuoles. The basal part of hydrocaul us penetrates
into host epidermal layer, which becomes deep hollow.
Epidermis of the basal part of the hydrocaul us is
covered with a parisarc facing closer to the micro¬
villous layer of the host epidermis in the deeper part.
Bundles of mesogleal fibers run to the parisarc from
the mesoglea where the hydrocaul us penetrates into
the soft coral. Hydrohizas branch off horizontally from
the basal parts of the hydrocaulus at the mesogleal
level. Hydrorhizas with the parisarc are surrourded
with extremely thin host epidermal cells.

EC 5

SYMBIOTIC BACTERIA OF APHIDS AND SOME INSECTS.
N. Muramoto. Hakodate La Salle High School,
Hakodate .

It is already known that heteropteran insects
have intracellular symbionts . Muller , H . J . ( 1 962 )
attempted to classify the Cleorrhyncha(Hemiptera)
by means of symbionts.

Aphids kept complicated ecology; i.e., not only
they have intracellular symbionts but also
sjmibiosis, predatism and parasitism among other
insects .

So far it is generally recognized that herbi¬
vorous insects have intracellular symbionts, but
carnivorous insects don't.

In this study, intracellular symbionts were
observed in every animal (ants, an ephemera,
ladybirds, a parasitic bee and a nematode) which
has some ecological interaction with aphids.

EC 6

PREVALENCE OF INTESTINAL HELMINTHS OF THE
FROG ( RAN A BREVIPODA POROSA AND RANA JAPONI-
CA) -II

Y. ocisaki and N. Makino. Dept, of Biology,
Tokyo Med. Coll., Tokyo.

Prevalence of intestinal helminths that
were parasitic in the frog ( Rana brevipoda
porosa_) were studied during the 9 -month peri¬
od except winter. Fifty frogs were collected
each month. Sex ratios of 450 frogs were
approximatly equal. A small number of Rana
j aponica ( about 20frogs per month) were also
investigated during the 14 -month period for
comparison. Both hosts were captured at
Inzai, Chiba Prefecture.

The nematodes, Cosmocerca i aponica and
Oswaldocruz ia sp . and the trematodes . Diplo-
discus japonicus and the cestodes .Ophiotae-
nia sp.were found in Rana b.p. .Total preva¬
lence was high. There was no prevalence of
Oswaldocruz ia sp.and Ophiotaenia sp.as they
were observed in only small number. Cosmo¬
cerca j . displayed a very high prevalence
through each month. In some hosts only new¬
born or larvae were found. In some hosts new¬
born , larvae and adults were present in all
periods. Diplodiscus j . showed a low preva¬
lence through each month. Above-mentioned
effects were not significantly different
between males and females. Prevalence of
Cosmocerca j . and Diplodiscus i . in Rana i .
showed the same tendency as the prevalence
in Rana b ,p ..Oswaldocruz ia sp.was peculiarly
parasitic in Rana j .during all month . Pseudo-
acanthocephalus bufonis was prevalent in
winter. In some tadpoles of Rana j . only a
few larvae of Cosmocerca i . were observed.

Ecology, Taxonomy and Systematics

1333

EC 7

INVOLVEMENT OF A SEX PHEROMONE IN PHYSIO¬
LOGICAL ISOLATION OF TWO RACES OF THE
STRIATED SPINED LOACH.

Y. Masaoka and T. Kawai , Dept. Biol., Fac.
Sci., Nara Women's University, Nara.

The striated spined loach (Cobitus tae¬

nia striata ) comprises several "races",
and two of them are synpatrically found in
our study site, situated in the middle
reach of the Uji River. Recently, they
were classified into different species
based on their morphological characters by
Saitoh and Aizawa (1987). However,

ecological evidence is not enough to argue
this problem. Thus, we investigated their
habitats, food habits and reproductive
periods to find a mechanism for the
physiological isolation. We could not
find any significant difference in these
parameters between the two races. We,
then, observed the behavioral responses of
males (pecking, quivering and coiling) to
existence of ovulated females or to the
incubating water in which an ovulated
female was kept. Positive responses of
males to the females and the water were
observed not in heterospecific combina¬
tions but only in conspecific ones. Thus,
a sex pheromone seems to be involved in
physiological isolation of the morpho¬
logically distinguishable synpatric races
of the striated spined loach.

EC 8

ROLE OF VOCAL COMMUNICATION AS A SPECIES
RECOGNITION IN TWO STORM-PETRELS,
OCEANODROMA LEUCORHOA AND O.MONORHIS.
M.Taoka and H.Okumura. Dept, of Biol.,
Japan Women's Univ., Tokyo.

Oceanodroma leucorhoa and Oceanodroma
monorhis . nocturnal seabirds, are closely
related species. Similarities are present
in calls of them. We studied role of vocal
communication as a species recognition in
their colonies. Both of Chatter-calls of
0.1. and Flight Calls of O^el. have sexual
dimorphism in frequency components and
function as a sex recognition. We played
back these calls to the birds of the other
species in each colony. Both species res¬
ponded to playback of the heterospecific
calls. However, they did not discriminate
the sexes of the other species calls and
recognised them as a female homospecific
call. This results from the different
mechanisms of sex recognition of both
species. The calls are differentiated to
function as a premating isolation. The
responsiveness to the heterospecific call
is significantly lower than that to homo¬
specific calls in (LL*.. However, 0, m.
responded to the hetero-specific call like
to homospecific calls. The mechanism of
species recognition by call is not fully
developed in the species of (L.HU- This
suggests that CL.nL. is speciated from 0.1..

TS 1

THE NUMBER OF OOCYST PRODUCED IN A
GAMETOCYST OF GREGARINA BLATTARUM
K. Hoshide and Y. Takeda Biol. Lab., Fac. of
Educ . , Yamaguchi Univ., Yamaguchi

The number of oocysts produced in a
gametocyst of _G_. blattarum was counted
with a light microscope. We mixed the
oocysts of G_. blattarum with baits and fed
Blattella germanica the mixture. The
gametocysts were eliminated in their feces
in 7 to 16 days after feeding. We
collected these gametocysts and kept
them in a moist chamber. After two
or three days, numerous oocysts were
dehisced through several sporeducts. The
oocysts were too many and too small to
count in the natural condition as it was.
Vie tried several methods and at last
succeeded in counting the number by using
a special mat of our own making ( having
1 320 squares and 800 pm squares in each)
and Neo-shigaral ( Shigakonchu Inc.), medi¬
cine for fixing and mounting oocysts. We
scattered the dehisced oocysts on the
mat, fixed them with Neo-shigaral solution,
and covered them both with a 24x54mm
coverglass. The number of oocysts produced
in one gametocyst is varies, from 6,000
to 60,000. The diameter of the gameto¬
cysts increase according to the time from
infection. The larger gametocyst contain
more oocysts. The number of oocysts
increases according to the time from
infection.

TS 2

A PHYLOGENETIC STUDY OF THE HYPOTRICH
CILIATES .

0. Noguchi, T. Matsusaka. Dept, of Biol.,
Fac. of Sci., Kumamoto Univ., Kumamoto.

Fission and encystment morphogenesis
was compared among several different
species of hypotrich ciliates . The results,
together with the earlier reports,
revealed that the genus Uroleptus (family
Urostylidae) showed considerably different
morphogenetic patterns from other
urostylid species:

1) During binary fission, cirral streaks
originated from only 1 primordium, a
part of dorsal bristle rows developed
from anterior end of right marginal
cirral primordium, and no kinetosome
proliferation occurred in the buccal
cavity .

2) Encysting specimens showed stub-like
cirri during cirral resorption.

3) Mature cyst was covered with 3-layered
cyst wall.

These morphogenetic characteristics
and mature cyst morphology of Uroleptus
resembled to those observed in the members
of family Oxytrichidae , indicating that
the genus Uroleptus may occupy a position
between family Urostylidae and family
Oxytrichidae .

1334

Taxonomy and Systematics

IS 3

THE NEW SMALL SPECIES OF THE LAND PLANARIAN
GENUS ; BIPALIUM ( TURBELLARIA , TRIG LAD IDA ) WHICH
BELONGS TO THE NON-FISSION TYPE AND THE
FOURTH TYPE OF THE COPULATORY ORGAN.

