4 aA ' ‘ ' Vr any (
? ’ eal ' : inbyansinaa ena
2 ae iat ‘ 1 avin '3:038 4)
A 1 vat nyd \ strait Wl sta
f ' % +4} Teiveaie4
; . vend ‘ tea MERRILY ESE] Vie antate 1
H ‘ pote a ‘ Lal fg ao a bse atath 6
i : 1 Lb A hemang i Pear , ty cn Gee cibed
‘ : park et ey glee aa fa gate kane
: an 1 ‘ vee yt nad MOA HER hes Say
i stabi essai ie titpataesee ina teal sit ti it
: wes , ' Wiis ety asd i Lea ar
- ‘ hs Ph) ty plosten? its tee hw
‘ bee Paha hat ta Ua eee 0 Tee CaN
: an a rata! iki wea
4 F ; ahmh tart a Fe cal ees ht Daal reper
, s 4 ovis ‘ yop ean a Cae ee ot
rk . ’ - , ar ia : af me eter 8
: . ok Age MA ‘ Aon che Goan
: i ; : Peeing Myatt of ' ,
: he tnan . eM a? are <a on TEEN Taty a oveticinaehe nas
2 poeta roar ie 5 ‘a A bg ME LI! Vary fet
4 A wa i Peta Ms SSI rae ri oe valli atte tenon A AA tes Netas Vole
. onan ae : | a { Corre bates ef ng Aen Sea
1 ae ¢ : : De itary got te Fesgug yatta iat
‘ A Yr en Vg not mat tea Wiettietae ptatie be Vaudnshaibyenehd Subleanme tome 9 Ral
areata whe ” Fc a bei ta a WRAP AE e's stale Be astray Re Bah oh
Aros : ‘ fiat Spaie : ' ! 4 Perm eenrcs
i art oe Rae ota Oags rei al ACHING N Uy AEA!
ays ‘a ba rar q ans
: : \ ‘ wale ' Pee ee AN PORN ICON BL He
¥ < : ene wah yo ee he NO Paced meetea an Fe HEN
‘ Fi Petia renew aes Fact WN
. . ae ' ey ek soe ered oA lie ted
; ee . is ag hungry : D Men et
: ney aa oy ea g Feit AML N the aca ote
tay ‘ Ahh, age (runt a F we errs
4 Pi : : Sela ta rah eae ae eC tu'8iys saad
: Ae aa a rab yta cil eiel
3 Macytat Geared Grrr. i Yard een en in
re ‘ A oe RY, SSCS
: - , “ stitqin Neh tte tre
4 a : astas ‘ Ronen Come nrtaterres
é Iie ie aaa . an QRS SN MMMM Tmt aie
R ‘ ‘ tytn pave Fuk :
; ' ; Faden ens A da tained asaera WATE KE
nee es eice wee ths Bath Feaadiy Vemal LL
; 2 ast way . SeEiamte S
Pune Bi ea oa be instant ark
i te Re ret? ay skis ithe a det
Oh a aa deine
oa oa , ie i? g Diemer ha ea
‘ Dba Ay wees
eS on 24 * ener
; : : pared : an at runt the ben
toy Soa Peg yh teas De RENEE Kar het
. any ' ny LI tei :
. S ames te sae aan sitar ed) a nitty
nee ey a aay? AA AGAON great Neti ete ae
3 . : : co Phineas a vince

pap
pupnrta

Revag

eh
on
San radcecriraiiea ase axe

Teoreciigaere ve ath

inverter va
Nae payee

cate
Sitig bacapeurtise

nee rqeté
Powweradiest
eeipect nyt dot any
gooniwen

Wee gas ue
eT an ant

neainey ag

fires
yet iba eee

wut t 9
od i BP Ae

desea Cerda ayes a
raursiaaest

Satis ISRO REY MEM A
presi paot tap sie we Ip" ace wed
Celia

Fara ip
He
Pepi lonlne 4a

eye Ling
ds bassinets

fsa
G99

Seren)

Uni eon peda
evista

nh Mee
aN Ye ilk Dee

ae pele Hanae
Creer ear on
yrange paren
eink MH WIE =e 5

aero
lytic, GR
aussie au
PAF DRS an reeds
oe tee OP EPR?

Pie
bs

pe woshll gies
Sh Nts ie eye

aber Ba 9
Or NIST are EAT

hate
sie NEM ROA

{ Eien ane et
Pe espe area Fd Wak NaH peaks
se Eaae Ni SUG pte Rey TF APL ACHR ee

i) ler Pr 2 AW Be Liter
asin i TY
s apenite phi ws ss tenarenye 9
RAL plow abe 07 Be EERIE
Wer Drennan a eons
alratase ner gai a CeNOrE ae aCih)
Rp adtnist ga esty teal puaytht

14 a
net a
ainiecris UPA Ih:
a

Sete
ert

has SS Ym % a
“2 Mea) = Ray 2 WY & Cy 2 Wp EGY 2
Po z S E a) 49 — NYS - <p Le) =— , GS he ie =
Coe wn Sasi m SN = Wasi’ rm NOSH = a
* ij is -e INS S3IYVUaT
NS Sa1NVNgIT LIBRARI FE EON ey eee ee 2 ,
w : za a 4 ee z =
= = 4 = = ¥ = i = =
Se 2
e 8 NOG a. ©: 2 :
= = ~ > = a > = >
i: IAN INSTITUTIO
MOREILITION p NOILNLILSNI_NVINOSHLINS Sal uWud ee BRARI ES | SMITHSON x
Zz u ,
é = & ! & " Wy, ?®
= : 5 < = < Gf
ye — oc + < <x a x Y; Y ds, c
i oat | <x fee s NN : By S ow for =
a = FZ i) I z ay ,
oa | NVINOSHLINS S3IY¥VUET
NS SJINVYGIT LIBRARIES SMITHSONIAN INST UTBTIOORE MOI Sense) a
- S a = AS re)
S a o ) o <a
° oo = be = = \\) eS oF
Ee a =a = > 2 WS 2
3 > — > = > e
- a a es E = =
D m 2 m 2 He z
: (29) = 4)
5 s ie SMITHSONIAN
AN” INSTITUTION pe enema as cane lie or 2
eG = 2 = ‘S AS = z
\ oe \\) Ne 4 Zz = S WN. : x 5 ;
$ NQais Se fe) a
3) 2 Ue B = g 2 NN g :
eu. s Z = 3 cine 2 ;
i ¢ | f Saluvuagl
197 NOILNLILSNI NVINOSHLIWS
NS/, S31 ¥V¥GIT_ LIBRARIES, SMITHSONIAN _ INSTITUTION a : o
us = N WS —
g = “3 2 Z ANS z
x S = ” c a Sis \. 3
o = ie = a = :
: 2 ~ ‘ 2 :: = HSONIAN™ INSTITUTIO
= . ¥ 1ES~ SMITHS )
AN INSTITUTION | NOILMLILSNI_NVINOSHLINS _S3/4VUGI7 LIBRAR : : .
= = : = o — L wo
z = S \ = 0 5 Gy 2
sad *, ae *
, E A WY = | wD = Y, id
5 , !
ie i ~ NS is = o" :
a ; er at: te a SHLINS S3INVYaI-
WS. LIBRARIES SMITHSONIAN ng _ NVINOSH “ '
2 ae 2 Zz z < =
> a > = me
a z : = z a 2 =|
\ z D NN \ o B a an uw
rs w” SANS O 2 o x 9
/ 2 = " NS Z E 2 iS Z|
5 = WS > = >" = >
> S S |
is LILSNI- ai NIAN. INSTITUTION
AN INSTITUTION NOLLALILSNI_NVINOSHLIWS LIBRARIES, SMITHSO “
. E Z = = = 2
Aa < 2 < a < =
3 . a = a =
ea m =| ea) 5 a 5
2 4 : e Fi - * saluvual
31 = NVINOSHLIWS
NS S3IMVYSIT LIBRARIES SMITHSONIAN INSTITUTION NOLLALILSNI &
© ow = Maes = cs =
be a > a = 2 =,
=) > = a - cad =
E - E %3 E -
m m z nf Z m Z
Ww aa
AN INSTITUTION NOILNLILSNI NVINOSHLINS, S31YVYGIT_ LIBRARIES, g
= ~ 2 = = .< NS = <
a ae 2 : 2 WS 5 é
A \ X; = f Me S xr : fe) : 4 aS a x 9
Oo Ly \ = fe) = 5 VX z 3
x S Gy, iy me z = ys = =
= > Ge = > = >

CfEO we" 2 NO 2 c Ca ae ee
s A BS ae 2 S s = Fg
RARIES SMITHSONIAN INSTITUTION = NOILALILSNI_ NVINOSHLIWS _LIBRARIES SMITHSONIAN INSTITUTION | NOILALILSNI_ NWINOSHLIL
JON us = eh Zz ae za os 2m us zZ
OS = n 2 n os ” ie oO = a
\ Ue ea o ot (acd au fac’ fem o a
al oo <x = pats ral <x aed = c
| Se S oc = o ‘S ox S a =
6&9 + xs = = = ac oO —

INOS ae (@) ee ‘Ss os (@) a (e) ere? Oo oes
5 Z = Zz Z af 7 — = = =
MLILSNI NVINOSHLINS S3I1YVHYEIT LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLINS S31YVYSIT LIBRARIES SMITHSON
= z S 2 i z= ie : z < = =

SEAR = 2) wo = «@ = ise] 2) @ S
ic ee = es 5 a > ae S > S

) > = > E 3 > = > = > =
ay = = a i 2 = a i= Es Fe
mane m 2) m @ m n = wo = =
7) = 7) = ty = ” = Y om
SMITHSONIAN NOILNLILSNI NVINOSHLINS S3IYVUSIT_LIBRARIES NOILALILSNI
G2 z ee) z a z feat 2) z wo
= < = ae = - . AS = < = :
= z = “ ae We 4 3 F
2 é 2 of fy 3 2 Gif i NS F 5 1 fy
G A L ¥ - st O° i, 4 WK Dao 2 0 GY
3 3 ae EE 2 ad 2 b ee

SMITHSONI/

fy
“iy

NVINOSHLIV

NLILSNI

NVINOSHLINS S3IYVYSIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSN! NVINOSHLINS S3IYVYSIT LIBRARIES

HSA

n4ise>

>

2AR1IES SMITHSONIAN

39 NG

NOLLNLILSNI
NOILALILSNI
LIBRARIES
NOILALILSNI

NOILNLILSNI
$f

INSTITUTION NOILNLILSNI INSTITUTION NOILNILILSNI

4,
7
hed
17 LIBRARIES

NVINOSHLINS S3Ji¥VUsIT LIBRARIES

&y
INOSASS

NLILSNI NVINOSHLINS SJIYVYEIT LIBRARIES

we

NVINOSHLINS S31¥Vudl1 LIBRARIES

SAIYVUGIT LIBRARIES
INSTITUTION NOILMLILSNI

INSTITUTION
Sasiuvyagii
INSTITUTION

INSTITUTION

INSTITUTION
NN
\
saluvua

SMITHSONIAN INSTITUTION SMITHSONI/

NVINOSHLINS SJ!IYVYSIT LIBRARIES SMITHSONIAN

NVINOSHLINS S3INVUSIT LIBRARIES
Lae NS

NYINOSHLINS
Nu
SMITHSONIAN

NVINOSHLIWS
SMITHSONIAN

NVINOSHLIWS
SMITHSONIAN
f

LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_ NVINOSHLII

Wa Ad

“MSM
RARIES SMITHSONIAN INSTITUTION NOILNLILSNI_NVINOSHLIWS

SMITHSONIAN

JIN

THR) — | bn,

RIES SMITHSONIAN

IES
RIES
ILSNI
ILSNI
RIES
ILSNI
RIES
TLSNI

ISNI

ZOOLOGICAL
SCIENCE

An International Journal

VOLUME 7
1990

published by

The Zoological Society of Japan

*

ni wol TéronRmSi nA

" AMUGOQV
oe

vo hervildug

io viewod In vigologt galt

CONTENTS OF VOLUME 7
NUMBER 1, FEBRUARY 1990

REVIEWS
Gotoh, T. and T. Suzuki: Molecular assembly
and evolution of multi-subunit extracellular

annelid hemoglobins ......................... 1
Lawrence, J.M.: The effect of stress and
disturbance on echinoderms................. 17

ORIGINAL PAPERS
Cell Biology and Morphology

Low, W. P., Y.K.Ip and D.J.W. Lane: A
comparative study of the gill morphology in
the mudskippers - Periophthalmus chrysospi-
los, Boleophthalmus boddaerti and Perioph-
thalmodon schlosseri ..............000..0005. 29

Kondo, H., Y. Yonezawa and T. A. Noma-
guchi: Difference in migratory ability be-
tween human lung and skin fibroblasts ...... 39

Immunology

Kono, H., A. Mizoguchi, H. Nagasawa, H.
Ishizaki, H. Fugo and A. Suzuki: A
monoclonal antibody against a synthetic car-
boxyl-terminal fragment of the eclosion hor-
mone of the silkworm, Bombyx mori: char-
acterization and application to immunohis-
tochemistry and affinity chromatography .... 47

Biochemistry
Michibata, H., T. Uyama and J. Hirata:
Vanadium-containing blood cells (vanado-
cytes) show no fluorescence due to the
tunichrome in the ascidian, Ascidia sydnei-
CHSISISGINIEG *..: REARAAT INA So sic lees see le ks: 55

Developmental Biology
Ma, Y. K. andS. B. Ramaswamy: Histochemis-
try of yolk formation in the ovaries of the
tarnished plant bug, Lygus lineolaris (Palisot
de Beauvois) (Hemiptera: Miridae) (COM-
MUNICATION) Fee eaten aerate. se 147
Kamishima, Y.: Organization and develop-
ment of reflecting platelets in iridophores of

the giant clam, Tridacna crocea Lamarck

Ishikawa, T.: Effects of puromycin and a-
amanitin on the activity of alkaline phosphat-
ase in early preimplantation mouse embryos
(COMMUNICATION) ee eee 153

Uchiyama, M., T. Murakami and H. Yoshiza-
wa: Notes on the development of the crab-
eating frog, Rana cancrivora ................ 73

Reproductive Biology
Asada, N. and T. Fukumitsu: Reaction mass
formation in Drosophila, with notes on a
phenoloxidase activation .................... 79
Jarosz,S.J. and W.R.Dukelow: Embryo
transfer and pregnancy rate in the golden
hamster (Mesocricetus auratus).............. 85

Endocrinology

Tagawa, M., S. Miwa, Y. Inui, E. G. de Jesus
and T. Hirano: Changes in thyroid hor-
mone concentrations during early develop-
ment and metamorphosis of the flounder,
Paralichthys olivaceus ............0...0..000- 93

Tasaki, Y. and S. Ishii: Effects of thyroidec-
tomy, hypophysectomy, temperature and
humidity on the occurrence of nocturnal
locomotor activity in th toad, Bufo japoni-
cus, during the breeding season............. 97

Uchiyama, M. and T. Murakami: Effects of
hypophysectomy and replacement therapy
with several hormones on plasma sodium
concentrations in bullfrog tadpoles......... 105

Taxonomy and Systematics
Abé, H.: Two species of the genus Actacarus
(Acari, Halacaridae) from Japan........... 111
Yunxia, T. and Y.Shaoyi: Comparative
study on LDH isozymes in different sub-
family of teleost fish-grass carp (Cteno-
phryngodon idellus) and blunt snout-bream
(Megalobrama amblycephala) .............. 127

il

Watabe, H., X.C. Liang and W. X. Zhang:
The Drosophila robusta species-group (Dip-
tera: Drosophilidae) from Yunnan Pro-
vince, southern China, with the revision of
its geographic distribution.................. 133

Ando, A., S. Shiraishi and T. A. Uchida:

Reexamination on the taxonomic position of
two intraspecific taxa in Japanese Eothe-
nomys: evidence from crossbreeding experi-
ments (Mammalia: Rodentia) .............. 141
Instructions to authors ......................- 159

NUMBER 2, APRIL 1990

REVIEWS
Yasugi, S. and T. Mizuno: Mesenchymal-
epithelial interactions in the organogenesis

Of digestive: tract tn ea eee reece 159
Tiedemann, H.: Cellular and molecular
aspects of embryonic induction............. 171

ORIGINAL PAPERS
Physiology

Ip, Y. K., S. F. Chew and R. W. L. Lim:
Ammoniagenesis in the mudskipper,
Periophthalmus chrysospilos...............- 187

Niida, A. and T. Ohono: Tectal visual affer-
ents from fish dorsolateral tegmental cells
(COMMUNICATION) ..................-. 327

Mizutani, A. and Y. Toh: Morphological
and physiological characterization of the
para-ocellar nerve of the cockroach, Peri-
planeta americana (COMMUNICATION)

Cell Biology
Fu, Y., K. Sato, K. Hosokawa and K. Shioka-
wa: Expression of circular plasmids which
contain bacterial chloramphenicol acetyl-
transferase gene connected to the promoter
of polypeptide IX gene of human adenovirus
type 12 in oocytes, eggs and embryos of
IX CHOPUS LACVIS 5s or0.s..acisie) Aurela esa avore erties 195
Kusakabe, T.: Ultrastructural studies of the
carotid labyrinth in the newt, Cynops pyr-
rhogaster.i:A058.. RG. PRS ox ee 201

Genetics
Niwa, M. and N. Wakasugi: Abnormal
development of preimplantation embryos
derived from intersubspecific hybrids be-

tween Mus musculus molossinus and M. m.
LO ANG) ee ee Cos 555 6005- 209

Immunology
Saad, A. H. and E. Cooper: Evidence for a
Thy-1-like molecule expressed on earth-
worm leucocytes ................-..ee eee eee 217

Developmental Biology
Sivasubramanian, P. and D. R. Nassel: Neu-
ral control of flight muscle differentiation in
the fly, Sarcophaga bullata ................. 223
Tonegawa, Y., E. Hojiro and K. Takahashi:
Effect of pH on the participation of calcium
ion in the cell aggregation of sea urchin

Reproductive Biology

Tachi, C. and S. Tachi: Mechanisms under-
lying regulation of local immune responses in
the uterus during early gestation of eutherian
mammals. III. Possible functional dif-
ferentiation of macrophages cultured
together with blastocyst in vitro, with special
tefrence to the cellular shape and produc-
tion of leukotriene Cy..................00- 235

Endocrinology
Tasaki, Y. and S. Ishii: Effects of thyroxine
on locomotor activity and carbon dioxide
release in the toad, Bufo japonicus......... 249
Yamada, C., S. Noji, S. Shioda, Y. Nakai and
H. Kobayashi: Intragranular colocalization
of arginine vasopressin- and angiotensin II-
like immunoreactivity in the hypothalamo-
neurohypophysial system of the goldfish,
CArGSSTUSTAUTALUS AI SO). Dees sterol 257

Polzonetti-Magni, A. M., R. Curini, O. Carne-
vali, C. Novara, M. Zerani and A. Gobbetti:
Ovarian development and sex steroid
hormones during the reproductive cycle of
Rana esculenta complex ...................5 265

Mugiya, Y.: Long-term effects of hypophys-
ectomy on the growth and calcification of
otoliths and scales in the goldfish, Carassius
auratus

Kobayashi, Y. and M. Okada: Urea stimula-
tion of pituitary pars intermedia cells of
suckling mice under copious drinking ...... 281

Morphology
Itow, T., T. Masuda and K. Sekiguchi: Forma-
tion of ganglions and stomodaeum in normal
and separate embryos of horseshoe crab,

Tachypleus tridentatus ...........0.0.00.000: 287

Taxonomy
Wynn, S., M. J. Toda and T. X. Peng: The
genus Phorticella Duda (Diptera: Drosophi-
lidae) from Burma and southern China ....297
Fukuda, Y.: Early larval and postlarval mor-
phology of the soldier crab, Mictyris brevi-
dactylus Stimpson (Crustacea: Brachyura:
Mictynidae) recohtees cen ad. ctectiacterd..can 303

Okajima, S.: Some Nesothrips (Insecta, Thy-
sanoptera, Phlaeothripidae) from east Asia
ASRS TIOGA Rc Staats Bee, eaters 2) 311

Miura, T. and L. Laubier: Nautiliniellid

polychaetes collected from the Hatsushima
cold-seep site in Sagami Bay, with descrip-
tions of new genera and species............ 319

NUMBER 3, JUNE 1990

REVIEWS

Plisetskaya, E. M.: Recent studies of fish
pancreatic hormones: Selected topics ...... 335

Suzuki, N.: Structure and function of sea
urchin egg jelly molecules.................. 355

ORIGINAL PAPERS
Physiology

Nakashima, H. and Y. Kamishima: Regula-
tion of water permeability of the skin of the
treefrog, Hyla arborea japonica ............ 371

Hori, K., Y. Furukawa and M. Kobayashi:
Regulatory actions of 5-hydroxytryptamine
and some neuropeptides on the heart of the
African giant snail, Achatina fulica Férussac

Kanui, T. I., K. Hole and J. O. Miaron:
Nociception in crocodiles: Capsaicin instilla-
tion, formalin and hot plate tests (COM-
MUNICATION) tee athagen tl. dex attoiebs 537

Cell Biology and Morphology
Tamamaki, N.: Evidence for the phagocyto-
tic removal of photoreceptive membrane by
pigment cells in the eye of the planarian,
Dugesia japonica ........... 0. ce cece eee ees 385

Sato, M., H. Mitani and A. Shima:
Eurythermic growth and synthesis of heat
shock proteins of primany cultured goldfish
COMSAT PENSE ccc EI i Oe hs 395

Iga, T., J. Kinutani and N. Maeno: Motility
of cultured iridophores from the freshwater
goby, Odontobutis obscura................. 401

Biochemistry
Ryuzaki, M. and M. Oonuki: Changes in
lipid composition in the tail of Rana catesbe-
lana larvae during metamorphosis.......... 409

Genetics
Phang, V. P. E., A. A. Fernando and E. W.
K. Chia: Inheritance of the color patterns
of the blue snakeskin and red snakeskin
varieties of the guppy, Poecilia reticulata

Developmental Biology
Irisawa, S., T. Iguchi and N. Takasugi: Criti-
cal period of induction by tamoxifen of
genital organ abnormalities in male mice
(COMMUNICATION) Maa-ks... Siencawnc 541

Reproductive Biology
Pandey, S. C. and S. D. Pandey: Photo-
periodic influences on pheromonal delay of
puberty in young female wild mice (COM-

MUNICATION) 24 73.5020) aiveenooks eer 547
Endocrinology
Oota, Y.: Immunocytochemical and ultra-

structural characterization of the cells in the
pars tuberalis of the turtle, Geoclemys
reevesii (COMMUNICATION) ...........- 551
Taniguchi, Y., S. Tanaka and K. Kurosumi:
Distribution of immunoreactive thyrotropin-
releasing hormone in the brain and hypo-
physis of larval bullfrogs with special refer-
ence to nerve fibers in the pars distalis

Takei, Y. and T. X. Watanabe: Vasodepres-
sor effect of atrial natriuretic peptides in the
quail, Coturnix coturnix japonica........... 435

Taxonomy and Systematics
Tanabe, T.: A new milliped of the genus
Riukiaria from Is. Yaku-shima, Japan (Diplo-
poda; Polydesmida; Xystodesmidae) ....... 443
Ito, A. and S. Imai: Ciliate Protozoa in the

rumen of Holstein-Friesian cattle (Bos
taurus taurus) in Hokkaido, Japan, with the
description of two new species ............. 449
Watabe, H., X. C. Liang and W. X. Zhang:
The Drosophila polychaeta and the D. quadri-
setata species-groups (Diptera: Drosophili-
dae) from Yunnan Province, southern
Chinayss).) ..sies. cdl.) 23h eee 459
Sawada, I. and M. Harada: Cestodes of field
micromammalians (Insectivora) from cen-
tral Honshu, Japan...................-..... 469
Takeda, M. and N. Shikatani: Crabs of the
genus Calappa from the Ryukyu Islands,
with description of a new species........... 477
Ito, T. and M. J. Grygier: Description and
complete larval development of a new spcies
of Baccalaureus (Crustacea: Ascothoracida)
parastitic in a zoanthid from Tanabe Bay,
Honshu, Japan .......... 6... 00s0 sce eee 485
Abé, H.: Three new species of the genus
Rhombognathus (Acari, Halacaridae) from
Japan. 0:2 c050/steck)sce oie oee oe eee Eee 517
Inger, R. F. and R. J. Wassersug: A cen-
trolenid-like anuran larva from southeast
Asia (COMMUNICATION ).............-. 557

NUMBER 4, AUGUST 1990

REVIEWS

Koolman, J.: Ecdysteroids.................. 563

Yoshizaki, N.: Functions and properties of
animal lectins Ae a Se 581

ORIGINAL PAPERS
Physiology
Lin, J. T., Toh, Y., Mizutani, M. and H.

Tateda: Putative neurotransmitter in the

ocellar neuropil of American cockroaches

Seti amiore a ave she erephe tere etg ONS «ars eRe 595
Okamoto, K. and T. Tagawa: Aminergic,

cholinergic and peptidergic innervation of
hepatic portal vein in the anuran amphibians

Cell Biology

Waku, Y., Koike, M. and N. Yoshida: Cell

culture of the antennal imaginal disc of the
silkworm, Bombyx mori L. and differentia-
tion of neurons from the culture ........... 613
Uchiyama, M., Yoshizawa, H., Wakasugi, C.
and C. Oguro: Structure of the internal
gills in tadpoles of the crab-eating frogs,
Rana cancrivoras..) 0... Ce 623

Biochemistry
Tamanoi, I., Fujii, N., Muraoka, S., Harada,
K., Joshima, H. and T. Ishihara: The
induction of D-aspartic acid in mouse lens
protein by continuous gamma-irradiation

(COMMUNICATION)) cn) .4k renee 763
Immunology
Saad, A. H.: Estradiol-induced lymphopenia

in the lizard, Chalcides ocellatus

Nunomura, W., Watanabe, H. K. and H.
Hirai: Interaction of C-Reactive protein
with macrophages in rat (COMMUNICA-
ION) Pertoes..akscwan. icine «dilate. Se 767

Developmental Biology
Yang C.-H. and R. Yanagimachi: Changes
in the rigidity of the hamster egg during
meiotic maturation and after fertilization . . 639
Kanayama, M. and Y. Kamishima: Role of
symbiotic algae in hatching of gemmules of
the freshwater sponge, Radiospongilla cere-

Tacke, L. and H. Grunz: Effect of cytochala-
sin B, nocodazol and procaine on binding
and fate of concanavalin A in competent
ectoderm of Xenopus laevis ................ 657

Tanaka, A. and M. H. Ross: Instability of
the number of segments of unoperated and
regenerated maxillary palpi in the maxillary-
palp-elongate (mpe) German cockroach

Reproductive Biology
Ueda, J., Hirano, T. and S. Fujimoto:
Changes in protein secretory patterns during
the development of the rat epididymis ..... 681
Ueda, H., Fukui, Y., Araki, H. and S.
Fujimoto: Protein secretory patterns dur-
ing the development of the rat ovary....... 691

Endocrinology
Endo, K., Fujimoto, Y., Masaki, T. and K.
Kumagai: Stage-dependent changes in the
activity of the prothoracicotropic hormone
(PTTH) in the brain of the Asian comma
butterfly, Polygonia c-aureum L............ 697
Yamanouchi, H. andS. Ishii: Effects of gonado-
tropin-specific antibodies on the interaction

v

of follicle-stimulating hormone and lutei-
nizing hormone with testicular receptors in
the bullfrog, Rana catesbeiana ............. 705
Kobayashi, M. and N. E. Stacey: Effect of
ovariectomy and steroid hormone implanta-
tion on serum gonadotropin levels in female
goldfishusseeuisc). detente Jaseeriie ie as 715
Holmes, W. N., Cronshaw, J. and J. L.
Redondo: Stress-induced adrenal steroido-
genesis in neonate mallard ducklings and
domestic chickens......................0065 723
Tomooka, Y., Edery, M., Mills, K. T., Bern,
H. A. and J. A. McLachlan: Effects of
androgen on mouse seminal vesicle epithelial
cells in serum-free culture.................. 731
Koshimizu, I. and Y. Oota: A scanning elec-
tron microscopic study of the blood vascular
architecture of the snake hypophysis (COM-

MUNICATION) seg epee 58 oh ocede shes of 771
Morphology
Ohtsuki, H.: Inner structures of the cerebral

vesicle in the ascidian larva, Styela plicata:
AMES Mi study Sepavae rs ceiere ore cis eels severe cetoved 739
Atoji, Y., Takayanagi, K., Suzuki, Y. and M.
Sugimura: Immunohistochemical demon-
stration of S-100 protein in the chick non-
MYETAVOUS WSUS 5 occccacocscasanecsonecaanuone 747
Atoji, Y., Takada, Y., Suzuki, Y. and M.
Sugimura: Immunocytochemical identifica-
tion of four cell types in the pancreatic islats
of the Japanese serow, Capricornis crispus
(COMMUNICATION) eases a oer een: 779

Taxonomy
Yoshimura, K.: Two new species of the
genus Monhystrium Cobb, 1920 (Monhys-
teridae: Nematoda) from terrestrial crabs of
subfamily Sesarminae (Brachyura) in Japan

NUMBER 5, OCTOBER 1990

REVIEWS

Omura, Y., K. Horst-W., M. Oguri and A.
Oksche: Properties of the blood-brain and
blood-cerebrospinal fluid (CSF) barrier in

the circumventricular organs of the

diencephalic roof of teleosts................ 783
Kobayashi, M. and Y. Muneoka: Structure
and action of molluscan neuropeptides ..... 801

vi

ORIGINAL PAPERS
Physiology

Burton, D. and B. A. Everard: In vitro charac-
teristics of K* and Na* induced melano-
phore responses in a cold ocean teleost,
Pseudopleuronectes americanus............. 815

Toh, Y.: Diurnal structural changes in rhab-
domeric microvilli of the compound eye in
the blue crab, Callinectes sapidus (COM-
MUNICATION) US) Aeron a eee 961

Uchiyama, M., T. Ogasawara, T. Hirano, S.
Kikuyama, Y. Sasayama and C. Oguro:
Serum and urine osmolyte concentrations
during acculimation to various dilutions of
seawater in the crab-eating frog, Rana
cancrivora (COMMUNICATION )......... 967

Cell Biology
Devi, S. A., S. Kan and S. Kawashima:
Effect of culture age on lipofuscin accu-
mulation and creatine phosphokinase activ-
ity in spontaneously beating rat heart cells,
and its modifications by tocopherol ........ 821

Biochemistry
Suyemitsu, T., Y. Tonegawa and K. Ishihara:
Similarities between the primary structures
of exogastrula-inducing peptides and pep-
tide B purified from embryos of the sea
urchin, Anthocidaris crassispina............ 831

Developmental Biology

Kurabuchi, S. and Y. Kishida: Comparative
study of the influence of head and tail grafts
on axial polarity in regeneration of the
freshwater planananien see ee teers 841

Fujishima, M., K. Nagahara and Y. Kojima:
Changes in morphology buoyant density and
protein composition in differentiation from
the reproductive short form to the infectious
long form of Holospora obtusa, a macro-
nucleus-specific symbiont of the ciliate Para-
WICCLUIN COUGGIUIN nent ee eee 849

Aoki, K., M. Nakamura, H. Namiki, S. Oki-
naga and K. Arai: The effect of glucose
and phosphate on mouse two-cell embryos
to develop in vitro (COMMUNICATION)

Reproductive Biology

Takahashi, N., N. Sato, N. Ohtomo, A.
Kondo, M. Takahashi and K. Kikuchi:
Analysis of the contraction-inducing factor
for gonadal smooth muscle contraction in
Sea UOChin 32.008 Wo. 4 Lise 861

Dukelow, W. R., C. S. T. Pow, J. H. Kennedy
and L. Martin: Stress effects on late preg-
nancy in the flying-fox, Pteropus scapulats

Endocrinology
Kato, Y., T. Ezashi, T. Hirai»and T. Kato:
Strain difference in nucleotide sequences of
rat glycoprotein hormone subunit cDNAs
and'gene fragment)... :..). “See 879

Behavior Biology
Hayashi, S.: Social condition influences sexu-
al attractiveness of dominant male mice

Chiba, Y., Y. Yamamoto, C. Shimizu, M.
Zaitsu, M. Uki, M. Yoshii and K. Tomioka:
Insemination-dependent modification of
circadian activity of the mosquito, Culex
pipiens pallens, .1.i...... eee 895

Yamanouchi, K.: Role of the medulary
raphe nucleus in regulating sexual behaviors
in‘femaleirats chs... .. J... o Heater 907

Morphology
Atoji, Y. and Y. Suzuki: Apocrine gland of
the infraorbital gland of the Japanese serow,
Capricornis crispus... ...¢ «.<.)eeeeee 913
Meyer, W. and A. Tsukise: Structural and
carbohydrate histochemical aspects of the
snout skin of the opossum, Didelphis virgi-
niana Kerr |... srorvosns ss aoe 923

Taxonomy
Shimazu, T.: Trematodes of the genus
Orientocreadium (Digenea: Orientocreadii-
dae) from freshwater fishes of Japan ....... 933
Higgins, R. P. and Y. Shirayama: Draco-
deridae, a new family of the cyclorhagid
Kinorhyncha from the Inland Sea of Japan

Mukai, H.: Systematic position of Stephanella

hina (Bryozoa: Phylactolaemata), with spe-
cial reference to the budding site and the
attachment of sessoblasts................... 947
Hirayama, A.: A new spepcies of the genus
Paramoera (Crustacea: Amphipoda) from
the intertidal zone of Hokkaido, northern

Vil

Ohkubo, N.: A new species of Ramusella
(Acari: Oribatei) from Japan............... 979

Sakagami, S. F. and M. Munakata: Lasiog-

lossum blackstoni sp. nov., the northern-

most representative of the palaeotropical

subgenus Ctenonomia (Insecta, Hymeno-
pteragblalictidae) Ryn. weeadte ss doe ees «s 985

NUMBER 6, DECEMBER 1990

REVIEWS

Fukumoto, M.: Morphological aspects of
ascidian fertilization.....................05. 989

Hill, R. B. and K. Kuwasawa: Neuromuscu-
lar transmission in molluscan hearts........ 999

Proceedings of the 61st Annual Meeting of the
Zoological Society of Japan.............. 1103

Announcements.........................0-- 1186
Acknowledgments.................... 1186, 1206
Autor Ind exe fata seat m asurcaci sancti neces 1187
| SHE RATAUUND is acre hd Sicko aS eee Reve re ora ene 1209
Contents of Zoological Science, Vol. 7, Nos.
ILESY ier etch Se vertier S ALcr ae re clare eeeer ence ee er i

ing ‘) epregt | bakit fu ul Ketan

wh

{ i sare | yer (We “i odode |
oi", oe)
7

a TA tiie, 4 Hnagh ase “ er.

nek aii PARR ne Ms ud i

; ‘iat ‘ (GHIDES) Lae Teen ~ a +H &
Hiss U) gion) sens soe mails gbiot ny s
PSasisk SIG i ea by ip ae 5 ie

2

tan met
4 ae)

histo’,

' rae ; @narngbolvwarthor
wit wna

nator

Anh ahi a te pt
. ‘oan

ity lat % Ae.

Per rey
uf the
wat Mea!

“ZOOLOGICAL
SCIENCE

An International Journal

ZOOLOGICAL SCIENCE

The Official Journal of the Zoological Society of Japan

Editors-in-Chief: The Zoological Society of Japan:
Seiichiro Kawashima (Tokyo) Toshin-building, Hongo 227-2, Bunkyo-ku,
Hideshi Kobayashi (Tokyo) Tokyo 113, Japan. Tel. (03) 814-5675
Managing Editor: Officers:
4 Chitaru Oguro (Toyama) President: Hiromichi Morita (Fukuoka)
ESS Saini Editors: Secretary: Hideo Namiki (Tokyo)
Yuichi Sasayama (Toyama) Treasurer: Tadakazu Ohoka (Tokyo)

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

Editorial Board:

Howard A. Bern (Berkeley) Walter Bock (New York) Aubrey Gorbman (Seattle)

Horst Grunz (Essen) Robert B. Hill (Kingston) Yukio Hiramoto (Chiba)
Susumu Ishii (Tokyo) Yukiaki Kuroda (Mishima) John M. Lawrence (Tampa)
Koscak Maruyama (Chiba) Roger Milkman (Okazaki) Kazuo Moriwaki (Mishima)
Tokindo S. Okada (Okazaki) Andreas Oksche (Giessen) Hidemi Sato (Nagoya)

Hiroshi Watanabe (Tokyo) Mayumi Yamada (Sapporo) Ryuzo Yanagimachi (Honolulu)

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

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

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

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

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

ZOOLOGICAL SCIENCE 7: 1-16 (1990) © 1990 Zoological Society of Japan

REVIEW

Extracellular Annelid Hemoglobins

Tosuio GotoH and Tomoniko Suzuki

Department of Biology, College of General Education, University of Tokushima,
Tokushima 770 and ‘Department of Biology, Faculty of Science,
Kochi University, Kochi 780, Japan

ABSTRACT— An extracellular annelid hemoglobin is a multi-subunit protein with a molecular weight
of 3-4 10° and exhibits a hexagonal bilayer of twelve submultiples. Recent advances in studies on
the molecular evolution and assembly of the huge annelid hemoglobins are summarized. Sequence
determinations of the eight polypeptide chains of the multi-subunit hemoglobins of the polychaete
Tylorrhynchus heterochaetus and the oligochaete Lumbricus terrestris have provided fundamental
information on the common molecular architecture and the phylogeny of these huge dioxygen-carrying
proteins. In addition, morphological studies using scanning transmission electron microscopy and
conventional transmission electron microscopy with image analysis have revealed the tetrahedeal
structure of the submultiple. A new nomenclature ‘a’, ‘A’, ‘b’, and ‘B’, is proposed for the four basic
constituent chains common to oligochaete and polychaete hemoglobins based on their homology.
Phylogenetically, these heme-containing chains can be separated into two strains ‘A’, and ‘B’.
According to the symmetrical ‘192-chain’ model, the multi-subunit hemoglobins might be represented
as ‘(aAbB)4g’. The minimum entity ‘aAbB’ that consists of a monomeric chain ‘a’ and a disulfide-
bonded trimer ‘AbB’ may correspond to one unit in the tetrahedral of the submultiple in electron
microscopic appearance. On the basis of recent information, earlier models are evaluated as well as
the ‘bracelet’ model, in which the minor subunits D1 and D2 without heme have a key role in linking
the complexes of subunit a and subunit AbB together.

INTRODUCTION

Ever since Svedberg [1-3] found that an ex-
tracellular annelid hemoglobin is a huge protein
and consists of either about 144 or 192 polypeptide
chains, an outstanding problem has been to con-
struct a common model for the molecular
architecture of annelid hemoglobins.

Ten years ago, we decided to determine the
molecular weight of each constituent chain of
hemoglobin of the polychaete Tylorrhynchus
heterochaetus by analyzing its amino acid sequence
[4, 5]. Our results have provided much informa-
tion on the molecular architecture and the evolu-
tion of giant annelid hemoglobins [6-11]. Similar
studies on hemoglobin of the oligochaete Lumbri-

Received October 16, 1989

cus terrestris have progressed concurrently in other
laboratories [12-14]. Four major species of con-
tituent chains with heme have been isolated from
each of these polychaete and oligochaete hemoglo-
bins and have been sequenced. Moreover, Vinog-
radov and his colleagues recently reported that
non-heme chains with molecular masses of 31-37
kDa act as a scaffold for the complexes of two
types of heme-containing subunits ‘monomers’ and
disulfide-bonded ‘trimers’ [15-17]. Morphological
studies by scanning transmission electron micros-
copy (‘STEM) [18-20] and conventional transmis-
sion electron microscopy (CTEM) with image

1 Abbreviations used: STEM, scanning transmission
electron microscopy; CTEM, conventional transmis-
sion electron microscopy; SDS-PAGE, sodium
dodecyl sulfate-polyyacrylamide gel electrophoresis;
SAXS, small-angle X-ray scattering.

2 T. GotoH AND T. SUZUKI

analysis technique [21, 22] have also provided
information for constructing of a common model
of giant hemoglobins.

These recent findings are all consistent with a
common model for the molecular architecture of
multi-subunit annelid hemoglobins consisting of
about 200 polypeptides including four species of
192 heme-containing chains and some non-heme
chains, although the detailed subunit assembly is
still controversial. In particular, it is still uncertain
whether annelid hemoglobins contain common
non-heme chains with molecular masses of about
double those of heme-containing chains. The 288
disulfide bonds present in the heme-containing
chains of a molecule must be important for inter-
loking the whole complex [11, 14] and perhaps the
non-heme chains may link the complexes of heme-
containing subunits together [16, 23]. With regard
to molecular phylogeny, two globin strains have
been found in both polychaete and oligochaete
hemoglobins [9, 23, 24]. In this article we intro-
duce a new nomenclature for the four main heme-
containing chains common to oligochaete and
polychaete hemoglobins so _ that future
‘monomeric’ and ‘trimeric’ globin chains can be
readily compared. Although non-heme chains are
expected to be ‘linkers’ that consist of two domains
of globin and conserve the drastic evolutionary
history of the hemoglobins, these chains must be
examined further, especially in terms of amino
acid or DNA sequences, to be identified as real
components. Since the major constituent chains
with heme have all been found to be homologous
with a vertebrate myoglobin, the old name for
giant annelid hemoglobins, ‘erythrocruorin’, is no
longer appropriate.

HISTORICAL BACKGROUND

Sixty years ago, Svedberg [1] developed a
method for determining the molecular weight of a
protein by centrifugation, and using this method
he found a variety of globins ranging in molecular
weight from 17,000 for myoglobin to about 3
million for annelid extracellular hemoglobins. As
the annelid extracellular hemoglobin is a huge
protein, Svedberg and Eriksson [2] thought that its
protein portion have completely different chemical

properties from those of vertebrate hemoglobins,
and they revived the name “erythrocruorin”, origi-
nally used for the red blood pigment of the inverte-
brates by Ray Lankester in 1868 [25]. No satisfac-
tory explanation of the evolutionary relationships
between different sizes and forms of globins was
proposed for many years. But in 1960, in X-ray
diffraction studies Perutz [26] observed similarity
in the three dimensional structures of vertebrate
myoglobin and hemoglobin. Now, more than 250
globin chains have been sequenced [27] and a
phylogenetic tree for their molecular evolution has
been constructed [28]. Thus the homologies of
diverse hemoglobins have been established and
these molecules have been shown to be members
of the globin family.

Kimura [29] proposed the neutral theory for
molecular evolution based largely on the primary
structures of vertebrate globin chains. Recently,
however, attention has been directed into the
structural diversity of invertebrate hemoglobins,
with the expectation of finding some drastic muta-
tions besides the point mutations that have already
been analyzed in vertebrate globin chains. Based
on the quarternary structures, Vinogradov [30] has
classified hemoglobins into four types: (a) Single-
domain, single-subunit molecules consisting of a
single polypeptide chain of about 16 kDa contain-
ing one heme group; (b) two domain, multi-
subunit hemoglobins, ranging in size from 250 to
800 kDa and consisting of 30-40 kDa chains, each
containing two heme-binding domains; (c) multi-
domain, multi-subunit hemoglobins, consisting of
two or more long polypeptide chains each contain-
ing 8-20 heme-binding domains connected lineal-
ly; (d) Single-domain, multi-subunit hemoglobins,
consisting of aggregates of several small subunits,
some of which are disulfide-bonded and not all of
which contain heme. The giant extracellular anne-
lid hemoglobins are of type d. Oligochaetes have
only this type of hemoglobin. On the other hand,
polychaetes have various types of pigments such as
monomeric and polymeric intracellular hemoglo-
bins, monomeric extracellular hemoglobins, and
multi-subunit extracellular hemoglobins and chlor-
ocruorins. In some cases, two types of hemoglo-
bins are present in a single species [31-33]. Chlor-
ocruorins are green, but they are considered to be

Multi-Subunit Annelid Hemoglobin 3

homologous with multi-subunit hemoglobins be-
cause of their similarities to the latter in size and
shape [3, 34]. Annelid multi-subunit hemoglobins
and chlorocruorins can also be characterized by
the appearance of a double-layered hexagonal
array of twelve submultiples [34-36]. These
hemoproteins appear to consist of ‘monomers’,
‘dimers’, disulfide-bonded ‘trimers’ and disulfide-
bonded ‘tetramers’ in various proportions and
combinations depending upon the species [30].
The molecular masses of unit chains in all these
forms except ‘dimers’ are comparable to those of
vertebrate myoglobins [4, 12, 30]. Most prepara-
tions of these hemoglobins and chlorocruorins
exhibit the presence of some ‘dimer’ components
with molecular masses of 31-37kDa, but they
have not yet been well characterized. The rela-
tionships between the oxygen-dissociation curve
and the structure of the annelid hemoglobins has
been studied extensively [37-46]. The present
review is, however, limited mainly to recent prog-
ress in studies on the molecular evolution and
assembly of the multi-subunit extracellular annelid
hemoglobins.

Many workers have studied different hemoglo-
bins and have proposed models for their molecular
assembly based on their results obtained by va-
rious methods [for earlier reviews, see Refs. 47-
52]. For analysis of the whole molecule, precise
measurements of basic parameters, such as the
molecular weight of the whole molecule and the
constituent chains, and the iron or heme content
are necessary. The values reported for the molecu-
lar masses of whole molecules vary from about
3,000 kDa to 4,000 kDa [3, 40, 53]. The molecular
masses of major constituent chains estimated by
sodium dodecyl sulfate-polyacrylamide gel elec-
trophoresis (SDS-PAGE) range from 11 kDa to 19
kDa [54-56], whereas the molecular masses of the
constituent chains estimated by centrifugation are
22-23 kDa [57, 58]. The minimum molecular mass
per heme observed also ranges from 17 kDa to 28
kDa [54, 59-61]. Thus, even very recently it was
difficult to justify the reported values of these
parameters and to assign the heme group to the
disulfide-bonded ‘trimer’ [15]. As pointed out by
many workers [62], it is also difficult to determine
the exact number of chains per whole molecule,

because in estimating the molecular weight of the
whole molecule any error in the minimum molecu-
lar weight is multiplied by about ‘144’ or ‘192’.

MOLECULAR SHAPE

Figure 1 shows typical STEM images of
polychaete and oligochaete hemoglobins, which
consist of double layered hexagonal submultiples
[11]. In the central cavity of the molecule, there
are faint indications of protein masses that appear
to be anchored to the surface of all 12 submulti-
ples. In the view of Tylorrhynchus hemoglobin
from the top (Fig. 1A), each submultiple appears
to be composed of three globular units with a tiny
hole at the center. The side view of a submultiple
(Fig. 1B) also shows three identical units. From
these findings the simplest steric model of a sub-
multiple is a tetrahedral structure, the fourth unit
being masked by the others both in the top and
side views. The size of Tylorrhynchus hemoglobin
is 28.4 nm in vertex-to-vertex diameter and 18.2
nm in height, as listed in Table 1 with the values
for Lumbricus hemoglobin for comparison [19].
The high resolution STEM method was developed
by Crewe of Chicago University [63] and first used
to examine Lumbricus hemoglobin by Kapp and
Crewe in the collaboration with Vinogradov of
Wayne State University [18]. Recently, Vinogra-
dov and his collaborators [15-17, 20] suggested a
role of the filament structure in the central cavity
of the hemoglobin molecule as a scaffolding or
linker for the submultiples. By two dimensional
image analysis and reconstruction by optical and
computed methods of electron micrographs,
Ghiretti Magaldi and her colleagues [21, 22, 64]
demonstrated the tetrahedral structure of the each
submultiple of an extracellular hemoglobin and a
chlorocruorin as shown is Figure 2. These electron
microscopic studies indicated the existence of 48
tetrahedral units in the whole molecule. However,
the materials found by STEM in the central cavity
could not be observed by CTEM with image
analysis [22].

Several multi-subunit annelid hemoglobins have
been examined by small-angle X-ray scattering
(SAXS) [65-70], which is a useful method for
obtaining information on the quarternary structure

4 T. GoTroH AND T. SUZUKI

Fic. 1. STEM images of Tylorrhynchus (A and B) and Lumbricus (C and D) hemoglobins. A and C, and B and D
are top and side views, respectively. Note the faint protein mass in the central cavity. (from Suzuki, T., Kapp, O.
H. and Gotoh, T. [11]. Reprinted with permission from the American Society for Biochemistry & Molecular

Biology.)

TABLE 1. Molecular dimensions of Tylorrhynchus and Lumbricus hemoglobins
determined by STEM

P Vertex-to-vertex Height Central hole
ISOS Cho diameter (nm) (nm) (nm Ref.
Tylorrhynchus 28.4 18.2 8.8 [11]
Lumbricus 30.7 20.1 8.8 [19]

(from Suzuki, T., Kapp, O. H. and Gotoh, T. [11]. Reprinted with permission
from the American Society for Biochemistry and Molecular Biology.)

Multi-Subunit Annelid Hemoglobin 5

r)
‘ ks SH:

cos
isealsieals

Fic. 2. Electron microscopic images of Spinographis chlorocruorin. 1 and 2 are bidimensional crystals in the axial
and lateral projections, respectively. 3 and 4 are the relative computed reconstructions for the axial and lateral
projections, respectively. Magnification electron micrographs 90,000. Note the tetrahedral structure of each
submultiple. (By courtesy of Dr. Ghiretti Magaldi, A. [22]. Reprinted with permission from Springer Verlag.)

of biological macromolecules in solution. For
instance, the radius of gyration and maximum
dimension of Tylorrhynchus hemoglobin have
been determined to be 10.8+0.2 nm and 29.6+0.5
nm, respectively, as listed in Table 2 in comparison
with values for Lumbricus hemoglobin [70]. These
molecular parameters are in good agreement with

the values obtained from the STEM images. A
model that fits the X-ray scattering curve of Tylor-
rhynchus hemoglobin is shown in Figure 3 [70].
This model indicates some protein masses in the
center of the molecule as well as betweem submul-
tiples consistent with STEM images. However, it
should be noted that in this model the small

6 T. GotoH AND T. SUZUKI

TABLE 2.

Radii of gyration and maximum dimensions of Tylorrhynchus and

Lumbricus hemoglobins determined by small-angle X-ray scattering

Maximum dimension

. Radius of gyration
Hemoglobin aa (nm) Ref.
Tylorrhynchus 10.8+0.2 29.5+0.5 [70]
Lumbricus 11.2+0.2 29.0+1.0 [67]

(from Pilz, I., Schwarz, E., Suzuki, T. and Gotoh, T. [70].

Reprinted with

permission of the publishers, Butterworth & Co. Ltd. ©.)

10 nm

A model of Tylorrhynchus hemoglobin based on
the X-ray scattering curve. A and B are top and side
views, respectively. Each submultiple indicated by a
large sphere is composed of 17 small spheres
arranged in five tiers containing one, four, seven,
four and one spheres, respectively. There are also
four small spheres per submultiple located between
the large spheres. Thus the whole model contains 12
<21=252 small spheres. (from Pilz, I., Schwarz,
E., Suzuki, T. and Gotoh, T. [70]. Reprinted with
permission of the publishers, Butterworth & Co.

Ltd. ©.)

Fic. 3.

spheres used to simulate the quarternary structure
have no relation to the real number and size of the
subunits or polypeptide chains forming the mole-
cule. Furthermore, the model in which each
submultiple consists of four tetramers of ellipsoids
in a tetrahedral arrangement also fits all the SAXS
data fairly well [70]. According to the SAXS
analysis, the protein mass in the central hole of
Tylorrhynchus hemoglobinis less than that Lum-
bricus hemoglobin [70]. In sharp contrast,
measurement of the pixel intensity of STEM im-
ages indicates more protein mass in the central

hole of Tylorrhynchus hemoglobin than in that of
Lumbricus hemoglobin [11]. Thus, the technical
limitations of STEM, CTEM and SAXS must be
taken into account in considering the filamentous
structure in the central cavity.

MOLECULAR PHYLOGENY

The hemoglobin of the polychaete Tylorrhyn-
chus was the first multi-subunit extracellular
hemoglobin to be sequenced completely [9]. Soon
afterwards, the corresponding chains of the oli-
gochaete Lumbricus hemoglobin were sequenced
[13, 14]. These sequencings provided much in-
formation on the phylogeny of giant annelid
hemoglobins. Two chains from other multi-
subunit oligochaete hemoglobins have subsequent-
ly been sequenced [71, 72]. Figure 4 shows the
amino acid sequences of the four chains of Tylor-
rhynchus hemoglobin [{5-7, 9]. Nine of the 25
invariant residues are also conserved in the human
8 chain. In vertebrate hemoglobins [73, 74], all
these residues except Val (A8) and Trp (A12)
belong to the central exonic regions, which are
known to be the minimal functional entity for Oz
binding [75]. It is noteworthy that the invariants of
Tylorrhynchus hemoglobin involve the residues
corresponding to the dista! (E7) His, distal (E11)
Val, and proximal (F8) His of vertebrate hemoglo-
bins, which are the most important residues for
maintaining the functional properties. The fifth
coordination position of the iron atom in the heme
is known to be histidine F8 (the proximal His); Oz
is bound at the sixth coordination position. In
sperm whale myoglobin, the distal (E7) His is
known to be the only residue capable of interacting
directly with bound dioxygen and stabilizing it
[76]. All eight chains of Tylorrhynchus and Lum-

Multi-Subunit Annelid Hemoglobin 7

bricus hemoglobins are clearly homologous with
those of vertebrate hemoglobins [9, 13, 14]. In
fact, each of the isolated chains shows a typical
absorption spectrum of a globin chain [46, 77, 62].
Jhiang et al. [78] found that the gene of a heme-
containing chain of Lumbricus hemoglobin has a
two intron-three exon structure like those of verte-
brate globin chains [73]. Therefore, there is no
reason to maintain the old name ‘erythrocruorin’
for invertebrate hemoglobins except familiarity
with this name, as pointed out by Garlick and
Riggs [12]. The heme moiety of ‘erythrocruorin’ is
the same as that of a vertebrate hemoglobin [79].

As the giant annelid hemoglobins differ from
other hemoglobins in possessing disulfide-bonded
‘trimers’ or ‘tetramers’, the sites of half-cystine
residues in Tylorrhynchus and Lumbricus hemog-
lobins are noted to be all located in the side exonic
regions [8, 9, 11, 14, 78], as shown in Figure 4.

These residues all participate in forming either
intra- or inter-chain disulfide bridges [11, 14], as
shown in Figure 5. There is one intra-chain
disulfide bridge in each chain of Lumbricus and
Tylorrhynchus hemoglobins including the monom-
ers [11, 14]. As the multi-subunit annelid hemog-
lobins dissociate completely into separate chains in
the presence of a reducing agent without any other
protein-denaturant [21, 81, 82], the disulfide bonds
appear to have a key role in the in vivo assembly of
the giant molecule [11]. Namely, point mutations
at the positions now having Cys residues must have
been a major factor in bringing about formation of
these huge proteins. The side exons can also be
considered to have evolved with a special role of
stabilizing the molecular assembly.

High homologies are seen between Tylorrhyn-
chus chains b and B (74 identical residues, 50%
homology), and Tylorrhynchus chains a and A (57

1 10 20 30 y 40
Chain a TDICIG IL QIRII KIVIKQQWAQVYS-VGESIR|-TDFAIDVFNNF
ChainA SSDHICIGPL ORILK\VIKQQWAKAYG-VGHE|JRI-VELGIALWKSM
Chain b DTCICIS 1 EDRIREVIQALIWRSIWSAEDTGRIRTLIGRLLFEEL
Chain B DDC|CISAADIRIHEWVILDNIWKGIWSAEFTGRIRVAIGQAIFQEL
e e
50 60 70 80
a RTNPD-RSLI/FINRVINGDNVYSPEFK/AHIMVIR ViF AIGIFIDILIS
A AQDNDARDLIFIKIRVHGEDVHSPAFE/AHMAIR VIF NIGILIDIRVIS
b EIDGATKGLIFIKIRVNVDDTHSPEE FIAH|VLIRVIVNGILIDITLIG
B ALDPNAKGVIFIGRVINVDKPSEADWKIAH|V I|R ViI NiGILIDJL AVN
ee e
—C— —— 0) ——_ — ) SS aoa
90 100 110 41 120
a VILIDIDIKPV/JLIDQALAJHIYAAFIHWJKKQFGTIPFKA/FIGQTMFQTIAE
A eisieatis LRQQWI-KLGITGHMFNL-MRTGLAY
b VILIGDISDTILINSLIDIH}LAEQHKARAGFKTVY|FKE-FGKALNH
B LILJEIDIPK AILJQEELKIHJ|LARQIWRERSGVKAVY|FIDE-MEKALLK
e e e e
-—_ § F———4 »#__ Fr 4 + — F 4
130 140 150
a HI---HG--ADIGAWRAICYAEQIVTGII|TA
A VLPAQLGRCFODK E/JAWA AICIWDEVIYPG/I/KHD
b VLPEVAS-CFNPEIAWNHCIFDGLVDV-|I|SHRIDG
B VLPQVSSH-FNSGAWDRIC}IFTRIADV-LWKAELP

Fic. 4. Amino acid sequences of the constituent chains of Tylorrhynchus hemoglobin. The alignment is based on the
assumption that the helical segments present in most other globins are also present in the constituent chains of the
multi-subunit annelid hemoglobin. The boxed residues indicate the 25 invariable residues in four globin chains.
The residues indicated by a dot are homologous with those of the human # chain. Arrows without a suffix indicate
the positions of half-cystine residues. Arrows with the suffix ‘J’ indicate intronic positions assuming that the
intronic positions in Tylorrhynchus hemoglobin are the same as those in vertebrate globins [73] and Lumbricus
hemoglobin [78]. (from Suzuki, T. and Gotoh, T. [9]. Reprinted with permission from the American Society for
Biochemistry and Molecular Biology.)

8 T. GoroH AND T. SUZUKI

chain A

Chain b

Fic. 5. A model for the steric assembly of the disulfide-bonded ‘trimer’ of chains A, b and B in Tylorrhynchus
hemoglobin. ‘Globin folding’ was cited from that of the human £ chain [80]. (from Suzuki, T., Kapp, O. H. and
Gotoh, T. [11]. Reprinted with permission from the American Society for Biochemisrty and Molecular Biology.)

He M2) 43)" 74) ebe 6) 7 28h ee9

10 11 12 13 14 15 16 17 18 19 20 21 22

Lum. a Glu} CYS|Leu Val Thr Glu Gly Leu}Lys . Lys} Leu Gin Ala Ser Ala
Lum. A Lys Lys GlIn}CYS|Gly Val Leu Glu Gly Leu|Lys Lys} Ser Glu Gly Arg Ala
Tyl. a Thr Asp}|CYS|Gly Ile Leu Gln Arg Ile|Lys Lys| Gin Gin Ala Gln Val
Tyl. A Ser Ser Asp CYS|Gly Pro Leu Gln Arg Leu|Lys Lys} Gln Gln Ala Lys Ala
Lum. b_ Asp Glu Ilis Glu His CYS Ser} Glu GluJAsp}Ilis Tyr Ile Gin Lys Gin Asp Ile Leu
Lum. B Ala Asp Glu Glu Ser CYS Ser} Tyr Glu|Asp} Arg Arg Glu Arg His Ile Asp Asp Val
Tyl. b Asp Thr CYS Ser| Ile GlujAsp| Arg Arg Glu Gin Ala Leu Arg Ser Ile
Tyl. B Asp Asp CYS Ser| Ala Alaj/Asp} Arg His Glu Leu Asp Asn Lys Gly Ile

NA A

Fic. 6. Alignment of NH-terminal sequences of Tylorrhynchus and Lumbricus chains. The boxed residues indicate
invariant residues in the eight chains and either the upper strain (strain A) or the lower strain (strain B). The
residues indicated by a dot are homologous with those of the human f# chain. Others are as for in Fig. 4. (from
Gotoh, T., Shishikura, F., Snow, J. W. Ereifej, K. I., Vinogradov, S. N. and Walz, D. A. [24]. Reprinted with
permission from the Biochemical Society and Portland Scientific Press.)

identical residues, 40% homology). Therefore,
phylogenetically, the ‘trimeric’ chain A is more
closely related to the ‘monomeric’ chain a than to
either of the other ‘trimeric’ chain, b and B [9].
This finding has been extended to the idea that
there are in general two distinct groups, strain A

and B, of chains in multi-subunit annelid hemoglo-
bins, as clearly seen in Figure 6 [24]. The separa-
tion of these strains must have been derived from
the gene duplication [24]. In fact, using the
unweighted pair-group clustering method [83],
Fushitani et al. [14] clearly demonstrated two

Multi-Subunit Annelid Hemoglobin 9

TABLE 3.

Proposed common nomenclature for the chains of multi-subunit annelid

hemoglobins, and the corresponding names for Tylorrhynchus and Lumbricus

hemoglobins used hitherto

Hemoglobin Chain name Ref.
Tylorrhynchus I IfA IIC IIB [6]
Lumbricus I II Ill IV [55]
Lumbricus d b c a [77]
Common name a A b B This paper
No. of Cys 2 3 4 3 [9, 14]

The arbitrary names were given according to either the order of mobility on SDS-PAGE

[6, 55] or the elution order on column chromatography [6, 77].

The proposed

nomenclature is based on the homology between different hemoglobins.

subfamilies in the phylogenetic tree for Lumbricus
and Tylorrhynchus hemoglobins. Furthermore,
they used the type of inter-chain disulfide bonding
and the number of half-cystine residues to disting-
uish two types of chains in the same strain [14].
Thus the correspondence of the four constituent
chains in these two hemoglobins has been estab-
lished.

Here, we would like to propose a new nomencl-
ature for common names of the sequenced chains
to facilitate comparison of the ‘monomeric’ and
‘trimeric’ globin chains of different species. We
have used the new nomenclature without defini-
tion in Figures 4, 5 and 6. The smaller and
‘monomeric’ chain in strain A is named chain ‘a’,
and the other one in strain A is named chain ‘A’.
Chain a (139 residues) of Tylorrhynchus hemoglo-
bin is smaller than all the other constituent chains:
chain A, 146 residues; b, 149 residues; B 148
residues. Lumbricus chain a is also the smallest
constituent and exists as a ‘monomer’ [13, 14, 55].
Chains a and A can also be distinguoshed by the
number of half-cystine residues [9, 14]: they have 2
and 3 half-cystine residues, respectively, as seen in
Figure 4. Similarly, chains ‘b’ and ‘B’ in strain B
can be defined as those having 4 and 3 half-cystine
residues, respectively. Chains b was shown ex-
perimentally to be situated in the center of the
trimer ‘AbB’, as shown in Figure 5 [11, 14].
Employing the myoglobin-fold for each chain in
the steric model of disulfide-bonded ‘trimer’, five
disulfide bridges can easily be located without any
bending or stretching [11]. Chain a exists as a
‘monomer’ because it has only two half-cystine

residues and forms one intra-chain disulfide bond
[9, 11, 14]. The distributions of half-cystine re-
sidues in Lumbricus and Tylorrhynchus hemoglo-
bins are the same [9, 14]. The relationships
between the new names and the arbitrary ones
used previously for the Tylorrhynchus and Lum-
bricus chains are summarized in Table 3. As the
polychaete Tylorrhynchus heterochaetus and the
oligochaete Lumbricus terrestris are very different
species each other in the phylum Annelida, the
proposed nomenclature can be extended to many
other annelid hemoglobins. A slight modification
of the proposed nomenclature will, however, be
necessary in the future bacause some annelid
hemoglobins and chlororuorins contain disulfide-
bonded ‘tetramers’ [30].

Figure 7 shows the phylogenetic tree of the nine
globin chains of the multi-subunit annelid hemog-
lobins from the polychaete Tylorrhynchus and the
oligochaete Lumbricus and Pheretima sieboldi
according to the unweighted pair-group clustering
method. Strains A and B are clearly separated.
According to the unweighted pair-group clustering
method [83], the amino acid substitution rate for
the ‘functionally essential’ central exonic region is
about 1.5 times slower than that for the ‘structural-
ly essential’ side exonic regions [71]. The observed
identities of the nine chains of Tylorrhynchus,
Lumbricus and Pheretima hemoglobins range from
32 to 51%, as shown in Figure 7. Some of these
identities are comparable to that (44%) between
the a and £ chains of human hemoglobin, which
were separated by gene duplication about 450
million years ago in the Ordovician period [84].

10 T. GotoH AND T. SUZUKI

OBSERVED IDENTITY (2)
T TUE Fe a
32 38 42 51

0.489

0.373

Strain B

0.472

Strain A

Phe.a

Fic. 7. Phylogenetic tree of the nine heme-containing
chains from multi-subunit annelid hemoglobins.
The tree was constructed from a homology matrix by
comparison of 137 amino acid residues common to
all chains [71], by an unweighted pair-group cluster-
ing method [83]. The standard errors at the bran-
ching points 1-8 are 0.045, 0.038, 0.042, 0.035,
0.041, 0.044, 0.036 and 0.042, respectively.

Assuming comparable evolutionary rates for ex-
tracellular and vertebrate hemoglobins, Fushitani
et al. [14] suggested that the divergence of the
Oligochaeta and Polychaeta occurred around the
time of the gene duplication that led to the a and £
gene families in vertebrates. On the other hand,
from the maximum parsimony tree, Goodman et
al. [28] calculated the times when the annelid
globin chains were separated more exactly. For
instance, they reported that the Tylorrhynchus
chain b and B were separated about 140 mililon
years ago in the Cretaceous period, the Tylorrhyn-
chus chain a and other Tylorrhynchus chain were
seperated about 380 million years ago in the Devo-
nian period, and the Glycera intracellular
monomeric chain and extracellular globin chains
were separated about 575 million years ago in the
Cambrian period. Although it is of great biological
interest to know the time when the Polychaeta and
Oligochaeta were separated, fossil records of
annelids, and particularly of oligochaetes are very
incomplete [85].

eS a oo |

FRACTION NUMBER

Fic. 8. Elution profile on chromatofocussing of Tylorrhynchus cyanomethemoglobin. Hemoglobin solution was
reduced with dithiothreitol and applied to a PBE94 column equilibrated with a imidazole buffer (pH 7.4).

Material was eluted with Polybuffer 74 (pH 5).

Peaks A, B, a and b are those of chains A, B, a and b,

respectively. Note that all four chains isolated carry a heme. (from Suzuki, T., and Gotoh, T. [6]. Reprinted with
permission from the American Society for Biochemistry and Molecular Biology.)

Multi-Subunit Annelid Hemoglobin 11

Chain a Chain A Chain b Chain B
‘192-CHAIN’ MODEL OF THE SUBUNIT CM, 16,321) (M, 17,218) (M, 17,411) (M, 17,926)
ASSEMBLY Monomer a i

Sequence analyses of the multi-subunit annelid
extracellular hemoglobins have indicated the pre-
cise molecular weight of protein moiety in each
constituent chain, which is essential for construc-
tion of a model of the molecular architecture. The
heme content of each chain has been clarified
directly by column chromatographic isolation pro-
cedures [6, 24, 77, 86]. Figure 8 shows the elution
profile of each constituent chain of Tylorrhynchus
hemoglobin on chromatofocussing [6]. The iso-
lated chains exhibit the typical absorption spec-
trum of a vertebrate myoglobin. Therefore, it is
concluded that all four chains, a, A, b and B,
contain one heme group per chain. In fact, these
four chains have a histidine residue, which corres-
ponds to the proximal one (F8) of vertebrate
hemoglobins [26], as shown in Figure 4.

The molar ratio of the four Tylorrhynchus
chains a:A:b:B was determined to be nearly
1:1:1:1 by statistical comparison of the exact
amino acid compositions calculated from the sequ-
ence of each chain and the observed composition
measured by amino acid analysis of the whole
molecule [10]. Considering the apparent molecu-
lar masses of the whole molecule (3, 370 kDa),
submultiple (250kDa), unstable tetramer (72
kDa) and disufide-bonded trimer (50 kDa) [9, 87,
88], we proposed that the formation of Tylorrhyn-
chus hemoglobin may be as follows [10].

Dislfide-bonded trimer AbB (M, 51,925)

Tetramer aAbB (M, 68,246)
Submultiple (4 tetramers; M, 272, 984)

Whole molecule
(12 submultiples, M, 3,275,808)

Therefore, our model proposed for Tylorrhynchus
hemoglobin consists of 192 polypeptide chains
containing heme. According to the new nomencla-
ture proposed above, Tylorrhynchus hemoglobin
can be represented as ‘(aAbB),g’. aAbB is the
minimum structural entity, which may correspond
to one unit in the tetrahedral structure of the

submultiple observed in the STEM image (Fig. 1).
This giant protein consists of 27,936 amino acid
residues and 192 heme groups, and is linked by 288
disulfide bridges, 192 intra- and 96 inter-chain
disulfide bonds. These are the contents of our
symmetrical ‘192-chain’ model. The salient point
of this model is that the subunit assembly is based
on the exact molecular weight and strict molar
ratio of each constituent polypeptide chain. The
‘192-chain’ model is in good agreement with elec-
tron microscopic observations for the tetrahedral
structure of a submultiple [21, 22, 61, 64, 89].

TABLE4. Earlier models of annelid multi-subunit hemoglobins

Hemoglobin Gaeahenee No. of units aa ane euiae) Ref.
Lumbricus 17,250 144 2.48 {1]

Lumbricus 17,600 192 3.38 [3]

Arenicola 27,000 192 2.59 [54]
Octolasium 23,000 144 3.30 [57]
Lumbricus 17,000 192 3.26 [90]
earthworm 22,000 192 3.84 [58]
Tylorrhynchus 17,062 192 3.275808 [10]

All models except those reported in refs. [10] and [54] consist of a single species of unit
protein containing a heme moiety.

? The M, of each constituent chain of Tylorrhynchus
hemoglobin including a heme group was recalculated.

12 T. GoroH AND T. SUZUKI

However, this model does not explain the filamen-
tous structure in the central hole of the molecule,
shown in Figure 1. Two further other problems
about the ‘192-chain’ model are that the measured
value of the minimum molecular mass per heme is
much higher than the value calculated from the
model, and that the SDS-PAGE pattern often
shows one or two extra minor components with
molecular masses of about double those of the
chains containing heme [88].

Table 4 shows some earlier models of annelid
hemoglobins compared with our ‘192-chain’ mod-
el. Chiancone et al. [57] and David and Daniel [58]
reported higher values for the minimum molecular
weight. These values might be reasonable for the
minimum molecular weight per heme, but not for
the molecular weight of a minimum unit, because
sequence analyses of Tylorrhynchus [9] and Lum-
bricus hemoglobins [14] revealed that the size of
each chain containing heme is comparable with
that of myoglobin. Chiancone et al. considered the
‘144-chain’ model based on the relationships of the
molecular weights of the minimum unit and whole
molecule. On the other hand, David and Daniel
proposed a ‘192-chain’ model by which the calcu-
lated molecular weight of the whole molecule
reached the measured value of 3.84 10°, in con-
trast to the others clustered in the first half of 3
million. Anyway, it is now obvious that these
models overestimated the molecular weight of a
minimum unit. On the other hand, in most cases,
the molecular mass of a constituent chain was
underestimated by SDS-PAGE. Waxman [54]
examined Arenicola cristata hemoglobin by SDS-
PAGE and he found two major components with
molecular masses of 13 kDa and 14kDa. As the
minimum molecular mass per heme was estimated
to be 27 kDa, he proposed a ‘192-chain’ model for
Arenicola hemiglobin with a relatively lower
molecular mass of 2,590 kDa, which consists of

two species of constituent chains, half of which do
not contain a heme group. Vinogradov and his
colleagues have maintained the idea that not all
chains contain a heme group [16, 55, 91].

‘BRACELET’ MODEL OF THE SUBUNIT
ASSEMBLY

Shlom and Vinogradov [55] noted fifth and sixth
components of multi-subunit hemoglobins that can
always be observed as faint bands V and VI on
SDS-PAGE of a preparation of Lumbricus hemog-
lobin. As these components are about twice the
size of the major components, chains a, A, b and
B, they are designated as ‘D1’ (M,=31,000) and
‘D2’ (M, =37,000) and called “dimers” arbitrarily
(15). Recently, Vinogradov’s group [15-17] prop-
osed a ‘bracelet’ model for the subunit assembly of
a giant annelid hemoglobin in which D1 and D2
have the remarkable role of linking 12 submulti-
ples. In the novel ‘bracelet’ model, subunits D1
and D2 are assumed to form a closed circular collar
or bracelet decorated with 12 complexes of several
‘monomers’ a and ‘trimers’ AbB, providing the
electron microscopic appearance of a symmetrical
hexagonal bilayer. There is no inter-chain di-
sulfide bridge between ‘dimers’ D1 and D2, and
these and heme-binding subunits a and AbB [5S].
The ‘bracelet’ model was deduced from extensive
studies on the dissociation and reassociation of
Lumbricus hemoglobin either at an extreme of pH
[17, 19] or in the presence of a dissociating agent at
neutral pH [16]. The existence of subunits D1 and
D2 appeared to be essential for complete reasso-
ciation of the molecule of Lumbricus hemoglobin
from the dissociated products under physiological
conditions. Although the experimental results did
not provide direct proof of the existence of a
bracelet structure, the filamentous structure in the
central hole of the giant hemoglobin may corres-

TABLE5. Models of annelid hemoglobins with none-heme chains
Hemoglobin No. of chain No. of heme M, (X10~°) Ref.
Arenicola 192 96 BESS [54]
Lumbricus 204 144 3)7// [91]
Lumbricus ca. 200 156 3.8 [16, 20]
Lumbricus 204 192 Sail] [23]

Multi-Subunit Annelid Hemoglobin 13

pond to the bracelet [20]. An alternative explana-
tion is that subunits D1 and D2 do not form a
continuous bracelet structure but may act as link-
ers between submultiples [16].