N. Makino and Y. Shirasawa. Dept, of Biol.,
Tokyo Med., Coll., Tokyo.

~~~~~Yhe . small BipaTIum sp.was collected in

the senior author's garden as the fifth
species from October inl987 to August in
1988. The large specimen in the winter was
115mg in body weight and 50mm in elongated
form. The collected worms were about ten
specimens. Their breeding season is the
winter and the infant was hatched out during
three weeks from the cocoon, with lor 2or 3
bodies in 18 °C. The adult individual has
dark brownish color in dorsal side and light
brown in ventral side and it has a blacky
narrow median line on the back. This species
is not fissionable .The large specimens have
genital pore, so the pharynx opens in the
middle position of total length( 0 .5 ) and the
genital pore's position is six-tenth(0.6)~
seven -tenth( 0.7) from anterior part of the
body. It has a dark point near the genital
pore. The reproductive apparatus of the worm
is formed using several months after the
hatching, but the forming of gonads , ovary and
testis is more early stage. The genital pore
seems likely to open with near half a year
after the hatching. The type of copulatory
organ belongs to the fourth type that
authors determined the four types to them.
The chromosome number was 2n=10 as same as
B.fuscatum of the non-fission type, and also
the fission type, B, nobile and B. multi-
lineatum .

TS 4

KARYOLOGICAL AND TAXONOMIC STUDIES OF THE
DUGESIA SPECIES IN SOUTHEAST ASIA. XIII.
ECOLOGY AND TAXONOMY OF DUGESIA BATUENSIS
FROM BATU CAVES IN MALAYSIA.

M.Kawakatsu R. W. Mitchell2 , S.Yussof 3, I.
Oki4 , and S.Tamura4. 1 Biol. Lab. , Fuji Wom¬
en's Coll., Sapporo, 2Texas Tech Univ.,
Lubbock, U.S.A., 3 Malayan Nat.Sci., Kuala
Lumpur, Malaysia, 4 Osaka Pref .Inst .Publ.
Health, Osaka.

Twenty species of freshwater planarians
of the genus Dugesia are now known from
Southeast Asia. Among them, Dugesia batu-
ensis Ball, 1970, the only cave inhabitant,
is known from Batu Caves near Kuala Lumpur,
Malaysia. This species is also known from
an epigean locality near the type locality
(Kawakatsu, 1972 a, b).

The Batu Caves are formed in limestone,
and the largest chamber is "Dark Cave",
which is approximately 560 m in length. On
Sept. 20, 1987, specimens of D. batuensis
were found in pools at several locations
in Dark Cave. They were especially abun¬
dant in the innermost pool. Numerous co¬
coons of this species were also observed.
For example, over 200 cocoons were found
on the surface of a flattened stone 6X8
cm in size. D. batuensis is one of the
prominent invertebrate species inhabiting
the waters of this cave.

Many color slides showing live specimens
of D. batuensis taken in the habitat were
exhibited. The taxonomy of this species
was also discussed.

TS 5

KARYOLOGICAL AND TAXONOMIC STUDIES OF THE
DUGESIA SPECIES IN SOUTHEAST ASIA. XIV.
CHROMOSOMES OF DUGESIA BATUENSIS FROM BATU
CAVES IN MALAYSIA.

I. Oki1, S.Tamura1 , R.W. Mitchell2 , S.Yussof
3, and M. Kawakatsu 4 . 1 Osaka Pref .Inst .Publ
Health, Osaka, 2Texas Tech Univ., Lubbock,
U.S.A., 3 Malayan Nat.Sci., Kuala Lumpur,
Malaysia, 4 Biol. Lab. , Fuji Women's Coll.,
Sapporo.

Specimens of Dugesia batuensis Ball,

1970, from Batu Caves were examined cyto-
logically . The chromosome numbers of this
species are n=7 and 2x=14 ; its karyotype
is 2m + 2m + 2m + 2m + 2sm + 2m + 2m.

Among the species of Dugesia. the follow¬
ing basic numbers are known: n=4, n=7, n=

8 and n=9. Most of the species have the
basic numbers of n=8. The species having
basic number of n=4 and n=9 are known only
from North and South America, including
the Caribbean area. The basic number of n=

7 is now known in the following 3 species:
D. hepta Pala, Casu et Vacca, 1981, from
Sardinia, Italy; D. japonic a ryukyuensis
Kawakatsu, 1976, from the Southwest Islands
of Japan; D. batuensis from Malaysia (re¬
ported in the present abstract). These 3
species belong to the "Dugesia gonocephala"
group. . .

Karyotypes of the 2 species from South¬
east Asia having the basic number of n=7,

D. batuensis and D. jj_ ryukyuensis . were
discussed in detail. A new simple method
of karyological analysis was also proposed.

TS 6

MARINE ENOPLID NEMATODES OF INTERTIDAL
SEDIMENTS IN HOKKAIDO, NORTHERN JAPAN, WITH
REFERENCE TO THE FAMILY ONCHOLA1MIDAE .
K.Kito, Dept, of Biology, Sapporo Medical
College, Sapporo _ _ _

Marine enoplid nematodes of intertidal

sediments were investigated at ten
localities on the coasts of the Pacific, the
Japan Sea, and the Okhotsk Sea in Hokkaido,
northern Japan. Twenty-three species of
eight families were recognized; Enoplidae (2
spp.) , Thoracostomopsidae (1 sp.) ,

Phanodermatidae (1 sp.) , Anticomidae (1
sp.), Leptosomatidae (1 sp - ) , Oncholaimidae
(12 spp.), Enchelidiidae (4 spp.), and
Tripyloididae (1 sp.) . Faunal feature of
enoplid nematodes in Hokkaido was
tentatively discussed with regard to the
members of the family Oncholaimidae showing
the highest diversity.* Adoncholaimus (2
spp.) , Oncholaimus (6 spp.) , Pontonema (1
sp.) , Pseudoncholaimus (1 sp.) , and Viscosia
(2 spp.) . Among the twelve oncholaimids,
three species are cosmopolitan and another
three are Pacific species, of which two are
restricted to the west side of the Pacific.
The remaining six are new to science.

Taxonomy and Systematics

1335

TS 7

TWO NEW SPECIES OF THE GENUS MONHYSTRIUM
(NEMATODA) INHABITING THE GILL CHAMBERS
OF CRABS OF THE SUBFAMILY SESARMINAE .

K. Yoshimura, Ube Junior College, Ube.

Two new species of the genus Monhyst-
rium ( Monhysteridae : Nematoda ) were found
from the gill chambers of crabs of the
subfamily Sesarminae (Grapsidae) collected
at the Koto River and the Ariho River,
Yamaguchi. All the three hitherto describ¬
ed species of the genus are also known to
occur on the gills of "land crabs" ( Ge-
carcinus and Cardisoma , Gecarcinidae ) dis¬
tributed along the Caribbean coasts.

One of the present new species is long¬
er (1361 - 1617 mm) in total body length
than the other (866 - 1041 mm). These two
species are also different from each other
in the shape of vulva, the structure of
spicule and gubernaculum , and, especially,
the relationship to their host crabs.

The examination of 1 1 species of
Sesarminae demonstrated that Chiromantes
haematocheir , Ch. dehaani , and Chasma-
gnathus convexus were associated with the
shorter species of Monhystrium , while
the longer one was found from Parasesarma
plicatum , P. pictum , P. erythrodactylum ,
and Clistocoeloma merguiense . The inter¬
relationship between these nematodes and
crabs seems to reflect the vertical dis¬
tribution of the crabs.

TS 8

PLANKTONIC ROTIFERS FROM TA I WAN , R . 0 . C .
M . Sudzuki

B i o 1 . Lab . , N i hon Daigaku Univ. , Omiya;

11 samples were collected from Wushan
tou dam(W), Cheng Ching pond(C),Jih Yueh
lake(Z) & prawn pond(P) during 29-31VII
1987 on the way back from Singapore. A
new record for Tai wan ( c f . Har ada , 1 9 3 3- 3 8
Ueno , '35 & Tien-Hsi Tan et al'80) is as
follows: — Brachionus a . bldens ( W , C , Z ) , B .
bidentata ( W , Z ) , B . c . per sonatus ( W , C ) , B . g .
cluniorbicuraris ( Z , C ) , B . sessilis(W ) , B .
nilsoni(W) , Anuraeopsis coelata(C) , A. na-
vicula(C) , Epiphanes macr ourus ( W , C ) , Lec-
ane aculeata(C) , L. elsa(Z) , L . haliclysta
( Z ) , L . hastata ( Z ) , L . proiecta ( W ) , Monosty-
la bulla(W,C,Z) ,M.closterocerca(C) ,M.
lunar i s ( W ) , M . quadridentata ( W ) , Colurel la
obtusa(W) , Lepadella latusinus(W) , L.ova-
11s (C , Z ) , L . patella ( C ) , L . rhomboides ( C ) ,
Euchlani s d . luckslana (W , C) , E.meneta(C) ,
E . sp ( C ) , D ip leuchlani s propatula(Z) , Tri-
pleuchlani s p 1 i c at a ( W ) , Lophoc har i s sal-
pina(W) ,Mytilina b i c ar i nat a ( C ) .M.v.iac-
r acant ha ( C , Z ) , M . v . r edunca ( C ) , Macrochae-
tus collinsi(W) , Tr ichotria t . c audat a ( W )
Asplanchna herr icki ( W , C ) , Asplanchnella
br ightwel 1 i ( C ) , Cephalodella g 1 bba ( W ) , C .
sp ( W , C ) , Trichocerca c.chattoni(WZ) , Diu-
rella bidens(C) , D. stylata(C) ,D.ruttneri
( C ) , Ascomorpha saltans(C,Z) , Synchaeta
stylata(C.Z) , S . sp ( C , Z ) .Polyarthra t . do-
1 i chopt er a ( Z ) ,P.minor(W) ,Hexarthra T.
brazil i ens i s ( C ) , Test udi ne 1 la pat i na ( W ) ,
Pompho lyx sulcata ( W , C , Z ) , Conoc h i 1 o i de s
coenobas i s ( W , C , Z ) , Collotheca sp(W,Z),
Ptygur a sp. (W,Z)and Rot ar i a t r i dens ( C ) .

TS 9

TARDIGRADES IN HOKKAIDO AREA.

K.Utsugi. Dept, of Biol., Tokyo Women's
Medical College, Tokyo. _

In order to study Japanese tardigrades,
118 samples of mosses and lichens were col¬
lected at 22 stations throughout Hokkaido.
These samples were collected from cities,
farms, fields and valleys during the period
from 1 984 to 1 988.

The presence of tardigrades were micro¬
scopically examined in each precipitate
after extricating them from the mosses or
lichens. Most of the specimens were mounted
with Gum-chloral for phase contrast micro¬
scopic observation. Some of them were fixed
in 5 % formalin after heat treatment and
prepared for SEM observation.

1 5 species of tardigrades were found in
26 samples collected at 14 stations.

3 species including Macrobiotus harmsworthi ,
M . huf elandi and M^_ intermedius were common¬
ly found in many stations. However, 4 common
species found (M^_ richtersi , Milnesium tar-
digradum , Hypsibius novemcinctus and Pseud-
echiniscus suillus ) were rare when compared
to other areas in Japan. Especially Echini-
scus j aponicus , which is widely distributed
in Japan, was not found in all 118 samples.

The other 8 species including M^_ echino-
genitus , M. maucci* , H. baumanni , H . macula-
tus , Isohypsibius eplenyensis* , Diphascon
alpinum , Echiniscus dreyf usi and E . virgini-
cus were also rarely found in Hokkaido.

Among these tardigrades, 2 species indicat¬
ed by * were reported for the first time
in Japan.

TS 10

ENDEIS NODOSA: A SEA SPIDER NEW TO JAPAN
K. Miyazaki and T.Makioka. Inst, of Biol.
Sci., Univ. of Tsukuba, Ibaraki.

A sea spider, Endeis nodosa Hilton
( Pycnogonida ; Endeidae), was collected from
the ascidian rearing baskets floating on
Nabeta Bay, near the University of
Tsukuba' s Shimoda Marine Research Center,
Shimoda, Shizuoka.

This species has previously been reported
only "from Hawaii and the Marshall Islands;
this is the first record from Japanese
waters .

E. nodosa is easily distinguished from
most other Endeis species, especially by
the well-branched gut-diverticulum in each
walking leg and the numerous and
irregularly arranged cement gland-pores on
male femurs. E. nodosa is most similar to
E. f laccida Caiman, a circumtropical
species, but differs from the latter in
having a ventral femoral projection on each
leg and more robust first and second
tibiae, and having no protuberance on the
sixth segment of ovigerous legs.
Observations by scanning electron
microscope revealed some peculiar features
on the genital pores and the cement gland-
pores in this species.

1336

Taxonomy and Systematics

TS 11

MARINE INTERSTITIAL OSTRACODA FROM
HOKKAIDO.

S. Hiruta, Biol. Lab., Kushiro College,
Hokkaido Univ. of Educ . , Kushiro.

Up to now, six species of four
different genera ( Psammocythere [ 1 ] ,
Microloxoconcha [ 1 ] , Cobanocy there [ 1 ] ,
Microcythere [ 3 ] ) of marine interstitial
Ostracoda were collected from the beaches
around Hokkaido (1985-1988). Materials
were extracted from intertidal sands of
various depths (20-50 cm) by means of
decanting and sieving method with tap-
water, using a sieve net with aperture
size of 0.04 mm. Granulometoric nature of
sands, from which ostracods were
detected, is very different in different
beaches and depths. The range of median
grain size is from 0.15 to 1.2 mm.

Microloxoconcha- and two Microcythere -
species are widely distributed around
Hokkaido, while Psammocythere- and one
Microcythere- species were collected only
from Kushiro, on the pacific coast of
eastern Hokkaido, and Cobanocy t her e-
species only from Hamatombetsu , on the
northern coast of Sea of Okhotsk.

Global distribution of these genera are
described, and means of dispersal were
briefly discussed, since some
interstitial ostracods show the
transoceanic or long distance distribu¬
tion of the related species.

TS 12

THE ASCIDICOLIDAE (COPEPODA, CYCLOPOIDA)
ASSOCIATED WITH COMPOUND ASCIDIANS FROM
AKKESHI BAY.

S. Ooishi1 , and P. L. Illg2. 1Fac. of Fish.,
Mie Univ., Tsu and 2Dept. of Zool., Univ.
of Wash., Seattle, U.S.A. _ _

In the studies of ascidicoles from Akke-
shi Bay, Hokkaido, in the western Pacific,

7 species of the family Ascidicolidae have
been found. These species are all associat¬
ed with compound ascidians collected in
1968 and 1980, and belong to 3 subfamilies,
Enteropsinae , Haplostominae and Botryllo-
philinae. In the Enteropsinae, the female
and the male fifth copepodid of Mychophilus
roseus have been observed for the first
time in the Pacific. In the Haplostominae,
the females of 3 species examined are fun¬
damentally comparable to those of Haplo-
stoma albicatum . Hap lost omi des luteolus and
Haplosaccus elongatus from other localities,
although there are some minor differences
in the Akkeshi material. Haplosaccus
elongatus from Akkeshi Bay involves both
sexes and the male of the genus has never
been studied so far. The male basically
has the same characteristics as the males
of Haplostoma and Haplostomides , bearing
the antennule ornamented with numerous
aesthetes. In addition, the maxillule is a
small but distinctly setiferous structure
as in that of the male of Haplostomides .

In the Botryllophilinae , the females of 3
species of Botryllophilus examined belong
morphologically to the type A of the
females of the genus.

TS 13

TWO NEW SPHAEROMATIDAE (CRUSTACEA : ISOPODA )
FROM UWAJIMA, SHIKOKU.

N.Nunomura, Toyama Science Museum, Toyama

Two kinds of qeer-looking sphaeromatid
isopods were collected from the intertidal
zone of Uwajima Port, Ehime Prefecture.

One of them is similar to the genus
Gnorimosphaeroma of the subfamily cassidi-
ninae, but is lacking in the exopod of uro-
pod. Further, it proved to belong to the
subfamily dynameninae and is considered to
be a new genus. The resemblance between the
present new genus and Gnorimosphaeroma is
only convergent. This genus seems to be
most closely related to the genera Dynamene
and Dynamenopsis , for they are lacking in
suture line of 3rd pleopod. This genus has
other advanced features such as less numer¬
ous teeth on appendages of the mouth part.

Another species belongs to the genus
Paracerceis and may be new to science.

This is the first record of the genus from
Japanese waters.

TS 14

THE LARVAL DEVELOPMENT OF PINNIXA TUMIDA
STIMPSON, 1858 UNDER LABORATORY CONDITIONS
(CRUSTACEA: DECAPODA: PINNOTHERIDAE) , WITH NOTES

ON LARVAL CHARACTERS IN THE GENUS PINNIXA.

1 o o

K. Konishi , S. Kiyokawa and S. Maruyama .

^Natn. Res. Inst ,_ Aquacult. , Mie and ^Hokkaido
Wakkanai Fish. Exp. Stn., Hokkaido.

Larvae of Pinnixa tumida Stimpson, a
commensal crab with the apodid holothurian
Paracaudina chilensis (J. Mtiller), were reared
from hatching to the crab 1 stage under
laboratory conditions: artificial sea wataer of
35 %o salinity, temperature 20 °C, and Artemia
nauplii as food. The larval development
consists of 5 zoeal and 1 megalopal stages with
a mean duration of 53.7 days after the hatching.
The larvae of Pinnixa are immediately
distinguished from other known brachyuran larvae
by their diagnostic features: lateral expansion
of abdominal somite 5 in zoeas, and adult-like
appearance in megalopas. Within the family
Pinnother idae , pinnixid zoeas do not exhibit
extreme morphological diversity as found in
Pinnotheres species. Differences between the
zoeas of 2 Japanese pinnixids, P. rathbuni Sakai
and P. tumida, are hardly recognized. On the
other hand, the megalopas and crab 1 stage of
these species are easily identified to species:
in P. tumida the merus of cheliped bears a
conspicuous hook-like spine (megalopa) or a knob
(crab 1) .

Taxonomy and Systematics

1337

TS 17

TS 15

ROLES OF CATCH CONNECTIVE TISSUE IN THE
EVOLUTION OF ECHINODERMS.

T . MOTOKAWA .

Dept, of Biol., Col. of Sci., Univ. of
Ryukyus , Okinawa.

Echinoderms have a unique connective
tissue, the catch connective tissue: it is
a collagenous connective tissue that
changes its mechanical properties rapidly
under nervous control. With this
tissue, echinoderms have developed a
unique supportive system, the supportive
system which is composed of small ossicles
bound together by catch connective tissue.
This system has properties which are well
suited for rheophilic suspension feeders:
it can change its rigidity; it is
economical in posture maintenance and in
construction of the body; it contributes
to the defense of the animal by providing
hard body and appendages such as spines
and cover shields. This supportive system
very likely contributed to the success of
early echinoderms as sessile rheophilic
suspension feeders. It is also likely
that this system contributed to the
evolution of motile forms from sessile
ancestors .

Echinoderms are designed to move a
little. Most of their unique features are
the results of adaptation to the epi-
benthic life with a little movement.

The supportive system has permitted
echinoderms to have a body which is
effective in mechanical defense without
loss of flexibility and mobility.

TS 16

THE VERTICAL DISTRIBUTION OF ECHINOIDS IN
THE SAGAMI SEA.

M. Shigei. Misaki Marine Biological Stati¬
on, Univ. of Tokyo, Kanagawa.

A total of over 3,000 specimens of echi-
noids collected from the intertidal zone to
a depth of 1,540 m of over 700 localities
in the Sagami Sea were examined systematic¬
ally and the collecting data were used for
the study of the vertical distribution.

It was cleared up that the most of all
the Sagami-Sea species had constant ranges
of the vertical distribution.

Interspecific segregations of the verti¬
cal habitats were found in such genera as
Clypeaster , Peronella , Lovenia and Echinoc-
ardium . The ranges of the vertical distrib¬
ution were constant even at the genus level
in some families.

The numbers of species found in benthic
vertical zones were as follows:

36 in the intertidal-upper sublittoral
zone (= coastal region; 0-30 m depth); 45
in the lower sublittoral zone (= shelf reg¬
ion; 30-200 m depth); 22 in the upper bath-
yal zone (200-500 m depth); 9 in the middle
bathyal zone (500-1,500 m depth).

The present study was greatly due to
His Majesty the Emperor of Japan, Hirohito
who repeatedly surveyed Sagami-Bay faunas
for over sixty years (1927-1987) and offer¬
ed me about 650 echinoid specimens collect¬
ed by himself at over 320 stations which
extended from the intertidal zone to a
depth of 450 m of Sagami Bay.

BIOCHEMICAL SYSTEMATICS OF THE SIX SPECIES OF
SEA-URCHINS OF THE FAMILY ECH IN0METR I DAE FOUND
IN THE JAPANESE WATERS.

N. Matsuoka1 and H. Suzuki2.

1 Dept, of Biol., Fac. of Sci., Hirosaki Univ.,
Hirosaki and 2 Res. Dept., Olympus Optical
Co. , Ltd. , Tokyo.

The family Echi nometridae of the order Echinoida
found in the Japanese waters is represented by six
sea-urchin species belonging to the five different
genera. They are Anthocidaris crassi spina ,

Ech i nometra mathaei , Echinostrephus aciculatus,
Ech inostrephus molari s, Colobocentrotus mertensii ,
and Heterocentrotus mammi 1 latus . The phylo¬

genetic and evolutionary relationships among these
six members of this family were investigated by
electrophoretic analyses of 18 different enzymes.
The biochemical dendrogram for the six species
constructed from the Nei’s genetic distances
between species for 26 genetic loci showed the
following: (1) H. mammi 1 latus and C. mertensii

are largely genetically differentiated from
the other four species and these two specie are
considered to be more primitive sea-urchins among
echinometrids from the molecular estimation of
their divergence time. (2) A. crassispina endemic
to Japan is more closely related to Ech inometra
than to Echinostrephus, and it is estimated that
Anthocidari s has speciated in relatively recent
geological age of the late Pliocene. (3) The two
morphologically very similar species of

Echinostrephus (E. aciculatus and E. molaris)
are little genetically differentiated from each
other, and therefore these two may belong to one
and the same species. These electrophoretic

results were discussed through the detailed
comparison with other molecular and non-molecular
results .

TS 18

WRINKLED BLASTULA OF THE SEA URCHIN, PERO¬
NELLA JAPQNICA

M. Komatsu, M. Murase and C. Oguro. Dept, of
Biol., Fac. of Sci., Toyama Univ., Toyama.

Peronella iaponica is one of A sea ur¬
chins which pass through the wrinkled blas-
tula stage. However, descriptions on the
wrinkled blastula are fragmentary. It seems
that little attention has been paid to this
stage in echinoids. In the present study,
changes in the cell arrangement during the
wrinkled blastula stage were observed with
a SEN. Microscopic observations on the pro¬
cess of the wrinkling were done on fresh
embryos and preparations made by routine
paraffin method. Early blastula before the
wrinkling is coeloblastula with a large
blastocoel surrounded by monolayered
blastomeres. Four hr after insemination,
they develop into wrinkled blastula stage
at about 21°C. In this time, folds occur on
the surface of the blastula by multipolar
invaginations of the blastoderm. Subsequent¬
ly, these invaginations become deeper and
more complex. At the most wrinkled stage,
they reach the center of the blastocoel. In
a late stage, they become shallower and
simpler. Two hr after the beginning of the
wrinkling, the surface of the blastula re¬
sumes smoothness. Six and half hr after in¬
semination, gastrulation takes place. Thus,
external and internal changes during the
wrinkling of P. iaponica is similar to
those observed in the sea stars, Asterina
pect inifera and Astrooecten scooarius ■

1338

Taxonomy and Systematics

TS 19

GENETIC VARIABILITY AMONG THREE TYPES IN
HALOCYNTHIA RORETZI.

K. Owada and M. Hoshi. Dept, of Life
Sci., Fac. of Sci., Tokyo Inst. of
Techno 1., Tokyo.

Halocynthia roretzi is classified into
three types(A, B, C) according to the
season and the time of spawning and
external characterizations. They seem
reproductively isolated and may represent
an early stage of sympatric speciation.
To test such possibilities at the level
of DNA, restriction fragment length
polymorphism (RFLP) was studied.

Nuclear DNAs extracted from
individuals were digested by Eco RI, Pst
I or Dra I, and degradation products were
analyzed by Southern hybridization using
three probes; l)mouse rDNA 18S, 2)mouse
rDNA 28S, 3)Halocynthia roretzi histone
H3. Polymorphism was found in the pattern
of 18S and 28S rDNAs but not in H3.
However, no difference between Type A and
C was observed.

No polymorphism was found in
mitochondrial(mt) DNA so far studied by
using BamH I and Dra I recognizing 6
bases and by Taq I and Hha I recognizing
4 bases.

RFLP observed in rDNAs indicates
genetic variability in Halocynthia
roretzi, but we have not yet found any
RFLP characterizing each types.

TS 20

THE UNUSUAL ORGANIZATION OF THE GOBY
PITUITARY AS THE RESULT OF PAEDOMORPHOSIS .
K. Tsuneki and Y. Koshida. Dept, of Biol.,
Coll, of Gen. Educ., Osaka Univ., Osaka.

The pituitary of many gobioid teleosts
is embedded partly in the hypothalamus.
This is in contrast to the hanging pitui¬
tary of most teleosts. In order to know
how the peculiar organization of the goby
pituitary was brought about, the develop¬
ment of the brain of the goby, Rhinogobius
f lumineus , was studied histologically. In
the embryos, the pituitary was embedded in
the hypothalamus as in the adults. The
pituitary of embryos of non-gobioid
teleosts also is embedded in the
hypothalamus to some extent. Therefore,
the organization of the goby pituitary
might be brought about by paedomorphosis .
In parallel with this study, the size of
nuclei of cerebellar Purkinje cells and
granular cells was measured on histolo¬
gical preparations of many adult perciform
teleosts including 16 species of gobies.
The PG ratio) the volume of nucleus of
Purkinje cells/that of granular cells) was