Table 5 summarizes the models of multi-subunit
annelid hemoglobins that include non-heme
chains. These models depend greatly upon the
estimated value for the heme content, which
varied significantly in different laboratories.
Vinogradov et al. [15, 91] assumed that one chain
of the ‘trimer’ of Lumbricus hemoglobin did not
contain heme. However, this idea was disproved
by the finding that all chains isolated by chroma-
tography under mild conditions contained a heme
group [24, 77]. Recently, Fushitani et al. [23]
determined the heme content of Lumbricus
hemoglobin to be one mole per 19,000 g of pro-
tein. Their model consists of 48 ‘aAbB’ and 12
chains of D1 and D2, as shown in Table 5. This
model appears to be the most consistent with all
data so far obtained on Lumbricus hemoglobin.
Fushitani and Riggs [23] assumed that D1 and D2
contain no heme, and reported that the amounts of
D1 and D2 constitute 11.2% of the total. On the
other hand, Mainwaring et al. [17] reported that
D2 exhibits a reduced capacity to bind heme. The
relative proportions of D1 and D2 in Lumbricus
hemoglobin are reported to be 25 and 10%, re-
spectively [17, 55]. D1 was separated into two
fractions [92]. There are no homologies in the
reported NH,-terminal sequences of D2 and two
D1 chains or between these and those of other
globins [23, 92]. If D1 and D2 have evolved as
heterogeneous components with the same function
of linking the complexes of subunits a and AbB
together, they shold exhibit high homology in
terms of amino acid sequences. The natures of
chains D1 and D2 are still unclear. Comparative
studies revealed diversity in the appearance of D1
and D2 components [30]. In Nephthys incisa
hemoglobin, no dimer component can be seen in
the absence of a reducing agent, but two dimers
are seen on SDS-PAGE in the presence of a
reducing agent [93]. On the other hand, in the case
of Potamia leptochaeta chlorocruorin, one band of
dimer is seen in the absence of a reducing agent,
but no dimer is seen in reduced conditions [94].
Waxman [54, 61] suggested that the faint bands of

D1 and D2 might be those of partially reduced
dimers derived from disulfide-bonded ‘tetramers’.
Pionetti and Pouyet [95] throught that the two
bands with molecular masses of about 30 kDa must
be due to contaminants with some non-heme
chains. At present, it can not be concluded that
D1 and D2 are common to all multi-subunit anne-
lid hemoglobins, although several lines of evidence
support the ‘bracelet’ model.

According to the ‘192-chain’ model and the
‘bracelet’ model, the values for the calculated
molecular weight of Lumbricus hemoglobin are
3,343,872 and 3,770,000, respectively [23].
Ghiretti Magaldi et al. [21] stated, “we have abso-
lutely no reason to believe that these M, values are
underestimated by even 25%”. They determined
the relative molecular weights of Opheria bicornis
hemoglobin and Spirographis sparanzanii chloroc-
ruorin to be about 310°. In contrast, Vinogra-
dov and Kolodziej [53] stressed that the recent
values reported for the molecular weight of Lum-
bricus hemoglobin are clustered at about 3.8 x 10°.
Furthermore, Vinogradov et al. [16] stated in the
‘bracelet’ model, “it is postulated that the
stoichiometries of some of the subunits need not
be constant”. Due to the symmetrical appearance
of 12 submultiples, however, we believe that there
are at least 6 or 12 units for each constituent in the
whole molecule. From an X-ray crystallographic
study, Royer and Hendrickson [96] reported that
self-rotation function calculations of Lumbricus
hemoglobin reveal D¢ symmetry to a resolution of
at least 6 A.

PERSPECTIVE

Which is correct, the ‘192-chain’ model or the
‘bracelrt’ model? At present neither can be disre-
garded. The discrepancy between these models
will be explained in part by the results of sequenc-
ing the D1 and D2 chains from at least two species
of hemoglobins. Immuno-electron microscopy
might be a suitable method to determine the loci of
D1 and D2 in the molecule, if it is possible to
distinguish D1 and D2 from the other chains a, A,
b and B immunologically [97]. The fine three-
dimensional structure of the multi-subunit hemo-
globin can of course be determined by X-ray

14 T. GoroH AND T. SUZUKI

crystallography, which is currently in progress us-
ing crystals of Lumbricus hemoglobin [96, 98].
Comparative study will also be of great importance
in considering D1 and D2. In fact, it is noteworthy
that the deep-sea tube worm Lamellibrachia sp.
contains two kinds of hemoglobins with molecular
masses of 440 kDa and 3,000 kDa, the latter hav-
ing the characteristic quarternary structure of mul-
ti-subunit annelid hemoglobins [99, 100]. Suzuki et
al. [101] found that the larger one contains compo-
nents corresponding in molecular size to ‘linkers’
D1 and D2, whereas the smaller one does not
contain ‘linker’ proteins. Although the tube worm
has been assigned to the phylum Vestimentifera
[102], the NH>-terminal amino acid sequences of
its four chains containing heme exhibit high
homology with those of polychaete and oli-
gochaete chains a, A, b and B [101, 103]. Even
more important, the ‘linker’ chain D1 of the tube
worm hemoglobin shows slight, but significant
homology with at least one of the four heme-
binding chains sequenced completely [103, 104].
The alignment of the D1 chain with the heme-
binding chain suggests that the linker resulted from
gene duplication and exon shuffling of a globin
chain with a three exon-two intron structure. The
D1 chain consists of two domains and the first exon
of domain 1 and the last exon of domain 2 are
deleted. Therefore, there is a high possibility that
in Lumbricus hemoglobin either D1 and D2 is also
a real member of the molecule. Anyway, to find a
rare protein, it is essential to scrutinize any possi-
ble candidates thoroughly. Conversely, when mor-
phologists agree with the idea of biochemists that
tube worms belong to the phylum Annelida, the
sequences of the D1 chain of Lamellibrachia
hemoglobin [104] will be recognized as the first
‘dimer’ chain of annelid hemoglobins sequenced.
In order to understand much clearly about the
evolution of multi-subunit annelid hemoglobins, it
is necessary to investigate the locus of each gene of
the constituent chains in the DNA molecules (S).
There are many challenging problems on inverte-
brate hemoglobin molecules to be solved in terms
of DNA sequences. In the near future, the drastic
mutations occurred in soluble proteins during the
course of evolution can be explained in large parts
by the knowledge of invertebrate hemoglobins

because they contain treasures of diversity not only
in the primary and quarternary structures [30, 32,
105-107] but also in the biosyntheses including the
regulation [108, 109]. On the other hand, it will
also be of great interest to design disulfide-bonded
trimers and tetramers such as ‘a B a’, ‘B a B’, and
‘ay f’ of vertebrate hemoglobins, which might be
synthesized with the technique of DNA recom-
binant [110] by inserting half-cystine residues at
sites corresponding to those of Cys-4, 5, 129 and
148 in multi-subunit annelid hemoglobins.

ACKNOWLEDGMENTS

This paper is dedicated to Professor Kazuhiko Konishi
of the Department of Biology, Faculty of Science,
Tohoku University on the occasion of his retirement. He
has encouraged us throughout this work.

REFERENCES

1 Svedberg, T. (1933) J. Biol. Chem., 103: 311-325.

2 Svedberg, T. and Eriksson, I. -B. (1933) J. Am.
Chem. Soc., 55: 2834-2841.

3 Svedberg, T. (1937) Nature, 139: 1051-1062.

4 Suzuki, T., Takagi, T., Shikama, K. and Gotoh, T.
(1981) Zool. Mag. (Tokyo), 90: 549. (Abstr.)

5 Suzuki, T., Takagi, T. and Gotoh, T. (1982)
Biochim. Biophys. Acta, 708: 253-258.

6 Suzuki, T., Furukohri, T. and Gotoh, T. (1985) J.
Biol. Chem., 260: 3145-3154.

7 Suzuki, T., Yasunaga, H., Furukohri, T., Naka-
mura, K. and Gotoh, T. (1985) J. Biol. Chem.,
260: 11481-11487.

8 Suzuki, T. and Gotoh, T. (1986) In “Invertebrate
Oxygen Carriers”. Ed. by B. Linzen, Springer
Verlag, Berlin/Heidelberg, pp. 69-72.

9 Suzuki, T. and Gotoh, T. (1986) J. Biol. Chem.,
261: 9257-9267.

10 Suzuki, T. and Gotoh, T. (1986) J. Mol. Biol., 190:
119-123.

11 Suzuki, T., Kapp, O. H. and Gotoh, T. (1988) J.
Biol. Chem., 263: 18524-18529.

12 Garlick, R. L. and Riggs, A. F. (1982) J. Biol.
Chem., 257: 9005-9015.

13 Shishikura, F., Snow, J. W., Gotoh, T. Vinogra-
dov, S. N. and Walz, D. A. (1987) J. Biol. Chem.,
262: 3123-3131.

14 Fushitani, K., Matsuura, M. S. A. and Riggs, A. F.
(1988) J. Biol. Chem., 263: 6502-6517.

15 Vinogradov, S. N. (1986) In “Invertebrate Oxygen
Carriers”. Ed. by B. Linzen, Springer Verlag,
Berlin/Heidelberg, pp. 25-36.

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32
33

34

35

36
37

Multi-Subunit Annelid Hemoglobin 15

Vinogradov, S. N., Lugo, S. D., Mainwaring, M.
G., Kapp, O. H. and Crewe, A. V. (1986) Proc.
Natl. Acad. Sci. U.S.A., 83: 8034-8038.
Mainwaring, M. G., Lugo, S. D., Fingal, R. A.,
Kapp, O. H. and Vinogradov, S. N. (1986) J. Biol.
Chem., 261: 10899-10908.

Kapp, O. H., Vinogradov, S. N., Ohtsuki, M. and
Crewe, A. V. (1982) Biochim. Biophys. Acta, 704:
546-548.

Kapp, O. H., Polidori, G., Mainwaring, M. G.,
Crewe, A. V. and Vinogradov, S. N. (1984) J.
Biol. Chem., 259: 628-639.

Kapp, O. H., Mainwaring, M. G., Vinogradov, S.
N. and Crewe, A. V. (1987) Proc. Natl. Acad. Sci.
U. S. A., 84: 7532-7536.

Ghiretti Magaldi, A., Zanotti, G., Tognon, G. and
Mezzasalma, V. (1985) Biochim. Biophys. Acta,
829: 144-149.

Ghiretti Magaldi, A., Ghiretti, F., Tognon, G. and
Zanotti, G. (1986) In “Invertebrate Oxygen Car-
riers”. Ed. by B. Linzen, Springer Verlag, Berlin/
Heidelberg, pp. 45-55.

Fushitani, K. and Riggs, A. F. (1988) Proc. Natl.
Acad. Sci. U.S.A., 85: 9461-9463.

Gotoh, T., Shishikura, F., Snow, J. W., Ereifej, K.
I., Vinogradov, S. N. and Walz, D. A. (1987)
Biochem. J., 241: 441-445.

Ray Lankester, E. (1868) J. Anat. Physiol., 2:
114-116.

Perutz, M. F., Rossmann, M. G., Cullis, A. F.,
Muirhead, H., Will, G. and North, A. T. C. (1960)
Nature, 185: 422-427.

Kleinschmidt, T. and Sgouros, J. G. (1987) Biol.
Chem. Hoppe Seyler, 367: 223-228.

Goodman, M., Pedwaydon, J., Czelusniak, J.,
Suzuki, T., Gotoh, T., Moens, L., Shishikura, F.,
Walz, D. A. and Vinogradov, S. N. (1988) J. Mol.
Evol., 27: 236-249.

Kimura, M. (1983) The Neutral Theory of Molecu-
lar Evolution. Cambridge Univ. Press, Cambridge.
Vinogradov, S. N. (1985) Comp. Biochem. Bio-
phys., 82B, 1-15.

Ochi, O. (1969) Mem. Ehime Univ. Sci. B., 6: 23-
91.

Terwilliger, R. (1980) Am. Zool., 20: 53-67.
Fushitani, K., Ochi, O. and Morimoto, H. (1982)
Comp. Biochem. Physiol., 72B: 267-273.

Roche, J. (1965) In “Studies in Comparative
Biochemistry”. Ed. by K. D. Munday, Pergamon
Press, Oxford, pp. 62-80.

Roche, J., Bessis, M. and Thiéry, J. P. (1960)
Biochim. Biophys. Acta, 41: 182-184.

Levin, O. (1963) J. Mol. Biol., 6: 95-101.
Haughton, T. M., Kerkut, G. A. and Munday, K.
A. (1958) J. Exp. Biol., 35: 360-368.

38

39

40

41

42

43

44

45

46

47

48

49

50

51

oy

53

54

55

56

57

58

59

60

61

62

Mangum, C. P., Lykkeboe, G. and Johansen, K.
(1975) Comp. Biochem. Physiol., 52A: 477-482.
Giardinia, B., Chiancone, E. and Antonini, E.
(1975) J. Mol. Biol., 93: 1-10.

Wood, E. J., Mosby, L. J. and Robinson, M. S.
(1976) Biochem. J., 153: 589-596.

Ochiai, T. (1984) Arch. Biochem. Biophys., 231:
136-143.

Igarashi, Y., Kimura, K. and Kajita, A. (1985)
Biochem. Int., 10: 611-618.

Imai, K. and Yoshikawa, S. (1985)
Biochem., 147: 453-463.

Fushitani, K., Imai, K. and Riggs, A. F. (1986) J.
Biol. Chem., 261: 8414-8423.

Bunn, H. F. and Forget, B. G. (1986) Hemoglo-
bin: Molecular, Genetic and Clinical Aspects. pp.
126-167. W. B. Saunders Co., Philadelphia.
Tsuneshige, A., Imai, K., Hori, H., Tyuma, I. and
Gotoh, T. (1989) J. Biochem. (Tokyo), 106: 406-
417.

Mungum, C. (1976) In “Adaptation to Environ-
ment: Physiology of Marine Animals”. Ed. by P.
C. Newell, Butterworth’s London, pp. 191-278.
Antonini, E., and Chiancone, E. (1977) Annu.
Rev. Biophys. Bioeng., 6: 239-271.

Chung, M. C. M. and Ellerton, H. D. (1979) Prog.
Biophys. Mol. Biol., 35: 53-102.

Weber, R. E. (1978) In “Physiology of Annelids ”.
Ed. by P. J. Mill, Academic Press, New York, pp.
393-478.

Weber, R. E. (1980) Am. Zool., 20: 79-101.
Vinogradov, S. N., Shlom, J. M., Kapp, O. H. and
Frossard, P. (1980) Comp. Biochem. Physiol.,
67B: 1-16.

Vinogradov, S. N. and Kolodziej, P. (1988) Comp.
Biochem. Physiol., 91B: 577-579.

Waxman, L. (1971) J. Biol. Chem., 246: 7318-
7327.

Shlom, J. M. and Vinogradov, S. N. (1973) J. Biol.
Chem., 248: 7904-7912.

Ochiai, T. and Enoki, Y. (1981) Comp. Biochem.
Physiol., 68B: 275-279.

Chiancone, E., Vecchini, P., Rossi-Fanelli, M. R.
and Antonini, E. (1972) J. Mol. Biol., 70: 73-84.
David, M. M. and Daniel, E. (1974) J. Mol. Biol.,
87: 89-101.

Partel, S. and Spencer, C. P. (1963) Comp.
Biochem. Physiol., 8: 65-82.

Yamagishi, M., Kajita, A., Shukuya, R. and Ka-
jiro, K. (1966) J. Mol. Biol., 21: 467-472.
Waxman, L. (1975) J. Biol. Chem., 250: 3790-
3795.

Terwilliger, R. C., Terwilliger, N. B. and Roxby,
R. (1975) Comp. Biochem. Physiol., 50B: 225-
232.

Eur. J.

16

63
64

65

66

67

68

69

70

71
V2

73

74

75

76

Ci

78

79

80

81

82

83

84

85

86

T. GoToH AND T. SUZUKI

Crewe, A. V. (1983) Science, 221: 325-330.
Mezzasalma, V., Di Stefano, L., Piazzese, S.,
Zagra, M., Salvato, B., Tognon, G. and Ghiretti
Magaldi, A. (1985) Biochim. Biophys. Acta, 829:
135-143.

Stockel, P., Mayer, A. and Keller, R. (1973) Eur.
J. Biochem., 37: 193-200.

Pilz, I., Schwarz, E. and Vinogradov, S. N. (1980)
Int. J. Biol. Macromol., 2: 279-283.

Wilhelm, P., Pilz, I. and Vinogradov, S. N. (1980)
Int. J. Biol. Macromol., 2: 383-384.
Messerschmidt, U., Whilhelm, P., Pilz, I., Kapp,
O. H. and Vinogradov, S. N. (1983) Biochim.
Biophys. Acta, 742: 366-373.

Theuer, M., Pilz, I., Schwarz, E., Wilhelm, P.,
Mainwaring, G. M. and Vinogradov, S. N. (1985)
Int. Biol. Macromol., 7: 25-29.

Pilz, I., Schwarz, E., Suzuki, T. and Gotoh, T.
(1988) Int. J. Biol. Macromol., 10: 356-360.
Suzuki, T. (1989) Eur. J. Biochem., 185: 127-134.
Stern, M., Snow, J. W., Ereifej, K., Mainwaring,
M. G., Vinogradov, S. N. and Walz, D. A. (1987)
Fed. Proc. Fed. Am. Soc. Exp. Biol., 46: 2266.
(Abstr.)

Nishioka, Y., and Leder, A. (1879) Cell, 18: 875-
882.

Go, M. (1981) Nature, 291: 90-92.

Craik, C. S., Buchman, S. R. and Beychok, S.
(1980) Proc. Natl. Acad. Sci. U.S.A., 77: 1384-
1388.

Phillips, S. E. V. and Schoenborn, B. P. (1981)
Nature, 292: 81-82.

Fushitani, K., Imai, K. and Riggs, A. F. (1986) In
“Invertebrate Oxygen Carriers”. Ed. by B. Linzen,
Springer Verlag, Berlin/Heidelberg, pp. 77-79.
Jhiang, S. M., Garey, J. R., and Riggs, A. F.
(1988) Science, 240: 334-336.

Keilin, D. and Hartree, E. F. (1951) Nature, 168:
266-269.

Dickerson, R. E. and Geis, I. (1983) Hemoglobin.
The Benjamin/Cumming Publishing Co., Menlo
Park, CA.

Fushitani, K., Morimoto, H. and Ochi, O. (1982)
Arch. Biochem. Biophys., 218: 540-547.

Suzuki, T., Takagi, T., Furukohri, T. and Gotoh,
T. (1983) Comp. Biochem. Physiol., 75B: 567-
570.

Nei, M., Stephens, J. C. and Saitou, N. (1985)
Mol. Biol. Evol., 2: 66-85.

Goodman, M., Weiss, M. L. and Czelusniak, J.
(1982) Syst. Zool., 31: 375-399.

Tasch, P. (1980) Paleobiology of the Invertebrates.
pp. 441-470. Wiley, New York.

Fushitani, K., Bonaventura, J. and Bonaventura,
C. (1986) Comp. Biochem. Physiol., 84B: 137-
141.

87

88

89

90

91

92

93

94

99

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

Gotoh, T, and Okada, K. (1975) J. Sci. Univ.
Tokushima, 13: 1-7.

Gotoh, T, and Kamada, S. (1980) J. Biochem.
(Tokyo), 87: 557-562.

Guerritore, D., Bonaci, M. L., Brunori, M., Anto-
nini, E., Wyman, J. and Rossi-Fanelli, A. (1965) J.
Mol. Biol., 13: 234-237.

Wiechelman, K. J. and Parkhurst, L. J. (1972)
Biochemistry, 11: 4515-4520.

Vinogradov S. N., Shlom, J. M., Hall, B. C.,
Kapp, O. H. and Mizukami. H. (1977) Biochim.
Biophys. Acta, 492: 136-155.

Walz, D. A., Snow, J., Mainwaring, M. G. and
Vinogradov, S. N. (1987) Fed. Proc. Fed. Am.
Soc. Exp. Biol., 46: 2266. (Abstr.)

Vinogradov S. N., Van Gelderen, J., Polidori, G.
and Kapp. O. H. (1983) Comp. Biochem. Physiol.,
76B: 207-214. :
Vinogradov, S. N. and Orii, Y. (1980) Comp.
Biochem. Physiol., 67B: 183-185.
Pionetti, J-M. and Pouyet, J. (1980)
Biochem., 105: 131-138.

Royer, W. E., Jr., Hendrickson, W. A. (1988) J.
Biol. Chem., 263: 13762-13765.

Lightbody, J. J., Ziaja, E. L., Lugo, S. D., Main-
waring, M. G., Vinogradov, S. N., Shishikura, F.,
Walz, D. A., Suzuki, T. and Gotoh, T. (1986)
Biochim. Biophys. Acta, 873: 340-349.

Royer, W. E., Jr., Hendrickson, W. A. and Love,
W. E. (1987) J. Mol. Biol., 197: 149-153.
Terwilliger, R. C., Terwilliger, N. B. and Schab-
tach, E. (1980) Comp. Biochem. Physiol., 65B:
531-535.

Arp, A. J., Childress, J. J. and Vetter, R. D.
(1987) J. Exp. Biol., 128: 139-158.

Suzuki, T., Takagi, T. and Ohta, S. (1988)
Biochem. J., 255: 541-545.

Jones, M. L., (1985) Biol. Soc. Wash. Bull., 6:
117-158.

Suzuki, T., Takagi, T.
Biochem. J., 266: 221-225.
Suzuki, T., Takagi, T. and Ohta, S. (1990) J. Biol.
Chem., 265: 1551-1555.

Iwaasa, H., Takagi, T. and Shikama, K. (1989) J.
Mol. Biol., 208: 355-358.

Usuki, I., Hino, A. and Ochiai, T. (1989) Comp.
Biochem. Physiol., 93B: 555-559.

Ochiai, T., Enoki, Y. and Usuki, I. (1989) Comp.
Biochem. Physiol., 93B: 935-940.

Antonie, M. and Niessing, J. (1984) Nature, 310:
795-797.

Kobayashi, M. and Hoshi, T. (1984) Zool. Sci., 1:
523-532.

Nagai, K., Perutz, M. F. and Poyart, C. (1985)
Proc. Natl. Acad. Sci. U. S. A., 82: 7252-7255.

Eur. J.

and Ohta, S. (1990)

ZOOLOGICAL SCIENCE 7: 17-28 (1990)

© 1990 Zoological Society of Japan

REVIEW

The Effect of Stress and Disturbance on Echinoderms

JOHN M. LAWRENCE

Department of Biology, University of South Florida, Tampa,
Florida 33620, U. S. A.

1. INTRODUCTION

Grime [1, 2] proposed that two types of external
factors limit biomass. One, stress, was defined as
any factor that limited production; the second,
disturbance, as any factor that partially or totally
caused destruction of biomass. Habitat productivi-
ty and habitat duration, and growth rate and
mortality (parameters of fitness) have been used in
a similar manner [3, 4]. Stress and disturbance can
be either biotic or abiotic. Grime’s definition of
stress is consistent with Levitt’s [5] analogy of
biological to physical stress: a stress produces a
strain. In biological systems, the strain would be a
decrease in production [6]. Levitt [5] pointed out
that stresses in biological systems differ from
physical ones in that the stress must be measured
not in units of force, but in units of energy.
Similarly, since production is best expressed in
units of energy [7, 8], biological stress would result
in a decrease in deposition of energy in the organ-
ism (a decrease in production). It is in this context
that low levels of resources, including food for
animals, may be stresses as they limit production.

Grime [1, 2] identified three primary strategies
in plants, corresponding to three of the four possi-
ble permutations of high and low stress and dis-
turbance (Figs. 1, 2). Distinct types of life-history
strategies would be associated with these three
strategies. Competitive plants (C-strategy) are
adapted to low stress and low disturbance in
habitats of high productivity and long duration;
stress-tolerant plants (S-strategy), to high stress and

Received December 7, 1989

low disturbance in habitats of low productivity and
long duration; ruderal plants (R-strategy, from
Latin meaning “rubble” and thus, growing in a
disturbed habitat), to low stress and high disturb-
ance in habitats of high productivity and short
duration. High stress would prevent recovery in
highly disturbed habitats, and this strategy would
not be possible.

The three primary strategies are the results of
extreme levels of stress and disturbance, and
Grime proposed that intermediate conditions
would result in secondary strategies. Competitive
ruderals (CR strategy) are adapted to low stress,
with competition being restricted by moderate
disturbance; stress-tolerant ruderals (SR strategy),
to lightly disturbed, unproductive habitats; stress-
tolerant competitors (CS strategy), to relatively
undisturbed conditions and moderate intensities of
stress; and CSR species, to habitats with moderate
intensities of both stress and disturbance that
restrict competition, or which have temporal or
spatial variation in intensities of competition,
stress, and disturbance.

A suite of characteristics would be associated
with each strategy (Table 1). According to Grime
[2, 3, 9-11], these characteristics involve the way
in which energy and matter are allocated for
somatic and gonadal production, maintenance,
and protection from disturbance.

Competitors have access to large amounts of
resources over considerable periods of time, and a
high potential for resource capture. In terrestrial
animals, competitors are active foragers over large
areas. In aquatic animals, competitors may also be
less active foragers or even stationary if productiv-

18 J. M. LAWRENCE

Intensity of

stress

(Grime 1979)

growth stress

(Sibly & Calow 1989)

stress-—
tolerator

competitor

ruderal

mortality stress
(Sibly & Calow 1989)

Intensity of
disturbance
(Grime 1979)

F

5
ie

Strategies resulting from combinations of different levels of stress and disturbance.

100Z

competition

1002

disturbance

1002

stress
Fic. 2. Triangular model indicating the strategies resulting from combinations of different levels of stress and
disturbance. C: competitor, S: stress-tolerator, R: ruderal, CS: stress-tolerant competitor, CR: ruderal
competitor, SR: stress-tolerant ruderal, CRS: CRS strategist. (modified from Grime 1977)

ity is high and food is brought to them by currents. _ cause of these characteristics, Grime [3] also refer-

Competitors are fast growing, long-lived, and iter- __ red to this as a “capitalistic” strategy.
Oparous. They reinvest captured resources in Stress-tolerators have access to small amounts of

continued somatic production and activity. Be- resources (either because of the small amounts

Effect of Stress and Disturbance

TaBLE 1. Characteristics of competitive, stress-tolerant, and ruderal echinoderms
Competitive Stree-tolerant Ruderal

Characteristic
Habitat
Potential high low high
productivity
Duration long long short
Stability high high low
Disturbance low low high

Life history
Longevity

Phenology of

long or
relatively short

annual, pronounced

long or very long

many types, may be

short, or very short

early in life, annual

reproduction cycle
Reproductive effort low
Reproductive output high
Developmental type* Dis (222;%)
Physiology

Maximal relative high
potential for

resource capture

Maximal relative rapid
potential growth

rate

Miscellaneous

Defense against relatively specialized,

predation often ineffective
(structural,

chemical,

bchavioral)

Palatability high

intermittent,

continuous, or asexual

low high

low low or high, depending on
body size

lee?) led,

low high

slow rapid

if present, relatively specialized,

constitutive often ineffective

if predation occurs, high

low

* Developmental types [107]: 1, Direct; 2, Indirect: 2.1, planktotrophic and pelagic; 2.2; lecithotrophic and
pelagic, 2.3, lecithotrophic and benthic; 2.4, lecithotrophic and brooded; 2.5, lecithotrophic and viviparous.

available or of their mode of resource capture). In
animals, stress-tolerators are passive in feeding or
have limited active foraging over limited areas.
Resources are captured at a low rate over a long
period or time, or in brief, infrequent, unpredict-
able pulses. They are slow growing, and long or
very long-lived. Although iteroparous, their re-
production may be irregular in time and nature; it
may be suspended with chronic, severe stress.
Ruderals have access to large amounts of re-
sources but their longevity is limited by disturb-
ance (food exhaustion, predation, abiotic factors).
They have rapid growth and early reproduction at
an early age. As they are short lived, they are

semelparous or have limited iteroparity. They
tend to sustain reproduction even when under
stress.

Grime [2, 9] suggested that heterotrophic organ-
isms also have the three primary strategies attri-
buted to autotrophic plants. He placed primary
significance on the response of the organism to
stress, and asserted that evolutionary responses to
stress fall into basic types that correspond to
widely-recurrent ecological strategies that would
allow all organisms, regardless of taxonomic or
trophic group, to be placed in a common
framework of basic functional types. His proposal
has not been systematically evaluated for any

20 J. M. LAWRENCE

major group of heterotrophs. This review applies
the triangular model to the Echinodermata, a
major animal phylum of great importance in the
ocean world.

2. PRIMARY STRATEGIES

2.1 Competitive echinoderms

The only echinoderns that clearly fit the crite-
rion of being active foragers are some asteroids.
Acanthaster planci of the Indo-Pacific coral reefs
may be an example. It is found in a region of high
resource availability and long duration, has a high
potential for resource acqusition, is long lived,
iteroparous with a high fecundity, and has relative-
ly ineffective protection against disturbance [12,
13]. Pycnopodia helianthoides and Meyenaster
gelatinosus are probaly other examples [14, 15].
These are very active species found on the highly
productive west coasts of North and South Amer-
ica respectively where abundant prey are avail-
able.

Among the regular echinoids, the diadematids
have the highest capacity for active foraging but
are not always found in habitats of high production
and low disturbance. The other regular echinoids
are less active. Although many are found in areas
of high productivity, the temporal and spatial
variation in intensities of competition, stress, and
disturbance suggests that they are not competitors
as a primary strategy.

Whether particulate-feeding echinoderms have
the potential for a sufficiently high rate of food
capture to be a competitor as a primary strategy is
debatable. Dense, persistent shallow-water epi-
faunal ophiuroid beds are found in various parts of
the world ocean [16]. These occur where flow and
levels of suspended organic matter are high and
the incidence of predation is low and sublethal.
The dendrochirotid holothuroid Aslia lefevrei
forms dense aggregates off the west coast of
Europe where moderately strong water-flow sup-
plies seston continuously [17, 18]. It can have a
longevity of ca. 10 years or more. As a result of
their susceptibility to predation, few crinoids are
found in shallow-water habitats of high productivi-
ty and low disturbance. Antedon rosaceus may be

an exception.

It is unlikely that deposit-feeding echinoderms
(spatangoids, and aspidochirotid and molpadid
holothuroids) are competitors. The ingestion of
inorganic sediment should reduce their potential
for a high rate of resource capture.

2.2 Stress-tolerating echinoderms

Although their feeding mode may limit the
potential for resource capture in many echino-
derms, those with stress-toleration as a primary
strategy are found where food availability is very
low. This occurs in the deep sea which receives
only 1 to 3% of the surface primary productivity
[19]. Many echinoderm species occur in the deep
sea, and no class is always dominant [20]. All
should be stress-tolerant strategists as a result of
the low availability of food. This is correlated with
their passive or low level of activity in feeding.

Perhaps the best known representatives are the
stalked crinoids, now restricted to deeper waters
[21]. Crinoids are passive, rheophilic suspension-
feeders restricted to low-energy water habitats [22]
and have little potential for capturing even the
limited resources in the deep sea. Meyer and
Macurda [23] suggested that the appearance of
predatory teleosts led to the restriction of stalked
crinoids to the deep sea. It would seem that the
combination of high stress (low capacity for pro-
duction) and disturbance (predation), Grime’s
fourth non-permissible permutation, would be re-
sponsible. Nothing is known of their life histories,
but one would predict that they would be slow-
growing and long-lived.

Growth of deep-sea holothuroids, echinoids and
ophiuroids is slow and the echinoids there are long
lived [24-32] as expected. Echinus affinis are
thought to live up to ca. 28 years. Several
holothuroid, ophiuroid, and asteroid species have
low fecundity [30, 33-36], but not necessarily less
than that characeristic for their taxon, size, and
reproductive mode. The deep-sea pelagic
holothuroid Scotoanassa has occurrence patterns
that have been interpreted to result from repro-
ductive responses to short-term environmental
changes (probably in food availability) [37].

The tropical shallow-water asteroids are micro-
phagous. They may be stressed by the low levels of

Effect of Stress and Disturbance 21

resources available. Blake [38] concluded that the
primarily tropical valvatids and echiniasterids have
extensive development of structural protection
that is related to the high predation pressure in the
region, and suggested that the microphagous mode
of feeding was the consequence of this. To the
contrary, the protection against predation may be
the consequence of the stress of the low rate of
resource capture and the advantage of allocating
resources to increase longevity.

The echinoids Colobocentrotus atratus and
Heterocentrotus mammillatus (Echinometridae)
are found in extremely high-energy water in the
tropics and have a low detrital food supply that
constitutes an additional stress to that of physical
exposure. These species have an extremely heavy
test and spines that constitute a greater proportion
of the body weight [39] and energy [40] than in
most regular echinoids. They also have a long
life-span [39, 41]. Ebert [39] noted that the
survivorship of regular echinoids increases with an
increase in the relative size of the body wall. He
related this to the degree of exposure to high-
energy habitats. The long life-span was associated
with a slow rate of growth. Ebert suggested that
these observations supported the hypothesis that
survival is related to allocation of resources to
maintenance. Rather than attributing the long
life-span and slow growth to the rigors of the
physical environment, one might attribute them to
low resource availability or acquisition.

The suspension-feeding ophiuroid Amphiura
filiformis would seem best classified as a stress-
tolerant species based on estimated life-spans of 20
to 25 years [42, 43]. Yet a number of reports
indicates life spans of only ca. 5 years [see 43].
Indeed, Buchanan [44] concluded that the species
was fast growing and short-lived, dying soon after
reproduction. He contrasted it to Amphiura chia-
Jei, which he found to be slow growing and long-
lived. This difference in characteristics reported
for A. filiformis has been attributed to under-
representation of juveniles in sampling [42]. It
would seem that the maximal longevity found best
indicates the strategy for the species. The high
incidence of sublethal predation of A. filiformis in
Galway Bay that should require a major allocation
of resources [45] is not that expected to be associ-

ated with such a long-lived species. In this regard,
it may be that the other susension-feeding
ophiuroids that form dense, persistent beds but
suffer little predation [16, 46, 47] are also stress-
tolerators. They would be comparable to the
stalked crinoids that apparently cannot exist in
shallow-water because of predation [23] as it com-
bines high stress and disturbance.

Deposit-feeding echinoids (Atelostomacea and
Gnathostomacea), ophiuroids (Ophiocomidae,
Amphiuridae, Ophiactida, | Ophiothrichidae),
holothuroids (Aspidochirotida, Elasipodida, Apo-
dia, Molpadida), and asteroids (some Paxillosida)
ingest sediment that should reduce nutrient levels
and should be stress-tolerators. The ophiuroid
Amphiura chiajei shows no evidence of lethal
predation, has a slow rate of growth, and is
long-lived and iteroparous [43]. The spatangoid
Echinocardium cordatum has a relatively long life-
span estimated to be ca. 15 years [48]. As it lives in
deep burrows [49], it should be relatively free from
predation, but mortality in shallow waters as a
result of storms occurs [50].

It is possible that the Antarctic echinoderms are
stress-tolerant strategists. Pearse et al. [51] char-
acterized the Antarctic shallow-water environment
as being extremely oligotrophic most of the year,
with the shallow benthos being seasonally very
productive and unstable and the deeper benthos
being very stable. They characterized the benthic
invertebrates as having very slow gonadal develop-
ment and somatic growth, and attributed this to
lack of adaptation to the low temperatures. These
characteristics may instead be responses to low
food availability or quality. Either would result in
a decrease in productivity.

Stressed echinoderms may have unusual repro-
ductive characteristics such as little seasonality
(asynchrony within and/or among individuals), in-
termittent reproduction, asexual reproduction,
and direct development over a long life-span. The
asteroid Asterina miniata, which feeds most com-
monly on detrital plant material that may limit its
production [52], seems to have no predators [53]
and has asynchronous reproduction [54]. The
asteroid Ctenodiscus crispatus, which feeds non-
selectively on deposits, shows asynchronous game-
togenesis and aseasonal, continuous reproduction

22 J. M. LAWRENCE

[55]. Shick et al. [55] noted that the reproductive
pattern in the eurybathic C. crispatus is similar to
that of deep-sea echinoderms and suggested that it
may be related to the constancy of the population’s
detrital food source. Similarly, Tyler [19, 29] and
Tyler et al. [56] suggested that the high frequency
of reproductive asynchrony in deep-sea ophiuroids
and asteroids results from the constant level of
food that provides no seasonal cues for reproduc-
tion. To the contrary, the low level of food, and
not its constancy, may be responsible.

Asexual reproduction by fission occurs in a few
species of asteroids, ophiuroids, and holothuroids
[57]. Asexual reproduction in these species seems
to be associated with low availability of food [57,
58] and thus would be a stress response.

2.3 Ruderal echinoderms

Several shallow-water, tropical regular echi-
noids of the family Toxopneustidae (Lytechinus,
Tripneustes) seem to be rudurals. They are found
in habitats of high primary production, have a
rapid growth rate, high reproductive output and
fecundity, and short life-spans that are associated
with both biotic and abiotic disturbance [59-62].

Among the ophiuroids, several species along the
French coast of the English Channel have life
spans of ca. 5 years or less [63-66]. These species
are found in current-washed habitats, and abiotic
disturbance could be the cause of the short longev-
ity in this productive area. The longevity of
Ophiura texturata is less in the Mediterranean Sea
than in the North Sea [66] which may result from a
difference in predation levels.

Some ophiuroids have longevities of only slight-
ly more than 1 year. Ophiothrix fragilis lives ca. 15
months [64]. Amphipholis squamata has a longev-
ity of ca. 1 to 2.5 years [see 47]. The shorter
longevities reported for this species were in tide-
pool populations and were thought to result from
higher stress. The simultaneous hermaphroditism
and brooding habit of this species may be related
to the small size of individuals and rigors of the
habitat. Somatic and gonadal growth occur simul-
taneously in A. squamata, which is not a character-
istic one would predict for a ruderal species as
most resources should be put into reproduction,
but whether the reproductive effort is actually

small is not known.

Amphipholis squamata is found in habitats with
high production [67]. High production would not
be associated with rubble habitats. Hendler and
Littman [68] noted that the rarity of small
ophiuroids in rubble remained to be explained.
The rubble habitat is probably the non-permissible
fourth permutation of Grime: high stress (low food
availability) and high disturbance.

If the potential for high resource capture is a
requisite for the ruderal strategy, it seems contra-
dictory to conclude that some ophiuroids could be
assigned to this primary strategy and not to the
competitive one which also requires that potential.
Differences in sizes or mode of feeding may be the
explanation.

Several species of asteroids have characteristics
of ruderals. Asterina phylactica is a small, simul-
taneous heramphrodite that lives in moderate to
strong currents along the coasts of the British Isles
[69]. It first reproduces at 2 years, produces 1 to 3
broods, and has a longevity of up to 4 years [70].
In contrast, the closely related Asterina gibbosa is
larger, a protandric hermaphrodite that lives in
more sheltered habitats although it overlaps with
A. phylactica. It first reproduces at 4 years,
produces 4 to 7 broods, and has a longevity of 7 or
more years. A. phylactica has a greater reproduc-
tive effort than A. gibbosa, a characteristic of a
ruderal species. Emson and Crump [69, 70] re-
lated the life-history characteristics of A. phylacti-
ca to a more stressful habitat than that of A.
gibbosa but they may be more related to a dis-
turbed habitat.

3. SECONDARY STRATEGIES

3.1 Competitive ruderal echinoderms

Echinoderms with this strategy would have a
high potential to obtain food but would be subject
to moderate disturbance. This restricts the possibi-
lities to asteroids and regular echinoids. Suspen-
sion-feeding ophiuroids and holothuroids in highly
productive habitats may also have this strategy.

Leptasterias hexactis coexists with Pisaster
ochraceus on the west coast of North America.
Whereas P. ochraceus has characteristics of a CS

Effect of Stress and Disturbance 23

strategy, L. hexactis has characteristics of a CR
strategy. It matures at a smaller size at ca 2 years
of age and has a moderate, but shorter longevity
[71, 74]. Menge [72] calculated that the energetic
reproductive output of P. ochraceus was less than
that of L. hexactis, but this is not indicative of the
reproductive effort. He suggested that the small
size and brooding habit of L. hexactis made it more
susceptible to disturbance than P. ochraceus in the
high water-energy intertidal habitat. As L. hexac-
tis is found in a productive region and grows larger
in the absence of P. ochraceus, it may be classified
as having a CR strategy.

Another forcipulatid that seems to experience
high disturbance is Asterias. Asterias amurensis
has a longevity of only ca. 3 years [75], and actually
becomes reproductively mature in less than 1 year
under favorable conditions [76]. Asterias vulgaris
and Asterias forbesi also mature at a small size in
less than 1 year [77-79]. Menge suggested that
massive mortality of these species results from
disease and storms, and that their density conse-
quently is less than the carrying capacity of their
habitat. As with other echinoderms, growth is
greatly affected by the availability of food. One
population of Asterias rubens off the coast of
Brittany grows slowly to a small size and has a
longevity of more than 5 years while another grows
rapidly to a large size and has a longevity of 3 years
[80]. Thus Asterias seems to be another asteroid
with a CR strategy.

3.2. Stress-tolerating ruderal echinoderms

Abiotic stress such as sublethal but non-optimal
levels of temperature, salinity, exposure to air and
wave-energy can affect intertidal and shallow-
water echinoderms. Much stress, however, may be
due to low levels of resources, either in their
availability and/or in the ability of the organism to
obtain it rapidly. The latter case can occur in those
taxa that feed passively or have a low level of
activity, amd also in those whose potential for
feeding is restricted by the potential for predation.

The reproductive characteristics of two species
of Caribbean ophiuroids differ according to their
level of exposure to abiotic disturbance and poss-
ibly to stress [81]. Ophiocomella ophiactoides in
protected areas reproduce asexually except in a

few large individuals. Those in exposed areas have
a high rate of asexual reproduction at all sizes.
Ophiatis savignyi found in sponges reproduce asex-
ually except for the few large individuals. Epiphy-
tic ones are all asexual. Both of these species are
small. The early reproduction at such a small size
could be related to the high level of disturbance.
The asexual reproduction could be an adaptation
to a high level of stress (either low levels of food,
the rigors of a shallow-water tropical habitat, or
from the small body-size itself that limits resource
capture). However, many echinoderms of similar
size reproduce sexually.

Shallow-water tropical crinoids also seem to be
SR-strategists. Crinoids in the tropical shallow-
waters not only are stressed by the low level of
resource availability and their passive mode of
acquiring the resource, but suffer disturbance from
predation as well. Although these crinoids have
some behavioral, structural, and chemical deter-
rents to predation, they are not effective [82].
Predation on the crinoids is usually sublethal and
the lost parts are regenerated.

Suspension-feeding ophiuroids have a similar
situation of stress from low resource availability or
acquisition and of disturbance from predation.
Species found on coral-reef flats, Caribbean la-
goons, and subtidal rocky reefs off the English
coast are cryptic and show high incidences of
sublethal predation [16, 46].

Representatives of several orders of echinoids
are stress-tolerating ruderals. In the clypeasteroids
Mellita and Dendraster the rate of feeding and
growth, longevity, and relative gonadal-somatic
production [83-87] are suggestive of stress-
tolerant ruderals. The regular echinoid Echi-
nometra is found in tropical intertidal and very
shallow-water habitats, and experiences the stress
of exposure and low food availability [88, 89] along
with predation [61, 90, 91].

Intertidal dendrochirotid holothuroids in areas
of low production can have SR characteristics.
Cucumaria curata has delayed sexual maturity (at 3
years), moderate longevity ca. 5 years and low
fecundity [92].

3:3:

These echinoderms would be subject to low

Stress-tolerating competitive echinoderms

24 J. M. LAWRENCE

resource availability but have an active mode of
resource capture. Several asteroids seem to fit this
category. Pisaster ochraceus is mobile, if not
highly active. It has delayed onset of reproduc-
tion, high fecundity, a potential high rate of
growth, and a long life-span [73, 93], but is usually
food limited [93, 94]. A related species Pisaster
giganteus allocates energy to the pyloric caeca and
to maintenance when stress is increased by low
food availability [95].

3.4. CSR echinoderms

These echinoderms should be found where re-
sources are relatively abundant and have an active
mode of feeding, but also experience disturbance.
As the potential for feeding is restricted by the
presence of predators, the availability of the re-
sources may be more apparent than real.

The asteroid Luidia clathrata is an active pre-
dator [96] that is subject to arm loss that affects its
reproductive capacity [97]. It has a nocturnal
feeding cycle, presumably predator-induced, that
results in its being resource limited in the field [96,
98]. With limited food, it allocates resources to
nutrient storage and reproduction rather than to
arm regeneration [99]. With abundant food, it
allocates resources to both arm regeneration and
growth of the internal organs [100]. With limited
food, not only is development of the gonads re-
duced, but also the capacity for egg fertilization
and development [101].

The regular echinoids associated with lush algal
floras of temperate waters are CSR strategists.
They show temporal and spatial variation in inten-
sities of competition, stress, and disturbance.
Strongylocentrotus species have been studied most
[see 53, 102-105]. These species have delayed
onset of reproduction followed by annual repro-
duction for a number of years, a low gonadal:
somatic production but a high potential maximal
rate of feeding and production. Their protection
against predation is not highly effective and their
life-history characteristics are greatly affected by
the level of predation. Predation by fish and sea
otters on Strongylocentrotus species can be great
on the west coast of North America. In this
situation, they have low stress and low or moder-
ate disturbance. Without this predation, stress

may become extreme as the strongylocentrotids
eliminate their food resource by overgrazing. In
this situation, they have high stress and low dis-
turbance, and adopt the foraging strategy of
“mobile-active search” rather than the “sit-and-
wait” one used when food is abundant. An in-
crease in the relative size of the Aristotle’s lantern
in Strongylocentrotus purpuratus in such situations
is thought to be an adaptive phenotypic response
[106]. Strongylocentrotids survive low food availa-
bility with no or minimal somatic and gonadal
production. Other Echinidae (Echinus, Loxechi-
nus, Paracentrotus, Parechinus, Psammechinus,
Sterechinus), and some Echinometridae (Helioci-
daris, Evechinus) also seem to be CSR-strategists.

4. DISCUSSION

Grime’s model predicts that species adapted to
different types of habitats will have different suites
of characters. These characters sort into primary
and secondary strategies involving differences in
life history and physiology (Table 1). Life history
characters involve reproduction and mortality.
Physiological characters involve feeding, growth,
and maintenance activities. Competitive and
ruderal species are similar physiologically in the
high potential for resource capture and growth
rates, but differ in life histories as competitive
species have greater longevity, a smaller reproduc-
tive effort, and a higher fecundity. Competitive
and stress-tolerant species overlap in longevity
(although the latter may be very long-lived) and
are similar in their small reproductive effort, but
stress-tolerant species have a lower potential for
resource capture and growth rate, and a lower
fecundity.

The criterion of longevity can be used to sepa-
rate ruderal echinoderms from the competitive and
stress-tolerant echinoderms. Ruderal, CR, andSR
echinoderms seem to have life-spans of up to 5
years; competitive, CR, and CS echinoderms,
from 5 to 15 years; and stress-tolerant and CS
echinoderms, of many more years. The overlap-
ping of longevity of the echinoderms in the differ-
ent groups except for the extremes makes use of
longevity as a criterion difficult.

Basic critera separating the groups should in-

Effect of Stress and Disturbance 25

volve size at first reproduction, reproductive
effort, reproduvtive output, and fecundity. It is
important to recognize that reproductive effort is
the proportion of the energy acquired by an organ-
ism that is allocated to reproduction. This is
different from the reproductive output, which is
simply the amount of energy allocated to repro-
duction, even if expressed in terms of body weight.
Competitive echinoderms have a high reproduc-
tive output and fecundity. Ruderal echinoderms
may have these characteristics also, but only in
those that attain a large size. CS, and CR echi-
noderms similarly vary in their reproductive out-
put and fecundity with individual size. Those few
values for CS and CR echinoids suggest that
reproductive effort is low. They continue to grow
asymptotically. Fecundity is the result of both egg
size and number, and is high in these groups. In
contrast, fecundity is low in SR echinoderms, often
involving the production of fewer, larger eggs and
culminating, in those groups where the potential
exists, in a few fissiparous species that produce a
single new individual with asexual reproduction.
The body form and location of the gonads may
limit fecundity in crinoids and ophiuroids and
affect their potential for the competitive or ruderal
strategies.

The potential to acquire resources is the result of
the quantity or quality of food and/or the ability to
acquire it. The latter can result from the level of
activity and/or effectiveness of the feeding
mechanism. A basic criterion that immediately
separates echinoderms is whether they are active
or passive or relatively inactive feeders. Active
feeders, defined as those that actively forage over
large areas, are found only in Asteroidea and
regular Echinoidea. Only they have the functional
morphology that enables this behavior. The other
groups range from the completely passive (suspen-
sion-feeding stalked Crinoidea and functionally
pelmatozoan Dendrochirotida and Ophiuroidea)
to the minimally vagile (suspension-feeding Com-
atulida, and some Ophiuroidea, Dendrochirotida,
and deposit-feeding Atelostomacea, Gnatho-
stomacea, Aspidochirotida, Molpadida, and
Ophiuroidea). This low level of activity correlated
with microphagous feeding limits the potential for
resource acquisition and often involves ingestion

of low quality food.

Although this major division can be made on a
broad scale, individual species within any taxon
with characteristic size, functional morphology, or
trophic type, will differ in their ability to acquire
food. Thus, species in groups that are passive or
relatively inactive in feeding can increase their
feeding ability and/or be found in habitats of high
production that increase their capacity to acquire
resources. This is the probable basis for the
appearance of some suspension- or deposit-feeding
echinoderms as competitors or ruderals. For ani-
mals in general, it is necessary to take into account
the particular characteristics of feeding behavior
and predation in relation to production [107].

The maximal potential relative growth rate is
another criterion that separates the competitive
and ruderal strategies from the stress tolerant
strategy. As echinoderms do not always recruit to
habitats in which a maximal growth rate results, a
wide range of growth rates for several species has
been reported. One generally concludes that the
fastest rate is the maximal potential rate.
Although growth rates are usually based on linear
dimensions or weights, comparisons of groups that
differ in body forms, structures, and composition
are difficult. If production is the criterion, growth
should be measured in energy units. In no other
way can the growth rate of a holothuroid be
compared to that of an asteroid, or a regular
echinoid to a spatangoid. Competitive, CS, CR,
and ruderal echinoderms within taxa do have more
rapid growth rates.

A basic characteristic of echinoderms important
to understanding their strategies is their overall
susceptibility to predation. Fish and crustaceans
are the primary predators of echinoderms although
some gastropods are also. Defense against preda-
tion can be structural, chemical, or behavioral.
Structural and chemical defense uses energy that is
not available for growth in size or for reproduc-
tion. Behavioral defense invarably reduces feed-
ing and consequently production. As great long-
evity is important to the strategy, the presence of
an effective defensive system against predation is
found in stress-tolerant echinoderms. Less effec-
tive systems are found CS, CR, and ruderal echi-
noderms. Those stress-tolerant echinoderms

26 J. M. LAWRENCE

found in habitats that have less predation, such as
the deep-sea, seem to lack structual defenses and
should lack chemical ones as well. Similarly, those
spatangoids burrowing at a depth sufficient to
avoid predation do not have deterrent spines.

5. CONCLUSION

Studies on echinoderms have not considered the
life-history and physiological characteristics in
terms of the relative levels of stress and disturb-
ance that echinoderms experience. Grime’s model
provides a context that can be used to interpret
these characteristics. Most of the data available
for echinoderms involve life-history characteristics
such as longevity and reproductive output.
Fecundity has not been documented for many
species. An important characteristic that has not
been studied for many species is reproductive
effort. Most studies of the pertinent physiological
characteristics involve the rate of growth, but even
these are primarily for regular echinoids and aster-
oids and almost invariably are not in energetic
units. Few studies except for regular echinoids and
asteroids report the potential for resource capture
and the quantify the availability of food. Data on
both the life-history and physiological characteris-
tics of a species are necessary before a complete
understanding of their way of life is possible.

REFERENCES

Grime, J. P. (1974) Nature, 250: 26-31.

Grime, J. P. (1977) Am. Nat., 111: 1169-1194.

Grime, J. P. (1989) Biol. J. Linn. Soc., 37: 3-17.

Sibley, R. M. and Calow, P. (1989) Biol. J. Linn.

Soc., 37: 101-116.

5 Levitt, J. (1972) Responses of Plants to Environ-
mental Stress, Academic Press, New York.

6 Fry, F. E., (1947) Publ. Ontario Fish. Res. Lab.,
68: 1-62.

7 Brody, S. (1945) Bioenergetics and Growth. Haf-
ner, New York.

8 Winberg, G. G. (1956) Fish. Res. Bd. Can. Trans.
Ser. No. 194.

9 Grime, J. P. (1979) Plant Strategies and Vegeta-
tion, John Wiley & Sons, Chichester.

10 Grime, J. P., Cuick, J. C. and Rincon, R. E. (1986)

In “Plasticity in Plants”. Ed. by D. H. Jennings and

A. J. Trewaras, Soc. Exp. Biol., Cambridge, pp.

WN Re

20

21

22

23

24

25
26

ay)

28

29

30

31

32

38

34

35

36

37

7-29.

Grime, J. P., Hodgson, J. G. and Hunt, R. (1988)
Comparative Plant Ecology. Unwin Hyman,
London.

Birkeland, C. (1989) Echinoderm Studies, 3: 1-79.
Moran, P. J. (1986) Oceanogr. Mar. Biol. Ann.
Rev., 24: 379-480.

Dayton, P. K. (1975) Ecol. Monogr., 45: 137-159.
Dayton, P. K., Rosenthal, R. J., Mahen, L. C.,
and Antezana, T. (1977) Mar. Biol., 39: 361-370.
Aronson, R. B. (1989) Ecology, 70: 856-865.
Costelloe, J. (1988) Mar. Biol. 99: 535-545.
Costelloe, J. and Keegan, B. F. (1984) Mar. Biol.
84: 135-142.

Tyler, P. A. (1988) Oceanogr. Mar. Biol. Ann.
Rev., 26: 227-258.

Sibuet, M. (1987) Structure des Peuplements
Benthiques en Relation avec les Conditions
Trophiques en Milieu Abyssal sans 1’°Ocean Atlan-
tique, Thesis, Univ. de Paris VI.

Roux, M. (1987) Echinoderm Studies, 2: 1-53.
Meyer, D. L. (1982) In “Echinoderm Nutrition”.
Ed. by M. Jangoux and J. M. Lawrence, Balkema,
Rotterdam, pp. 25—42.

Meyer, D. L. and Macurda, D. B., Jr. (1977)
Paleobiology, 3: 74-82.

Gage, J. D. (1984) J. Mar. Biol. Ass. U. K., 64:
157-170.

Gage, J. D. (1987) Mar. Biol., 96: 19-30.

Gage, J. D. and Tyler, P. A. (1981) Mar. Biol., 64:
163-172.

Gage, J. D. and Tyler, P. A. (1982) Oceanol.
Acta, 5: 73-83.

Gage, J. D. and Tyler, P. A. (1985) Mar. Biol., 90:
41-53.

Tyler, P. A., (1980) Oceanogr. Mar. Biol. Ann.
Rev., 18: 125-153.

Tyler, P. A. and Gage, J. D. (1979) In “Cyclic
Phenomena in Marine Plants and Animals”. Ed. by
E. Naylor and R. G. Hartnoll, Pergamon, Oxford,
pp. 215-222.

Tyler, P. A. and Gage, J. D. (1980) Oceanol. Acta,
3: 177-185.

Tyler, P. A. Muirhead, A., Billett, D. S. M., and
Gage, J. D. (1985) Mar. Ecol. Prog. Ser., 23: 269-
277.

Tyler, P. A. and J. D. Gage (1982) J. Mar. Biol.
Ass. U. K., 62: 45-55.

Tyler, P. A. and J. D. Gage (1983) J. Mar. Biol.
Ass. U. K., 63: 609-616.

Tyler, P. A., Billett, D. S. M. and Gage, J. D.
(1987) J. Mar. Biol. Ass. U. K., 67: 385-397.
Tyler, P. A. and Pain, S. L. (1982) J. Mar. Biol.
Ass. U. K., 62: 869-887.

Childress, J. J., Gluck, D. L., Carney, R. S. and

38
39
40

4
42
43
44
45
46
47
48
49
50
51

52

53

54

35

56

57

58

59

60

61

62

63

64

Effect of Stress and Disturbance 27

Gowing, M. M. (1989) Limnol. Oceanogr., 34:
913-930.

Blake, D. B. (1983) Bull. Mar. Sci., 33: 703-712.
Ebert, T. A. (1982) Ecol. Monogr., 52: 353-384.
Lawrence, J. M. (1985) In “Echinoderms”. Ed. by
B. F. Keegan and B. D. S. O'Connor, Balkema,
Rotterdam, pp. 47-67.

Ebert, T. A. (1975) Am. Zool., 15: 755-775.
Muus, K. (1981) Ophelia, 20: 153-168.
O’Connor, B., Bowmer, T. and Grehan, A. (1983)
Mar. Biol., 75: 279-286.

Buchanan, J. (1964) J. Mar. Biol. Ass. U. K., 44:
565-576.

Bowmer, T. and Keegan, B. F. (1983) Mar. Biol.,
74: 65-71.

Aronson, R. B. (1987) Paleobiology, 13: 187-192.
Aronson, R. B. (1988) Bull. Mar. Sci., 43: 93-97.
Buchanan, J. (1966) J. Mar. Biol. Ass. U. K., 46:
97-114.

Nichols, D. (1959) Pubs. System. Ass. Lond., 3:
61-80.

Beukema, J. J. (1985) Neth. J. Sea Res., 19: 129-
134.

Pearse, J. S., J. B. McClintock and I. Bosch.(1990)
Amer. Zool.(In press)

Gerard, V. A. (1976) Some Aspects of Material
Dynamics and Energy Flow in a Kelp Forest in
Monterey Bay, California, Thesis, Univ. of Cali-
fornia, Santa Cruz.

Harrold, C., and Pearse, J. S. (1987) Echinoderm
Studies, 2: 137-233.

Farmanfarmaian, A., Giese, A. C., Boolootian, R.
A. and Bennett, J. (1958) J. Exp. Zool., 138: 355-
367.

Shick, J. M., Taylor, W. F. and Lamb, A. N.
(1981) Mar. Biol., 63: 51-66.

Tyler, P. A., Pain, S. L. and Gage, J. D. (1982) J.
Mar. Biol. Ass. U. K., 62: 57-69.

Emson, R. H. and Wilkie, I. E. (1980) Oceanogr.
Mar. Biol. Ann. Rev., 18: 155-250.

Lawrence, J. M. (1987) A Functional Biology of
Echinoderms, Croom and Helm, London.
Bacolod, P. T. and Dy, D. T. (1986) Philip. Sci.,
23: 1-14.

Dafni, J. and Tobol, R. (1986/87) Isr. J. Zool., 34:
191-204.

Hendler, G. (1977) Proc., Third Internat. Coral
Reef Symp., Univ. of Miami, pp. 217-223.
Scheibling, R. E. and Mladenov, P. V. (1988)
Mar. Fish. Rev., 49: 62-69.

Bourgoin, A. (1987) Ecologic et Demographie
d’Acronida branchiata (Montague) (Echinoderma-
ta: Ophiuroidea) en Baie de Douarnenez (Bre-
tagne), Thesis, Univ. de Bretagne Occidentale.
Davoult, D. (1989) Vie Mar., Hors serie no. 10:

65

66

67
68
69
70
71
12
73
74

75

76

77

78

79)

80

81

82
83

84

85

86

87

88

89

90

91

92
93

116-127.

Gentil, F. and Zakardjian, B. (1989) Vie Mar.,
Hors serie no. 10: 148.

Guillou, J. and Robert, R. (1979) In “Echi-
noderms: Present and and Past”. Ed. by M. Jan-
goux, Balkema, Rotterdam, pp. 171-177.

Emson, R. H. and Whitfield, P. J. (1989) J. Mar.
Biol. Ass. U. K., 69: 27-41.

Hendler, G. and Littman, B. S. (1986) Coral
Reefs, 5: 31-42.

Emson, R. H. and Crump, R. G. (1984) J. Mar.
Biol. Ass. U. K., 64: 35-53.

Emson, R. H. and Crump, R. G. (1979) J. Mar.
Biol. Ass. U. K., 59: 77-94.

Menge, B. A. (1972) Ecology, 53: 635-644.
Menge, B. A. (1974) Ecology, 55: 84-93.
Menge, B. A. (1975) Mar. Biol., 31: 87-100.
Menge, J. L. and Menge, B. A. (1974) Ecology,
44: 189-209.

Hatanaka, M. and Kosaka, M. (1958) Tohoku J.
Agricul. Res., 9: 159-178.

Nojima, S., Soliman, F. E1-S., Kondo, Y., Kuwa-
no, Y., Nasu, K. and Kitajima, C. (1986) Pub.
Amakusa Mar. Biol. Lab., 8: 90-112.

Galtsoff, P. S. and Loosanoff, V. L. (1939) Bull.
Bur. Fish. U. S., 49: 75-132.

Menge, B. A. (1979) Oecologia, 41: 245-272.
Menge, B. A. (1986) Bull. Mar. Sci., 39: 467-476.
Guillou, M. and Guillaumin, A. (1985) In “Echi-
nodermata”. Ed. by B. F. Keegan and B. D. S.
O’Connor, Balkema, Rotterdam, pp. 513-521.
Mladenov, P. V. and Emson, R. H. (1988) Mar.
Erol. Prog. Ser., 42: 181-194.

Meyer, D. L. (1985) Paleobiology, 11: 154-164.
Birkeland, C. and Chia, F. -S. (1971) J. Exp. Mar.
Biol. Ecol., 6: 265-278.

Ebert, T. A. and Dexter, D. M. (195) Mar. Biol.,
32: 397-407.

Lane, J. M. (1977) Bioenergetics of the Sand
Dollar, Mellita quinquiesperforata, Theses, Univ.
of South Florida, Tampa.

Morin, J. G., Kastendiek, J. E., Harrington, A.
and Davis, N. (1985) Mar. Ecol. Prog. Ser., 27:
163-185.

Timko, P. L. (1976) Biol. Bull., 151: 247-259.
Lawrence, J. M. and Kafri, J. (1979) Mar. Biol.,
52: 87-91.

Russo, A. R. (1977) Aust. J. Mar. Freshwater
Res., 28: 693-702.

McClanahan, T. R. and Muthiga, N. A. (1989) J.
Exp. Mar. Biol. Ecol., 126: 77-94.

Menge, B. A. and Lubchenko, J. (1981) Ecol.
Monogr., 51: 429-450.

Rutherford, J. C. (1973) Mar. Biol., 22: 167-176.
Feder, H. (1970) Ophelia, 8: 161-185.

28
94

95

96

97

98

99
100

J. M. LAWRENCE

Castilla, J. C. and Paine, R. T. (1987) Rev.
Chilena Hist. Nat., 60: 131-151.

Harrold, C. and J. S. Pearse (1980) J. Exp. Mar.
Biol. Ecol., 48: 169-183.

McClintock, J. B. (1984) An Optimization Study
on Feeding Behavior of Luidia clathrata (Say)
(Echinodermata: Asteroidea), Univ. of South Flor-
ida, Tampa.

Lawrence, J. M. and Dehn, P. F. (1979) Fla. Sci.,
42: 9-12.

Lawrence, J. M., Klinger, T. S., McClintock, J. B.,
Watts, S. A., Chen, C. -P., Marsh, A. and Smith,
L. (1986) J. Exp. Mar. Biol. Ecol., 102: 47-53.
Lawrence, J. M., and Ellwood, A. (unpubl.)
George, S., Lawrence, J. M. and Fenaux, L.
(unpubl. )

101

102

103

104

105

106

107

Lawrence, J. M. (1975) Oceanogr. Mar. Biol.
Ann. Reyv., 13; 213-286.

Lawrence, J. M. and Sammarco, P. W. (1982) In
“Echinoderm Nutrition”. Ed. by M. Jangoux and
J. M. Lawrence, Balkema, Rotterdam, pp. 331-
371.

Estes, J. A. and Harrold, C. (1988) In “The
Community Ecology of Sea Otters”. Springer-
Verlag, Berlin, pp. 116-150.

104 Ebeling, A. W. and Laur, D. R. (1988) In
“The Community Ecology of Sea Otters”. Sprin-
ger-Verlag, Berlin, pp. 169-191.

Ebert, T. A. (1980) Bull. Mar. Sci., 30: 467-474.
Dickie, L. M., Kerr, S. R., Boudreau, P. R.,
(1987) Ecol. Monogr., 57: 233-250.

Chia, F. S. (1974) Thal. Jugoslav., 10: 121-130.

ZOOLOGICAL SCIENCE 7: 29-38 (1990)

A Comparative Study of the Gill Morphometry in the
Mudskippers-Periophthalmus chrysospilos, Boleophthalmus
boddaerti and Periophthalmodon schlosseri

Wai P. Low, YuEN K. IP AND Davip J. W. LANE

Department of Zoology, National University of Zoology, Kent Ridge,
Singapore 0511, Republic of Singapore

ABSTRACT—The gill and skin morphometries of three mudskippers-Periophthalmus chrysospilos,
Boleophthalmus boddaerti and Periophthalmodon schlosseri were studied and compared. Correlations
were made between the morphometric parameters of these respiratory surfaces and the different
terrestrial and aquatic affinities of the mudskippers with special reference to their capabilities to respire
terrestrially.

The natural preference of B. boddaerti for an aquatic environment can be explained by their having
the longest and greatest number of gill filaments amongst the three mudskippers studied. It also has the
largest gill area and gill area: skin area ratio, indicating a greater role of its gills than its skin in
respiration.

Both P. chrysospilos and P. schlosseri have relatively lesser affinity to water as their gills are not well
adapted for aquatic respiration. P. chrysospilos has the smallest total number of filaments and shortest
total filament length. It also has smaller gill area: skin ratio than B. boddaerti, and exhibits the most
rapid increase in skin area with respect to body weight. Thus, its skin appears to have a greater role in
gaseous exchange than its gills. In P. schlosseri, the gill area : skin area ratio increases as the fish grows,
suggesting that small fish depends more on its skin for gaseous exchange whilst branchial respiration is

© 1990 Zoological Society of Japan

more important in bigger individuals.

INTRODUCTION

Much of the information available on_air-
breathing fish generally deals with their diverse
respiratory adaptations to their mode of life.
Accessory breathing organs of air-breathing fish
eg. Channa, Saccobranchus, Clarias, Anabas [1]
and Heteropneustes fossilis [2] have been examined
and described. The various respiratory areas of
air-breathing fish have also been measured [1, 3].
In general, the gill area in air-breathing fish is
smaller than non air-breathing aquatic fish. Even
air-breathing fish which never leave water have
relatively small gill area [4, 5].

Much interest has also been centered on amphi-
bious fishes, mainly the mudskippers. Mudskip-
pers are unique in being some of the most terrest-
rial fishes [6]. Hence, most of the information

Accepted February 20, 1989
Received October 11, 1988

collected on mudskippers had placed major emph-
asis on their physiological [7-10] and biochemical
[11-15] adaptations to terrestrial exposure.

The three major genera of mudskippers in Sing-
apore are Periophthalmus chrysospilos, Boleoph-
thalmus boddaerti and Periophthalmodon schlos-
seri. They live in the same vicinity at Pasir Ris
estuary, off the east coast of Singapore, but differ
markedly in behavior and microhabitat. B. bod-
daerti and P. schlosseri are found on the intertidal
zone of the mudflats whereas P. chrysospilos in-
habits the littoral zone of the seashore nearby. Of
the three mudskippers, B. boddaerti has the
greatest affinity for water. At low tide, it is found
on the mudflats and it periodically enters the
water. But, as the tide rises, it retreats into its
burrows which are on the lower region of the
mudflats and remains submerged until the tide
ebbs. P. schlosseri and P. chrysospilos have less
affinity for water compared to B. boddaerti. At
low tide, P. schlosseri comes onto the mudflats; at

30 Low, W. P., Y. K. Ip AnD D. J. W. LANE

high tide, they are usually found swimming along
the water’s edge with their snout above water. P.
chrysospilos is almost invariably found on land
next to water at both high and low tides. When
agitated, B. boddaerti dives and remains sub-
merged for some time whereas P. schlosseri and P.
chrysospilos skim away at the water surface
through several bounces. Low et al. [16] studied
the gill morphologies of these three genera of
mudskippers in Singapore by scanning electron
microscopy and reported their structural adapta-
tions to be very different.

One of the problems mudskippers face upon
terrestrial exposure is that their gills may collapse
and their secondary lamellae tend to coalesce,
resulting in a major reduction in functional respira-
tory area available for gaseous and ionic exchange.
The gill morphometry of the mudskippers B. bod-
daerti [17, 18], B. chinensis and P. cantonensis [19]
have been studied. Although the gill mor-
phometry of B. boddaerti from the Arabian Gulf
has been studied by Hughes and Al-Kadhomiy
[18], the behavior reported in their investigation
was very different from that observed in our local
species. In contrast to the local B. boddaerti, the
specimens studied by Hughes and Al-Kadhomiy
[18] were found at the water’s edge during high
tide. Moreover, the local mudskipper is herbivor-
ous [20] whereas the Arabian Gulf B. boddaerti is
reported to be omnivorous. Since no detailed
information on the gills of P. schlosseri is available
and the most recent report on those of P. chrysos-
pilos concerned more about the gill morphology
than morphometry [16], the presence of the three
local mudskippers in Pasir Ris therefore presented
the authors with a unique opportunity to compare
their gill morphometries and skin areas in relation
to body sizes and their variable capabilities to
respire terrestrially. It is hoped that this study can
explain the very different behavioral strategies of
these mudskippers in their natural habitats.

MATERIALS AND METHODS

Mudskippers ranging from young to fully grown
adults, (2 to 13 g for P. chrysospilos, 2 to 35 g for
B. boddaerti and 3 to 111 g for P. schlosseri) were
captured from August to September in 1986, at

Pasir Ris, Singapore. These mudskippers were
identified according to Khoo [20]. Normal breed-
ing period of the mudskippers in Singapore was
observed to be between May and July. No
measurement was made on gravid specimens and
no attempt was made to separate the sexes. They
were maintained in the laboratory in 50% seawater
(18% salinity). Fish were killed by pithing, lightly
blotted dry and their weights recorded by a Shi-
madzu Libror EB 280M electronic animal balance
to the nearest milligram.

Skin morphometry

The skin area was obtained by rolling one side of
the mudskipper flat onto a piece of paper and
tracing its outline. This outline was retraced onto a
piece of paper of even thickness, cut out, and its
area determined. This value was doubled to obtain
the bilateral skin area of the specimen.

Gill morphometry

Dissected gills were rinsed with 0.85% NaCl
solution and immersed in 5% formalin made with
50% sea water. Measurements were made on all
four right gill arches. Exposure of the gills to
formalin for various periods of time might cause
artificial changes to some of the morphometric
parameters measured, especially the secondary
lamellar area. Therefore, attempt was made to
standardise the period of exposure of all the gill
samples to formalin before measurement of a
specific gill parameter was made. Such procedure
ensured that comparison of the specific gill para-
meters between specimens was valid. The total
number of filaments were counted after preserva-
tion for one day. On the second day, the total
filament length was measured, whilst the number
of secondary lamellae/mm and lamellar area were
obtained on the third and fourth day respectively.

The method of filament sampling was according
to Hughes [21]. The gill parameters were obtained
from measurements of every fifth filament in P.
chrysospilos and B. boddaerti. In P. schlosseri,
however, a substantial number of branched fila-
ments occurred at the centre of the arch [16]. If
measurements of the filament length were made on
every fifth filament, some of them being branched
and others not, it would give rise to a considerable

Gill Morphometry in Mudskippers 31

amount of error. Therefore, measurements of
filament length were made on every filament on all
four right arches in this mudskipper. To measure
the number of secondary lamellae/mm, two counts
were made near the base and tip of the unbranched
or branched filaments. This was performed on
every other filament on all four arches. These
values were averaged and doubled as there were
secondary lamellae on both sides of the filament.
They were next multiplied by the total filament
length of that gill arch to obtain the number of
secondary lamellae.

To determine the secondary lamellar area, the
arches of all three species were stained with
methylene blue. Four lamellae were excised from
the base, four from the tip and four from the
middle of the filaments. These were mounted on
glass slides. Their outlines were traced by a WILD
camera lucida onto even thickness paper which
was cut, and the average bilateral areas calculated.
This was performed on every fifth filament of P.
chrysospilos and B. boddaerti from which the
filament length was obtained. For P. schlosseri,
three filaments were chosen for this determination-
one from the centre (usually branched), and two
from the mid-point between the centre and the
dorsal and ventral aspects of the arch. The aver-
aged bilateral lamellar area was then multiplied by
the number of secondary lamellae to give the gill
area of that particular arch. This value was next
doubled to account for the left arch as well. The
total gill area was obtained by summimg the gill
area of all the gill arches.

RESULTS

The results of the skin and gill parameters fit the
logarithmic equation:

Log Y=log a+b log W
or Y=aw?

where Y is the parameter measured; W=weight of
the fish; a=intercept on the Y axis giving the
parameter for a 1 g fish; b=regression coefficient
(slope), with correlation coefficients (r) greater
than 0.95 for most of the parameters measured
(Table 1). The standard deviations of the Y inter-
cept (S,) and the slope (S,), are also given in the

same table. Unlike most gill morphometric stu-
dies, various dimensions for 1, 10 and 100g fish
and 95% confidence limits are not given as these
values can be calculated from the equations in
Table 1. Bilogarithmic plots of the skin and
various gill parameters against body weight are
presented in Figures 1, 2, 3 and 4.

The total filament number and total filament
length decrease in the order B. boddaerti>P.
schlosseri >P. chrysospilos (Fig. 1, Table 1). From
the bilogarithmic plots of secondary lamellae/mm
against body weight, it can be seen that the de-
crease in this parameter is slight in P. schlosseri
and B. boddaerti but marked in P. chrysospilos
(Fig. 2, Table 1). Although the total number of
secondary lamellae (N) decrease in the order B.
boddaerti > P. schlosseri>P. chrysospilos (Fig. 2),
the slope of increase in this gill parameter with
body size is the greatest in P. schlosseri (Table 1).
The regression coefficient in enlargement of bi-
lateral secondary lamellar area (bl) with body

®
a
E
S
ec -_
~~ ©
c O
o—
— x
sé
cS
IS
3 \0
o
Cc
x
‘=
o -_

fo)
zo
QE

| 10 100
Body weight (g)
Fic. 1. Bilogarithmic plot of total filament number and

total filament length of P. chrysospilos (@), B.
boddaerti (4) and P. schlosseri (@) against body
weight.

32 Low, W. P., Y. K. Ip anp D. J. W. LANE

Secondary
lamellae 7mm

Total secondary lamellae
(x 10,000)

area (mm@)

Secondary lamella

| He)

Kexe)

Fic. 2. Bilogarithmic plot of number of secondary
lamellae/mm on one side of the filament, total
secondary lamellae and average secondary lamella
area of P. chrysospilos (@), B. boddaerti (4) and P.
schlosseri (@) against body weight.

TABLE 1.
P. schlosseri

fe)
fe)
fe)

100

Gill and weight-specific gill areas (xlO mm*)

l 10 100
Body weight (g)

Fic. 3. Bilogarithmic plot of gill area P.
chrysospilos(@), B. boddaerti (4) and P. schlosseri
(™) against body weight. Open symbols represent
the corresponding weight-specific gill areas of the
mudskippers.

Summary of equations obtained from regression analysis of the

Parameter

Skin area (mm?)

Weight-specific skin
area (mm7/g)

Total number of
filaments

Total filament length
(mm)

Number of secondary
lamellae/mm (one side)

Total number of
secondary lamellae

Average bilateral secondary
lamella area (mm?)

Gill area (mm?)