usually higher than 10 in non-gobioid
perciforms, but it was lower than 10 in
most gobies. In contrast to granular
cells, Purkinje cells increase in size
during ontogeny. Therefore, the small PG
ratio of adult gobies also may be the
results of paedomorphosis. It is unclear
whether the paedomorphosis of gobies was
brought about by progenesis or neoteny.

TS 21

INTERPOPULATION VARIATION IN LARVAL LIFE
HISTORY OF HYNOBIUS NIGRESCENS
Hisashi Takahashi and Hisaaki Iwasawa.
Biol. Inst., Facul. Sci., Niigata Univ.,
Niigata

In studying the relationship between
larval growth and population-specific egg
size, larval growth was compared among 9
populations of H_;_ nigrescens from various
habital altitudes, breeding seasons and
breeding ponds. Interpopulation varia¬
tion in egg size affected the size of the
hatchlings in each population. Larvae in
low-altitudinal and early-breeding

populations (females deposit smaller
eggs) had a longer larval period and
larger size at metamorphosis. On the
other hand, larvae in high-altitudinal
and later-breeding populations (females
deposit larger eggs) had a shorter larval
period and smaller size at metamorphosis.
Among populations at similar altitudes,
larvae of the populations having stable
breeding ponds and low larval density,
derived from smaller eggs, had a longer
larval period, and were larger at
metamorphoses. Interpopulation variation
in larval morphology were .clearly seen.
Larvae of one population having a tem¬
porary pond and high larval density had
larger head width than those of the
populations having low larval density.
In laboratory experiments, larvae from
the populations having larger eggs were
larger at metamorphosis in comparison
with the larval period.

TS 22

GEOGRAPHIC VARIATION IN ALLOZYMES OF
CYNOPS PYRRHOGASTER DISTRIBUTED IN EASTERN
JAPAN.

T. Hayashi1 and M. Matsui2.

TDept. of Zool., Fac. of Sci., Kyoto
Univ., Kyoto and 2Biol. Lab., Yoshida
Coll., Kyoto Univ., Kyoto.

Using standard horizontal starch gel
electrophoresis, we analyzed 16 loci of 12
enzymes in 19 populations of CL_ pyrr-
hogaster collected from Tohoku and Kanto
Districts and from Niigata Prefecture. At
three of 12 polymorphic loci (S-Aat-A,
Pgm-A and Pgdh-A), electromorph frequen¬
cies showed latitudinal dines, and espe¬
cially, at Pgdh-A locus, predominant
electromorphs clearly differed between
Kanto populations and the others. In the
UPGMA phenogram based on Nei ' s genetic
distance (D), the first major dichotomy
separated Kanto populations from Tohoku
and Niigata populations (mean D value -
0.083). The latter group showed a very low
(D = 0 - 0.038) level of intragroup genet¬
ic differentiation in spite of its rela¬
tively large range. On the contrary, Kanto
populations showed a higher genetic varia¬
bility within the group (D = 0 - 0.115).
C. pyrrhogaster has been morphologically
divided into six local races, of which two
(Tohoku and Kanto races) occur in eastern
Japan. Geographic pattern of allozymic
variation agrees with the distribution of
the two local races and supports the
division between them.

Taxonomy and Systematics

1339

TS 23

ORIGIN AND EVOLUTION OF A BROWN FROG, RANA

SP. FROM HONSHU, JAPAN

M. Matsui, Biol. Lab., Yoshida Coll.,

Kyoto Univ., Kyoto

An undescribed brown frog Rana sp. was
discovered in 1978 from Tokyo, HOnshu. The
species is unique among brown frogs in its
lotic-breeding habit, but the larva lacks
dark body pigmentation, relatively large
body size, and specialized mouth parts that
are often seen in other lotic species. The
larva spends a relatively short life unlike
other lotic larvae among pebbles on the
bottom of the stream, and seldom appears in
the open water. These morphology and ecolo¬
gy are nearly identical with those of alli¬
ed R. tagoi , which is also unique in the
habit of underground breeding. The common
ancestor of R. sp . and R. tagoi, derived
from lentic breeder, seems to have attained
the habit of underground breeding and have
lost sexual dimorphism in the body size and
full webbing in the toe, as seen in the
extant R. tagoi. The latter two character¬
istics seem to have secondarilly developed
in R. sp. in adaptation for a torrential
breeding. R. sp . and R. tagoi seem to have
speciated sympatrically within Japan
through the separation of the place of
breeding by a parental population of the
common ancestor that overwintered in the
stream. The one retained the habit of land¬
ing and breeding in the underground, and
the other developed the habit of remaining
and breeding on the bottom of the stream.

TS 24

THE NUMBERING OF THE DIGITS OF THE CHICK
WING ON THE BASIS OF THE MOLECULAR
MARKER, AV-1.

K. Ohsugi. Biol. Inst., Fac. of Sci.,
Tohoku Univ., Sendai.

There is a considerable disagrrement
between palaeontologists and embryo¬
logists considering the numbering of
birds' fingers. Palaeontologists are
labelling the wing digits as 1, 2 & 3.
But embryologists are labelling that as
2, 3 & 4.

The AV-1 antigen is the molecule
which shows position specific expression
during the determination stage of the
cartilage pattern in chick limbs. In
stage 28 leg buds, the expression of the
AV-1 antigen was observed between the
precartilaginous regions of tibia and
fibula, and those of the metatarsal 1 and
2. In stage 28 wing buds, the expression
of the AV-1 antigen was observed between
the precartilaginous regions of radius
and ulna, and those of anterior two
metacarpals. On the basis of such
expression pattern of this molecular
marker, the numbering of the chick wing
digits seems to be 1,2 & 3.

TS 25

MATSUMORI TANEYASU'S RYOU KINRUI ZUFU .
N.Isono^ and Y.Uchida. 1 Dept, of Biol.,
Keio Univ. , Yokohama.

Matsumori Taneyasu (1825-1892), an
amateur naturalist, wrote Ryou Kinrui
Zufu (Birds of Yamagata and Akita Prefec¬
tures) in 14 volumes during his last
ten years of life. It constitutes the
most important part of his lifework,
Ryou Hakubutsu Zufu (Natural History
of Yamagata and Akita Prefectures) in
59 volumes, and contains about 650 pic¬
tures of birds, all in color. The almost
all of them are so accurate that we could
identify 257 species and subspecies of
Japanese birds, including many rare visi¬
tors and accidentals, such as Phaethon
lepturus , Buteo hemilasius , Aegolius
funereus, Nyctea scandiaca, L anius spheno-
cercus, Phylloscopus fuscatus , Erithacus
svecicus and Remiz pendulinus . We also
found a picture of a male of an extinct
species, Tadorna cristata, in Oizumi
Shocho Shashin Gafu Furoku: Kyogai Shocho
no Bu (Illustrations of the birds he
saw in Edo), which is one of his works
precedent to Ryou Kinrui Zufu. He sketch¬
ed it at a pet shop in 1868.

Ryou Kinrui Zufu also gives much
useful information on the distribution
of birds. For example, it was shown
that Cyanopica cyana had already reached
the east part of the Tohoku District
as far north as Sendai, and that Ciconia
ciconia was still breeding in Edo.

ZOOLOGICAL SCIENCE 5: 1340 (1988)

© 1988 Zoological Society of Japan

[Book Review]

Invertebrate and Fish Tissue Culture

Yukiaki Kuroda, Edouard Kurstak, and Karl Marmorosch, editors.

Japan Scientific Societies Press, Tokyo, and Springer-Verlag, Berlin, Heidelberg, New
York, London, Paris & Tokyo.

1988, 294 pp.

To bring tissues and cells into culture conditions
is an apparent advantage for ones who study
cellular functions at molecular basis. Moreover,
cultured tissues and cells can provide with a good
experimental system in pathology and microbiolo¬
gy. In those aspects, there should be nothing
different between vertebrates and invertebrates.
Here is a book entitled “Invertebrate and Fish
Tissue Culture”, the edited proceedings of the VII
International Conference on Invertebrate and Fish
Tissue Culture, held May 10-16, 1987 at Ohito,
Izu Peninsula, Japan. The readers would learn
from this volume that a good store of information
that has been accumulated from studies using
vertebrate cells and tissues is not always applicable
to invertebrates, and may understand why the
particular field of the invertebrate tissue and cell
culture has been established.

Among 66 papers contributed to this volume 48
deal with insect cells and only 5 papers deal with
invertebrates other than insects. (Since the VII
Conference was broadened by fish cell culture
being added, 13 papers are devoted to fish cells.)
Those papers are arranged accordingly under 7
Parts, corresponding to the 7 Sessions of the
Conference: Pt.l. New cultures and techniques,
Pt.2. Physiology and endocrinology of cells in
vitro, Pt. 3. Production of specific proteins by
cultured cells in vitro, Pt. 4. Cell differentiation
and gene expression, Pt. 5. Application of cell
cultures to pathology and microbiology, Pt. 6. Fish
cell cultures, and Pt. 7. New cell lines and their
characterization.

Papers reporting newly developed media for
insect cell cultures as well as the newly established
cultures of insect cells and tissues are in Parts 1 and
7. Since characteristics of insect cells are
tremendously variable due to the variety of spe¬
cies, it seems to be inevitable to develop
equivalently variable types of culture media. Until
a totipotent culture medium (Mitsuhashi, at p. 15

of this volume) is developed, many new media will
have to be formulated for specific species and
tissues of insects. Three papers report 3 different
new cell lines from Bombyx, the important animal
with extensive commercial and experimental usage
here in Japan. The paper that interests me most in
Part 7 is one from Mitsubishi-Kasei Institute of
Life Sciences. It presents successfully established
20 cell lines from Drosophila imaginal discs of
several different kinds. In addition, those cell lines
are shown to retain the capacity to differentiate as
adult tissues according to their original imaginal
discs under an appropriate culture condition.
These cell lines would make strong tools for
studies asking what cell determination is. Other
papers along the line of developmental biology
appear in Pt. 4., where readers may find interest¬
ing examples of the demonstration of cell dif¬
ferentiation in cultures at morphological, molecu¬
lar, and genetic bases.

Applications of in vitro culture techniques to
viral incubation for harvesting a quantity of the
virus or for testing pathological reactions of cells
against viral infection are beautifully presented in
13 papers in Pt. 5. At present those authors may
be ones who appreciate insect cell cultures most.
Furthermore, in many fields of biological sciences,
in particular in the developmental biology, the
contribution of insects and other invertebrates as
materials for studies at molecular and genetic basis
is becoming more and more important. Inverte¬
brate cell and tissue cultures are therefore more
needed than any time in the past to become
laboratory’s routine techniques, as mammalian cell
cultures are now. The present volume may give
may readers an idea about the points they need to
work on in this specific field of scinece.

Masukichi Okada

Tsukuba University

Tsukuba, Japan.

ZOOLOGICAL SCIENCE 5: 1341 (1988)

© 1988 Zoological Society of Japan

[Book Review]

Biology of Horseshoe Crabs

Koichi Sekiguchi, editor
Science House Co., Tokyo.

distributed by Business Center for Academic Societies Japan, Tokyo.
1988, 428 pp.

This comprehensive volume covers a wide range
of biology of the horseshoe crab, one of the well
known “living fossils”, and is the revised and
enlarged edition of the former book in Japanese
(Science House, 1984, 346 pp.). In 1930’s the late
Professor Hidemiti Oka of the Tokyo University of
Education studied experimental embryology of the
Japanese horseshoe crab and Dr. Sekiguchi en¬
tered this field under Prof. Oka’s guidance. After
more than 30 years’ extensive work Professor
Sekiguchi edited this voluminous book assisted by
a number of his disciples and friends. This book
consists of 11 chapters: History, Taxonomy,
Biogeography, Ecology, Morphology, Normal De¬
velopment, Experimental Embryology, Ex¬
perimental Hybridization, Karyology, Biochemis¬
try, and General Discussion, written by the experts
in each field. The Editor himself contributed the
chapters, 1, 3, 4, 6, 7, 8, 9, 10 and 11, except for
the chapters 2 and 5. Therefore, this is almost his
own book.

There are only 4 species alive in the world:
Limulus polyphemus Linnaeus 1758 (USA),
Tachypleus gigas Muller 1785 (South Asia, India),
T. tridentatus Leech 1819 (Southeast Aia, China,
Japan) and Carcinoscorpius rotundicauda Latreille
1802 (Southeast Asia, India).

It was generally assumed that Limulus and
Tachypleus first diverged at the Jurassic, and
Carcinoscorpius bifurcated at the Quaternary,

based on morphological evidence. However,
embryological investigations including artificial
hybridization and biochemical and molecular
biological approach have revealed that T. tridenta¬
tus is more closely related to C. rotunticauda than
to T. giga. This conclusion has led not only to a
new phylogeny of the living horseshoe crab but
also to a new taxonomy: all the three species
belong to the genus Tachypleus in place of
Carcinoscorpius.

Recently, an entirely new problem has arisen for
the survival of the horseshoe crab. It comes from
the use of horseshoe crab lysate as a reagent to
detect bacterial endotoxins. The Limulus lysate
test was developed in USA, where the Limulus
lysate is mass-produced, and its use has spread all
over the world. Biological knowledge of the
horseshoe crab is certainly required as the basis of
an attempt to save this living fossil from extinction.
This book is extremely useful from this point of
view.

This book is very fascinating, and most highly
recommended to all the biologists. The reviewer
sincerely extends “congratulations” to the Editor
for his monumental life work.

Koscak Maruyama
Chiba University
Chiba, Japan.

1342

ANNOUNCEMENT

The editors express their gratitude to the following reviewers, who
ZOOLOGICAL SCIENCE from September 1, 1987 to August 31, 1988.
ance the journal could not function.

Agui, N.

Ai, N.

Akai, H.
Aketa, K.
Amano, M.
Amemiya, S.
Ando, K.
Ando, M.
Aoki, J.

Aoki, K.

Aoto, T.

Arai, Y.
Asami, K.
Ashida, M.

Baba, K.

Chiba, Y.
Chino, H.

Ebara, S.
Egami, N.
Eguchi, E.
Eguchi, H.
Ehara, S.
Endo, K.

Fujii, Y.

Fujii, R.
Fujimoto, M.
Fujisawa, H.
Fujishima, M.

Gorbman, A.

Hamaguchi, S.
Hanaoka, Y.
Hara, T.
Hayashi, K.-I.
Hayashi, S.

Hidaka, T.
Hihara, F.
Hiramoto, Y.
Hirano, T.
Hisada, M.
Honma, Y.
Horiuchi, S.
Hoshi, M.
Hoshino, M.

Ichikawa, M.
Iga, T.

Ikebe, C.
Ishibashi, N.
Ishihara, K.
Ishii, S.
Ishikawa, H.
Ishizaki, FI.
Isono, N.

Ito, T.

Iuchi, I.
Iwamatsu, T.
Iwasawa, FI.

Kakizawa, Y.
Kambara, S.
Kamegai, S.
Kamishima, Y.
Katagiri, C.
Kato, K.

Kato, S.
Kawahara, A.
Kawanabe, H.
Kifune, T.
Kikuyama, S.
Kimura, T.
Kobayashi, K.
Kobayashi, Y.
Kohno, S.
Kohno, S.-I.

evaluated papers for
Without their assist-

Kojima, M.
Kondo, H.
Konishi, K.
Konishi, T.
Koshihara, H.
Kubota, T.
Kurokawa, H.
Kuwasawa, K.

Mabuchi, I.
Makino, N.
Maruyama, K.
Matsui, M.
Matsumura, S.
Michibata, H.
Miki-Noumura, T.
Milkman, K.
Mohri, H.

Mori, S.

Mori, T.
Moriwaki, K.
Moriya, T.
Muneoka, Y.
Murakami, A.

Nagahama, Y.
Naito, N.

Naito, T.
Nakauchi, M.
Nakazawa, T.
Namiki, H.
Nishioka, M.
Nishizuka, M.
Nomaguchi, T.
Noumura, T.
Numakunai, T.

Obinata, T.
Ogawa, M.

Oguro, C.

Ohoka, T.

Ohta, S.

Ohta, Y.
Ohtaki, T.
Oishi, T.

Oka, Y.

Okada, M.
Okada, T.

Okai, Y.
Okajima, A.
Omata, S.
Ooishi, S.
Ooka, H.

Ooka, S.

Oota, Y.
Osanai, K.
Osanai, M.
Oshima, M.

Sakagami, S. F.
Sakai, H.
Sakaizumi, M.
Sakurai, S.
Sasaji, H.
Sasaki, M.
Sasaki, N.

Sato, H.

Satoh, N.
Satou, M.
Sawada, H.
Sawada, I.
Sekiguchi, K.
Shibuya, T.

Shigei, M.
Shigenaka, Y.
Shimada, H.
Shimazu, T.
Shimozawa, A.
Shingyoji, C.
Shirai, H.
Sioda, M.
Sugiyama, K.
Suhama M.
Suzuki, M.
Suzuki, N.

Tachi, C.
Taguchi, Y.
Takahashi, H.
Takahashi, K.
Takahashi, S.
Takasugi, N.
Takeda, M.
Takei, Y.
Takeuchi, H.
Takeuchi, N.
Tamate, H.
Tanaka, S.
Tateda, H.
Terauchi, K.
Tochinai, S.
Toda, M.
Tomino, S.
Tomita, H.
Tsuchiya, K.
Tsuneki, K.

Uchida, T.
Uematsu, T.
Uemura, H.
Ueno, S.-I.
Urano, A.

Uwa, H.

Wada, M.
Washio, H.
Watanabe, T.
Watanabe, Y.

Yagi, K.

Yagi, S.

Yajima, H.
Yamada, K.
Yamada, M.
Yamagami, K.
Yamagishi, H.
Yamaguchi, T.
Yamamoto, M.
Yamamoto, T.
Yamaoka, I.
Yamasaki, T.
Yanagimachi, R
Yanagisawa, T.
Yasumasu, I.
Yoshida, M.
Yoshida, Y.
Yoshizaki, N.
Yoshizato, K.
Yoshizawa, T.

1344

ANNOUNCEMENTS

THE 60TH ANNUAL MEETING OF THE ZOOLOGICAL
SOCIETY OF JAPAN

The 60th Annual Meeting of the Zoological Society of Japan will be held at Kyoto Kaikan and
Kyoto-shi Dento Sangyo Kaikan from October 4 to 6, 1989. Further information and applica¬
tion forms will be sent to the domestic members in the April issue of ‘Biological Science News’
(No. 209). The deadline for application is July 15, 1989.

For application from abroad, please contact:

Professor Mikita Kato

Organizing Committee of the 60th Annual Meeting
of the Zoological Society of Japan
Zoological Institute, Faculty of Science,

Kyoto University,

Kitashirakawa-Oiwake-cho , Sakyo-ku ,

Kyoto 606, Japan.

Phone: 075-753-4087.

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 1988 was given to the following four papers.

Yoshiomi Takagi, Takako Suzuki and Chiaki Shimada: Isolation of a Paramecium tetraurelia
mutant with short clonal life-span and with novel life-cycle features. Vol. 4, No. 1, pp.
73-80, 1987.

Hisao Honda, Tadashi Oishi and Takao Konishi: Comparison of reproductive activities
between two Japanese quail lines selected with regard to photoperiodic gonadal response.
Vol. 4, No. 1, pp. 135-144, 1987.

Norio Suzuki, Masanori Kurita, Ken-ichi Yoshino, Hiroko Kajiura, Kohji Nomura and
Masaaki Yamaguchi: Purification and structure of Mosact and its derivatives from the egg
jelly of the sea urchin Clypeaster japonicus. Vol. 4, No. 4, pp. 649-656, 1987.

Shin-Ichi Abe and Shoji Asakura: Meiotic divisions and early-mid-spermiogenesis from
cultured primary spermatocytes of Xenopus laevis. Vol. 4, No. 5, pp. 839-847, 1987.

ACKNOWLEDGEMENT

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 5 and constant support for
Zoological Science Awards for volumes 1-5.

1345

AUTHOR INDEX

A

Abdulaev, N .

. 1202

Arnold, A. P .

. 1317

Abe, H .

. 1227

Artamonov, I. D .

. 1203, 1204

Abe, H .

. 1278

Asada, N .

. 915

Abe, K .

. 1292

Asada, S .

. 1281

Abe, M .

. 1231

Asakura, E .

. 1263

Abe, S .

. 1291

Asakura, S .

. 1291

Abe, T .

. 1242

Asami, K. .

. 1245

Adachi, T .

. 1278

Asao, T .

. 1294

Adachi, T .

. 1306

Asari, T .

. 1211

Agui, N .

. 1292

Asashima, M .

. 1283

Ai, H .

. 1196

Asha Devi, S .

. 1284

Ai, N .

. 1196, 1328

Ashida, M .

. 1242, 1243

Aida, K .

. 1303

Ashley, C. C .

. 1213

Aigaki, T .

. 1292

Atomi, Y .

. 1214

Aihara, M .

. 1210

Awaji, M .

. 1059

Aikawa, M .

. 1283

Azuma, K .

. 33

Aizawa, H .

. 1236

Azuma, K .

. 1207

Aizawa, M .

. 1327

Azuma, M .

. 1202

Akaike, M .

. 1222

Azuma, M .

. 1239

Akasaka, K .

. 1295, 1296

Azumi, K .

. 1242

Aketa, K .

. 1259

Akiyama, T .

. 1282

B

Amakaw, T .

. 1193

Amanai, K .

. 1242

Baba, S. A .

. 1217

Amano, T .

. 1259, 1260

Begg, D. A .

. 699

Amanuma, H .

. 1246

Bern, H. A .

. 267

Amasaki, H .

. 205

Bonder, E. M .

. 699

Amemiya, S .

. 85, 1293

Brawley, J. B .

. 603

Amemiya, Y .

. 1201

Bryan, J .

. 623

Amikura, R. M. . . ,

. 1275

Ando, H .

. 1298

c

Andreassen, J .

. 1221

Aoki, K .

... 1196, 1199, 1268

Caprio, J .

. 1194

Aoki, M .

. 1241

Cernichiari, E .

. 1248

Aoto, T .

. 833, 1308

Chiba, A .

. 1065

Aotsuka, T .

. 1239

Chiba, K .

. 1254

Arai, J .

. 1326

Chiba, Y .

. 1316, 1328

Arai, K .

. 1290

Chino, H .

. 1244

Arai, S .

. 1292

Clark, W. H. Jr .

. 573

Arai, Y .

... 1267,1318,1319

Clarkson, T. W .

. 1248

Arakawa, N .

. 1210

Coffe, G .

. 585

Araki, Y .

. 1289

Cotran, N. M .

. 699

1346

Fujimoto, Y. . .

. . . 1305

D

Fujino, Y. ......... _ _ _ _ _ _

..... 93,1265,1296

Dafgard, E . .

. . 133

Fujisawa, H. . . . . . . .

. . 85, 1332

Daigo, M .

............... 205

Fujisawa, T. _ _ _ ...... _

. . 1254

Dan, K . . . .

. . . 507

Fujisawa, Y . .

. . . 1212

Dan-Sohkawa, M. . .

.............. 1266

Fujishima, M. ........... _

. . 791, 1272

Daumae, M . . .

. . 855, 1330

Fujishita, S. . . . . . . . . .

......... 1200, 1202

David, E. ...........

............... 291

Fujisue, M. . . .

. . 1273

De Jesus, E. G. T. . .

.............. 1311

Fujita, S. . . . .

............... 1234

De Pomerai, D. I. . . .

................. 1

Fujiwara, A. . . .

... 1261,1262,1294

De Santis, R. .......

............... 919

Fujiwara, M. ......... _ _ _ _

.......... 397, 1033

Deno, T. _ .... _

.............. 1266

Fujiwara, S. . . . . .

............... 1275

Diaz, M. R. M. . . . . .

........ 1262, 1263

Fukada, Y . . .

. . . 1202, 1203, 1204

Dohke, K. . .

........ 1239, 1240

Fukamachi, H . .

. . . 1281

Dudgeon, D. ........

.............. 1224

Fukamizu, A. . . . . .

. . . 1247

Fukazawa, Y. ...............

. . 1314

E

Fukuda, H. . . . . . . .

............... 1251

Fukui, Y. . . . . .

. . 1269

Ebara, A .

. . . 1215

Fukunaga, S. . . . . . .

. . 1192

Ebina, Y. . .

.............. 1280

Fukushi, T . . . . .

. . 1206

Ebino, K. Y .

. . . 863

Funahashi, H. ...............

............... 1330

Ebitani, N .

. 183, 1282

Furukawa, Y. . .

............... 1215

Egami, N. . . . .

. 897, 1149

Furukohri, T . .

. . 69, 1232

Eguchi, E. . .

. . 1200

Furuta, E. ..... . . .

............... 1220

Eguchi, M. . .

. . 1239

Fusetani, N . . . ...

. . . 1253

El wing, H . . .

. . 959

Endo, K. . .

... 145,1051,1305

(1

Endo, S. . .

. 613, 1287

Ij

Endo, T. . . .

. 905

Cause, G. G. . .

. . . 727

Endo, T . . . . .

. 1228

Gibbons, I. R. ..............

. . . 1217

Endo, Y. . . .

. . 905,965

Gomi, T. . . . .

......... 1322, 1323

Engstrom, W. .

. . . 133

Gonzalez, M. A. R. .........

. . 1083

Enomoto, Y .

. . 1319

Goto, A. ....................

. . 1213, 1214

Ezure, T .

. . 1245

Gremigni, V. ... _ _ _ ...

............... 1153

Grundstrom, N. .............

. . . . 959

F

Falkmer, S . . .

. . 133

e

Fishkind, D. J . . . .

. 699

Haino-Fukushima, K. .........

......> . 1251

Fujii, K . . .

. . . . 1308

Hamaguch, S. . .

. 1290

Fujii, M . . .

. . 1294

Hamaguchi, Y. 545, 1219, 1262, 1285, 1286, 1287

Fujii, R .

. 1210, 1211

Hamanaka, T . . .

. . . 1201

Fujii, T .

. 1265, 1267

Hamazaki, T. S. . . .

. . . 1252

Fujii, Y . . .

. . . 1270

Hanada, T . . .

. . 1193

Fujikura, K .

. 415

Hanai, K. . . . . .

............. 1192

Fujimori, M . . . .

. 201, 1310

Hanaoka, Y. . . . .

. 1298, 1312

Fujimoto, M .

. . . 1207

Hanyu, I . .

. . 1059

Fujimoto, S .

. 1289

Hanzawa, N. . . . .

. . . 449

1347

Hara, E .

Hara, H .

Hara, M .

Hara, R .

Hara, T .

Harada, T .

Hara-Nishimura, I. .

Hardy, D. M .

Harris, P. J .

Harumi, T .

Hasegawa, H .

Hasegawa, K .

Hasegawa, M .

Hasegawa, S .

Hasegawa, T .

Hashimoto, N .

Hashimoto, N .

Hashimoto, T .

Hashimoto, Y .

Hasumi, M .

Hatanaka, K .

Hato, M .

Hatoh, M .

Hatta, H .

Hattori, A .

Hattori, K .

Hattori, S .

Hayakawa, T .

Hayakawa, Y .

Hayama, T .

Hayashi, H .

Hayashi, S .

Hayashi, T .

Hedrick, J. L .

Henson, J. H .

Hidaka, M .

Hidaka, T .

Higaki, Y .

Higashinakagawa, T.

Higashitani, A .

Hihara, F .

Hikida, T .

Hino, A .

Hinsch, G. W .

Hirabayashi, T .

Hirakawa, M .

Hiramoto, Y .

Hirano, K .

. 1237

. 1328

. 1321

. 1201

. 1201

. 1210

. 1201

. 1278

. 639

. 1258

. 1269

. 1306

. 901

. 1311

. 1290

. 1276

. 1277

. 1322, 1323

. 1270

. 1316

. 1274

. 1236

. 1296

. 1214

. 1303

. 1301

. 1315

. 1270

. 1241, 1244

. 1194

1296, 1298, 1312

. 1330

. 1121, 1338

. 1278

. 699

. 1219, 1331

. 947, 1195

. 1268

. 1296, 1298

. 1246

. 1325

. 901

. 1260

. 573

. 1232, 1265

. 1241

. 539, 1285

. 1262

Hirano, S .

Hirano, T .

Hirano-Ohnishi

Hirao, Y .

Hiraoka, T. . . .

Hirata, J .

Hirayama, A. .
Hirohama, T. .
Hironaka, N. .
Hirono, M.

Hiruta, S .

Hisada, M .

Hisatomi, O. ..
Hiyoshi, H. ...

Hoda, Y .

Hokano, M. . .

Honda, K .