Weight-specific gill
area (mm/7/g)

P. chrysospilos Sa Sp Tr
Log SA=2.6423+0.8077 log W 0.9985
SA=439 W807 1.0319 0.01582
Log S=2.6206—0.1696 log W 0.9659
S=417 w-° 169 1.0345 0.01712
Log F=2.3376+0.0629 log W 0.4517
F=218 Ww? 1.0975 0.04689
Log L=2.0568+0.5033 log W 0.9788
L=114 Ww? 1.0819 0.03968
Log n=1.4802—0.2367 Log W 0.9844
n=30.2 W-°?367 1.0322 0.01599
Log N=3.8209+0.2933 log W 0.8933
N=6620 Ww??? 1.1171 0.05584
Log bl=—1.8275+0.6456 log W 0.9990
bl=0.01488 w®-6° 1.0211 0.01051
Log GA=1.9895+0.9577 log W 0.9813
GA=97.6 W977 1.1518 0.07126
Log G=1.9679—0.0198 log W 0.1095
G=92.9 W~ 2.0198 1.1436 0.06765

Gill Morphometry in Mudskippers

fo)
fe)

Skin and weight-specific skin areas (xl0O mm?)

Body weight

(g)

Fic. 4. Bilogarithmic plot of skin area of P. chrysospi-
los (@), B. boddaerti (4) and P. schlosseri (@)
against body weight. Open symbols represent the
corresponding weight-specific skin area of the mud-
skippers.

33

weight is smallest in B. boddaerti and greatest in P.
chrysospilos (Fig. 2, Table 1).

B. boddaerti has larger gill area than P. schlos-
seri and P. chrysospilos (Fig. 3), but smaller slopes
of increase in N and bl with respect to weight.
These result in its smallest slope of increase in gill
area, and hence its largest slope of decrease in
weight-specific gill area with body size among the
three mudskippers (Fig. 3, Table 1). In contrast,
P. chrysospilos and P. schlosseri have regression
coefficients of increase in gill area with body
weight approaching 1, indicating only slight change
in weight-specific gill area as the fish grows larger
(Fig. 3, Table 1).

The slopes of increase in skin area with body
weight increase in the order B. boddaerti>P.
schlosseri>P. chrysospilos (Fig.4, Table 1).
Calculated gill area: skin are ratios are plotted
against body weight in Figure 5. The ratios for P.
chrysospilos, which range from 0.27 to 0.36, are
smaller than those for B. boddaerti (0.67 to 0.77).
P. schlosseri is unusual in having ratios that in-
crease from 0.22 in small specimens to 0.50 in
larger ones (Fig. 5).

A comparison has been made between some gill
parameters of the mudskippers with those of va-
rious fish species (Table 2) and mudskippers
(Table 3) reported elsewhere.

skin areas and the various gill parameters with body weight of P. chrysospilos, B. boddaerti and

B. boddaerti Sa Sb r P. schlosseri Sh Sb I
Log SA=2.9250+0.5299 log W 0.9990 Log SA=2.8015+0.6889 log W 0.9985
SA=841 w?°29° 1.0197 0.00753 SA =633 W?-688? 1.0559 0.01627

Log S=2. 9252— -0.4701 log W 1.0197 0.00754 0.9990 Log S=2.8015—0.3112 log W 0.9920
S=842 w-°4 S=633 W 03112 1.0559 0.01626

Log F=2.6109+0.0521 log W 0.9000 Log F=2.4423+0.0615 log W 0.9247
F=408 w?-021 1.0221 0.00843 F=277 wes 1.0347 0.01021

Log L=2.7553+0.2567 log W 0.9659 Log L=2.1629+0.4600 log W 0.9910
L=569 Ww. 1.0614 0.02301 L=146 w?46° 1.0898 0.02570

Log n=1. 3636— -0. 0307 log W 0.4025 Log n=1.6803—0.0518 log W 0.8432
n=23.1 Ww °° 1.0620 0.02323 n=47.9 W °%18 1.0506 0.01475

Log N=4.4162+0.2290 log W 0.9482 Log N=4.0289+0.4759 log W 0.9950
N=26074 Ww??? 1.0685 0.02557 N=10687 W°47%? 1.0723 0.02088

Log bl=—1.6002+0.2650 log W 0.9970 Log bl=—2.0262+0.4491 0.9955
bl=0.02511 W?:26° 1.0197 0.00718 bl=0.00942 w°:41 1.0646 0.01870

Log GA=2.8319+0.4812 log W 0.9783 Log GA=2.0012+0.9312 log W 0.9965
GA=679 W?-*812 1.1032 0.03618 GA=100 w?32 1.1257 0.03539

Log G=2.8325—0.5194 log W 0.9798 Log G=2.0012—0.0688 log W 0.6557
G=680 Ww °>!4 1.1029 0.03607 G=100 w°°88 1.1257 0.03540

34 Low, W.P., Y. K. Ip AND D. J. W. Lane

TABLE 2. Comparison of the computed values of total filament number (F), total filament
length (L) and gill area (GA) obtained in the present studies for 10 g mudskippers and
various species of fish of equivalent weight reported elsewhere

Fish species F

at 10¢ (x 100 mm) (xX 1060 mm?) Reference
Aquatic fishes

Cyprinus carpio 1110 23.2 Soil 7/ 25

Katsuwonus pelamis — 137 262 26

Thunnus sp. — 135 200 26
Air-breathing fishes

Anabas testudineus 1049 11.2 Dd) 1

Channa punctuata 1111 14.9 1.84 3
Mudskippers

P. chrysospilos 251 3.63 0.89 Present study

B. boddaerti 460 10.28 2.06 Present study

P. schlosseri 547 4.21 0.86 Present study

— value is not available as regression analyses were not performed on this gill parameter

TABLE 3.

Comparison of gill measurements of the local mudskippers P. chrysospilos (P. chry.), B.

boddaerti (B. b.) and P. schlosseri (P. s.) with the B. boddaerti studied by Niva et al. (17) and Hughes

and Al-Kadhomiy (18).

The measurements of 8.8 g Periophthalmus cantonensis (P. can.) and 53 g

Boleophthalmus chinensis (B. c.) by Tamura and Moriyama (19) are also given

Tamura & Moriyama Present study Niva Hughes &
et al. Al-Kadhomiy
Gill 8.8 2 53 g 8.8 8.8 53 53 53 53
measurement P. can. B. c. 12, hey. P. e B. B. 12. = B. b. B. B
Total filament number 306 486 249 317 502 354 630 —
Total filament length 495— 2500- 340 396 1577 904 2179 3678
(mm) 510 2969
Number of secondary 21-26 12-16 18.1 42.8 11.8 39.0 17.7 10.7
lamellae/mm (one side)
Average bilateral 0.040 0.080 0.061 0.025 0.072 0.056 0.061 0.076
secondary lamella area (mm7*)
Total secondary lamellae 23617 77532 12529 30085 64723 70706 77975 78870
Gill area (mm) 1050- 5020- 783.5 759.8 4587 4044 4695 5979
1150 5551
necessarily leads to a smaller role of branchial
DISCUSSION

The amphibious nature of all three genera of
mudskippers is undisputable. However, their
relative success in terrestrial adaptation and their
strategies for terrestrial respiration may be very
different. B. boddaerti enters and submerges itself
in water frequently whereas P. chrysospilos and P.
schlosseri stay on land most of the time. Thus, the
mudskippers have marked differences in their de-
pendence on an aquatic environment. Tamura and
Moriyama [19] pointed out that a small gill area

respiration. Hence, the relationship between the
skin and gill area may explain the behavioral
differences in these mudskippers.

P. chrysospilos has a smaller gill area than B.
boddaerti (Fig. 3) and the fewest and shortest
filaments of the three mudskippers (Fig. 1). As its
filaments are bent and twisted [16], not all of its
secondary lamellae will be oriented parallel to the
respiratory water current, making the counter-
current distribution mechanism for oxygen absorp-
tion inefficient. Therefore, oxygen uptake in this

Gill Morphometry in Mudskippers 35

mudskipper has to be supplemented by other
surfaces such as the skin, be it in water or on land.
In general, the respiratory medium-blood distance
of skin is much thicker than that of gill epithelia.
Nevertheless, cutaneous respiration in P. chrysos-
pilos could be more efficient in air where the
medium is less dense and the diffusion rate of
oxygen would be higher as relatively more oxygen
is available (200 1 oxygen/ml air compared to 5 pl
oxygen/ml water). The importance of cutaneous
respiration in P. chrysospilos is further reflected in
its small gill: skin area ratio (Fig. 5) as compared
to B. boddaerti. Since cutaneous respiration is
important for this mudskipper, it exhibits the
greatest slope of increase in skin area and the
smallest slope of decrease in weight-specific skin
area with respect to weight (Fig. 4, Table 1). The
results of Tamura ef al. (22) are in support of our
hypothesis of the importance of cutaneous respira-
tion in this genera of mudskipper. Their studies
showed that P. cantonensis relies on its skin for
76% and its gills for 27% of its oxygen uptake on
land. When confined to water, the oxygen uptake
by the skin and gills were about equal (48% and
52% respectively). Thus, for respiratory reasons,
P. chrysospilos may remain on land most of the

time. Furthermore, the gills of P. chrysospilos
o 0-75

=

So

_

fo]

o

_

te}

=

~~ 0-50

7)

i]

<>)

—

fo}

G 0-25

| 10 100
Body weight (q)
Fic. 5. Plot of gill area: skin area ratio of P. chrysospi-

los (@), B. boddaerti (4) and P. schlosseri (@)
against logarithmic body weight.

have adaptive features to withstand aerial expo-
sure. Graham [23] suggested that greater spacing
between secondary lamellae reduces collapse of
the respiratory surfaces in air. Secondary lamellar
frequency in P. chrysospilos is indeed lower and
decreases markedly body weight as compared to
the other two mudskippers (Fig. 2, Table 1).

In contrast, B. boddaerti has the longest and
greatest number of filaments (Fig. 1) of the three
mudskippers. Long filaments are more likely to
collapse when removed from water. Although B.
boddaerti of body weights greater than 4 g have
smaller secondary lamellar area than P. chrysospi-
los of similar weight (Fig. 2), the former have
comparatively greater number of lamellae/mm
(Fig. 2). Thus, the secondary lamellae may tend to
coalesce upon removal from water. All these
factors lead to the reduced preference of B. bod-
daerti for land compared to the other two mudskip-
pers. However, its filaments are more aligned [16],
hence, most of the secondary lamellae may be in
the optimal position to fully utilise the counter-
current mechanism for aquatic gaseous exchange.
Also, since it has the largest gill area (Fig. 3),
weight-specific gill area (Fig. 3) and gill area: skin
area ratio of the three mudskippers (Fig. 5), its
greater affinity for water can be easily appreciated.
The results of the present studies are in agreement
with those reported by Tamura and Moriyama [19]
for B. chinensis. Since the gills of B. boddaerti are
playing a greater role in respiration, therefore
measurements reveal that the slope of increase in
skin area with respect to weight in this fish is the
smallest of the three mudskippers (Table 1, Fig.
4).

The bilogarithmic plots of gill area and weight-
specific gill area of B. boddaerti against body
weight are notably different from those of P.
chrysospilos and P. schlosseri. The gill area of a
fish is determined by its total number of secondary
lamellae and average secondary lamellar area. As
the regression coefficients of the increase in these
two gill parameters with body weight in B. bod-
daerti is the smallest of the three mudskippers
(Table 1, Fig. 2), it thus results in its smallest slope
of increase in total gill area with body weight (Fig.
3, Table 1). A consequence of its small slope of
increase in gill area is the rapid reduction in its

36 Low, W.P., Y. K. Ip anp D. J. W. LANE

weight-specific gill area as the fish grows larger
(Fig. 3, Table 1). A greater slope of increase in gill
area may restrict the terrestrial capability of B.
boddaerti, for it necessarily result from a greater
increase in either total number of secondary lamel-
lae, or average bilateral lamellar area, or both of
these gill parameters with body weight. If its
regression coefficient of increase in secondary
lamellar area with body weight is similar to that of
P. chrysospilos, its secondary lamellae will be so
large that they may fold over removed from water
and thus reduce the respiratory area. An increase
in the total number of secondary lamellae without
a concomitant increase in filament length would
lead to a high lamellar frequency. A high lamellar
frequency implies closely spaced secondary lamel-
lae which will tend to coalesce upon aerial expo-
sure. If the filament length were to increase to
accomodate these secondary lamellae, these lon-
ger filaments will have a greater tendency than
shorter ones to collapse when the fish emerges
from water.

The gills of P. schlosseri are not well adapted for
aquatic respiration. Low et al. [16] reported the
presence of intrafilamentary secondary lamellar
fusions in the gills of P. schlosseri. This may
prevent water from flowing between the lamellae
and impede oxygen uptake by the counter-current
mechanism of the gills. Its total filament length
(Fig. 1) and gill area (Fig. 3) are similar to those of
P. chrysospilos. P. schlosseri is also frequently
found out of water in its natural habitat. In
addition, the gills of P. schlosseri can withstand
desiccation as the fenestrae formed by the tissue
fusions between secondary lamellae may trap wa-
ter, reducing the risk of gill dehydration upon
terrestrial exposure [16].

Small P. schlosseri have smaller gill area: skin
area ratio (Fig.5) as compared to bigger ones.
This suggests that small P. schlosseri, like P.
chrysospilos, may have greater dependence on
cutaneous than branchial respiration. P. schlos-
seri, on the other hand, can attain a size of greater
than 100 g. At 100 g, presumably its skin would be
much thicker than those of the other two mudskip-
pers at their respective maximum sizes. Furth-
ermore, the surface area: volume ratio would be
smaller in a large fish compared ‘to a small one.

Cutaneous respiration in bigger P. schlosseri may
thus be relatively inefficient. However, it has
branched filaments (16) which naturally have more
filament tips than unbranched ones. As secondary
lamellae are added towards the tip of gill filament,
more tips would therefore increase the potential of
generating a greater total number of secondary
lamellae and, hence, gill area as the fish grows.
This potential is clearly shown in its large slopes of
increase in total secondary lamellae and gill area
with weight (Figs. 2,3, and Table 1). The gill area:
skin area ratio of P. schlosseri increases with size
from 0.22 until it reaches 0.50 in larger specimens
(Fig. 5). This indicates that its gills play a bigger
role than its skin in respiration as this mudskipper
grows.

Comparison with aquatic fishes, air-breathing fishes
and other mudskippers

Schottle [24] and Graham [23] found that air-
breathing and amphibious fishes have reduced and
length of filaments as well as gill area relative to
aquatic fishes. Indeed, these paramenters in the
mudskippers are smaller than both aquatic and
air-breathing fishes (Table 2). The exception is B.
boddaerti, which has slightly longer filament length
and greater gill area than Channa punctuata (Table
2). These air-breathing fishes spend most of their
time in water and occasionally surface to breathe
whereas mudskippers come on land often. Thus,
the fewer and shorter filaments of mudskippers
may be correlated to their greater terrestrial capa-
bilities [23].

When similar weight Periophthalmus (8.8 g)
were compared, it is notable that P. chrysospilos
has many smaller gill parameters than P. can-
tonensis [19], resulting in a smaller gill area (Table
3). It is possible that the B. boddaerti studied by
Niva et al. [17] and Hughes and Al-Kadhomiy [18]
are of a dissimilar species from the local counter-
part as they have different observed swimming and
feeding behaviors. Most of the gill paramenters of
the local species are considerably divergent from
those of B. boddaerti in previous studies (Table 3),
pointing also towards this possibility. No detailed
data has been collected on the gill morphometry of
Periophthalmodon.

P. schlosseri at 8.8g have comparable total

Gill Morphometry in

filament length and gill area to P. chrysospilos
whilst its total filament number approximates that
of P. cantonensis of similar weight (Table 3).
However, P. schlosseri has fewer and shorter
filaments, and smaller gill area than a similar sized
Boleophthalmus (Table 3). In addition to its uni-
que gill morphology of branched filaments and
intrafilamentary secondary lamellar fusions, P.
schlosseri has smaller lamellae but higher lamellar
frequency than Periophthalmus and Boleoph-
thalmus.

ACKNOWLEDGMENTS

This study was supported by grants RP 70/85 and RP
860337 from the National University of Singapore.

REFERENCES

1 Hughes, G. M., Dube, S. C. and Munshi, J. S. D.
(1973) Surface area of the respiratory organs of the
climbing perch, Anabas testudineus (Pisces: Ana-
bantidae). J. Zool., 170: 227-243.

2 Munshi, J. S. D. (1962) On the accessory respira-
tory organs of Heteropneustes fossilis (Bloch). Proc.
R. Soc. Edinb. (B), 68: 128-146.

3 Hakin, A., Munshi, J. S. D. and Hughes, G. M.
(1978) Morphometrics of the respiratory organs of
the Indian green snake-headed fish, Channa punc-
tata. J. Exp. Zool., 184: 519-543.

4 Carter, G. S. (1957) Air-breathing. In “The Physiol-
ogy of Fishes”, Vol.1. Ed. by Brown, M. E.,
Academic Press, New York/London, pp. 65-79.

5 Hughes, G. M. and Morgan, M. (1973) The struc-
ture of fish gills in relation to their respiratory
function. Biol. Rev., 48: 419-475.

6 Gordon, M. S., Gabaldon, D. J. and Yip, A. Y. W.
(1985) Exploratory observations on microhabitat
selection within the intertidal zone by the Chinese
mudskipper Periophthalmus cantonensis. Mat.
Biol., 85: 209-215.

7 Gordon, M. S., Boetius, J., Evans, D. H., McCar-
thy, R. and Oglesby, L. C. (1969) Aspects of
physiology of the terrestrial life in the amphibious
fishes. I. The mudskipper Periophthalmus sobrinus.
J. Exp. Biol., 50: 141-149.

8 Gordon, M. S., Ng, W. W. S. and Yip, A. Y. W.
(1978) Aspects of the physiology of terrestrial life in
amphibious fishes. III. The Chinese mudskipper
Periophthalmus cantonensis. J. Exp. Biol., 72: 57-
TS:

9 Lee, C. G. L., Low, W. P., and Ip, Y. K. (1987)
Na*, K* and volume regulation in the mudskipper,

10

11

12

13

14

15

16

17

18

19

20

21

22

Mudskippers 37

Periophthalmus chrysospilos. Comp. Biochem. Phy-
siol., 87A: 439-448.

Lee, C. G. L. and Ip, Y. K. (1987) Environmental
effect on plasma thyroxine (T4), 3, 5, 3’-triido-L-
thyronine (T3), prolactin and cyclic adenosine 3’,
5’,-monophosphate (cAMP) content in the mudskip-
pers Periophthalmus chrysospilos and Boleophthal-
mus boddaerti. Comp. Biochem. Physiol., 87A:
1009-1014.

Gregory, R. B. (1977) Synthesis and total excretion
of waste nitrogen by fish of the Periophthalmus
(mudskipper) and Scartelaos Families. Comp.
Biochem. Physiol., 57A: 33-36.

Morii, H., Nishikata, K. and Tamura, O. (1978).
Nitrogen excretion of mudskipper fish Periophthal-
mus cantonensis and Boleophthalmus pectinirostris
in water and on land. Comp. Biochem. Physiol.,
60A: 189-193.

Morii, H. (1979) Changes with time of ammonia
and urea concentrations in the blood and tissue of
mudskipper fish, Periophthalmus cantonensis and
Boleophthalmus pectinirostris transferred from land
to water. Comp. Biochem. Physiol., 63A: 23-28.
Chew, S. F. and Ip, Y. K. (1987) Ammoniagenesis
in mudskippers Boleophthalmus boddaerti and
Periophthalmodon schlosseri. Comp. Biochem. Phy-
siol., 87B: 941-948.

Siau, H. and Ip, Y. K. (1987) Activities of enzymes
associated with phosphoenolphyruvate metabolism
in the mudskippers, Boleophthalmus boddaerti and
Periophthalmodon schlosseri. Comp. Biochem. Phy-
siol., 88B: 119-125.

Low, W. P., Lane, D. J. W. and Ip, K. Y. (1988) A
comparative study of terrestrial adaptations of the
gills in three mudskippers-Periophthalmus chrysos-
pilos, Boleophthalmus boddaerti and Periophthal-
modon schlosseri. Biol. Bull. (In press).

Niva, B., Ojha, J. and Munshi, J. S. D. (1981)
Morphometrics of the respiratory organs of an
estuarine goby, Boleophthalmus boddaerti. Jpn. J.
Ichthyol., 27: 316-326.

Hughes, G. M. and Al-Kadhomiy, N. K. (1986) Gill
morphometry of mudskipper Boleophthalmus bod-
daerti. J. Mar. Biol. Ass. U. K., 66: 671-682.
Tamura, O. and Moriyama, T. (1976) On the
morphological feature of the gill of amphibious and
air-breathing fishes. Bull. Fac. Fish. Nagasaki
Univ., 41: 1-8.

Khoo, K. G. (1966) Studies on the biology of
Periophthalmidae fishes in Singapore. Honours
thesis. National University of Singapore. Singapore.
Hughes, G. M. (1984) Measurement of gill area in
fishes: Practices and problems. J. Mar. Biol. Ass. U.
K., 64: 637-655.

Tamura, S. O., Morii, H. and Yuzuriha, M. (1976)
Respiration of the amphibious fishes Periophthal-

23

24

38 Low, W.P., Y. K. Ip AND D. J. W. LANE

mus cantonensis and Boleophthalmus chinensis in
water and on land. J. Exp. Biol., 65: 97-107.
Graham, J. B. (1976) Respiratory adaptations of
marine air-breathing fishes. In “Respiration of
Amphibious Vertebrates” Ed. by Hughes, G. M..,
Academic Press, New York, pp. 165-187.

Schottle, E. (1932) Morphologie und Physiologie
der Atmung bei wassers-, schlamm- und landleben-

25

26

den Gobiiformes. Z. Wiss. Zool., 140: 1-114.
Oikawa, S., and Itazawa, Y. (1985) Gill and body
surface areas in the carp in relation to body mass,
with special reference to the metabolism-size rela-
tionship. J. Exp. Biol., 117: 1-14.

Muir, B. S., and Hughes, G. M. (1969) Gill
dimensions for three species of tunny. J. Exp. Biol.,
51: 271-285.

ZOOLOGICAL SCIENCE 7: 39-46 (1990)

Difference in Migratory Ability between Human Lung
and Skin Fibroblasts

HirosHI KoNDO, YUMIKO YONEZAWA and TAKASHI A. NOMAGUCHI

Department of Biology, Tokyo Metropolitan Institute
of Gerontology, Tokyo 173, Japan

ABSTRACT—TIG-3 human lung fibroblasts were found to differ in their migratory ability from TIG-3S
human skin fibroblasts derived from the same fetus: (1) TIG-3 cells migrated in medium supplemented
with 10% fetal bovine serum (FBS) more slowly than TIG-3S cells. (2) TIG-3 cells migrated in
serum-free medium as effectively as in medium supplemented with 10% FBS, whereas TIG-3S cells
migrated in serum-free medium much more slowly than in medium supplemented with 10% FBS. (3)
The migration of TIG-3S cells was changed more markedly by the pH of the culture medium than that of
TIG-3 cells. The second was the most striking difference in migratory ability between the TIG-3 and
TIG-3S cells, and was also the case when several human fetal lung fibroblasts (TIG-1, TIG-7, WI-38,
IMR-90, MRC-5), and skin fibroblasts from adult and elderly donors were tested. The monovalent
ionophore, monensin, inhibited the migration of TIG-3 and TIG-3S cells in our experimental system
according to monensin concentration, and immunofluorescence staining for fibronectin demonstrated
that monensin inhibited the secretion of fibronectin. This implies that secreted substances including
fibronectin regulate cell migration. However, the migration of TIG-3S cells was not decreased to the
same degree as that of TIG-3 cells after monensin treatment. The role of the extracellular matrix in this

© 1990 Zoological Society of Japan

difference of migratory ability between human lung and skin fibroblasts is discussed.

INTRODUCTION

Since human fibroblasts have an intrinsic limit of
cell division potential, they are often used as a
model system of cellular aging in vitro [1, 2].
Human fibroblasts derived from different tissues
seem to possess similar features. However, it has
been reported that human fibroblasts show heter-
ogeneity and tissue differences [3-8]. Our pre-
vious study on the effects of serum from human
subjects of various ages on cell migration demon-
strated a difference in migratory ability between
human fetal lung fibroblasts (TIG-1) and skin
fibroblasts from adult donors: Adult donor skin
fibroblasts migrated in serum-free medium much
more slowly than in medium supplemented with
10% FBS, whereas TIG-1 cells migrated in serum-
free medium as effectively as in medium sup-
plemented with 10% FBS [9]. There has been no
previous report of any tissue difference in the

Accepted March 29, 1989
Received January 7, 1989

migratory ability of human fibroblasts. The pre-
sent study was therefore carried out to confirm and
generalize the differences in migratory ability be-
tween human lung and skin fibroblasts.

MATERIALS AND METHODS

Cells

Several human fetal lung fibroblast lines (TIG-1,
TIG-3, TIG-7, WI-38, IMR-90, MRC-5) were
used. TIG-1, TIG-3 and TIG-7 cells were estab-
lished at a project team of the Tokyo Metropolitan
Institute of Gerontology [10, 11]. WI-38 and
IMR-90 cells were obtained from the Institute for
Medical Research (Camden, USA), and MRC-5
cells were obtained from the American Type Cul-
ture Collection (Rockville, USA). Several human
skin fibroblast lines (TIG-3S, ASF-4, ASF-5, ASF-
3, ASF-2) were also used. Human fetal skin
fibroblasts (TIG-3S) were established at a project
team of the Tokyo Metropolitan Institute of
Gerontology, and skin fibroblasts from adult(ASF-

40 H. Konpbo, Y. YONEZAWA AND T. A. NOMAGUCHI

4, ASF-5) and elderly (ASF-2, ASF-3) donors
were kindly supplied by Drs. K. Kaji and M.
Matsuo [12]. The cells were cultured in Eagle’s
basal medium (BME) (GIBCO) supplemented
with 10% FBS and antibiotics, as described in a
previous paper [13].

Mycoplasma contamination in these cell cultures
was measured by the method of Kihara et al. [14],
but none was detected.

Serum

One lot of FBS (Hyclone, #100394) was used
throughout all experiments.
Measurement of cell migration

Cell migration was measured by a slight mod-
ification of the method of Stenn [15]. Before cell
preparation, cover glasses (22x22mm, No. 1,
Matsunami Glass Ind., Ltd, Japan) were cleaned,
and coated by dipping them in 1,2-dichloroethane
(Dojindo Lab., Japan) solution containing 1%
Formvar powder (Polyvinyl Formal, Oken Shoji
Co., Japan) and 0.2% Scarlet red (Sudan IV,
Chroma-Gesellshaft Schmid GMBH & Co.,
DDR). The coated cover glasses were placed in
35-mm plastic Petri dishes (Falcon, 3001) and
secured to the bottom of the dishes with sterile
silicone. Confluent cells, obtained 1 week after
subcultivation with a split ratio of 1:4, were har-
vested from 60-mm plastic culture dishes (Falcon,
3002) by treatment with 0.25% trypsin (Difco,
1:250) in Mg*t-, Ca*t- free phosphate-buffered
saline (PBS), pH 7.4. Confluent cells (split ratio
1:1) or 1X 10° cells per dish were poured into the
35-mm plastic culture dishes containing the pre-
viously prepared cover glasses. The cells covered
the bottom of the dish as a monolayer. The cell
numbers were determined with a Coulter counter
(Coulter Electronics, Hialeah, FL). After culture
for 1 day at 37°C in a 5% CO, incubator, the
cultures were washed once with BME. Each cover
glass was placed on a sterile glass slide and the cell
sheet on the cover glass was marked by cutting off
the cell-Formvar coat along the midline of the
cover glass with a sterile stainless steel blade
(Disposable Dermatome, Feather Ind., Ltd.,
Japan). Then, the cell-coated cover glass was
secured to the bottom of a new 35-mm culture dish
containing culture medium with or without 10%

FBS. On the second day after re-culture, the
outgrowths were stained by removing the culture
medium and flooding the cover glass for 2 min with
a staining solution composed of 0.73% toluidine
blue and 0.27% basic fuchsin in 30% ethanol.
Outgrowths were quantified using a calibrated
ocular micrometer by measuring the maximal
linear distance of cell movement from the cut
edge.

Immunofluorescence staining of fibronectin

In order to detect extracellular fibronectin, im-
munofluorescence staining of human fibroblasts
was performed by the method described previously
[16]. Cell sheets on Formvar-coated cover glasses
were washed three times with PBS, pH7.2, and
incubated with rabbit antibody to human
fibronectin (E-Y Laboratories, Inc., USA) at 1:20
dilution without fixation or drying. Control cul-
tures were incubated with nonimmune rabbit
serum or PBS. After incubation for 30 min at
room temperature, the cell sheets were washed
three times with PBS, and then fluorescein
isothiocyanate (FITC)-conjugated goat anti-rabbit
IgG (E-Y Laboratories, Inc., USA) as a second
antibody at 1:20 dilution was poured over the cell
sheets and incubated at room temperature for 30
min. The cell sheets on Formvar-coated cover
glasses were washed three times with PBS and then
embedded with glycerol solution (glycerol/PBS : 9/
1).

RESULTS

Migratory ability of human fetal lung and skin
fibroblasts derived from the same fetus

The previous study showed that human fetal
lung fibroblasts (TIG-1) migrated linearly during 3
days of incubation [9]. Figure 1 also demonstrates
that human lung (TIG-3) and skin (TIG-3S)
fibroblasts derived from the same fetus migrated
linearly for 3 days in culture medium sup-
plemented with 10% FBS. TIG-3 cells migrated in
serum-free medium as effectively as in medium
supplemented with 10% FBS (Fig. 1A), although
the number of migrating cells seemed to be much
larger in serum-supplemented medium than in

Human Lung and Skin Fibroblast Migration 41

E
=
ne}
w
o
ro’
”
o
O
0 1 2 3
Days in culture
Fic. 1.

3

0) 1 2 3

Days in culture

Time course of migration of human lung (TIG-3) and skin (TIG-3S) fibroblasts derived from the same fetus.

TIG-3 (A) and TIG-3S (B) cells were cultured in BME supplemented with 10% FBS in a humidified CO,
incubator (5% CQO,) at 37°C and cell migration experiments were carried out as described in Materials &
Methods. TIG-3 cells at PD29 and TIG-3S cells at PD25 were treated with 0.25% trypsin and suspended in
culture medium. Two ml (1X 10° cells) of each cell suspension was inoculated into a 35-mm plastic culture dish
containing a Formvar-coated cover glass. After culture for 1 day, each culture was washed once with BME. The
cell sheets on cover glasses were cut off, and the cell-coated cover glasses were re-cultured in BME with or

without 10% FBS and stained at regular time intervals.

Values represent the means of at least triplicate

determinations. Vertical bars show standard deviations of means. O——O, medium supplemented with 10%

FBS; 4------ A, serum-free medium.

serum-free medium (Fig. 2). On the other hand,
the migration rate of TIG-3S cells decreased re-
markably when serum was removed from the
culture medium (Figs. 1B, 2). TIG-3 cells mi-
grated more rapidly than TIG-3 cells.

A study was carried out to determine whether
the difference in the migratory ability between
TIG-3 and TIG-3S cells was observed in culture
media containing various concentrations of FBS
(Fig. 3). TIG-3S cells migrated in media contain-
ing 2.5-50% FBS much more rapidly than TIG-3
cells, although the opposite result was obtained in
serum-free medium. The migratory ability of

TIG-3 and TIG-3S cells gradually decreased
according to increased serum concentration, and
the slope of the decline in migration rate was more
rapid for TIG-3S cells than for TIG-3 cells. Next,
the effects of culture medium pH on the migration
of TIG-3 and TIG-3S cells were measured (Fig. 4).
When culture medium containing 10% FBS was
used, the migration rate of TIG-3 and TIG-3S cells
was lower at pH 6.8 than at pH7.4-8.2. The
migration rates of TIG-3 cells were the same
within a pH range of 6.8-8.2, even when serum
was removed from the culture medium. On the
other hand, the migration rate of TIG-3S cells at

42

D

H. Konno, Y. YONEZAWA AND T. A. NOMAGUCHI

Fic. 2. Photomicrographs of migrating TIG-3 (A, B) and TIG-3S (C, D) cells.

Cell migration experiments were carried out using TIG-3 cells at PD29 and TIG-3S cells at PD25, as described for
Fig. 1. Cells were stained after 2 days of re-culture and photographs of the cells were taken. Arrow designates
cut edge from which cells began to migrate. Bar shows 200 um. A, C: medium supplemented with 10% FBS; B,

Cell spread (mm)

D: serum-free medium.

(0) Pa hee T ieee Sele sae yee rks
ft) 10 20 30 40 #50

Serum concentration (%)

pH 6.8 was much lower in serum-free medium
than in medium supplemented with 10% FBS.
However, TIG-3S cells migrated rapidly at pH 7.8
and 8.2, although the migration of TIG-3S cells
was lower in serum-free medium than in medium

Fic. 3. Effects of serum concentration on migration of

human lung (TIG-3) and skin (TIG-3S) fibroblasts
derived from the same fetus.

The migration of TIG-3 cells at PD29 and TIG-3S
cells at PD25 was determined as described for Fig. 1.
Two ml of each cell suspension (1X 10° cells) was
inoculated. After 1 day of culture, the cell sheets on
cover glasses were washed with BME and cut off,
and cell-coated cover glasses were re-cultured in
BME containing various concentrations of FBS and
stained on the second day. Values represent the
means of at least triplicate determinations. Vertical
bars show standard deviations of means. O——O,
TIG-3 cells; @——®, TIG-3S cells.

supplemented with 10% FBS.

Comparison of migratory ability of several fi-
broblast cell lines cultured from lung and skin

When FBS was removed from the culture

( mm )

Cell spread

6.8 7.4

Human Lung and Skin Fibroblast Migration

7.8 8.2

pH

control

74 78

pH

43

-—_—
control

Fic. 4. Effects of pH of culture medium on migration of human lung (TIG-3) and skin (TIG-3S) fibroblasts derived

from the same fetus.

The migration of TIG-3 cells (A) at PD29 and TIG-3S cells (B) at PD25 was determined as described for Fig. 3.
Two ml of each cell suspension (1 x 10° cells) was incubated in a 5% CO, atmosphere. After 1 day of culture, the
cell sheets on cover glasses were washed and cut off, and cell-coated cover glasses were re-cultured in medium
containing 30 mM Hepes buffer adjusted to each pH (6.8, 7.4, 7.8 and 8.2) in a CO; incubator (0% CO,), and
stained on the first day. For the control experiment, cell sheets on cover glasses were washed and cut off, and
cell-coated cover glasses were re-cultured in medium without Hepes buffer in a CO, incubator (5% CO). Values
represent the means of at least triplicate determinations. Vertical bars show standard deviations of means.

O——O, medium supplemented with 10% FBS; 4

A, serum-free medium.

TABLE 1.. Migration of human lung and skin fibroblast lines in media with or without FBS

Cells PD used/
(Donor age, Sex) Total PD
Lung

TIG-1 (Fetus, F) (26/67)
TIG-3 (Fetus, M) (25/75)
TIG-7 (Fetus, M) (23/63)
WI-38 (Fetus, F) (39-41/52)
IMR-90(Fetus, F) (24/62)
MRC-5(Fetus, M) (4244/55)
Skin

TIG-3S(Fetus, M) (25-29/69)
ASF-5 (21Y, M) (26/66)
ASF-4 (36Y, M) (32-34/74)
ASF-2 (65Y, F) (25/56)
ASF-3 (77Y, M) (24-26/43)

Cell spread (um)

10% FBS/

Ne») No serum 10% FBS IN@ sexta
(7) 1126+99 1223 +84 (+ 9%)
(1) 1026+ 198 1212+84 (4+ 18%)
(1) 912+90 939 +85 (+ 3%)
(2) 902 +7 1159+1 (4+ 28%)
(1) 758 + 85 1052+81 (4+ 39%)
(2) 928 +53 1047 +61 (+ 13%)
(2) 677 + 162 1723 +60 (+155%)
(1) 519+ 11 1420+ 64 (+174%)
(2) 450+7 1330+33 (+196%)
(1) 587474 1642 +140 (+180%)
(2) 407+37 1050+75 (+158%)

Migration experiments were carried out as described for Fig. 1.
each 35-mm culture dish were 110° cells for TIG-1 cells and confluent cells (split ratio 1 : 1) for other
cell lines. After 1 day of culture, the cell sheets on cover glasses were washed and cut off. The
cell-coated cover glasses were re-cultured in BME with or without 10% FBS, and stained on the second

day. Values represent the means (+S.D.) of at least triplicate determinations.

The numbers of cells inoculated onto

44 H. Konpbo, Y. YONEZAWA AND T. A. NOMAGUCHI

medium, the migration rate of TIG-3S cells de-
creased whereas that of TIG-3 cells did not change
(Figs. 1, 2). A study was carried out to clarify
whether the same result was obtained when many
other fibroblast lines derived from lung and skin
were tested. Table 1 shows that the migratory
abilities of five human fetal lung fibroblast lines
were the same as that of TIG-3 cells, whereas the
migratory abilities of four human skin fibroblast
lines derived from adult and elderly donors were
the same as that of TIG-3S cells. These results
imply that human lung fibroblasts differ from
human skin fibroblasts in migratory ability.

Effects of monensin on migration of human fetal
lung and skin fibroblasts

In order to clarify whether substances in the
extracellular matrix contribute to the difference in
migratory ability between human lung and skin
fibroblasts, the effects of monensin on cell migra-
tion were measured (Fig. 5). Migration of both

z 2
E
= t
no] [Osa /
i}
o
a 1
0
@
(S)
0 m) \7 +6
0 10 10 10
Monensin concentration ( M )
Fic. 5. Effects of monensin on migration of human fetal

lung (TIG-3) and skin (TIG-3S) fibroblasts.

The migration of TIG-3 cells (A) at PD35 and
TIG-3S cells (B) at PD25 was determined as de-
scribed for Fig. 3. Two ml of cell suspension (1x 10°
cells) was inoculated. After 1 day of culture, the cell
sheets on cover glasses were washed and cut off, and
cell-coated cover glasses were re-cultured in BME
containing 10% FBS and various concentrations of
monensin, and stained on the second day. Values
represent the means of duplicate determinations.
Vertical bars show standard deviations of means.
O——O, TIG-3 cells; @——@, TIG-3S cells.

cell types was inhibited according to increased
monensin concentration (510~°-1x10~°M).
However, the migration rate of TIG-3S cells did
not become the same as that of TIG-3 cells even
when monensin at 1x10~°M was used. Since it
has been reported that monensin inhibits the secre-
tion of substances constituting the extracellular
matrix, such as procollagen and fibronectin [17,
18], an attempt was made to detect the secretion of
fibronectin after monensin treatment, using im-

Fic. 6.

Immunofluorescence staining of the monolayers
of TIG-3 human fetal lung fibroblasts with anti-
bodies to human fibronectin.

Cell migration experiments were carried out using
TIG-3 cells at PD37, as described for Fig. 5. Two ml
of cell suspension (1X10° cells) was inoculated.
After 1 day of culture, the cell sheets on cover
glasses were washed and cut off, and the cell-coated
cover glasses were re-cultured in culture medium
with (B) or without (A) 5X10~’M monensin. On
the second day, the cell sheets on cover glasses were
incubated with antibodies to human fibronectin, and
extracellular fibronectin was stained, as described in
Materials and Methods. Monolayer cells were mic-
roscopically photographed. Control cultures which
were incubated with nonimmune rabbit serum or
PBS were not stained (data not shown). X50.

Human Lung and Skin Fibroblast Migration 45

munofluorescence staining of fibronectin. The
result showed that monensin strongly inhibited the
secretion of fibronectin from TIG-3 cells (Fig. 6).
The same result was also obtained when TIG-3S
cells were tested (data not shown).

DISCUSSION

When measuring the migration of human lung
and skin fibroblasts, we observed a difference in
migratory ability between the two fibroblast lines
employed (Table 1). It has been reported that
human lung fibroblasts differ from human skin
fibroblasts in cell morphology, growth rate and cell
density at confluence [6], the capacity to change
cortisone into hydrocortisone [7], and the Kd value
for the binding reaction of dexamethasone to cells
[8]. However, a difference in migratory ability
between these two fibroblast lines has not been
reported. Therefore the findings of our present
study seem to be the first evidence of a tissue
difference in the migration of human fibroblasts.

The migration rate of human skin fibroblasts
(TIG-3S) changed greatly upon removal of FBS
from the culture medium, range of pH of serum-
free medium and change in the serum concentra-
tion of culture medium, unlike the case of human
lung fibroblasts (TIG-3) derived from the same
fetus (Figs. 1, 3, 4). This result was also obtained
when skin fibroblasts from adult and elderly
donors were used. This implies that human skin
fibroblasts seem to be sensitive to stimuli or en-
vironmental change. For skin fibroblasts to per-
form their role in wound healing, a high sensitivity
to many forms of stimulus may be an essential
feature. However, the mechanism by which hu-
man skin fibroblasts migrate more rapidly than
human lung fibroblasts is unclear.

Using relatively early passaged cells, we demon-
strated that human lung fibroblasts (TIG-3) differ
from human skin fibroblasts (TIG-3S) with regard
to migration rate and pH and serum dependency.
A study was therefore carried out to clarify
whether this phenomenon changed during the in
vitro aging of human fibroblasts. The same result
was obtained, although the migratory ability of
cells decreased with successive passages (unpub-
lished data). These results show that TIG-3 cells

differ from TIG-3S cells in migratory rate at all
stages of passage in addition to showing serum
dependency (and perhaps pH dependency).

The number of migrating TIG-3 cells seemed to
be much greater in serum-supplemented medium
than serum-free medium (Fig. 2A, B). This may
have been due partly to cell proliferation because
dividing cells were often observed in serum-
supplemented medium. However, the migration
rate of TIG-3 cells did not change when serum was
removed from the culture medium (Fig. 1). These
results are consistent with our previous study, i.e.,
a loss of cell division potential, which was induced
by exposure to “°Co-y-rays, did not change the
migration rate of TIG-1 cells [9].

It has been reported that monensin inhibits the
secretion of procollagen and fibronectin from cul-
tured human fibroblasts but does not inhibit pro-
tein synthesis [17, 18]. Also, since monensin
inhibits the spreading of human fibroblasts [19-
21], we determined the effects of monensin on cell
migration using our present experimental system.
Monensin inhibited the migration of TIG-3 and
TIG-3S cells according to its conecentration (Fig.
5). Immunofluorescence staining for extracellular
fibronectin revealed that monensin strongly inhi-
bited the secretion of fibronectin from TIG-3 (Fig.
6) and TIG-3S cells (data not shown). This implies
that monensin inhibits the secretion of secretory
proteins including fibronectin. The finding that a
relatively low concentration of monensin produced
an effect is compatible with the action of monensin
on cell attachment and spreading [17, 18]. The
monensin concentration effective for inhibition of
cell migration was the same in both TIG-3 and
TIG-3S cells. However, the migration of TIG-3S
did not decrease to the same extent as that of
TIG-3 cells even when 1X10~°M monensin was
used. In other words, the difference in migratory
ability between TIG-3 and TIG-3S cells could not
be explained by the quantitative and qualitative
differences in extracellular matrix secreted from
the two cell types. Rather, it may reflect differ-
ences in the cell-specific events necessary for cell
migration.

46

ACKNOWLEDGMENTS

We wish to thank Dr. M. Osanai for helpful discussion
during the course of this work. We also thank Drs. H.
Okumura and K. Kihara, National Institute of Health,
Tokyo, for the mycoplasma testing of cultures.

10

REFERENCES

Hayflik, L. and Moorhead, P. S. (1961) The serial
cultivation of human diploid cell strains. Exp. Cell
Res., 25: 585-621.

Hayflik, L. (1965) The limited in vitro lifetime of
human diploid cell strains. Exp. Cell Res., 37: 614—
636.

Pinsky, L., Finkelberg, R., Straisfeld, C., Zilahi,
B., Kaufman, M. and Hall, G. (1972) Testosterone
metabolism by serially subcultured fibroblasts from
genital and nongenital skin of individual human
donors. Biochem. Biophys. Res. Commun., 46:
364-369.

Griffin, J. E., Punyashthiti, K. and Wilson, J. D.
(1976) Dihydrotestosterone binding by cultured
human fibroblasts. J. Clin. Invest., 57: 1342-1351.
Kaufman, M., Pinsky, L., Straisfeld, C., Shanfield,
B. and Zilahi, B. (1975) Qualitative differemces in
testosterone metabolism as an indication of cellular
heterogeneity in fibroblast monolayers derived from
human preputial skin. Exp. Cell Res., 96: 31-36.
Schneider, E. L., Mitsui, Y., Au, K. S. and Shorr,
S. S. (1977) Tissue specific differences in cultured
human diploid fibroblasts. Exp. Cell Res., 108: 1-6.
Aronson, J. F., McClaskey, J. W. and Cristofalo, V.
J. (1983) Human fetal lung fibroblasts: Observation
on origin and stability in culture. Mech. Ageing
Dev., 21: 229-244.

Kondo, H., Kasuga, H. and Noumura, T. (1985)
The heterogeneity of human fibroblasts as deter-
mined from the effects of hydrocortisone on cell
growth and specific dexamethasone binding. Exp.
Cell Res., 158: 342-348.

Kondo, H., Nomaguchi, T. A. and Yonezawa, Y.
(1989) Effects of serum from human subjects of
different ages on migration in vitro of human fibro-
blasts. Mech. Ageing Dev., 47: 25-37.

Ohashi, M., Aizawa, S., Ooka, H., Ohsawa, T.,
Kaji, K., Kondo, H., Kobayashi, T., Noumura, Te,

11

12

13

14

15

16

17

18

19

20

21

H. Konno, Y. YONEZAWA AND T. A. NOMAGUCHI

Matsuo, M., Mitsui, Y., Murota, S., Yamamoto,
K., Ito, H., Shimada, H. and Utakoji, T. (1980) A
new human diploid cell strain, TIG-1, for the re-
search on cellular aging. Exp. Gerontol., 15: 539-
549.

Matsuo, M., Kaji, K., Utakoji, T. and Hosoda, K.
(1982) Ploidy of human embryonic fibroblasts dur-
ing in vitro aging. J. Gerontol., 37: 33-37.

Kondo, H., Nomaguchi, T. A., Sakurai, Y., Yone-
zawa, Y., Kaji, K., Matsuo, M. and Okabe, H.
(1988) Effects of serum from human subjects of
various ages on proliferation of human lung and skin
fibroblasts. Exp. Cell Res., 178: 287-295.

Kondo, H., Kasuga, H. and Noumura, T. (1983)
Effects of various steroids on in vitro lifespan and
cell growth of human fetal lung fibroblasts (WI-38).
Mech. Ageing Dev., 21: 335-344.

Kihara, K., Ishida, S. and Okumura, H. (1981)
Detection of mycoplasmal contaminants in sera. J.
Biol. Stand., 9: 243-251.

Stenn, K. S. (1980) Quantitative assay of dissoci-
ated tissue cell motility in vitro. In Vitro, 16: 357-
360.

Kuhl, U., Timple, R. and von der Mark, K. (1982)
Synthesis of type IV collagen and fibronectin from
cultured human fibroblasts. Dev. Biol., 93: 344-354.
Uchida, N., Smilowitz, H. and Tanzer, M. L. (1979)
Monovalent ionophores inhibit secretion of procol-
lagen from cultured human fibroblasts. Proc. Natl.
Acad. Sci. USA, 76: 1868-1872.

Uchida, N., Smilowitz, H., Ledger, P. W. and
Tanzer, M. L. (1980) Kinetic studies of the in-
tracellular transport of procollagen and fibronectin
in human fibroblasts. J. Biol. Chem., 255: 8638-
8644.

Virtanen, I., Vartio, T., Badley, R. A. and Lehto,
V.-P. (1982) Fibronectin in adhesion, spreading and
cytoskeletal organization of cultured fibroblasts. Na-
ture, 298: 660-663.

Pizzey, J. A., Bennet, F. A, and Jones, G. E. (1983)
Monensin inhibits initial spreading of cultured hu-
man fibroblasts. Nature, 305: 315-317.

Pizzey, J., Witkowski, J. and Jones, G. (1984)
Monensin-induced inhibition of cell spreading in
normal and dystrophic human fibroblasts. Pro. Natl.
Acad. Sci., USA, 81: 4960-4964.

ZOOLOGICAL SCIENCE 7: 47-54 (1990)

© 1990 Zoological Society of Japan

A Monoclonal Antibody against a Synthetic Carboxyl-Terminal
Fragment of the Eclosion Hormone of the Silkworm,
Bombyx mori: Characterization and Application
to Immunohistochemistry and
Affinity Chromatography

TAKAHARU Kono, AxkiRA Mizocucnt', Hiromichi Nagasawa, Hironori

Ishizaki', HasimeE Fuco* and Akinori Suzuki

Department of Agricultural Chemistry, Faculty of Agriculture, The University of Tokyo,
Tokyo 113, ‘Biological Institute, Faculty of Science, Nagoya University,
Nagoya 464, *Faculty of Agriculture, Tokyo University of Agriculture
and Technology, Tokyo 183, Japan

ABSTRACT— Monoclonal antibodies were produced using as antigen a synthetic fragment correspond-

ing to the C-terminal portion of the eclosion hormone (EH) of the silkworm, Bombyx mori.

The

characterization of these antibodies using ELISA revealed that one of them recognized specifically both

the synthetic fragment and native eclosion hormone.

Immunohistochemistry using this antibody

indicated that EH was produced in two pairs of median neurosecretory cells of the brain. Affinity
chromatography of a partially purified EH using a column on which this antibody had been immobilized
showed that EH activity was completely adsorbed to this column and eluted with the synthetic fragment

with about 60% recovery.

INTRODUCTION

The adult eclosion in the silkworm, Bombyx
mori, occurs during a specific period of the day [1]
as in other lepidopteran insects. The timing of
pupal-adult eclosion is controlled by the
neurosecretory hormone designated eclosion hor-
mone (EH) [2] which triggers a series of eclosion
behaviors. EH acts not only at the adult ecdysis
but also at the larval and pupal ecdyses [3]. In B.
mori, EH might also be involved in the egg hatch-
ing behavior [4]. Recently, using the tobacco
hornworm, Manduca sexta, Copenhaver and Tru-
man have succeeded in identifying a cluster of
ipsilaterally projecting cells (group Ia) that contain
EH by a sensitive behavioral bioassay and, more
specifically, five cells in this group in each brain
hemisphere by immunological techniques using an
anti-EH antiserum [5].

Accepted March 15, 1989
Received January 27, 1989

We have recently determined the amino acid
sequence (61 residues) of the Bombyx EH [6]. At
almost the same time, Marti et al. [7] and Kataoka
et al. [8] determined the whole amino acid sequ-
ence (62 residues) of the Manduca EH indepen-
dently, showing 80% sequence homology with the
Bombyx EH.

The clarification of the amino acid sequence
permitted us to attempt to make monoclonal anti-
bodies against a synthetic peptide fragment. In
this paper we describe the characterization of a
highly specific monoclonal antibody raised against
a synthetic peptide corresponding to the C-
terminal portion of EH and its application to
immunohistochemical study and affinity chroma-
tography.

MATERIALS AND METHODS

Peptide synthesis

A C-terminal synthetic peptide corresponding to

48 T. Kono, A. Mizocucut et al

EH (49-61), H-Cys-Glu-Ser-Phe-Ala-Ser-Ile-Ser-
Pro-Phe-Leu-Asn-Lys-OH, was synthesized as fol-
lows. The protecting groups for the functional side
chains of the amino acids were cyclohexyl ester for
glutamic acid, benzyl ether for serine, and 2-
chlorobenzyloxycarbonyl (2CIZ) for lysine. Start-
ing with Boc-Lys(2CIZ)-OCH-Pam resin, the
stepwise solid-phase synthesis was performed on
an Applied Biosystems model 430A _ peptide
synthesizer using dicyclohexyl carbodiimide/1-
hydroxybenzotriazole as a coupling reagent. After
removal of the protecting groups and resin in HF,
the residue was washed several times with dieth-
ylether and chloroform by turns. Then the de-
blocked peptides were extracted with 2 M acetic
acid, lyophilized, and finally purified by prepara-
tive reverse phase HPLC. Amino acid sequence of
the synthetic peptide was checked by an Applied
Biosystems model 470 A protein sequencer.

Preparation of antigen

Four mg (0.06 zmol) of bovine serum albumin
(BSA) was mixed with 0.3mg (1 wmol) of N-
hydroxysuccinimidyl 3-(2-pyridyldithio) propion-
ate (SPDP, Pharmacia Fine Chemicals) in 0.1M
phosphate-buffered saline (pH 7.5) at 23°C for 30
min, allowing the amino groups of BSA to react
with N-hydroxysuccinimide ester moiety of SPDP.
After removal of reagents by gel filtration on
Sephadex G-25, synthetic EH (49-61) (1.3 mg, 1
ymol) was added to the resulting BSA-SPDP
conjugate fraction at 25°C for 20min. The free
sulfhydryl group of cysteine of the synthetic pep-
tide was treated with the 2-pyridyldisulfide moiety
of SPDP [9]. After removal of the reagents with
Sephadex G-25, about 5.3 mg of EH(49-61)-BSA
conjugate were obtained. N-Terminal sequence
analysis after purification by HPLC indicated that
more than 5 mol of EH(49-61) were coupled with
1 mol of BSA.

Production of monoclonal antibodies

Three female BALB/c mice were immunized
four times at a 2-week interval by the intraperi-
toneal injection of EH(49-61)-BSA (20 ~g/mouse)
in Freund’s complete adjuvant. Three days after
each injection, the mice were bled from the tail
vein and the blood was centrifuged to remove

cells. The antibody detection in each mouse serum
was carried out by the dot-immunobinding assay,
essentially according to Hawkes ef al. [10]. After
four times of antigen injection, the antibody activ-
ity of one mouse detected by the dot immunobind-
ing assay using the synthetic EH(49-61) fragment
was positive at 5,000 fold dilution of the antiserum,
and this mouse was used to produce monoclonal
antibodies.

The splenocytes collected (8X10’ cells) were
fused with mouse myeloma NS-1 cells using
polyethyleneglycol. Hybridoma cells obtained
were seeded into 96-well microplates (Falcon,
3072) and cultured in the presence of 1X10*
peritoneal macrophages per well as a feeder layer.
The supernatant of each well was primarily
screened by dot-immunobinding assay using
EH(49-61)-BSA conjugate or EH(49-61) frag-
ment as an antigen. Positive colonies on this assay
were again cultured in 24-well plates, and further
screened in the same manner, and positive hybri-
domas were cloned by limit dilution.

The cloned hybridomas secreting anti-EH(49-
61) antibody were intraperitoneally injected into a
mouse previously injected with 0.5 ml of pristane,
and after two weeks, about 5 ml of an ascites fluid
was collected. The monoclonal antibody was par-
tially purified from this ascites by ammonium
sulfate precipitation at 20-33% saturation,
Sephadex G-25 gel filtration, and DEAE-
Sepharose CL-6B ion exchange chromatography.

Competitive enzyme-linked immunosorbent assay
(competitive ELISA)

Wells of a 96-well ELISA plate (Sumitomo
bakelite, MS-3596F) were coated with 50 al of 5x
10-'°M_ EH(49-61)-BSA conjugate in 0.1M
sodium carbonate buffer (pH 9.6) for 2 hr at 25°C.
After washing with 50mM _ Tris-buffered saline
(pH 7.4) (TBS) and blocking with 3% gelatin (Bio
Rad), 50 ul of the monoclonal antibody solution
corresponding to 1:200,000 dilution of the ascites
and 50 ul of serially diluted test materials were
added into each well, and the plate was incubated
overnight at 4°C.

The plate was washed with TBS containing
0.05% Tween-20 (TTBS), and reincubated with 50
ul of 1:1,000 diluted horseradish peroxidase

Antibody against Bombyx Eclosion Hormone 49

(HRP)-linked anti-mouse immunoglobulin goat
serum (AMS, Kirkegaard & Perry Laboratories
Inc.) at 25°C for 2 hr. Wells were developed with
o-phenylenediamine as a substrate for HRP. The
enzyme reaction was stopped by addition of 50 yl
of 2M H»SQOx,, and the absorbance at 492 nm was
measured.

Affinity chromatography

Partially purified monoclonal antibody obtained
from ascites was linked to CNBr-activated Sephar-
ose 4B (Pharmacia Fine Chemicals) by the stan-
dard method. About 5 mg of the partially purified
antibody was immobilized to the gel (1g dry
weight) and 3.5 ml of affinity gel was obtained.
EH was partially purified from 80,000 heads of B.
mori through the 11 step purification procedure
according to the methods as described previously
[11]. This partially purified EH in 100 ml of 0.1 M
ammonium acetate (pH 8.5) was applied to the
affinity column. The column was sufficiently
washed with 0.2 M ammonium acetate (pH 8.5)
until the absorbance at 280nm became below
0.005. Adsorbed materials were eluted successive-
ly with 0.2 M ammonium acetate containing 100
#M of S-carboxamidomethyl EH(49-61) frag-
ment, 0.25 M sodium carbonate containing 0.5 M
NaCl (pH 8.5), and 0.2 M Gly-HCl (pH 2.5).

The active fraction after the affinity chroma-
tography was acidified to pH 2, and directly sub-
jected to a reverse phase HPLC using VP-318
(Senshu Kagaku). Chromatography was per-
formed by applying a linear gradient of acetonitrile
(0.5%/min) in the presence of 0.1% trifluoroacetic
acid.

Immunohistochemistry

The Bombyx brains from freshly ecdysed pupae,
3-, 5-, and 7- day developing adults were dissected
out and fixed in Bouin’s solution for 4hr. The
specimens were dehydrated with graded ethanol
and embedded in paraffin. Serial sections (7 ~m
thick) were cut on a rotary microtome and affixed
on slide glasses. The sections were deparaffinized
by xylene, washed with absolute ethanol, soaked
in methanol containing 0.03% HO, for 30 min to
inhibit the endogenous peroxidase activity in the
tissue, and washed with TBS. The anti-EH(49-61)

IgG of 1:500 diluted solution was applied to the
rehydrated tissue section after blocking with 3%
gelatin, and incubated overnight at 4°C. Subse-
quently, the extra antibody solution was removed
by washing with TTBS. The tissue sections were
incubated with the 500-fold diluted solution of
HRP-linked AMS for 2 hr, and rinsed again with
50 mM Tris-buffered saline (pH 7.4) (TBS). Then
the sections were mounted with 2.2mM of 4-
chloro-1-naphthol in TBS containing 0.03% H,Op.

Whole mount staining was performed using
Bombyx pharate adult brains. The procedure
followed essentially the protocol of Bollenbacher
et al. (personal communication). The brain was
fixed with Bouin’s solution for 4 hr. After washing
with TTBS, the fixed brain was desheathed and
exposed to the 500-fold diluted antibody solution
containing 2% Triton X-100 overnight at 4°C.
After washing with TTBS, the tissue was incubated
with 500-fold diluted solution of HRP-linked AMS
for 2 hr at room temperature. The tissue was
washed with TBS, and incubated with 1.3 mM of
diaminobenzidine in TBS containing 0.02% H,O,
for 10 min. Then the reaction was terminated by
removing the enzyme substrate solution and
washed successively with TBS, water, 70%
ethanol,95% ethanol, and 100% ethanol. Finally,
the tissue was cleared by methylsalicylate over-
night and mounted on a slide.

RESULTS

Generation of the monoclonal antibody recognizing
EH

During the first screening of the hybridoma
colonies using EH(49-61)-BSA conjugate as an
antigen, 10 colonies gave positive immunoreac-
tion. On the second screening using EH(49-61)
synthetic fragment, however, only one hybridoma
colony among them appeared to produce antibody
that recognized this fragment. This hybridoma was
cloned and the antibody was produced by the
intraperitoneal injection of the hybridoma cells
into a mouse. The monoclonal antibody obtained
from the ascites fluid was purified partially and
characterized.

The class of the immunoglobulin produced by

50 T. Kono, A. Mizocucut et al

this clone was identified as IgG by its reactivity to
the class-specific goat anti-mouse immunoglobulin
sera. The characterization of the immunobinding
specificity of this antibody was further accom-
plished by the competitive ELISA using BSA-
EH(49-61) conjugate as a coated antigen on a

solid phase. The binding of the antibody to the
solid phase was inhibited in a dose-dependent
manner when BSA-EH(49-61), EH(49-61) frag-
ment, and “highly purified EH” [11] were used as a
competitive antigen added to the liquid phase (Fig.
1), but this antibody did not recognize BSA.

1005

% Maximal Binding

BSA-EH(49-61 )

19710 1079

Fic. 1.

EH(49-61)

“Highly purified EH”

10-8

1076
Antigen (M)

1077

Binding activity of an EH(49-61) monoclonal antibody as assessed by competitive ELISA to EH(49-

61)-BSA. ELISA plate was precoated with 50 pl of 5 x 10~ 1° M of EH(49-61)-BSA as a competitive antigen and
competitive ELISA was performed on this plate with the presence of the monoclonal antibody corresponding to 1

: 200,000 dilution of ascites in liquid phase.

0,54

=

Absorbance at 280 nin

|

(1)
Fic. 2.

Affinity chromatography using an EH(49-61) monoclonal antibody-Sepharose 4B column.

Ifo ora

(2) (3) (4)

As a sample

solution, 80,000 head equivalents of the partially purified EH preparation was used. After application of the
sample, the column was washed with (1) 0.2M CH3;COONH, (pH 8.5), (2) 100 “M S-carboxamidomethyl
.EH(49-61)/0.2 M CH;COONH, (pH 8.5), (3) 0.5 M NaCl/0.25 M NazCO; (pH 8.5) and (4) 0.2 M Gly-HCl (pH
2.5) in order. Flow rate was 3 ml/hr and EH activity detected is shown as a solid bar.

Antibody against Bombyx Eclosion Hormone 51

Therefore, this antibody not only bound to
EH(49-61) portion at a concentration of 107’ M,
but also cross-reacted with partially purified native
EH. The content of EH in this “highly purified
EH” sample was estimated to be about 0.1%.

Affinity chromatography

Partially purified EH was obtained by the usual
11-step purification procedure [11], and this mate-
rial was applied to the affinity column. Biological
activity was completely adsorbed to this column
and eluted with 100“M of synthetic S-
carboxamidomethyl EH(49-61) fragment in 0.2 M
ammonium acetate buffer (pH 8.5) with about
60% recovery. The column was eluted successive-
ly with 0.5 M NaCl containing 0.25 M NazCO3 (pH
8.5) and 0.2 M Gly-HCl (pH 2.5), but EH activity
was not detected in either of these two fractions
(Fig. 2).

The active fraction was acidified and directly
subjected to reverse phase HPLC. Activity was
detected in a single peak with a shoulder shown in
the painted peak in Figure 3, and Table 1 shows
the summary of the purification efficiency. Sequ-
ence analysis of the peptide in the main part of the
peak for N-terminal portion (up to the 5th residue)
assessed the substance in this fraction to be EH.

Immunohistochemical localization of EH in the
Bombyx brain

Immunohistochemical staining was performed
on various stages of Bombyx pupal brains. In the
usual immunohistochemical procedures using se-
rial sections with 7 ~m thick, two pairs of median
neurosecretory cells were found to be im-

=
E e
S E
: 5
oO
é c
ig 5
x ts
oO
=
5
Cc
vo
1S)
=
8
10 20 30 40
Retention time (min)
Fic. 3. Reverse phase HPLC of the affinity-purified

EH. The result from 10,000 head equivalents of
active fraction is shown. The dotted line shows the
concentration of acetonitrile in 0.1% trifluoroacetic
acid. The shaded peak at 35.84 min indicate fraction
containing EH activity and the peak at 22 min is
S-carboxamidomethyl EH(49-61).

munoreactive to the antibody in all stages (Fig.
4A). These cells were about 10 um in diameter,
and their cytoplasm was densely filled with im-
munoreactive material.

To assess the topographical distribution of these
cells further, whole mount staining was carried out
on the brains from the pharate adult 2 days before
ecdysis. As shown in Figure 4B, two pairs of
median neurosecretory cells were stained, and
some immunoreactive nerve fibers were derived
from these cells. From various angles of observa-

TaBLeE 1. Summary of the purification of EH from 80,000 head equivalents of EH sample
. : Weight Total activity Specific activity
PUBCON SHEP (ug (EH units) (ag/EH unit)
8th Ppt. with 80% acetone 4,400,000 110,000 40,000
(“Crude EH”)
9th Sephadex G-50 (fine) 1,960,000 100,000 19,600
1ith SP-Sephadex C-25 332,000 90,000 3,700
12th EH(49-61) antibody-Sepharose 4B* 460 55,000 8.4
13th VP-318 (TFA) 4 20,000 0.2

* Activity was eluted with 100M S-carboxamidomethyl EH(49-61) synthetic fragment/0.2M

CH;COONH, (pH 8.5).

52 T. Kono, A. Mizocucui et al

tion, these four cells were supposed to localize in
the anterior portion of the brain as illustrated in
Figure 4C and 4D.

DISCUSSION

In our previous study to isolate EH, we obtained

Dorsal

Ventral

Anterior

Posterior

only about 30 yg of EH from the extract of 770,000
Bombyx pharate adult heads. Because of the
limited availability of pure EH, it seemed to be
quite difficult to use natural EH for raising anti-
EH antibody. Therefore, we took the strategy to
make a fragment peptide of EH and to immunize
mice with this synthetic peptide after conjugation
with BSA. EH has six cysteine residues and these
residues were supposed to make three intra-
molecular disulfide bonds in the molecule to form
a globular tertiary structure. Therefore, we de-
cided to make synthetic fragments corresponding
to the N-terminal 14 residues and the C-terminal
13 residues, respectively, because these parts
would be localized at the surface of the molecule
and, therefore, have a rather flexible structure.
The study to make monoclonal antibodies against
the N-terminal fragment is now in progress.

The attempt to raise the monoclonal antibody
against the C-terminal synthetic fragment suc-
ceeded in getting one hybridoma clone which
secreted an antibody capable of recognizing the
native EH as well as EH(49-61) fragment.
According to the results from the competitive
ELISA, the antibody recognized a partially
purified native EH designated “highly purified
EH” in a dose-dependent manner. The results
showed that our antibody detected the EH mole-
cule in “highly purified EH” at concentrations
higher than 10-° M. This assay needs only 50 wl of
the sample solution, and so, the amount of EH in
10~° M of the sample solution is about 3 ng. That
is to say, this assay system can detect a several
hundred fent mole level of EH molecule consider-
ing the molecular weight of EH to be about 7,000.
This sensitivity is quite satisfactory for immunolo-

Fic. 4. Immunohistochemical localization of EH in
Bombyx brain. (A) Transverse section of day-0
pupal brain. In the median part, two pairs of
immunoreactive cells are seen. (B, C, D) Whole
mount staining of pharate adult brain. (B) Two
pairs of neurosecretory cells are stained, and im-
munoreactive nerve fibers are seen to start from
these cells. Anterior view (C) and dorsal view (D)
of the pharate adult brain-suboesophageal ganglion
complex show that the immunoreactive four cells are
supposed to localize in the anterior portion of the
brain. AL: antenal lobe. OL: optic lobe. SG:
suboesophageal ganglion.

Antibody against Bombyx Eclosion Hormone 53

gical assay, but is still less sensitive than biological
assay using Bombyx pharate pupa, which can
detect 0.1 ng of EH.

In our previous study, 18 steps of purification
procedure were necessary to isolate EH from the
extract of Bombyx heads. However, by the use of
immunoaffinity chromatography with this anti-
body, the purification procedure could be sim-
plified considerably by omitting several steps of
open column chromatographies and HPLCs.

The EH activity was adsorbed to the affinity
column completely. Elution of the active material
with the eluant of pH 2.5 did not give a good yield,
and the solution containing the synthetic EH(49-
61) fragment was concluded to be the best in
recovery. Consecutive elution with the solutions
of 0.5 M NaCl or of pH 2.5 did not give any active
material, indicating that most of the activity was
eluted by substitution with the synthetic fragment.
In purification by the affinity chromatography us-
ing the sample of 80,000 head equivalents, the
recovery estimated from total activity was about
60%, and about 440 fold purification could be
attained. The specific activity after this affinity
chromatography was about 8.4 ng/unit. Consider-
ing that specific activity of pure EH is 0.1-0.2
ng/unit, an about 80-fold purification was neces-
sary to isolate EH after this step. By the use of an
ODS column, EH was separated from a large
amount of S-carboxamidomethyl EH(49-61) and
the activity was detected in a single peak. Because
the specific activity of this peak was estimated to
be 0.2 ng/EH unit and, in addition, the amino acid
sequence analysis showed that a peptide in this
peak coincided with EH at least for the N-terminal
portion, we think EH was isolated by this HPLC
step. This immunoaffinity chromatography fol-
lowed by one-step reverse phase HPLC can con-
siderably simplify the isolation procedure for EH.

Immunohistochemistry using the antibody re-
vealed that two pairs of brain median neurosecre-
tory cells had immunoreactive material in the
perikarya (Fig. 4A). Therefore, it is highly possi-
ble that these four cells produce EH. Whole
mount staining showed that the immunoreactive
nerve fibers were originated from these cells as
shown in Fig. 4B, and these fibers did not seem to
cross at the middle part of the brain. Fugo et al.

examined the distribution of EH activity in the
brain-suboesophageal ganglion (SG) complex of
pharate adult Bombyx by the surgical cutting into
three pieces (median and lateral pieces of brain
and SG) and showed that EH activity was highest
in the brain median part [12]. Thus, the present
immunohistochemical results agree well with this
previous report.

Copenhaver et al. have already identified EH-
producing cells in the moth, Manduca, sexta [5].
By an immunohistochemical study using an anti-
serum against Manduca EH, they revealed that 5
pairs of group Ia cells that project to ipsilateral
corpus cardiacum and corpus allatum contained
EH. It is of interest that there is such a great
difference in the number of EH producing cells
between the two lepidopteran insects, Bombyx
and Manduca.

Recently, Mizoguchi et al. [13] reported four
pairs of Bombyx median neurosecretory cells con-
tained bombyxin, a Bombyx neurosecretory pep-
tide that activates the prothoracic glands of the
saturniid moth Samia cynthia ricini, by immunohis-
tochemistry using a monoclonal antibody against a
synthetic fragment of bombyxin. By a whole
mount immuno-staining study using the anti-
bombyxin antibody, the bombyxin cells were lo-
cated in the dorsal-posterior position and proven
different from the EH immunoreactive cells
(photographs not shown).

ACKNOWLEDGMENTS

We are grateful to Drs. M. Nagata and A. Takenaka of
The University of Tokyo, for their technical advice in
immunohistochemistry, and to Mr. K. Soma and Miss I.
Kubo for their technical assistance. This work was partly
supported by Grants-in-Aid for Scientific Research (Nos.
61560135, 62560117 and 63430021) from the Ministry of
Education, Science and Culture of Japan.

REFERENCES

1 Fugo, H. (1982) The eclosion behaviour of the
silkworm, Bombyx mori and its hormonal control. J.
Seric. Sci. Jpn., 51: 523-527.

2 Truman, J. W. (1985) In “Comprehensive Insect
Physiology, Biochemistry, and Pharmacology,” Vol.
8, Ed. by G. A. Kerkut and L. I. Gilbert, Pergamon
Press, Oxford, pp. 413-440.

54 T. Kono, A. Mizocucui et al

Truman, J. W., Taghert, P. H., Copenhaver, P. F.,
Tublitz, N. J. and Schwartz, L. M. (1981) Eclosion
hormone may control all ecdysis in insects. Nature,
291: 70-71.

Chen, J. H., Fugo, H., Nakajima, M., Nagasawa,
H. and Suzuki, A. (1986) The presence of neuro-
hormonal activities in embryos of the silkworm,
Bombyx mori. J. Seric. Sci. Jpn., 55: 54-59.
Copenhaver, P. F. and Truman, J. W. (1986) Iden-
tification of the cerebral neurosecretory cells that
contain eclosion hormone in the moth Manduca
sexta. J. Neurosci., 6: 1738-1747.

Kono, T., Nagasawa, H., Isogai, A., Fugo, H. and
Suzuki, A. (1987) Amino acid sequence of eclosion
hormone of the silkworm, Bombyx mori. Agric.
Biol. Chem., 51: 2307-2308.

Marti, T., Takio, K., Walsh, K., Terzi, G. and
Truman, J. W. (1987) Microanalysis of the amino
acid sequence of the eclosion hormone from the
tobacco hornworm Manduca sexta. FEBS Lett., 219:
415-418.

Kataoka, H., Troetschler, R. G., Kramer, S. J.,
Cesarin, B. J. and Schooley, D. A. (1987) Isolation
and primary structure of the eclosion hormone of
the tobacco hornworm, Manduca sexta. Biochem.

10

11

12

13

Biophys. Res. Commun., 146: 746-750.

Carlsson, J., Drevin, H. and Axen, R. (1978) Pro-
tein thiolation and reversible protein-protein con-
jugation. Biochem. J., 173: 723-737.

Hawkes, R., Niday, E. and Gordon, J. (1982) A
dot-immunobinding assay for monoclonal and other
antibodies. Anal. Biochem., 119: 142-147.
Nagasawa, H., Kamito, T., Takahashi, S., Isogai,
A., Fugo, H. and Suzuki, A. (1985) Eclosion
hormone of the silkworm, Bombyx mori: purifica-
tion and determination of the N-terminal amino acid
sequence. Insect Biochem., 15: 573-578.

Fugo, H. and Iwata, Y. (1983) Change of eclosion
hormone activity in the brain during the pupal-adult
development in the silkworm, Bombyx mori. J.
Seric. Sci. Jpn., 52: 79-84.

Mizoguchi, A., Ishizaki, H., Nagasawa, H.,
Kataoka, H., Isogai, A., Tamura, S., Suzuki, A.,
Fujino, M. and Kitada, C. (1987) A monoclonal
antibody against a synthetic fragment of bombyxin
(4K-prothoracicotropic hormone) from the silk-
worm, Bombyx mori: characterization and immuno-
histochemistry. Mol. Cell. Endocrinol., 51: 227-
235.

ZOOLOGICAL SCIENCE 7: 55-61 (1990)

Vanadium-Containing Blood Cells (Vanadocytes) Show No
Fluorescence Due to the Tunichrome in the Ascidian,
Ascidia sydneiensis samea

Hitosui MicuiBaATA, TARO UyaMA and Junko Hirata!

Biological Institute, Faculty of Science, Toyama University,
Toyama 930, Japan

ABSTRACT—Ascidians belonging to the family Ascidiidae are known to accumulate vanadium ions
from seawater to levels in excess of one million times the level in seawater and to maintain the vanadium
ions in a reduced form. A tunichrome which appears to be involved in the accumulation and reduction
of vanadium ions produces an autonomous fluorescence upon excitation with blue-violet light. Among
six different types of ascidian blood cell, the morula cell, which emits fluorescence brightly, has been
thought to be the vanadium-containing blood cell (vanadocyte) and, consequently, it has been suggested
that the intensity of fluorescence is indicative of the concentration of vanadium ions in the blood cell.