Honma, Y .

Hori, I .

Hori, K .

Horikami, H. .
Horiuchi, S. . . .
Horiuchi, S. . . .

Hoshi, M .

Hoshide, K. . . .
Hoshino, Z. . . .
Hosokawa, K.

Hosoya, H .

Hosoya, M. . . .

Hosoya, N. _

Hotta, Y .

Howard, C .

Hu, D. H .

Hung, F .

Hyodo, S .

Hyoto-Taguchi,

. 1247

. 1297, 1299, 1311, 1312

,J . 1231

. 1277

. 1244

. 1219

. 1089

. 833, 1308

. 999

. 1231

. 1336

. 1189, 1197, 1198

. 1202

. 1291

. 1241

. 1321

. 291

. 1065

. 1256

. 1215

. 1327

. 1202, 1267

. 1234, 1270

1250, 1253, 1254, 1259, 1260, 1338

. 791, 1333

. 103

. 1253

. 1226, 1287, 1288

. 1265

. 1236, 1287

. 1246

. 585

. 1229

. 809

. 397, 1033, 1299, 1300

Y . 1322, 1231

I

Ichikawa, M

Ide, H .

Iga, T .

Igarashi, J. .
Igata, H. . . .
Iguchi, R. . .
Iguchi, T. . .
Ihara, S. ...
lino, T .

. 1300

1268, 1269

. 1211

. 1317

. 1195

. 1331

1314, 1315

. 1256

. 1240

1348

Iizuka, Y .

. 1293

Isomura, M .

. 179

Ikawa, Y .

. 1246

Isono, N .

. . 1339

Ikeda, T .

. 1212

Itayama, T .

. 1194

Ikegami, S .

. 1295

Ito, A .

. 1221

Illg, P L .

. 1336

Ito, K .

1246, 1267, 1282

Imada, M .

. 973

Ito, M. . .

. . 1246

Imae, Y .

. 1218

Ito, S .

. 1249

Imahori, K .

. 1277

Itogawa, K .

. 1216

Imakado, K .

. 1291

Itoh, K .

. 1247

Imamoto, Y .

. 1203, 1204

Itoh, M .

. 1232

Inaba, K .

. 1235, 1236

Itoh, M .

. 1316, 1330

Inaba, M .

. 1141

Itoh, N. .

. . 1295

Inamura, H .

. 1324

Itoh, T .

. 1208

Inoda, T .

. 939, 1218

Iuchi, I .

. 191,

1230, 1252, 1281

Inoue, C .

. 1291

Iwahori, N .

. 1283

Inoue, M .

. 1249

Iwamatsu, T .

. 353,

1248, 1249, 1314

Inoue, S .

. 1249

Iwami, M .

. 1306

Inoue, S .

. 415

Iwamoto, H .

. 1213

Inoue, S .

. 529

Iwamuro, S. .......

. 1296

Inoue, S .

. 291

Iwao, Y .

. 1149, 1278

Inoue, T .

. 1232

Iwasa, T .

. 1202

Inoue, Y .

. 1249

Iwasaki, S .

. 61,331

Inouye, M .

. 1284

Iwasawa, H. 1007,

1073, 1289,

1314, 1316, 1338

Inui, Y .

. 1311

Iwashita, S .

. 1277

Irisawa, S .

. 1315

Iwatsuki, K .

. 1326

Ishibashi, T .

. 1208

Izumi, A .

. 1285

Ishida, H .

, 179,973,1212,1213,1319

Izumi, S .

. 1243, 1244

Ishida, M .

. 1245

Ishiguro, M .

. 733

I

Ishihara, H .

. 1321

J

Ishii, K .

. 1320, 1327

Jaana, H .

. 1324

Ishii, N .

. 1213

Jaffe, L. A .

. 1254, 1278

Ishii, S .

. 1255

Johnson, K. A .

. . 1236

Ishii, S .

. 1291, 1298, 1316, 1330

Ishijima, S .

. 1218

K

Ishijima, T .

. 1310

Ishikawa, H .

. 1190, 1237, 1238, 1296

Kabasawa, H .

. 431,437,1329

Ishikawa, K .

. 1254

Kageura, H .

. 1273

Ishikawa, M .

. 1202

Kaji, K .

1221, 1272, 1284

Ishikawa, S .

. 1195

Kajikawa, H .

. 1258, 1326

Ishimoda-Takagi, T.

. 1232

Kajiwara, S .

. 103

Ishine, M .

. 979, 1288

Kajiwara, Y .

. 1284

Ishiura, M .

. 1254

Kamata, K .

. 1141

Ishizaka, S .

. 1288

Kamata, K .

. 1286

Ishizaki, H .

. 1306, 1307

Kamata, Y .

. 1295

Ishizaki, T .

. 1200

Kambegawa, A .

. 1297, 1309

Ishizaki, Y .

. 1240

Kameda, T .

. 1257

Ishizuka, H .

. 1043

Kamiguchi, Y .

. 1222

1349

Kamimura, J .

Kamimura, S .

Kamishima, Y. . . .

Kamizato, Y .

Kan, S .

Kanai, C .

Kanamori, A .

Kanda, T .

Kandori, H .

Kaneko, H .

Kaneko, M .

Kano, H .

Kano, Y .

Kanou, M .

Kariya, M .

Karlsson, J.-O. G.

Kasai, Y .

Kasuga, H .

Kasuga, H .

Kasukawa, H .

Kasuya, Y .

Katada, J .

Katada, T .

Katagiri, C .

Katagiri, N .

Katagiri, Y .

Katakura, Y .

Kataoka, K .

Katayama, H .

Katayama, J .

Kato, H .

Kato, K. H .

Kato, M .

Kato, Y .

Kato, Y .

Katoh, K .

Katoh, S .

Katou, Y .

Katow, H .

Katsube, Y .

Katsura, S .

Kawada, J .

Kawahara, Y .

Kawai, G .

Kawai, T .

Kawakami, A.
Kawakatsu, M. . . .
Kawamoto, K. ...

. 1292

. 1237

. 1219

. 1332

. 1284

. 1252, 1255

. 1299, 1313, 1314

. 1212

. 1203, 1204

. 1266

. 1318

. 1222

. 1319

. 1209

. 1282

. 959

. 1277

. 1276

. 1238, 1292

. 1210

. 1199

. 1217

. 1254

1244, 1267, 1276, 1279

. 1200

. 1200

. 1307

. 1294

. 1211

. 1207

. 1290

. 1280

. 1248

. 1222

. 1253

. 1217

. 1240

. 1298

. 53, 1258

. 1244

. 1043

. 1043

. 1266

. 1236

. 1333

. 1306

. 1334

. 1019

Kawamura, K . 1257

Kawamura, K . 677, 1286

Kawamura, K . 1313

Kawamura, S . 801

Kawashima, S . 911, 1019, 1284, 1300, 1301

Kawashima, T . 1281

Kawata, A . 1203

Kawauchi, H . 1297, 1302

Kelly, K. M . 267

Khin, Maung, Saing . 781

Kido, 1 . 1326

Kihara, A . 1320, 1327

Kihara, H. K . 1235

Kikuchi, K . 1212

Kikuchi, M . 1330

Kikuchi, S . 1256

Kikuchi, S . 1320

Kikuchi, Y . 1322

Kikuta, A . 1224

Kikuta, T . 1324

Kikuyama, S . 407, 1296

1297, 1298, 1308, 1312, 1313, 1316

Kimura, A . 1322, 1323

Kimura, K . 1321

Kimura, M . 1306

Kimura, S . 1228

Kimura, T . 767,1196,1197

Kimura, T . 855, 1330

Kimura, Y . 1325

Kinoshita, H . 1272

Kinoshita, S . 431

Kinoshita, T . 1270

Kiriishi, S . 1304

Kisamori, T . 69

Kishi, K . 1322, 1323

Kishida, M . 1299

Kishimoto, T . 1254, 1277

Kita, T . 1195

Kitajima, K . 1249

Kilo, K . 1334

Kito, Y . 1201

Kiyokawa, S . 1336

Kiyomoto, M . 1293

Kobayakawa, Y . 1255

Kobayashi, A . 1225

Kobayashi, H . 1308

Kobayashi, K . 61, 331

Kobayashi, K . 1234, 1270

1350

Kobayashi, M. .
Kobayashi, M. .
Kobayashi, S.
Kobayashi, S.
Kobayashi, T. ...
Kobayashi, T. . . ,
Kobayashi, T. . . .
Kobayashi, T. . . .
Kobayashi, T. . . .
Kobayashi, W. . .
Kobayashi, Y. . .
Kobayashi, Y. . .
Koenuma, A.

Koga, K .

Koga, T .

Kohama, K .

Kohda, Y .

Koizumi, O .

Koizumi, S .

Koizumi, T .

Kojima, M. K.

Kojima, S .

Kojima, T .

Komatsu, M .

Kominami, T. . . .
Komiyama, E.
Komukai, M.

Kondo, H .

Kondo, N .

Kondo, S .

Kondo, Y .

Kondoh, H .

Kondoh, Y .

Konishi, K .

Konishi, K .

Konno, S .

Kosaka, T .

Koshida, Y .

Koshida, Y .

Koshimizu, I.

Kotani, S .

Koyama, E .

Koyama, H .

Kozono, S .

Kristensen, R. M

Kubo, T .

Kubokawa, K.
Kubokawa, R. . .

, 179, 1212, 1215

. 1303

. 1274

. 1272

. 1214

. 1231

. 1283

. 1312

. 1314

. 1250

. 911, 1301

. 1304

. 1305

. 375

. 1192

. 1229

. 501

. 1255

. 1272

. 1309

. 645, 1261

. 1203

. 1227

. 1337

. 1294

. 1329

. 1293

. 1283

. 1215

. 1321

. 1318

. 1246

. 1197

. 1231

. 463, 1336

. 1222

. 139, 1137

. 1246, 1267

. 1338

1013, 1302, 1304

. 1236

. 1266

. 1232

. 397

. 875

. 183, 1282

1291, 1298, 1330
. 1321

Kudo, S .

Kudo, Y .

Kudoh, M. . . .
Kumada, N. .
Kumagai, K. .
Kunitomo, Y.

Kuno, S .

Kurabuchi, S.
Kurahashi, H.
Kurahashi, T.
Kurita, M.
Kuriyama, R.
Kuroda, H.
Kuroda, M. . .
Kurokawa, M.
Kurosumi, K.
Kusaka, H.
Kusunoki, S. .
Kuwana, T. . .
Kuwasawa, K.
Kuwata, O.
Kwon, Oh-Yu.
Kyozuka, K. .

. 1249

. 1218

. 1326

. 1220

145, 1051, 1305

. 1238

. 1293

. 415, 1302

. 1305

. 1194

. 1258

. 1286

. 733, 1261

1212, 1213, 1319

. 1195

. 1301

. 1194

. 1325, 1326

. 1263, 1284

1195, 1214, 1215

. 1202, 1203

. 1238

. 1277

L

Lasnitzki, 1 . 999

Lin, J. T . 1206

Lindstrom, M . 743

Longo, F . 573

Lukoschus, F. S . 883

M

Mabuchi, I .

Maeda, Y .

Maekawa, S .

Maeno, N .

Makabe, K. W .

Makino, N .

Makioka, T .

Marchese-Ragona, S.

Maruo, F .

Maruyama, K .

Maruyama, M .

Maruyama, S .

Masaki, T .

Masaoka, Y .

691, 1225, 1226, 1227, 1288

. 1331

. 1226

. 1211

. 1266, 1275, 1295

. 1324, 1325, 1332, 1334

. 1335

P . 1236

. 1274

. 347, 1228, 1229, 1342

. 1195

. 1336

. 145, 1051, 1305

. 1333

1351

Masho, R .

Masuda, H .

Masuda, Y .

Masuyama, E .

Masuyama, Y .

Matsuda, J .

Matsuda, K .

Matsuhisa, A .

Matsui, M .

Matsui, N .

Matsui, S .

Matsui, T .

Matsumoto, A .

Matsumoto, J .

Matsumoto, T .

Matsumura, K .

Matsumura, S .

Matsunaga, T .

Matsuno, A .

Matsuno, T .

Matsuoka, N .

Matsusaka, T .

Matsushima, T .

Matsushita, S .

Matsuura, T .

Matsuzaki, T .

Mazda, E .

Mazia, D .

McLachlan, J. A .

Meguro, N .

Menrai, N .

Meserve, L. A .

Meusy, J.-J .

Meyer-Rochow, V. B.

Micevych, P. E .

Michibata, H .

Michinomae, M .

Mikamo, K .

Mikami, K .

Mikami-Takei, K .

Miki-Noumura, T.

Mimura, K .

Mimura, T .

Mishima, N .

Mita, M .

Mita, T .

Mitani, H .

Mitchell, R. W .

. 1273

. 1201

. 1212, 1279

. 1237

. 1233

. 1264

. 1297

. 1314

901, 1121, 1338, 1339

. 1301

. 1201

. 1306

. 1317

. 1282

. 1145

. 1259

. 1215

. 1235

. . . . 1212, 1228, 1319

. 1213

. 1337

. 1319, 1333

. 1199

. 1264

. 1228

. 1220

. 1295, 1296

. 519

. 1281

. 205

. 1250

. 1083

. 217

. 743

. 1317

. 1219

. 1201

. 1222

. 1272

. 1259

. 563, 1237

. 1206

. 1213

. 1331

.... 1253, 1258, 1313

. 1295

. 1225

. 1334

Mitsui, H . 1290

Mitsunaga, K . 93, 1293, 1296

Mitsuno, T . 1295

Miura, K . 1244

Miura, T . 1313

Miwa, S . 1311

Miyagawa, A . 1263

Miyagawa, J . 1264

Miyaji, K . 1210

Miyakawa, H . 1243

Miyakawa, M . 1319

Miyashita, Y . rr^~. . 1329

Miyata, K . 61

Miyata, S . 1235

Miyata, T . 1202

Miyawaki, M . 1307

Miyazaki, H . 1326

Miyazaki, J . 1232

Miyazaki, K . 1335

Miyoshi, K . 1322

Mizobe, Y . 1272

Mizoguchi, A . 1307

Mizoguchi, H . 1256

Mizokami, A . 1295

Mizumoto, H . 1255

Mizunami, M . 1206

Mizuno, T . 385, 823, 999, 1235, 1264

Mizuta, C . 1331

Mizutani, A . 1205

Mochii, M . 1250

Mochizuki, F . 1196

Mogami, Y . 1217

Mohri, H . 1165, 1235, 1236, 1237, 1287

Mohri, T . 1262

Molan, A. L . 483

Mori, H . 1201

Mori, T . 1252, 1292

Moribayashi, A . 1305

Morioka, M . 1245

Morisawa, M . 939, 1218

Morishita, F . 1211

Morishita, H . 1197

Morita, H . 1192

Morita, T . 1246, 1267, 1282

Moriwaki, K . 1223, 1224, 1290

Moriya, T . 1329

Motobayashi, Y . 1256

Motokawa, T . 1337

1352

Motono, M. .

. . 1307

Mukai, M. ......

. . 1309

Munekata, E. . . . ,

. . . 1241

Muneoka, Y .

.................. 1212

Murakami, K. . . . ,

. . . 1247

Murakami, M. . .

. . . 801

Murakami, M. . . ,

. . . 1281, 1283

Murakami, S .

. . 1318

Murakami, T .

. . . 1310

Muramatsu, R. . .

. . . 69

Muramoto, A. . . .

. . . 1198

Muramoto, N. . . .

. . . 1332

Muraoka, A .

. . 1213

Murase, M . . .

.................. 1337

Murata, K. .......

.................. 1252

Murata, M . . .

. . . 1307

Murayama, T. . . .

. . . 1228

Murofushi, H. . . .

. . . . 1236

Murofushi, K. . . . .

N

.................. 1294

Nadamitsu, S. . . .

. . 1282

Nagahama, Y. . .

1299,

...... 1250, 1253, 1254

1302, 1313, 1314, 1316

Nagai, K . . .

.................. 1316

Nagai, T. .......

. . . . 1193

Nagaishi, H. . . . .

. . 1210, 1211

Nagao, E. . . .

. . . . 1256

Nagasawa, H. . . .

. 1306. 1307

Nagata, S. .

. . . 77,493

Naito, N. .

. . . 1302

Naitoh, T. ......

. . . . 759

Naitoh, Y. .

. . 1216, 1217

Naka, K. ....... .

. . 1198

Nakagawa, H. . .

. . . 1197

Nakai, Y. .......

. . 1302

Nakajima, T. . . .

.................. 1241

Nakajima, Y. . . .

. . 1323

Nakamura, H. . .

. . . 1260

Nakamura, J. . . .

. . . . 1231

Nakamura, K. . .

. . 1232, 1237

Nakamura, M. . .

. . 1141, 1290

Nakamura, M. . .

. . . 1246

Nakamura, S. . . .

. . . . 833, 1308

Nakamura, S. . . .

. . . . 1261

Nakamura, T. . . .

. 187

Nakamura, T. . . .

. . . . . 1319

Nakamura, Y. .....

Nakanishi, T. ......

Nakanishi, Y. ......

Nakano, H. . .

Nakano, Y. .... _

Nakashima, H.
Nakasone, Y. ......

Nakata, K. ..... _

Nakatani, I. ........

Nakato, H .

Nakauchi, M. ......

Nakauchi, Y. .

Nakayama, S. _ _ _

Nakazawa, T .

Nambu, Z. .........

Nambu- Akiyama , F.

Namiki, H. .

Naruse, I. . . .

N anise, K. .........

Naruse, M. .

Naruse, M. _ _

Negishi, S. . .

Nemoto, S. ........

Niida, A. ..........

Niinuma, K. .......

Niki, T. . . .

Nilson, H. . . .

Ninomiya, H. ......

Ninomiya, K. ......

Nishi, H. .......... .

Nishi, T. ...........

Nishida, M. . . . .

Nishihira, M. . .

Nishikata, T. ... _

Nishikawa, A.
Nishikawa, K. ......

Nishikawa, M.
Nishimura, M.
Nishio, H. .........

Nishioka, R. S.

Nishitani, Y . .

Nishiyama, I. ......

Nishizawa, Y. ......

Nishizuka, M. ......

Noda, K. . . .

Noda, Y. .......... .

Nogawa, H. . . .

Noguchi, M. . .

........... 1216

........... 1250

. . 1270

........... 1331

. . 539

. . 1219

...... 165, 1105

............ 883

........... 1329

. . 1243

........... 1257

. 1228