In the present experiments, after ascidian blood cells were fractioned into the various subpopulations
by means of Ficoll density gradient centrifugation, the level of vanadium in each subpopulation was
determined to ascertain which type of blood cell is the true vanadocyte in Ascidia sydneiensis samea.
The autonomous fluorescence from the vanadocyte was also monitored with a fluorescence microscope.
Consequently, we found that the subpopulation of morula cells that fluoresced brightly did not contain
vanadium, whereas the subpopulation of signet ring cells, which did not emit fluorescence, contained

© 1990 Zoological Society of Japan

high levels of vanadium.

INTRODUCTION

The ability of ascidians to concentrate vanadium
ions, to levels in excess of one million times the
level in seawater, is a source of special fascination
[1, 2]. A tunichrome, which can be extracted from
the blood cells of Ascidia nigra, has been proposed
to be involved in the accumulation of vanadium
ions from seawater [1, 3-5]. This substance has
been reported to emit a specific autonomous
fluorescence upon excitation with blue-violet light
[4, 6]. Among several different types of ascidian
blood cell examined, the strongest fluorescence
that could be ascribed to the tunichrome was
observed in the morula cell. Thus, it was suggested
that the tensity of fluorescence is indicative of
the concentration of vanadium ions in the cells [4].

Accepted February 20, 1990
Received December 21, 1989

' Present address : Faculty of Pharmaceutical Sciences,
University of Tokushima, Tokushaima 770.

However, we have already verified that the
morula cell contains no vanadium, whereas the
signet ring cell contains a very high level of vana-
dium ions, in the case of A. ahodori. We based our
conclusions on the results of cell fractionation
techniques, neutron activation analysis and elec-
tron spin resonance spectrometry (ESR) [7].

In the present experiments, we examined
whether the signet ring cell (vanadocyte), sepa-
rated by Ficoll density gradient centrifugation,
emits an autonomous fluorescence due to the
tunichrome, in order to verify any participation by
this substance in the accumulation of vanadium
ions in ascidian blood cells from seawater.

MATERIALS AND METHODS

Ascidia sydneiensis samea wete collected in the
bay of Nanao, Ishikawa Prefecture, Japan, and
were maintained in an aerated seawater aquarium
at 18° C. Blood, drawn by making an incision
through the lower part of the tunic and puncturing

56 H. Micumpata, T. UYAMA AND J. HirATA

the mantle, was suspended in artificial seawater
(ASW) that contained 460 mM NaCl, 9mM KCl,
33mM Na SO,, 6mM NaHCO3, 1mM
EDTA(ethylenediamine-tetraacetic acid) and 5
mM HEPES (N-2-hydroxyethylpiperazine-N’-2-
ethanesulfonic acid) buffer (pH 7.0) to avoid clot-
ting. The suspension was separated into blood
cells and plasma by centrifugation at 300 x g for 10
min at 10°C. The pellet obtained was resuspended
at a concentration of about 107 cells/ml in ASW
and is referred to as “washed cells”. Thereafter,
cell fractionation by Ficoll density gradient centri-
fugation was carried out in a manner similar to that
described previously [7]. Ficoll type 400 (Pharma-
cia Fine Chemicals) was dissolved in ASW to final
concentrations of 34.0, 18.0, 14.5 and 4.0% (w/v)
and discontinuous gradients were prepared in 10-
ml centrifuge tubes. One ml of washed cells was
layered onto each gradient and tubes were centri-
fuged at 300g for 20 min at 10°C. Layers of cells
were gently pipeted from the top of the tubes.
Each layer of cells obtained in this way was washed
twice with ASW by centrifugation at 350 x g for 10
min in order to remove Ficoll 400 and was resus-
pended in a small amount of ASW. The fractioned
populations of cells were subsequently used for
determination of levels of vanadium by neutron
activation analysis at the Institute for Atomic
Energy of Rikkyo University, Yokosuka, Japan
[2]. The blood cells which were resuspended in a
small amount of ASW were observed with a stan-
dard bright field microscope and a fluorescence
microscope (Nikon). They were also observed
after vital staining with neutral red, nile blue and
Janus green.

RESULTS

Morphology and fluorescence of blood cells

We were able to recognize six different types of
cell: the giant cell, signet ring cell, morula cell,
compartment cell, pigment cell and hyaline leuco-
cyte. These types of cell were classified mainly
according to the criteria of Wright [8] and Rowley
[9].

Subpopulations of the blood cells are present at
variable proportions in individuals in this species.

The giant cell was the second most abundant cell
type accounting for 21 to 30% of the total cells.
This cell was very large and spherical or irregularly
shaped. It was 40 to 804m in diameter and
contained a single, very large, fluid-filled vacuole
which occupied most of the cell (Fig. 1A). The
vacuole was coloured faint red and faint violet
after staining with neutral red and nile blue, re-
spectively. The cell weakly emitted a pale green
fluorescence (Fig. la). The giant cell is probably
analogous to the nephrocyte [8]. It seems however
reasonable that the cell should be designated as a
giant cell because it is not apparent that the giant
cell is involved in excretion.

The morula cell was round to ovoid, 8 to 10 ~m
in diameter. As shown in Figure 1B, the morula
cell in this species exhibited refractive cytoplasm
under bright field illumination and a few cells
appeared typical berry-like shape, differing from
the morula cell in the other species. It was
accounted for 6 to 12% of the total population.
This cell appeared red, blue-green and green after
staining with neutral red, nile blue and Janus
green, respectively, and emitted autonomous
fluorescence with a yellow green colour upon
excitation with blue-violet light (Fig. 1b).

The signet ring cell, 10 to 12 ~m in diameter,
which comprised about 32 to 44% of the total
population of cells, predominated. This cell was
characterized by a single and fluid-filled vacuole
which displaced the nucleus and cytoplasm to the
periphery of the cell (Fig. 1C). A single, small,
refractive vesicle was suspended in the vacuole.
Such vacuole was coloured a faint red after stain-
ing with neutral red. The small vesicle was dyed a
red and green with neutral red and Janus green,
respectively. No fluorescence was detected from
this cell upon excitation with blue-violet light (Fig.
Ic).

The compartment cell, was ovoid, 6 «m in dia-
meter, and accounted for 15 to 21% of the total
population of cells (Fig. 1D). This cell was dyed a
red with neutral red. The granules in the cyto-
plasm appeared deep red and green after staining
with neutral red and Janus green, respectively. No
fluorescence was detected from the compartment
cell (Fig. 1d).

Pigment cells (Fig. 1E) were relatively rare. The

Fic.

Vanadocytes Show No Fluorescence Si

1. Blood cells of Ascidia sydneiensis samea were observed with a bright field microscope (A-F) and a
fluorescence microscope (a-f). The giant cell, the second most abundant cell type, was 40 to 80 4m in diameter
and contained a single, very large, fluid-filled vacuole (A). This cell weakly emitted a pale green fluorescence
upon excitation with blue-violet light (a). The morula cell had refractive cytoplasm (B) and a few cells appeared
typical berry-like shape. This cell emitted autonomous fluorescence with a yellow green colour (b). The signet
ring cell, which contained large amounts of vanadium, was characterized by a single and fluid-filled vacuole and
refractive vesicle in the vacuole (C). This cell type was predominant in the blood cells. From this cell no
fluorescence was detected upon excitation with blue-violet light (c). The compartment cell was small and ovoid,
accounting for 15 to 21% of the total population of cells (D). No fluorescence was also detected from this cell (d).
The pigment cell shows red, dark orange or brown colour (E) and emitted bright coloured fluorescence (e). The
hyaline leucocyte (F) fluoresced most brightly among the blood cells in this species (f). Scale bar indicates 10 pm.

58 H. Micureata, T. UYAMA AND J. HIRATA

pigment cell emitted bright orange coloured
fluorescence (Fig. le). The cells shown in Figure
1F seem to be hyaline leucocytes and fluoresced
most brightly among the blood cells in this species

Cell flactionation

The blood cells were partitioned into four dis-
crete layers that contained various subpopulations,

after Ficoll density gradient centrifugation. Most

(Fig. 1f).
TaBLE 1. Distribution of each subpopulation of blood cells prior to and following separation by
centrifugation on a Ficoll density gradient
Total cell . Signet rin Compartment
Raranee Giant cells 8 calls 8 Morula cells Pat e
Washed 79,422 + 16,250 20,545 + 10,789 29,398 + 7,274 8,344 +1,210 13,936+2,370
cells (100.0) (63.9) (14.0) (G2) (8.5)
Layer 1 2,294 + 808 1,466+555 BY) Gis Oh) 73 +66 194+59
(100.0) (63.9) (14.0) (3.2) (8.5)
Layer 2 3,096 + 1,189 — 2,816+1,119 — 217+70
(100.0) (91.0) (7.0)
Layer 3 875 + 164 — 465 +171 — 396 +56
(100.0) (53.1) (45.3)
Layer 4 517+ 167 — — 471 +159 —
(100.0) (91.1)
Total number 6,782 1,466 3,602 544 807
of cells
recovered

Each number of blood cells is shown in 1,000 cells and as the mean+standard error. Figures in the parenthesis

express % of total number of blood cells in the washed cells and each layer.

(*/o)
100

Percentage of distribution in each layer

12 3 4& (Layer)

Vanadium

VEZ

Compartment
cell

234

Morula cell

1234 1

Signet ring
cell

1234

Giant cell

Fic. 2. Comparison of the patterns of distribution of giant cells, signet ring cells, morula cells and compartment cells
with that of vanadium ions after density gradient centrifugation of blood cells of A. sydneiensis samea.
Histograms depict the relative numbers of each type of cell and the amounts of vanadium distributed in the four
layers of cells as percentages of the total number of cells and the total amount of vanadium. Each bar represents
the average of results of five trials. The pattern of distribution of vanadium is similar to that of the signet ring cells

but is different from those of the other cell types.

Vanadocytes Show No Fluorescence 59

of the separated blood cells in each layer did not
take up eosin Y, indicating that they were still
alive. The results obtained represent the average
of results of five trials ++S.E. (standard error) and
are shown in Table 1. About 6 to 12% of each
subpopulation were recovered from the various
layers, which low rates were due to sacrifice the
recovery rate to obtain purer subpopulation of
blood cells.

The percentages of giant cells, signet ring cells,
morula cells and compartment cells distributed
among each of four layers, are presented graphi-
cally in Figure2. All of the giant cells were
gathered in layer 1, suggesting that they were of
low density. About 79% of the signet ring cells
were present in layer 2 and the remaining 8% and
13% were found in layer 1 and 3, respectively.
Almost all of the morula cells (88%) were present
in layer 4, indicating that this cell type has highest
density among the blood cells in this species. Half
of the subpopulation of the compartment cells was
found in layer 3 and the other half was divided
between layer 1 and 2.

Vanadium content

Neutron activation analysis revealed that 127.5
vg of vanadium was contained in the washed blood
cells and 14.7 yg of the metal, which corresponded
to about 12% of the initial amount in the washed
cells, was recovered from the fractioned blood

TABLE 2. Vanadium content of layers sepa-
rated by Ficoll density gradient centri-
fugation

Total content (ng)

Washed cells 127,540 + 32,240
Layer 1 1,370+640
(9.3%)
Layer 2 10,220+3,640
(69.4%)
Layer 3 2,900 + 820
(19.7%)
Layer 4 230+ 150
(1.6%)
Recovered vanadium 14,720

after separation

Each value represents the average of five trials +
standard error. Figures in the parenthesis ex-
press % of recovered vanadium.

cells as shown in Table. 2. The highest percentage
of vanadium distributed among four layers was
found in layer 2, in which about 70% of the
vanadium was present. The remaining 9 and 20%
of vanadium were distributed in layer 1 and 3,
respectively.

The pattern of distribution of vanadium was
compared with those of blood cells in Figure 2.
This pattern was very similar to that of the signet
ring cells, but was clearly different from that of the
morula cells and of the other cell types. Furth-
ermore, the following data strengthen evidence
that the signet ring cell is the vanadocyte. As has
been pointed out, the proportion of each type of
blood cell varied in individuals in this species.
Hence, the signet ring cells recovered from layer 2
also varied in number through 1,664,000 to
4,867,000 cells and vanadium content in layer 2
rose and fell in proportion to the cell number.
When the vanadium content per 1000 signet ring
cells in layer 2 was calculated in each trial, the
values were of a very narrow range of 3.04 to 4.11
ng/1000 cells (average value+S.E. was 3.71+0.4
ng/1000 cells). Such tight correlation of vanadium
content with the cell number could not be found
out in the other type of blood cell.

DISCUSSION

The vanadium ion dissolved in seawater is in the
+V oxidation state at concentrations of about 35
nM [10, 11]. Some ascidians concentrate these
ions 10°-fold in their blood cells and store the
metal ion in its reduced +III and/or +IV states [7,
12-16]. Macara et al. [1] isolated a tunichrome
from the ascidian bllod cells, which serves as a
good agent for forming complexes with vanadium
ions and which reduces the metal ion to its reduced
form. The tunichrome emits fiuorescence when it
is excited with blue-violet light and is present with
vanadium at appoximately equimolar concentra-
tions [1]. The concentration of vanadium has,
therefore, been estimated from the intensity of
fluorescence of the tunichrome in each type of
blood cell. In consequence, concentrations of both
vanadium and tunichrome were thought to be in
the order, morula cell >compartment cell > signet
ring cell [4].

60 H. Micuipata, T. UYAMA AND J. HIRATA

Although the morula cell unequivocally emitted
fluorescence in this experiment (Fig. 1), the com-
bind results of cell fractionation and neutron
activation analysis have revealed that the morula
cell dose not contain vanadium, while the signet
ring cell which dose not fluoresce contains large
amounts of vanadium (Tables 1 and 2). In fact, the
pattern of distribution of vanadium among the
fractioned layers corresponded clearly to that of
the signet ring cells as shown in Figure 2. Moreov-
er, the vanadium content per 1000 signet ring cells
in layer 2 was almost consistent in each trial.
Based on these results, it could be concluded that
the actual vanadocyte involved in the accumula-
tion of vanadium must be the signet ring cell in A.
sydneiensis samea.

If the tunichrome is involved in the reduction
and accumulation of vanadium ions in the ascidian
blood cells, it would be necessary that the
tunichrome should be contained in the signet ring
cell, which is the vanadocyte. However, no
fluorescence due to the tunichrome was detected in
this cell (Fig.1). In a different species of A.
ahodori, a similar finding that the signet ring cell
did not emit fluorescence was obtained from pre-
liminary experiments in which the blood cells were
not fractioned [17]. Tunichrome B-1, one of the
tunichromes, has been isolated from non-
separated blood cells of A. nigra and its chemical
structure has been determined. It consists of three
units of hydroxy-DOPA (3, 4-dehydroxy-
phenylalanine) [18], and it must have the ability to
reduce an oxide to its reduced state, as shown in
the report by Macara et al. [1]. There is, however,
no evidence that this substance is involved in the
reduction and accumulation of vanadium ions in
the vanadocytes in ascidian blood.

The fluorescence emitted by ascidian blood cells
[1, 3, 4, 6] cannot be attributed solely to the
tunichrome. It is well known that there are several
kinds of autonomous fluorescent substance, for
example, lipids, vitamins and porphyrins, in living
cells. Therefore, the finding of fluorescence does
not always provide evidence for the presence of a
tunichrome.

We have extracted a vanadium-binding subst-
ance which we have called vanadobin from the
blood cells of A. sydneiensis samea. This substance

is colourless and can maintain the vanadium ion in
the vanadyl form (VO(IV)), even under aerobic
conditions. Moreover, this substance has an affini-
ty for exogenous vanadium ions (V) and contains a
reducing sugar [16]. Taking all the above data into
account, we suggest that it is not the tunichrome
but rather the vanadobin that is the substance
involved in the accumulation of vanadium ions
from seawater in ascidian blood cells.

ACKNOWLEDGMENTS

We are grateful to Profs. M. Yoneda and N. Satoh of
Kyoto University for providing help in the use of
fluorescence microscope. This work was supported in
part by a grant-in-aid from the Ministry of Education,
Science and Culture, Japan (No. 62540540) and was also
supported financially by the Japan Securities Scholarship
Foundation and the Ito Science Foundation. Neutron
activation analysis was carried out under the Cooperative
Programs of the Institute for Atomic Energy of Rikkyo
University.

REFERENCES

1 Macara, I. G., McLeod, G. C. and Kustin, K.
(1979) Isolation, properties and structural studies
on a compound from tunicate blood cells that may
be involved in vanadium accumulation. Biochemical
J., 181: 457-465.

2 Michibata, H., Terada, T., Anada, N., Yamakawa,
K. and Numakunai, T. (1986) The accumulation
and distribution of vanadium, iron, and manganese
in some solitary ascidians. Biol. Bull., 171: 672-681.

3 Macara, I. G., McLeod, G. C. and Kustin, K.
(1979) Tunichromes and metal ion accumulation in
tunicate blood cells. Comp. Biochem. Physiol., 63B:
299-302.

4 Robinson, W. E., Agudelo, M. I. and Kustin, K.
(1984) Tunichrome content in the blood cells of the
tunicate, Ascidia callosa Stimpson, as an indicator of
vanadium distribution. Comp. Biochem. Physiol.,
78A: 667-673.

5 Oltz, E. M., Bruening, R. C., Smith, M. J., Kustin,
K. and Nakanishi, K. (1988) The tunichromes. A
class of reducing blood pigments from sea squirts:
Isolation, structures, and vanadium chemistry. J.
Am. Chem. Soc., 110: 6162-6172.

6 Oltz, E. M. (1987) Biorganic studies of the
tunichromes: A class of reducing blood pigments
obtained from sea squirts. Thesis of Columbia Uni-
versity.

7 Michibata, H., Hirata, J., Uesaka, M., Numakunai,
T. and Sakurai, H. (1987) Separation of vanado-

10

11

12

13

Vanadocytes Show No Fluorescence 61

cytes: Determination and characterization of vana-
dium ion in the separated blood cells of the ascidian,
Ascidia ahodori. J. Exp. Zool., 244: 33-38.
Wright, K. R. (1981) Urochordates. In “Inverte-
brate Blood Cells”. Ed. by Ratcliffe, N. A. and
Rowley, A. F., Academic Press, London, Vol. 2,
pp. 565-626.

Rowley, A. F. (1981) The blood cells of the sea
squirt, Ciona intestinalis: Morphology, differential
counts, and in vitro phagocytic activity. J. Invertebr.
Pathol., 37: 91-100.

Cole, P. C., Eckert, J. M. and Williams, K. L.
(1983) The determination of dissolved and particu-
late vanadium in sea water by X-ray fluorescence
spectrometry. Anal. Chim. Acta, 153: 61-67.
Collier, R. W. (1984) Particulate and dissolved
vanadium in the North Pacific Ocean. Nature, 309:
441-444.

Swinehart, J. H., Biggs, W. R., Halko, D. J. and
Schroeder, N. C. (1974) The vanadium and selected
metal contents of some ascidians. Biol. Bull., 146:
302-312.

Dingley, A. L., Kustin, K., Macara, I. G. and
McLeod, G. C. (1981) Accumulation of vanadium
by tunicate blood cells occurs via specific anion

15

16

17

18

transport system. Biochim. Biophys. Acta, 649:
493-502.

Bell, M. V., Pirie, B. J. S., McPhail, D. B.,
Goodman, B. A., Falk-Petersen, I. -B. and Sargent,
J. R. (1982) Contents of vanadium and sulphur in
the blood cells of Ascidia mentula and Ascidiella
aspersa. J. Mar. Biol. Ass. U. K., 62: 709-716.
Frank, P., Carlson, R. M. K. and Hodgson, K. O.
(1986) Vanadyl ion EPR as a non-invasive probe of
pH in intact vanadocytes from Ascidia ceratodes.
Inorg. Chem., 25: 470-478.

Michibata, H., Miyamoto, T. and Sakurai, H.
(1986) Purification of vanadium binding substance
from the blood cells of the tunicate, Ascidia syd-
neiensis samea. Biochem. Biophys. Res. Commun.,
141: 251-257.

Michibata, H., Hirata, J., Terada, T. and Sakurai,
H. (1988) Autonomous fluorescence of ascidian
blood cells with special reference to identification of
vanadocytes. Experientia, 44: 906-907.

Bruening, R. C., Oltz, E. M., Furukawa, J., Naka-
nishi, K. and Kustin, K. (1985) Isolation and
structure of tunichrome B-1, a reducing blood pig-
ment from the tunicate Ascidia nigra L. J. Am.
Chem. Soc., 107: 5298-5300.

ZOOLOGICAL SCIENCE 7: 63-72 (1990)

© 1990 Zoological Society of Japan

Organization and Development of Reflecting Platelets in
Iridophores of the Giant Clam, Tridacna crocea Lamarck

Y OSHIHISA KAMISHIMA

Department of Biology, Faculty of Science, Okayama University,
Okayama 700, Japan

ABSTRACT— Giant clams show brilliant coloration on the mantle. The color comes from iridophores
which are distributed in the mesenchyme. Each iridophore contains thin reflecting platelets which are
aligned uniformly in rows. The platelet is bound with a membrane and has a fine substructure with a 7
nm lattice. A flat cistern intervenes two neighbouring platelets to keep the interspace constant.

In developing iridoblast, the reflecting platelets are formed in a confined area around the nucleus.
Various types of vacuoles representing transitional froms from the ER to a mature reflecting platelet are
seen in the area. Golgi vesicles are involved in the platelet formation. They are incorporated onto the
vacuolar membrane while accumulation of the dense reflecting substance takes place in the vacuolar
lumen. The dense substance is condensed in the vacuoles. The vacuoles are then fashioned into thin
rectangular platelets and aligned in rows to form an alternating reflecting surface. The intervening
cistern is formed from vesicles which fuse with one another to become a flat, thin cistern.

INTRODUCTION

In molluscs, two types of iridophores (irido-
cytes) have been identified. The first type is. the
iridophore which contains small granular or vesicu-
lar organelles by which the incident light is split
and sent backwards to effect the Tyndall phe-
nomenon. This type of iridophore has been
observed in the mantle of opisthobranchiate gas-
tropods [1]. The second type is the iridophore
which shows a color by reflection and interference
through multilayered platelets arranged uniformly
in the cytoplasm. Cells of the second type are
observed in the skin of cephalopods [2, 3], and in
the mantle tissue of some bivalved shells [4, 5].
Because of their poorly arranged platelets, some of
these cells display less effective coloration and
have been referred to as reflector cells [6].

Iridophores of the giant clam show a clear
monochromatic coloration in various spectral
ranges according to the anatomical location of the
cell. These iridophores contain multiple rows of
reflecting platelets, each of which is uniform in
thickness [4, 7]. The precision with which the

Accepted March 22, 1989
Received February 6, 1989

platelets are arranged is reflected in a narrow
range of spectrum, which leads to the giant clam
producing one of the prominent colorations among
molluscs.

In spite of their high efficiency as chroma-
tophores, no detailed studies have so far been
reported on the structure and development of the
clam iridophores. This study intends to show the
ultrastructural organization and the devolopment
of reflecting platelets in iridophores of the giant
clam.

MATERIALS AND METHODS

Giant clams (Tridacna crocea) were collected
from the Ryukyu Islands, the south-western
archipelago of Japan. The brilliantly colored por-
tion of the mantle tissue was excised and minced
into small blocks in 3% glutaraldehyde fixative
buffered to pH=7.4 with 0.1 M phosphate solu-
tion. Tissue blocks were further fixed in the same
solution at room temperature for 2 hr and then
transferred into 1% osmium tetroxide solution
buffered with 0.1 M phosphate to pH 7.4. After
the post osmification for 1.5 hr in the solution at
the room temperature, tissue blocks were dehy-
drated through the ethanol series and embedded in

64 Y. KAMISHIMA

Iridophore Formation in Giant Clam 65

epoxy resin. Thin sections obtained on the LKB
ultramicrotome 4800 A or the Porter-Blum MT-I
ultratome mounted with glass knives were ex-
amined under the Hitachi Electron Microsope
HU-11E.

RESULTS

Morphology of the iridophore

Mantle iridophores of the giant clam were
observed in the mesenchyme. They were distri-
buted mostly in clusters among muscle cells and
positioned to cover the outer layer of digestive
glands in which zooxanthellae, the algal sym-
bionts, were colonized (Fig. 1). The iridophore
was spherical or oval in shape. A round nucleus
was usually lacated in the peripheral cytoplasm
and the rest of the cell was filled with rows of
reflecting platelets (Fig. 2). Mitochondria, vesicles
and ribosomes were observed around the nucleus.
A small number of vesicles and ribosomes were
also found in the interspaces between the platelets.

The reflecting platelets were rectangular in
shape (Fig.5). Thickness of the platelet was
uniform within each cell(Fig. 2), although it dif-
fered from cell to cell, ranging from 80 nm to 120
nm. The platelet was enveloped with a single
limiting membrane, which measured 7 nm in thick-
ness and was slightly thinner than the plasma or
ER membrane (Figs. 3 and 4). At the end of the
platelet, vesicles and ribosomes were often seen
closely associated with the platelet (Figs. 3~6).
The reflecting body mass of the platelet was elec-
tron dense and had a substructure of fine lattice
(Fig. 6). The lattice consisted of an alternate
arrangement of dense and light lines (2.0 nm and
5.0 nm in thickness, respectively) at 7.0 nm inter-
vals. There was a narrow marginal space of 3 nm

between the envelope and the inner reflecting
body mass (Fig. 6). All platelets in a mature
iridophore faced one direction and were aligned in
parallel rows with a set interval (Figs. 2~3). In
each row, the platelets were arranged end to end,
forming a broad reflecting plane (Fig. 5).

Flat and long cisterns lay between the platelet
rows. The cistern was tightly secured between two
platelets, ensuring a constant interspace (Figs. 2~
4). The width of the cistern was uniform within
each cell but differed slightly according to the cell.
The width ranged from 50 nm to 60 nm in most
cases, so that the distance between the dense
reflecting masses of the platelets measured from 80
nm to 100nm. The lumen of the cistern appeared
to be empty, although a few granular substances
were detected on the internal surface of the cister-
nal membrane. Cisterns were found only between
the platelet rows, so that no cisterns were found
along the outer side of the platelets at the extrem-
ity of the row (Figs. 3 and 11). Cisterns were often
seen fusing with the cell membrane at the marginal
end, so that their lumina were directly opened to
the extracellular space (Fig. 4). The opening was
always covered with the solid basal lamina which
did not invaginate concomitantly with the plasma
membrane.

Development of the iridoblast

In iridoblasts, especially in those in earlier de-
velopmental stages, many endoplasmic reticula
and Golgi complexes were seen around the nucleus
(Fig. 7). Numerous vacuoles and vesicles of va-
rious shapes and sizes were also observed in this
area (Figs. 7~13). Developing platelets were
observed in the vicinity of this area. The develop-
ing platelets were smaller than mature ones and
often lacked the associating cisterns along them
(Figs. 8, 11 and 12).

Fic. 1.

Electron micrograph of the mantle tissue of a giant clam. Iridophores (I) are observed in clusters among

musule cells (M). The iridophore is spherical and has a nucleus in the periphery. Most part of the cytoplasm is
occupied with reflecting platelets. Each cell contains 20 to 30 rows of reflecting platelets which are aligned in
parallel one another and are arranged around the nucleus. The orientation of the platelets differs in each cell.

Bar indicates 1 wm. 3,500

Fic. 2. Transverse profile of the platelet in a giant clam iridophore. Each row consists of a series of platelets (P)
arranged end to end. Interspace between the platelet rows is kept in uniform distance by an intervening cistern
(C). Only few organelles such as mitochondria, small vesicles and ribosomes are seen around the nucleus (N).
Thick basal lamina (Bl) is seen outside of the iridophore plasma membrane (Pm). Bar indicates 1 um. 12,000

66

Fic.

Fic.

Y. KAMISHIMA

3. A portion of platelet forming area near the cell surface showing developing cisterns. A flat cistern (C) with 50
nm to 60 nm width is tightly secured between two platelets (P). No cisterns are seen along the outerside of the
marginal platelets. Distended vesicles (Vs) are seen along the platelets. Ribosomes are seen among the platelets.
Bar indicates 1 ~. «54,000

4. A peripheral portion of an iridophore sectioned transversely to the platelet row. Some of cisterns fuse with
the plasma membrane at the periphery, so that the cisternal lumnen opens directly to the extracellular space
(arrows). Note dilated vesicles (Vs) at the position of the cistern. Thick basal lamina (Bl) is seen outside of the
plasm membrane (Pm). Bar indicate 1 “~m. x 43,700.

Iridophore Formation in Giant Clam 67

Fic. 5. Para-horizontal section through the platelets (P). Platelets are rectangular in shape and differ in size. They
are arranged closely side by side to form a single plane of the reflecting surface. Bar indicates 1 um. <21,000

Fic. 6. Higher maginification of reflecting platelets in transverse profiles. Each platelet (P) is enveloped with a
membrane (E). The envelope is separated from the inner platelet mass by 3nm. The platelet shows fine
substructure of 7 nm lattice. The cisterns (C) with 60 nm width are observed between neighbouring platelets. Bar
indicates 0.1 ~m. X 150,000.

Y. KAMISHIMA

68

Iridophore Formation in Giant Clam 69

In the platelet forming area, three types of
vacuoles were observed (Figs. 7~12). The first
type of vacuole (marked as V1 in Figs. 7, 9 and 10)
was irregular in shape and resembled distended
rough surfaced endoplasmic reticulum. Some of
these vacuoles contained various cytoplasmic com-
ponents, such as vesicles and ribosome-like gra-
nules. The second type of vacuole (marked as V2
in micrographs) were round in profile and con-
tained fluffy materials in the lumen. The second
type of vacuole had a complex internal membrane
structure and often appeared as a double walled
vacuole (Figs. 8, 10~12). The last type (marked
as V3 in micrographs) often had an elipsoidal or
even elongated shape and contained a dense amor-
phous substance which was similar in appearance
to the internal mass of the reflecting platelet.

A cluster of vesicles or smaller cisterns (marked
as Vs in Figs. 3, 4 and 13) were observed along the
developing platelets. These vesicles were aligned
in a line at the place of the cistern between
neighbouring platelets. Newly formed platelets
were often seen in close contact, because of the
absence of an intervening cistern (Fig. 12). Golgi
bodies were often found in the platelet forming
area in iridoblasts (Fig. 7). The outer lamella of
the Golgi stack was distended in these cells. Golgi
vesicles were also seen associated with the
vacuoles or the devoloping platelets (Figs. 9 and
11). Some of the Golgi vesicles were seen directly
fusing with the limiting membrane of the develop-
ing platelet (Fig. 11). The lumina of the Golgi
lamellae and vesicles were filled with opaque mate-

rial. Centrioles were often detected in close asso-
ciation with the Golgi complex. Microtubules
were seen radiating from the centriole.

DISCUSSION

The overall appearance of the clam iridophores
resembled those in cephalopods [2] and verte-
brates [4, 8, 18]. The cytoplasm of the cell is fully
occupied with tightly packed reflecting platelets.
Platelets are aligned in rows and form multiple
reflecting planes at each interface with the cyto-
plasm. Each platelet is bound with a membrane:
the envelope. The membrane measures 7 nm in
thickness and its dimensions are virtually identical
to those of the plasma or cisternal membrane. The
platelet is rectangular in shape and it has uniform
thickness in each cell. The dimension, however,
differs from cell to cell, ranging from 80 nm to 120
nm. The reflecting body of the platelet is electron
dense and has a fine lattice that resembles the
paracrystalline structure of proteins.

There are long and flat cisterns tightly secured
between the rows of membrane bound platelets.
Since the cistern is fairly uniform in width measur-
ing 50 nm to 60 nm, the space between platelets is
also kept in uniform. The intervening cistern
illustrates the mechanism that secures the inter-
platelet space at a definite distance which is re-
quired for the efficient coloration of the iri-
dophores [18]. In this respect, the clam iri-
dophores differ from the cephalopod iridophores,
in which the space between the platelets is the

Fic.

Fic.

Fic.

Fic.

7. The platelet forming area near the cell periphery of a developing iridoblast. Diverse forms of vasuloes (V1
and V3) are seen around the Golgi body (G) and the centriole (*). Golgi vesicles (g) are seen associated with
vacuoles. Some of vacuoles (V3) show the same internal density with the developing platelelts (P). A small
cistern (c) is observed between developing platelets on the left of the micrograph. Bar indicates 1 wm. 35,800.
8. Platelet forming area where some of vacuoles are seen double walled (V2). Fluffy material in the vacuoles
appears similar to the cytoplasmic matrices. Developing cistern (C) formed by fusion of vesicles are seen between
two developing platelets (P). Bar indicates 1 ~m. 32,000

9. Another platelet forming area showing condensation of dense material in the vacuole (V1 and V2). Vesicles,
ribosome-like granules and fluffy materials are seen in the lumina of the vacuoles. Some vacuoles (V2) are double
walled. Golgi vesicles (g) are seen accumulating around the vacuole with dense materials. Bar indicates 0.5 ~m.
x 45,000

10. Vacuoles showing transitinal stages of the platelet formation. Condensation process of the dense materials
is seen in the developing platelets (marked V1, V2, and V3, which indicate the transitional vacuoles in the
numerical order). Vacuole at the final stage (V3) sppears almost similer to the platelet, except its loose envelope
and less dense internal mass. Cytoplasmic components are seen inside of all vacuoles. Transverse sections
through the dense mass of the V3 vacuoles are shown as V3s in Figs. 7 and 13. Bar indicates 1 wm. 23,000

70 Y. KAMISHIMA

yo oe
a,

EF

aan

Iridophore Formation in Giant Clam 71

extracellular, so that the platelet is not bound to
membrane and is free in the cytoplasm [3, 9, 6, 10].

As well as the structural organization, the
physical or chemical nature of the platelet may
influence the coloration of the cell. Clam iri-
dophores are sectioned smoothly on the glass
knives, suggesting pliablity of the platelet element,
while purine (such as guanine) platelets in verte-
brate iridophores are brittle, so that well sectioned
profiles for electron microscopy are not easily
obtainable. Thus, purines do not seem to be the
major component of platelets in the clam iri-
dophore. The platelets in cephalopod iridophores
are also readily sectioned and appeared similar to
those of clam iridophores under the electron
microscope [2, 3, 7]. There have been diverse
reports on the nature of platelets in cephalopod
iridophores, such as guanine [3], purines [7], chitin
[11] or protein [6, 9]. The paracrystalline structure
with fine lattice observed in this study may indicate
that the platelet in clam iridophores is mainly of
proteineous rather than purine nature, although
the latter cannot totally be excluded.

Clam iridophores display various colorations
ranging from blue to yellow-green, or sometimes
even to red, depending on their distribution. This
indicates that the thickness and/or arrangement of
reflecting platelets differ accordinng to the cell. As
mentioned previously, the thickness differs from
cell to cell ranging from 80 nm to 120 nm, although
it is fairly uniform within a single cell. Providing
that the optical path of the reflecting layer in
efficient iridophores equals a quarter wavelength
of the reflected spectrum [12, 18], the platelet in
clam iridophores which has blue to red interfer-
ence colors and a thickness ranging from 80 nm to
120 nm seems to have a refractive index (n) of

around 1.5. This is almost the same value as that
of chitin [11].

In developing iridoblast, platelets are formed in
a confined area around the nucleus. In this platelet
forming area various organelles, such as endoplas-
mic reticula, vesicles and vacuoles are found. The
vacuole seems to be derived from the endoplasmic
reticulum and finally become a platelet envelope,
because various forms that suggest a gradual tran-
sition from the distended endoplasmic reticulum
(V1) to the envelope (V3) are observed in the
area. Vacuoles containing vesicles and ribosome
granules (V1) are morphologically similar to the
“vesiculo-globular bodies” or the “multivesicular
bodies” observed in developing melanoblasts in
mouse skin [13] or goldfish fin [14], respectively.
These vacuoles seem to be in the earlier stages of
the platelet formation in the giant clam iridophores
and are considered to be equivalent forms to the
primordial vesicle proposed by Bagnara [15, 16]
for the common precursor to all pigment
organelles in vertebrate chromatophores. Double
walled vacuoles are frequently seen in the platelet
forming area (V2 in Figs.9, 11 and 12). The
similar structure in vertebrate iridoblasts (double
walled saccule) is shown to be formed by an
invagination or infolding of a vacuole (7). Howev-
er, this is not demonstrated in the clam iridoblasts.

The accumulation process of the dense reflecting
material is also shown in transitional internal struc-
tures of the vacuoles. Since the lumina of the
earlier vacuoles are slightly denser than the cyto-
plasm, the accumulation of the reflecting material
in these vacuoles seems to have a low concentra-
tion (V1 and V2). In the later stage of the vacule,
dense materials appear at the middle of the
vacuole and are gradually condensed into a rec-

Fic. 11.

A portion of the platelet forming area. No cistern are observed along th developing platelets (P) at the

margin of the row. Golgi vesicles are seen fusing with the limiting membrane of the developing platelets (arrows).
Double walled vacuole (V2) with fluffy material in the lumen is seen close to the developing platelets. Bar

indicated 1 wm. 32,500

Fic. 12. Newly formed platelets (P) which are not yet arranged into a row are seen near the nucleus. These platelets
show clear limiting membrane, but have no intervening cistern, so that they are closely contacted each other. Bar

indicates 1 ym. x 46,000
Fic. 13.

Platelet forming area where developing platelets (P) dispose at the margin of the platelet row. A series of

vesicles (Vs) are seen at the position of the cistern. Newly formed cisterns (c) which still appear as flattened
vesicles are positioned both sides of the developing platelets. Developing platelets show transitional forms from
spherical one (V3), to spindle (S), or rectanglar (P) one. Bar indicates 1 um. x 32,000

7 Y. KAMISHIMA

tagular platelet mass (Fig. 11, V3). When the
vacuole in this stage is sectioned through the dense
accumulation at the middle, it appears as a dark
vacuole (Fig. 7). The incorporation of Golgi vesi-
cles onto the developing platelets is frequently
observed (Figs. 7 and 11). The involvement of the
Golgi vesicles may indicate the possibility of pro-
tein and/or carbohydrate (chitinous) as compo-
nents of the reflecting platelet in the giant clam.
The intervening cisterns are formed from vesi-
cles (Vs in Figs. 3 and 4). These vesicles are also
dilated, but are somewhat smaller than the
vacuoles involved in the platelet formation. Di-
lated vesicles are seen to be depressed between
small developing platelets which are not yet assem-
bled in rows (Figs. 8 and 13). These vesicles are
opaque in appearance and contain flocculent mat-
ters in the lumina. They appear in a line between
developing platelets and seem to fuse with each
other to form a long and flat cistern (Figs. 3 and 4).

ACKNOWLEDGMENTS

The author is greatly indebted to Professor Siro Kawa-
guti of Kawasaki Paramedical School for kindly provid-
ing the materials, and is also very much grateful for his
valuable suggestions during the work.

REFERENCES

1 Kawaguti, S. and Kamishima, Y. (1964) Electron
microscopic study on the iridophores of opisthob-
ranchiate mollusks. Biol. J. Okayama Univ., 10: 93-
103.

2 Kawaguti, S. and Ohgishi, S. (1962) Electron mic-
roscopic study on iridophores of a cuttlefish, Sepia
esculenta. Biol. J. Okayama Univ., 8: 115-129.

3 Arnold, M. (1967) Organellogenesis of the cephalo-
pod iridophore: Cytomembranes in development. J.
Ultrastruct. Res., 20: 410-421.

4 Kawaguti, S. (1966) Electron microscopy on the
mantle of the giant clam with special references to
zooxanthellae and iridophores. Biol. J. Okayama
Univ., 12: 81-92.

5 Kawaguti, S. (1968) Electron microscopy on zoox-

10

11

12

13

14

15

16

17

18

anthella in the mantle and gill of the heart shell.
Biol. J. Okayama Univ., 14: 1-11.

Brocco, S. L. and Cloney, R. A. (1980) Reflector
cells in the skin of Octopus dofleini. Cell Tissue
Res., 205: 167-186.

Kamishima, Y. (1981) Reflecting platelet formation
in iridophores. In “Phenotypic Expression in Pig-
ment Cells”. Ed. by M. Seiji, Univ. Tokyo Press,
Tokyo, pp. 279-284.

Kawaguti, S. (1965) Electron microscopy on iri-
dophores in the scale of the blue wrasse, Proc. Japan
Acad., 41: 610-613.

Mirow, S. (1972) Skin color in the squids, Loligo
pealii and Loligo opalescens. II. Iridophores. Z.
Zellforsch., 125: 176-190.

Cooper, K. M. and Hanlon, R. T. (1986) Correla-
tion of iridescence with changes in iridophore
platelet ultrastructure in the squid, Lolliguncula
brevis. J. Exp. Biol., 121: 451-455.

Denton, E. J. and Land, M. F. (1971) Mechanism
of reflexion in silvery layers of fish and cephalopod.
Proc. Roy. Soc. Lond., A175: 43-61.

Land, M. F. (1966) A multilayer interference reflec-
tor in the eye of the scallop, Pecten maximus. J.
Exp. Biol., 45: 433-447.

Takeuchi, T. and Ishiguro, S. and Tamate, H. B.
(1981) Gene expression in melanosome formation.
In “Phenotypic Expression in Pigment Cells”. Ed.
by M. Seiji, Univ. Tokyo Press, Tokyo, pp. 139-
144.

Turner, Jr., W. A., Taylor, J. D. and Tchen, T. T.
(1975) Melanosome formation in the goldfish: the
role of multivesicular bodies. J. Ultrastruct. Res.,
51: 16-31.

Bagnara, J. T., Turner, W. A., Rothstein, J., Ferris,
W. and Taylor, J. D. (1979) Chromatophore
organellogenesis. In “Pigment Cell”. Ed. by S. N.
Klaus, S. Karger, Basel, Vol. 4, pp. 13-27.
Bagnara, J. T. (1983) Developmental aspects of
vertebrate chromatophores. Amer. Zool., 23: 465—-
478.

Kamishima, Y. (1979) Electronmicroscopic study
on reflecting platelets in the dorsal iridophores of
the sand eel, Ammodytes personatus GIRARD.
Proc. Japan Acad., B54: 634-639.

Kasukawa, H., Oshima, N. and Fujii, R. (1987)
Mechanism of light reflection in blue dameselfish
motile iridophore. Zool. Sci., 4: 243-257.

ZOOLOGICAL SCIENCE 7: 73-78 (1990)

© 1990 Zoological Society of Japan

Notes on the Development of the Crab-Eating Frog,
Rana cancrivora

Minoru UcuiyaMA, TosHiki Murakami and Hipexi Yosuizawa!

Department of Oral Physiology, School of Dentistry at Niigata, The Nippon
Dental University, Niigata 951, and ‘Department of Oral Histology,
Matsumoto Dental College, Shiojiri 399-07, Japan

ABSTRACT—Fertilized eggs were obtained from a pair of crab-eating frogs collected in a mangrove
swamp in Thailand at the end of April. The tadpoles grew well when both parents and eggs were
maintained in 10% seawater, although eggs from a pair of frogs kept in 50% seawater did not develop in

10% seawater.

Clutch size was about 1800. Each egg was 1.2-1.3 mm in diameter.

Embryonic

development was fairly rapid. Hatching took place 27 hr after fertilization at 24.5-26.0°C. Individual
variation in the progress of embryonic and larval development was large. In the most rapidly growing
tadpoles, metamorphosis took place 44 days after spawning in 10% seawater. On the other hand, at
higher salinities (40-100% seawater) development tended to be delayed and tadpoles remained between
stages V and XV (Taylor-Kollros stages) 55 days after fertilization.

INTRODUCTION

The tadpole of the crab-eating frog, Rana can-
crivora, is the only amphibian larva which lives
naturally in brackish water [1, 2]. However, it is
not clear what mechanisms make this possible. It
is, therefore, important to raise tadpoles of this
species in the laboratory as a first step toward
elucidating the physiological mechanisms of salt
water adaptation. Alcala [1] described the de-
velopment of this tadpole raised from eggs col-
lected in the field with two tables and two figures.
However, he did not show the correct timetable
subsequent to fertilization. Later, Gordon and
Tucker [2] reported failure to raise artificially
fertilized eggs of this frog in various dilutions of
seawater.

In the present study, spawning was induced by
injection of pituitary homogenates and develop-
ment was observed carefully under laboratory con-
ditions. The embryonic stages were judged
according to the developmental stages for R. pi-
piens described by Witschi [3] and the subsequent
larval stages were those described by Taylor and
Kollros [4].

Accepted April 26, 1989
Received April 13, 1989

MATERIALS AND METHODS

Adult males and females of the crab-eating frog,
Rana cancrivora, were captured around prawn
culture-ponds (salinity 33%o) located in a man-
grove swamp at Ang-Sila near Bangkok, Thailand,
in late April 1987. They were shipped by air to the
laboratory in Niigata, Japan and maintained in 10-
50% seawater (3.5-18%0) at 25.0-26.0°C. Body
weight was 20-60 g for both sexes. On June 15
1987, two pairs of the adult frogs kept in 10%
seawater were injected with pituitary homogenate
(one pituitary gland per frog) of R. brevipoda
porosa and kept in a plastic tank containing 10%
seawater. A pair of frogs was found to be clasping
the following day and repeat injections were given
to them. At 6 hr after the 2nd injection, spawning
took place and the eggs were fertilized simul-
taneously by the male. The fertilized eggs were
maintained in 10% seawater at 24.5-26.0°C. Some
of the eggs were kept in small dishes for detailed
observations on development. When the embryos
reached stage 23, they were divided into small
groups (5-20 tadpoles). All observations on de-
velopmental stage were made using a binocular
dissecting microscope. Tadpoles were anesthe-
tized by means of ice and body length was mea-
sured. Pictures were taken through the dissecting

74 M. UcuiyAMA, T. MuRAKAMI AND H. YOSHIZAWA

microscope. After the tadpoles had started to
swim, they were transferred into tall-skirted dish-
es kept at 28.0°C. The tadpoles were fed on freshly
bioled spinach twice a week, and the water (10%
seawater) was changed every day.

Acclimation to various dilutions of seawater

Eggs were transferred into aged tap-water, 10%
and 20% seawater. Tadpoles of stages 21-XV were
acclimated directly to various dilutions of seawater
(tap-water to 100% seawater). Tadpoles of stages
24-XV were also acclimated in steps with 10-20%
changes of salinity every 2-7 days.

OBSERVATIONS

Breeding behavior and spawning

As noted above, injection with the pituitary
homogenates induced breeding behavior: the
clasping is axillary and the male having pigmented
vocal sacs pushed rhythmically the female’s side
and called. This breeding behavior was also
observed in mature frogs kept in 50% seawater
upon injection of pituitary homogenate and trans-
fer to 10% seawater. Six hr later, eggs were laid.
Eggs from parents that had been kept in 10%
seawater developed rapidly. On the other hand,
eggs from parents that had been kept in 50%
seawater did not develop.

Eggs and early developmental stages

Fertilized eggs formed egg masses of 6 to several
hundreds of eggs and floated on the surface of the
water. Eggs were about 1.2-1.3 mm in diameter
and were encapsulated within two transparent
layers, the outer layer being very sticky. Clutch
size was about 1800. The color of the eggs was
brown in the animal hemisphere and yellowish-
white in the vegetal one. Embryonic development
proceeded fairly rapidly and hatching was

Fic. 2.

Fic. 1.

Stage 20. An embryo moving its tail actively
just before hatching. Arrows indicate two layers of
the capsule.

observed 27 hr after fertilization. Before hatching,
embryos actively moved their tails within the cap-
sule (Fig. 1). After hatching, embryos lay on the
bottom of the tank. Thirty hr after fertilization,
the external gills were completely developed and
the tadpoles swam around. Thereafter, the right
external gills began to be covered with the opercu-
lar fold, and 96 hr after fertilization both sides of
the operculum were completed with a respiratory
pore in the left side. The tadpoles at stage 25 were
ventro-dorsally compressed. The dorsal and ven-
tral fins were relatively wide. The abdomen was
convex and the cloaca opened to the right side.
These observations are summarized in Table 1 and
Figure 2A-Q.

Larval development (limb bud stage-juvenile)

Fourteen days after fertilization, a pair of hind-
limb buds emerged. In tadpoles at stage IV, the

tooth row was fully developed with the
dental formula, 4+? (Fig. 3).

The tadpoles usually stayed on the bottom of the
dish. Before stage I, the skin of the tadpoles was
translucent, the color of the back brown, and
melanophores were deposited on the dermis of the
abdomen. Tadpoles after stage I were yellowish-

A, Stage 1; B, Stage 5; C, Stage 6; D, Stage 7; E, Stage 8; F, Stage 13; G, Stage 14; H, Stage 15; I, Stage 16;

J, Stage 18; K, Stage 19; L, Stage 20; M, Stage 21; N, Stage 22; O, Stage 23; P, Stage 24; Q, Stage 25; R, Stage
IV; S, Stage XV; T, Stage XX; U, Stage XXII, V, Stage XXIV; W, Stage XXV.

Scale bars indicate 1 mm (A-Q), and 1 cm (R-W).

Arrow indicates deposit of guanophores on the abdominal skin in tadpole of stage XV.

Development of Crab-Eating Frog TS

76 M. UcuiyAmMA, T. MURAKAMI AND H. YOSHIZAWA
TaBLe 1. Early larval development of Rana cancrivora in 10% seawater (water temperature 24.5-26°C)
Stage Time ie age Notes

1 0:05 Egg is encapsulated by two layers.
Diameter of egg is 1.2-1.3 mm (Fig. 2A).

4 1:20 Period of cleavage (Fig. 2B-D).

6 3:20

8 4:30 Blastula stage (Fig. 2E).

11 7:30 Period of gastrulation.

12 8:30 Primitive streak stage.

13 10:30 Embryo elongated. Flattened on dorsal
surface of embryo (Fig. 2F).

14 11:30 Blastopore closed (Fig. 2G).

15 12:30 Neural plate stage (Fig. 2H).

16 14:00 Neural tube stage (Fig. 21).

17 15:30 Tail bud stage. *Total length 1.8 mm.

18 16:30 Tail bud elongated (Fig. 2J).
Oral sucker distinct. Total length 1.9 mm.

19 18:00 The Ist and 2nd external gill buds become
visible (Fig. 2K). Total length 2.05 mm.

20 25:00 Tail of embryo is curved (Fig. 2L).

21 27:00 Spontaneous hatching (Fig. 2M).
Lavae lie on the bottom. Total length
3.4-4.4 mm.

22 30:00 Larvae bigin to swim. Body is asymmetrical (Fig. 2N).
Total length 4.0—4.2 mm.

23 51:00 Abdomen becomes round (Fig. 20).
Total length 4.0-5.5 mm.

24 62:00 Right external gills are covered by
opercular fold. Total length 6.0-6.2 mm.

25 96:00 Left external gills are covered (Fig. 2P, Q).

Total length 6.2-6.5 mm.

* Total length is defined as the length from tip of snout to tip of tail.

Fic. 3. Oral part of tadpole at stage X, dental formula

being

Ll+ 1
aod

gray in color on the back and silver on the abdo-
men due to guanophore deposition. Development
proceeded fairly rapidly. It took 44 days after
fertilization to begin metamorphosis in the most
rapidly growing tadpole. There were large indi-
vidual differences in the progress of larval develop-
ment, but no cannibalism was observed. The
froglets at the metamorphic climax (stage XXIII)
could not climb the glass wall of the dish. These
observations are summarized in Table 2 and Fig-
ure 2R-W.

Acclimation to various dilutions of seawater

Fertilized eggs developed well in tap-water, and
10% and 20% seawater. When hatchlings (stage
22) were transferred directly from 10% seawater to

Development of Crab-Eating Frog 77

TaBLE2. Larval development of Rana cancrivora in 10% seawater (water temperature 28°C)
Stage EERIE leugth** (mm) Notes
I 1S 6.6-10.0*** Limb buds become visible.
Ill 33 6.6-12.5 Length of limb bud equal to its diameter.
IV 20 17.0 Horny teeth fully developed (Fig. 2R).
Vv 24 21.1 Pigmentation evident on abdomen.
VI 27-43 26.0-29.0 Limb buds paddle-shaped.
x 27-52 31.1-32.0 Five toes distinct.
XII 44 28.0
XII 50 32.0
XIV 44 30.0-35.0 Toe pads appear.
XV 30-50 33.0—45.0 Hindlimbs elongate (Fig. 2S).
XVIII 44 37.0-42.0 Cloacal tail-piece disappears.
XIX 39 33.0-40.0 Skin windows become clear.
XX 36 Forelimbs appear (Fig. 2T).
XXI 39 Tail fins absorbed.
XXII 50 26.0-32.0 Tail shorter than hindlimbs (Fig. 2U).
XXII 42-46 19.0
XXIV 43 13.0 Stub of tail remains (Fig. 2V).
XXV 44 Metamorphosis accomplished (Fig. 2W).

* Individual variation in the progress of development was large.
** Total length is defined as the length from tip of snout to tip of tail and to vent in tadpole and frog,

respectively.

*** Number of animals used for measurements was 1-5.

various dilutions of seawater (tap-water to 100%),
they became well acclimated to environments up
to 40% seawater. However, tadpoles transferred
to 50% seawater or higher concentrations could
not acclimated and all died within 8 hr. When
environmental salinity was increased stepwise, tad-
poles (stages I-XV) were able to adapt to higher
salinities (tap-water to 100% seawater). Meta-
morphosis took place in tadpoles kept in tap-water
and 10% seawater. On the other hand, at 55 days
after fertilization, tadpoles kept in 40-100% sea-
water still remained between stages V and XV,
without any indication of metamorphosis.

DISCUSSION

This is probably the first report on fertilized eggs
of R. cancrivora being obtained by artificial induc-
tion of spawning and on young frogs being raised
in the laboratory. The characteristics, including
the size and time of development, of eggs and

tadpoles observed in the present study were fairly
consistent with those reported by Alcala [1].
Therefore, the present observations seem to reflect
the normal breeding behavior and development of
this species in the field. The developmental pro-
cess of this species is similar to that of R. pipiens,
except that development proceeds very rapidly. In
the present study, metamorphosis took place in the
most rapidly growing tadpoles 44 days after spawn-
ing at about 28°C. On the other hand, Taylor and
Kollros [4] reported that R. pipiens larvae meta-
morphosed at an age of over 90 days at room
temperature (about 20°C).

Gordon and Tucker [2] reported that no de-
velopment was observed when eggs from frogs
kept in 60% seawater were artificially fertilized in
20% seawater. However, the first few cleavages
occurred when eggs from adults kept in 20%
seawater were artificially fertilized and placed in
either fresh water or 20% seawater. They [2] also
suggested from their field data that spawning might

78 M. UcuiyAma, T. MURAKAMI AND H. YOSHIZAWA

occur only during and soon after heavy rains when
the salinity of the spawning pools becomes low. In
the present study, frogs kept in 50% seawater
showed breeding behavior, but eggs were unde-
veloped even when removed to 10% seawater after
fertilization. These results, therefore, suggest that
it is necessary for frogs and eggs to be kept in
hypoosmotic media in order for them to develop.

In the tolerance experiments, although the tad-
poles at developmental stages III-XIX were able to
survive well at all salinities from fresh water to
full-strength seawater (32%), eggs and tadpoles
before stage 25 could not stay in high salinity [2, 5].
Early development (stages 1-25) proceeded rapid-
ly until 4 days after fertilization, and it then took
about 10 days until the appearance of limb buds
(stage I). During the period between stage 25 and
stage I, the skin grew to be thick. This morpholo-
gical change may be one of the factors for salt
tolerance of tadpoles at these stages. Gordon and
Tucker [2] observed that metamorphosis is hin-
dered by salinities higher than 20% seawater, and
suggested from the field data that metamorphosis
is delayed as long as the pond salinity remains
high. The present observations are consistent with
their suggestion. Metamorphosis took place in
tadpoles kept in tap-water and 10% seawater and
development of tadpoles kept in higher salinities
(40-100% seawater) was delayed. These results
may suggest that a low-salinity environment is
necessary for the induction of both metamorphosis
and spawning.

According to the previous observations [1, 2]
and the present study, it can be speculated that the
following may occur in nature. After heavy rain-

fall during the rainy season, spawning occurs and
early embryonic development proceeds rapidly in
low-salinity and high-temperature water. Then,
when the tadpoles acquire salinity tolerance they
can survive in higher-salinity water. Thereafter,
the development of tadpoles proceeds further and
metamorphosis is accomplished when water salin-
ity becomes low after the next heavy rainfall. The
rapid advance of development in this species must
therefore be favored by an environment where the
salinity changes markedly.

ACKNOWLEDGMENTS

This study was supported in part by a Grant-in-Aid for
Overseas Scientific Research (No. 62041035) from the
Ministry of Education, Science and Culture of Japan.
The authors with to thank Prof. Chitaru Oguro of
Toyama University (project leader) for his thoughtful
arrangement of the project.

REFERENCES

1 Alcala, A. C. (1962) Breeding behavior and early
development of frogs of Negros, Philippine Islands.
Copeia, 4: 679-726.

2 Gordon, M. S. and Tucker, V. A. (1965) Osmotic
regulation in the tadpoles of the crab-eating frog
(Rana cancrivora). J. Exp. Biol., 42: 437-445.

3 Witschi, E. (1956) Development of Vertebrates. W.
B. Saunders. Co., Philadelphia, pp. 78-84.

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

5 Uchiyama, M., Murakami, T., Wakasugi, C. and
Sudara, S. (1987) Development and salinity toler-
ance in tadpoles of the crab-eating frog. Zool. Sci., 4:
1077.

ZOOLOGICAL SCIENCE 7: 79-84 (1990)

© 1990 Zoological Society of Japan

Reaction Mass Formation in Drosophila, with Notes on a
Phenoloxidase Activation

Nosuuiko AsApA! and TAKASHI FUKUMITSU

Biological Laboratory, Faculty of Science, Okayama University of Science,
Okayama 700, Japan

ABSTRACT— In the supernatant fraction of the adult
Drosophila, phenoloxidase activity was detected by the
use of SDS-PAGE and spectrophotometry. The level of
phenoloxidase activity increased within six minutes after
the initiation of copulation. The length of time of
reaction mass formation corresponded to the increase of
phenoloxidase activity after the initiation of copulation in
a copulated female. The reaction mass formation in
mated female Drosophila was, in part, considered to be
in relation to phenoloxidase activation after copulation.

Abbreviations

Phenoloxidase: o-diphenol: O> oxidoreductase, EC
1. 10. 3.1

SDS-PAGE: Sodium dodecyl
amide gel electrophoresis

p-NPGB: p-nitrophenyl-p’-guanidinobenzoate

EDTA: Ethylenediaminetetraacetic acid

Dopa: L-3, 4-dihydroxyphenylalanine

Dopa chrome: 2, 3-dioxyindole-5, 6-quinone-2-car-
boxylic acid

MW: Molecular weight

K: Kilodalton

sulfate-polyacryl-

INTRODUCTION

In Drosophila, reaction mass, which is to be
found in the female uterus, is formed immediately
after copulation associated with the insemination
reaction [1-6]. It takes place just after the
initiation of copulation, within 10 min, in both
intra- and interspecific crosses. Early studies
revealed that it played a role that led to prepara-
tion of the reproductive tract for oviposition and
had a bearing on the problem of speciation process
in intraspecific crosses, because reaction mass
remained soft and unmelanized then disappeared,

Accepted March 3, 1989
Received August 22, 1989
To whom all correspondence should be addressed.

probably by proteolysis, in vivo prior to oviposi-
tion. On the other hand, in interspecific crosses,
reaction mass that remained in the uterus pre-
vented oviposition resulting in a failure of hybrid
production. Therefore reaction mass seems to play
an important role not only in fecundity, but also as
a primitive self-defense response in Drosophila.
Biochemical studies revealed that reaction mass
formation was likely to be a consequence of
polymerization and/or conformational change(s)
of phenol-containing substance(s) involving the
same course of the melanization cascade reaction
[6-7]. In this article, the relationship between the
reaction mass formation and the key enzyme for
melanization, phenoloxidase, is discussed.

MATERIALS AND METHODS

1. Flies

Wild type flies of Drosophila nasuta (ADM-1
strain, Andaman, India; MYS-23 strain, Mysore,
India; SEZ-2 strain, Seychelles) and D. pallidif-
rons (PNI-74 and PNI-110 strains, Ponape, Caro-
line Islands) were used in this study. All flies were
iso-female strains caught in nature. D. pallidifrons
has been regarded as an ancestral species and D.
nasuta has been thought to be a derived species
from D. pallidifrons [8]. Culture condition and
preparation of flies were performed in the same
manner as described in Insemination reaction in
the Drosophila nasuta subgroup [6].

2. Temperature shift

Two experiments were performed. Virgin
females and males, four-day-old, were put
together in a glass vial (30 mm in diameter, 110

80 N. ASADA AND T. FUKUMITSU

mm in depth) at 25°C. To observe the copulation,
approximately 30 pairs were placed together in the
vial. In Experiment-1 (Ex-1), the vials containing
the copulating pairs were removed from 25°C and
replaced in 2°C within two min after the initiation
of copulation. After five hr at 2°C, females were
dissected in the Drosophila Ringer solution [9] and
were examined to determine whether reaction
mass was produced in the uterus. In Ex-2,
copulating pairs were placed at 2°C in the same
manner as Ex-1; then the females were separated
to avoid remating, and the culture temperature
was raised from 2 to 25°C. After four hr at 25°C,
females were dissected and examined. Fifty pairs
were run for each experiment.

3. Preparation of samples

A crude extract was prepared to test for the
presence of phenoloxidase and to study the kine-
tics of the activity. Five individuals of four-day-old
adult flies were homogenized in the solution
containing 50 ul of a 0.1 M phosphate buffer
solution (PBS), pH 6.0, at 0°C. Homogenate was
centrifuged at 12,000 round per min (rpm) for five
min at 4°C, then supernatant fraction was sub-
jected to SDS-PAGE. To assay the phenoloxidase
activity, adult flies were homogenized in the PBS
mentioned above adding 25 ul of a specific serine
protease inhibitor, p-NPGB dissolved in dimethyl-
formamide and diluted with acetonitrile (final
concentration 0.01M) and 251 of a 10mM
EDTA.

4. SDS-PAGE

The crude extract was dissolved in 50 pl of a
sample buffer solution containing 0.0625 M Tris,
2% SDS, 5% 2-mercaptoethanol, pH 6.0, and a
small amount of bromothymol blue. Extracts
boiled for two min were also prepared. The
samples were separated by 10% SDS-PAGE
according to the buffer system of Laemmli [10] for
the slab gel (2 mm gel thickness). Electrophoresis
was run for about 10 hr at a constant current of 20
mA, 65 V at the initial stage. Phenoloxidase
activity was detected by staining with a 0.013 M
dopa in a 0.1 M PBS, pH 6.0 used as a substrate
buffer solution for several min at 37°C.

5. In vitro melanization and _ phenoloxidase
activity

In vitro melanogenesis of reaction mass was
performed. A single reaction mass collected from
the uterus of a copulated female was rinsed three
times in a PBS, pH 6.8, and then soaked at 25°C in
a substrate of 5x10~°M dopa. Melanization of
the reaction mass was observed at regular inter-
vals, and the score was expressed in the same
manner as Hiruma and Riddiford [11].

Phenoloxidase activity was determined at 2, 5,
10 and 15 min from the beginning of the initiation
of copulation. The reason for these specific times
being that the average duration of copulation in
intraspecific crosses of D. nasuta and D. pallidif-
rons and interspecific crosses between D. pallidif-
rons females and D. nasuta males was approx-
imately 16, 10 and 12 min, respectively [6]. Pheno-
loxidase activity was assayed from the supernatant
fraction just after adding 1 ml of a PBS and 1 ml of
a 30% acetic acid. The color intensity of dopa
chrome was measured colorimetrically by a
Hitachi Model-101 Spectrophotometer at 475 nm.
Two replications were run at each observation
time.

RESULTS AND DISCUSSION

Temperature dependency and in vitro melanization
of reaction mass

Results of Ex-1 and Ex-2 are summarized in
Table 1. In the controled experiment, high
frequencies of reaction mass formation was
obtained in intraspecific crosses of D. nasuta and
there were no significant differences in the t-test
among the three strains. The frequency being so
low, approximately 20.0% in the intraspecific
crosses of D. pallidifrons, no further analysis was
performed.

In Ex-1, the difference in the frequency of the
reaction mass formation was highly significant in
all crosses as compared with the control group.
Reduction ranged from 60.0%, ADM-strain, to
33.0%, SEZ-strain, in the intraspecific crosses of
D. nasuta as compared to the corresponding
control figures of 86.0-90.0%. No complete

Phenoloxidase in Drosophila 81

TaBLE1. Percent of reaction mass formation in temperature shift experiments
Cross type % of formation of reaction mass
Strain Number
Species of Control Ex-1 Ex-2
Female Male females
—— *** —— rt *** —4
Drosophila nasuta ADM-1 ADM-1 50 90.0 36.0 70.0
r * KK 1 ns: 1
Xx MYS-23 MYS-23 50 90.0 52.0 70.0
r * 1 I ns: 1
D. nasuta SEZ-2 SEZ-2 50 86.0 58.0 68.0
D. pallidifrons PNI-74 PNI-74 50 12.0 nd nd
x
D. pallidifrons PNI-110 PNI-110 50 20.0 nd nd
[> ts 1 fi * 1
D. pallidifrons PNI-74 SEZ-2 50 52.0 24.0 46.0
x
[RE Sa
D. nasuta PNI-110 SEZ-2 50 94.0 46.0 82.0

* p<0.05, ** p<0.01, *** p<0.001.
ns: not significant, nd: not done.

pe
@-®@ Dd. nasuta xX D. nasuta r

rs .
g O--O D. pallidifrons xX D. nasuta
3 3
oO
N
‘d
is)
Lo}
qd
cy)
=)
4
0 2
a
i)
4
oO
0)
)

1

0 6 12 18 24 30 36

Time after incubation

(hours)

Fic. 1. Melanization of reaction mass incubated with dopa in vitro. Score of melanization is expressed after Hiruma
and Riddiford (1984).

inhibition, however, was observed at the low
temperature. As shown in Ex-2, recovery of the
reaction mass formation observed in all crosses
was especially significant in intraspecific crosses of
D. nasuta, ADM.-1 strain, and interspecific crosses
between D. pallidifrons females, PNI-110 strain,

and D. nasuta males, SEZ-2, strain when the
copulated females were re-transferred from 2 to
25°C. The frequencies, however, were lower than
the control values for all crosses.

In vitro melanogenesis was confirmed after
incubation with dopa as shown in Figure 1. Length

82 N. ASADA AND T. FUKUMITSU

of time necessary for melanization was much
longer in intraspecific than in interspecific crosses.
The larger the score increase, the darker the color
of the reaction mass. Length of time of melaniza-
tion reflects that in vivo; the reaction mass is
somewhat rigid, then colored dark only in inters-
pecific crosses as shown in the previous study [6].
The present work demonstrates that stimulation by
copulation itself of some sexual substances ejacu-
lated by the male play an important role in
activation of phenoloxidase in vivo as the trigger in
a copulated female. Additionally, phenoloxidase
plays a key role in the formation of reaction mass.
Firstly, because melanization of the reaction mass
occurred both in vivo [6] and in vitro (Fig. 1)
especially in interspecific crosses. Secondly,
formation of reaction mass was significantly inhi-
bited by a direct injection of phenoloxidase inhibi-
tors, thiourea and sodiumdiethyldithiocarbamate

[7].

Detection of phenoloxidase by SDS-PAGE and
spectrophotometry

Phenoloxidase activity in the extracts from adult
flies were examined by SDS-PAGE. As presented
in Figure 2, the arrow indicates active bands for

Fic. 2. SDS-polyacrylamide gel electrophoresis of phe-
noloxidase in adult fly.

Lane-1: marker protein, 2: marker protein, 3: D. nasuta
(SEZ-2 strain), non-heated, 4: D. nasuta (SEZ-2
strain), heated, 5: D. pallidifrons (PNI-74 strain),
non-heated, 6: D. pallidifrons (PNI-74 strain), he-
ated, 7: D. pallidifrons (PNI-110 strain), non-
heated, 8: D. pallidifrons (PNI-110 strain), heated.

phenoloxidase in lanes-3, 5, and 7 appeared within
six minutes and then darkened during incubation
with dopa at 37°C. Three active bands were
distinguishable and molecular weights of each
protein were estimated with marker proteins; the
major protein was 323,000 daltons, the minor
proteins were 339,000, and 302,000 daltons (data
not shown). In comparison with active bands in D.
nasuta and two strains of D. pallidifrons, banding
patterns were generally similar to one another. No
active bands, however, could be detected when the
extracts were heated at 100°C as shown in lanes 4,
6, and 8.

Insect phenoloxidase exists as a proenzyme that
is processed to become active by protein denatura-
tion and proteases [13]. In Drosophila, there are
some proenzymes including three A components
activated by a natural activator isolated from
pupae [14] and one P activating component by
2-propanol [15]. In D. melanogaster, three pheno-
loxidases have been found; one with a monophe-
noloxidase activity (tyrosinase) and two with dopa
oxidase activities [16-17]. The active bands in the
gels in this study seems to be dopa oxidase rather
than tyrosinase because the extracts obtained from
adults were incubated not in tyrosine but in dopa,
resulting in the three active bands as given in
Figure 2.

Concerning the previous study in Drosophila,
phenoloxidase activity could be detected in the
first-instar-larva, and began to rise, then reached a
maximum level at the time of puparium formation.
Activity decreased at the stage of pupa, then
recovered after eclosion [18-21]. Electrophero-
gram of the present study showed the developmen-
tal profile of phenoloxidase activity, that is, the
highest activity of phenoloxidase was detected at
the late third-instar-larva and not at pupa (data not
shown).

Phenoloxidase activity in a supernatant fraction
from the whole body of female flies is given in
Figure 3. A virgin female was used as the control.
The highest level of phenoloxidase activity was
presented at five minutes after the initiation of
copulation, and no significant differences were
observed in all cross types. Then the phenolox-
idase activity decreased gradually. The data
suggests that the positive correlation between the

Phenoloxidase in Drosophila 83

Absorbancy at 475 nm

0 2 5)

Time after copulation

Fic. 3.

formation of reaction mass in the uterus and
phenoloxidase activation in the supernatant frac-
tion, and that phenoloxidase seems to make a
certain contribution to reaction mass formation.
The reaction mass formation was temperature
dependent; in other words, the frequency was
significantly reduced at a low temperature (2°C),
but it recovered when the copulated females were
placed again under suitable conditions (Table 1).
However, no phenoloxidase activity appeared
when the extracts heated at 100°C showed a lack of
mass formation. The length of time of phenolox-
idase activity was quite similar to that of the
reaction mass formation (Fig. 3).

The phenoloxidase system in insects is quite
important not only in melanin synthesis and the
formation of the cuticle, but also in primitive
defense responses in Drosophila [7]. In D. mela-
nogaster, some mutants which are known to
influence melanogenesis, for example, the rela-
tionship between lozenge (Iz) and various compo-
nents of phenoloxidase complex; that is, the A;
activity is absent in the mutant lozenge-grossy

(minutes)

Phenoloxidase activity in copulated female in vitro.
D. nasutaX D. nasuta

D. nasuta (2, control)

D. pallidifrons x D. nasuta

D. pallidifrons (., control)

(/z®), the Ag activity is reduced in /z [20] and some
electrophoretic variants of phenoloxidase in Jz
mutant have been identified [17]. The phenolox-
idase system, however, might be related to reac-
tion mass formation in a manner that may play an
important role in preventing gene exchange and
interspecific hybridization. Molecular analysis of
the phenoloxidase system is needed for further
understanding of this system.

ACKNOWLEDGMENTS

The authors wish to express their hearty thanks for
invaluable criticism to Prof. Narise, S., Josai University.
They also wish to express many thanks for proof reading
this manuscript to Dr. Kimble, D. M., Okayama
University of Science. This work was partly supported by
The National Institute of Genetics, Japan as a coopera-
tive study.

REFERENCES

1 Patterson, J. T. (1946) A new type of isolation
mechanism in Drosophila. Proc. Natl. Acad. Sci.
U.S.A., 32: 202-208.

10

11

12

84

Patterson, J. T. (1947) The insemination reaction
and its bearing on the problem of speciation in the
mulleri subgroup. Univ. Texas Publ., 4720: 42-77.
Wheeler, M. R. (1947) The insemination reaction in
the intraspecific matings of Drosophila. Univ. Texas
Publ., 4720: 78-115.

Patterson, J. T. and Stone, W. S. (1952) Evolution
in the Genus Drosophila, Macmillan, New York.
Takanashi, E. (1983) Genetic and reproductive
differentiation in Drosophila sulfurigaster. Jpn. J.
Genet., 58: 463-485.

Asada, N. and Kitagawa, O. (1988) Insemination
reaction in the Drosophila nasuta subgroup. Jpn. J.
Genet., 63: 137-148.

Asada, N. and Kitagawa, O. (1988) Formation and
the inhibition of reaction plug in mated Drosophila-
Study of a primitive defense reaction. Dev. Comp.
Immunol., 12: 521-529.

Asada, N. and Kitagawa, O. (1982) Reproductive
isolating mechanisms in the D. nasuta subgroup I.
Jpn. J. Genet., 57: 658-659.

Ephrussi, B. and Beadle, G. W. (1936) A technique

of transplantation for Drosophila. Amer. Natl., 70:
218-225.

Laemmli, U. K. (1970) Cleavage of structural
proteins during the assembly of the head of bacter-
iophage T4. Nature, 227: 680-685.

Hiruma, K. and Riddiford, L. M. (1984) Regulation
of melanization of tobacco hornworm larval cuticle
in vitro. J. Exp. Zool., 230: 393-403.

Ashida, M. (1971) Purification and characterization
of prephenoloxidase from hemolymph of the silk-
worm Bombyx mori. Arch. Biochem. Biophys., 144:
749-762.

13

14

15

16

7

18

19

20

21

N. ASADA AND T. FUKUMITSU

Ashida, M. and Dohke, K. (1980) Activation of
pro-phenoloxidase by the activating enzyme of the
silkworm, Bombyx mori. Insect Biochem., 10: 37—
47.

Mitchell, H. K. and Weber, U. M. (1965) Drosophi-
la phenol oxidase. Science, 148: 964-965.
Batterham, P. and McKechnie, S. W. (1980) A
phenol oxidase polymorphism in Drosophila mela-
nogaster. Genetika, 54: 121-126.

Rizki, T. M. and Rizki, R. M. (1984) The cellular
defense system of Drosophila melanogaster. In
“Insect Ultrastructure”. Ed. by R. C. King, and H.
Akai, Plenum Publishing Co., New York, pp. 579-
604.