........... 1247

..... 1253, 1281

........... 1235

........... 1235

1282, 1325, 1326
........... 1285

...... 489,1225

........... 1308

........... 1329

. . 951

........... 1276

........... 1208

. 1298, 1312

. . 1243

............ 743

........... 1321

........... 1330

........... 1211

............ 179

........... 1043

. . 1331

. . . . . 1275, 1294

..... 1265, 1313

........... 1265

........... 1208

........... 1201

. . 1330

............ 267

. . . 1267

. . 1267

........... 1266

..... 1318, 1319

. . . . . 1252, 1255

........... 1253

. . 1270

........... 1216

1353

Noguchi, M .

Noguchi, O .

Noji, S .

Nomaguchi, T. A

Nomura, K .

Nomura, K .

Nonoyama, K. . .

Nose, T .

Noumura, T .

Noyori, M .

Numachi, K .

Numakunai, T. . .

Numata, H .

Numata, O .

Nunomura, N. . .

. 1290

. 1319, 1333

. 1266

. 1230, 1283

1258, 1279, 1280

. 1300

. 1269

. 1298

. 1318

. 1291

. 449

. 103

. 1305

119, 1230, 1271
. 1336

O

Oami, K . 1216

Obara, Y . 815,1223

Obata, M . 1137

Obata, S . 733

Obika, M . 311

Obinata, T . 1227, 1228, 1229, 1265

Ochiai, M . 1242, 1243

Ogasawara,T . 1309,1311,1312

Ogawa, H . 1246, 1247

Ogawa, K . 1320

Ogawa, M . 1309

Ogino, K . 1238

Ogino, K . 1305

Ogiso, M . 1247

Oguro, C . 201, 1297, 1309, 1310, 1337

Ohashi, K . 347, 1228

Ohba, E . 1310

Ohba, S . 1237

Ohgushi, R . 869

Ohinata, H . 1267

Ohkawa, M . 1313

Ohmori, A . 1195

Ohmuro, H . 1228, 1229

Ohnishi, E . 1304

Ohnishi, K . 1199

Ohnishi, K . 1231

Ohno, T . 1324

Ohnuma, M . 691, 1226

Ohoka, T . 1245

Ohotoh, K . 1315

Ohoyama, H. .
Ohshima, S. . .

Ohsugi, K .

Ohsumi, K. ...

Ohta, I .

Ohta, K .

Ohta, K .

Ohta, K .

Ohta, M .

Ohta, N .

Ohta, T .

Ohtaka, C .

Ohtake, H. ...

Ohtaki, T .

Ohtsu, J .

Ohtsu, K .

Ohtsuka, E. . . .
Ohtsuki, H.
Ohyama, T. . . .

Oikawa, T .

Oishi, T .

Oka, Y .

Okada, J .

Okada, M .

Okada, M .

Okada, Y .

Okamoto, M. .

Okano, T .

Okano, Y .

Oki, I .

Okinaga, S. . . .
Okinaga, T. . . .

Okita, Y .

Okubo, J .

Okuma, J .

Okumura, H. .

Okuno, M .

Omata, S .

Onitake, K. . . .

Ono, T .

Onozato, H.

Ooishi, S .

Ooiwa, K .

Ooka, H .

Ooka, S .

Ooka-Souda, S

Ookata, K .

Oota, I .

. 1285

. 1227

1268, 1269, 1339

. 1276

. 1260

. 1192

. 613, 1287

. 1286

. 1247

. 1308

. 353, 1249

. 1238

. 939

1242, 1268, 1305

. 1219

. 1207

. 1255

. 1323

. 1208

. 1278

. 1203, 1329

. 1199, 1300

. 1328

1191, 1274, 1341

. 1304

. 1197

. 337

. 1203, 1204

. 291

. 1334

. 1290

. 1259

. 1259, 1260

. 43

. 1197

. 1333

. 1218, 1237

. 1248

1221, 1248, 1278

. 1223

. 1250

. 1336

. 1214

. 1302, 1303

. 1329

. 431,437

. 1222

. 1212

1354

Oota, Y .

... 1013,1302,1304

Saigusa, M .

. 1327

Osanai, K .

. 1277

Saito, K .

. 1241

Osanai, M .

. 1238, 1292

Saito, M. .

. . 1250

Osawa, S .

. 1262, 1263

Saito, T .

. 1207

Osawa, S .

. 1263

Saito, T. R .

. 863, 1137, 1141

Oshima, E .

. 353

Saitoh, M .

. 815

Oshima, N .

. 1210, 1211

Saiki, T. . .

. . . . . 1285

Oshima, Y .

. 1316

Sakagami, S. F. ...

. 869

Oshimi, A .

. 1314

Sakaguchi, B .

. . . 375

Oshimi, Y .

. 1214

Sakai, H .

. . . 505, 613, 1226, 1236, 1287

Ota, H. ..........

. . . 901

Sakai, H .

. 1299

Ota, Y .

. 1234

Sakai, M .

. 1273, 1276

Otokawa, M .

. 1195

Sakai, M .

. 1298, 1312

Otto, J. J .

. 713

Sakai, N .

. 353, 1314

Ouji, M .

. 847

Sakai, Y .

. 1303

Owada, K .

. 1338

Sakairi, K .

. 1288

Ozaki, K .

. 1200, 1201

Sakaizumi, M .

. 1223, 1224

Ozaki, M .

.... 281, 1193, 1201

Sakamoto, T .

. 1311

Ozawa, S .

. 1315

Sakurai, H .

. 1219

Sakurai, S .

. 1224, 1239

Sakurai, S .

. 1242, 1304

P

Sakurai, T .

. 1323

Palazzo, R. E .

. 603

Sakurai, Y .

. 1230

Pandey, S. C .

. . . . 153

Sano, K .

. 1227

Pandey, S. D .

. 153

Saotome, K .

. 1222

Payen, G. G .

. 217

Sasaki, F .

. 1270

Petrow, V .

. 1315

Sasaki, M .

. 1205

Pinto, M. R .

. 919

Sasaki, T .

. 1230, 1283, 1308, 1322

Poulson, D. F .

. 375

Sasaki, Y .

. 1332

Poung-In, S .

. 1331

Sasayama, Y .

. 201, 1297, 1309, 1310

Sato, H .

. 623, 733, 1225

Sato, M .

. 397, 1033

Q

Sato, M .

. . . 1198

Quintana, R .

. 463

Sato, M .

. 1244

Sato, M .

. 1299, 1315

Sato, N .

. . . 1212

R

Sato, S .

. 1238

Rani, L .

. 893

Satoh, K .

. 1279

Rebhun, L. I .

. 603

Satoh, M .

. 1265

Riihimaa, A .

. 1209

Satoh, N .

. 1266, 1275, 1294, 1295

Ross, M. H .

. 1268

Satoh, T .

. 1196

Ryuzaki, M .

. 1248

Satou, M .

. 1199

Sawada, H .

. 1239

Sawada, H .

. 1250, 1253

S

Sawada, I .

. 483

Sadakane, K .

. 1208

Sawada, T . . .

. 667

Saga, T .

. 1304

Sawada, Y .

. 1194

Saiga, H .

. 1295

Sawai, T .

. 685

1355

Schatten, G .

Schatten, H .

Schroeder, T. E.

Seidou, M .

Seki, T .

Seki, T .

Sekiguchi, S .

Sekizawa, Y .

Sendai, M .

Sendai, Y .

Senoo, M .

Sensui, N .

Seo, H .

Seo, M.-S .

Seo, N .

Serizawa, O,

Shaoyi, Y .

Shibahara, S.

Shibata, M .

Shibata, M .

Shibata, N .

Shibuya, T .

Shichida, Y .

Shigei, M .

Shigenaka, Y. . . .

Shima, A .

Shimada, A .

Shimada, H .

Shimada, I .

Shimada, K .

Shimada, K .

Shimada, Y .

Shimamoto, K. . .

Shimaoka, S .

Shimatani, Y. . . .

Shimidzu, T .

Shimizu, H .

Shimizu, I .

Shimizu, K .

Shimizu, K .

Shimizu, T .

Shimizu, T .

Shimohigashi, M.
Shimozawa, A. . .

Shimura, Y .

Shinagawa, A. . .

Shinbo, K .

Shingyoji, C .

. 585

. 585

. 713

. 1201

. 407, 1267

. 1200, 1202

. 1285

. 1308

. 1235

. 1259

. 1301

. 1276

. 1301

. 1247

. 1325

. 1260

. 809

. 1246

. 1228

. 1245

. 1290

... 1193, 1194, 1196
... 1202,1203,1204

. 1337

. 179,973

489, 897, 1220, 1225

. 489, 897, 1225

... 1245,1295,1296

. 1328

. 1209

. 1283

. 1284

. 1249

. 1229

. 1200

. 1258

. 1325

... 1206,1244,1306

. 1233

. 1322

. 1236

. 1292

. 1205

. 1220

. 1203

. 1273

. 1227

. 1217

Shinkai, T .

Shinkawa, Y .

Shinozawa, T .

Shiojiri, N .

Shirai, H .

Shiraishi, A .

Shiraishi, S .

Shirakawa, T .

Shirane, T .

Shirasawa, M. ......

Shirasawa, N .

Shirasawa, Y .

Shirayama, Y .

Shishikura, F .

Shitara, M .

Shoji, R .

Shozawa, A .

Simerly, C .

Simpson, A. W. M.
Sivasubramanian, P.

Sluder, G .

Someno, T .

Srivastav, A. K .

Srivastav, S. P .

Strohman, R. C.

Sudzuki, M .

Suematsu, N .

Sueoka, T .

Sugai, T .

Suganuma, Y .

Sugi, H .

Sugiyama, K .

Sugiyama, S .

Sumi, Y .

Sumida, H .

Sundgren, H .

Suto, C .

Sutoh, K .

Suyemitsu, T .

Suzuki, A .

Suzuki, A. S .

Suzuki, H .

Suzuki, H .

Suzuki, J .

Suzuki, K .

Suzuki, M .

Suzuki, N .

Suzuki, N .

. 1302, 1303

. 1328

. 1192

. 1266

. 1291

. 1192

. 1252

. 1252

. 1275

. 1324

. 1325

. 1324, 1334

. 875

. 1242

. 1251

. 1285

. 1220

. 585

. 1213

. 497

. 653

. 1253

. 197,893

. 197

. 1214, 1230

. 1335

. 385,999

. 1240

. 119

. 323

. 1213, 1214

. 1221

. 1274

. 1321

. 1319

. 959

. 1220

. 1226

. 1289

. 1306, 1307

. 1263, 1264

. 1337

. 1223, 1224

. 1266

. 1302

. 407

1258, 1279, 1280, 1314
. 1297, 1310

1356

Suzuki, N. ...............

Suzuki, R . .

Suzuki, S . . —

Suzuki, S . .

Suzuki, T. . . .

Suzuki, T. ................

Suzuki, T. ... _ .........

Suzuki, T. ................

Suzuki, Y. ...............

Suzuki, Y. . . . . .

Suzuki, Y. ...............

T

Tabata, S . .

Tabuchi, M . .

Tachi, C. . .

Tachibana, K. . . .

Tada, M. .................

Tadenuma, H. ..........

Tagawa, M . .

Tagawa, T. ..............

Tagi, Y. . .

Taguchi, K. .............

Taguchi, M . .

Tahara, U. ........... —

Taka, Y. ................

Takabatake, I. ...... -

Takabatake, T. . .

Takada, M. . .

Takagi, I. ... — . . .

Takagi, T. ..............

Takagi, Y. ..............

Takagi, Y. . . . .

Takagi-Sawada, M. _ _

Takahama, H. ..........

Takahashi, H. _ .......

Takahashi, K. ...........

Takahashi, K. . .

Takahashi, K. W. .......

Takahashi, M. _ _ _ ....

Takahashi, M . .

Takahashi, M. M .

Takahashi, N. ...........

Takahashi, N. .... _ ...

Takahashi, S .

Takahashi, S . .

Takahashi, S .

........... 1194 Takahashi, S. Y. ............... 1239, 1304, 1316

. — ...... 1297 Takahashi, T. C. ..................... 1262, 1263

........... 1214 Takahashi, Y. _ _ ... 1321

........... 1307 Takahata, M. ........ _ .................. 1198

....... 69, 1232 Takano-Ohmuro, H. ................. 1229, 1288

........... 1200 Takasugi, N. ......................... 1314, 1315

........... 1239 Takata, K. ........................... 1262, 1263

. . . . . 1303, 1321 Takeda, H. ............................. 385,999

........... 1306 Takeda, S. ................................. 1239

........... 1316 Takeda, Y. ................................ . 1333

........... 1326 Takei, K. .................................. 1199

Takei, T. . . . . 1214

Takei, Y. .............................. 43, 1309

Takeichi, M. ............................... 1202

........... 1246 Takemasa, T. ............. _ .... _ ..... 1231

............ 415 Takemoto, K. ......................... 733, 1261

........... 1221 Takeshima, K. . . 1262, 1263

. . 1254 Taketomi, Y. .............................. 1307

........... 1247 Takeuchi, I. ............................... . 1277

........... 1254 Takeuchi, K. ............................... . 759

........... 1312 Takeuchi, S. ............................... 1257

........... 1320 Takeuchi, T. ......................... 1246, 1267

. . 1241 Takeuchi, Y. . . 1277

. . . . . . 1262 Takezawa, T. . . 1299, 1300

. 1270, 1285 Takiguchi-Hayashi, K . . 1264

............ 109 Takimor i, J. . . 1303

........... 1239 Takiya, S. . . . 1306

. . 759,1212 Takizawa, M. . . 1330

........... 1263 Takizawa, S. ...................... _ ..... 1250

........... 1312 Tamaki, F. . . . 1288

. . . . . 1230, 1271 Tamura, K. ................................ 1269

. . . . . 1226, 1231 Tamura, S. ................................ . 1334

. . . . . 1220, 1246 Tanaka, A. ................................ 1268

........... 1272 Tanaka, H. ................................ 1269

........... 1253 Tanaka, H. .......................... 1279, 1280

........... 1270 Tanaka, K. ................................ . 1214

. 1073, 1338 Tanaka, K. . . 1242

........... 1217 Tanaka, M. . . 1312

. . 1311 Tanaka, R. . . . 1231

. 139,863,1137 Tanaka, S. ......... 1007,1298,1301,1302,1312

........... 1212 Tanaka, S. . . 1237

. . 1198 Tanaka, S. ...................... . 1256

. . . . . 1302, 1303 Tanaka, S. . . 1325

. . 1212 Taneda, K. . . . 1210, 1216

1262, 1263, 1296 Taneda, Y. ................................ . 1331

. 1229 Taniguchi, N. . . 449

. 1247 Taniguchi, S. . . . 1266

. 1300, 1301 Taniguchi, Y. .............................. 1202

1357

Taniguchi, Y .

. 1214

Tsuyuki, S .

. 1224

Taniguchi, Y .

. 1217

Tukahara, J .

. 1293

Tanimoto, S .

. 1218

Tanimura, A .

. 1289

U

Tanimura, T .

. 1192

Taoka, M .

. 1333

Uchida, D .

. 1297

Tateda, H .

. 1206

Uchida, T .

. 1252, 1292

Tateno, H .

. 1222

Uchida, T .

. 1254

Tazaki, K .

. 299

Uchida, Y .

. 1339

Tazawa, E . . .

. 1261

Uchikawa, K .

. 883

Tazawa, H .

. 1309

Uchiyama, H .

. 823

Terakita, A .

1200, 1201

Uchiyama, M .

. 1310

Terao, O .

. 1149

Uda, H .

. 1325

Teshirogi, W .

. 1252

Ueda, H .

. 139

Tochinai, S .

. 1267

Ueda, H .

1289, 1313, 1314

Toda, M. J .

. 1095

Ueda, K .

. 1199

Toga, T .

. 1198

Ueda, M .

. 1232

Toh, Y .

. 1205,

1206, 1208

Ueda, T .

. 1296

Tokunaga, F .

. 1202

Uemura, H .

. 833, 1308

Tominaga, Y .

1205, 1208

Uemura, I .

. 1286

Tominaga, T .

. 1216

Uesaka, H .

. 1215

Tomino, S .

1243, 1244

Uesugi, Y .

. 1315

Tomioka, K .

1316, 1328

Ueta, N .

1253, 1258, 1313

Tomita, H .

. 1225

Umebachi, Y .

. 1240, 1241

Tomiyama, H .

. 1248

Uno, S .

. 1291

Tonegawa, Y .

. 1292

Urano, A .

... 397,925,1033,1299,1300

Tonosaki, A .

. 1202

Usui, N .

. 1251

Toribara, T. Y .

. 1248

Usukura, J .

. 1204

Toriyama, M .

. 613,

1226, 1287

Uto, N .

. 1260

Toriyama, M .

. 1244

Utsugi, K .

. 1335

Tosuji, H .

. 1253

Uwa, H .

. 1224

Toyoizumi, R .

. 1257

Toyoshima, Y. Y .

. 1236

\ j

Tozawa, K .

. 1245

V

Tsuchimori, N .