Rizki, T. M., Rizki, R. M. and Bellotti, R. A.
(1985) Genetics of Drosophila phenoloxidase. Mol.
Gen. Genet., 201: 7-13.

Ohnishi, E. (1953) Tyrosinase activity during pupar-
ium formation in Drosophila melanogaster. Jpn. J.
Zool., 11: 69-74.

Geiger, H. R. and Mitchell, H. K. (1966) Salivary
gland function in phenol oxidase production in
Drosophila melanogaster. J. Insect Physiol., 12:
747-754.

Mitchell, H. K. (1966) Phenol oxidase and Dro-
sophila development. J. Insect Physiol., 12: 755-
765.

Peeples, E., Geisler, A., Whitcraft, C. J. and
Oliber, C. P. (1969) Activity of phenol oxidase at
the puparium formation stage in development of
nineteen /ozenge mutant of Drosophila melanogas-
ter. Biochem. Genet., 3: 563-569.

ZOOLOGICAL SCIENCE 7: 85-91 (1990)

Embryo Transfer and Pregnancy Rate in the Golden Hamster
(Mesocricetus auratus)

S.J. Jarosz! and W. R. DuKELow”

‘Akademia Rolnicza im. Eugena Kellqtaja, W. Krakowie, Zaklad Hodnell
Zwierzat Futerkowych, Al. Mickiewicza 24/28, 30-059 Krakow,
Poland, and 7Endocrine Research Center, Michigan State
University, East Lansing, MI 48824, USA

ABSTRACT—Two series of experiments were carried out to determine the reason for the low
pregnancy rate after embryo transfer in the hamster. Two culture media, TC-199 and TALP plus 20%
FCS, were tested for flushing and transfer of embryos. The 4- to 8-cell embryos were recovered from
mated females (donors) and transferred to female recipients synchronized by the same hormonal
regimen. The pregnancy rate after the use of TC-199 and TALP plus 20% FCS were similar, 36.4% and
39.1% respectively, and compared with control females (80.0%). Only two of eight pregnant females in
the TC-199 group and three of nine pregnant females in the TALP group delivered live young. The
second series of experiments was carried out on three groups of females. Those in the first group were
subjected to embryo transfer (ten, 4- to 8-cell embryos into right uterine horn). Those in the second
group were mated and received only medium in the right uterine horn. In the third group, also mated,
sham injections were performed in the right uterine horn. All females were autopsied on Day 14 of
gestation. The pregnancy rate in females of Groups I, II, and III were 50.0, 59.1, and 60.0%
respectively, and the percent of pregnancies with at least one normal developed fetus from the right
uterine horn of these three groups were 20.0, 46.2, and 58.3% respectively. In females of Groups IT and
III, the level of pregnancies in the right uterine horn were 30.8 and 17.7 percentage units less than in the
left uterine horn. The number of normally developed fetuses in pregnant females of Group I was
24.4%, similar to that in the first experiment. The numbers of all recovered fetuses and of normally
developed fetuses in the right uterine horn of females of Group II were significantly lower (25.8 and 19.5
respectively) than in the left uterine horn (74.2 and 56.6% respectively). A similar tendency was found
in females of Group III. Of 174 recovered fetuses from both uteri, 30.5% were from the right uterine
horn and 69.5% from the left uterine horn and the levels of normally developed fetuses were 14.9 and
55.2%, respectively.

These results shown that the main reason for a decreased pregnancy rate after embryo transfer in the
hamster is due to a trauma of the endometrium of the uterus and medium introduced into uteri, which
may induce secretion of prostaglandins.

© 1990 Zoological Society of Japan

INTRODUCTION

The technique of embryo transfer has been
increasingly used in domestic animals to enhance
genetical performance and, improve productivity
[1-5]. In humans, this procedure is used in con-
junction with in vitro fertilization to overcome
infertility due to impaired tubal function [6-10].
Embryo transfer has been widely used in labora-
tory animals for fundamental studies of fertiliza-

Accepted April 13, 1989
Received December 22, 1988

tion, blastomere separation, or nuclear trans-
plantation. Primarily these types of experiments
have been performed in mice [11-17]. Consider-
ably fewer experiments on embryo transfer have
been done in hamsters, a species which seems to be
especially suitable for studies on reproduction be-
cause of their early maturity, stable cycle length,
high prolificacy, and short gestation. Blaha [18]
reported that 49.2% of 6- to 8-cell embryos from
young hamsters developed to term when transfer-
red into young recipients, but that only 8.3% of
the embryos of the same developmental stage
developed into fetuses when transferred into old

86 S.J. JARoSZ AND W. R. DUKELOw

recipients. Orsini and Psychoyos [19] transferred
hamster blastocysts into ovariectomized and prog-
esterone treated females, and found that some of
the embryos could develop into live fetuses (7-12
days of gestation). Sato and Yanagimachi [20]
transferred 1- to 2-cell embryos into the oviduct or
4- to 8-cell embryos into the uterus (using TC-199
medium) and found that 50 to 100% of the females
receiving 4- to 8-cell embryos did become pregnant
while none of the females receiving 1- to 2-cell
embryos did so. Ridha [21] transferred embryos at
different stages of development (1-cell to 8-cell)
into the oviduct or uterus of hamsters after in-
duced and natural ovulation and had 53-62%
implantation rates in superovulated females. He
was not able to produce full term live young. Fan
et al [22] reported that the culture medium had a
strong influence on the success of the embryo
transfer. Increased survival rates are obtained in
hamster embryo culture with the use of tissue
culture medium 199 (TC-199) or Tyrode’s solution
supplemented with fetal calf serum (FCS) or pre-
gnant hamster serum (PHS). According to Bavis-
ter et al. [23], the modified Tyrode’s solution
designated TALP (tyrodes-albumin-lactate-
pyruvate) is the most suitable for embryo culture
and transfer in hamster.

The objective of the present experiments was to
determine whether the type of medium or other
factors influence the pregnancy outcome results
when 4- to 12-cell embryos are transferred to the
hamster uterus.

MATERIALS AND METHODS

Mature golden hamsters (Mesocricetus auratus)
8 to 15 weeks old were maintained on a 12 h light:
12h dark cycle. The estrous of each female was
determined according to Orsini [24]. The day of
vaginal discharge was designated as Day 1 of the
estrus cycle. Hamster females (prospective ovum
donors and recipients) were superovulated with an
i.p. injection of 30 I.U. of PMSG (Serotropin,
Teizo Ltd, Tokyo) on Day 1 of estrus at 0900 h
followed by an i.p. injection of 30 I.U. hCG
(Sigma Chemical Co., St. Louis, MO) at 1400 h on
Day 4 (77 hr after PMS). Females to be used as
donors were mated with fertile males and female

recipients with vasectomized males 7-8 hr after
the hCG injection. On Day 4 of pregnancy the
donor females were killed by cervical dislocation
and then the uteri and oviducts were excised from
the donors and, during the first series of experi-
ments, were flushed with 0.8 ml of TC-199 or
TALP with 20% FCS. Recovered embryos were
rinsed once and stored no longer than 35 min at
37°C before transfer to the recipient. FCS was
from GIBCO Co., (Grand Island, NY) and TALP
was prepared by the method of Bavister et al. [23].

Recipient females were anesthetized with 0.12-
0.15 ml sodium pentobarbital (60 mg/ml) i.p. and
their uteri exposed through a dorsolateral incision.
During the first exploratory series of experiments,
eight to twenty 4- to 12-cell embryo were drawn
into a micropipette with 20 ul of TC-199 or TALP
and injected into the uterine lumen near the
utero-tubal junction. Control animals were either
a) mated to fertile males, or b) mated and under-
went a sham embryo transfer wtih injection of 20
ul of medium only. Control animals received the
same hormonal regime as the experimental
females. All females were allowed to deliver term
fetuses, or if no pregnancy was evident by 17 days
after mating, the females were killed by cervical
dislocation and the uteri were recovered for ex-
amination.

During the first series of experiments, it was
found that after sham transfer (medium only) into
the uterus no pregnancies resulted in the injected
horn. Therefore, a second series of experiments
were performed on females divided among three
groups. The same hormonal regime as described
above was used. The right uterine horn of females
in Group I (embryo transfer group) received ten 4-
to 8-cell embryos in 20 ul of TALP. In Group II
females (medium only transfer group), the right
uterine horn received 20 ul of TALP alone and in
Group III females (sham injection group) the right
uterine horn received only the tip of micropipette.
On Day 14 after hCG and mating, the females
were anesthetized, killed by cervical dislocation,
and examined for pregnancy. The fetuses were
counted and measured for crown-rump (C-R)
length, weight, and stage of development.

Student’s t-test was used for statistical evalua-
tion.

Embryo Transfer in Hamsters 87

RESULTS

The results of the first series of experiments
demonstrated that the pregnancy rate was 36.4%
(8/22) with the use of TC-199 and 39.1% (9/23)
with TALP plus 20% FCS. A significantly higher
pregnancy rate was found in naturally mated,
control females (80.0%, 16/20) which received the
same regimen for superovulation (P<0.05). Only
two of eight pregnant females with TC-199 and
three of nine pregnant females with TALP plus
20% FCS delivered live young. In control females,
the delivery rate was 93.8% and was significantly
higher than in experimental females (P<0.05).
The percent of transferred embryos that developed

TaBLE 1.
horns of hamsters

to fetuses of the two groups was 54.3 (51/94) and
36.9% (55/149) respectively. The numbers of
embryos transferred to each recipient varied from
8 to 20. As mentioned earlier, no implantations
occurred in uterine horns of naturally mated
females which received an injection of medium
alone.

The results of the second series of experiments
to determine whether the trauma of transfer had a
role in the restriction of pregnancy rate are shown
in Tables 1 and 2. Only two of ten pregnant Group
I females (20.0%) were pregnant 14 days after
embryo transfer. In the females with a medium
only transfer (Group II) or a sham injection
(Group III) the total pregnancy rate in the right

Pregnancy rate after embryo transfer, medium only transfer, or sham injection in the right uterine

No. of pregnant females with recovered fetuses on Day 14 of

gestation

Right uterine horn (treated) Left uterine horn (untreated)

Embryos in

with viable fetuses With viable fetuses

Treatment Females both uterine CO on Day 14 ray on Day 14

horns (%) 2 (%) z (%)

Group I 20 — 10 2/10 0 0

(embryo transfer) (recipients) (50.0) (20.0)

Group II 22 13 9/13 6/13 13/13 11/13

(medium only transfer) (mated) (59.1) (69.2) (46.2) (100.0) (84.6)

Group II 20 12 10/12 7/12 12/12 10/12

(sham injection) (mated) (60.0) (83.3) (58.3) (100.0) (83.3)

All females were sacrificed on Day 14 of gestation for examination of the fetuses.

TABLE 2. Number of recovered fetuses after embryo transfer, medium only transfer, and sham injection in
the right uterine horns of hamsters
ee ee ee ee ee
No. of fetuses recovered from females on Day 14

In the right uterine horn In the left uterine horn

Treatment In both Total With viable fetuses Total With viable fetuses
uterine horns (%) (%)
Group I (embryo transfer) 44 44 11 — —
from 10 pregnant (100.0) (25.0)
females
Group II (medium only transfer) 159 41 31 118 90
from 13 pregnant (25.8)* (19.5)* (74.2) (56.6)
females
Group III (sham injection) 174 53 26 121 96
from 12 pregnant (30.5)* (14.9)* (69.5) (55.2)

females
ee ee a ee PR Ae PMS SS STE me we MOE Teel Pe See ee Cee oe

* Significantly different from control levels (p<0.05).

88 S.J. JARoszZ AND W. R. DuUKELOW

TABLE 3.

Crown-rump length and weights of viable hamster embryos on Day 14 of gestation

Right uterine horn

Left uterine horn

Treatment

(right uterine horn) Crown-rump length
(mm
Group I (embryo transfer) 14.3+2.8'
Group II (medium only transfer) 18.1+1.4
Group III (sham injection) 16.2+1.8

Weight Crown-rump length Weight
(mg) (mm (mg)
589 + 246 — —
816+178 18.6+2.1 679 +243
655 + 198 15.8+2.9 7344177

' Values expressed as mean+(S.D.)

TABLE 4.
14 of gestation

Right uterine horn (treated)
No. of fetuses in relation
to no. of corpora lutea

Ovarian weight and relationship of the number of fetuses to the number of corpora lutea on Day

Left uterine horn (control)
No. of fetuses in relation
to no. of corpora lutea

Treatment Ovarian weight Total Viable fetuses Ovarian weight Total (%) Viable fetuses
(mg) (%) (%) (mg) (%)
Group II 58.1+0.0 4.8/24.3 3.8/24.8 62.5+0.0 10.4/28.0 9.6/28.0
(medium only transfer) (19.4)* (15.5) (37.1) (34.3)
Group III 41.3+0.0 4.1/22.0 2.7/22.0 61.5+0.0 11.4/24.0 9.7/24.0
(sham injection) (18.8) (12.5) (47.4) (40.6)

Group I (embryo transfer) not evaluated.

* Significantly different from control levels (p<0.05).

uterine horn in comparison to the left was de-
creased by 30.8 and 16.8% respectively. The
viable pregnancy rate (embroys judged capable of
being delivered live two days later) was reduced by
38.4 and 25.0 percentage units in Groups II and III
respectively.

Forty-four of the fetuses in 10 pregnant females
were identified after embryo-transfer (Group I)
but only 25.0% of these were classified as viable
(Table 2). This result was similar to that observed
in the first series of experiments.

The crown-rump length and weight of normally
developed embryos at 14 days of gestation are
shown on Table 3. No significant differences were
found between groups.

Ovarian weight, the number of corpora lutea,
and the relation of the number of fetuses to the
number of corpora lutea are shown in Table 4. No
Statistically significant differences were found be-
tween groups for ovarian weight or numbers of
corpora lutea as expected. In both Group II and
III, the relationship between the number of all
fetuses to the number of corpora lutea were, in the

right side, 19.4 and 18.8% respectively and, in the
left side, 37.1 and 47.4% respectively, a statistical-
ly significant difference reflecting the decreased
embryo survival in the treated right side.

DISCUSSION

In the first series of experiments, attention was
directed to two culture media, which according to
Bavister et al. [23] play an important role in the
culture of hamster embryos and their successful
development after transfer to pseudopregnant
females. Work reported by Fan et al. [22] has
shown that the best results in embryo development
were with media TC-199 or Tyrode’s solution
supplemented with 20-30% of PHS or FCS.

The present results confirm these findings. Pre-
gnancy rates were lower in our experiments than
those reported by Blaha [18] who reported 49.2%
developed fetuses after transfer of 6- to 8-cell
embryos into the uterus, and were also lower than
those of Sato and Yanagimachi [20] who reported
a 66% pregnancy rate with 48% live fetuses after

Embryo Transfer in Hamsters 89

transfer of 4- to 8-cell embryos. Ridha [21] re-
ported a 58-59% rate of implantation after un-
ilateral transfer of ten 4- to 8-cell embryos per
female. These values are similar to our pregnancy
rate (50%) in the second series of experiments, but
Ridha was not able to produce live offspring.

Significantly lower pregnancy rates and number
of live fetuses were recovered after embryo trans-
fer than in control females. This suggested that
factors other than the type of medium, such as
trauma to the uterus during the transfer proce-
dure, may play an important role in the restriction
of the implantation rate and further development
of transferred embryos.

The results of the second series of experiments
with sham transfer of medium or the introduction
of only the tip of micropipette into the uterine
horn show very clearly that these procedures
adversely affect the pregnancy rate and the num-
ber of implantations.

The pregnancy rates in females with the sham
transfer of medium (Group II) and in females with
sham micropipette introduction (Group III) are
similar- 59.1 and 60.0% respectively (Table 1), but
about 20% lower than in control females in the
first series of experiments. Comparing the number
of pregnancies in the right (treated) horn and the
left (untreated) uterine horn to the number of
pregnant females, it can be seen that pregnancy
rates in term pregnancies with normal and de-
generated fetuses as well as those with at least one
normal fetus were significantly lower in the treated
uterine horn. Significant differences between the
treated and untreated uterine horns are seen in the
numbers of recovered fetuses on Day 14 of gesta-
tion (Table 2). The total number of fetuses (nor-
mal and degenerated) in the treated horns were
less by about 48 percentage units in females with
medium only transfer and by about 39 percentage
units in females with sham injection transfer. Simi-
lar significant differences were found in the num-
ber of normally developed fetuses at Day 14.
Thses results demonstrate that the main factor
restricting the implantation and normal develop-
ment of transferred embryos is trauma of the
uterus.

The physiological mechanism of this trauma
effect is not known. Pharriss et al. [25] noted that

trauma of the endometrium causes release of pros-
taglandins which, according to Horton et al. [26]
and Kirton et al. [27], stimulate uterine contractil-
ity in rabbits and primates, and can induce abor-
tion [28]. Spilman et al. [29] studied the effects of
two isomers of PGF, (19(R)-19(S)-OH) and
found that they were considerably less effective
than PGF, in stimulating motility of the rabbit and
monkey reproductive tracts, and that there are
different species sensitivities for prostaglandins.
Gutknecht et al. [30] found that PGF, at a dose
level of 2 mg/day over any consecutive three-day
period from Day 4 after coitus were 100% effective
in preventing or terminating pregnancies in the rat.
The same author [31], reported that PGF, admi-
nistered subcutaneously at a dose of 0.1 mg/day on
Days 5 through 7 post-coitus in the hamster lo-
wered both plasma and ovarian progesterone
levels and terminated pregnancy in all animals. He
also reported histological evidence of luteal degen-
eration on Days 6 and 7 post-coitus in treated
animals, but that exogenous progesterone main-
tained pregnancy in PGF, treated females.
According to Pharriss et al. [25] the hamster is the
most sensitive species to prostaglandin-induced
luteolysis of any examined animals. Thomas et al.
[32], taking blood and tissue samples from hams-
ters during estrous cycle and for the first four days
of pregnancy, found that maximum steroid and
prostaglandin concentrations occurred around
ovulation and after that declined to the lowest
level on Days 3 and 4 of pregnancy. They reported
that a close relationship exists between steroids
and prostaglandins. These findings suggest that
the trauma which probably releases prostaglandins
during embryo transfer, is the main factor decreas-
ing implantation and embryo development.

ACKNOWLEDGMENTS

These studies were supported by REED funds from
the state of Michigan and grants from the National
Institutes of Health. Appreciation is expressed to Mr. W.
E. Roudebush for assistance in the procedures.

REFERENCES

1 Williams, T. J., Elsden, R. P. and Seidel, G. E. Jr.
(1983) Bisecting bovine embryos: Method applica-

90 S.J. Jarosz AND W. R. DUKELOW

tions and success rates. Proc. Ann. Conf. on A. I.
and Embryo Transfer in Beef Cattle, Denver, Co.
pp. 45-51.

Rorie, R. W., Voelkel, S. A., McFarland, C. W.,
Southern, L. L. and Godke, R. A. (1984) Another
split-embryo transplant in pigs. Louisiana Agric.,
28: 3.

Allen, W. R. and Pashen, R. L. (1984) Production
of monozygotic (identical) horse twins by embryo
micromanipulation. J. Reprod. Fertil., 71: 607-613.
Willadsen, S. M. and Godke, R. A. (1984) A simple
procedure for the production of identical sheep
twins. Vet. Rec., 114: 240-243.

Tsunoda, Y., Tokunaga, T., Sugie, T. and Katsuma-
ta, M. (1985) Production of monozygotic twins
following the transfer of bisected embryos in the
goats. Theriogenology, 24: 337-342.

Mettler, L., Michelman, H. W., Riedel, H. H.,
Grillo, M., Weisner, D. and Semm, K. (1984) In
vitro fertilization and embryo replacement at the
Department of Obstetrics and Gynecology, Uni-
versity of Kiel, FRG. J. In Vitro Fertil. and Embryo
Transfer, 1: 250-262.

Mohr, L. R., Trounson, A. and Freeman, L. (1985)
Deep-freezing and transfer of human embryos. J. In
Vitro Fertil. and Embryo Transfer, 2: 1-10.
Testart, J., Lassalee, B., Belaisch-Allart, J., Haz-
out, A., Forman, R., Rainbor, J. D. and Frydman,
R. (1986) High pregnancy rate after early human
embryo freezing. Fertil. Steril., 46: 268-272.
Reinthalter, A., Riss, P., Deutinger, J., Muller-Tyl,
E., Fischl, F. and Janish, H. (1986) Factors influenc-
ing successful in vitro fertilization and embryo trans-
fer: A matched pair study. Fertil. Steril., 46: 511-
513.

Poindexter, A. N., Thomson, D. J., Gibbons, W.
E., Findley, W. E., Dodson, M. G. and Young, R.
L. (1986) Residual embryos in failed embryo trans-
fer. Fertil. Steril., 46: 262-267.

Tarkowski, A. K. and Wroblewska, J. (1967) De-
velopment of blastomeres of mouse eggs isolated at
the 4- and 8-cell stage. J. Embryol. Exp. Morphol.,
18: 155-180.

Hoppe, P. C. and Illmensee, K. (1977) Microsurgi-
cally produced homozygousdiploid uniparental
mice. Proc. Natl. Acad. Sci. USA, 74: 5657-5661.
Illmensee, K. (1982) Experimental genetics of the
mammalian embryo. J. Cellular Physiol. Suppl., 2:
117-129.

O’Brien, M. J., Crister, E. S. and First, N. L. (1984)
Developmental potential of isolated blastomeres
from early murine embryos. Theriogenology, 22:
601-607.

Nakagawa, A., Takanashi, Y. and Kanagawa, H.
(1985) Studies on developmental potential of
bisected mouse embryos in vitro and in vivo. Jpn. J.

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

Vet. Res., 33: 121-133.

Robl, J. M., Gilligan, B., Critser, E. S. and First, N.
L. (1986) Nuclear transplantation in mouse
embryos: Assessment of recipient cell stage. Biol.
Reprod., 34: 733-739.

Papaioannou, V. E. and Ebert, K. M. (1986) De-
velopment of fertilized embryos transferred to ovi-
ducts of immature mice. J. Reprod. Fertil., 76: 603-
608.

Blaha, G. C. (1964) Effect of age of the donor and
recipient on the development of transferred hamster
ova. Anat. Rec., 150: 413-416.

Orsini, M. W. and Psychoyos, A. (1965) Trans-
plantation of blastocysts transferred into progester-
one treated virgin hamsters previously ovariecto-
mized. J. Reprod. Fertil., 10: 300-301.

Sato, A. and Yanagimachi, R. (1972) Transplanta-
tion of preimplantation hamster embryos. J. Re-
prod. Fertil., 30: 329-332.

Ridha, M. T. (1983) Embryo production, freezing
and transfer in the golden hamster (Mesocricetus
auratus). Ph. D. Dissertation. Michigan State Uni-
versity, 95 pp.

Fan, B., Ridha, M. T., Demayo, F. J. and Dukelow,
W. R. (1986) Culture and transplantation of preim-
plantation hamster embryos. J. Agric. Sci., Suppl.,
2: 69-77.

Bavister, B. D., Leibfried, M. L. and Lieberman,
G. (1983) Development of preimplantation embryos
of the golden hamster in defined culture medium.
Biol. Reprod., 28: 235-247.

Orsini, M. W. (1961) The external vaginal phe-
nomena characterizing the stages of the estrous
cycle, pregnancy, pseudopregnancy, lactation, and
the anestrous hamster, Mesocricetus auratus Water-
house. Proc. Animal Care Panel, 11: 193-206.
Pharriss, B. B., Tillson, S. A. and Erickson, R. R.
(1972) Prostaglandins in luteal function. Recent
Progress in Hormone Research, 28: 51-80.
Horton, E. W., Main, I. H. M. and Thompson, C. J.
(1965) Effect of prostaglandins on the oviduct,
studied in rabbits and ewes. J. Physiol., 180: 514-
528.

Kirton, K. T. and Farbes, A. D. (1972) Activity of
15(S) 15-methyl prostaglandin E, and F, as stimu-
lants of uterine contractility. Prostaglandins, 1: 319-
323.

Fuchs, F., Prieto, M. and Marcus, S. (1971) Effect
of prostaglandins on uterine activity in pregnant
Rhesus monkeys. Ann. New York Acad. Sci., 180:
531-536.

Spilman, C. H., Bergstrom, K. K. and Forbes, A.
D. (1977) Effect of 19-hydroxy-prostaglandins on
oviductal and uterine motility. Prostaglandins, 13:
795-805.

Gutknecht, G. D., Cornette, J. C. and Pharris, B.

31

Embryo Transfer in Hamsters 91

B. (1969) Antifertility properties of prostaglandin
F,. Biol. Reprod., 1: 367-371.

Gutknecht, G. D., Wyngarden, L. J. and Pharris, B.
B. (1971) The effect of prostaglandin F, on ovarian
and plasma progesterone levels in the pregnant
hamster. Proc. Soc. Exp. Biol. Med., 136: 1151-
1157.

32 Thomas, C. M. G., Bastiaans, L. A. and Rolland,
R. (1982) Concentrations of conjugated oestradiol
and progesterone in blood plasma and prostaglan-
dins F-2 and E-2 in oviducts of hamsters during

oestrous cycle and in early pregnancy. J. Reprod.
Fertil., 66: 469-474.

QM Qsheedt 1€f
moro FANT fA

1 neta 1 Cen

pitpvy ¢ ‘Daa. Doe girilay,

. ee oe a
ib eb sig ims
ir = ie yemind * vier hy’
z a ‘
ar ain wna 4 WR |
| ind
’ ve 2 Gay
a) vi ' a peer af
_ ~ = 2» » oe: -'
7 rir
= -
; wa 7.7 ‘
tes i) ow
¢ 4 “is jae
j ui f
f vi , t t wt ee
tot ee w 4s
a : 4
isk
v i
—
"ity
aM
4 "
"-OtCi
is li Oeil

ZOOLOGICAL SCIENCE 7: 93-96 (1990)

© 1990 Zoological Society of Japan

Changes in Thyroid Hormone Concentrations during Early
Development and Metamorphosis of the Flounder,
Paralichthys olivaceus

MasatTomo TaGAwa, SatTosHi Miwa!, Yasuo Inui!,

EVELYN GRACE DE JESUS and TETSUYA HIRANO

Ocean Research Institute, University of Tokyo, Tokyo 164, and
‘Inland Station, National Research Institute of Aquaculture,
Tamaki, Mie 519-04, Japan

ABSTRACT—Changes in the whole body concentrations of thyroid hormones were examined during
early development and metamorphosis of the flounder (Paralichthys olivaceus). Thyroxine (T4) as well
as triiodothyronine (T3) were detected in eggs just after fertilization; concentration of T3 was 6-7 ng/g
and that of T4 was about 1 ng/g. Concentration of T3 declined gradually until hatching, decreased
sharply from 5 ng/g to 0.5 ng/g within one day after hatching and became non-detectable (below 0.1
ng/g) thereafter. T4 concentration did not show marked changes until 10 days after fertilization. Until
the climax of metamorphosis, T3 was undetectable and T4 concentration was less than 1 ng/g. During
the climax, T4 concentration increased markedly to 10-15 ng/g, and T3 concentration increased to a
detectable level (1-1.5 ng/g). After the completion of metamorphosis, T4 concentration decreased to
about a half of the peak level. T3 concentration also decreased slightly. These observations were
discussed in relation to the role of T4 and T3 in early development and metamorphosis of the flounder.

INTRODUCTION

Thyroid hormones are known to play an impor-
tant role in amphibian metamorphosis [1]. Activa-
tion of thyroid gland was also observed during
metamorphosis in the conger eel [2] and flounder
[3], and treatments with thyroid hormones induced
metamorphosis in these species [4-6]. Recently, a
radioimmunoassay procedure was applied to esti-
mate whole body concentrations of thyroxine (T4)
during the metamorphosis of flounder larvae [7, 8],
and a significant surge was observed at the climax
of metamorphosis. A similar procedure was ap-
plied to examine the changes in whole body con-
centration of triiodothyronine (T3) during early
development of chum salmon [9]. Although T3
has been shown to be more effective than T4 in
inducing flounder metamorphosis [6], there is no
report on the changes in T3 concentrations during
the flounder metamorphosis. In this paper, we

Accepted March 24, 1989
Received February 15, 1989

report the changes in whole body concentrations of
T3 as well as T4 during early development and
metamorphosis of the flounder.

MATERIALS AND METHODS

Fertilized eggs of the flounder ( Paralichtys oli-
vaceus) were obtained from spawning aquarium
containing several females and males, and incu-
bated in seawater at 15°C. They hatched 4 days
after fertilization. About 600 eggs or larvae were
sampled daily until 8 days and on 10th day after
fertilization. Larvae were offered rotifers starting
from 8 days after fertilization. Another batch of
larvae was obtained from a commercial source and
reared in a 5001 aquarium. They were fed rotifers
and brine shrimp nauplii. Twenty to 40 larvae
were sampled at intervals starting from 18 days
after hatching until 57 days after hatching. They
were stored at —80°C until analyses. Metamor-
phic stages were identified by the eye migration
stage and the length of the 2nd fin ray [7]. Samples
of eggs and larvae until 10 days after fertilization

94 M. Tacawa, S. Miwa et al.

were processed as previously described for the
distribution of thyroid hormones in developing
chum salmon [9]. The hormone concentrations of
metamorphosing larvae were measured from one
homogenate made from at least 12 individuals as
follows; frozen samples were homogenized in 4 ml
of methanol, and the homogenates were divided
into halves, one for T4 determination and another
for T3 determination, using the half volume of
solutions of agents as used in previous studies [9,
10]. The least detectable concentrations of thyroid
hormones were 0.1 ng/g for eggs and larvae until
10 days after fertilization and 0.4 ng/g for meta-
morphosing larvae.

RESULTS

Figure 1 shows the changes in T4 and T3 concen-
trations until 10 days after fertilization. Both T4
and T3 were detected in eggs just after fertiliza-
tion; the concentration of T3 was 6.6 ng/g and that

7

0)

Ol

£{

Ww

hatching

Hormone concentration (ng/g)

of T4 was 0.8ng/g. T3 and T4 concentration
decreased gradually toward hatching. After hatch-
ing, T3 concentration decreased sharply to about
1/10 of the level before hatching within one day,
and became non-detectable thereafter. On the
other hand, T4 concentration did not show such a
marked change at the time of hatching, but a low
level less than 1 ng/g was maintained thereafter.

As shown in Figure 2, T4 concentration during
the premetamorphosis was less than 1 ng/g, and
tended to increase during the prometamorphosis.
T3 concentration was still non-detectable (less
than 0.4 ng/g) during these periods. During the
climax of metamorphosis, when the dorsal fins are
being resorbed, T4 concentration increased
markedly to 10-15 ng/g. T3 became detectable
during the climax, but the level (1-1.5 ng/g) was
still lower than the T4 level. During the postclimax
of the metamorphosis, T4 concentration decreased
to a half of the peak level.

feeding

oT SS

1 ailenee e
0)
) 2 4

6 8 10

Days after fertilization

Fic. 1. Changes in T4 and T3 concentrations until 10 days after fertilization in the flounder. Vertical bars represent
standard errors of the means of 3 pooled samples. In some cases, the variation was extremely small and within the
size of the circle. N: non-detectable (below 0.1 ng/g).

T4 and T3 in Metamorphosing Flounder 95

15
=
cal
2
240
:
ne)
©
e
Gd)
(6)
Cc
[e)
(6)
2 5 pre- pro-
2 metamorphosis | metamorphosis |
5
ae
0)

20 30

ostclimax

climax p

T3

ni ae ee aI a
a Tiina tSasir ta Ul SiCliia chaiediaa Gas... TL ee

40 50
Days after hatching

Fic. 2. Changes in T4 and T3 concentrations during metamorphosis of the flounder. Each point represents the
average of the duplicate determination of one pooled sample. N: non-detectable (below 0.4 ng/g).

DISCUSSION

Significant quantities of both T4 and T3 have
been detected in unfertilized or fertilized eggs of
chum salmon (T4; 5-15 ng/g, T3; 4-9 ng/g) [9, 10],
coho salmon (T4; 15-30 ng/g, T3; 20 ng/g) [11, 12],
chinook salmon (T4; 9 ng/g, T3; 22 ng/g) [12] and
striped bass (14; 4 ng/g, T3; 5 ng/g) [13]. In the
flounder eggs, T3 concentration (6.6 ng/g) was
higher than T4 (0.8 ng/g) as in chinook salmon
[12]. The relative quantities of T4 and T3 in fish
eggs may be related to the mechanisms of utilizing
thyroid hormones, such as the onset of S’-
deiodinase activity, during early developmental
stages.

During the embryonic development of the floun-
der, whole body concentrations of both T4 and T3
decreased gradually toward hatching. After hatch-
ing, however T3 concentration decreased sharply
to about 1/10 within 1 day, whereas no marked
change was seen in T4 concentration. According
to Fukuhara [14], a half of the yolk was resorbed
within one day after hatching in the flounder larvae

kept at 14.2°C, and remnant yolk was linearly
resorbed to completion in succeeding 3 days.
Therefore, the rate of decrease in T3 concentra-
tion in the flounder larvae just after hatching
seems to be greater than that of the yolk resorp-
tion, indicating some selective absorption mechan-
isms of T3 from the yolk. Selective absorption of
thyroid hormones during yolk resorption has also
been suggested in salmonid fishes and striped bass
[9-13].

Although there is no information on the onset of
thyroid hormone receptors during early develop-
ment of fishes, the higher concentrations of T3
than T4 and the sharp decrease in T3 concentra-
tion just after hatching seem to indicate a primary
role of T3 in early development of the flounder
embryo and larva. According to Eales [15], T3
seems to be the effective thyroid hormone not only
in juvenile and adult fishes but also even in early
developmental stages.

During the climax of the metamorphosis, T4
concentration increased markedly to 10-15 ng/g,
consistent with previous observations [7, 8]. T3

6 M. TaGawa, S. Miwa et al.

concentration was non-detectable (below 0.4 ng/g)
during pre- and prometamorphosis, increased to
1-1.5ng/g at the climax when T4 surge was
observed, and the same level was maintained
during postclimax. According to Miwa and Inui
[6], T3 was several times more effective than T4 in
inducing metamorphosis in the flounder. There-
fore, a small increase in T3 concentration might be
enough to stimulate the metamorphosis, although
it is still possible that T4, but not T3, plays
important roles in the flounder metamorphosis.
Distribution and development of the thyroid hor-
mone receptors are to be studied in the “metamor-
phosing” organs or tissues, not only in flounders
but also in fishes in general.

ACKNOWLEDGMENTS

This study was supported in part by grants-in-aid from
the Ministry of Agriculture, Forestry and Fisheries
(BMP-88-II-2-3) to Y. I. and T. H. and also from the
Ministry of Education, Science and Culture, Japan
(61300014 and 62480022) to T.H.

REFERENCES

1 White, B. A. and Nicoll, C. S. (1982) Hormonal
control of amphibian metamorphosis. In “Meta-
morphosis”. Ed. by L. I. Gilbert and E. Frieden,
Plenum Press, New York, pp. 363-396.

2 Kubota, S. (1961) Studies on the ecology, growth
and metamorphosis in conger eel, Conger myriaster
(Brevoort). J. Fac. Fish. Prefectural Univ. of Mie,
5: 190-329. (In Japanese)

3 Miwa, S. and Inui, Y. (1987) Histological changes in
the pituitary-thyroid axis during spontaneous and
artificially-induced metamorphosis of larvae of the
flounder Paralichthys olivaceus. Cell. Tissue Res.,
249: 117-123.

4 Kitajima, C., Sato, T. and Kawanishi, M. (1967) On
the effect of thyroxine to promote the metamorph-
osis of a conger eel—preliminary report. Bull.

10

11

12

13

14

15

Japan. Soc. Sci. Fish., 33: 919-922. (In Japanese
with English summary)

Inui, Y. and Miwa, S. (1985) Thyroid hormone
induces metamorphosis of flounder larvae. Gen.
Comp. Endocrinol., 60: 450-454.

Miwa, S. and Inui, Y. (1987) Effects of various
doses of thyroxine and triiodothyronine on the
metamorphosis of flounder ( Paralichthys olivaceus).
Gen. Comp. Endocrinol., 67: 356-363.

Miwa, S., Tagawa, M., Inui, Y. and Hirano, T.
(1988) Thyroxine surge in metamorphosing flounder
larvae. Gen. Comp. Endocrinol., 70: 158-163.
Tanangonan, J. B., Tagawa, M., Tanaka, M. and
Hirano, T. (1989) Changes in tissue thyroxine level
of metamorphosing Japanese flounder Paralichthys
olivaceus reared at different temperatures. Nippon
Suisan Gakkaishi, 55: 485-490.

Tagawa, M. and Hirano, T. (1989) Changes in
tissue and blood concentrations of thyroid hormones
in developing chum salmon. Gen. Comp. Endocri-
nol., 76: 437-443.

Tagawa, M. and Hirano, T. (1987) Presence of
thyroxine in eggs and changes in its content during
early development of chum salmon, Oncorhynchus
keta. Gen. Comp. Endocrinol., 68: 129-135.
Kobuke, L., Specker, J. L. and Bern, H. A. (1987)
Thyroxine content of eggs and larvae of coho sal-
mon, Oncorhynchus kisutch. J. Exp. Zool., 242: 89-
94.

Greenblatt, M., Brown, C. L., Lee, M., Dauder, S.
and Bern, H. A. (1989) Changes in thyroid hor-
mone levels in eggs and larvae and in iodide uptake
by eggs of coho and chinook salmon, Oncorhynchus
kisutch and O. tschawytscha. Fish Physiol.
Biochem., 6: 261-278.

Brown, C. L., Sullivan, C. V., Bern, H. A. and
Dickhoff, W. W. (1987) Occurrence of thyroid
hormones in early developmental stages of teleost
fish. Trans. Am. Fish. Soc. Symp., 2: 144-150.
Fukuhara, O. (1986) Morphological and functional
development of Japanese flounder in early life stage.
Bull. Japan. Soc. Sci. Fish., 52: 81-91.

Eales, J. G. (1985) The peripheral metabolism of
thyroid hormones and regulation of thyroidal status
in poikilotherms. Can. J. Zool., 63: 1217-1231.

ZOOLOGICAL SCIENCE 7: 97-103 (1990)

Effects of Thyroidectomy, Hypophysectomy, Temperature and
Humidity on the Occurrence of Nocturnal Locomotor
Activity in the Toad, Bufo japonicus,
during the Breeding Season

YoKo TASAKI and Susumu IsHII

Department of Biology, School of Education, Waseda University,
Tokyo 169, Japan

ABSTRACT—Toads collected in October 1987 were thyroidectomized or hypophysectomized, and
then released into an outdoor pen. Observations were made once a day between 2030 and 2130 hr for 19
days from February 27 to March 16, 1988. Toads found completely exposed above ground at the time of
the observation were regarded as active individuals. In males thyroidectomy significantly reduced both
the mean number of active individuals each day and the mean number of active days of each individual.
Hypophysectomy significantly reduced both of these parameters in males. In females, the effects of the
operations were not clear, since the number of active individuals was extremely small in all groups.
Significant positive correlations (r=0.58—0.76) were observed between the number of active individuals
and temperature in both sexes, and between the number of active individuals and the humidity of the air
in males. When the temperature and humidity were combined as an independent variable, a highly
significant multiple correlation (r>0.8) in both sexes was observed. The present results suggest that the
pituitary gland plays some role in the migration of the toad toward the pond but thyroid hormone
suppresses the migratory activity, and also that the combination of temperature and humidity is the

© 1990 Zoological Society of Japan

external factor which initiates the migration.

INTRODUCTION

In early spring, adult toads, Bufo japonicus,
come out of hibernation and begin migration to a
particular pond for breeding. This migratory activ-
ity occurs only on warm and humid nights and
hence, not every day. Temperature and humidity
have been proposed as the atmospheric factors
which initiate the migration [1, 2]. A possible
candidate for the endocrine factor initiating the
migration for breeding in amphibians was prolac-
tin, since a number of investigators have shown
that prolactin is the water drive factor in newts and
salamanders [3-8]. However, one of the present
authors and his associates have found that the
plasma prolactin level of toads just before or just
beginning migration toward a breeding pond was
low, and hence, prolactin can not be the migration

Accepted April 27, 1989
Received March 13, 1989

inducing factor in the toad [9, 10]. Tasaki et al.
[11], observing the elevation of plasma thyroxine
and triiodothyronine levels during the breeding
season in Bufo japonicus, suggested the possibility
that thyroid hormone is the migration inducing
factor in the toad. However, they recently found
that thyroxine administration to normal and thy-
roidectomized male toads suppressed their loco-
motor activity when measured in a small chamber,
and thyroidectomy of male toads increased the
activity [12]. These results, though under ex-
perimental conditions, suggest that thyroid hor-
mone also can not be the migration inducing factor
of the toad.

The purpose of the present study, is to confirm
our previous results under conditions which
approximate the natural environment, and also to
determine the role of the pituitary gland in the
migration. The effects of thyroidectomy and
hypophysectomy on the occurrence of nocturnal
locomotor activity were studied in toads kept in an

98 Y. TASAKI AND S. IsHI

outdoor pen during the breeding season.

MATERIALS AND METHODS

Material

Adult male and female toads (Bufo japonicus)
were used. They were captured in the suburbs of
Tokyo in October 1987. The mean body weight
was 145.2 g with a standard deviation of 42.0 g.

Operations

After capture, male and female toads were put
in separate plastic boxes (55X40 X43 cm?) with
loose fitting tops. Wet pieces of plastic sponge
were put in the boxes to maintain humidity. No
feeding took place, since toads abstain from food

Males e—— temperature

ae wan Wa 1 May ine! EC LGC0> humidity

80
60
40
20
Oo
2)
mo)
oO
°
~
—
°
_
®o
a
E
3
4
27 1 5 10 15
February March
Date
Fic. 1.

each experimental day in each group.

during the hibernation and breeding periods.
Toads were kept in the boxes outdoors for about 2
weeks before the operations. Eight males and
eight females were thyroidectomized. Sham-
operations were performed on the same number of
animals of each sex. Hypophysectomy by the oral
approach and corresponding sham-operations
were also performed with the same number of
male and female toads. MS-222 was used for
anesthetization of toads. Each animal was indi-
vidually marked with a small, numbered plastic
sheet which was adhered to its back, and a num-
bered plastic band which was tied around a fore-
limb.

Observation of toads

Fourteen to eighteen days after the thyroidec-
tomy and related sham operation or 8 to 11 days

Females «—— temperature

Oe de ed Oa humidity

Number of toads

es

27 1 5 10 15

February March

Date

Atmospheric factors and the number of male (left) and female (right) toads which appeared above ground on

TX and HX Effects on Toad Locomotion 99

after the hypophysectomy and related sham oper-
ation, toads were released into an outdoor pen (2
x26 m7). As soon as they were released, they
buried themselves under the ground. Observa-
tions were made in the pen once a day between
2030 and 2130 for 19 days from February 27 to
March 16, 1988. When eggs were laid by a female
in a pool (7042 x20 em?) in the pen on March
16, observations were terminated. Toads which
were found completely exposed above ground
were regarded as active individuals, and their
identification numbers were recorded. The
temperature and relative humidity 10 cm from the
ground surface in a corner of the pen were re-
corded every day at the time of the observation.

Statistical methods

The significant over-all difference in the number
of active toads among the groups was determined
by Friedman’s test. The significant difference in
the number of active toads between two groups
was determined by the signed-rank test. The
randomization test was used to compare the two
groups as to the number of days on which toads

Males
n
§ Tx
4
2
fo)
6
Tx-sham
4
2
”
To
0
ww 6
es Hx
04
o
= 2
= (0)
e Hx-sham
4
2
(0)
6
Intact
4
2

fo}

(0) 1 2 3 4 5 6 7 8 9
Number of active days days

showed activity. For these tests, personal compu-
ter programs [13] were employed.

RESULTS

Effects of thyroidectomy and hypophysectomy on
the occurrence of activity

The daily number of male and female toads of
each group which appeared above ground for
locomotor activity is shown in Figure 1. The
temperature and relative humidity are indicated in
the figure. Using the same data, the number of
days on which each toad showed activity was
calculated, and its distribution is shown in Figure
Dp

Males: In males, the locomotor activity was
observed relatively frequently, i.e., 15 out of 19
days. However, the number of active toads fluctu-
ated widely over the course of the days and also
among the groups (Fig. 1). The over-all difference
in the nuinber of active toads each day among the
five groups was tested by using Friedman’s test.
The difference was highly significant (p<0.01).

Females
n
§ Tx
4
2
(0)
e Tx-sham
4
2
7)
To
9
= «6
= Hx
O04
o
2 2
5 0
= 6
Hx-sham
4
2
(0)
6
Intact

nN»

0 1 2 3 4 5 6 7 8

9
Number of active days days

Fic. 2. The distribution of the number of days in which male (left) and female (right) toads showed activity. The
open column indicates the number of toads which were not recovered after the observation period.

100 Y. TASAKI AND S. IsHm

Then, the two-sided signed-rank test was used for
paired comparisons between two selected groups.
The differences in the number of active males
between the thyroidectomized and its sham-
operation control groups and also between the
hypophysectomized and its sham-operation con-
trol groups were highly significant (p<0.01). The
differences between each of the sham-operation
groups and the intact group were not significant (p
>0.05).

The distribution of the number of days on which
the toad showed activity is indicated in Figure 2.

=
a
®
o

r=0.687 (P<0.05)

LS)
a
e

y=2.02x-9.10

3 .
02°

(e)

ta

w 15

°

= 10

o

E

= ©

z

4 8 12 16
Temperature (°C)

The difference in the average number of active
days between two groups was tested by the two-
sided randomization test. Thyroidectomized males
showed activity more frequently than correspond-
ing sham-operation males (p<0.008), but
hypophysectomized males showed it less frequent-
ly than the corresponding sham-operation males (p
<0.002). The frequency did not differ significantly
(p>0.05) between each of the sham-operation
groups and the intact group.

Females: Compared to males, a fewer number
of females showed locomotor activity (cf. Fig. 1).

Females
16
of r=0.757 (P<0.05) °
3 y=1.11x-6.79
3 12
°
Pa
—_
°
hn
o
a
£
3
za

4 8 12 16
Temperature (°C)

Fic. 3. The single correlation between temperature and the number of active male (left) and female (right) toads.
There was a significant (p<0.05) positive correlation in both males and females.

Males
25 °
e
1) r=0.576 e
xe}
© (P<0.05)
°
aa y=0.27x-8.09
e—
°
he
rT)
a
E
2
rad

20 40 60 80 100
Humidity (%)
Fic. 4.

Females
16
r=0.294 e
SL (P>0.05)
x oj 4
wo e
° y=0.07x-1.42 e
—
ee
° 8
hn
o
Q
£4
=}
za

20 40 60 80 100
Humidity (%)

The single correlation between humidity and the number of active male (left) and female (right) toads. Only

in males did the humidity show a significant (p<0.05) positive correlation.

TX and HX Effects on Toad Locomotion 101

The mean number of active days for females was
also less than that for males (cf. Fig. 2). Due to the
small number of active individuals, no significant
difference was obtained or no statistical test was
valid for the comparison of the number of active
toads or the number of active days among the

Males
24

Observed number (Z)
o rs ©

(o)

groups.

Correlations between atmospheric parameters and
toad activity

The effect of temperature or humidity on the
number of active toads was studied by simple

r=0.887 (P<0.01)

12 18 24

Calculated number
(Z=1.99X+0.26Y-24.70)

Females
15

—_—
oO

Observed number (Z)
Oo

(o)

-4 (0)

r=0.806 (P<0.01)

4 8 12

Calculated number
(Z=1.10X+0.07Y-10.65 )

Fic. 5. The multiple correlation among temperature, humidity and the number of active male and female toads. A
highly significant multiple correlation was observed between the temperature plus humidity and the number of
active toads. On the horizontal axis(Z’), the expected numbers of active toads, which were calcuated by using the
regression formula with observed temperature(X) and humidity(Y) values, are plotted. On the vertical axis(Z),

the observed numbers of active toads are plotted.

102 Y. TASAKI AND S. ISH

correlation analysis. The combined effect of both
temperature and humidity was also studied by
multiple correlation analysis. Females and males
were separately analyzed.

As depicted in Figure 3, temperature showed a
significant (p<0.05) positive correlation with the
number of active toads in both males and females.
Humidity showed a significant positive correlation
in males only (Fig. 4). In females, the correlation
coefficient was positive, but it was too small to be
statistically significant (p >0.05, Fig. 4).

When both temperature and humidity were used
as independent variables, a highly significant mul-
tiple correlation (p<0.01) with the number of
active toads was observed in males as well al in
females (Fig. 5).

DISCUSSION

A number of investigations have been published
on the role of thyroid hormones in the migration of
vertebrates: in fish [14-15], amphibians [5, 7, 16,
17] and birds [18]. Our observations on the annual
cycle of plasma thyroxine and triiodothyronine
levels in the toad, Bufo japonicus, seemed to
support the idea that thyroid hormone plays a role
in toad migration. On the contrary, we recently
observed that the treatment of male Bufo japoni-
cus with thyroxine decreased the distance of
locomotion of a toad kept in a small experimental
chamber [12]. This effect of thyroxine was
observed in both normal and thyroidectomized
males, and thyroidectomy increased the distance
of locomotion. However, we were cautious in
concluding that the sedative effect of thyroxine
was physiological, since the effect was observed
under artificial conditions.

Our former results of the thyroidectomy experi-
ment [12] were confirmed under the quasi-natural
conditions in the present study. The thyroidec-
tomy increased both the number of toads which
showed activity and the number of days on which
the toads showed activity. Accordingly, we con-
clude that thyroxine decreased both parameters
representing the occurrence of locomotor or
migratory activity during the breeding season even
under the quasi-natural conditions. However, it is
not known whether the effect of thyroxine is direct

or indirect. Furthermore, the physiological mean-
ing of this effect is still obscure, although we
postulated that it might be related to post-breeding
inactiveness [12].

Hypophysectomy clearly suppressed the activity
of male toads. This suggests that some hormone(s)
of the pituitary gland activates(s) migratory activ-
ity in male toads. However, it is difficult to
conclude whether this suppressive effect of
hypophysectomy is due to the ablation of a specific
hormone controlling the migration or the ablation
of hormone(s) regulating the general metabolism.

Another problem was the inactiveness of the
female toads, which caused difficulty in observing
the effect of the operations in females. Only a few
females showed activity in the present study as well
as in the previous study in the chamber [12]. This
may be related to the fact that not all female toads
participate in breeding activity every spring [19-
21).

It is revealed in the present study that the
combination of temperature and humidity is the
external factor which initiates locomotor activity in
early spring. The high humidity may be advan-
tageous for gas exchange through the skin of the
toad and the relatively high temperature, for ener-
gy metabolism. A high rate of this external and
internal respiration is necessary for muscular activ-
ity during the migration of toads in early spring.
We can now predict the occurrence of migration in
toads from the temperature and humidity by using
the regression formula obtained in the present
study.

ACKNOWLEDGMENTS

The authors are grateful to Prof. Yasuyuki Oshima for
his valuable advice and suggestions. This study was
supported by a Grant-in-Aid from the Japanese Ministry
of Education, Science and Culture.

REFERENCES

1 Hisai, N. and Sugawara, T. (1978) Ecological stu-
dies of Bufo bufo japonicus Schlegel (V) The rela-
tion between appearances and the climatic condi-
tions at breeding season. Miscellaneous Reports of
the National Park for Nature Study, 8: 135-149. (In
Japanese)

10

11

TX and HX Effects on Toad Locomotion

Okuno, R. (1985) Studies on the natural history of
the Japanese toad, Bufo japonicus japonicus. VIII.
Climatic factors influencing the breeding activity.
Jap. J. Ecol., 35: 527-535. (In Japanese)

Reinke, E. E. and Chadwick, C. S. (1940) The
origin of the water drive in Triturus viridescens. J.
Exp. Zool., 83: 223-233.

Chadwick, C. S. (1944) Further observations on the
water drive in Triturus viridescens. J. Exp. Zool.,
86: 175-187.

Crim, J. W. (1975) Prolactin-induced modification
of visual pigments in the eastern red-spotted newt,
Notophthalmus viridescens. Gen. Comp. Endocri-
nol., 26: 233-242.

Duvall, D. and Norris, D. O. (1977) Prolactin and
substrate stimulation of locomotor activity in adult
tiger salamanders (Ambystoma tigrinum). J. Exp.
Zool., 200: 103-106.

Duvall, D. and Norris, D. O. (1980) Stimulation of
terrestrial-substrate preferences and locomotor
activity in newly transformed tiger salamanders
(Ambystoma tigrinum)by exogenous or endogenous
thyroxine. Anim. Behav., 28: 116-123.

Moriya, T. (1982) Prolactin induces increase in the
specific gravity of salamander, Hynobius retardatus,
that raises adaptability to water. J. Exp. Zool., 223:
83-88.

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.

Ishii, S., Yoneyama, H., Inoue, M., Yamamoto, K.
and Kikuyama, S. (1989) Changes in plasma and
pituitary levels of prolactin in the toad, Bufo japoni-
cus, throughout the year with special reference to
the breeding migration. Gen. Comp. Endocrinol.,
74: 365-372.

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.

12

13

14

15

16

17

18

19

20

21

103

Tasaki, Y. and Ishii, S. (1989) Effects of thyroxine
on locomotor activity and carbon dioxide release in
the toad. Bufo japonicus. Zool. Sci. (In press)
Ishii, S. (1983) “Programs of statistical methods for
biologists by N88-BASIC”, Baifukan, Tokyo.
Woodhead, A. D. (1975) Endocrine physiology of
fish migration. Oceanogr. Mar. Biol. Annu. Rev.,
13: 287-382.

Dickhoff, W. W., Folmar, L. C. and Gorbman, A.
(1978) Changes in plasma thyroxine during smol-
tification of coho salmon, Oncorhynchus kisutch.
Gen. Comp. Endocrinol., 36: 229-232.

Grant, W. C., Jr. and Cooper, G. IV (1965) Be-
havioral and integumentary changes associated with
induced metamorphosis in Diemictylus. Biol. Bull.,
129: 510-522.

Dent, J. N. (1985) Hormonal interaction in the
regulation of migratory movements in urodele
amphibians. In “The Endocrine System and the
Environment”. Ed. by B. K. Follet, S. Ishii and A.
Chandola, Japan Sci. Soc. Press, Tokyo/Springer-
Verlag, Berlin, pp. 79-84.

Berthold, P. (1985) Migration: Control and metabo-
lic physiology. In “Avian Biology”, Ed. by D. S.
Farner and J. R. King, vol. 5, Academic Press, New
York, pp. 77-128.

Hisai, N. (1981) Ecological studies of Bufo bufo
formosus Boulenger (V1) Differences of postmeta-
morphic growth rate and the sexual maturity be-
tween sexes in the natural population. Miscel-
laneous Reports of the National Park for Nature
Study, 12: 103-113. (In Japanese)

Okuno, R. (1985) Studies on the natural history of
the Japanese toad, Bufo japonicus japonicus. V.
Post-metamorphic survival and longevity. Jap. J.
Ecol., 35: 93-101. (In Japanese)

Okuno, R. (1986) Studies on the natural history of
the Japanese toad, Bufo japonicus japonicus. IX.
Male behaviors during breeding season. Jap. J.
Ecol., 35: 621-630. (In Japanese)

ne |

(lees

tnd o Pitot
at’ POR)
Ug ut bate call
ee it

cys ru vires

ze setts PW
is i |
OAK hele

¢ howphok’s 062 4)

er! at tiie

peeieealiry
eS A oct

> Ww nasi? >

oe WW ey

7 aw Eon
no 4 i eli '
ima)

oh

b

fi}

- Swe

i hed) tanta «aa ote bow £T :

=
te

fal,

‘

4

1 P
— x 7 | a ae
a é ’ , -
ab i a
a
in
i

“el Se Tl
ie

mis

{ aol ome.” Vies sal

nérten gitviaeniins a La
WOCe Bua ( imnealarg

uypiiate Woe uae Bat
anor yee
pies ae,

aes ic nav
Tm AP Te Di he
emaaarecetie e
Pcehany MOD et ag Sana ROL:
Tht ok a ge mt

Wé ft
Sika.) & ‘A .
eb Ry eee tale ORD i a
eronrgihae For wove
4 F F140

14 10 Genre eae

Ret aL oh =
i : Wns H “eq ,

64 2 > et

und!) 3° ett “ia tad
cher Weyl bey! saps ohne wrens Thy
atl ahemertbncs ali ae genie

y's

P, x es i
iiibethe « , [4 paves i) co
7 + ee aaes® (Oey

rs yh! WN My nyt) tiliohth
2) eiatrogeil te Waegiudty
prints Apeped earns tine piney

a) ‘an ih? fete 1
=< Ai a » now hat lewertvill sh
® noreobns qd a i

\ my [
-
é * wt
~
, 7

ZOOLOGICAL SCIENCE 7: 105-109 (1990)

© 1990 Zoological Society of Japan

Effects of Hypophysectomy and Replacement Therapy with
Several Hormones on Plasma Sodium Concentrations
in Bullfrog Tadpoles

MINorU UCHIYAMA and TosHIKI MURAKAMI

Department of Oral Physiology, School of Dentistry at Niigata,
The Nippon Dental University, Niigata 951, Japan

ABSTRACT—Hypophysectomy significantly decreased plasma sodium concentrations in bullfrog

tadpoles kept in low-sodium media (2.5 megq/liter).

A study on replacement therapy revealed that

administration of tadpole pituitary homogenate corrected this hyponatremia in hypophysectomized
tadpoles. These results show that in a low-sodium environment where the tadpoles live in nature, the
pituitary gland is important for plasma sodium regulation. AVT and ovine prolactin did not maintain
normal plasma sodium levels in hypophysectomized tadpoles, whereas ACTH and corticosteroids
(especially cortisol) restrained the sodium decrease. Therefore, in pre-metamorphic tadpoles it seems
that the pituitary-adrenal axis plays an important role in plasma sodium regulation.

INTRODUCTION

Among the lower vertebrates, the amphibians,
as a group, have been studied extensively with
regard to their mechanisms of osmotic and ionic
regulation. It is well known that the skin of
amphibians can actively take up salt from the
surrounding aqueous environment. In these ani-
mals, corticosteroids especially aldosterone, and
neurohypophysial hormone, arginine vasotocin
(AVT), generally show potent natriferic and
osmotic actions [see 1, 2]. It is known that
tadpoles also actively transport sodium and chlor-
ide ions into their body fluids from a low-
concentration external solution [3-6]. The integu-
ment of tadpoles, however, has been thought to be
inessential for absorption of ions from the environ-
ment [7-9]. Jn vitro studies have shown that the
main organ for active uptake of ions in tadpoles is
not the skin but the gills [4, 10]. On the other
hand, in unfed freshwater fish, accumulation of
sodium depends almost entirely on the active
uptake of sodium across the gill epithelium. It is
also known that hypophysectomy results in an
increased rate of sodium loss, and that in many

Accepted March 20, 1989
Received December 11, 1988

cases, this can be prevented by injection of prolac-
tin [reviewed by 11]. However, injection of ovine
prolactin into normal tadpoles acclimated to low
aqueous calcium and sodium or high aqueous
calcium and sodium does not produce consistent
hypernatremia [12], and it has been observed that
ovine prolactin does not correct hyponatremia in
hypophysectomized tadpoles [13]. In tadpoles,
therefore, the role of prolactin in plasma sodium
metabolism seems to be minor.

The present study was undertaken to investigate
hypernatremic hormone(s) in bullfrog tadpoles,
with special reference to the effects of replacement
therapy with both pituitary gland and pituitary
hormone and corticosteroid on plasma sodium
concentration in hypophysectomized tadpoles.

MATERIALS AND METHODS

Tadpoles of the bullfrog, Rana catesbeiana
Shaw, at T-K stages VI to XIV were obtained from
commercial dealers in Niigata [14]. They were
kept for at least 1 week in low-calcium and low-
sodium water (Ca: 0.85 meq/liter, Na: 2.5 meq/
liter) at room temperature and maintained unfed
during the experiments. Tadpoles were anesthe-
tized in 0.05% MS 222 (m-aminobenzoic acid
ethylester methanesulfonate, Sankyo) prior to

106 M. UcnrivAMA AND T. MuRAKAMI

manipulation, hormone injection and blood sam-
pling. The procedures of blood sampling and the
method of plasma sodium determination have
been described previously [12]. All data are pre-
sented as means+SE and statistical analysis of
differences in means was performed using Stu-
dent’s f-test.

Hypophysectomy and autotransplantation or pitui-
tary homogenate injection

Hypophysectomy and autotransplantation or
pituitary homogenate injection were done by the
methods reported previously [12, 15]. In the
autotransplantation experiment, tadpoles were
hypophysectomized and then the removed pitui-
tary gland was transplanted under the skin of the
head region of the same individual (HYPX+
autotransp.). A piece of muscle from each tadpole
was transplanted into sham-transplanted tadpoles
(HYPX-+m. transp.). In the pituitary homoge-
nate injection study, hypophysectomized tadpoles
were injected intraperitoneally with a homogenate
of tadpole pituitary gland. Each animal received a
homogenate of 2 pituitary glands per injection
daily for 6 days.

In these experiments, treated tadpoles were
maintained in low-sodium water (Na: 2.5 meq/
liter) for one more week and then sacrificed for
blood sampling.

Hypophysectomy and hormonal treatment

In these experiments, hypophysectomized tad-

poles were divided into several groups and main-
tained in low-sodium water (Na: 2.5 meq/liter).
Then the effects of certain hormones were ex-
amined. The following hormones were used: ovine
prolactin (NIH-p-S8), aldosterone (Sigma),
ACTH (Armour Pharmaceutical Co.), corticoste-
rone (Sigma), cortisol (Merck Sharp & Dohme),
and AVT (Sandoz). These hormones were dis-
solved in 0.6% saline and/or 95% ethanol. Injec-
tions (20-25 pl/animal) were made with a Hamil-
ton microsyringe into the lymph sinus beneath the
skin of the back, passing through the tail muscles
to prevent leakage. Injections were started one
day after the operation. Each group received one
of the above-mentioned hormones or the combina-
tions of them daily for 6 days. Two series of
experiments were performed.

RESULTS

Hypophysectomy followed by autotransplantation
or injection of pituitary homogenates

The results obtained in these experiments are
shown in Tablel. It was confirmed that
hypophysectomy caused significant hyponatremia,
showing that the pituitary gland is important for
plasma sodium regulation in bullfrog tadpoles.
There was a significant difference between the
plasma sodium concentration in the sham opera-
tion group and that in the HYPX-+muscle trans-
plant group (P<0.001). In the second experiment,

TaBLe1. Effect of hypophysectomy and replacement therapy on plasma sodium in bullfrog tadpoles
Significance compared to:
Experiment Treatment Number Time after Na sham HYPX*+m. —
treatment (meq/liter) operation __transp. or saline
sham operation 9 1 week 97.9+1.4° — P<0.001
1° HYPX-+m. transp. 10 1 week 87.5+1.3 P<0.001 —
HYPX-+ autotransp. 9 1 week 92.2+2.0 P<0.05 NS
sham operation 6 1 week 96.0+1.0 — P<0.001
2 HYPX-+ saline 9 1 week 86.8+1.4 P<0.001 _
RSE piautary 12 1 week 93.0+1.7 NS P<0.05
omogenate

“ Abbreviations used: HYPX, hypophysectomy; m. transp., muscle transplantation; saline, saline injection.

> Values are means+SE.

© Number of experiments conducted: 1, HYPX and autotransplantation; 2, HYPX and injection with pituitary

homogenates.

Plasma Na in Tadpole 107

plasma sodium concentration in the sham opera-
tion group was significantly higher than that in the
HYPX-+saline group (P<0.001). The plasma
sodium concentration in the HYPX-+ pituitary
homogenate group was also significantly higher
than that in the HYPX-+saline group (P<0.05).

Hypophysectomy and hormonal treatment

In the first experiment, the effects of different
doses of a hormone or a combination of two
hormones were examined on plasma sodium con-
centration in hypophysectomized tadpoles. Table
2 shows the results obtained from this experiment.

Plasma sodium concentration in the sham opera-
tion+saline group was significantly different from
those in the HYPX-+saline, HYPX+AVT (0.02
ng/g and 0.2 ng/g) and HYPX-+ ovine prolactin (5
peg/g) groups (P<0.05). Injections of ACTH (50
ng/g), cortisol (0.5 ug/g, 5 ug/g), ACTH and ovine
prolactin in combination, or aldosterone (50 ng/g)
into hypophysectomized tadpoles produced signif-
icantly higher plasma sodium concentrations than
those in hypophysectomized tadpoles (P<0.05),
whereas no statistically significant difference was
found between the HYPX-+ saline group and the
other experimental groups. In the second experi-

TaBLE2. Effect of hypophysectomy and hormonal replacement therapy on plasma sodium in bullfrog

tadpoles
Significance compared to:
Tien men AIGRDSS Gea Mas ee ere
sham operation -+ saline 14 99.9+1.0 — P<0.05
HYPX + saline 10 97.0+0.5 P<0.05 —
HYPX+ ACTH (50 ng/g) 10 100.4+0.8 NS P<0.01
HYPX+ACTH (500 ng/g) 10 99.3+1.1 NS NS
HYPX-+ aldosterone (5 ng/g) 8 99.1+2.1 NS NS
HYPX-+ aldosterone (50 ng/g) 12 100.2+0.9 NS P<0.01
HYPX-+ corticosterone (0.5 yg/g) 12 97.5+0.9 NS NS
HYPX-+ corticosterone (5 g/g) 10 98.8+1.2 NS NS
HYPX-+ cortisol (0.5 yg/g) 12 99.8+1.1 NS P<0.05
HYPX-+ cortisol (5 g/g) 10 100.8+1.1 NS P<0.05
HYPX+o0PRL (5 pe/g) 9 95.9+1.4 P<0.05 NS
HYPX-+o0PRL (5 pg/g)+ ACTH (150 ng/g) 12 100.8+0.9 NS P<0.01
HYPX+AVT (0.02 ng/g) 12 OS, /s Il P<0.05 NS
HYPX+AVT (0.2 ng/g) 12 96.9+1.0 P<0.05 NS

* Values are means+SE.

TaBLeE3. Effect of hypophysectomy and hormonal replacement therapy on plasma sodium in bullfrog
tadpoles

Significance compared to:

Na sham operation HYPX

Gixcatment Nitistloe (meq/liter) + saline + saline

sham operation+saline 10 98.4+0.8* — P<0.05
HYPX-+ saline 8 91.0+2.4 P<0.05 —
HYPX-+ ACTH (50 ng/g) 8 96.2+2.0 NS NS
HYPX + aldosterone (50 ng/g) 5 94.8+3.0 NS NS

HYPX-+ cortisol (5 g/g) 10 100.0+1.8 NS P<0.01
HYPX+o0PRL (5 pg/g) 8 94.4+2.8 NS NS

* Values are means+SE.

108 M. UcCHIYAMA AND T. MURAKAMI

ment, ACTH, aldosterone and ovine prolactin
failed to maintain normal plasma sodium concen-
trations. However, cortisol corrected the plasma
sodium concentration after hypophysectomy (P<
0.01). These results are summarized in Table 3.

DISCUSSION

In bullfrog tadpoles kept in low-sodium water
(Na: 2.5 meq/liter), the previous observation that
hypophysectomy brought about a significant de-
crease in plasma sodium concentration was con-
firmed [12, 13]. In the replacement therapy study,
administration of tadpole pituitary homogenate
corrected the hyponatremia and pituitary grafts
partially prevented this in hypophysectomized tad-
poles. These results suggest that the pituitary
gland is important for maintaining plasma sodium
concentrations in bullfrog tadpoles. The next
question to be answered was which pituitary hor-
mones are important for the control of plasma
sodium concentration in bullfrog tadpoles.

In the present study, AVT was not able to
maintain a normal plasma sodium level after
hypophysectomy. Bentley and Greenwald [16]
reported that the pituitary gland of the bullfrog
tadpole contains only about one-fifth of the
neurohypophysial peptides (activity/kg body
weight) present in the adult frog. It is also
reported that anuran tadpoles exhibit little or no
response to AVT, although AVT is an antidiuretic
and natriferic hormone in adult amphibians [8, 16].
Therefore, the present result might be explained
by assuming that AVT administration is not suf-
ficient for maintaining a normal plasma sodium
level and/or AVT might not be a potent agent for
the control of hydromineral balance in young
tadpoles. In the present study, prolactin was also
unable to restore plasma sodium concentrations in
hypophysectomized tadpoles. This result is consis-
tent with the previous report [13]. Brown and
Brown [17] suggested that the major role of prolac-
tin in amphibians might be water retention. If this
is so in bullfrog tadpoles, it is possible that the
increase in plasma sodium concentration caused by
prolactin treatment might be masked by hemodilu-
tion. On the other hand, Clemons and Nicoll [18]
reported that the circulating plasma level and

pituitary content of prolactin were increased sig-
nificantly during metamorphic climax in bullfrog
tadpoles. It was also shown that during the climax
stage, prolactin is involved in regulation of the
active sodium transport system in the ventral skin
of bullfrog tadpoles [19, 20]. The bullfrog tadpoles
used in the present study were pre-metamorphic
larvae (stages VI-XIV). Therefore, there are two
possible explanations for the minimal effect of
prolaction on sodium regulation in young tadpoles.
One is that prolactin secretion is insufficient in
young tadpoles and the other is that the target
systems of prolactin sodium transport might be
undeveloped. The latter concept is probably true
in the case of ACTH (presumably acting through
the adrenal gland), since Krug et al. [21] showed
that a single injection of ACTH failed to elevate
the serum level of aldosterone, corticosterone and
cortisol in pre-metamorphic larvae (stages X-
XIV), whereas the same treatment caused a strik-
ing increase in the levels of corticosterone and
cortisol in more advanced tadpoles (stages XV-
XIX). In the present study, however, it is possible
that long-term ACTH treatment might have stimu-
lated the production of corticosteroid in young
tadpoles.

It is generally accepted that aldosterone and
corticosterone are produced in the interrenal
organ, and that corticosteroids, especially aldo-
sterone, are involved in the regulation of electro-
lytes in adult amphibians. However, little in-
formation is available on the influence of corticos-
teroids on electrolyte regulation in larval anurans.
Krug et al. [21] reported that serum aldosterone
was maintained at fairly low levels in bullfrog
tadpoles until a signifcant increase occurred in the
metamorphic climax stage. In the present study,
injection of aldosterone was inconsistent in restor-
ing hyponatremia caused by hypophysectomy.
Therefore, aldosterone might not be a major
sodium-regulating agent in pre-metamorphic bull-
frog tadpoles. Krug et al. [21] detected very low
levels of corticosterone in serum during stages V to
X, and then serum corticosterone concentration
increased steadily until stage XVII. In the present
study, which examined the effects of corticoster-
one and cortisol in bullfrog tadpoles (stages VII-
XII) kept in low-sodium water, the hypernatremic

Plasma Na in Tadpole

effect of corticosterone did not exceed that of
cortisol at the same dose. It is not yet known
whether cortisol is produced in larval amphibians.
However, according to Krug et al. [21] a cortisol-
like substance is present in the serum of young
tadpoles (stages V-XXV). In the present study,
cortisol was the most potent hypernatremic agent
in young tadpoles. In pre-metamorphic bullfrog
tadpoles, a cortisol-like substance was detectable
[21] and their gills were important site for sodium
uptake [4, 10]. Therefore, a cortisol-like substance
might be involved in the uptake of sodium by the
gills of bullfrog tadpoles. This may also influence
the metabolic processes in the body, thus indirectly
affecting sodium homeostasis and osmoregulation.
In conclusion, it seems that the pituitary-adrenal
axis 1s important for plasma sodium regulation in
pre-metamorphic tadpoles.

ACKNOWLEDGMENTS

The authors wish to thank Prof. C. Oguro of Toyama
University for his valuable discussions and revision of the
manuscript. The authors’ thanks are also extended to
Prof. P. K. T. Pang, University of Alberta, for his
suggestions and encouragement.

REFERENCES

1 Bentley, P. J. (1971) The Amphibia. In “Endoc-
rines and Osmoregulation”. Ed. by P. J. Bentley,
Springer-Verlag, Berlin Heidelberg, New York, pp.
161-192.

2 Bentley, P. J. and Baldwin, G. F. (1980) Compari-
son of trans-cutaneous permeability in skins of larval
and adult salamanders (Ambystoma tigrinum). Am.
J. Physiol., 239: R505—-508.

3 Kawada, J., Taylor, R. E. and Barker, S. B. (1968)
Measurement of Na-K ATPase in the separated
epidermis of Rana catesbeiana frogs and tadpoles.
Comp. Biochem. Physiol., 30: 965-975.

4 Alvarado, R. H. and Moody, A. (1970) Sodium and
chloride transport in tadpoles of the bullfrog Rana
catesbeiana. Am. J. Physiol., 218: 1510-1516.

5 Casada, J. H. and Nichols, J. R. (1986) Interrela-
tionships among epidermal Na-K ATPase, develop-
mental stage and length of Rana catesbeiana tad-
poles. Comp. Biochem. Physiol., 3: 429-433.

6 Robinson, D. H. and Mills, J. W. (1987) Ouabain
binding in tadpole ventral skin. I. Kinetics and effect
on intracellular ions. Am. J. Physiol., 253: R402-
409.

7 ‘Taylor, R. E. and Barker, S. B. (1965) Trans-

10

11

12

13

14

15

16

17

18

19

20

21

109

epidermal potential difference: development in
anuran larvae. Science, 148: 1612-1613.

Alvarado, R. H. and Johnson, S. R. (1966) The
effects of neurohypophysial hormones on water and
sodium balance in larval and adult bullfrogs (Rana
catesbeiana). Comp. Biochem. Physiol., 18: 549-
561.

Cox, T. C. and Alvarado, R. H. (1979) Electrical
and transport characteristics of skin of larval Rana
catesbeiana. Am. J. Physiol., 237: R74-79.

Dietz, T. H. and Alvarado, R. H. (1974) Na and Cl
transport across gill chamber epithelium of Rana
catesbeiana tadpoles. Am. J. Physiol., 226: 764-770.
Ball, J. N. (1969) Prolactin and osmoregulation in
teleost fishes: a review. Gen. Comp. Endocrinol.,
Suppl., 2: 10-25.

Uchiyama, M. and Pang, P. K. T. (1981) Endocrine
influence on hypercalcemic regulation in bullfrog
tadpoles. Gen. Comp. Endocrinol., 44: 428-435.
Sasayama, Y. and Oguro, C. (1982) Effects of hy-
pophysectomy and replacement therapy with pitui-
tary homogenates or ovine prolaction on serum cal-

cium, sodium, and magnesium concentrations in bull-
frog tadpoles. Gen. Comp. Endocrinol., 46: 75-80.
Taylor, A. C. and Kollros, J. J. (1946) Stages in the
normal development of Rana pipiens larvae. Anat.
Rec., 94: 7-23.

Uchiyama, M. and Pang, P. K. T. (1982) Replace-
ment therapy and plasma calcium concentration in
hypophysectomized bullfrog tadpoles, Rana cates-
beiana. Gen. Comp. Endocrinol., 47: 351-356.
Bentley, P. J. and Greenwald, L. (1970) Neurohy-
pophysial function in bullfrog (Rana catesbeiana)
tadpoles. Gen. Comp. Endocrinol., 14: 412-415.
Brown, P. S. and Brown, S. C. (1987) Osmoregula-
tory actions of prolactin and other adenohypophy-
sial hormones. In “Vertebrate Endocrinology: Fun-
damentals and Biomedical Implications, Vol. 2”.
Ed. by P. K. T. Pang and M. P. Schreibman,
Academic Press, San Diego, pp. 45-84.

Clemons, G. K. and Nicoll, C. S. (1977) Develop-
ment and preliminary application of a homologous
radioimmunoassay for bullfrog prolactin. Gen.
Comp. Endocrinol., 32: 531-535.

Eddy, L. J. and Allen, R. F. (1979) Prolactin action
on short circuit current in the developing tadpole
skin: a comparison with ADH. Gen. Comp. Endo-
crinol., 38: 360-364.

Takada, M. (1986) The short-term effect of prolac-
tin on the active Na transport system of the tadpole
skin during metamorphosis. Comp. Biochem. Phys-
iol., 85A: 755-759.

Krug, E. C., Honn, K. Y., Battista, J. and Nicoll, C.
S. (1983) Corticosteroids in serum of Rana cates-
beiana during development and metamorphosis.
Gen. Comp. Endocrinol., 52: 232-241.

oe
oa

relnak , Chana

outs

Lopate any Tenaga =
heegg~ Ant Tet oteners i
ce ereylAy 8

baiey prey toga to atyalhy
ae \G¥ vel v2 DUsTE i h fit a

vant in Sy 1 q ad ite

{ae

oinpity tives whee
catrond yay hem ahr oo
‘ruining bw

pease SDD hn 2 gal

if ohare prema ey

hogy, ban OT 260 2 Cadyae lina woe t
prac rads. hoganat bey geste bet ead
hy lM |b (te vie OMA tg Aaah tot jester i
& Obey ly Dit Fi T skis 410 Ui i
get tite Nise ‘ BY ft ‘Pecvenierpest (PanbabNisiae wet Horta og:
wv. % FLvepeuaet iA\ Ay naa Situenn deeb API vy a
nie etal YONOE RT POO OUMpe aire tamil tng enim
. (8, sion (tate pena +4 Sell ni cada er gah
11 pee MIA \ Vib ao veh alert fitted) eat
»! ) } ‘ 4 t ake { > i ; "
ORE RRS LAL RTM ON ee Hipage wits CAE eH
i Wigs vavl -f ue, ll ae z es ~
aan aie eae a eet f abt 4 Feat: ot
bon ay, We Rep Ae 4 iweaey idl
“ tals 4 Hei dow cotrgeg " j iguwt
sist riot ie 4 3
ae I A Hw. tort hy ' “ ‘ete a
=) - foohe) aaa i
iy een eaoeinene os aera
, vats levered t Bit vorgews | aT PH ia f
Win tr
‘hee ia, «Sem Mia lait. ie “eae
\ f bat Ht
7 ‘ *) mbalyy sl ny
My el nla ae okt sau '
Weel (1 nt OF. "noiieeasnans
ym iw fn, oe Neild H hfe’ a)
j esi oneh Han iy
: bate 7S Mieye! x} iy Pe ei
rut ctl Sewit C\)le TaD
’ 1 une voiny cokpen yea } 7
YR ©
eh) dtehl Daw ry
Mewu aa eo
- ferytrad) Ries, gia? nacho ed "
= ul ' (0 tive oath ie
jor Wegeteo | (TV). Al er Miealt
ents | hy CHO tig .
Jia ‘7 art PA my taf ie ,
he Pty 4 alate
1 A811 A eS knelge seen
y tay wteahnd “ali a4 dist hg yf}
dg Le O08 8 owt Cadet
al

ttucjecrt 3 re Wen de A Gi,
2 Getta tn! 108 | Oidaestney chatter Py

HY AD. toladt Sy hii Ae Age
>
WEF Sachi Se

fast)

niet RR eS

ZOOLOGICAL SCIENCE 7: 111-126 (1990)

© 1990 Zoological Society of Japan

Two Species of the Genus Actacarus (Acari, Halacaridae)
from Japan

HirosuHi ABE

Department of Systematic Zoology, Division of Environmental Structure,
Graduate School of Environmental Science, Hokkaido University,
Sapporo 060, Japan

ABSTRACT— A new arenicolous halacarid mite, Actacarus karoensis sp. nov., is described and another

species, Actacarus illustrans Newell, 1951, is newly recorded from Japan.

A. karoensis is easily

distinguishable from congeners on the basis of several characters including the shape of the genitoanal
plate and length of the ovipositor. As for A. illustrans, it was noted that the Japanese specimens differ
from the North American specimens in some morphological characters. These differences are regarded

here as intraspecific variations.

INTRODUCTION

Taxonomic knowledge of marine halacarid mites
in Japan is limited. The first record of the family in
Japan was in 1927, when Halacarus spongiphilus
Kishida, 1927 was described from the abyssal zone
in Sagami Bay [1]. An additional species, Agauop-
sis okinavensis, was described by Bartsch [2] from
Okinawa, southern Japan. The present paper
describes two species of the genus Actacarus newly
found in Japan.

MATERIALS AND METHODS

Specimens were fixed with modified Imamura’s
fluid [3], dissected in a drop of pure lactic acid, and
mounted in gumchloral medium. Observation was
made under a phase-contrast microscope using oil
immersion, figures were drawn with the aid of a
camera lucida, and measurements were made with
an ocular micrometer. Sizes of idiosoma and
gnathosoma were measured before dissection,
while other parts were measured on dissected
specimens.

Terms: The terms for body parts of halacarid
mites follow Newell [4-7].

Presentation of numerical data: Metric charac-

Accepted January 26, 1989
Received November 5, 1988

ters are always given in micrometers (um). Meris-
tic characters are sometimes given with ranges.
Presentation of leg chaetotaxy and arrangement of
subgenital setae follow Newell [7]. In describing
positions of certain structures on a plate, the
decimal system developed by Newell [6-9] is em-
ployed; for example, the statement ‘setae at 0.44
on the genitoanal plate’ means that the setae are
located at a level 0.44 of the interval between the
anteromedian point (0.00) and the posteromedian
point (1.00) on the plate.

Measured parts (letters in parentheses refer to
those given in Fig. 1): Idiosoma: Length (a), from
the anteriormost margin of the anterior dorsal
plate to the terminal end of the anal papilla; width
(b), at the level of the lateral coxal margin of leg
III. Plate, genital foramen, and spermatophor-
otype: Length (c), from the anterior margin to the
posterior margin; width (d), at the widest level.
Gnathosoma: Length (e), from the posterior mar-
gin of the base of the gnathosoma to the anterior
tip of the rostrum; width (f), at the widest level of
the gnathosoma. Base of gnathosoma: Length (g),
from the level of the base of the palpal insertions
to the posterior margin of the gnathosoma. Ros-
trum: Length (h), from the level of the base of the
palpal insertions to the anterior tip of the rostrum;
width (i), at the widest level. Basal cheliceral
segment: Length (j), from the level of the most
proximal end of the segment to the level of the tiny

112 H. ABE

b f

Fic. 1.
parts explained in the text.

ventral gap; height (k), at the highest level. Mov-
able digit: Length (1), from the level of the tiny gap
to the distal end of the digit. Fixed digit: Length
(m), from the level of the tiny gap to the distal end
of the digit. Palpal segment: Length (n), from the
proximal end to the distal end along the ventral
margin; height (0), at the highest level. Leg:
Length (p), from the proximal end of the trochan-
ter to the distal end of the tarsal claw fossa along
the ventral margin.

Abbreviations: AD, anterodorsal plate; PD,
posterodorsal plate; OC, ocular plate; AE, anter-
ior epimeral plate; PE, posterior epimeral plate;
GA, genitoanal plate; ds, dorsal setae; aes-i,
anterior epimeral setae; aes-ii-lat (-v), lateral (ven-
tral) setae of coxae II; pes-iii-lat (-v), lateral
(ventral) setae of coxae III; pes-iv-a (-P), anterior
(posterior) setae of coxae IV; P-1 to P-4, first to
fourth segment of palp; L/W, the ratio of length to
width; Hal-55/etc., specimen codes of the author’s
personal system. In the present paper the codes
are given only to the described specimens.

In addition, the following abbreviations are used
in the figure legends: Ds, dorsal view; Vr, ventral
view; Lat, lateral view; R, right appendage (or
part); L, left appendage (or part).

Fic. 2. Actacarus illustrans.

C D
o
k
l E ;
p

Diagram of body parts measured. A, idiosoma; B, gnathosoma; C, chelicera; D, palp; E, leg. a-p: Measured

Family Halacaridae Murray, 1987
(Japanese name: Ushiodani-ka)

Subfamily Actacarinae Viets, 1939
(Japanese name: Nagisadani-aka, new)

Genus Actacarus Schulz, 1937
(Japanese name: Nagisadani-zoku, new)

Actacarus illustrans Newell, 1951
(Japanese name: Kita-nagisadani, new)
(Figs. 2-6)

Actacarus illustrans Newell, 1951 [4], pp. 33-36,
Figs. 126-140. (Holotype: Male, in American
Museum of Natural History; type locality, Unalas-
ka Island, Alaska.)

Actacarus illustrans: Krantz, 1976 [10], pp. 255-
257, Tab. 5, Figs. 30-32. (Descriptions of imma-
ture stages with leg chaetotaxy in adult, record
from Oregon.) (nec A. illustrans sensu Vorob’yeva
and Yaroshenko, 1979 [11].)

Specimens examined. 3 males, 3 females: Inter-
tidal, in coarse sand along shore line at low tide,
Uchikabuto, Oshoro Bay (43°12°N, 140°51’B),
Hokkaido, Sea of Japan, Japan, 26-VI-1986, H.
Abé coll.—1 deutonymph, 1 protonymph, 1 larva:

Male (Hal-55): A, idiosoma (Ds); B, idiosoma (Vr); C, idiosoma (Lat, R); D,

genitoanal region. Female (Hal-1): E, idiosoma (Vr); F, genitoanal region. Males & Females: G, posterior
margin of PE (a, b: R; c, d: L); H, anterior lateral platelet (a-c; L). Scale bars=20 um.

Two Actacarus Species from Japan 113

Ovipositor

Spermatophorotype as.
ds-vi—|! & (] igs Terminal pit

,»- Lateral platelets ~~ ee

114 H. ABE

Intertidal, in coarse sand and gravel along shore
line at low tide, Shamodomari, Oshoro Bay, Hok-
kaido, Sea of Japan, Japan, 19-V-1987, H. Abé
coll.—1 male, 1 deutonymph, 2 protonymphs:
Intertidal, in coarse sand among boulders along
shore line at high tide, Ebisu Rock, Oshoro Bay,
Hokkaido, Sea of Japan, Japan, 23-VI-1987, H.
Abé coll. The following American specimens were
also examined for the purpose of comparison: 1
male, 1 female, 1 deutonymph, 1 protonymph: In
coarse sand near bank, mouth of Schooner Creek,
Pacific coast of Oregon, U.S.A., 18-VI-1974, G.
W. and V. J. Krantz coll.

Description. Male (Hal-55): Idiosoma 256 um
long, 128 um wide, color in life semitransparent
with dark brown specks and a longitudinal white
dorsal line medially.

Dorsum (Fig. 2A) almost completely covered
with two dorsal plates, which are strongly ribbed,
uniformly punctate and with small scattered
alveoli. AD approximately 1/3 length of PD, L/W
0.66, furnished with a large pore on anterolateral
corner on each side, a few minute canaliculi on
anterolateral site, and paired weak areolae on
surface of posterior half. PD 182 um long, 114 ~m
wide, furnished with weak areolae anterolaterally,
a cluster of three weak panels medially, and scat-
tered minute canaliculi laterally on each side; two
pores near lateral margin. OC (Fig. 2C) 16 um

long, 8 zm wide, lateromarginally placed. Two

A

lateral platelets (Fig. 2C) lying marginally on each
side of idiosoma alongside AD and PD, quite
narrow and very weakly sclerotized; the first
platelet divided into anterior microplatelet, 3 ~m
long, 3 um wide, located dorsally to OC, and
posterior microplatelet, 30 4m long, 2 ~m wide,
extending posteriorly to level about midway be-
tween anterior margin of PD and insertion of leg
III; the second platelet lying parallel to PD, near
insertion of leg IV.

Chaetotaxy of dorsal region: Setae ds-i on AD;
ds-ii on anterior microplatelets, seen as if located
on membranous cuticle from dorsal view; ds-iii,
-iV, -v, -vi (adanal setae) on PD.

Venter (Fig. 2B) covered with four plates which
are weakly ornamented in a manner similar to
dorsal plates. AE 110 4m long, 106 um wide,
ornamented with tiny triangular epimeral proces-
ses, anteriorly with a thin membranous collar, with
five sets of lateral and medial subsurface pores of
various shapes. PE (Fig. 2C) 136 um long, 28 um
wide, elongate, marked with some series of mar-
ginal subsurface pores ventrally.

Chaetotaxy of epimeral region: Setae aes-i, aes-
li-lat, aes-ii-v on AE; pes-iii-lat, pes-iii-v, pes-iv-a,
pes-iv-p each on PE.

Genitoanal region (Fig.2B, D): GA 180 um
long, 108 ~m wide, reaching anteriorly just post-
erior to level of pes-iv-a, ornamented with two
anterolateral subsurface pores and cluster of post-

Fic. 3. Actacarus illustrans. Male (Hal-55): A, gnathosoma (Ds); B, gnathosoma (Vr); C, chelicera (L); D, palp

(R); E, tarsus I (L). Scale bars=20 um.

Two Actacarus Species from Japan 115

erolateral weak panels on each side. Genital
foramen about 1/4 length of GA, L/W 1.95,
triangular; anterior margin at 0.68 relative to GA
length. Anal papilla terminal, well separated from
genital foramen. A terminal pit (Fig. 2B) located
laterally just adjacent to anal papilla on each side.
Spermatophorotype (Fig. 2B) massive, approx-
imately 1/2 length of GA, L/W 1.06.

Chaetotaxy of genitoanal region: One pair of
outlying setae at 0.44 on GA; a group of eleven
perigenital setae on each lateral side of genital
foramen as illustrated in Fig. 2D; two subgenital
setae on each genital sclerite, arranged 1-1. No
setae on anal papilla.

Gnathosoma (Fig.3A, B) 70 um long, 56 um
wide, gnathosomal length/idiosomal length 1.06.
Anterior margin of tectum weakly convex. Ros-
trum 36 ym long, 26 ~m wide, furnished with four
pairs of delicate rostral setae. Chelicera (Fig. 3C)
with basal segment 42 um long, 16 ~m wide, partly
punctate, with oblique proximal end. Movable
digit approximately 1/2 length of basal segment,
slightly inclined dorsally, bearing 16-18 minute
denticles along dorsal edge. Fixed digit nearly 1/2
length of movable digit. Palp (Fig. 3D) 61 um

long, slightly inclined ventrally, with four free
segments.

Legs: Length of legs I, II, Ill, IV=194, 162,
174, 192 wm, respectively. Leg chaetotaxy as
follows: Trochanters I-IV, 0-—0-1-1; basifemora,
2-2-2-2; telofemora, 2—2—2-2; genua, 5—4-3-3;
tibiae, 7-S—5—5. Tarsus I (Fig. 3E) with strongly
developed posterior lamella, with three dorsal
setae (one at intermediate level on basidorsal limb,
others on claw fossa), one solenidion, one famu-
lus, three filiform ventral setae (one intermediate-
ly, others distally), and two parambulacral setae
(single euphathidai); solenidion fine, setiform, at
base of fossary lamella; famulus minute, blade-like
in form, with fine canaliculus, lying distally to
solenidion; lateral claws small compared with
those on other legs, with indistinct combs; median
claw bidentate.

Female (Hal-1): Idiosoma 242 4m long, 124 um
wide, resembling male in essential details except
for the sculpture of PD and characters of genitoan-
al region. PD furnished with only one weak panel
at about mid-level on each side. Genitoanal region
(Fig. 2E, F): Genital foramen 32 4m long, 24 um
wide, located terminally and covering anal fora-

B

aN ae ae
Sa

Fic. 4. Actacarus illustrans. Deutonymph (Hal-59): A, idiosoma (Ds); B, idiosoma (Vr). Scale bar=20 pm.

116 H. ABE

men, furnished with three pairs of perigenital
setae; the first pair at 0.40, the second at 0.62
(level of anterior margin of genital foramen), and
the third at 0.87 (behind genital foramen occupy-
ing terminal concavity). Subgenital setae absent.
Three pairs of genital acetabula (Fig. 2F) lying
inside of genital foramen. Ovipositor (Fig. 2E)
short, funnel-like, medially placed.

Immature stages. Deutonymph (Hal-59): Idioso-
ma 228 yum long, 108 um wide. Dorsum (Fig. 4A):
AD slightly concave posteriorly. PD extending
anteriorly to level about midway between inser-
tions of legs II and III. Setae ds-iii inserted in
striated membranous cuticle between AD and PD.
Venter (Fig. 4B): AE narrowed posteriorly from
level of insertion of leg II to level of pes-iv-a.
Lateral platelets short and weakly sclerotized. A
number of subsurface pores present on membra-
nous cuticle between AE and PE. GA 78 pm long,
60 «m wide, extending anteriorly to level of pes-iv-
p, furnished with two pairs of perigenital setae and
internal genital acetabula. Legs: Basifemur IV has
only one seta.

Protonymph (Hal-58): Idiosoma 236 um logn,

A

Koy’

Fic. 5.

108 um wide. Dorsum (Fig. 5A): AD not reaching
posteriorly to level of ds-ii. PD extending anterior-
ly to level of insertion of leg III. Venter (Fig. 5B):
AE strongly narrowed posteriorly from level of
insertion of leg II, with truncated posterior end at
level about midway between pes-iii-v and pes-iv-p.
Setae pes-iv-a lacking. Several subsurface pores
located on membranous cuticle between AE and
PE. GA 54m long, 46 um wide, not reaching
anteriorly to level of trochanter IV, without setae,
bearing one pair of genital acetabula. Legs:
Basifemur III with only one seta; trochanter IV
without setae; femur IV undivided.

Larva (Hal-61): Idiosoma 168 ~m long, 92 ~m
wide. Dorsum (Fig. 6A): AD and PD separated
from each other by about the same length of AD.
Lateral pores on AD and PD relatively large,
distinct. Two subsurface pores placed on membra-
nous cuticle dorsoposterior to OC. Venter (Fig.
6B): AE lacking aes-ii-lat. PE (Fig. 6C) small,
with only one seta ventrally. Two very weakly
sclerotized lateral microplatelets lying posterior to
each of ds-ii. GA 28 um long, 32 wm wide, square
in outline, lacking both genital setae and genital

Actacarus illustrans. Protonymph (Hal-58): A, idiosoma (Ds); B, idiosoma (Vr). Scale bar=20 am.

Two Actacarus Species from Japan 117

Fic. 6. Actacarus illustrans. Larva (Hal-61): A, idiosoma (Ds); B, idiosoma (Vr); C, PE with trochanter of leg III

(R). Scale bars=20 san.

slit. Legs three pairs; basifemur and telofemur
fused. Leg chaetotaxy as follows: Trochanters
I-III, 0-0-1; femora, 3-3-3; genua, 4-4-3; tibiae,
5-5-5.

Morphological variation and abnormality: The
number of panels comprising a cluster at about
mid-level on each side of PD was from two to four
in the male, and one to two in the female. The
outline of the posterior margin of PE (Fig. 2G;
a-d) differed among specimens and even between
sides of one specimen. The microplatelets forming
the first lateral platelet on each side of the idioso-
ma usually were separate, but in some cases were
fused (Fig. 2H; a-c). Several specimens had three
subsurface pores on one side of GA. A few male
specimens had only ten perigenital setae on one
side of the genital foramen. Basifemur IV had two
setae in the described male specimen, but most of
the adult specimens examined had only one seta.
One deutonymph specimen retained the pro-
tonymphal condition in leg segmentation, with leg
IV being five-segmented instead of six.

Distribution: Hitherto recorded from the north-
eastern Pacific coasts of Alaska and Oregon in
U.S.A. This is the first record of A. illustrans from

the western part of the northern Pacific.

Actacarus illustrans sensu Vorob’yeva and
Yaroshenko [11] from the northwestern part of the
Black Sea is identical with A. bacescui Konnerth-
Ionescu, 1970 (=A. illustrans sensu Monniot, 1968
=A. monniotae Krantz, 1974) which was originally
described from the Black Sea as a subspecies of A.
illustrans and has been confused with A. illustrans
Newell, 1951.

Remarks: Actacarus illustrans Newell, 1951 was
originally described from the intertidal zone at
Dutch Harbor, Unalaska Island [4], and its imma-
ture stages were described by Krantz [10]. The
adult specimens from Hokkaido, northern Japan,
accord well with the original description in the
following characters: Setae ds-ii on membranous
cuticle (on microplatelets forming the first lateral
platelets); three pairs of setae on AE; four setae
on PE (one dorsally, three ventrally); GA fur-
nished with one pair of outlying setae and eleven
perigenital setae in the male; ovipositor funnel-
shaped and occupying nearly one third of the
distance between each anterior margin of genital
foramen and GA.

On the other hand, Japanese adult specimens

118 H. Asé

deviate from the original description in the follow-
ing points (corresponding conditions in the original
description in parentheses if necessary): (1) Some-
what larger body size: Male 252-260 um, n=4
(240-246 um, n=3); female 242-248 um, n=3
(227-240 um, n=3); (2) idiosoma with weakly
sclerotized lateral platelets (no reference to the
platelets); (3) anterior margin of tectum weakly
convex (bilobed, from illustration); (4) leg chaeto-
taxy; (5) median claw on tarsus I bidentate (un-
identate); (6) tarsus I with ten setae including
solenidion and famulus (seven setae; solenidion
and famulus lacking); (7) arrangement of
perigenital setae in the male: Anterior two pairs of
perigenital setae located anteriorly (posteriorly) to
level of anterior margin of genital foramen, and
one of the remaining nine pairs lying laterally and
distant from lateral sides of genital foramen (all
nine pairs located in the vicinity of genital fora-
men); and (8) three pairs of distinct perigenital
setae in the female (two pairs).

Of these discrepancies between the specimens
under study and the original description, the dif-
ference in body size (1) is probably attributable to
the individual variation or sampling errors due to
the small sample size. Other inconsistencies may
reflect the insufficient original description, as
shown below: (i) As for the leg chaetotaxy (4), the
present material accords well with the description
later provided by Krantz [10] on the basis of
specimens from Schooner Creek, Oregon; (ii) as
regards the anterior margin of the tectum (3) and
bidentate median claw on tarsus I (5), the present
specimens correspond well with the later descrip-
tion by Krantz [12] based on the paratype series;
(iii) as for the setae on tarsus I (6) and female
perigenital setae (8), Newell [7] later treated these
features as found in the Japanese material as the
generic characters of the genus Actacarus,
although he did not specifically emend his original
description of these characters in A. illustrans; (iv)
the lateral platelets of the idiosoma (2) are very
slender and lateromarginal in position so that they
are not clearly visible in dorsal and ventral views;

these platelets are actually present both in imma-
tures [10] and adults [Abé, the present study] of A.
illustrans from Schooner Creek. Newell [13] might
have overlooked them; (v) the comparison be-
tween specimens from Hokkaido and those from
Schooner Creek shows the close resemblance in
the arrangement of perigenital setae (7) between
specimens from these two localities, so that it is
probable that Newell [13] figured perigenital setae
somewhat insufficiently in his original description.

Krantz [10] described for the first time immature
stages of A. illustrans on the basis of specimens
from Schooner Creek. According to him,
deutonymphs have very poorly developed dorsal
plates, and this author verified his observation in
the deutonymph specimen from Schooner Creek.
The Japanese deutonymph specimens have well
developed dorsal plates.

Actacarus karoensis sp. nov.
(Japanese name: Karo-nagisadani, new)
(Figs. 7-11)

Type series. Holotype: Male, intertidal, in fine
sand along shore line, Karo Beach (35°33'N,
134°13’E), Tottori Pref., Sea of Japan, Japan,
6-X-1987, N. Tsurusaki coll.—Allotype: Female,
data same as the holotype.—Paratypes: 1 male, 3
protonymphs, 1 larva, data same as the holotype; 5
males, 3 females, 1 deutonymph, same locality as
the holotype, 10-VI-1987, N. Tsurusaki coll.

Type deposition: The type series is deposited in
the collections of the National Science Museum,
Tokyo, the Zoological Institute, Faculty of Scien-
ce, Hokkaido University, Sapporo, the National
Museum of Natural History, Smithsonian Institu-
tion, Washington, DC, U.S.A., and in my private
collection.

Description. Male (holotype): Idiosoma small,
compact, 174 um long, 82 ~m wide, color semi-
transparent after a few weeks preservation in
fixative. Color in life unknown.

Dorsum (Fig. 7A) almost completely covered
with two dorsal plates which are uniformly punc-

Fic. 7. Actacarus karoensis sp. nov. Male (holotype): A, idiosoma (Ds); B, idiosoma (Vr); C, idiosoma (Lat, L); D,
genitoanal region; E, gnathosoma (Ds); F, gnathosoma (Vr); G, chelicera (R); H, palp (R). Female (allotype): I,
idiosoma (Vr); J, genitoanal region. Scale bars=20 um.

Two Actacarus Species from Japan

119

120 H. ABE

tate with fine sparse alveoli. Dorsal setae very
fine. AD and PD separated from each other by a
narrow strip of finely striated membranous cuticle,
opposing portion of AD and PD rectangular. AD
approximately 1/3 length of PD, L/W 0.66, mod-
erately convex anteriorly, very slightly concave
posteriorly, furnished with a large pore on antero-
lateral corner on each side, ornamented with
paired weak areolae and a number of laterally
scattered minute canaliculi at about mid-level. PD
124 wm long, 76 ~m wide, very weakly convex
anteriorly, narrow, concave and faintly ribbed
posteriorly, ornamented with tiny anterolateral
. areolae; a pair of weak intermediate panels and a
series of lateral minute canaliculi on each side; two
pores on each lateral margin; the first pore at 0.50
and the second at 0.88. OC (Fig. 7C) 15 um long, 8
ym wide, lateromarginally placed, lying at dorsal
side of anterior angle of PE, nearly triangular in
outline with pointed anterior end, furnished with a
small pore medially, a larger pore near post-
eroventral margin, and an oval apodeme at post-
erior end. A series of many microplatelets (Fig.
7C) lying marginally on each side of idiosoma
alongside AD and PD, of which three anterior and
two posterior microplatelets are relatively large
and elliptical; anterior three larger microplatelets
lying at level between OC and insertion of leg III,
all approximately equal in size (4 um long, 3 ~m
wide); two larger posterior microplatelets lying at
level of insertion of leg IV, each approximately 6
ym long, 3 um wide.

Chaetotaxy of dorsal region: Setae ds-i on PD at
0.43; ds-ii on the anteriormost larger micro-
platelets, seen as if placed on membranous cuticle
from dorsal view; ds-iii on PD, each separated
posteriorly from anterior margin of PD by approx-
imately four alveolar diameters; ds-iv on PD at
0.43; ds-v on PD, at 0.50; ds-vi (adanal setae) on
PD, each separated from posterior margin of PD
by about five alveolar diameters.

Venter (Fig. 7B) covered with four plates weakly
ornamented in a manner similar to dorsal plates.
AE 68 um long, 64 um wide, subrectangular in
outline, reaching posteriorly to level about midway
between insertions of legs III and IV, very slightly
concave posteriorly, furnished with tiny triangular
epimeral processes, with a thin membranous collar

anteriorly, and five sets of lateral and medial
subsurface pores of various shapes. PE (Fig. 7C)
82 «wm long, 24 um wide, elongate, tapering post-
eriorly from 0.57, moderately convex anteriorly,
and strongly outcurved ventrally, with bluntly
pointing terminal end, marked with a series of
marginal subsurface pores ventrally.

Chaetotaxy of epimeral region: Setae aes-i on
AE, each separated posteriorly from anterior mar-
gin of AE by bout six alveolar diameters; aes-ii-lat
on AE, posterior and somewhat medial to inser-
tion of leg III; aes-ii-v on AE, at level of insertion
of leg III; pes-iii-lat each on PE, on dorsolateral
margin of PE at level about midway between
anterior margin of PE and insertion of leg III;
pes-ili-v each on PE, at level slightly anterior to
insertion of leg III, separated from ventral margin
of PE by approximately six alveolar diameters;
pes-iv-a each on PE, at ventral apex of PE, sepa-
rated from margin by about three alveolar dia-
meters; aes-iv-p each on PE, at level anterior to
insertion of leg IV, separated from ventral margin
of PE by about three alveolar diameters.

Genitoanal region (Fig. 7B, D): GA 72 um long,
54 wm wide, almost truncated anteriorly, extend-
ing slightly anterior to level of pes-iv-a, almost
touching posterior margin of AE, moderately ex-
panded intermediately and narrowed posteriorly
to terminal end; two subsurface pores found anter-
olaterally and a cluster of lateral weak panels at
about mid-level on each side. Genital foramen
about 1/4 length of GA, L/W 1.50, pyriform in
outline; anterior margin of foramen at 0.60 relative
to GA length. Anal papilla terminally placed, well
separated from genital foramen. A terminal pit
(Fig. 7B) located laterally on each side of anal
papilla. Spermatophorotype (Fig. 7B) L/W 1.17,
massive, complex in structure, approximately 1/2
length of GA.

Chaetotaxy of genitoanal region: One pair of
outlying setae located at 0.42 on GA; a group of
eleven perigenital setae on each side of genital
foramen as illustrated in Fig. 7D; two subgenital
setae on each genital sclerite, arranged 1-1. No
setae on anal papilla.

Gnathosoma (Fig. 7E, F): 52 um long, 36 «m
wide, gnathosomal length/idiosomal length 0.30;
base of gnathosoma L/W 0.78, slightly expanded

Two Actacarus Species from Japan 121

laterally, lacking setae, entirely ornamented with
fine punctations, and with a few round panels on
dorsolateral and ventroproximal sites. Pharyngeal
plate fusiform, with eight visible panels. Anterior
margin of tectum weakly convex. Rostrum 24 um
long, 18 zm wide, subtriangular with round tip,
just reaching to level of distal end of P-2, bearing
four pairs of delicate filiform setae as follows:
Protorostral setae minute, near tip; deutorostral
setae short, posterior to protorostral setae; tritor-
ostral and basirostral setae long, approximately
four times as long as deutorostral setae, located at
0.14 and 0.21 relative to rostral length, respective-
ly. Rostral sulcus reaching to about 2/3 level
relative to rostral length. Chelicera (Fig. 7G) with

basal segment 30 um long, 14 um wide, strongly
convex anterodorsally, with oblique proximal end
and indistinct ornamentation. Movable digit
approximately 2/3 length of basal cheliceral seg-
ment, strongly inclined dorsally, with 16-18 minute
denticles along dorsal edge. Fixed digit nearly 1/2
length of movable digit. Palp (Fig. 7H) 48 um
long, inserted dorsolaterally, slightly inclined ven-
trally, with four free segments as described below:
P-1 short, cylindrical, L/W 1.25; P-2 longest and
robust, exceeding combined length of P-3 and P-4,
L/W 1.43, slightly expanded dorsoproximally,
ornamented with fine faint punctations and a few
faint panels, with one distidorsal filiform seta; P-3
about the same length as P-1, L/W 0.75, with one

Fic. 8. Actacarus karoensis sp. nov. Male (holotype): A, leg I (R); B, leg II (R); C, leg III (L); D, leg IV (L); E,
tarsus I (R); F, tarsus II (R); G, tarsus III (L); H, tarsus IV (L). Scale bars=20 um.

122 H. Asé

short spiniform seta anterodistally; P-4 conical,
slightly curved ventrally, furnished with three slen-
der filiform setae proximally, one slender filiform
seta and one short spiniform seta intermediately,
and one distal bacilliform seta parallel to terminal
blunt spiniform seta.

Legs (Fig. 8A-D): Length of legs I, II, 11, lV=
138, 112, 128, 130 um, respectively, thin, with fine
faint punctations on all segments. Each tarsus with
claw fossa. Lateral claws with indistinct accessory
processes. Median claw of each leg minute, biden-
tate only in leg I, unidentate in others. Carpite and
cavity in claw not clear. Parambulacral setae all
single euphathidia.

Leg chaetotaxy as follow: Trochanters I-IV, 0-
0-1-1; basifemora, 2—2—2-—2; telofemora, 2—2—2-
2; genua, 5—4—3-3; tibiae, 7-5-5-5. Tarsus I (Fig.
8E) with strongly developed posterior lamella,
with three dorsal setae (one intermediate seta on
basidorsal limb, others on claw fossa), one soleni-
dion, one famulus, three filiform ventral setae (one
intermediately, others distally), and two parambu-
lacral setae; solenidion fine setiform, at the base of
fossary lamella; famulus minute, blade-shaped,
with fine canaliculus, lying distally to solenidion;
lateral claws small compared with those on other
legs, and combs not in visible. Tarsus II (Fig. 8F)
with three dorsal setae (one filiform intermediate
seta on basidorsal limb, two filiform setae on claw
fossa), one distally swollen solenidion at the base
of fossary lamella, two parambulacral setae; lateral
claws with weakly developed combs. Tarsus III
(Fig. 8G) with four dorsal setae (one filiform in-
termediate seta, one filiform distal seta on basidor-
sal limb, and two filiform setae on claw fossa), two
parambulacral setae; lateral claws with well de-
veloped combs. Tarsus IV (Fig. 8H) with three
dorsal setae (one filiform distal seta on basidorsal
limb, two filiform setae on claw fossa), two para-
mbulacral setae; lateral claws as in tarsus III.

Female (allotype): Idiosoma 190 um long, 94
ym wide, resembling male in essential details
except for the characters of weak panels on PD
and genitoanal region. Dorsum: PD furnished
with only one weak panel near lateral margin on
each side.

Genitoanal region (Fig. 71, J): Genital foramen
26 wm long, 14 ~m wide, located terminally and

overlying anal foramen, furnished with three pairs
of perigenital setae; the first pair at 0.39, the
second at 0.69 (level of anterior margin of fora-
men), and the third at 0.84 (behind the foramen
occupying terminal concavity). Subgenital setae
lacking. Genital acetabula (Fig. 7J) internal, three
pairs. Ovipositor (Fig. 7I) long, tubular, extend-
ing to level near insertion of leg IV, slightly shifted
to one side from idiosomal longitudinal median
axis, although not reaching to lateral side of
idiosoma.

Immature stages: Three immature stages of A.
karoensis are distinguished. They differ from
adults in that 1) plates are less developed and more
widely separated from each other by distinctly
striated membranous cuticle, 2) smaller lateral
microplatelets are almost indistinct, 3) AD and PD
more ribbed and furnished with more canaliculi, 4)
PD lacks a cluster of weak panels at mid-level on
each side, and 5) legs are shorter and more weakly
punctate.

Deutonymph (paratype, Hal-63): Idiosoma 180
ym long, 84 um wide. Dorsum (Fig. 9A): Setae
ds-iii placed on striated membranous cuticle be-
tween AD and PD. PD truncated anteriorly,
extending to level of slightly posterior to ds-iii.
Venter (Fig. 9B): AE strongly narrowed posterior-
ly from level of insertion of leg II to level of
anterior to pes-iv-a. GA 62 um long, 52 um wide,
reaching anteriorly to level about midway between
pes-iv-a and pes-iv-p, furnished with two subsur-
face pores on each anterolateral corner, bearing
two pairs of perigenital setae; the first pair at 0.27;
the second at lateral sides of genital field. Primor-
dial genital slit very weakly sclerotized, reaching
anteriorly to 0.58 relative to GA length. Subgenit-
al setae lacking. Genital acetabula internal, two
pairs. Legs (Fig. 9C-F): Basifemur IV with only
one seta.

Protonymph (paratype, Hal-90): Idiosoma 160
yum long, 78 um wide. Dorsum (Fig. 10 A): Post-
erior margin of AD not reaching to mid-level
between insertions of legs II and III. Only two
lateral microplatelets located dorsoposterior to
OC. Venter (Fig. 10 B): Posterior margin of AE
extending to level about midway between pes-iii-
lat and pes-iv-p. Setae pes-iv-a lacking. GA 46 «m
long, 34m wide, reaching anteriorly to level

Two Actacarus Species from Japan 123

Fic. 9. Actacarus karoensis sp. nov. Deutonymph (paratype, Hal-63): A, idiosoma (Ds); B, idiosoma (Vr); C, leg I
(R); D, leg II (R); E, leg III (R); F, leg IV (R). Scale bars=20 pm.

slightly posterior to pes-iv-p, lacking setae, bearing
one pair of internal genital acetabula that flank the
primordial genital slit. Anterior margin of genital
slit at 0.61 relative to GA length. Legs (Fig.
10C-F): Basifemur III with only one seta, trochan-
ter IV without setae, femur IV undivided.

Larva (paratype, Hal-93): Idiosoma 140 um
long, 72 um wide. Dorsum (Fig. 11A): AD and
PD separated from each other by interval of
approximately 2/3 length of PD. Lateral pores on

AD and PD relatively large and distinct. Lateral
microplatelets very faint, lying dorsoposteriorly to
OC. Venter (Fig. 11B): AE lacking aes-ii-lat. PE
small, with only one seta and marginal subsurface
pores ventrally. Two subsurface pores located on
membranous cuticle between AE and PE. GA 22
ym long, 22 wm wide, trapezoidal, weakly prot-
ruding anteriorly, lacking both genital setae and
genital slit. Legs (Fig. 11C-E) three pairs;
basifemur and telofemur fused. Leg chaetotaxy as

124 H. ABE

Fic. 10. Actacarus karoensis sp. nov. Protonymph (paratype, Hal-90): ie idiosoma (Ds); B, idiosoma (Vr); C, leg I
(R); D, leg II (R); E, leg III (R); F, leg IV (R). Scale bars=20 ym.

follows: Trochanters I-III, 0-0-1; femora, 3-3-3;
genua, 4-4-3; tibiae, 5—5—S.

Morphological variation and abnormality: The
number of panels comprising a cluster at the
intermediate level on each side of PD was from
two to three in the male. The shape and the
arrangement of the lateral microplatelets were
variable even between sides of one specimen,

although the larger microplatelets were more
stable. Two male specimens of the type series had
only ten perigenital setae on one side of the genital
foramen.

Distribution: Tottori Prefecture, Sea of Japan,
Japan.

Remarks: Actacarus karoensis is distinguished
from congeners on the basis of the following

Two Actacarus Species from Japan 125

Fic. 11. Actacarus karoensis sp. nov. Larva (paratype, Hal-93): A, idiosoma (Ds); B, idiosoma (Vr); C, leg I (L); D,

leg II (L); E, leg III (R). Scal bars=20 pm.

characters: Setae ds-ii on lateral microplatelets;
three setae on PE (one dorsally, three ventrally);
GA expanded medially, furnished with one pair of
outlying setae and eleven pairs of perigenital setae
as illustrated in Fig. 7D; anterior margin of tectum
very weakly convex; ovipositor extending to level
of insertion of leg IV; a series of weakly sclerotized
lateral microplatelets laterad from AD and PD on
each side of idiosoma.

Morselli and Mari [14] mentioned that A.
clipeolatus may be distinguished from congeners
by the presence of two lateral platelets laterad
from AD and PD on each side of the idiosoma. In
the present study, however, both the Actacarus
species examined have the lateral platelets laterad
from AD and PD. Therefore, the presence of
these platelets cannot be regarded as a critical
specific character. It is even possible that the

existence of the lateral platelets or the lateral
microplatelets is one of the generic characters of
the genus Actacarus, and the shape and the size of
these platelets might be of specific significance.

The specific epithet is derived from the type
locality, “Karo”.

ACKNOWLEDGMENTS

The author wishes to express his appreciation to Dr.
Haruo Katakura (Hokkaido Univ.) for his valuable
advice and revision of the manuscript. Cordial thanks
also are due to Professor G. W. Krantz (Oregon State
Univ.) for supplying the author with his private collec-
tion from Schooner Creek, helpful suggestions on the
manuscript, and the correction of English. Dr. Nobuo
Tsurusaki (Tottori Univ.) kindly gave the author an
opportunity to examine material of value.

126

REFERENCES

Kishida, K. (1927) Ushio-dani (Halacarus spon-
giphilus Kishida). In “Nippon Doobutsu Zukan”.
Hokuryukan, Tokyo, p. 978 (in Japanese).
Bartsch, I. (1986) Zur gattung Agauopsis (Acari,
Halacaridae), Beschreibung zweier neuer Arten und
Ubersicht iiber Verwandtschaftsgruppen. Zool.
Scr., 15: 165-174.

Imamura, T. (1965) Hydrachnellae. In “Mites. An
Introduction to Classification, Bionomics and Con-
trol of Acarina”. Ed. by M. Sasa, Univ. of Tokyo
Press, Tokyo, pp. 29-30, 34, 216-251 (in Japanese).
Newell, I. M. (1947) A systematic and ecological
study of the Halacaridae of eastern North America.
Bull. Bingham Oceanogr. Collect. New Haven, 10:
1-232.

Newell, I. M. (1953) The natural classification of the
Rhombognathinae (Acari, Hfalacaridae). Syst.
Zool., 2: 119-135.

Newell, I. M. (1967) Abyssal Halacaridae (Acari)
from the southeast Pacific. Pac. Insects, 9: 693-708.
Newell, I. M. (1984) Antarctic Halacaroidea.
Antarc. Res. Series, 40: 1-284.

Newell, I. M. (1957) Studies on the Johnstonianidae

H. ABE

9

10

11

12

13

14

(Acari, Parasitengona). Pac. Science, 11: 396-466.
Newell, I. M. and Ryckman, R. E. (1966) Species of
Pimeliaphilus (Acari, Pterygosomidae) attacking in-
sects with particular references to the species para-
sitizing Triatominae (Hemiptera, Reduviidae). Hil-
gardia, 37: 403-436.

Krantz, G. W. (1976) Arenicolous Halacaridae
from the intertidal zone of Schooner Creek, Oregon
(Acari: Prostigmata). Acarologia, 18: 251-258.
Vorob’yeva, L. V. and Yaroshenko, N. A. (1979)
Halacaridae in the north-western Black Sea. Hydro-
biol. Jour., 15: 25-28 (originally in Russian, English
translation in 1980).

Krantz, G. W. (1974) Actacarus monniotae n. sp.
(=A. illustrans sensu Monniot 1968), an arenicolous
mite (Acari: Halacaridae) from the Mediterranean
region. Vie et Milieu, 24: 115-118.

Newell, I. M. (1951) Further studies on Alaskan
Halacaridae (Acari) . Amer. Mus. Novitates, 1536:
1-56.

Morselli, I. and Mari, M. (1986) Researches on the
coast of Somalia. The shore and the dune of Sar
Uanle. 39. On three interstitial species of Halacar-
idae (Acari). Italian Jour. Zool., N. S. Suppl., 21:
137-148.

ZOOLOGICAL SCIENCE 7: 127-132 (1990)

© 1990 Zoological Society of Japan

Comparative Study on LDH Isozymes in Different Subfamily of
Teleost Fish - Grass Carp (Ctenopharyngodon idellus) and
Blunt Snout-Bream (Megalobrama amblycephala)

Tao YunxiA! and YAN SHAoyI2

Institute of Developmental Biology, Academia Sinica,
Beijing, China

ABSTRACT— Electrophoretic, physico-chemical and immuno-analysis of lactate dehydrogenase isozy-
mes of grass carp (Ctenopharyngodon idellus) and blunt snout-bream (Megalobrama amblycephala)
indicated that although they were rather conservative in evolution, however, some divergences in their
gene activities and molecular structures were still remained. So, LDH isozyme can be used as a genetic

marker to distinguish these two kinds of fish.

INTRODUCTION

Markert and Faulhaber [1], using starch gel
electrophoresis, examined the LDH isozyme pat-
terns in 30 kinds of fish. They found out that the
LDH isozymes in fish were more complicated than
those in mammals and birds. One to twenty LDH
isozymes were found in different fish examined.
They were found distributed differently in various
tissues in different kinds of fish and were classified
into three systems. Among them a major LDH
isozyme system was distributed in most tissues
corresponding to the A and B gene systems of
mammals and birds. Two minor systems were
restricted to eye and gonads. Through elec-
tophoresis and immuno analysis, Shaklee et al. [2]
found that the two minor systems were probably
coded by the same gene locus, corresponding to
the C gene of mammals and birds. In comparison
of the kinetic properties, amino acid content and
electrophoretic zymograms of LDH isozymes of
Brook trout, lake trout and their hybrid, Splake
trout, Wuntch and Goldberg [3] found out that
there were some differences of all these properties.

Accepted March 20, 1989
Received September 6, 1988

! Present address: The Medical College of Pennsylvania,
3300 Henry Ave., Phila., Pa.19129, U.S.A.

? Requests for reprints should be addressed to Yan
Shaoyi.

Yan et al. [4] reported that the electrophoretic
zymograms of LDH isozymes can be used for
distiguishing four kinds of subfamily teleost fish in
the Family Cyprininae. This means that the LDH
isozymes not only behave differently in tissue
distributions, but also could be performed as a
species-specific marker in fish taxonomy.

For accurate comparative study of isozymes in
different organisms, purified LDH isozyme frac-
tions must be obtained. In 1967, Okabe et al. [5]
seperated and purified five human LDH isozyme
fractions by using the ammonium sulfate fractiona-
tion, calcium phosphate gel absorption and
DEAE-cellulose separation methods. In 1970,
Cuatrecasas [6] using a very simple and rapid
method—blue dextran affinity chromatography
separated and purified human LDH 1 and LDH 5.
Later, Fulton et al. [7] using HPLC separated five
LDH isozyme fractions in rat. But up to now, the
report about the LDH isozymes separation in fish
has not been seen.

In this investigation, we have tried to purify
LDH isozyme and analyze their divergences in
respect to their electrophoretic zymograms, im-
muno-properties, molecular structures as well as
physico-chemical properties in two kinds of fish
which belong to a different subfamily—grass carp
and blunt snout-bream. Since it has also been
known that the LDH 1 is a gene B product and
sometimes the divergence of gene B is larger than

128 T. YUNXIA AND Y. SHAOYI

gene A [8], so, the LDH 1 was preferentially to use
in this investigation.

MATERIALS AND METHODS

1. Experimental animals: Grass carp (Cte-
nopharyngoden idellas) and blunt snout bream
(Megalobrama amblycephala) were used for these
experiments. Both of them belong to the same
family (Cyprininae) but to different subfamilies
and genera. Grass carp belongs to the Subfamily
Leucinae and blunt-snout bream belongs to the
Subfamily Abramidinae. They were purchased
from Wan Quan Zhuang Fishery Farm, Beijing,
and were two years old.

2. Chemicals: Blue dextran and sepharose 4B
were purchased from Pharmacia Chemical Co.,
Sweden; Nitro blue tetrazolium (NBT) and Phena-
zine methosulfate (PMS) from Buchs Chemical
Co., Switzerland; Starch (Lot 387-1) from Can-
naught Laboratories Limited Co., Canada;
NADH from Boehringer Mannheim Chemical
Co., Western Germany: NAD from Yeast Plant of
Shanghai, China.

3. Preparation of tissue extracts: Tissues were
taken from freshly killed fish, washed with cold
0.75% saline, then homogenized and centrifuged
at 15,000 rpm (MSE-18) for 30 min. The super-
natants were used for starch electrophoresis.

4. Preparation of blood samples: blood was
collected from caudal vein, washed with 0.75%
saline containing heparin three times. The red
blood cells were hemolysed with 2 ml double
distiled water, then centrifuged at 3,000 rpm (k 70,
Eastern Germany). The supernatants were used
for electrophoresis and affinity chromatography
analysis.

5. Electrophoresis and specific stain: Vertical
starch gel electrophoresis and LDH isozymes
staining were carried out following Xue’s improved
method [9]. The starch gel electrophoresis was
carried out at 4°C for 16hr. And then the gels
were immersed in a specific staining solution at
37°C for 1-2hr. Every 100 ml staining solution
contains 50 mg NAD, 30 mg NBT, 2 mg PMS, 15
ml 0.5M Tris-HCl buffer, pH7.2, 10ml 1M
sodium lactate, and 5 ml 0.1 M NaCl.

6. Purification of LDH: LDH 1-5 and LDH 1

were purified by blue dextran affinity chroma-
tography mainly following Cuatrecasas’ method [6]
and only the NAD and NADH concentrations of
the eluted buffer were changed. The elution buffer
for LDH 1-5 was 0.35 mg/ml of NADH, 10 mM of
Tris, 0.5mM of mercaptoethanol, pH 8.6. The
elution buffer for LDH 1 was 0.05 mg/ml of NAD,
0.1 mg/ml of lithium lactate, 10 mM of Tris, 0.5
mM mercaptoethanol, pH 8.6.

7. Assay for LDH activity and purity: LDH
activity was determined spectrophotometrically by
monitoring the formation of NADH at 340 nm in 1
cm quartz cuvettes following the procedure of
Holmes et al. [10]. LDH purity was determined by
enzyme specificity staining with the starch gel, and
by measuring the international units (I.U.) per
milligram of the extracted LDH. LDH concentra-
tion was determined as protein concentration by
the method of Lowry et al. [11].

8. Amino acid content assay: About 0.5 mg of
enzyme were hydrolyzed in 6 N HCl at 110°C for
24 hr. The resultant hydrolysate was washed and
evaporated to dryness and then resuspended in 0.2
ml of double distilled water. Amino acid content
was determined by HPLC (Waters Company,
Model AAA) according to the ion exchange
separation method described in Waters Associates
Operator’s Manual [12].

9. Kinetic parameter determination: The Km
value of LDH 1 was calculated from Lineweaver-
Burk plots, with sodium lactate as substrate at pH
8.6 and temperature 25°C. Kinetic of heat activa-
tion: The LDH 1 activity was measured at the
temperatures ranging from —4°C to 80°C, at pH
8.6. Kinetics of acid and alkali treatment: The
LDH 1 activity was measured in the pH range of 6
to 13, at temperature 25°C.

10. Preparation of antibody and investigation
on LDH immuno properties: Rabbits were immu-
nized with purified LDH 1-5 isozymes prepared
from blunt snout-bream red blood cells following
the method of Clausen [13]. A mixture of grass
carp and blunt snout-bream LDH 1-5 isozymes
and antisera were incubated for 30 min at 30°C
prior to starch gel electrophoresis and LDH stain-
ing for neutralization or inhibition tests. The
immunoprecipitated bands left in the double im-
muno diffusion agar gels were oberved.

LDH Isozymes in Fish Taxonomy 129

RESULTS

LDH isozymes electrophoresis zymograms

The zymograms of LDH isozymes from different
tissues of grass carp and blunt snout-bream were
displayed by starch gel electrophoresis and they
are shown in Figure la and 1b. Among them, in
the tissues of cardiac muscle, skeletal muscle and
eye of both kinds of fish, there existed five LDH
isozyme bands, migrating towards the anode, and
in liver, there was the C band migrating towards
the cathode in both kinds of fish. In kidney tissue
there were still five bands in the sample of blunt
snout-bream, but seven bands in that of grass carp.
The two ‘additional bands’ also migrated towards
the anode in between LDH 2 to LDH 3 and LDH 3
to LDH 4.

Fic. 1.
grass carp(Ct) and blunt snout-bream(Me) revealed

LDH isozyme zymograms of various tissues of

by starch gel electrophoresis. Ct(a), Me(b). 1.
Cardiac muscle, 2. Skeletal muscle, 3. Eye, 4.
Kidney, 5. Liver. Arrows show 2 additional bands.

Purification of LDH 1-5 and LDH 1 from blood
cells

The optimum concentration of NADH for
separating LDH 1-5 was 0.39 mg/ml and the opti-
mum concentration of NAD for separating LDH 1
is 0.05 mg/ml. Figure 2a and 2b shows the purified
LDH 1-5 and LDH 1 isozymes of grass carp and
blunt snout-bream. The specific activities of these
purified LDH isozymes were shown in Table 1 by
calculating the international units (I.U.) per milli-
gram for purified LDH protein. The data indi-
cated that all the purified LDH isozymes were over
500 1.U. per milligram.

- iby

Fic. 2. Purified LDH 1-5 (left) and LDH, (right) ISOZy-
me zymograms of red blood cell of grass carp(Ct)
and blunt snout-bream(Me) revealed by starch gel
electrophoresis. Ct(a), Me(b).

TABLE 1. The specific activities of LDHI-5
and LDHI isozymes extracted and
purified from the red blood cells of grass

carp (Ct) and blunt snout-bream (Me)
_——— EE eee

I.U./mg
4O.D. 340 20*
Fish LDH 6.2m
Ct LDHI-5 510
Me LDHI-5 1035
Ct LDHI 515
Me LDHI 1225

—— ee eee
* Average value of three times of measurement.

Immuno properties

The antiserum against blunt snout-bream red
blood cell LDH 1-5 isozyme were used. After the
double diffusion on agar gel with blood sample of
grass carp and blunt snout-bream, a complete
crossing precipitation line was clearly observed
(Fig. 3). When the antisera were mixed with the
blood samples of both fish, all the activities of
LDH isozymes disappeared. When the antisera
were mixed with the extracts from livers and
kidneys of the two fish, it was found that the C
band remained in both fish, but all the other
bands, including two ‘additional bands’ of grass
carp disappeared in kidney extracts (Fig. 4a and
4b).

130 T. YUNXIA AND Y. SHAOYI

Fic. 3. Double diffusion precipitation lines of the red
blood cell LDH 1-5 of Grass carp(A), and Blunt
snout-bream(B) aganist the antiserum of red blood
cell LDH 1-5 of Blunt snout-bream(C).

LDHe a ~ LDHc
LDHs @ w PTR Ga

= cd - LDH

, bd

-” He

™ aes:

> = LDH]

LDHy -

Fic. 4. Starch gel electrophoresis zymograms show, af-
ter added the antiserum of blunt snout-bream(Me)
blood cell LDH 1-5 to the grass(Cp) and blunt
snout-bream(Me) kidney(middle) and liver(right)
extracts, the LDHc bands of liver of both fish are
still remained, but all the kidney bands disappeared.
Both left rows show the Ct and Me kidney LDH 1-5
zymograms without antiserum treatments. A addi-
tional bands of Ct kidney LDH isozyme are also
showed by arrows.

Physico-chemical properties of LDH I isozyme

The starch gel electrophoresis showed that the
LDH 1 isozymes of both grass carp and blunt

snout-bream migrated towards the anode but they
had different mobility. The LDH 1 isozyme of
blunt snout-bream carried more positive charges
than the grass carp LDH 1 did. The blue dextran
affinity chromatography showed that both of them
almost had the same affinity to the blue dextran
and could be released from the blue dextran by
0.05 mg/ml of NAD. The optimum pH value,
optimum temperature of reaction and denaturing
concentration by urea of LDH 1 isozyme are also
the same in both fish except their Km values
remained different. The Km value of Grass carp
LDH 1 is 1.1107! and the km value of blunt
snout-bream LDH 1 is 5.6X10~*.

The amino acid content of LDH 1 of the two fish
are shown in Table 2. It can be seen that some
kinds of amino acid content are different in the
LDH 1 isozyme of the two fish, i.e. grass carp
LDH 1 has more val and blunt snout-bream LDH
1 has more lys and arg.

DISCUSSION

In vertebrates, the LDH isozymes exist as tetrad
forms resulting from the random polymerization of
different peptides, and having different distribu-
tion in different tissues [1]. Obviously, this diffe-
rent distribution is due to the different expression
of genes. Comparative studies on fish LDH isozy-
mes in earlier years were carried out mainly by the
methods of electrophoresis and they only provided
limited evidence for deducing its molecular struc-
ture and enzymatic properties. In this paper, some
evidences obtained from the kinetic, immuno and
amino acid analysis of purified LDH isozymes
were observed for indicating the divergences be-
tween LDH isozymes of grass carp and blunt
snout-bream.

The results of starch gel electrophoretic zymo-
grams indicated that five LDH isozymes existed in
most tissues of both grass carp and blunt snout-

TABLE 2. Amino acid contents of red blood cells LDHI of grass carp (Ct) and blunt snout-bream (Me)
Fish LDH Amino acid content
Asp Thr Ser Glu Pro Gly Ala Val Met Ile Len Phe His Lys Arg Tyr
Ct LDH | 116 80 108 132 68 108 144 248 12 28 48 12 «40 80 44 0
Me LDH | 132 88 124 148 80 124 164 100 12 20 = 28 8 44 136 92 0

LDH Isozymes in Fish Taxonomy 131

bream, and in liver tissue of both fish, there existed
the cathode bands-LDHc. Whitt et al. [14] noted
that the B gene of LDH isozyme is produced by
the duplication of A gene, and the C gene is
produced by the duplication of B gene subsequent-
ly. Odense et al. [15] also found out that, in the
evolution of fish, the further LDH gene duplica-
tions also existed. For example, in carp, apart
from the existence of A, B and C gene loci, there
were also B’ and C’ gene loci. But in grass carp
and blunt snout-bream, it seems that a typical
major LDH isozyme system encoded by A and B
gene loci as well as a minor LDH isozyme system
coded by a C gene locus were observed. Carp,
grass carp and blunt snout-bream belong to same
family, Cyprininae. However, as compared with
carp, no B’ and C’ gene duplication could be found
in either grass carp or blunt snout-bream. In
consideration of the chromosome number differ-
ences existing in carp (2n=100), grass carp (2n=
48) and blunt snout-bream (2n=48), it could be
explained that the less chromosome numbers of
grass carp and blunt snout-bream may decrease the
possibility of gene duplications on those fish as
compared with the carp. However, in the kidney
tissue of grass carp, there were seven LDH isozy-
me bands while in the blunt snout-bream kidney
tissue only five LDH isozyme bands were found.
Although, at present, it is not clear how the two
‘additional bands’ of LDH isozyme in grass carp
kidney tissue arose, we believe that some minor
divergence in the LDH isozymes occurred during
the divergent evolution of both fish, even though
they have the same chromosome number.

The results obtained from physico-chemical
properties analysis of the B gene product—LDH 1
of both fish show that they have almost the same
blue dextran affinity, same optimum pH value,
same optimum temperature of reaction as well as
the same concentration of urea for denaturation
except their Km value are different which indicates
that an enzymatic property difference exists in the
two fish. Table 2 also shows that some amino acid
content of LDH 1 have changed in the two fish, i.e.
grass carp has more val and blunt snout-bream has
more lys and arg. This means that minor molecu-
lar structural differences also exist in the LDH
isozymes of both fish.

The results of immuno-experiments show that
the antiserum against LDH 1-5 isozymes of blunt
snout-bream red blood cell not only can precipitate
the red blood cell LDH 1-5 isozymes of grass carp
and blunt snout-bream but also can neutralize the
LDH 1-5 isozyme components of different tissues,
for example, the LDH 1-5 isozymes of red blood
cell, liver and kidney in both fish even including
the two ‘additonal bands’ of grass carp kidney
LDH isozyme. However, it can not neurtalize the
LDH C isozyme either in grass carp or in blunt
snout-bream liver tissue. It means that the LDH
1-5 isozymes, the products of A and B gene, in
grass carp and blunt snout-bream have a very
common immuno property and the cathod band of
liver LDH isozyme components in both fish is the
product of C gene [14]. It can be also proposed
that the two ‘additional bands’ of grass carp kidney
LDH 5 isozyme might be recognized as the sub-
bands of its LDH 3 and LDH 4 components, as
revealed by electrophoretic zymograms, rather
than as the products of other genes, because they
can also be neutralized by the antiserum against
the purified LDH 1-5 isozyme of blunt snout-
bream.

It can be concluded that the LDH isozymes of
grass carp and blunt snout-bream have many simi-
larities in general, but some divergeneces were
observed in gene activities, molecular structures as
well as some physico-chemical properties accord-
ing to our above experiments. Therefore, it was
confirmed that, as revealed by other authors in fish
[3, 4], birds and mammals [1], the LDH isozyme
might be used as a genetic maker to distinguish
grass carp and blunt snout-bream in addition to
their morphological criteria in taxonomy.

ACKNOWLEDGMENTS

The authors wish to thank Professor Huang Gefang of
the Institute of Developmental Biology, the Chinese
Academy of Sciences, Beijing, China for his reading of
this English maniscript. Thanks are also given to Profes-
sor Kenjiro Yamagami of the Life Science Institute,
Sophia University, Japan and Professor Yoshitaka Naga-
hama of the National Institute for Basic Biology, Okaza-
ki, Japan for their valuable comments and advices of this
maniscript.

The present study was supported by the Important
Research Project Grant of the Chinese Academy of

182;

Sciences and the RF 84031 Grant of the Rockefeller
Foundation, USA.

REFERENCES

Markert, C. L. and Faulhaber, I. (1965) Lactate
dehydrogenase isozyme patterns of fish. J. Exp.
Zool., 159: 319-322.

Shaklee, J. B., Kepes, K. L. and Whitt, G. S. (1973)
Specialized lactate dehydrogenase isozyme: the
molecular and genetic basis for the unique eye and
liver LDHs of teleost fishes. J. Exp. Zool., 185:
217-240.

Wuntch, T. and Goldberg, E. (1970) A comparative
physicochemical characterization of lactated dehyd-
rogenase isozymes in brook trout, lake trout and
their hybrid splake trout. J. Exp. Zool., 174: 233-
252.

Yan, J., Wang G. and Yan, S. (1986) Analysis of
starch gel electrophoresis patterns of hemoglobin
and red blood cell LDH isozymes of four kinds of
fresh water teleost (Vylopharyngodon piceus, Cte-
nopharyngodon idellus, Hypophthalmichtys molit-
rix, Aristichys nobilis). Hereditas (Beijing), 8: 25—
ip

Okabe, K., Hayakawa, Hamada, M. and Koike, M.
(1968) Purification and comparative properties of
human lactate dehydrogenase isozymes from uterus,
uterine myoma and cervical cancer. Biochemistry, 7:
79-90.

Cuatrecasas, P. (1970) Protein purification by affini-
ty chromatography. J. Biol. Chem., 245: 3059-3065.
Fulton, J. A., Schlabach, T. D., Kerl, J. E., Toren,

10

11

12

13

14

15

T. YUNXIA AND Y. SHAOYI

E. and Clifford, J. (1979) Dual-detector-post-
column reactor system for the detection of isozymes
separated by high-performance liquid chromatogra-
phy. II. Evaluation and application to lactate dehyd-
rogenase isozymes. J. Chromatogr., 175: 283-291.
Pesce, A., McKay, R. H., Stolzenback, F. R. D.
and Kaplan, N. O. (1964) The comparative en-
zymology of lactate dehydrogenases. I. Properties of
the crystalline beef and chicken enzyme. J. Biol.
Chem., 239: 1753-1761.

Xue, G. (1978) Lactate dehydrogenase isozyme.
Shengwu Kexue Dongtai, 3: 10-17.

Holmes, R. S. and Soopes, R. K. (1974) Im-
munochemical homologies among vertebrate lactate
dehydrogenase isozymes. Eur. J. Biochem., 43:
167-177.

Lowry, O. H., Rosebrough, N. J., Farr, A. L. and
Randall, R. J. (1951) Protein measurement with the
phenol reagent. J. Biol. Chem., 193: 265-275.
Amino Acid Analysis System, Operator’s Manual.
(1983) Waters Associates Publications, Milford,
MA, U.S.A. pp. 4-14-11.

Clausen, J. (1981) Immunochemical Techniques for
the Identification and Estimation of Macro-
molecules. Elsevier North-Holl and Biomedical
Press, Amsterdam.

Witt, G. S., Shaklee, J. B. and Markert, C. L.
(1975) Evolution of the lactate dehydrogenase
isozymes of fishes. Isozymes, 4: 381-400.

Odense, P. H., Allen, T. M. and Leung, T. C.
(1966) Multiple forms of lactate dehydrogenase and
aspartate aminotransferase in herring (Clupea
harengus L.) Can. J. Biochem., 44: 1319-1326.

ZOOLOGICAL SCIENCE 7: 133-140 (1990)

© 1990 Zoological Society of Japan

The Drosophila robusta Species-group (Diptera: Drosophilidae)
from Yunnan Province, Southern China, with the Revision
of its Geographic Distribution

Hipe-Aki WATABE, XING CHal LiANG! and WEN XIA ZHANG!

Biological Laboratory, Sapporo College, Hokkaido University of Education,
Sapporo 002, Japan and ‘Kunming Institute of Zoology,
Academia Sinica, Kunming, Yunnan, China

ABSTRACT—Two known and four new species of the Drosophila robusta species-group are reported
from Yunnan, southern China, and the geographic distribution of the group is revised.

INTRODUCTION

The present paper deals with two known and
four new species of the Drosophila (Drosophila)
robusta species-group, in Yunnan Province, south-
ern China.

All the holotypes and a part of paratypes are
deposited in the Kunming Institute of Zoology,
Academia Sinica, Kunming, China, and the re-
maining paratypes in the Biological Laboratory,
Hokkaido University of Education, Sapporo,
Japan.

COLLECTION SITES AND METHODS

Dali district covering collection sites of Xian-
guan, Dabochin and Butterfly-spring is about 2300
meters above the sea level in northern parts of
Yunnan Province, Anning and Kunming in the
center of Province, and Simao in subtropical cli-
mate. Most of drosophilid flies described here
were collected in watersides by using cup-traps
baited with fermenting bananas.

D. ROBUSTA SPECIES-GROUP

D. robusta species-group: Sturtevant, 1942, Univ.
Texas Publ., 4213: 31.

Accepted April 11, 1989
Received February 6, 1989

Diagnosis. Dark brown or black species with 2
pairs of dorsocentrals, body length ca. 3.5-4.0 mm
(ca. 2.5mm in D. cheda Tan et al., 1949). Palpus
with several long bristles besides numerous tiny
hairs. Acrostichal hairs in 6 regular rows. Preapic-
als on all three tibiae; apicals on fore and mid
tibiae. Wing hyaline, slightly fuscous. Veins dark
brown; crossveins clear. Rj+3 straight; R4+5 and
M parallel. C, bristles 2, subequal. Cercus fused
to pubescent epandrium. Aedeagus curved ven-
trally. Anterior paramere rudiment or absent,
posterior paramere absent.

Drosophila (Drosophila) lacertosa Okada

Drosophila (Drosophila) lacertosa Okada, 1956,
Syst. Study, 158.

Specimens examined. China: 1 #, 12, Anning,
15. X. 1987, 309%, 289, Xianguan, 18. X. 1987,
49, 82, Butterfly-spring, 17-18. X. 1987, 19,
Dabochin, 22. IX. 1988, 14 g, 182, Kunming, 16.
IX. 1987, 3 7’, Simao, 4. XI. 1987. Collectors: H.
Watabe and X. C. Liang.

Distribution. Japan, Korea, India, Nepal, Bur-
ma; China: Taiwan, Guangdong, Yunnan (n.
loc.).

Remarks. In Kunming and Dali, D. lacertosa
has been collected in abundance not only in water-
sides but also in restaurants and kitchens as a
domestic species.

The color patterns of abdominal tergites are

134 H. WatasBE, X. C. LIANG AND W. X. ZHANG

quite variable; yellowish brown with black caudal
bands in most specimens [1], but entirely black in
Simao specimens.

Drosophila (Drosophila) neokadai
Kaneko et Takada

Drosophila (Drosophila) neokadai Kaneko et
Takada 1966, Annot. Zool. Japon., 39:55.

Specimens examined. China: 23, Dabochin,
21. IX. 1988; 1h, 1%, Xianguan, 22. IX. 1988.
Collectors: H. Watabe and X. C. Liang.

Distribution. Japan, China (n. loc.): Yunnan.

Remarks. Tis species is related to the following
new species D. gani in the external morphology
and in the shape of aedeagus, but distinguishable
from the latter by the shapes of surstylus and
spermatheca [2].

Drosophila (Drosophila) gani

Liang et Zhang, sp. nov.
(Figs. 1-6)

Diagnosis. Body black, largest in this species-
group. Arista with ca. 4 upper and ca. 2 lower
branches. C index ca. 4.5, C3-fringe ca. 3/4. Epan-
drium posteriorly pubescent except lower portion
(Fig. 1). Surstylus rectangular, roundish on lower
margin (Fig. 2). Spermatheca large, with apical
hollow (Fig. 6).

JS, %. Body length ca. 4.25 mm (range: 3.7—
4.8), thorax length (including scutellum) ca. 1.88
mm (1.8-1.9), wing length ca. 4.40 mm (4.1—4.6).

Head: Eye dark red, with thick piles. Second
joint of antenna dark gray, with 2 stout setae; 3rd
black, with numerous tiny hairs. Arista with ca. 4
(3-4) upper and ca. 2 lower branches in addition to

10

Fics. 1-6. Drosophila (Drosophila) gani Liang et Zhang, sp. nov. 1: Periphallic organs. 2: Surstylus. 3: Phallic
organs. 4: Aedeagus (lateral view). 5: Ovipositor. 6: Spermatheca.

Fics. 7-12. Drosophila (Drosophila) yunnanensis Watabe et Liang, sp. nov. 7: Periphallic organs. 8: Surstylus. 9:
Phallic organs. 10: Aedeagus (lateral view). 11: Ovipositor. 12: Spermatheca. Signs: a, anterior paramere; c,
surstylus; e, aedeagus; n, novasternum; 0, aedeagal apodeme; t, cercus; v, ventral fragma. Scale-line=0.1 mm.

D. robusta Group from China 135

a moderate terminal fork. Frons dark brown, ca.
0.50 (0.47-0.53) as broad as head, anteriorly with
sparce frontal hairs. Anterior reclinate orbital
(Orb 2) ca. 0.38 (0.30-0.47) length of posterior
reclinate orbital (Orb 1); proclinate orbital (Orb 3)
ca. 0.56 (0.47—0.67) length of Orb 1. Face reddish
brown; carina brown, darker on margin, high,
wider below. Clypeus blackish brown. Cheek
brown, ca. 0.21 (0.17—0.26) as broad as maximum
diameter of eye, with ca. 3 bristles at lower corner.
Second oral (Or 2) ca. 0.72 (0.64—-0.78) length of
vibrissa (Or 1); third oral (Or 3) ca. 4/9 length of
Or 2. Palpus grayish brown, flattened laterally,
with % ca. 7 long bristles and ? ca. 4 bristles.

Thorax: Mesoscutum dark brown, with a longi-
tudinal darker stripe running to scutellum in mid-
dle and 1 pair of obscure stripes along notopleural
region. Scutellum dark brown; its lateral sides
black. Thoracic pleura dark brown. Lower
humeral ca. 0.83 (0.78-0.91) length of upper one.
Length distance of dorsocentrals ca. 0.64 (0.60-
0.72) cross distance; anterior dorsocentral (DcA)
ca. 0.71 (0.68-0.73) length of posterior dor-
socentral (DcP). Anterior scutellars (SctAs) near-
ly parallel and posterior scutellars (SctPs) conver-
gent. SctA ca. 1.03 (0.98-1.10) length of SctP;
distance between SctPs ca. 0.50 (0.44—0.53) dis-
tance between SctAs. Relative length of anterior/
posterior sternopleural (Sterno-index) ca. 0.70
(0.53-0.78).

Legs brown; fore femur posteriorly with ca. 4
long bristles. Number of small stout bristles on 3rd
costa (3Cfr) ca. 33 (28-36). Wing indices: C ca.
4.47 (3.54-5.21), 4V ca. 1.48 (1.43-1.56), 4C ca.
0.52 (0.42-0.64), 5x ca. 1.07 (0.92-1.17), Ac ca.
1.53 (1.17-2.00), C3-fringe ca. 0.75 (0.73-0.77).
Haltere whitish yellow, basally darker on anterior
margin.

Abdomen: Tergites blackish brown, slightly
paler in middle of 2nd to 6th tergites. Sternites
dark brown; ¥ Sth large, rectangular, with ca. 11
long and ca. 43 short bristles.

Periphallic organs (Figs. 1 and 2): Epandrium
dark brown, paler on lower half, with ca. 5 long
bristles on upper half and ca. 18 bristles on middle
to lower half. Surstylus pale brown, with ca. 9
primary teeth and ca. 3 bristles; basal portion
connected to epandrium very narrow. Decaster-

num pale yellow, darker on margin, nearly qua-
drate. Cercus dark brown, oval, entirely pubes-
cent, with ca. 30 long bristles.

Phallic organs (Figs. 3 and 4): Aedeagus orange,
bilobed, ventrally curved moderately; apodeme
ca. 1/4 as long as aedeagus. Anterior paramere
pale brown. Novasternum yellowish brown, nearly
triangular, with sparce tiny spines in middle of
outer surface but without submedian spines.

reproductive organs (Figs. 5 and 6): Lobe of
Ovipositor brown, with ca. 5 light orange discal
teeth and ca. 21 orange marginal teeth which
gradually decrease in size. Spermatheca dark
brown, cylindric in lateral view, basally wrinkled;
introvert deep; inner duct expanding at 1/4 portion
from tip.

Holotype jg, China: Xianguan, Yunnan Pro-
vince, 19. IX. 1988. Collector: X. C. Liang.

Paratypes, China: 2 J\, same data as holotype
except 18. X. 1987, 12, Dabochin, Yunnan Pro-
vince, 21. IX. 1988. Collector: X. C. Liang.

Distribution. | China: Xianguan, Dabochin,
Kunming, Yunnan Province, and Meitan, Guichou
Province (Prof. Gan, personal. comm.). Dro-
sophila gani is the same species as Watabe and
Nakata [3] described as D. sp. 2, which was
collected in Tsugaru district of Honshu Is., north-
ern Japan.

Relationships: This species is related to D.
pullata Tan et al. [4] in the external morphology,
but distinguishable from the latter species by the
diagnostic characters.

Remarks: This species is dedicated to Prof. Yun
Xing Gan (Kunming Institute of Zoology,
Academia Sinica), who has introduced the descri-
bers (X. C. L., W. X. Z) into the field of dipteran
taxonomy.

Drosophila (Drosophila) yunnanensis
Watabe et Liang, sp. nov.
(Figs. 7-12)

Diagnosis. Arista with ca. 4 upper and ca. 2
lower branches. C index ca. 4.6, C3-fringe ca. 5/9.
Tergites with black caudal bands interrupted at
middle. Lower margin of epandrium anteriorly
convexed, posteriorly rounded (Fig. 7). Aedeagus
ventrally curved heavily (Fig. 10). Lobe of ovipo-

136 H. WataBeE, X. C. LIANG AND W. X. ZHANG

sitor broaden caudodorsally (Fig. 11).

dS, ?-. Body color blackish brown. Body length
J ca. 3.93 mm (3.6-4.2), 2 ca. 4.22 mm (3.3-
4.8). Thorax length # ca. 1.63 mm (1.5-1.8), ?
ca. 1.78mm (1.7-1.8). Wing length ca. 3.68
mm (3.4-3.9), 2 ca. 4.22 mm (3.4—4.8).