. 1295

Vale, R. D .

. 1236

Tsuchiya, K .

1223, 1224

Verrell, P. A .

. 159

Tsuchiya, T .

. 1214

Tsuchiyama-Omura, S .

. 375

W

Tsuda, M .

. 1202

Tsukii, Y .

. 1320

Wada, M .

. 1317

Tsukita, S .

. 1288

Wada, S .

. 1237

Tsukuda, H .

. 1209

Wakabayashi, K. . . .

. 1007, 1298

Tsuneki, K .

. 21,847,1338

Wakabayashi, S. . . .

. 1299

Tsunemoto, M .

. 119

Wakabashi, K .

. 1201

Tsunenari, T .

. 1317

Wakana, S .

. 1223, 1224

Tsuneoka, M .

. 347

Wakasugi, C .

. 1310

Tsusue, M .

1239, 1240

Washida, S .

. 1261

Tsuyoshi, H .

. 1317

Washio, H .

. 1195

1358

Watanabe,A . 1193

Watanabe, C . 1290

Watanabe, H . 1289

Watanabe, M . 501

Watanabe, N . 1246

Watanabe, T . 1309

Watanabe, T . 1231

Watanabe, Y . 1251

Watanabe, Y . 119, 1230, 1231, 1271

Watanabe, Y . 1303

Weldon, P. J . 443

Win, Win, Yee . 1301

X

Xia, Z. W . 1095

Y

Yada,T . 1311

Yagi, S . 1292

Yamaai, T . 1266

Yamada, C . 1299, 1300

Yamada, J . 1219, 1220

Yamada, K . 1211

Yamada, K . 1295, 1296

Yamada, M . 1207

Yamada, S . 1214, 1230

Yamada, T . 1285

Yamagami, K . 191, 1230, 1252, 1281, 1322

Yamagata, N . 1221

Yamagishi, H . 1215

Yamaguchi, K . 43

Yamaguchi, K . 1220

Yamaguchi, M . 1258, 1279

Yamaguchi, S . 1325

Yamaguchi, T . 767, 1196, 1197, 1208

Yamaguchi, T . 1331

Yamaguchi, Y . 1273

Yamamoto, H . 323

Yamamoto, H . 1229

Yamamoto, H . 1246, 1267, 1271

Yamamoto, K . 1276

Yamamoto, K . 1297, 1298

Yamamoto, M . 979, 989, 1288

Yamamoto, N . 1289

Yamamoto, T. S . 1250

Yamamoto, Y . 1233

Yamamoto, Y . 1316

Yamanaka, M . 1288

Yamane, S . 869

Yamanobe,T . 1213

Yamanouchi, H . 1330

Yamanouchi, K . 1318

Yamao, W . 563

Yamasaki, H . 1279

Yamasaki, K . 1243, 1307

Yamase, K . 1300

Yamashiki, N . 1286

Yamashiro, H . 1331

Yamshita, M . 1250

Yamashita, S . 1204, 1205

Yamashita, T . 1019

Yamauchi, K . 1313

Yamazaki, H. 1 . 1243

Yamazaki, K . 1321

Yamazaki, S . 1205

Yamazato, K . 1332

Yanagi, A . 1271

Yanase, H . 1207

Yano, K . 1244

Yano, Y . 1252

Yashima, Y . 1271

Yashiro, T . 1303

Yasuda, A . 1297

Yasuda, K . 1203

Yasugi, S . 1235, 1264

Yasuhara, T . 1241

Yasui, T . 1241

Yasumasu, 1 . 93, 1233, 1259,

1261, 1262, 1293, 1294, 1295, 1296

Yasumasu, S . 191, 1280

Yasutomi, M . 1271

Yasuyama, K . 767,1196,1197

Yazawa, Y . 1233,1234

Yodono, M . 1228

Yokohari, F . 1208

Yokosawa, H . 1242

Yokota, E . 1225

Yokota, Y . 1280

Yokouchi. Y . 1268

Yokoya, S . 1280

Yokoyama, S . 1236

Yoneda, M . 553

Yonehana, T . 1209

Yonei, S . 1248

1359

Yonemura, S .

. 1288

Yoshino, H .

. 1327

Yonezawa, S .

. 1234

Yoshino, K .

. 1258

Yonezawa, Y .

. 1283

Yoshioka, E .

. 1327

Yoshida, A .

. 1268

Yoshizaki, N .

. 1279

Yoshida, M .

. 979, 1289

Yoshizato, K .

... 1233, 1265, 1313

Yoshida, N .

. 1278

Yoshizawa, H .

. 1310

Yoshida, T .

. 1298, 1315

Yoshizawa, T .

... 1202,1203,1204

Yoshihama, K .

. 1298

Yukiyama, S .

. 947

Yoshihara, K .

. 1201

Yussof, S .

. 1334

Yoshihara, M .

. 1310

Yoshii, M .

. 1316

7

Yoshikuni, M .

. 1254

Li

Yoshimoto, Y .

. 1249

Zama, N .

. 53

Yoshimura, K .

. 1335

Yoshinaga, K .

. 863

■

Published by

the Japanese Society of Developmental Biologists

gbevefopmentr

The journal is devoted to the publication of original papers dealing with any aspects of developmental
phenomena in all kinds of organisms, including plants and micro-organisms. Papers in any of the following fields
will be considered: developmental genetics, growth, morphogenesis, cellular kinetics, fertilization, cell division,
dormancy, germination, metamorphosis, regeneration and pathogenesis, at the biochemical, biophysical and
analytically morphological levels ; reports on techniques applicable to the above fields. At times reviews on subjects
selected by the editors will be published. Brief complete papers will be accepted, but not preliminary reports.

Members of the Society receive the Journal free of charge. Subscription by institutions is also welcome.

Papers in Vol. 30, No. 6. (December, 1988)

REVIEW: S. Miyazaki: Fertilization potential and calcium transients in mammalian eggs.

V-W. P. Lin, H. B. Kwon, T. R. Petrino and A. W. Schuetz: Studies on the mechanism of action of estradiol
in regulating follicular progesterone levels: Effects on cAMP mediated events and 3/?-hydroxysteroid
dehydrogenase.

J. S. Wolenski and N. H. Hart: Sperm incorporation independent of fertilization cone formation in the danio

egg-

M. Myohara and M. Okada: Activation of heat-shock genes by digitonin is selectively repressed in preheated
Drosophila salivary glands.

K. Katoh, M. Yokoyama, S. Kimura, Y. Hiramoto and H. Kondoh: Analysis of cellular mosaicism in a
transgenic mouse by histological in situ hybridization.

N. Yoshimatsu and R. Yanagimachi: Effects of cations and other medium components on the zona-induced
acrosome reaction of hamster spermatozoa.

B. J. Crawford: Regional ultrastructural differences in basal lamenae isolated from the strafish Disaster
ochraceus.

A. Matsuno and M. Ouji: Ultrastructural studies on development of the tail gland of a cuttlefish. Sepiella
japonica.

A. Tanimura and H. Iwasawa: Ultrastructural observations on the orgin and differentiation of somatic cells
during gonadal development in the frog Rana nigromaculata.

K. Kawamura, H.. Fujita and M. Nakauchi: Helper function of follicle cells in sperm-egg interactions of the
ascidian, Ciona intestinalis.

E. K. Boon-Niermeyer, A. M. de Waal, J. E. M. Souren and R. van Wijk: Heat-indiced changes in thermo¬
sensitivity and gene expression during development.

H. Nakamura, S. Takagi, T. Tsuji, K. A. Matsui and H. Fujisawa: The prosencephalon has the capacity to
differentiate into the optic tectum: Analysis by chick-specific monoclonal antibodies in quail-chick-chimeric
brains.

U. Ulelkind and B. J. Crawford: Influence of donor age on establishing well-differentiated clonal cultures
from embryonic chicken retinal pigmented epithelium.

Development, Growth and Differentiation (ISSN 0012-1592) is published bimonthly by The Japanese Society of
Developmental Biologists, Department of Biology, School of Education, Waseda University, Tokyo 160, Japan.
1988: Volume 30. Annual subscription U. S. $ 110.00 including air speed delivery except Japan. Application to
mail at second class postage rate is pending at Jamaica, NY 11431, U. S. A.

Outside Japan : Send subscription orders and notices of change of address to Academic Press, Inc., Journal
Subscription Fulfillment Department, 6277 Sea Harbor Drive, Orlando, FL 32887, 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, 6277 Sea Harbor Drive, Orlando, FL 32887, 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
Refrigerator

Extra-Quiet

Operation

Ease of Loading/
Unloading
The Rotors

Quick Start/
Quick Stop

TOiviy

High Quality

Triple Safety
Design

Corrosion

Resistance

HIGH SPEED
REFRIGERATED
MICRO CENTRIFUGE

TOIVIY

SOLE AGENT

1002 SOLEIL NARiMASU BLDG.. 31-8. NARIMASU 1-CHOME,
ITABASHI-KU. TOKYO 175 JAPAN

TEL:(03)976-341 1 TLX. 027231 1 1 TOMYCO J

CABLE: TOMYSHO TOKYO FAX: (Gill G II ) (03)930-70 1 O

TOMY SEIKO CO., LTD

MANUFACTURER

2-2-12. ASAHICHO NERIMA-KU. TOKYO 176 JAPAN
TEL:(03)976-3 1 1 1

NARISHIGE

THE ULTIMATE NAME IN MICROMANIPULATION

OUR NEW MODELS MO-102 and MO-103
MAKE PRECISION MICROMANIPULATION SO EASY!

(Photo: by courtesy of Olympus Optical CO., LTD.)

SOME FEATURES of MO-102 and MO-103:

* The manipulator head is so small that it can be mounted directly on the
microscope stage. There is no need for a bulky stand.

* Hydraulic remote control ensures totally vibration-free operation.

* 3-D movements achieved with a single joystick.

MicromanSpulators Microelectrode pullers Stereotaxic instruments

r—TZ—TTTL \

NARISHIGE SCIENTIFIC INSTRUMENT
LABORATORY CO., LTD.

4-9-28, Kasuya, Setagaya-ku, Tokyo 157 JAPAN
Telephone: 03-308-8233 Telex: NARISHG J27781

ZOOLOGICAL SCIENCE

VOLUME 5 NUMBER 6

DECEMBER 1988

CONTENTS ,

REVIEWS

Gremigni, V.: Planarian regeneration: an overview of some cellular mechanisms . . 1153

Mohri, H. and N. Hosoya: Two decades since the naming of tubulin — the multi-facets of tubulin

. . . . . 1165

PROCEEDINGS OF THE 59TH ANNUAL MEETING OF THE ZOOLOGICAL SOCIETY OF JAPAN

Abstracts of papers read by the Zoological Society Prize winners

Hisada, M. : Structure and function of nonspiking interneurons ......................... 1189

Ishikawa, H.: The intracellular symbiont of insects as a genetic element . . 1190

Okada, M.: Developmental and molecular analyses of ooplasmic factors required for

pole cell formation _ ................ . . . 1191

Abstracts of papers presented at the 59th Annual Meeting of the Zoological Society of Japan

Physiology . . . . . . . . 1192

Genetics . . . . . . . . ...... 1222

Biochemistry . . . . . . . . 1225

Developmental Biology . . . . . . . 1248

Endocrinology ........................... — . . . 1296

Morphology . . . . . ........................... _ ....... _ .... 1319

Behavior Biology . . . . . . . . . . . 1326

Ecology . . . . . . . . . . . . ........................ 1331

Taxonomy and Systematics ....... . . . . 1333

Book reviews . . . . . . . . . . . 1340

Announcements _ _ ............... . . . . . . . .... 1342

Author index . . . . . . . 1345

Contents of ZOOLOGICAL SCIENCE, Vol. 5, Nos. 1-6 . . . . . i

INDEXED IN: Issued on December 15

Current Conlerits/LS and AB & ES, Printed by Daigaku Printing Co., Ltd.,

Science Citation Index , Hiroshima, Japan

ISI Online Database,

CABS Database, INFOBIB

NVINOSHlItAJS S3 I BVB 8 11 LIBRARIES SMITHSONIAN INSTITUTION NOlIfUIJLSNI NVINOSHill

w z *5 •>?r § > S

ES SMITHSONIAN _ INSTITUTION NOIinillSNI_ NVINOSHillMS00 S3 I a VB 8 Ilf LI B RAR I ES^SMITHSONI/

y//. Q LlI ^CTsTIT/7>v ?? ii I >!”Tc.tit77^v. Z

jNI NVINOSH1IIAIS S3l8Vaan LIBRARIES SMITHSONIAN INSTITUTION NOlinillSN! NVINOSH1M
r- v, Z „ Z / — ^ 5 <“ ^ ^ Z r-

m 2 2 m rn

ES^SMITHSONIAN "INSTITUTION NOU.niU.SNI~NVINOSHi.HMS S3 I a Vd a n~LI B RAR I ES^SMITHSONU

,rt Z CO Z _ _ CO z ... CO

«r ? «r ><v4SOaXs. <f \X 5*

^ U) 2 ,*r' 2

Nl NVINOSHlIIMS^SaiaVaan LIBRARIES SMITHSONIAN INSTITUTION^NOlinillSNrNVINOSHilll

~ CO 9 _ CO ~ _ CO —

CO

O \ — x^UiiSt^ CJ _ -v/a O

E S^ SMITHSONIAN^ INSTITUTION _ NOliniliSNI^NVINOSHillAIS^Sa IBVdan^LIBRARI ES^ SMITHSON 1/

m

3NI NVINOSHillMS S3 I a V a a 11 ~L I B R A R I E S SMITHSONIAN'lNSTITUTION^NOUniliSNl'NVINOSHlIV

ES^SMITHSONIAN lNSTITUTION^NOlinillSNI NVINOSHillMS^Sa I a Va 8 II^LI B RAR I ES^SMITHSONIA
w — _ <s> _ ^ > to _ = <n

SNI NVINOSHillMS S3iavaai1 LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHim

' m z. rn g m — xgmtfBy m

i es, smithsonian'institution NoiinnisNi'NviNOSHiiiMs^sa lava aiuiB rar i es^smithsonia

— CO Z _ CO Z .... CO

Jr i ^ ^

UVd a n2U B RAR I ES(/jSMITHSONIANJNSTlTUTlONt/>NOIifUliSNrNVINOSHllWS</1S3 I BVd 9 IT ~L I B R/

1/A -*■ * -X _ - — 2 » (/) 7 m ->■

2 XT' > 5 xv > S __

jtuTION MOUniliSM^NVINOSHilWS S3 ldVaan_ LIBRARIES SMITHSONIAN INSTITUTION ^NOIlflj

BVd a n B RAR I ES ^SMITHSONIAN^INSTITUTION NOIlfULLSNI NVINOSH1IWS ^3 I d Vd B 11 ^ L I B HA

2 H r-

v> - ±Z - w 2 Niam2S^ m -v - xi2i!iSx m

itution w NoiiniiiSNi nwnoshiiwsmS3 I a va a n l i b r ar i es"smitosonian “institution " Noun.

LT>

m

I

^ yj 42LSS-V > -<»■ ^ X § 5 ^

IVaan2UBRARIES SMITHSONIAN^NSTITUTION^NOlinillSNI^NVINOSHlIWS^SStavaail^UBRA

2 °° 01 _ _ ZZ CO —

O X^vqx/ — o

•7 ' ' 2 ~ _ _ _ ^ „ _

^UTlON2NOUfUllSNI^NVI_NOSHi.lWS2;S3iavaan^LIBRARIES2SMITHSONIAN^INSTITUTION2NOU.n.

m • vt> ^ m — ^ —

ban LIBRARI ES SMITHSONIAN INSTITUTION NOlinillSNI ~ NVINOSHlIWS^SS I B V B 8 IT ~ L I B R A

1 — ^ . 2 * CO 2 cn 2:

< - 2 i s 1

A Z -1 dwfct,.. 2 /Skt*£X =i 2

V.

rUTION NOIJ.nilXSNI_NVjNOSHJ.IWS S3 I d Vd 8 IT LIBRARI ES^SMITHSONIAN INSTITUTION ^NOlin.

— " " ~ </> — 10

vaan libraries SMITHSONIAN institution NouniiiSNi nvinoshjiws saiavaan^LiBR/

— r z r* 2: r* 2 <-

_ ™ 2 Nguusgx m

iiniiiSNi nvinoshjiws S3 1 a va a 1 i“l i b r a r i e s ^smithsonian “institu Noun

~ 5” 2 C ft -y?

— &****&*> — x ^ '* = Amv\ — 5 ,w

&/ ^ 2 v*W\\X “4 /O^fWX^X 2