Head: Eye dark red, with thick piles. Antenna
dark brown: 2nd joint with 2-3 stout setae; 3rd
with numerous tiny hairs. Arista with ca. 4 (4-5)
upper and ca. 2 lower branches in addition to a
terminal fork. Frons blackish brown, { ca. 0.47
(0.44-0.51), 2 ca. 0.51 (0.50-0.52) as broad as
head, anteriorly with a few frontal hairs. Orb 2 ca.
0.36 (0.24—-0.53) length of Orb 1; Orb 3 ca. 0.69
(0.48—0.84) length of Orb 1. Face reddish brown;
carina dark brown, wider below. Cheek light
brown, ca. 0.18 (0.15—0.22) as broad as maximum
diameter of eye. Or 1 long and stout; Or 2 thin, ca.
0.58 (0.42—0.75)length of Or 1; Or 3 ca. 1/2 length
of Or 2. Palpus dark brown, with ca. 2-3 long and
a few of middle bristles.

Thorax: Mesoscutum dull brown, with 3 longi-
tudinal darker stripes; wide stripe in middle, post-
eriorly bifurcated, a lateral pair of stripes just
outside dorsocentrals, interrupted at transverse
suture. Scutellum dark yellow, medially with a
broad brown stripe running from mesoscutum.
Thoracic pleura dark brown. Humeral plate pale
yellow, with 2 humerals; lower one ca. 0.71 (0.65-
0.87) length of upper one. DcA ca. 0.58 (0.51-
0.65) length of DcP; length distance of dor-
socentrals ca. 0.45 (0.41-0.55) cross distance.
SptAs nearly parallel; SctPs convergent. SctA ca.
0.89 (0.77-0.98) length of SctP; distance between
SctPs ca. 0.41 (0.36-0.47) distance between SctAs.
Sterno-index ca. 0.69 (0.47-0.92).

Legs dark brown; fore femur anteriorly with 1
long bristle, posteriorly with ca. 4 bristles. Num-
ber of 3Cfr ca. 20 (16-25). Wing indices: C # ca.
4.55 (4.23-4.95), ? ca. 4.66 (4.36-5.43), 4V ff
ca. 1.59 (1.45-1.79), & ca. 1.62 (1.47-1.72), 4C
ca. 0.57 (0.47-0.63), 5x f ca. 1.51 (1.33-1.65), $
ca. 1.31 (1.14-1.58), Ac # ca. 1.63 (1.50-1.78),
? ca. 1.58 (1.40-1.90), C3-fringe ca. 0.57 (0.44-
0.65). Haltere yellowish white.

Abdomen: Tergites dark brown, with black
caudal bands interrupted at middle; some speci-
mens have entirely black tergites owing to mela-

nization by low temperatures. Sternites brown;
Sth rectangular, posteriorly concaved slightly, with
ca. 104 bristles; ? 6th quadrate, with ca. 34
bristles.

Periphallic organs (Figs. 7 and 8): Epandrium
posteriorly pubescent except lower half, with ca. 7
long bristles on upper half and ca. 17 bristles on
middle to lower margin. Surstylus distally con-
caved, with ca. 10 apically pointed primary teeth
and ca. 8 bristles. Decasternum pale yellow,
darker on dorsal margin, dorsally rounded and
caudally broaden. Cercus pubescent, with ca. 40
long bristles; caudoventral corner somewhat
pointed, with tuft of several short bristles.

Phallic organs (Figs.9 and 10): Aedeagus
robust, broaden at tip, distally 1/3 bilobed; tip
brownish orange, pointed like fook; apodeme ca.
2/5 as long as aedeagus. Novasternum broad, with
1 pair of prominent submedian spines on inner
margin.

? reproductive organs (Figs. 11 and 12): Lobe
of ovipositor light orange, apicocaudally broaden,
with ca. 5 discal teeth and ca. 21 marginal teeth;
ultimate marginal tooth large, ca. 2 times as long
as penultimate. Spermatheca dark brown, oval,
basally narrowing and wrinkled, with small dark
patches on outer surface of capsule, without apical
indentation; inner duct narrowed just below tip.

Holotype jg, China: Dabochin, Yunnan Pro-
vince, 21. IX. 1988. Collector: X. C. Liang.

Paratypes, China: 5, 52, same data as
holotype.

Distribution. Widely distributed from central to
northern parts of Yunnan Province (thus species
name).

Relationships. D. yunnanensis is closely related
to D. lecertosa in the external appearance, but
distinguishable from the latter species by the
shapes of aedeagus and spermatheca.

Remarks. In the robusta group, novasternum
with submedian spines has been found only in D.
lacertosa, and so D. yunnanensis is the second
species having such novasternum.

Drosophila (Drosophila) bai
Watabe et Liang, sp. nov.
(Figs. 13-20)

D. robusta Group from China 137

Fics. 13-20. Drosophila (Drosophila) bai Watabe et Liang, sp. nov. 13: Antenna. 14: Palpus. 15: Periphallic
organs. 16: Surstylus. 17: Phallic organs. 18: Aedeagus (lateral view). 19: Ovipositor. 20: Spermatheca.
Fics. 21-26. Drosophila (Drosophila) medioconstricta, Watabe, Zhang et Gan, sp. nov. 21: Periphallic organs. 22:

Surstylus. 23: Phallic organs. 24: Aedeagus (lateral view). 25: Ovipositor. 26: Spermatheca. Signs and scales as

in Figs. 1-12.

Diagnosis. Arista with ca. 4 upper and ca. 1
lower branches (Fig. 13). Palpus with 2 long
bristles at tip besides numerous tiny hairs (Fig.
14). Or 2 thin, ca. 1/3 length of Or 1. C index ca.
4.5, C3-fringe ca. 4/9. Primary teeth on surstylus
sparce, usually separated into two parts (Fig. 15).
Aedeagus nearly rectangular (Fig. 18). Lobe of
ovipositor slender (Fig. 19); spermatheca less-
sclerotized (Fig. 20).

dé’, *. Body color black. Body length ¥ ca.
3.35mm (3.2-3.6), 2 ca. 3.80mm (3.44.1).
Thorax length ~ ca. 1.48 mm (1.4-1.6), 2 ca.
1.61 mm (1.5-1.8). Wing length # ca. 3.83 mm
(3.8-3.9), 2 ca. 4.25 mm (4.0-4.6).

Head: Eye dark red, with thick piles. Antenna
black: 2nd joint with 2 stout setae; 3rd joint with
numerous tiny hairs. Arista with ca. 4 (3-5) upper
and ca. 1 (0-2) lower branches in addition to a

small terminal fork. Frons black, ca. 0.50 (0.43-
0.52) as broad as head, anteriorly with a few
frontal hairs. Orb 2 ca. 0.34 (0.25-0.41) length of
Orb 1; Orb 3 ca. 0.68 (0.58-0.91) length of Orb 1.
Face brown; carina high, wider below. Cheek dark
brown, ca. 0.20 (0.13-0.27) as broad as maximum
diameter of eye. Or 2 thin, short, ca. 0.36 (0.22-
0.43) length of Or 1; Or 3 subequal to Or 2. Palpus
dark brown, club-shaped.

Thorax: Mesoscutum dark brown, medially with
an obscure longitudinal darker stripe; scutellum
blackish brown. Thoracic pleura black. Lower
humeral ca. 0.63 (0.53-0.73) length of upper one.
DcA ca. 0.67 (0.60-0.73) length of DcP; length
distance of dorsocentrals ca. 0.46 (0.37—0.53) cross
distance. SptAs slightly and SctPs heavily conver-
gent. SctA ca. 1.02 (0.93-1.14) length of SctP;
distance between SctPs ca. 0.40 (0.31-0.45) dis-

138 H. WatTaBE, X. C. LIANG AND W. X. ZHANG

tance between SctAs. Sterno-index ca. 0.72 (0.54-
0.84).

Legs dark brown; fore coxa much darker, anter-
iorly with ca. 2 long bristles. Number of 3Cfr ca.
15 (11-20). Wing indices: C § ca. 4.51 (4.33-
4.63), % ca. 4.44 (4.06-4.66), 4V f ca. 1.57
(1.48-1.72), $ ca. 1.53 (1.43-1.66), 4C ca. 0.56
(0.51-0.62), 5x f ca. 1.30 (1.18-1.44), $ ca. 1.17
(0.92-1.36), Ac & ca. 1.62 (1.55-1.70), $ ca.
1.75 (1.58-1.80), C3-fringe ca. 0.45 (0.40-0.51).
Haltere white.

Abdomen: Tergites black, somewhat paler at
middle. Sternites brown, paler at middle; / 3rd to

5th large.
Periphallic organs (Figs. 15 and 16): Epandrium
black, pubescent except lower margin,

caudoventrally convexed, with ca. 4 long bristles
on upper portion, 1 bristle on middle and ca. 13
bristles on lower. Surstylus dark brown, rectangu-
lar, medially pubescent on outer surface, with ca.
11 primary teeth and ca. 2 bristles; basal part
connected to epandrium narrow. Decasternum
pale brown, darker on margin, medially con-
stricted, ventrally broaden. Cercus black, entirely
pubescent, slightly convexed at lower part, with
ca. 28 long bristles and with several short bristles at
caudoventral apex.

Phallic organs (Figs.17 and 18): Aedeagus
yellowish brown, distally bilobed; cross distance
ca. 3/7 length distance. Novasternum medially
pubescent, with submedian spines on inner mar-
gin. Ventral fragma blackish brown.

reproductive organs (Figs. 19 and 20): Lobe
of ovipositor brown, much darker on ventral mar-
gin, with 4-5 discal and ca. 18 marginal teeth.
Spermatheca small, transparent, without introver-
sion, embedded by adipose tissue.

Holotype #, China: Dabochin, Yunnan Pro-
vince, 19. IX. 1988. Collector: X. C. Liang.

Paratypes, China: 5 #1, 52, same data as holoty-
pe, except 21. IX. 1988.

Distribution. The distribution of this species has
been restricted to Dali (the Autonomous District
of Bai Minority, thus species name).

Relationships. D. bai is somewhat related to D.
virilis group species in the external appearance and
in having submedian spines, but distinguishable
from these species by the diagnostic characters.

Drosophila (Drosophila) medioconstricta
Watabe, Zhang et Gan, sp. nov.
(Figs. 21-26)

Diagnosis. Arista with ca. 3 upper and ca. 2
lower branches. C index ca. 2.82, C3-fringe ca. 4/5.
Ventral margin of epandrium roundish (Fig. 21).
Aedeagus ventrally curved heavily (Fig. 24). Ven-
tral fragma laterally very wide (Fig. 23). Sper-
matheca gourd-shaped (Fig. 26).

JS, %. Body color black. Body length ca. 3.57
mm (3.4-3.6), thorax length ca. 1.70mm (1.6-
1.8), wing length ca. 3.95 mm (3.5—4.2).

Head: Eye dark red, with thick piles. Second
joint of antenna black with 2 stout setae; 3rd black,
rounded at tip. Arista with ca. 3 upper and ca. 2
lower branches in addition to a moderate terminal
fork. Frons blackish brown, ca. 0.51 (0.49-0.52)
as broad as head, anteriorly with a few frontal
hairs. Orb 2 ca. 0.34 (0.24-0.39) length of Orb 1;
Orb 3 ca. 0.62 (0.58-0.64) length of Orb 1. Face
reddish brown; carina high, wider below. Clypeus
blackish brown. Cheek brown, ca. 0.21 (0.15-
0.26) as broad as maximum diameter of eye, with
ca. 3 bristles at lower corner. Or 2 thin, ca. 0.56
(0.45-0.67) length of Or 1; Or 3 ca. 1/2 length of
Or 2. Palpus dark brown, with ¥ ca. 8 long bristles
and ca. 4 bristles.

Thorax: Mesoscutum dark brown, medially with
a longitudinal darker stripe and laterally 1 pair of
obscure broad stripes along notopleurals.
Scutellum black, with 1 pair of brown stripes in
lateral sides. Thoracic pleura black. Lower
humeral ca. 0.75 (0.71-0.77) length of upper one.
DcA ca. 0.67 (0.52-0.73) length of DcP; length
distance of dorsocentrals ca. 0.49 (0.40-0.62) cross
distance. SptA slightly and SctP heavily conver-
gent. SctA ca. 0.89 (0.77-1.09) length of SctP;
distance between SctPs ca. 0.41 (0.40-0.43) dis-
tance between SctAs. Sterno-index ca. 0.75 (0.55-
0.92).

Legs dark brown; fore coxa and femur darker.
Number of 3Cfr ca. 41 (35-48). Wing indices: C
ca. 2.82 (2.61-3.20), 4V ca. 1.59 (1.55-1.63), 4C
ca. 0.84 (0.76-0.88), 5x ca. 1.23 (1.09-1.40), Ac
ca. 2.27 (2.08-2.55), C3-fringe ca. 0.81 (0.80-
0.83). Haltere whitish yellow; basal stalk gray.

Abdomen: Tergites black, medially with small

D. robusta Group from China 139

U-shaped dark yellow area in 2nd to Sth; 1st
entirely black. Sternites brown, darker on margin;
gf 5th nearly rectangular, ? 6th quadrate, each
with ca. 10-15 long bristles on margin.

Periphallic organs (Figs. 21 and 22): Epandrium
dark brown, much darker at antero-dorsal corner,
pubescent except anterior margin, with ca. 5 bris-
tles on upper half and ca. 20 bristles on middle to
lower half. Surstylus brown, pubescent on distal
half of outer surface, with ca. 12 apically pointed
primary teeth and ca. 5 bristles at caudoventral
corner; basal part connected to epandrium broad,
ca. 1/2 maximum width of surstylus. Decasternum
pale brown, darker on margin, ladder-shaped.
Cercus black, entirely pubescent, with ca. 23 bris-
tles and ca. 9 relatively short bristles on ventral
apex.

Phallic organs (Figs. 23 and 24): Aedeagus pale
yellow, distally bilobed, ventrally curved strongly,
distally swollen in lateral view; apodeme short, ca.
1/4 as long as aedeagus. Anterior paramere pale
brown, rectangular in lateral view. Novasternum
broad, with small wart-like spines on inner half of
surface and with 1 pair of prominent submedian
spines on inner margin. Ventral fragma laterally
broaden, medially narrowing.

reproductive organs (Figs. 25 and 26): Lobe
of ovipositor brown, with ca. 4 light orange discal
teeth and ca. 25 orange marginal teeth; ultimate
marginal tooth blackish brown, ca. 2.5 times as
long as penultimate. Spermatheca constricted at
middle (thus, species name), with small dark
patches on upper half of outer surface, without
apical indentation; introvert ca. 6/7 height of outer
capsule; duct distally expanded slightly.

surope

OD. pullata
D. unimaculata

D yunnanensis
D.medioconstricta
D. bai

Holotype #, China: Dabochin, Yunnan Pro-
vince, 21. IX. 1988. Collector: X. C. Liang.

Paratypes, China: 3 , same data as holotype.

Distribution. China: Dabochin, Kunming, Yun-
nan Province.

Relationships. D. medioconstricta is related to
D. lacertosa, D. yunnanensis and D. bai in having
novasternum with submedian spines, but disting-
uishable from the latter three species by the shapes
of aedeagus and ovipositor. Further, this species
seems to be related to the D. melanica and D.
virilis species-groups in having a relatively small
value of C-index.

Remarks. D. medioconstricta inhabits water-
sides and their surrounding forests.

THE GEOGRAPHIC DISTRIBUTION

The robusta species-group is considered to have
evolved in the virilis-repleta Radiation, and all
members have been recorded in temperate to cool
regions of Asia and North America: 6 species in
Japan, D. okadai Takada, D. neokadai, D. mori-
wakii Okada et Kurokawa, D. sordidula Kikkawa
et Peng, D. pseudosordidula Kaneko et al. and D.
lacertosa; 2 species in the mainland of China, D.
cheda and D. pullata; 2 species in North America,
D. colorata Walker and D. robusta Sturtevant.
Narayanan [5] examined chromosomes and cross-
abilities of these American and Japanese members
except for D. okadai and D. neokadai, and prop-
osed an evolutionary phylogeny of the robusta
group.

From a comparative study of genitalia, however,
Beppu [6] has recently transferred D. moriwakii

D. robusta’

D.sordidula
D. pseudosordidula

Fic. 27. The geographic distribution of the Drosophila robusta species-group.

140 H. WataBE, X. C. LIANG AND W. X. ZHANG

and D. colorata from the robusta group to its allied
melanica species-group, and vice versa D. unim-
aculata Strobl distributed in Europe. The geog-
raphic distribution of the robusta group is redrawn
in Fig. 27, on the basis of recent knowledge includ-
ing the present new species. The figure shows that
China is richest in this fauna and includes three
endemic species having submedian spines common
to the melanica and virilis species-groups.

These information indicates that the evolution-
ary history of the robusta group should be reconsi-
dered, and that China is a very important area
when considering this.

ACKNOWLEDGMENTS

The authors are grateful to Dr. Masanori J. Toda of
Hokkaido University for his advice during this study.
This work was supported by a Grant-in-Aid for Overseas
Scientific Survey from the Ministry of Education, Science
and Culture, Japan (Nos. 62041085, 63043060).

REFERENCES

Okada, T. (1956) Systematic Study of Drosophilidae
and Allied Families of Japan. Gihodo Co. Ltd.,
Tokyo, pp. 183.

Kaneko, A. and Takada, H. (1966) Drosophila
Survey of Hokkaido XXI. Description of a new
species, Drosophila neokadai sp. nov. (Diptera, Dro-
sophilidae). Annot. Zool. Japon., 39: 55-59.
Watabe, H. and Nakata, S. (1989) A comparative
study of genitalia in the Drosophila robusta and D.
melanica species-groups (Diptera: Drosophilidae). J.
Hokkaido Univ. of Education, Ser. IIB., 40: 1-18.
Tan, C. C., Hsu, T. C. and Sheng, T. C. (1949)
Known Drosophila species in China with descriptions
of twelve new species. Univ. Texas Publ., 4920: 196-
206.

Narayanan, Y. (1973) The phylogenetic relationships
of the members of the Drosophila robusta group.
Genetics, 73: 319-350.

Beppu, K. (1988) Systematic positions of three
Drosophila species (Diptera: Drosophilidae) in the
virilis-repleta radiation. Proc. Japan. Soc. Syst. Zool.,
37: 55-58.

ZOOLOGICAL SCIENCE 7: 141-145 (1990)

Reexamination on the Taxonomic Position of Two Intraspecific
Taxa in Japanese Eothenomys: Evidence from Crossbreeding
Experiments (Mammalia: Rodentia)

Axiro ANDO, SATOSHI SHIRAISHI) and TeRu Aki UcHIDA

Zoological Laboratory, Faculty of Agriculture,
Kyushu University 46-06, Fukuoka 812, Japan

ABSTRACT—As a part of the study in which the taxonomic validity of E. kageus is inclusively
reexamined, crossbreeding experiments were made between Eothenomys smithii with six mammae and
E. kageus with four mammae, which have been separated by the difference mainly in number of the
mammae and in shape of the baculum. Consequently, these two species readily interbred in crosses of
both E. smithii? x E. kageus § and E. kageus? XE. smithiif\. F, hybrids obtained from the two
crosses possessed normal breeding ability. Moreover, F; hybrids were produced from the line of the
former cross. Some daughters were different from their mothers in number of the mammae, and some
litters included both females with four mammae and females with six mammae within a litter. These
facts suggest that both species may share an intercommunicating gene pool with each other, and mean
that the difference in number of the mammae is an intraspecific variation or a polymorphism. It has
been also said that the shape of the bacula is a poor taxonomic character. Therefore, taking the
reproductive compatibility between them and the unreliability of the taxonomic characters into account,
together with the almost entire identity in their karyotypes previously reported by us, it is concluded that

© 1990 Zoological Society of Japan

E. kageus is synonymous with E. smithii.

INTRODUCTION

Taxonomy must be grounded on integrated con-
clusions which were clarified by investigations
from various aspects. Accordingly, when two
related forms, which had been separated only by
the morphological characters, have given rise to a
taxonomic dispute, more extensive studies are
desired. In particular, a crossbreeding experiment
occupies a great important position in this field and
is a useful method of assessing the relationship
between such allied forms.

The Smith’s red-backed vole with six mammae
(Eothenomys smithii) and “Kage” red-backed vole
with four mammae (E. kageus) are said to occur in
the western and central parts of Japan, respective-
ly [1]. However, a dispute on the taxonomic
position of these two “species” has still remained

Accepted February 20, 1989
Received October 14, 1988
" To whom reprint requests should be addressed.

unsolved. The dispute began with the separation
of E. kageus from E. smithii by the difference
mainly in number of the mammae and in shape of
the baculum. Afterwards, many studies have been
made on this problem, but restricted to morpholog-
ical [2-7] and karyological (a conventional staining
method) [8, 9] aspects. In this connection, as a
part of the broader study in which the taxonomic
validity of E. kageus is inclusively reexamined,
comparisons of the growth and development pat-
terns and the karyotypes (G- and C-band patterns)
between E. smithii and E. kageus revealed that the
characteristics of these patterns in E. kageus were
basically identical with those in E. smithii [10, 11].
Then, in order to gain new information on this
problem, crossbreeding experiments between
them were carried out, although the sample size
was small. The aim of the present study is to
examine the reproductive compatibility between
E. smithii and E. kageus and the reliability of the
number of the mammae as a diagnostic character,
and to discuss the taxonomic position of both

142 A. ANbo, S. SHIRAISHI AND T. A. UCHIDA

species.

MATERIALS AND METHODS

E. smithii (2 males, 6 females) and E. kageus (3
mles, 2 females), used as parental generations,
were obtained from laboratory colonies which
were derived from wild voles live-trapped in
Kyushu (Fukuoka Prefecture) and central Honshu
(Nagano and Yamanashi Prefectures), respective-
ly. All hybrid generations were originated from
four pairs of E. smithii? x E. kageus § and two
pairs of E. kageus 2 XE. smithii J (see Table 1).

All animals were housed in stainless steel cages
(43 x25 x23 cm), and given ad libitum a commer-
cial diet (NMF or CMF, Oriental Yeast Co., Ltd.,
Tokyo) and water, and sometimes fresh cabbages.
The experimental colonies were maintained at
temperatures of 20+1°C on photoperiods of 12 hr
light: 12 hr dark. Humidity was not controlled
throughout the experimental period. Voles pro-
duced litters throughout the year under such rear-
ing conditions. Gravid females were inspected
daily for deliveries so that their litters were discov-
ered within 24hr post partum. The day when
neonates were found by checking was designated
the day of parturition and day 0 of the newborn
young. The number of young on the day when
they were found was regarded as the litter size.
Since the mammae of female young became de-
tectable as about day 7 in both E. smithii and E.
kageus [12, 10], the number of the mammae of
female hybrids was determined at this time.

RESULTS

The results of the breeding experiments are
given in Table 1. Eleven types of crosses, consist-
ing of eight in the line of E. smithii? XE.
kageus § and three in the line of E. kageus 2 X E.
smithii f’, were attempted in this study. Litters
were produced from all types of crosses. Mean
litter sizes showed 2.5-3.0 in most of the crosses,
although the smaple size was very small in some
cases. The maximum and minimum mean litter
sizes were 4.8 and 1.9, respectively.

Regarding the number of the mammae, there
were two kinds of crosses; i.e. one was the cross in

which all females of the progeny had six mammae,
and the other was the cross in which females of the
progeny possessed four or six mammae. Out of the
eleven types of crosses, eight types of crosses
leading to the former case bore 71 litters en bloc,
whereas the remaining three types of crosses re-
sulting in the latter case gave birth to 29 litters as a
whole. Out of the above 29 litters, five litters
contained both females having four mammae and
females having six mammae (6 males and 12
females in total; out of 12 females, five had four
mammae and six had six mammae, and one was
undetermined); the remaining 24 litters involved
only females with six mammae. However, no cross
existed in which all female hybrids had four mam-
mae. The maximum longevity was 1,354 days in F,
hybrids between E. smithii? and E. kageus %.

DISCUSSION

1) Reproductive compatibility between E. smithii
and E. kageus.

Isolating mechanisms which are necessary for
maintenance of speices integrity are classified into
two categories, premating and postmating
mechanisms; the former category contains season-
al and habitat, ethological and mechanical isola-
tion, while the latter category includes gametic and
zygote mortality, hybrid inviability and hybrid
sterility [13].

In connection with attempts to obtain inter-
specific hybrids, crossbreeding (or artificial insemi-
nation) experiments between different species
have been carried out in some orders of Mamma-
lia, including Rodentia. However, there have
been few hybrids with fertility in such cases be-
cause of operation of the above postmating
mechanisms. As examples of hybrids which die
during embryogenesis, the following combinations
have been well known: goats (Capra hircus) xX
sheep (Ovis aries) [14], ferrets (Mustela furo) Xx
minks (Mustela vison) [15], rabbits (Oryctolagus
cuniculus) Xhares (Lepus americanus) [15] and
black rats (Rattus rattus) x brown rats (Rattus nor-
vegicus) [16,17]. On the othr hand, in crosses of
horses (Equus caballus) X donkeys (Equus asinus)
[18], horses (E. caballus) x zebras (Equus grevyi)

Crossbreeding in Japanese Eothenomys 143
TABLE 1. Breeding results
No. of
young Mean No. of
Types of crosses Negot Ne, ok ——— litter Range mammae of
P df £ Ud size 9 young
I E. smithii (%)XE. kageus (f) 4 i PAY) 2 WD 1-5 6
(smithii x kageus)F, X (smithii x kageus)F, 2 16 2425 4 3.3 1-5 6
(smithii X kageus)F > X (smithii x kageus)F > 5 17 2419 2.5 2-4 6
(smithii X kageus)F3 X (smithii x kageus)F3 3 12 1415 1 2.5 1-5 4 or 6*
(smithii X kageus)F 4 X (smithii x kageus)F 4 il 2 DB 2.5 2-3 6
E. smithii (2) xX (smithiix kageus)F, (J) 2 7 6 7 1.9 1-3 6
(smithii X kageus)F3 ( 9.) X(smithii x kageus)F2 (3) 1 1 2 3.0 3 4 or 67
(smithii x kageus)F, (2) X (smithii x kageus)F3 (3) 1 2 3m 2 DES) 2-3 6
II E. kageus ($.)XE. smithii (f) 2 SiS RiSip 2 eS:8 2-6 6
(kageus X smithii)F, X (kageus X smithii)F, 3 146 4033 3 4.8 3-6 4 or 64
(kKageus X smithii)F X (kageus X smithii)F > 2 33 3.0 3 6

Ud, undetermined.

* Two and eight females have 4 and 6 mammae, respectively, and five have mammae of undetermined number.
+ One and the other female have 4 and 6 mammae, respectively.
+ Four and 23 females have 4 and 6 mammae, respectively, and six have mammae of undetermined number.

[19], Syrian hamsters (Mesocricetus newtoni) x
golden hamsters (Mesocricetus auratus) [20] and
Shaw’s jirds (Meriones shawi) xLibyan. jirds
(Meriones libycus) {21], their hybrids survive to
adulthood, but males and/or females are sterile.
As to the present experiments, in crosses of both
E. smithii?. XE. kageus § and E. kageus ? XE.
smithii {, these two species readily interbred.
Both sexes of F; hybrids obtained from the above
two crosses seem to possess, at least under labora-
tory conditions, normal breeding ability. Further,
F; hybrids were produced from the line of the
former cross. Therefore, these facts demonstrate
that almost no barrier caused by postmating isolat-
ing mechanisms exists between both species, so far
as the crossbreeding experiments are concerned.
In our study mean litter sizes in the crosses were
1.9-4.8, being smaller than the mean litter size
(4.5) of E. smithii [22], except for the value 4.8. E.
smithit has wide individual variations in litter size,
and the lowest and highest prolificacies were 2.8
and 6.8 young per litter, respectively [22]. Accord-
ingly, when the litter size is discussed, it is neces-
sary to get a larger sample size just in such
crossbreeding experiments. Although there is still
much to be investigated, very important is the fact

that progeny between E. smithii and E. kageus
were fertile. This fact suggests the probability that
both species may share an intercommunicating
gene pool with each other, but detailed analyses at
the gene level are indispensable with respect to
this.

It must be noted that fertility or sterility of
hybrids is not the sole criterion of species [13, 23].
For example, coyotes (Canis latrans) x dogs (Canis
familiaris) hybrids are fertile [24]: premating iso-
lating mechanisms seem to play a significant role in
such cases [13]. There has been no direct evidence
showing the presence or absence of barriers due to
premating mechanisms between E. smithii and E.
kageus under natural conditions. However, judg-
ing from our success in interbreeding between both
species in the laboratory, and from existence of
females with four and six mammae under both
laboratory and field (as mentioned below) condi-
tions, there is a high possibility that both forms are
crossbred under natural conditions.

2) Taxonomic validity of E. kageus.

When discussing the taxonomic status of E.
smithii and E. kageus, it is of importance to
examine the reliability of the taxonomic characters

144 A. ANDO, S. SHIRAISHI AND T. A. UCHIDA

(the number of the mammae and the shape of the
baculum) which were described by Imaizumi [1].

Imaizumi [1] has stated that the difference in the
mammary formula of these two species seems to be
fairly important as one of the main diagnostic
characters. However, our crossbreeding experi-
ments revealed that some daughters were different
from their mothers in number of the mammae, and
that some litters included both females with four
mammae and females with six ones within a litter.
These facts apparently indicate that the number of
the mammae is not a constant character at the
individual level. Furthermore, in Mt. Yatsugatake
(Nagano Pref.), Mt. Kamegamori (Ehime Pref.),
Mt. Tsurugi (Tokushima Pref.) and Mt. Hakusan
(Ishikawa Pref.), both females with four mammae
and females with six mammae are captured at the
same locality [3-5, 25]. The difference in number
of the mammae has been considered to be indi-
vidual variation [3, 4]. In this context, it is worthy
of note that a pair of voles from Nagano Prefecture
(the female with four mammae) produced a litter
including a female with four mammae and a female
with six mammae in our laboratory [unpublished];
this fact agrees with the results of the above field
studies.

In murid mammals, the swamp rat (Rattus lut-
reolus) and the bush rat (Rattus fuscipes) inhabit-
ing Australia are known as species in which the
numbers of the mammae are different between
geographically isolated populations and between
subspecies, respectively [26]. In the former spe-
cies, mainland females have five pairs of teats,
while females from Tasmania have only four pairs.
Rattus fuscipes is divided into four subspecies (R. f.
fuscipes, R. f. greyi, R. f. assimilis and R. f.
coracinus); only R. f. coracinus females have four
pairs of teats, whereas females of the other three
subspecies have five pairs. Thus, the difference in
number of the mammae is an intraspecific varia-
tion or a polymorphism, being unreliable as a
diagnostic character by which E. kageus was sepa-
rated from E. smithii. With respect to this prob-
lem, Kaneko [7] also has drawn the same conclu-
sion from an investigation on the number of the
mammae in pregnant or postpartum wild females.

Regarding the baculum, Imaizumi [1] has men-
tioned that there are differences in general outline

of the posterior border of its body (semicircular in
E. smithit, while concave in E. kageus) and in
shape of the lateral prongs (curved as the letter “c”
in E. smithii, while double curved as the letter “s”
in E. kageus). However, Jameson [2] has stated
that the difference in the bacula of E. kageus and
E. smithii may be due to individual variation. On
the basis of a detailed analysis of the bacula in 71
wild males, Kaneko [7] also has concluded that the
above differences in the bacula cannot be em-
ployed as a diagnostic character between these two
species in question.

Furthermore, the following facts have been re-
ported: in an analysis of the skull, the relative
growth coefficient shows no difference among five
localities including Kyushu and central Honshu
[6]; geographical clines are recognized in the hind
foot length, the tail length and the parietal width
[4, 6]; the growth and development patterns of
both species are basically identical with each other
[10]; except for a slight variation in size of the short
arm of the Y chromosome, no detectable differ-
ence is found in the karyotype (G- and C-band
patterns) between both species [11]. All these
facts may indirectly adduce negative evidence for
the taxonomic validity of E. kageus.

From the above consideration, it can be said that
E. kageus may not be reproductively isolated from
E. smithii, even under natural condition, and that
the taxonomic characters (the number of the mam-
mae and the shape of the bacula) pointed out by
Imaizumi [1] are not regarded as diagnostic. Our
conclusion reached, therefore, is that E. kageus is
synonymous with E. smithii.

ACKNOWLEDGMENTS

We thank Professor E. W. Jameson, Jr., University of
California, for comments on the manuscript.

REFERENCES

1 Imaizumi, Y. (1957) Taxonomic studies on the
red-backed voles of Japan. Part I. Major divisions of
the vole and description of Eothenomys with a new
species. Bull. Nat. Sci. Mus. (Tokyo), (40): 195-
216.

2 Jameson, E. W., Jr. (1961) Relationships of the
red-backed voles of Japan. Pacific Sci., 15: 594-604.

10

11

12

13

14

Crossbreeding in Japanese Eothenomys

Miyao, T., Morozumi, T., Morozumi, M., Hana-
mura, H., Akahane, H. and Sakai, A. (1964) Small
mammals on Mt. Yatsugatake in Honshu. III.
Smith’s red-backed vole (Eothenomys smithi) in the
subalpine forest zone on Mt. Yatsugatake. Zool.
Mag., 73: 189-195. (In Japanese with English ab-
stract).

Miyao, T. (1967) Studies on the geographical varia-
tion of the small mammals in Japanese islands. I.
Geographical variation of Smith’s red-backed vole,
Eothenomys smithi. (2) Hind-foot length, tail
length, number of sacro-caudal vertebrae and breed-
ing activity. J. Growth, 6: 7-18. (In Japanese with
English abstract).

Tanaka, R. (1971) A research into variation in
molar and external features among a population of
the Smith’s red-bakced vole for elucidation of its
systematic rank. Jap. J. Zool., 16: 163-176.

Aimi, M. (1980) A revised classification of the
Japanese red-backed voles. Mem. Fac. Sci. Kyoto
Univ., Ser. Biol., 8: 35-84.

Kaneko, Y. (1985) Examinations of diagnostic
characters (mammae and bacula) between Eotheno-
mys smithi and E. kageus. J. Mamm. Soc. Japan, 10:
221-229. (In Japanese with English abstract).
Tsuchiya, K. (1970) Classification of Japanese
cricetid and murid rodents based on their karyotypes
(1). Yama to Hakubutsukan (Mountains and Muse-
um), 15: 2—3. (In Japanese).

Tsuchiya, K. (1981) On the chromosome variations
in Japanese cricetid and murid rodents. Honyurui
Kagaku (Mammalian Science), 21: 51-58. (In
Japanese).

Ando, A. and Shiraishi, S. (1988) Reproduction,
growth and development of the so-called “Kage”
red-backed vole, Eothenomys kageus. Honyurui
Kagaku (Mammalian Science), 28: 13-22. (In
Japanese with English abstract).

Ando, A., Shiraishi, S., Harada, M. and Uchida, T.
A. (1988) A karyological study of two intraspecific
taxa in Japanese Eothenomys (Mammalia: Roden-
tia). J. Mamm. Soc. Japan, 13: 93-104.

Ando, A., Shiraishi, S. and Uchida, T. A. (1987)
Growth and development of the Smith’s red-backed
vole, Eothenomys smithi. J. Fac. Agr., Kyushu
Univ., 31: 309-320.

Mayr, E. (1964) Animal Species and Evolution.
Belknap Press of Harvard Univ. Press, Cambridge.
Alexander, G., Williams, D. and Bailey, L. (1967)

15

16

17

18

19

20

21

22

23

24

25

26

145

Natural immunization in pregnant goats against red
blood cells of their sheep X goat hybrid foetuses.
Aust. J. Biol. Sci., 20: 1217-1226.

Chang, M. C., Pickworth, S. and McGaughey, R.
W. (1969) Experimental hybridization and chromo-
somes of hybrid. In “Comparative Mammalian
Cytogenetics”. Ed. by K. Benirschke, Springer-
Verlag, Berlin, pp. 132-145.

Hiraiwa, Y. K. and Yoshida, H. (1955) Conception
by the cross between Rattus norvegicus and R.
rattus. II. Breeding experiments by the artificial
insemination. Sci. Bull. Fac. Agr., Kyushu Univ.,
15: 267-273. (In Japanese with English résumé).
Yosida, T. H. (1980) Cytogenetics of the Black Rat.
University of Tokyo Press, Tokyo.

Benirschke, K., Brownhill, L. E. and Beath, M. M.
(1962) Somatic chromosomes of the horse, the
donkey and their hybrids, the mule and the hinny. J.
Reprod. Fert., 4: 319-326.

King, J. M., Short, R. V., Mutton, D. E. and
Hamerton, J. L. (1966) The reproductive physiolo-
gy of male zebra-horse and zebra-donkey hybrids. In
“Comparative Biology of Reproduction in Mam-
mals”. Ed. by I. W. Rowlands, Academic Press,
New York, pp. 511-527.

Raicu, P. and Bratosin, S. (1968) Interspecific
reciprocal hybrids between Mesocricetus auratus and
M. newtoni. Genet. Res., 11: 113-114.

Lay, D.M. and Nadler, C. F. (1969) Hybridization
in the rodent genus Meriones: I. Breeding and
cytological analyses of Meriones shawi (?)xX
Meriones libycus( {) hybrids. Cytogenetics, 8: 35-
50.

Ando, A., Shiraishi, S. and Uchida, T. A. (1988)
Reproduction in a laboratory colony of the Smith’s
red-backed vole, Eothenomys smithii. J. Mamm.
Soc. Japan, 13: 11-20.

Futuyma, D. J. (1986) Evolutionary Biology.
Sinauer Associate, Inc. Publishers, Sunderland.
Mengel, R. M. (1971) A study of dog-coyote
hybrids and implications concerning hybridization in
Canis. J. Mamm., 52: 316-336.

Shida, T. (1983) Notes on the small mammal fauna
in the northern slope of Mt. Hakusan. Annu. Rep.
Hakusan Nat. Conserv. Cent., (9): 57-65. (In
Japanese with English summary).

Watts, C. H. S. and Aslin, H. J. (1981) The
Rodents of Australia. Angus & Robertson Pub-
lishers, London.

q

eer rE

sw

bw prentas fie”

Via ste

erry gre

ak 3 ennai

ai sre
ty Asi een

vis oe

a canoe ina ~ é

hae F debtgaet. si Mosaic a ne

: Perec in Ue tite bul ope
cw

Ae? teaie

cana ep

7 ae Lhd etre
adh Sor anilte

nil Yo x
ar My paren Lae LAR

comerer tes (ple wpe
acy h/) See Minot) wack
one : ies : f 6
ates) ORR ae war
coe) tiie SGiel’ te
uk Eine | dal

cope ov b>» Regain

haan onl Jo 1ozegrn
‘he et sailed 1p ee A ;
: th. oe woe WENN
; Sityhs FAAES

f -atiaeel fiw Bhs ye ee
Ke Aetna se meed ly Ae loka ln lotrel “

iabeont} (heutipiyat av eye | #5

ZOOLOGICAL SCIENCE 7: 147-151 (1990)

[COMMUNICATION]

© 1990 Zoological Society of Japan

Histochemistry of Yolk Formation in the Ovaries of the
Tarnished Plant Bug, Lygus lineolaris
(Palisot de Beauvois) (Hemiptera: Miridae)

W. K. Ma! and S. B. RAMASWAMY

Department of Entomology, Drawer EM, Mississippi State University,
Mississippi State, MS 39762, USA

ABSTRACT—The ovaries of Lygus lineolaris were
studied using histochemical techniques. Three types of
yolk bodies, YB I, YB II and YB III, are recognizable in
vitellogenic oocytes. YB I granules were determined to
be lipid by Sudan black B staining and YB II and YB III
were protein/carbohydrate complexes based on brom-
phenol blue and periodic acid-Schiff staining. During
early vitellogenesis, YB I and YB II are predominant,
while in mature oocytes, YB I and YB III predominate
suggesting that YB II may be a precursor for YB III. The
germarium remains unchanged histochemically through-
out the gonotropic cycle.

INTRODUCTION

Insect eggs contain a large amount of yolk which
is incorporated during vitellogenesis. In most
insects, yolk consists mainly of lipid and protein
(with or without conjugated carbohydrates); and
in some insects, glycogen deposits are also found.
In many cases, yolk is of an extraovarian origin
and the fat body is the most common site of
storage and synthesis of yolk components [1-3]. In
addition, the nurse cells in meroistic ovarioles and
follicle cells may contribute some yolk material.
Nurse cells may be associated with each develop-
ing oocyte as in polytrophic ovarioles or housed
exclusively in the germarium as in telotrophic
ovarioles common to Hemiptera. The develop-
ment of nutritive cords in the latter type has made
this an interesting developmental system to study.

Accepted February 23, 1989
Received June 30, 1988

' Present address: Department of Entomology, Cor-
nell Univeristy, Ithaca, NY 14853, USA.

There have been a few histochemical studies of
vitellogenesis in the Hemiptera (for example,
Acanthocephala, Coreidae [4]; Oncopeltus,
Lygaeidae [5-7]; Gerris, Gerridae [8]).

A recent study [9] showed that the gonotropic
cycle in Lygus lineolaris requires seven days to be
completed. The current paper reports of observa-
tions on the histochemical changes during vitel-
logenesis in L. lineolaris.

MATERIALS AND METHODS

Insects were obtained and raised as described
previously [9]. Virgin females aged 1-7 days were
used throughout the study. Insects were anesthe-
tized under CO, and dissected in insect Ringer
[10]. Ovarioles with part of their associated lateral
oviducts were removed, fixed and stored in 10%
buffered formalin until use. Tissues used for
detection of carbohydrate and protein were dehy-
drated through graded series of ethanol, passed
successively through 1/2, 1/1, 2/1 Sorvall® embed-
ding medium (E. I. Dupont Co. Newtown,
Conn.)/absolute ethanol, infiltrated with three
fresh changes of 12 hr each of embedding medium
and embedded in the same at room temperature
under nitrogen. Plastic sections (3-4 ~m) were cut
on a LKB ultratome using glass knives.

Periodic acid-Schiff (PAS) techinique with and
without periodic acid oxidation was employed as a
general test for carbohydrate [11]. Dimedone was
used as a free aldehyde blocking agent [12]. To
demonstrate glycogen deposits, plastic sections

148 W. K. Ma anp S. B. RAMASWAMY

were incubated in 0.5% diastase at 37°C for 30 min
[13] followed by PAS staining. Control sections
were similarly processed except for the enzyme
incubation.

Treatment of sections with 0.1% bromphenol
blue in 70% ethanol was used to demonstrate
protein [13]. For detection of lipid, tissues pre-
viously fixed in 10% buffered formalin were
washed overnight in water and passed through 5%
and 10% gelatin for 2 hr each at 37°C in a vacuum
oven. Tissues were then infiltrated with 25%
gelatin overnight and embedded in the same at
4°C. Gelatin blocks with tissues were frozen with
dry ice and 4 um cryosections were cut on an IEC
cryostat (IEC Co. Ltd., Needham Heights,
Mass.). Sections were picked up on a warm glass
slide, stained with Sudan black B in propylene
glycol [14] and mounted in glycerin jelly.

Tissue sections were examined under a Zeiss
compound microscope and photographed with
Panatomic X film (Kodak, 32 ASA).

OBSERVATIONS

Three types of yolk bodies, YB I, II and III are
distinguishable in vitellogenic oocytes in L.
lineolaris. Results from bromphenol blue and PAS
staining suggested that YB II and III are protein
carbohydrate complexes (Figs. 1, 10, 11 and 12).
Combination of PAS staining with diastase treat-
ment further confirmed that the carbohydrate
moiety of YB II and III is not glycogen (Fig. 1).
YB II is more intensely stained (dark blue) with
bromphenol blue and is relatively smaller (ca. 2-8
ym) than YB III (pale blue and ca. 5—25 xm) (Fig.
12). YB I is not preserved in plastic sections but is
stained bluish black with sudan black B in cryosec-
tions suggesting its lipoidal nature (Figs. 2 and 3).

Throughout the gonotropic cycle, neither nurse
cells nor follicle cells were observed to have
undergone any histochemical changes. PAS posi-
tive and bromphenol blue stained cellular inclu-
sions were not observed in the trophic core and the
nutritive cord (Figs. 4 and 5).

At early vitellogenesis, PAS positive flocculent
material accumulates in the extraovarian space
beneath the follicular epithelium (Fig. 6, 7 and 8).
Small droplets of YB I and II appear at the

periphery of the oocyte and gradually proceed into
the central region of the oocyte (Fig. 9). Later in

ot é 2 ON © * e on

Ne Oe)

eS a a a F ed
FG? © » % gl : o.. 28
[One oS

Fics. 1. a-e. Sections of a mature egg of L. lineolaris.
Arrow heads indicate protein/carbohydrate yolk
stained by PAS (1b), bromphenol blue (le), PAS
treated with dimedone (la), PAS without periodic
acid oxidation (1c) and PAS with diastase digestion
(1d). Clear areas between the protein/carbohydrate
yolk are lipid yolk. Note the overall decrease in
staining intensity of the section after blocking of free
aldehyde groups with dimedone. Staining of pro-
tein/carbohydrate yolk in 1c is due to the counter
stain picro-aniline blue, no PAS-positive reaction
appears in these sections without going through
periodic acid oxidation. Diastase treatment has no
effect on PAS staining. All protein/carbohydrate
yolk is stained with similar intensity by bromphenol
blue. Plastic sections. (Bar=10 um).

Ovarian Histochemistry in Lygus 149

vitellogenesis, the relatively larger YB III begins
to appear in the oocyte (Fig. 12). The end of
vitellogenesis is indicated by the formation of a
PAS positive vitelline membrane and the oocytes
are filled up mainly with YB I and III (Fig. 1). This
remains unchanged throughout choriogenesis and
no histochemical changes are observed inside the
mature oocytes even after ovulation.

DISCUSSION
The origin of lipid yolk in eggs of hemipteran

insects has been a controversial subject. Three
different origins for lipid yolk have been reported
i.e., extraovarian, trophic core and follicle cells [5,
7, 15]. Previously we showed in Lygus that minute
amounts of lipid from the trophic core enter the
developing oocyte during vitellogenesis [9]. Cur-
rent observations suggest that part of the lipid yolk
in vitellogenic oocytes of Lygus originates from the
hemolymph while the follicle cells do not contri-
bute any lipid to the oocytes.

As in other insects, large amounts of carbohy-
drate (in association with protein) are incorpo-
rated into the vitellogenic oocytes of L. lineolaris.
The role(s) of the carbohydrate component(s) in
the yolk precursor in insects is still unknown.
Whether it is for maintenance of protein structure
requisite for yolk precursor recognition or has
additional nutritive value for the embryos remains
to be determined [16].

In L. lineolaris, two types of protein/carbohy-
drate yolk spheres may be differentiated in vitel-
logenic oocytes by bromphenol blue staining.
Since bromphenol blue staining intensity reflects
the number of dye binding groups [17], it is
possible that conversion of YB II into YB III is a
result of molecular modification of the yolk precur-
sor after incorporation into the oocytes similar to

Fic. 2. Distribution of lipid yolk (arrow head) in a
mature egg of L. lineolaris. Clear area around lipid
yolk is due to protein/carbohydrate complexes which
are very slightly stained. Cryosection, Sudan black
B. (Bar=10 um).

Fic. 3. Vitellogenic oocyte in L. lineolaris showing
accumulation of lipid deposits on the oocyte surface
(arrow) and also in the intercellular spaces (arrow
heads) between follicle cells (FC). Cryosection,
Sudan black B. (Bar=10 am).

Fic. 4. Ovariole of L. lineolaris showing germarium
that houses the nurse cells (T). Note the trophic
core (TR) at the central region of the germarium
which is bounded by nurse cells. The trophic core
has a homogeneous overall staining and discrete
cellular inclusions are not found. Plastic sections,
PAS. (Bar=10 pum).

Fic. 5. The nutritive cord (NR) connects the trophic
core (not shown) and the developing oocyte (O)
throughout the gonotropic cycle until chorion
formation. Note the uniform staining of the nutri-
tive cord and the oocyte and also the absence of
inclusions in the former. Plastic section, PAS. (Bar
=5 pam).

150 W. K. Ma anp S. B. RAMASWAMY

“

Fics. 6-8. Terminal oocyte at early vitellogenesis showing accumulation of yolk precursor (arrow heads) in the space
between follicular epithelium and oocyte. This yolk precursor is PAS-positive (Fig. 6) (Bar=10 sm) and stained
dark blue by bromphenol blue (Fig. 7) (Bar=10 «m). In cryosection, the yolk precursor is stained bluish black by
Sudan black B (Fig. 8) (Bar=5 «m). Note the large round germinal vesicle (GV) in the oocyte of Fig. 8.

Fic. 9. Terminal oocyte at early vitellogenesis (right) showing appearance of yolk droplets at the cortex of the
oocyte. At later development, these yolk droplets coalesce and are seen to migrate into central core of the oocyte

Ovarian Histochemistry in Lygus 151

that reported in cockroaches [18]. However, this is
currently unknown in Hemiptera and further
biochemical studies are necessary to validate the
above hypothesis.

ACKNOWLEDGMENTS

We thank Dr. Gordon Snodgrass, USDA-ARS,
Stoneville, Miss. for help in obtaining the insects used in
this study and Drs. Erwin Huebner, Univ. of Manitoba,
Gerald Baker, Howard Chambers and James Heitz for
reviewing an earlier draft of this paper. This study was
supported in part by a grant form Albany International
(now Scentry, Inc., Buckeye, AZ). Miss. Agr. For. Exp.
Stn. Paper No. 6333.

REFERENCES

1 Telfer, W. H. (1965) Annu. Rev. Entomol., 10:
161-184.

2 Engelmann, F. (1968) Annu. Rev. Entomol., 13: 1-
26.

3 Kunkel, J. G. and Nordin, J. H. (1984) In
“Comprehensive Insect Physiology, Biochemistry
and Pharmacology”. Vol.1. Ed. by G. A. Kerkert
and L. I. Gilbert, Pergamon Press, New York, pp.

4

ONAN

17

18

Schrader, F. and Leuchtenberger, C. (1952) Exp.
Cell Res., 3: 136-146.

Bonhag, P. F. (1955) J. Morphol., 96: 381-439.
Bonhag, P. F. (1955) J. Morphol., 97: 283-311.
Schreiner, B. (1977) J. Morphol., 151: 81-101.
Eschenberg, K. M. and Dunlap, H. L. (1966) J.
Morphol., 118: 297-316.

Ma, W. K. and Ramaswamy, S. B. (1987) Int. J.
Insect Morphol. Embryol., 16: 309-322.

Pringle, J. W. S. (1938) J. Exp. Biol., 15: 101-113.
Drury, R. A. B. and Wallington, E. A. (1976)
Carleton’s Histological Technique. (4ed). Oxford
Univ. Press, London. pp. 204-206.

Cannon, M. S., Kapes, E. D. and Cannon, A. M.
(1982) Lab. Med., 13: 102-105.

Klungness, L. M. and Peng, Y. S. (1984) J. Insect
Physiol., 30: 511-521.

Culling, C. F. A. (1975) Handbook of Histopatholo-
gical and Histochemical Techniques. (3ed). Butter-
worths, London. pp. 360-361.

Huebner, E. and Anderson, E. (1972) J. Morphol.,
138: 1-40.

Hagedorn, H. H. and Kunkel, J. G. (1979) Annu.
Rev. Entomol., 24: 475-505.

Mazia, D., Brewer, P. and Alfert, M. (1953) Biol.
Bull., 104: 57-67.

Brookes, V. J. and Dejmal, R. K. (1968) Science,
160: 999-1001.

(left). The PAS-positive flocculent material between follicular epithelium and oocyte is found to be more diffuse
in oocyte at the right than the one at the left. Plastic section, PAS. (Bar=10 um).

Fic. 10. Higher magnification of an early vitellogenic oocyte indicates that both protein/carbohydrate yolk (open
arrows) and lipid yolk (arrow heads) appear on the periphery of the oocyte. Plastic section, PAS. (Bar=5 um).

Fics. 11-12. Vitellogenic oocyte at later development is found to incorporate large amount of protein/carbohydrate
yolk which is PAS-positive (Fig. 11). However, when stained with bromphenol blue (Fig. 12), this yolk can be
distinguished into two types: YB II stained darker (open arrows) and ca. 2-8 pam in size and YB III stained lighter
(arrow heads) and ca. 5—25 am in size. Lipid yolk (YB I) is represented in these figures as clear areas around the

protein/carbohydrate yolk. Plastic sections. (Bar=10 um).

i
ot
t
" : Dy ie qe Pi
WHOL! mk sfapbenelioa EWAN areNeapeaaD
~ 7 i
a
wel ae ae be
nos i tie?
a te" D st: ad
‘ eS ee hat.
* a i ’ iter bane c
nk i ee Te Seeil Pe ap ay
| PRM. 9 Set ies
i ; \ Y ¥ Hh 7 if 7 Poti J
. . S easy -
; 4 , wry ya
{ '
ie
, H
' rk’
#4
ty
: é

hia tar and
| links (iy one BAA ae
4) BoD bh rr Bea

ae 2 2 ph te diese

ZOOLOGICAL SCIENCE 7: 153-157 (1990)

[COMMUNICATION]

© 1990 Zoological Society of Japan

Effects of Puromycin and a-Amanitin on the Activity
of Alkaline Phosphatase in Early Preimplantation
Mouse Embryos

TOMOICHI IsHikawa!

Department of Anatomy, Kochi Medical School,
Nankoku, Kochi 781-51, Japan

ABSTRACT—The activity of nonspecific alkaline phos-
phatase (ALPase) was cytochemically and biochemically
investigated in early preimplantation mouse embryos.
The ALPase activity was cytochemically detected in
embryos from 2- to 8-cell stage. The activity was
exclusively localized on the adjacent cell membranes of
the two blastomeres. This activity was biochemically first
detected at the very low level in the 2-cell embryos.
Biochemical assay revealed that in 4- to 8-cell embryos
the ALPase activity increased dramatically. In these
embryos, intense activity of ALPase was also detected
cytochemically. When 2-cell embryos were treated with
puromycin (15 yg/ml) in culture, the embryos could not
develop to advanced stages and the expression of
ALPase activity was inhibited. However, a-amanitin (2
and 10 g/ml) suppressed the further cleavage, although
it did not interfere with the expression of ALPase
activity. The results suggest that mRNA molecules for
ALPase of maternal origin exist in 2-cell embryos.

INTRODUCTION

Non-specific alkaline phosphatase (ALPase,
E.C.3.1.3.1) is one of the well-known enzymes in
early mammalian embryos. Cytochemical studies
have shown the presence of its activity in 8-cell and
more advanced mouse embryos [1-5]. Most of
these investigators could not detect ALPase activ-
ity in fertilized and 2-cell embryos, except Mulnard
and Huygens [4] who succeeded in detecting its

Accepted April 4, 1989
Received January 30, 1989

" Deceased on February 1, 1988. Reprint requests and
correspondece should be addressed to Dr. T. Hirobe
of National Institute of Radiological Sciences, Divi-
sion of Biology, Anagawa, Chiba 260, Japan.

activity in 2-cell embryos. They showed the
precipitation of its reaction products in the adja-
cent cell membranes of the two blastomeres [4].

In the present study the expression of ALPase
activity in the early mouse embryos was investi-
gated in detail by employing both cytochemical
and biochemical methods. Few information is
available on the mechanism of ALPase expression
except that concerning ascidian embryos, where
ALPase has been employed as a histochemical
marker for gut endodermal differentiation. This
enzyme is considered to be maternal origin [6]. In
the present investigation, inhibitors of transcrip-
tion and translation were used to clarify whether
the expression of ALPase activity in 2-cell embryos
requires the de novo transcription and translation
or not. The results suggest that the expression of
ALPase activity requires de novo translation, but
not transcription.

MATERIALS AND METHODS

Embryos

Female mice of strain 129/Sv were mated with
male mice of the same strain after the superovula-
tion by intraperitoneal injection of 5 i.u. pregnant
mare serum gonadotropin (PMSG) followed by 5
iu. human chorionic gonadotropin (hCG) 48 hr
later. The 0 hr was defined as the time when hCG
was injected. Two-cell stage and 4- to 8-cell stage
embryos were flushed out from the oviduct 40-42
hr and 65-67 hr after hCG injection, respectively.

154 T. ISHIKAWA

Embryos were cultured in the standard egg culture
medium [7]. The embryos of the experimental
groups were cultured with medium containing
either 15 ug/ml of puromycin or 2-10 ug/ml of
a-amanitin for 3-24 hr.

Cytochemistry of alkaline phosphatase

Embryos were prefixed with ice-cold 1% glutar-
aldehyde in 0.1 M cacodylate buffer (pH 7.3) for
20 min. They were then washed in the same buffer
for more than 30 min. The lead citrate method [8]
was used for the cytochemical demonstration of
ALPase activity. The incubation was carried out at
37°C for 30-60 min in pH range of 9.2-9.3. No
reaction was observed in control specimens which
were incubated without substrate. After the
incubation, specimens were washed with the buffer
and embedded in agar blocks [9] and were fixed
with ice-cold 1% OsO, solution for 60 min. They
were dehydrated through a series of graded etha-
nols and embedded in epoxy resin.

For light microscopy, some of the specimens
were washed with distilled water after the ALPase
reaction and then dipped into diluted ammonium
sulfide solution to detect the reaction products.

Biochemical assay of alkaline phosphatase

Fresh unfertilized eggs (>100), 2-cell embryos
(>100) and 4- to 8-cell embryos (50-100) were
collected for each assay. The blastomeres were
broken by freezing and thawing. The ALPase

a

Fic. 1.

activity was assayed by using Chung’s method [10].
The reaction was carried out at 37°C for 4 hr. The
amount of p-nitrophenol released from _ p-
nitrophenol phosphate by the enzyme reaction was
determined by spectrophotometry at the wave
length of 410 nm. The specific activity of the
enzyme was expressed in terms of nmol p-
nitrophenol released in 1hr per embryo. No
enzyme activity was observed in the controls
incubated in the same reaction mixture without the
substrate.

RESULTS

Cytochemistry of alkaline phosphatase

ALPase activity was detected exclusively in the
adjacent cell membranes of the two blastomeres in
2-cell embryos (Fig. la). Free cell surfaces of the
blastomeres completely lacked ALPase activity.
No ALPase activity was detected in the cell
membranes of unfertilized eggs.

Biochemical assay of alkaline phosphatase activity

Embryos of 2-cell stage showed a weak ALPase
activity (0.005+0.007 nmol/hr/embryo, mean+
standard deviation, Table 1). In contrast, unfertil-
ized eggs completely lacked its activity. In 4- to
8-cell embryos, ALPase activity drastically in-
creased. The enzyme activities (0.15+0.070 nmol/
hr/embryo, Table 1) in 4- to 8-cell embryos were

b eas

Localization of ALPase activity in 2-cell mouse embryos and effects of puromycin on its activity. Mouse

embryos were cultured with standard medium for 3 hr (a) and with medium containing 15 g/ml of puromycin (b).
Alkaline phosphatase activity is shown on the adjacent cell membranes of the two blastomeres in 2-cell embryos
of control (a). However, the enzyme activity was not observed when 15 g/ml of puromycin was added to the

culture medium (b). Scale, 10 «m.

Alkaline Phosphatase of Mouse Embryo 155

about 30 times higher than those in 2-cell embryos.

Effects of puromycin and a-amanitin on the de-
velopment of 2-cell embryos in culture

When 2-cell embryos were cultured with stan-
dard medium for 24 hr, more than 50% embryos
developed to advanced stages (Table2). The
effects of puromycin, an inhibitor of translation,
on the development of 2-cell embryos were investi-
gated. Puromycin added at a concentration of 15
yg/ml completely blocked the development of

TABLE 1.
mouse embryos

2-cell embryos in culture. Also, the effects of
a-amanitin, an inhibitor of transcription, on the
development of 2-cell embryos were investigated.
When a-amanitin was added to the medium at the
dose of 10 g/ml for 24 hr, most of embryos could
not develop further (Table 2). However, more
than 35% of the embryos could develop to
advanced stages when 2-cell embryos were cul-
tured with medium containing 2 ug/ml of a-
amanitin (Table 2).

Biochemical assays of alkaline phosphatase activity in preimplantation

Developmental No. of No. of Alkaline

stage experiments eggs or phosphatase
embryos activity*

Unfertilized eggs 130-150 0+0

2-cell embryos 70-130 0.005 + 0.007

(41-43 hr after hCG)

4- to 8-cell embryos 3 15— 35 0.150+0.070

(65 hr after hCG)

* nmol p-nitrophenol released/hr/embryo (mean+standard deviation)

TaBLE2. Effects of puromycin and a-amanitin on the development of 2-cell embryos cultured for 24 hr

Group Dose No. of 2-cell 3-cell 4-cell 5- to 8-cell
embryos

Control — 125 52 12 50 11
(41.6%) (9.6%) (40%) (8.8%)

Puromycin 15 ng/ml 64 64 0 0 0
(100%) (0%) (0%) (0%)

a-amanitin 2 pg/ml 53 34 3 12 4
(64.1%) (5.7%) (22.6%) (7.6%)

a-amanitin 10 pg/ml 60 56 3 1 0
(93.3%) (5%) (1.7%) (0%)

TABLE 3.
2-cell embryos cultured for 24 hr

Effects of puromycin and a-amanitin on alkaline phosphatase activity in

Alkaline phosphatase activity*

Group Dose No. of + + —
embryos
Control — 15 12 2 1
Puromycin 15 pg/ml 9 0 4 5
a-amanitin 2 pg/ml 6 5 1 0
a-amanitin 10 4g/ml 22 10 7 5
* +: positive reaction
+: weakly positive reaction

: Negative reaction

156 T. ISHIKAWA

Effects of puromycin and a-amanitin on alkaline
Phosphatase activity

Most of the 2-cell embryos cultured with normal
medium for 24hr expressed ALPase activity
(Table 3). When 2-cell embryos were treated in
culture with 15 g/ml of puromycin for 24 hr, most
of the embryos showed a very weak or no ALPase
activity (Fig. 1b and Table3). In contrast to
puromycin, a-amanitin (2 and 10 g/ml) did not
inhibit the expression of ALPase activity (Table
3). Although 10 g/ml of a-amanitin blocked the
development of 2-cell embryos in culture, it did
not suppress ALPase activity.

DISCUSSION

In the present study, ALPase activity was
demonstrated in 2-cell mouse embryos both
cytochemically and biochemically. Although
ALPase activity was detected cytochemically in
2-cell mouse embryos by Mulnard and Huygens
[4], their observations have not been supported by
the biochemical analysis [5]. The present report,
for the first time, describes the presence of ALPase
activity in 2-cell embryos as demonstrated by
biochemical means. The reason for the success in
the detection of ALPase activity in 2-cell embryos
probably due to the fact that the author used more
than 100 fresh embryos for the assay.

When 2-cell embryos were treated with puromy-
cin (15 g/ml) for 24 hr, they could neither under-
go further cleavage nor express ALPase activity. It
is probable that the suppression of ALPase activity
is due to the inhibition of de novo synthesis of
ALPase molecules, since puromycin is known to
inhibit translation. It has been known that
puromycin inhibits in the concentration ranges of
4.7-94 ug/ml, the protein synthesis in rabbit reti-
culocyte [11]. When 2-cell embryos were treated
with a-amanitin (2 or 10 g/ml) for 24hr, the
development was inhibited, but the expression of
ALPase activity was not suppressed. From the
results it was proposed that the ALPase activity in
2-cell embryos was expressed without de vovo
synthesis of mRNA, since a-amanitin is an inhibi-
tor of RNA polymerase II. a-amanitin (0.4—40
pyeg/ml) is known to completely inhibit RNA

polymerase II activity in calf thymocytes [12].
These results suggest that mRNA molecules for
ALPase of maternal origin exists in 2-cell embryos.
In contrast, the increase of ALPase activity in 4- to
8-cell embryos may be due to the embryonic
transcription and translation, since ALPase activ-
ity increased drastically from 2-cell to 8-cell stage.

The 1- and 2-cell stages of development of a
mouse embryo are thought to be dependent on the
use of inherited maternal mRNA and on the
operation of post-transcriptional regulators [13]. It
is suggested that changes in the protein synthetic
profile occur between the early 2-cell stage and the
late 2-cell stage, and much of the maternally
inherited mRNA is inactivated rapidly [14]. The
stage of the 2-cell embryos used in the present
study corresponds to the late 2-cell stage, since the
embryos were collected from the oviduct 40-42 hr
after hCG injection. Therefore, it is conceivable
that the changes in the ALPase synthesis occur
between the late 2-cell stage and the 4-cell stage.
Although the difference between their results and
the present findings cannot be fully explained, they
might be attributed to difference in kind of
proteins used. Molecular analyses of mRNA for
ALPase in the mouse embryos at the late 2-cell
stage remain to be investigated in a future study.

ACKNOWLEDGMENTS

The author expresses his thanks to Dr. T. Hirobe of
National Institute of Radiological Sciences for his help in
preparing the manuscript.

REFERENCES

1 Izquierdo, L. and Marticorena, P. (1975) Alkaline
phosphatase in preimplantation mouse embryos.
Exp. Cell Res., 92: 399-402.

2 Johnson, L. V., Calarco, P. G. and Siebert, M. L.
(1977) Alkaline phosphatase activity in the preim-

plantation mouse embryo. J. Embryol. Exp.
Morph., 40: 83-89.
3. Vorbrodt, A., Konwinski, M., Solter, D. and

Koproski, H. (1977) Ultrastructural cytochemistry
of membrane-bound phosphate in preimplantation
mouse embryos. Dev. Biol., 55: 117-134.

4 Mulnard, J. and Huygens, R. (1978) Ultracytoche-
mical localization of nonspecific alkaline phospha-
tase during cleavage and blastocyst formation in the

Alkaline Phosphatase of Mouse Embryo 157

mouse. J. Embryol. Exp. Morphol., 44: 121-131.
Izquierdo, L., Lopez, T. and Marticorena, P. (1980)
Cell membrane regions in preimplantation mouse
embryos. J. Embryol. Exp. Morph., 59: 89-102.
Whittaker, J. R. (1977) Segregation during cleavage
of a factor determining endodermal alkaline phos-
phatase development in ascidian embryos. J. Exp.
Zool., 202: 139-154.

Biggers, J. D., Whitten, W. K. and Whittingham,
D. G. (1971) The culture of mouse embryos in vitro.
In “Methods in Mammalian Embryology”. Ed. by J.
C. Daniel, Jr., Freeman, Co., San Francisco, pp.
86-116.

Mayahara, H., Hirano, H., Saito, T. and Ogawa, K.
(1967) The new lead citrate method for the ultracy-
tochemical demonstration of activity of non-specific
alkaline phosphatase (orthophosphoric monoester
phosphohydrase). Histochemie, 11: 88-96.
Ishikawa, T. and Seguchi, H. (1984) Mg**-
dependent adenosine triphosphatase activity in
preimplantation mouse embryos. Acta Histochem.

10

11

12

14

Cytochem., 17: 269-278.

Chung, A. E., Esters, L. E., Shinozuka, H.,
Braginsky, J., Lorz, C. and Chung, C. A. (1977)
Morphological and biochemical observations on
cells derived from the in vitro differentiation of the
embryonal carcinoma cell line PCC4-F. Cancer
Res., 37: 2072-2081.

Allen, D. W. and Zamecnik, P. C. (1962) The effect
of puromycin on rabbit reticulocyte ribosomes.
Biochim. Biophys. Acta, 55: 865-874.

Kedinger, C., Gniazdowski, M., Mandel, J. L. Jr.,
Gissinger, F. and Chambon, P. (1970) a-Amanitin:
a specific inhibitor of one of two DNA-dependent
RNA polymerase activities from calf thymus.
Biochem. Biophys. Res. Commun., 38: 165-171.
Davidson, E. H. (1986) Gene Activity in Early
Development. 3rd ed. Academic Press, New York.
Johnson, M. H. (1981) The molecular and cellular
basis of preimplantation mouse development. Biol.
Rev., 56: 463-498.

Par o

+ creel Niessen etl

ee ae ere. PR ey
pig 5 a git) ay (Owes) ¥ ate Aap aT aM
Pot aan 4 7 als oe ete ee } ih.

on On ;
‘1 ly Wea : jan raph 3 annp® an or) a 4 Mi

‘ y tes Qs aie ah eS liintebetin ye =
* any en ee 7 > UR cae ‘salted at ORat 7
fa b eylestl®. 3 “ eraeticolhy freeads
ne 5 line-in

Dane giiveg Aw Ty Mf Is Bits

isflbeast' t sf Ct ce: breve oH eERe
arith - aa GAP oS aia

vine PR sient oir AP ting 50 tee | .
ba) a0 wut) YO tacenen cote Let

ale Ms Lye fiat, fa, ng

oN a¥ ra] Fs eh i oaeheseey
nenmok obi. ae ve TT BNIBI Sah As:

1 laa 7) | SAS Sep ely 7 OK
m Hy eteviti aha peAaen a

rel wei tine abana seeen

¥

Tic i

159

INSTRUCTIONS TO AUTHORS

ZOOLOCIGAL SCIENCE publishes contri-
butions, written in English, in the form of (1)
Reviews, (2) Articles, and (3) Communications of
material requiring prompt publication. A Review
is usually invited by the Editors. Those who
submit reviews should consult with the Editor-in-
Chief or the Managing Editor in advance. Articles
of less than 6 printed pages and Communications
less than 3 printed pages will be published free of
charge. Charges will be made for extra pages
(10,000 yen/page). A Communication cannot
exceed 4 printed pages. No charge will be imposed
for invited reviews up to 15 printed pages. No free
reprints of Articles and Communications are avail-
able. To the author of an invited review 50
reprints are provided gratis. Submission of papers
from nonmembers of the Society is welcome.
However, page charges (10,000 yen/page) will be
made to nonmembers.

A. SUBMISSION OF MANUSCRIPT

The manuscript should be submitted in tripli-
cate, one original and two copies, each including
all illustrations. Rough copies of line drawings and
graphs may accompany the manuscript copies, but
the two copies of continuous-tone prints (photo-
micrographs, etc.) should be as informative as the
original. The manuscript should be sent to:

Dr. Cxiraru Ocuro, Managing Editor,
ZOOLOGICAL SCIENCE, Department of Biology,
Faculty of Science, Toyama University,
Toyama 930, Japan.

B. CONDITIONS

All manuscripts are subjected to editorial re-
view. A manuscript which has been published or
of which a substantial portion has been published
elsewhere will not be accepted. It is the author’s
responsibility to obtain permission to reproduce
illustrations, tables, etc. from other publications.
Accepted papers become the permanent property
of ZOOLOGICAL SCIENCE and may not be
reproduced by any means, in whole or in part,
without the written consent of both the Zoological

Society of Japan and the author(s) of the article in
question.

C. ORGANIZATION OF MANUSCRIPT

The desirable style of the organization of an
original paper is as follows: (1) Title, Author(s)
and Affiliation (2) Abstract (3) Introduction (4)
Materials and Methods (5) Results (6) Discussion
(7) Acknowledgments (8) References (9) Tables
(10) Illustrations and Legends (11) Footnotes. The
author is not obliged to adhere rigidly to this
organization. He or she may modify the style
when such modification makes the presentation
clearer and more effective. In a Communication,
combination of some of these sections is recom-
mended. There is no restriction on the style of
review articles.

D. FORM OF MANUSCRIPT

Manuscripts should be typewritten and double
spaced throughout on one side of white type-
writing paper with 2.5cm margins on all sides.
Abstract not exceeding 250 words, tables, figure
legends and footnotes should be typed on separate
sheets. All manuscript sheets must be numbered
successively. The use of footnotes to the text is not
recommended.

1. Title page

The first page of manuscript should contain title,
authors’ names and addresses of university or
institution, abbreviated form of title (40 characters
or less, including spaces), and name and address
for correspondence. Authors with different affilia-
tions should be identified by the use of the same
superscript on name and affiliation. If one or more
of the authors has changed his or her address since
the work was carried out, the present address(es)
to be published should be indicated in a footnote.
In addition, a sub-field of submitted papers to be
used as heading of an issue may be indicated in the
first page. Authors are encouraged to choose one
of the following: physiology, cell biology, molecu-
lar biology, genetics, immunology, biochemistry,
developmental biology, reproductive biology,

160

endocrinology, behavior biology, ecology, phy-
logeny, taxonomy, or others (specify). However,
the Editors are responsible for the choice and
arrangement of headings.
2. Introduction

This section should clearly describe the objec-
tives of the study, and provide enough background
information to make it clear why the study was
undertaken. Lenghty reviews of past literature are
discouraged.
3. Materials and Methods

This section should provide the reader with all
the information that will make it possible to repeat
the work. For modification of published methodol-
ogy, only the modification needs to be described
with reference to the source of the method.
4. Results

Results should be presented referring to tables
and figures, without discussion.
5. Discussion

The Discussion should include a concise state-
ment of the principal findings, a discussion of the
validity of the observations, a discussion of the
findings in the light of other published works deal-
ing with the same subject, and a discussion of the
significance of the work. Redundant repetition of
material in Introduction and Results, and extensive
discussion of the literature are discouraged.
6. Statistical analysis

Statistical analysis of the data using appropriate
methods is mandatory and the method(s) used
must be cited.
7. References

References should be cited in the text by an
Arabic numeral in square parentheses and listed at
the end of the paper in numerical order. For
example:

1 Takewaki, K. (1931) Oestrus cycle of female
rat in parabiotic union with male. J. Fac. Sci.
Imp. Univ. Tokyo, Sec. IV, 2: 353-356.

2 Shima, A., Ikenaga, M.., Nikaido, O.,
Takabe, H. and Egami, N. (1981) Photoreac-
tivation of ultraviolet light-induced damage in
cultured fish cells as revealed by increased
colony forming ability and decreased content
of pyrimidine dimers. Photochem. Photo-
biol., 33: 313-316.

3 Hubel,O. and Wiesel, T.N. (1986) The

functional architecture of the striated cortex.
In “Physiological and Biochemical Aspects of
Nervous Integration”. Ed. by F. D. Carlson,
Prentice-Hall, New Jersey, pp. 153-161.

4 Campbell, R. C. (1974) Statistics for Biolo-
gists. Cambridge Univ. Press, London, 2nd
ed., pp. 59-61.

Titles of cited papers may be omitted in Reviews
and Communications. The source of reference
should be given following the commonly accepted
abbreviations for journal titles (e,g., refer to
‘International List of Periodical Title Abbrevia-
tions’). The use of “in preparation”, “submitted
for publication” or “personal communication” is
not allowed in the reference list. “Unpublished
data” and “Personal communication” should
appear parenthetically following the name(s) in
the text. Text citations to references with three or
more authors should be styled as, e.g., Everett et
al. [7].

E. ABBREVIATIONS

Abbreviations of measurement units, quantity
units, chemical names and other technical terms in
the body of the paper should be used after they are
defined clearly in the place they first appear in the
text. However, abbreviations that would be rec-
ognized by scientists outside the author’s field
may be used without definition, such as SD, SE,
DNA, RNA, ATP, ADP, AMP, EDTA, UV, and
CoA. The metric system should be used for all
measurements, and metric abbreviations (Table)

Table. Abbreviations for units of measure which

may be used without definition

Length: km, m, cm, mm, “«m, nm, pm, etc.

Area: ' km?, m?, cm?, mm? yan”, nm?, pm?, etc.

Volume: km?*, m°, cm*, mm*, «m?, nm’, pm’, kl,
liter (always spellout), ml, sl, nl, etc.

Weight: kg, g, mg, “g, ng, pg, etc.

Concentration: M, mM, «M, nM, %, g/l, mg/l,
ug/l, etc.

Time: hr, min, sec, msec, psec, etc.

Other units: A, W, C, atm, cal, kcal, R, Ci, cpm,

dB, v, Hz, lx, Xg, rpm, S, J, IU, ete.

should, in general, be expressed in lower case
without periods.

F. PREPARATIONS OF TABLES

Tables should only include essential data needed
to show important points in the text. Each table
should be typed on a separate sheet of paper and
must have an explanatory title and sufficient
explanatory material. All tables should be referred
to in the text, and their approximate position
indicated in the margin of manuscript.

G. PREPARATION OF ILLUSTRATIONS

All figures should be appropriately lettered and
labelled with letters and numbers that will be at
least 1.5 mm high in the final reproduction. Note
the conventions for abbreviations used in the
journal so that usage in illustrations and text is
consistent. All figures should be referred to in the
text and numbered consecutively (Fig. 1, Fig. 2,
etc.). The figures must be identified on the reverse
side with the author’s name, the figure number and
the orientation of the figure (top and bottom). The
preferred location of the figures should be indi-
cated in the margin of the manuscipt. Illustrations
that are substandard will be returned, delaying
publication. Illustrations in color may be pub-
lished at the author’s expense.

1. Line drawings and graphs

Original artwork of high quality, glossy prints
mounted on appropriate mounting card (less than
25 X 38 cm) should be submitted for reproduction.
Author(s) may indicate size preference by making
on the back of figures, such as “Do not reduce”,
“Two-column width” (no wider than 14cm), or
“One-column width” (no wider than 7 cm). Lines
must be dark and sharply drawn. Solid black,
white, or bold designs should be used for histo-
grams. Xerox or any other copying mean may be
used for the two review copies.

161

2. Continuous-tone prints

Three sets of continuous-tone prints (photo-
micrographs, etc.) must be submitted. One set for
reproduction should be mounted on appropriate
mounting card, and the other two for reviewers
may be unmounted prints. Xerox or similar copies
of photomicrographs are not acceptable for review
purposes. The continuous-tone prints should be
submitted preferably at the exact magnification
which is to be used in the published papers and
trimmed to conform to the page size (in no case
should it exceed 14X20cm). Press-on numbers
should be applied to the lower right corner of
individual prints. Letters (a, b, c, etc.) should be
used for multiple parts of a single figure. If
important structures will be covered by use of the
lower right corner, identification may be applied in
the lower left corner.

Reproduction of color photographs will have to
be approved by the Editors. The extra costs of
color reproduction will be charged to the authors.
3. Figure legends

Each figure should be accompanied by a title
and an explanatory legend. The legends for
several figures may be typed on the same sheet of
paper. Sufficient detail should be given in the
legend to make it intelligible without reference to
the text.

H. PROOF AND REPRINTS

A galley proof and reprint order will be sent to
the submitting author. The first proofreading is
the author’s responsibility, and the proof should be
returned within 72 hours from the date of receipt
(by air mail from outside Japan). The minimum
quantity for a reprint order is fifty. Manuscript,
tables and illustrations will be discarded after the
editorial use unless their return is requested when
the manuscript is accepted for publication.

AS

7
’
a
Zz
?
1 ’
ft
é
- TL eed
7 Tite +” BT ees. i
: (Pe ah ‘ nc
1 hoa ‘ Patan
bietth SASS opines
A .
: J ; “ai Levin pei? Sev Wie
hey ye tration arty.
; :
eh OPT. cpu)
ries wt This tt
i / re ts ry (al bani
5 " ib fy Stet
ry y! ve 4 msn ? Whit
srontr<e 94 Cited
G r Al de gh it
; Ryo) e|
; orrnernit
are
j
]
- : 4
“

yab-xsis nasbridkole pre
A i ee bephisbose :
"huis Wong ip Vigad
EO Cima pila vite
feat Shan by} LAI lie birt; rile:

i oa, svete anes iw

peo Ae a a

vi ;
7 4Q 1] ‘
a PPPS Anan; — oth
Woot Aire leit, Mei aven tae é
Nig wort PACIEN Song r Voted offiga
Fe ON, Pitney peta ieee
PP oni ebieermert ely
eth) oe Ae) ero! binateedix
ati, 4 SghT) einvitiedenus
10 wie bellegi. 7] wore ame

PA he Mi vie wee

a oo .
Twlignisty PIuyi 2a) , RIE 4

a9 Cit thw vy vert aid
ine @ Said ove iT sr eye of
wtriytpa ater cke a Rem
FORFAR <P 48: att Hw av it r

at) 1 Spar j ait iy

sep eae
tnceag oN vp

i eye ito tale
ee were Star fee co
apnat ; rivtoe Ge
enbar te

J jae Le

Warts He Vey OR .

Vit Lat “CE

: ca rede 8 od
f SIME? € Oy esgby a ec

O03. 19 ty Vite King

loo waehre?

oy Nees

4
}

D | nt Published Bimonthly by the Japanese Society of
eve opm e Developmental Biologists
Distributed by Business Center for Academic

Growth & Differentiation Societies Japan, Academic Press, Inc.

Papers in Vol. 32, No. 1. (February 1990)

1. REVIEW: H.Ide: Growth and differentiation of limb bud cells in vitro: Implications for
limb pattern formation

2. K. Akasaka, T. Ueda, T. Higashinakagawa, K. Yamada and H. Shimada: Spatial patterns of
arylsulfatase mRNA expression in sea urchin embryo

3. I. Kaneda, E. Kamitsubo and Y. Hiramoto: The mechanical structure of the cytoplasm of the
echinoderm egg determined by “Gold Particle Method” using a centrifuge microscope

4. P.D.Nieuwkoop and B. Albers: The role of competence in the cranio-caudal segregation of
the central nervous system

5. H.Katow: A new technique for introducing antifibronectin antibodies and fibronectin-related
synthetic peptides into the blastulae of the sea urchin, Clypeaster japonicus

6. K.H. Kato, S. Washitani-Nemoto, A. Hino and S. Nemoto: Ultrastructural studies on the
behavior of centrioles during meiosis of starfish oocytes
M. Fukumoto: The acrosome reaction in Ciona intestinalis (Ascidia, Tunicata)
R. Masho: Close correlation between the first cleavage and the body axis in early Xenopus
embryos

9. H. Yamasaki, C. Katagiri and N. Yoshizaki: Selective degradation of specific components of
fertilization coat and differentiation of hatching gland cells during the two phase hatching of
Bufo japonicus embryos

10. Y.K. Maruyama and M. Shinoda: Archenteron-forming capacity in blastomeres isolated from
eight-cell stage embryos of the starfish, Asterina pectinifera

11. H.Ohtsuki: Statocyte and ocellar pigment cell in embryos and larvae of the Ascidian, Styela
plicata (Lesueur)

12. K.Asamoto, Y. Nojyyo and H. Aoyama: Do peanut agglutinin receptors on somites control
the behavior of neural cells?

Development, Growth and Differentiation (ISSN 0012-1592) is published bimonthly by The Japanese Society of
Developmental Biologists, Department of Developmental Biology, Mitsubishi Kasei Institute of life Science,
Minami-ootani 11, Machida, Tokyo 194, Japan. 1989: Volume 31. Annual subscription for Vol. 32, 1990: U. S.$
148,00, U.S. and Canada: U. S. $ 163,00, all other countries except Japan. A\ll prices include postage, handling
and air speed delivery except Japan. Second class postage paid at Jamaica, N.Y. 11431, U.S. A.

Outside Japan: Send subscription orders and notices of change of address to Academic Press, Inc., Journal
Subscription Fulfillment Department, 1 East First Street, Duluth, MN 55802, U.S. A. Send notices of change of
address at least 6-8 weeks in advance. Please include both old and new addresses. U.S.A. POSTMASTER:
Send changes of address to Development, Growth and Differentiation, Academic Press. Inc., Journal Subscription
Fulfillment Department, 1 East First Street, Duluth, MN 55802, U.S. A.

In Japan: Send nonmember subscription orders and notices of change of address to Business Center for
Academic Societies Japan, 16-3, Hongo 6-chome, Bunkyo-ku, Tokyo 113, Japan. Send inquiries about
membership to Business Center for Academic Societies Japan,4—16, Yayoi 2-chome, Bunkyo-ku, Tokyo 113, Japan.

Air freight and mailing in the U.S. A. by Publications Expediting, Inc., 200 Meacham Avenue, Elmont, NY
11003, U.S.A.

NARISHIGE

THE ULTIMATE NAME IN MICROMANIPULATION

OUR NEW MODELS WR-88 and MO-102M
MAKE PRECISION MICROMANIPULATION SO EASY!

SOME FEATURES of THE WR-88 WATER ROBOT MICROMANIPULATOR (3-DIMENSIONAL)
* Drift-free, the new WR-88 has a DRIFT movement of less than 2 microns.
* The new WR-88 has a SMOOTH MICRODRIVE MECHANISM.
* An Aqua Purificate remote control ensures totally vibration-free operation.

THARISHIGE) NARISHIGE SCIENTIFIC INSTRUMENT LAB.

akan, 9-28 KASUYA 4-CHOME SETAGAYA-KU, TOKYO 157, JAPAN
PHONE (INT-L) 81-308-8233, FAX (INT-L) 81-3-308-2005

CABLE : NARISHIGE LABO, TELEX, NARISHIGE J27781

(Contents continued from back cover)

Taxonomy and Systematics
Abe, H.: Two species of the genus Actacarus
(Acari, Halacaridae) from Japan ......... 111
Yunxia,T. and Y.Shaoyi: Comparative
study on LDH isozymes in different subfami-
ly of teleost fish - grass carp (Ctenophryngo-
don idellus) and blunt snout-bream (Mega-
lobrama amblycephala)
Watabe, H., X.C. Liang and W. X. Zhang:

The Drosophila robusta species-group (Dip-
tera: Drosophilidae) from Yunnan Province,
southern China, with the revision of its geo-
graphic distribution
Ando, A., S.Shiraishi and T. A. Uchida:
Reexamination on the taxonomic position of
two intraspecific taxa in Japanese Eothe-
nomys: evidence from crossbreeding experi-
ments (Mammalia: Rodentia) ............ 141

Instructions to authors

ZOOLOGICAL SCIENCE

VOLUME 7 NUMBER 1

FEBRUARY 1990

CONTENTS

REVIEWS
Gotoh, T. and T. Suzuki: Molecular assembly
and evolution of multi-subunit extracellular
annelidhemoglobins
Lawrence, J. M.: The effect of stress and dis-
turbance on echinoderms

ORIGINAL PAPERS

_ Cell Biology and Morphology
Low, W. P., Y.K-Ip and D.J.W. Lane: A
comparative study of the gill morphology in
the mudskippers - Periophthalmus chrysospi-
los, Boleophthalmus boddaerti and Perioph-
oe thalmodon schlosseri
Kondo, H., Y. Yonezawa and T. A. Noma-
guchi: Difference in migratory ability be-
‘tween human lung and skin fibroblasts .... 39

Immunology

Kono, H., A. Mizoguchi, H. Nagasawa, H.
Ishizaki, H.Fugo and A. Suzuki: A
monoclonal antibody against a synthetic car-
boxyl-terminal fragment of the eclosion hor-
mone of the silkworm, Bombyx mori: char-
acterization and application to immunohis-
tochemistry and affinity chromatography .. 47

Biochemistry
Michibata,H., T.Uyama and _ J. Hirata:
Vanadium-containing blood cells (vanado-
cytes) show no fluorescence due to the
tunichrome in the ascidian, Ascidia sydneien-
sis samea

Developmental Biology
Ma, Y. K. andS. B. Ramaswamy: Histochemis-
try of yolk formation in the ovaries of the
tarnished plant bug, Lygus lineolaris (Palisot

de Beauvois) (Hemiptera: Miridae) (COM-
MUNICATION)
Kamishima, Y.: Organization and develop-
ment of reflecting platelets in iridophores of
the giant clam, Tridacna crocea Lamarck

Ishikawa, T.: Effects of puromycin and a-
amanitin on the activity of alkaline phosphat-
ase in early preimplantation mouse embryos
(COMMUNICATION)

Uchiyama, M., T. Murakami and H. Yoshiza-
wa: Notes on the development of the crab-
eating frog, Rana cancrivora

Reproductive Biology

Asada, N. and T. Fukumitsu: Reaction mass
formation in Drosophila, with notes on a
phenoloxidase activation
Jarosz,S.J. and W.R.Dukelow: Embryo
transfer and pregnancy rate in the golden
hamster (Mesocricetus auratus)

Endocrinology
Tagawa, M., S. Miwa, Y. Inui, E. G. de Jesus
and T. Hirano: Changes in thyroid hor-
mone concentrations during early develop-
ment and metamorphosis of the flounder,
Paralichthys olivaceus
Tasaki, Y. and S. Ishii: Effects of thyroidec-
tomy, hypophysectomy, temperature and
humidity on the occurrence of nocturnal
locomotor activity in th toad, Bufo japoni-
cus, during the breeding season
Uchiyama, M. and T. Murakami: Effects of
hypophysectomy and replacement therapy

-with several hormones on plasma sodium
concentrations in bullfrog tadpoles

(Contents continued on inside back cover)

INDEXED IN: ,
Current Contents/LS and AB & ES,
Science Citation Index,

ISI Online Database,
CABS Database, INFOBIB

Issued on February 15
Printed by Daigaku Letterpress Co., Ltd.,
Hiroshima, Japan

ZOOLOGICAL SCIENCE

The Official Journal of the Zoological Society of Japan

Editorssin-Chict: ; The Zoological Society of Japan:
Seiichiro Kawashima (Tokyo) Toshin-building, Hongo 2-27-2, Bunkyo-ku,

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

Chitaru Oguro (Toyama)

5 President: Hiromichi Morita (Fukuoka)
Assistant Editors:

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

Hitoshi Michibata (Toyama) Librarian: Masatsune Takeda (Tokyo)
Miéko Komatsu (Toyama) ——___________—________ an

Editorial Board:

Howard A. Bern (Berkeley) Walter Bock (New York) Aubrey Gorbman (Seattle)

Horst Grunz (Essen) Robert B. Hill (Kingston) Yukio Hiramoto (Chiba)
Susumu Ishii (Tokyo) Yukiaki Kuroda (Mishima) John M. Lawrence (Tampa)
Koscak Maruyama (Chiba) Roger Milkman (Okazaki) Kazuo Moriwaki (Mishima)
Tokindo S. Okada (Okazaki) Andreas Oksche (Giessen) Hidemi Sato (Nagoya)

Hiroshi Watanabe (Tokyo) Mayumi Yamada (Sapporo) Ryuzo Yanagimachi (Honolulu)

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

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

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

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

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

ZOOLOGICAL SCIENCE 7: 159-170 (1990)

REVIEW

Mesenchymal-Epithelial Interactions in the
Organogenesis of Digestive Tract

SADAO YasuGt and TakKEOo MizuNo~

‘Zoological Institute, Faculty of Science, University of Tokyo, Tokyo 113,
*Faculty of Pharmaceutical Sciences, Teikyo University,
Kanagawa 199-01, Japan

ABSTRACT—It is now well established that the mesenchymal-epithelial interactions are essential in
the spacio-temporally regulated differentiation processes in organogenesis of higher vertebrates. The
mesenchymal-epithelial interactions in the digestive tract (DT) in avian and mammalian embryos have
presented precious information on the nature of interaction, because the endoderm that later gives rise
to the DT epithelium is a simple sheet consisited of undifferentiated cells at the early stages and with
the development of DT the epithelium gradulally acquires region-specific morphology and cytodif-
ferentiation under the influence of the underlying mesenchyme. The studies carried out have
demonstrated that the mesenchyme is dominant over the epithelium in determining the morphology of
the latter and that the morphogenesis and cytodifferentiation are not always associated. Also the
importance of epithelial competence against mesenchymal inductive influence has been repeatedly
stressed. In this article we summarize these results and our recent findings to clarify the present status

© 1990 Zoological Society of Japan

of the problem and to suggest the direction of further investigations.

INTRODUCTION

Regional differentiation of tissues in the animal
development has been shown to be often brought
about by the influences generated by other tissues
within the embryo and transmission of signals
between tissues which evoke specific response in
the target tissue has been called ‘induction’ [1-3].
Induction has been postulated to work on the
levels of morphology and cytodifferentiation, but
the exprimental dissociation of the two levels has
been by no means easy. Moreover, studies have
revealed that the induction processes are reciproc-
al in many systems, and that we can distinguish
‘instructive’ and ‘permissive’ inductions ex-
perimentally [4]. The nature of inductive process,
however, has been not fully elucidated so far,
partly because of time-consuming experimental
systems and partly because of our lack of know-

Received November 25, 1989

ledge of events evoked in responding tissues. Re-
cently more refined experimental systems have
been developed as to the mesodermal induction in
amphibian embryos with using the specific gene
transcripts as a differentiation marker of respond-
ing tissue [5, 6]. However, the epithelial-
mesenchymal interactions working in the orga-
nogenesis of many organs are still in the vague,
although their importance is fully recognized.

The developping digestive tract (DT) of higher
vertebrates such as mammals and birds provides us
the opportunity to distinguish morphological and
functional differentiation and offers us definitive
markers of epithelial differentiation such as pepsin-
ogens for the stomach and disaccharidases for the
intestine. From these reasons, the DT is among
the well-studied organ systems in the field of
research on tissue interactions.

In this article, we will confine ourselves to the
problems of effects of DT mesenchymes on the
morphogenesis and cytodifferentiation of DT

160 S. YASUGI AND T. MizuNo

epithelia, not dealing with the accessary glands of
DT, i.e. the salivary glands, liver and pancreas,
although our laboratory has been deeply con-
cerned in studies of development of these organs
[7-10].

THE DEVELOPMENT OF DIGESTIVE
TRACT IN AVIAN AND MAMMALIAN
EMBRYOS

The chicken DT consists of esophagus, proven-
triculus (glandular stomach), gizzard (muscular
stomach), small and large intestines. Anatomically
these organs can be recognized from day 3 to 4 of
incubation [11]. The epithelial tissue is composed
of simple columnar cells throughout the DT at this
stage, and no organ-specific structures such as
stomach glands or intestinal villi are seen.

In the chicken, the invagination of proventricu-
lar glands into the surrounding mesenchyme be-
gins at day 6 [12]. Glands penetrate deep into the
mesenchyme and repeat branching, forming the
complex glands [13]. At day 13-15, the epithelial
cells in the glands become cuboidal, while the
luminal mucous epithelial cells and duct cells re-
main columnar. The embryonic pepsinogen, the
zymogen of embryonic pepsin, appears first at day
9 [14-18]. The production of pepsinogen reaches
maximum at day 15, and ceases to almost zero
level in the latest period of embryonic life. The
adult-type pepsinogens appear just after the
hatching.

The epithelium of embyonic chicken gizzard
does not form any complex glands. The cells are
high columnar in form, contain many glycogen
particles and actively secrete mucus. After day 12
of incubation, proliferation of epithelial cells be-
come active and at the same time there occur the
vertical striations of the epithelium. As a result of
these changes, gizzard epithelium comes to possess
finger-like projections in the later period of
embyonic life. The cells still actively secrete mucus
and secreted substances form a horny layer at the
luminal surface. In the gizzard mesenchyme the
muscle develops well [12].

The avian intestine can be divided into duode-
num, small intestine, caecum and large intestine.
At day 4 of incubation, the intestine is a very

simple tube, but in the following days, it elongates
rapidly and form loops at duodenum level and
small intestine. Large intestine can be recognized
from day 5 on and always shorter and thicker than
other part of intestine. The histological examina-
tion revealed that the villi formation begins at day
9 to 12, depending on the level of intestine, first at
duodenum and later in more caudal parts. The
brush border can be observed from day 12 and
alkaline phosphatase activity from day 9-17 [19].
Disaccharidase (sucrase) can be detected by im-
munoelectron microscopy from day 8-10, histoche-
mically and biochemically from day 14 [20-22].

The stomach of mouse and rat consists of two
parts: the forestomach and glandular stomach. In
the forestomach, epithelium has 2 to 3 layers of
cells till the day 17 of gestation, becomes pluristra-
tified at day 18, and first keratinization can be
observed on day 19. In glandular part, pluristra-
tification occurs at day 15-16, and intraepithelial
vacuoles communicate with stomach lumen and
become primitive pits of the stomach. At day 18,
the epithelium becomes simple columnar, and
villus-like structures develop. Morphologically
identifiable parietal cells appear just before birth,
while chief cells can be distinguished before wean-
ing [23, 24]. In the mouse stomach a pepsinogen
can be detected biochemically from day 16 of
gestation. In the period just before birth another
pepsinogen appears and the ratio of the two pepsin-
ogens varies as the development proceeds [25].

The intestinal epithelium of the rat and mouse is
pluristratified until day 15-18 of gestation, but
from day 19 onward the formation of villi begins
and the epithelium becomes simple columnar.
Disaccharidases in the mouse intestine are low in
activity before birth and become active after va-
rious periods depending on the kind of emzymes
[26].

From this brief description of development of
stomach and intestine of bird and mammal, it is
obvious that these organs have their specific mor-
phologies and functional specificities even in
embryonic and fetal period. Considering that
these organs are derived from a single, continous
tube, it is easily conceivable that the differentia-
tion of these organs provides us very suitable
model for the analysis of tissue interactions.

Tissue Interactions in Gut Development 161

DETERMINATION OF MORPHOLOGICAL
FEATURES OF EPITHELIUM BY
ASSOCIATED MESENCHYME

Balinsky first studied the experimental altera-
tion of developmental fate of DT epithelium by the
mesenchyme: He tranplanted the presumptive
stomach region of axolotl to the presumptive re-
gion of liver of Triton and found that the stomach
epithelium became liver parenchymal cells [27].
Okada [28] also analyzed the developmental fate
of presumptive digestive organ epithelium trans-
planted heterotopically in the amphibian embyos
and reported the presence in mesoderm of anter-
iorizing effect against endoderm, and stated that
the developmental fate of endoderm is determined
by both mesodermal factors and endodermal reac-
tivity.

After these pioneering works, the studies of
epithelial-mesenchymal interactions in the DT de-
velopment have been carried out mainly in two
groups, one in France led by Etienne Wolff and
another in Japan led by one of the present authors
(T.M.).

The exact location of fate map of DT is prere-
quisite for the experimental studies. This was
achieved in chicken with the use of carbon marking
method and X-ray-induced lesion [29] and with the
use of *H-labelled tissues [30]. In the mammals the
presumptive fate of the endodermal cells was not
studied systematically, because of the difficulty in
treating early embyos, and almost all the experi-
ments were carried out on the DT of fetuses of
latter half of gestation period.

Until about the mid 1970s, the experiments on
the epithelial-mesenchymal interaction in DT were
dependent exclusively on the morphological
criteria. The situation was the same in other
systems such as skin [31], tooth [4], kidney [32] and
so on. Such morphological studies indefinitely
showed the dominance of mesenchymal tissue over
epithelial tissue in the determination of morpholo-
gical traites of epithelial tissue in many instances.
We will describe some examples below.

When the chicken epidermis was associated and
cultured with mesenchyme of chicken gizzard,
epidermal cells, which are determined to kerati-
nize in the normal development, showed mucous

metaplasia [33], thus showing the inducing effect
of gizzard mesenchyme. In the chicken stomach,
the epithelial-mesenchymal interaction between
proventriculus and gizzard was first studied by
Sigot [34, 35]. He demonstrated that proventricu-
lar epithelium of 5- to 6-day embryos differentiates
into gizzard-type epithelium when it was cultivated
in vitro alone or with gizzard mesenchyme. On the
other hand, proventricular mesenchyme induces
gland formation in associated gizzard and intestin-
al epithelia, and the inductive ability of proven-
tricular mesenchyme is highest on day 5 and dis-
appears at day 7. David [36-38], who carried
out the similar experiments with fetal rabbit stom-
ach, concluded that there exist two inducing fac-
tors in the mesenchyme; first one is highly duffusi-
ble and induces epithelial morphogenesis, the
second one is lightly diffusible and evokes cytodif-
ferentiation. Naturally the second factor is pro-
duced in the later stage of development.

These studies demonstrated that the develop-
mental fate of DT epithelium can be changed in
some cases by the association of heterologous
mesenchymes. But in the mouse DT, Soriano [39]
and Fukamachi et al. [40, 41] reported the early
determination of developmental fate of the DT
epithelium. Fukamachi et al. carried out extensive
tissue-recombination experiments in vitro using
forestomachs and glandular stomachs and con-
cluded that the developmental fate of the epithe-
lium was not affected by the associated heterotypic
and heterochronic mesenchymes, but the degree of
differentiation was affected by the type of mesen-
chyme.

A group of researchers led by Mizuno studied
extensively the inductive ability of DT mesen-
chyme on the epithelia of DT and embyonic mem-
branes in the chicken embryos. Yasugi and Mizu-
no [42] associated the allantoic endoderm of 3- to
4-day embryos with various mesenchymes of DT of
day 6 and older [43] and cultivated in vitro [44] or
on the chorio-allantoic membrane (CAM) of 9-day
embyos [45]. They demonstrated that morpholo-
gical characters of allantoic endoderm were largely
determined by the mesenchyme: when it was
associated with esophageal mesenchyme it diffe-
rentiated into pluristratified epithelium; with
proventricular mesenchyme, it formed very com-

162 S. YASUGI AND T. MIZUNO

plex glands; with gizzard mesenchyme, it diffe-
rentiated into highly columnar epithelium with
high glycogen content; with intestinal mesenchyme
it formed villus-like structures with simple colum-
nar cells. Also, when the allantoic endoderm was
implanted into the presumptive digestive area of
young chicken embryos and hosts were kept alive
until the latter half of the embryonic period or
even after hatching, the endoderm differentiated
heterotypically under the influence of surrounding
mesenchymes [46, 47].

The inductive influence of DT mesenchyme
seemed not restricted against allantoic endoderm.
The DT epithelium could also more or less re-
spond to the influence of DT mesenchyme
[48](Table 1). Most remarkably, gizzard epithe-
lium differentiated into proventricular epithelium
and formed complex proventricular glands when it
was associated with proventricular mesenchyme.
In contrast, in the association of proventricular
epithelium and gizzard mesenchyme, the epithe-
lium never formed glands found in the normal

TasLeE 1. Homotypic and heterotypic dif-
ferentiation of epithelia associated with
various mesenchymes and cultivated on the
chorio-allantoic membrane. Upper figure
in each combination represents the percen-
tage of homotypic differentiation, lower
figure heterotypic differentiation. The
sum of upper and lower figures is not
necessarily 100 because some explants
underwent both homotypic and heterotypic
differentiation and some others remained
undifferentiated. From Yasugi and Mizu-

no [48]
Mesenchyme associated

Epithelium
ES PV GZ SI
ES 86 50 88 100
0 100 24 0
PV 0 100 0 0
7) 0 100 50
GZ 100 0) 100 0

0) 100 0 71
SI 75 100 100 100
34 0 0 0

ES: Esophagus, PV: Proventriculus, GZ: Gizzard,
SI: Small intestine.

proventriculus, but is differentiated into the
epithelium rather similar to the gizzard epithelium.
Moreover, intestinal mesenchyme showed strong
inductive influence against gizzard epithelium [48,
49].

Masui [50, 51] and Yasugi and Mizuno [52]
respectively demonstrated the flexibility of dif-
ferentiation potency of yolk sac endoderm and
hypoblast of young chicken embryos under the
influence of DT mesenchymes.

As to self-differentiation potency of the DT
epithelium, studies of in vitro cultivation of iso-
lated fragments of the endoderm have been per-
formed extensively at Mizuno’s laboratory, and we
found that region-specific self-differentiation
potency appears in young presumptive endoderms
of DT epihtelium [53-57]. Based on these and the
above-mentioned results, Mizuno [58] postulated a
hypothesis that the developmental fate of DT
epithelium may be determined by both self-
differentiation potency of the endoderm and the
inductive influences of the mesenchyme.

Epithelial-mesenchymal interactions were de-
monstrated also with isolated and cultured cells.
Fukamachi et al. [59, 60] showed the gland forma-
tion of human colon cancer cells (LS174T) main-
tained in vitro when the cells were recombined
with fetal rat mesenchyme. They concluded that
the desmosome formation in epithelial cells is
important for the establishment of epithelial polar-
ity and gland formation [61].

In France, Haffen and her collegues studied
actively the differentiation of intestinal epithelial
cells and have demonstrated that fetal rat intestinal
mesenchyme could permit the normal differentia-
tion of cultured rat intestinal epithelial cells so that
the latter: formed villus-like structures, and that
fibroblasts obtained from suckling rat duodenum
could induce intestinal differentiation in
embryonic chicken gizzard epithelium [62-65].
Also, epithelioid cultures derived from rat and
quail intestine were able to differentiate into intes-
tinal epithelium when they were associated with
chicken embryonic intestinal mesenchyme [66].

These results mentioned above unequivocally
demonstrate the importance of mesenchymal in-
fluence in determining the epithelial morphology
in the development of avian and mammalian DT

Tissue Interactions in Gut Development 163

on one hand, and that of the endodermal epithe-
lium as a responding tissue on the other hand.

REGULATION OF ORGAN-SPECIFIC GENE
EXPRESSION AND PROTEIN SYNTHESIS
BY ASSOIATED MESENCHYME

The most important aim of the current develop-
mental biology is to clarify the mechanisms reg-
ulating the differential expression of genes during
the course of development. Although the morpho-
logical differentiation of an organ is based on the
local expression or suppresion of sets of genes, we
have at present no valid information about the
genes involved in the morphogenesis. The role of
mesenchyme in the epithelial differentiation must
also be analyzed from the view point of its effects
on the expression of specific genes. As stated
above, digestive organs have advantage in that
they possess many organ-specific marker proteins.
The use of these markers as criteria of epithelial
cell differentiation, that began at mid 1970s,
brought about fruitful information.

It was already mentioned that the allantoic
endoderm can differentiate morphologically into
the epithelia of organs from which associated
mesenchymes were derived. Yasugi [10] made use
of anti-glucagon antibody to examine whether the
allantoic endodermal cells can synthesize glucagon
under the influence of pancreatic or intestinal
mesenchyme after the recombinants were culti-
vated on the CAM. He obtained the positive
results: both mesenchymes induced the anti-
glucagon antibody-positive cells in the epithelium.

As for the induction of endocrine cells in the
epithelium, Andrew and Rawdon carried out va-
rious recombination experiments and reported
that the chicken mesenchymes of DT often induce
ectopic endocrine cells in associated epithelia. For
example, a wide variety of endocrine cell types
differentiated in the gizzard and allantoic epithelia
which are normally devoid of such cells [67-69].

The intestinal epithelial cells are characterized
by the presence of brush border on the apical cell
surface and the presence of disaccharidases on the
brush border membrane. Among various di-
saccharidases, sucrase was purified and antibodies
were raised by several workers, mainly because of

the abundance and ease of purification procedure
of this enzyme.

In Japan, Matsushita studied the ontogeny of
sucrase in chicken embryo and purified it and
obtained the polyclonal antibody [21]. Using this
antibody, he could show that the allantoic en-
doderm expresses sucrase in recombination with
DT mesenchyme including non-intesinal ones. So
it was concluded that the allanotic endoderm has
an intrinsic tendency to differentiate into intestinal
epithelium [70].

In France, Haffen and coworkers used wide
range of brush border enzymes as markers (suc-
rase-isomaltase, maltase-glucoamylase, lactase,
aminopeptidase, alkaline phosphatase and so on)
that can be detected by monoclonal antibodies
raised against each enzyme. Also they measured
enzymatic activity of the recombinants biochemi-
cally. These experiments confirmed the results
obtained in morphological studies. For example,
when chicken embryonic gizzard epithelium was
associated with duodenal fibroblasts derived from
newborn rat, the epithelium differentiated into
typical intestinal epithelium with brush border
detected by electron microscopy, and the bioche-
mical analysis showed the sucrase, maltase and
alkaline phosphatase activity comparable to those
of intact chicken intestine [65].

Pepsinogens, zymogens of pepsins, are proteins
specifically synthesized in stomach gland cells, and
have been most extensively studied in relation to
epithelial-mesenchymnal interactions. Yasugi and
Mizuno [15-18] have investigated first the normal
development of pepsinogen in chicken and quail
proventriculus and demonstrated that embryonic
proventriculus synthesizes embryo-specific pepsin-
ogen (EPg) which is no longer expressed at the end
of the embryonic life. Later the study was ex-
tended to ontogeny and phylogeny of pepsinogens
among vertebrates and even among protochor-
dates [71-74].

The use of antibody to ECPg [75] stimulated
much the studies of tissue interaction in the avian
DT (Table 2), and revealed some unexpected re-
sults: In some instances where the mesenchyme
seemed to exert instructive induction on the associ-
ated epithelium, the cytodifferentiation was found
not to be accompanied. For example, intestinal

164 S. YASUGI AND T. Mizuno

and allantoic epithelia formed very complex glands
that were never observed in the normal develop-
ment of these epithelia when they were associated
with proventricular mesenchyme and cultivated on
the CAM. Nevertheless the application of mono-
clonal and polyclonal antibodies to EPg cleary
demonstrated that those epithelia never express
EPg. Biochemical analysis with specific inhibitor
of pepsin also supported the results [76]. The
allantoic endoderm implanted into the presump-
tive proventricular area of young chicken embryo
formed complex glands, as stated above, but it
never expressed EPg and made a quite clear
contrast to the host proventricular glands which
actively synthesized EPg, although the grafted
allantoic endoderm and host glnad cells were sepa-
rated by only 50 ~m of mesenchymal cell layers
[47].

These facts are very important in considering the
ralationship between morphogenesis and cytodif-
ferentiation. Many worker in this field generally
believed that the morphological differentiation is
always accompanied by the biochemical dif-
ferentiation. The results obtained from the asso-
ciation of intestinal and allantoic epithelia and
proventricular mesenchyme clearly showed that
morphogenesis and cytodifferentiation are separ-
able events and achievement of morphological
differentiation is not necessarily accompanied by
cytodifferentiation. So far a few examples of
separation of morphological and cytochemical dif-
ferentiation have been reported: In recombinants
of salivary gland mesenchyme and mammary gland
epithelium [77], of adult rat urinary bladder
epithelium and fetal urogenital sinus mesenchyme
[78], of fetal urogenital sinus epithelium and
glandular stomach mesenchyme (Mizuno et al.,
Roux’s Arch. Develop. Biol., in press), and
others. The significance of these phenomena will
be discussed in detail in another review paper
(Mizuno and Yasugi, Cell Differ. Develop., in
preparation).

Then a question arose whether the inability of
intestinal and allantoic epithelia to synthesize EPg
protein under the influence of proventricular
mesenchyme is due to the lack of EPg gene
transcription or due to the defects in posttrascrip-
tional or translational processing. To answer this

question, Hayashi et al. [79-81] obtained cDNA of
EPg and examined the expression of EPg gene
transcripts in recombinants consisted of
esophageal, proventricular, gizzard, intestinal or
allantoic epithelia with proventricular mesen-
chyme, with cDNA of EPg as a probe. The
Northern blot analysis after gel electrophresis of
mRNA extracted from recombinants cleary de-
monstrated that intestinal and allantoic epithelia
could not express EPg gene while other epithelia
synthesized mRNA of EPg the length of which was
exactly the same as in normal embryonic proven-
triculus [81]. So we can now affirm that the
inductive influence generated by proventricular
mesenchyme could not evoke EPg gene transcrip-
tion in intestinal and allantoic epithelia. This may
be an importnat clue in anaylzing the molecular
basis of induction.

The appearance of EPg in various recombinants
of chicken DT epithelia and mesenchymes derived
from 6-day embryos is summarized in Table 2.
From this it is obvious that proventricular and
gizzard epithelia respond in just the same manner
against the mesenchymal influences. Thus they
synthesize small amont of EPg on the esophageal
and intestinal mesenchymes, synthesize EPg very
actively on the proventricular mesenchyme, and
EPg synthesis was completely inhibited on the
gizzard mesenchyme. This fact led Takiguchi er al.
to analyze precisely the chronological changes of
reactivity of gizzard epithelium to proventricular

TaBLE2. The expression of embryonic chick-
en pepsinogen (EPg) in various associa-
tions culitvated on the chorio-allantoic
membrane (% of explants possessing EPg-
positive gland cells)

Mesenchyme associated

Epithelium
ES PV GZ SI
ES 0 100 0 63
PV 75 100 0 77
GZ 22 100 0 71
SI 0 0 0 71
i AL 0 0 0 0

ES: Esophagus, PV: Proventriculus, GZ: Gizzard,
SI: Small intestine AL: Allantois.

Tissue Interactions in Gut Development 165

mesenchymal influence [82, 83]. The reactivity
decreased rapidly after day 7 of incubation onward
but still remained at very low level at day 12. The
result indicates that gizzard epithelium does not
express EPg in normal development because of
both a decrease in ability to express the enzyme in
the course of development and a suppressive in-
fluence of gizzard mesenchyme.

The induction of EPg in gizzard epithelium by
proventricular mesenchyme was confirmed also in
in vitro system, and the dissociated and reaggre-
gated mesenchymal cells are also effective [84].
This opens the way to analyze the nature of
induction by proventricular mesenchyme in in vitro
culture system.

THE MEDIATOR SYSTEM OF INDUCTIVE
INFORMATION IN THE DEVELOPMENT
OF DIGESTIVE TRACT

The results described so far strongly suggest the
transfer of some inductive information from
mesenchyme to epithelium. Generally speaking,
quite many factors may be involved in the transfer
of informations. Among them are: informative
substances emitted from mesenchymal cells, ex-
tracellular substances of mesenchyme, basement
membrane, cell adhesion molecules, receptor of
epithelial cells, intracellular information transfer
systems, and so on. In the epithelial-mesenchymal
inteactions of DT development, some progresses
have been made recently.

Structural organization of mesenchyme and the
inductive action

First, whether the structural organization of
mesenchyme is required for its inductive action
was tested with using the isolation and cultivation
in vitro of mesenchymal cells. As already men-
tioned, in the case of recombination of chicken
gizzard epithelium and rat intestinal fibroblasts,
fibrablastic cells cultivated in vitro still maintain
inducing activity [65]. Dissociated and instantly
reaggregated proventricular mesenchymal cells
could elicit proventricular epithelial differentiation
in associated proventricular and gizzard epithelia
[84]. These studies showed that the structural
integrity in situ is not necessary for the inductive

ability although there is a possibility that these
mesenchymal cells reconstituted the normal
structural integrity soon after the onset of cultiva-
tion with epithelium.

Quite unexpected results were obtained when
proventricular epithelium of 6-day embryos was
associated with cultured rat intestinal mesenchym-
al cells or cultured human or rat skin fibroblasts
and recombinants were implanted into coelomic
cavity of chicken embryo [85] (Table 3). These
cells supported full differentiation of epithelium
toward proventricular epithelium. The epithelium
formed complex glands the cells of which synthe-
sized EPg very actively. Under the same condi-
tions, the intact chicken and rat intestinal mesen-
chymes suppressed the proventricular differentia-
tion both morphologically and functionally. One
can explain the results as these mesenchymal cells
cultivated in vitro acted as dedifferentiated, ’neut-
ral’ cells lacking organ-specific inducing ability,
and permitted the development of intrinsic tenden-
cy toward the proventricular epithelium.

The molecluar differences of mesenchymes

The mesenchymes of proventriculus and gizzard
exert contradictory effect on the proventricular
and gizzard epithelia (see above). The elucidation
of molecular difference between the two mesen-
chymes will be an important step for the under-
standing of inductive influence of mesenchyme.
To detect the molecular difference, Takiguchi and
Yasugi (unpublished result) used monoclonal anti-
body technique. The crude extract of embryonic
chicken gizzard was injected into mouse and anti-
bodies were screened with attention to the tissue
distribution of antigen. Among many clones, we
picked up an antibody designated as GM-1. The
localization of GM-1 antigen varies between
proventriculus and gizzard at day 6 of incubation
when these mesenchymes show differential effect
on the epithelium. GM-1 antigen was scarcely
detected in 6-day proventriculus except the very
weak signal in basement membrane, whereas in
the gizzard GM-1 antigen localized in the mesen-
chymal layer just underneath the epithelium. The
molucular weight of GM-1 antigen calculated from
immunoblotting after SDS-PAGE was about 42
kDa and immunolocalization was quite different

166 S. YASuGI AND T. MizuNo
TABLE 3. Differentiation of proventricular epithelium associated with mesenchyme or

cultivated mesenchymal cells, derived from digestive organs, or with skin fibroblasts.
From Yasugi et al. [85]

Type of associated No. of No. of grafts showing No. of grafts expressing

mesenchyae bene Complex Simple Pepsinogen Sucrase

peers Ba glands glands

cPV mesenchyme 8 8 0 8 0

cPV mes cells 5 0 11 11 0

cSI mesenchyme 11 0 11 11 0

rSI_ mesenchyme 6 0 0 0 0

rSI mes cells 4 4 0 4 0

r skin fibrobl 5 5 0 5 1

h skin fibrobl 4 4 0 4 1

c: Chicken, r: Rat, h: Human, PV: Proventriculus, SI: Small intestine, mes: Mesenchymal,

fibrobl: Fibroblasts.

from those of well-known ECM, such as larger than 0.6 um, and even in case of 0.2 um

fibronectin, laminin, and tenascin. By the way, the
distribution of these common ECM was almost the
same in proventriculus and gizzard.

GM-1 antigen represents one of organ-specific
substances and may be thought to be involved in
the establishment of regionality of mesenchyme,
although we do not know the exact biological
significance of it which is to be elucidated in the
near future.

In mouse small intetine, tenascin appears in
mesenchyme when epithelial cell differentiation
begins and its expression is dependent on the
epithelial-mesenchymal interactions [86].

The diffusibility of inductive action of mesenchyme

It is now well established that in many systems of
tissue interactions the direct contact of inducer
cells and responding cells is necessary [87, 88].
Takiguchi and Yasugi (Roux’s Arch. Devlop.
Biol., in press) examined the diffusibility of induc-
ing effect of proventricular mesenchyme with
Nuclepore filter interposed between mesenchyme
and epithelium. With Nuclepore filter of pore size
less than 0.4 um, the induction never occurred,
while with filter of pore size larger than 0.6 «m the
epithelium expressed EPg and the amount of EPg
expressed increased as the pore size became lar-
ger.
pass through the Nuclepore filter of pore size

It must be noted that mesenchymal cells can

filter the cell processes of mesenchyme reached the
epithelial surface of filter. So we suppose that the
direct contact of eithelial cells with mesenchymal
cells or with mesenchymal extracellular substances
is necessary also in the induction in these cases,
although it must be confirmed by electron micros-
copic study.

Basement membrane

The important role of basement membrane (bas-
al lamina) can be easily understood since basement
membrane not only mechanically separates epithe-
lium and mesenchyme but also regulates the flow
of substances between these two tissues. In de-
veloping DT, the thinning of basement membrane
was reported. In the proventriculus and intestine,
cytoplasmic processes of epithelial cells penetrate
the gaps made in basal lamina and make direct
contact with mesenchymal cells [12, 89, 90]. Also
basement membrane components (laminin,
nidogen, type IV-collagen) are present at the
epithelial-mesenchymal interface early in develop-
ment and the spatial distribution of some ex-
tracellular matrix proteins is closely associated
with morphogenesis [91]. Moreover, the produc-
tion of basement membrane is thought to require
epithelial-mesenchymal interaction, at least in ex-
perimental in vitro system [92].

Tissue Interactions in Gut Development 167

Cytoskeletal system of epithelial cells and the ex-
pression of organ-spcific proteins

Cytoskeletal systems are involved in the mor-
phogenesis of cells, in intracellular transmission of
information received at cell surface, and in phy-
siological functions such as secretion. However,
the studies on the relations of cytoskeletal systems
and organ-specific gene expression in the organo-
genesis are very limited [93].

Epithelial cells contain cytokeratins as in-
termediate filaments, and the cytokeratins of
chicken DT epithelial cells are partially characte-
rized with antibodies [94].

Takiguchi et al. (unpublished results) studied the
expression of cytokeratin immunoreactive to an
antibody PKK-1 which is reactive to human
cytokeratins 8, 18, and 19 and recognizes 52kDa
protein in chicken DT. Interestingly enough, there
is an inverse correlation between expression of
PKK-1 antigen and of EPg in chicken proventricu-
lus: In EPg-producing gland cells PKK-1 antigen
disappears, while EPg-non producing luminal
epithelial cells and duct cells are always PKK-1-
positive. Epithelial cells of DT except the proven-
tricular gland cells are also positive to PKK-1, and
in the stomach of all vertebrates so far examined
except mammals, the pepsinogen-producing gland
cells are negaive to PKK-1 [95]. Although the
precise relation between pepsinogen expression
and disappearance of PKK-1 antigen is not clear at
present, disappearance of PKK-1 antigen is a very
specific marker of appearance of pepsinogen-
producing cells in the developing stomach.

Various epithelia were associated with proven-
tricular mesenchyme and cultivated on the CAM
to see whether the disappearacne of PKK-1 anti-
gen is also regulated by the mesenchymal in-
fluences. The results cleary indicated that
esophageal, proventricular and gizzard epithelia,
which expressed EPg, lost PKK-1 antigen without
exception, while intestinal epithelium, which could
not synthesize EPg, continuously expressed PKK-1
antigen. So the expression of PKK-1 antigen is
also under the control of mesenchyme and the
disappearance of PKK-1 antigen is well linked to
EPg expression.

This raises very interesting questions: does the

proventricular mesenchyme emit two independent
factors, one for the suppression of PKK-1 antigen
and another for the induction of EPg? or does the
mesenchyme release only one sort of signal that
evokes at the same time the two events indepen-
dently? or does the mesenchyme release a signal
that induces EPg expression via the suppression of
PKK-1 antigen expression? The fact that the
intestinal epithelium did not lose the PKK-1 anti-
gen in the presence of proventricular mesenchyme
seems to exclude the first hypothesis.

PERSPECTIVES

We reviewed here various aspects of epithelial-
mesenchymal interactions working in the organo-
genesis of DT in avian and mammalian embryos
and fetuses. Although we are still far from the
complete understanding of the nature of the pro-
cesses, we have now ample clue to pursue the
molecular basis of tissue interactions.

First and most important subject is to identify
directly the substances that are responsible for the
organ-specific inducing influence of mesenchyme.
Evidence suggests that the inducing substances are
not freely diffusible ones, and may be sought
among extracellular or basement membrane com-
ponents. There is a possibility, however, that
some kinds of growth factors or their derivatives
are involved in the process, since in some instances
they act as potent inducing factors [96-98].

Next problem is to find out the nature of reactiv-
ity of epithelial cells in relation to the activation of
specific gene expression. To approach the prob-
lem, it is necessary to elucidate the control
mechanisms of expression of relevant genes such
as EPg, sucrase or cytokeratin genes. The finding
of trans-acting factors for these genes will provide
much information of signal transfer routes in tissue
interactions. As for the reactivity of various
epithelia against the influence of the same mesen-
chyme, i. e., the reactivity of gizzard and intestinal
epithelia against proventricular mesenchyme, it is
prerequisite to know the structural differences of
genes that are induced or not induced to be
transcribed by the mesenchymal influence. One
possibility is a methylated state of the genes, since
the methylation of the genes often brought about
the inactivation of them. In fact, in mammais,

168

pepsinogen genes are hypermethylated when they
are out of work and become hypomethylated when
they are transcribed [99]. This problem is current-
ly under investigation with EPg gene in the labora-
tory of one of the present authors (S.Y.).

ACKNOWLEDGMENTS

We wish to express our deepest gratitude to Emeritus
Professor Takashi Fujii, University of Tokyo, and Emer-
itus Professor Etienne Wolff, College de France, for
introducing us to this problem and for their encourage-
ment during the course of the studies.

REFERENCES

Wessells, N. K. (1977) Tissue Interactions and
Development. W. A. Benjamin, Inc., California/

Massachusetts/London/Amsterdam/Ontario/Sydney.

Nieuwkoop, P. D., Johnen, A. G. and Albers, B.
(1985) The Epigenetic Nature of Early Chordate
Development. Inductive Interaction and Compe-
tence. Cambridge University Press, Cambridge/
London/New York/Rochelle/Mellbourne/Sydney.
Raff, R. A. and Kaufman, T. C. (1983) Embryos,
Genes, and Evolution. Macmillan Company, New
York.

Saxén, L., Karikinen-Jaaskelainen, M., Lehtonen,
E. Nordling, S. and Wartiovaara, J. (1976) Induc-
tive tissue interactions. In “The Cell Surface in
Animal Embryogenesis and Development”. Ed. by
G. Poste and G. L. Nicolson, North-Holland Publ.
Comp., Amsterdam/New York/Oxford, pp. 331-
407.

Gurdon, J. B., Fairman, S., Mohun, T. J. and
Brennan, S. (1985) Cell, 41: 913-922.

Gurdon, J. B. (1987) Development, 99: 285-306.
Nogawa, H. (1981) J. Embryol. Exp. Morphol., 62:
229-239.

Fukuda-Taira, S. (1981)
phol., 63: 111-125.
Shiojiri, N. (1984) J. Embryol. Exp. Morphol., 79:
24-39.

Yasugi, S. (1976) C.R.Acad.Sci., Paris, 283: 383-
385.

Romanoff, A. L. (1960) The Avian Embryo, The
Macmillan Company, New York.

Hayashi, K. (1987) Develop. Growth Differ., 29:
285-295.

Calhoun, M. L. (1954) Microscopic Anatomy of the
Digestive System of the Chicken. Iowa State Uni-
versity Press, Ames.

Yasugi, S., Mizuno, T. and Esumi, H. (1979) Ex-
perientia, 35: 814-815.

J. Embryol. Exp. Mor-

S. YASUGI AND T. Mizuno

Yasugi, S. and Mizuno, T. (1979) C.R.Soc. Biol.,
173: 973-978.

Yasugi, S. and Mizuno, T. (1981) J. Exp. Zool.,
216: 331-335.

Yasugi, S. and Mizuno, T. (1981) J. Biochem., 89:
311-315.

Yasugi, S. and Mizuno, T. (1982) C.R.Soc. Biol.,
176: 575-579.

Moog, F. (1950) J. Exp. Zool., 115: 109-129.
Matsushita, S. (1981) J. Fac. Sci., Univ. Tokyo,
Sect. IV, 15: 81-92.

Matsushita, S. (1983) Comp. Biochem. Physiol.,
76B: 365-470.

Matsushita, S. (1985) J. Exp. Zool., 233: 377-383.
Hellander, H. F. (1969) Gastroenterology, 56: 53-
70.

Fukamachi, H. (1978) J. Fac. Sci., Univ. Tokyo,
Sect. IV, 14: 85-94.

Yasugi, S., Matsumoto, F. and Mizuno, T. (1987)
C. R. Soc. Biol., 181: 450-453.

Moog, F. Denes, A. E. and Powell, P. M. (1973)
Develop. Biol., 35: 143-159.

Balinsky, B. I., (1948) Roux’ Arch. Entwicklungs-
mech. Organ., 143: 365-395.

Okada, T. S. (1960) Roux’ Arch. Entwichlungs-
mech. Organ., 152: 1-21.

Le Douarin, N. (1964) Bull. Biol. France Belg., 98:
543-676.

Rosenquist, G. C. (1971) Develop. Biol., 26: 323-
335.

Sengel, P. (1971) Adv. Morphol., 9: 181-230.
Ekblom. P. (1981) Medical Biol., 59: 139-160.
McLoughlin, C. B. (1960) Xe Congrés Internation.
Biol. Cell., Paris, pp. 38-39.

Sigot, M. and Marin, L. (1970) J. Embryol. Exp.
Morphol., 24, 43-63.

Sigot, M. (1971) Arch. Anat. Microsc. Morphol
Exp., 60: 169-204.

David, D. (1971) Roux’ Arch. Entwicklungsmech.
Organ., 168: 304-319.
David D. (1972) J. Embryol. Exp. Morphol., 27:
177-197.

David, D. (1972) Roux’ Arch. Entwicklungsmech.
Organ., 170: 1-12.

Soriano, L. (1965) J. Embryol. Exp. Morphol., 14:
119-128.

Fukamachi, H., Mizuno, T. and Takayama, S.
(1979) Anat. Embryol., 157: 151-160.

Fukamachi, H. and Takayama, S. (1980) Experien-
tia, 36: 335-336.

Yasugi, S. and Mizuno, T. (1974) Roux’ Arch.
Entwicklungsmech. Organ., 174: 107-116.

Yasugi, S. (1979) Develop. Growth Differ., 21:
343-348.

Wolff, Et. and Haffen, K. (1952) Texas Rep. Biol.

45

46

47

48

49

50

51

oy

53

54

55

56

57

58
5)

60

61

62

63

64

65

66

67

68

69

70

Tissue Interactions in Gut Development

Med., 10: 463-472.

Yasugi, S. (1976) C.R.Acad. Sci., Paris, 283: 179-
183.

Gumpel-Piot, M., Yasugi, S. and Mizuno, T. (1978)
C.R.Acd.Sci., Paris, 286: 117-120.

Yasugi, S. (1984) J. Embryol. Exp. Morphol., 80:
137-153.

Yasugi, S. and Mizuno, T. (1978) Develop. Growth
Differ., 20: 261-267.

Ishizuya-Oka, A. and Mizuno, T. (1984) J.
Embryol. Exp. Morphol., 82: 163-176.

Masui, T. (1981) J. Embryol. Exp. Morphol., 62:
277-289.

Masui, T. (1982) J. Embryol. Exp. Morphol., 72:
117-124.

Yasugi, S. and Mizuno, T. (1984) C. R. Soc. Biol.,
178: 580-583.

Sumiya, M. and Mizuno, T. (1974) C.R. Acad.Sci.,
Paris, 278: 1529-1532.

Sumiya, M. (1976) J. Fac. Sci., Univ. Tokyo, Sect.
IV, 13: 363-381.

Sumiya, M. (1976) Roux’s Arch. develop. Biol.,
179: 1-18.

Sumiya, M. and Mizuno, T. (1976) Proc. Japan
Acad., 52: 587-590.

Ishizuya-Oka, A. (1983) Roux’s Arch. Develop.
Biol., 192: 171-178.

Mizuno, T. (1975) C. R. Soc. Biol., 169: 1096-1098.
Fukamachi, H., Mizuno, T. and Kim, Y. S. (1986)
Experientia, 42: 312-315.

Fukamachi, H., Mizuno, T. and Kim, Y. S. (1987)
J. Cell Sci., 87: 615-621.

Fukamachi, H. and Kim. Y.S. (1989) Develop.
Growth Differ., 31: 307-313.

Kedinger, M., Haffen, K. and Simon-Assmann, P.
(1987) Differentiation, 36: 71-85.

Kedinger, M., Simon-Assmann, P., Bouziges, F.
and Haffen, K. (1988) Scand. J. Gastroenterol., 23,
Suppl. 151: 62-69.

Kedinger, M., Simon-Assmann, P.M., Lacroix, B.,
Marxer, A., Hauri, H.P. and Haffen, K. (1986)
Develop. Biol., 113: 474-483.

Haffen, K., Lacroix, B., Kedinger, M. and Simon-
Assmann, P. M. (1983) Differentiation, 23: 226-
233.

Haffen, K., Kedinger, M., Simon, P. M. and Raul,
F. (1981) Differentiation, 18: 97-103.

Kramer, B., Andrew, A. Rawdon, B. B. and Beck-
er, P. (1987) Development, 100: 661-671.
Rawdon, B. B. (1988) Inaugural Lecture of Uni-
versity of Cape Town, 138: 1-16.

Rawdon, B. B. and Andrew, A. (1988) Cell Differ.
Develop., 25: 135-144.

Matsushita, S. (1984) Roux’s Arch. Develop. Biol.,
193: 211-218.

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

169

Yasugi, S. and Mizuno, T. (1982) C. R. Soc. Biol.,
176: 880-884.

Yasugi, S. (1987) Comp. Biochem. Physiol., 86B:
675-680.

Yasugi, S., Matsunaga, T. and Mizuno, T. (1988)
Comp. Biochem. Physiol., 91A: 564-569.

Yasugi, S., Matsunaga, T. and Mizuno, T. (1989)
Zool. Sci., 6: 283-288.

Yasugi, S., Hayashi, K., Takiguchi, K., Mizuno, T.,
Mochii, M., Kodama, R., Agata, K. and Eguchi, G.
(1987) Develop. Growth Differ., 29: 85-91.
Yasugi, S., Matsushita, S. and Mizuno, T. (1985)
Differentiation, 30: 47-52.

Sakakura, T., Nishizuka, Y. and Dawe, C. J. (1976)
Science, 194: 1439-1441.

Suematsu, N., Takeda, H. and Mizuno, T. (1988)
Zool. Sci., 5: 385-395.

Hayashi, K., Agata, K., Mochii, M., Yasugi, S.,
Eguchi, G. and Mizuno, T. (1988) J. Biochem., 103:
290-296.

Hayashi, K., Yasugi, S. and Mizuno, T. (1988)
Biochem. Biophys. Res. Commun., 152: 776-782.
Hayashi, K., Yasugi, S. and Mizuno, T. (1988)
Development, 103: 725-731.

Takiguchi, K., Yasugi, S. and Mizuno, T. (1986)
Roux’s Arch. Develop. Biol., 195: 475-483.
Takiguchi, K., Yasugi, S. and Mizuno, T. (1988)
Roux’s Arch. Develop. Biol., 197: 56-62.
Takiguchi, K., Yasugi, S. and Mizuno, T. (1988)
Develop. Growth Differ., 30, 241-250.

Yasugi, S., Kedinger, M., Simon-Assmann, P.,
Bouziges, F. and Haffen, K. (1989) Roux’s Arch.
Develop. Biol., 198: 114-117.

Aufderheide, E. and Ekblom, P. (1988) J. Cell
Biol., 107: 2341-2349.

Slavkin, H. C. and Bringas, P., Jr. (1976) Develop.
Biol., 50: 428-442.

Lehtonen, E., Wartiovaara, J., Nordling, S. and
Saxén, L. (1975) J. Embryol. Exp. Morphol., 33:
187-203.

Mathan, M., Hermos, J. A. and Trier, J. S. (1972)
J. Cell Biol., 52: 577-588.

Ishizuya-Oka, A. and Mizuno, T. (1985) Roux’s
Arch. Develop. Biol., 194: 301-305.
Simon-Assmann, P., Kedinger, M. and Haffen, K.
(1986) Differentiation, 32: 59-66.
Simon-Assmann, P., Bouziges, F., Arnold, C., Haf-
fen, K. and Kedinger, M. (1988) Development, 102:
339-347.

Lesot, H., Karcher-Djuricic, V., Kubler, M. D. and
Ruch, J. V. (1988) Differentiation, 37: 62-72.
O’Guin, W. M., Galvin, S., Schermer, A. and Sun,
T.-T. (1987) Curr. Topics Develop. Biol., 22: 97-
125.

Takiguchi-Hayashi, K. and Yasugi, S. (1988) C.R.

170 S. YASUGI AND T. MizuUNo
Soc. Biol., 182: 445-448. tion, 34: 18-24.
96 Slack, J. M. W., Isaacs, H. V. and Darlington, B. 99

Ichinose, M., Miki, K., Furihata, C., Tatematsu,
G. (1988) Development, 103: 581-590.

M., Ichihara, Y., Ishihara, T., Katsura,I., Soga-

97 Ekblom, P., Thesleff, I., Saxén, L., Miettinen, A. wa,,K., Fujii-Kuriyama, Y., Tanji, M., Oka, H.,
and Timpl, R. (1983) Proc. Natl. Acad. Sci., USA, Matsushima, T. and Takahashi, K. (1988) Cancer
80: 2651-2655. Res., 48: 1603-1609.

98 Partanen, A.-M. and Thesleff, I. (1987) Differentia-

ZOOLOGICAL SCIENCE 7: 171-186 (1990)

REVIEW

Cellular and Molecular Aspects of
Embryonic Induction

HEINZ TIEDEMANN

Institut fiir Molekularbiologie und Biochemie, Freie Universitat Berlin,
Arnimallee 22, D-1000 Berlin (West)33,
Federal Republic of Germany

ABSTRACT— A homogeneous protein, which induces mesodermal and endodermal tissues in amphi-
bian gastrula ectoderm (“vegetalizing factor”) has been isolated from chicken embryos. Inducing
factors with similar chemical properties have been found in the Xenopus XTC-cell line and in calf
kidney. The factors belong to evolutionary conserved proteins, which may also have regulatory
functions in later differentiation processes or the maintenance of differentiation. They are related to the
TGE- protein superfamily. Members of this and of the FGF protein family induce mesodermal tissues.
In early embryos mesoderm inducing factors show a graded distribution. Masked maternal neural
inducing factors are in part activated after the cleavage stages. They have been partially purified from
Xenopus embryos. We could show that phorbolester evokes neural differentiation, suggesting a signal

© 1990 Zoological Society of Japan

transduction mechanism which may include phospholipases and protein kinases.

INTRODUCTION

Amphibian eggs and embryos have been widely
used to study the development of vertebrates.
Early stages which can be handled with relative
ease, are endowed with a high regulatory capacity.
They are favorite objects for the study of tissue
determination and induction. Determination has
been defined as a process which initiates a specific
pathway of development by singling it out from
various possibilities for which the system is compe-
tent [1]. In very early stages determination is not
yet definitive (except for endoderm) and can be
changed by inductive tissue interactions. Dorsal
ectoderm isolated before gastrulation forms
epidermis like cells as does ventral ectoderm [2].
In 1924 Spemann and Hilde Mangold discovered
that the ectoderm can be induced by the presump-
tive mesoderm [3]. When the induced dorsal
ectoderm is isolated after gastrulation it forms the
different parts of the nervous system. The induc-

Received February 2, 1990

tion of the nervous system is not an all or none, but
a progressive process. Isolation experiments [4]
and grafting experiments [5] have shown that at
successive stages of gastrulation neural differentia-
tions are obtained in the following order: Neural
crest derivatives, archencephalic (forehead) and
deuterencephalic (hindhead) structures.

Test of inducing activity on omnipotent
gastrula ectoderm

The inducing activity of tissues or isolated fac-
tors can be tested on omnipotent gastrula
ectoderm. The differentiation of ectoderm can up
to the early gastrula stage still be channelled into
other pathways by the addition of inducers. In the
implantation method devised by Mangold [6, 7] a
piece of tissue or a pellet which contains inducing
factors is implanted through a slit in the ectoderm
into the blastocoelic cavity of an early gastrula of
Triturus or Ambystoma. By the gastrulation move-
ments the implanted tissue is brought into contact
with the ventral ectoderm (Fig. 1). Purified factors

172 H. TIEDEMANN

Fic. 1. Test of inducing factors. A. Implantation
method a) Insertion of the implant into the blasto-
coelic cavity of an early gastrula, b) Position of the
graft after gastrulation, facing the ventral ectoderm.
B. Test on isolated ectoderm in solution. Ec=
ectoderm; B=blastocoelic cavity; Bl=blastopore.
From Knéchel et al., Blut, 59: 208 (1989), Springer-
Verlag, Berlin.

can be tested by the implantation method in differ-
ent dilutions, if the inducing substance is mixed
with a non-inducing protein (i.e. y-globulin) be-
fore the pellet for implantation is prepared. Xeno-
pus gastrulae cannot be used for the implantation
method [8] due to the different architecture of
Xenopus embryos. Inducing factors can be tested
in solution on isolated ectoderm of Xenopus, Tri-
turus or Ambystoma embryos [7, 9, 10]. Bovine
serum albumin is added to the solution to prevent
the adsorption of the small amount of highly
purified inducing factors to glass or plastic surfaces
[11]. The test on isolated ectoderm has the advan-
tage that Triturus and Ambystoma as well as
Xenopus embryos which are available all the year
round can be used, but the disadvantage that all
induced explants must be examined histologically.
Biochemical markers are at present only available
for a limited number of tissues.

The response of ectoderm from different species
especially to neural inducers is different. Amby-
stoma ectoderm is most susceptible to neural in-
Isolated ectoderm cultivated in a phy-
siological salt solution forms neural tissue without

duction.

the addition of an inducing factor (so called auto-
neuralization). Triturus alpestris ectoderm does on

the contrary not form any neural tissue at physiolo-
gical salt concentration, but forms neural tissues
when the Na*~ ion concentration is increased 1.5
fold [7]. Xenopus ectoderm is the least susceptible
and does not form neural tissues at an 1.5 fold
increased salt concentration. A strong autoneura-
lization, however, has been observed if the Xeno-
pus ectoderm is dissociated into single cells and
reassociated after a lag period [12].

Neural induction and neural inducing factors

The temporal changes of the neural inducing
capacity of the presumptive dorsal mesoderm as
well differences in the regional inductive capacity
of the invaginated mesoderm (the archenteron
roof) have been extensively investigated. The
experiments have been carried out on Triturus and
Ambystoma embryos which are more suitable for
this type of investigation than the smaller and
more rapidly developing Xenopus embryos [review
14]. The presumptive dorsal mesoderm of early
cleavage stages induces neural tissues at a very low
rate. The inducing activity starts at the morula
stage [13]. Saxén and Toivonen [15] discovered
that the simultaneous implantation of a mostly
forehead inducing heterologous tissue (guinea pig
liver) and a mostly mesoderm inducing tissue
(guinea pig bone marrow) induced preferentially
forehead, hindhead or neural tube depending on
the proportions of neural and mesodermal induc-
ers. This did lead to the theory that a double
gradient of a neuralizing and a mesodermalizing
agent are in the embryo responsible for the induc-
tion of the different regions of the nervous system.
By other experiments (“the fold method”) Nieuw-
koop et ai. [16] came to the conclusion that in
neural induction a wave of activation (resulting in
archencephalic structures) is followed by a wave of
transformation (resulting in deuterencephalic
structures). Attempts by Tiedemann ef al. [17] to
purify a deuterencephalic (hindhead) inducer led
to the separation of a neural-archencephalic (fore-
head) and a mesoderm inducing fraction. Com-
bination of the two fractions in different propor-
tions did lead to hindhead and neural tube induc-
tion [18]. Toivonen and Saxén combined dispersed
cells of the mesoderm and the forebrain anlage of

Embryonic Induction 173

gastrulae in different ratios [19, 20]. They
observed that a larger proportion of mesodermal
cells causes formation of more caudal parts of the
central nervous system. The ability of cells of the
forebrain anlage to be transformed to more caudal
parts of the brain is lost after about 10-12 hr, the
transforming ability of the mesoderm lasts some-
what longer [21]. Isolated dorsal ectoderm of early
Triturus gastrulae (which in normal development is
induced to the nervous system) as well isolated
ventral ectoderm can be induced to neural struc-
tures [22, 23]. In Xenopus the ventral ectoderm is,
however, induced less frequently. This could be
due to an earlier loss of competence in the ventral
ectoderm but can not be considered as neural
“predetermination” of dorsal ectoderm, which as
ventral ectoderm forms neural tissues only after a
neuralizing stimulus. The homeotic gene XHox 6
which is expressed early in the development of the
neural plate and restricted to the middle and
posterior part of the neural axis was used as a
marker for neural induction in these experiments
[24]. An antigen related to the Drosophila homeo-
tic gene invected (which is involved in segmenta-
tion) is expressed in the anterior neural plate [25].

The experiments on the transmission of the
neural stimulus and the chemical nature of neural
inducing agents are described in an excellent re-
view by Saxén [26] and will only briefly be men-
tioned. Saxén and Toivonen have shown that the
separation of the inducing mesoderm and the
reacting ectoderm by nucleopore filters with pore
diameters down to 0.05 » did not prevent neural
induction. Cytoplasmic bridges were not observed
by electronmicroscopy, so that direct cell contacts
in the transfilter experiments can be ruled out [27,
28]. The induction depends on short range interac-
tions. Tacke and Grunz [29] observed that a close
juxtaposition of the ectoderm and the inducing
chorda mesoderm (distance of plasma membranes
of opposite cells less than 50 nm) is needed for
neural induction. This is correlated with an in-
crease of the number of coated pits, a feature of
receptor-mediated endocytosis, in the ectoderm.
Connecting cell projections between ectoderm and
mesoderm but no cytoplasmic bridges, which could
allow a free transfer of inducing factors through
intercellular channels, are formed during the mid-

gastrula stage [30].

The neural inducing activity of the blastoporal
lip is diminished when the secretion of the factor
from the blastoporal lip is impaired by treatment
with actinomycin D or cycloheximide, substances
which could inhibit the synthesis of components of
the export system [31]. The factor may be secreted
by a mechanism similar to that found in oocytes.
Oocytes can secrete proteins which are injected or
synthesized on foreign m-RNA. The proteins are
probably sequestered into vesicles and exported by
exocytosis [review: 32].

A small amount of neural inducing protein was
isolated from the extracellular space between
mesoderm and the neural plate, whereas proteo-
glycans from the extracellular space, isolated from
the aqueous phase after phenol extraction, did not
induce [33]. The proteoglycans did on the contrary
prevent the autoneuralization of Triturus alpestris
ectoderm cultured in Flickinger salt solution with
1.5 fold Na* ion concentration (Hildegard Tiede-
mann, unpublished experiments). The extracellu-
lar material did not contain RNA indicating little
contamination from damaged cells. Duprat and
Gualandris have shown that the extracellular
material on the inner surface of the ectoderm is not
implicated in neural determination [34]. The elec-
tric coupling of ectoderm and mesoderm which
occurs 3-6 hr after combination of the two tissues
[35] as well as the change of Na* and K~ ion
concentrations in induced ectoderm [36] may not
be a prerequisite but a consequence of induction.

The induction of kidney tubules in meta-nephric
mesenchyme in transfilter experiments depends on
the other hand on direct cell to cell contact by cell
processes penetrating the filter [37]. The cell to
cell interaction leading to induction is dependent
on protein glycosylation [38] and may be due to
factors which are integrated into the plasma mem-
brane or extracellular matrix proteins [for an ex-
ample, 39]. Neural inducing protein factors have
been separated from the soluble fraction of chick-
en embryo brain and retina by electrophoresis and
isoelectric focusing [40] or by chromatography on
DEAE-cellulose [15].

From Xenopus embryos neural inducing factors
have been partially purified by Janeczek et al. [31,
41, 45 and unpublished experiments]. The factors

174 H. TIEDEMANN

are found in oocytes and in gastrula stages in
ribonucleoprotein particles of about 110A dia-
meter, which are different from ribosomes or their
subunits, in the high speed supernatant and in
small vesicles. In addition a very small mesodem
inducing activity (as shown by the induction of
hindheads; 15, 18) was found in these fractions. A
forehead (archencephalic) induction is shown in
Fig. 2. The factors are inactivated by proteolytic
enzymes. The neural inducing factor in the RNP-
particles is a basic protein with an apparent
molecular weight >70,000 for the undegraded
factor. Smaller proteins with neural inducing
activity arise probably by enzymatic cleavage of
the larger ones.

Fic. 2.

A (upper). Forehead with eye induced on the
ventral side of a Triturus alpestris larva by the
implantation method. B (lower). Section through
the forehead induction. L=lens; T=tapetum; B=

brain; N=nose. The lumen of the induced nose is
found in other serial sections of this induction.

The neural inducing factor in the supernatant
has been partially purified by DEAE-cellulose

chromatography [53] or by size exclusion HPLC
[41]. The factor from gastrulae elutes at several
size classes (Mr 16,000, Mr 35-50,000 and Mr
130,000—150,000) whereas the factor from oocytes
is preferentially found in the largest size class. The
factor is not inactivated after reduction with mer-
captoethanol. Its molecular weight is not changed
after reduction, but many contaminating proteins
are shifted to smaller size. To take advantage of
this fact the high speed supernatant from gastrulae
was prepared in the presence of the protease
inhibitors a -macroglobulin and leupeptin, re-
duced by mercaptoethanol and subjected to size
exclusion HPLC. About 80-90% of the factor is
then eluted at an apparent Mr of 100,000-150,000.
When this protein fraction was subjected to SDS-
polyacrylamide gel electrophoresis besides large
proteins with an apparent Mr up to 150,000 also
molecules of much smaller size were found. They
obviously constitute a complex which is stable in
50% formic acid. Proteins of different size were
then electroeluted from the gel and tested. About
20-30% of the neural inducing activity is found in
proteins of an apparent molecular weight of
100,000-90,000, 70-80% of the activity in smaller
proteins of Mr 15,000-25,000. Up to this step the
smaller factor is purified about 800-1,000 fold.
The experiments suggest that the larger factor
could be a precursor of the smaller ones. The
complex is not artificially formed in 50% formic
acid. When the high speed supernatant was centri-
fuged on a sucrose gradient, most of the neural
inducing activity was found in proteins larger than
100,000 Dalton.

It is possible, but has not been proven, that the
factor in the RNP-particles is related to the super-
natant factor. The factor in the ribonucleoprotein
particles is, as was already mentioned, a basic
protein, the factor complex in the supernatant, as
the factor extracted from a fraction of small vesi-
cles, an acidic protein (isoelectric point pH 5.5).
Incubation with neuraminidase, hydrolysis with
sulfuric acid under conditions where neuraminic
acid is completely split from glycoproteins [44] or
chemical deglycosylation with fluoromethansulfo-
nic acid [Hoppe er al. unpublished experiments]
did not change the isoelectric point. Treatment
with phenol at 60°C does, however, convert a part

Embryonic Induction 175

of the acidic neural inducing protein to a basic
neural inducing protein perhaps by partially dis-
sociating protein complexes.

Previous experiments have shown that the
neuralizing factor is a maternal protein which is
present in oocytes in ribonucleoprotein particles as
well as in the supernatant in a masked biologically
not active state [31, 44]. The neural inducing
activity of the presumptive dorsal mesoderm in-
creases from the morula stage onward [13, 46, 47].
This could depend on a partial activation of a
maternal factor. The dorsal cortex has no inducing
capacity [46]. The masked factors can artificially
be activated by precipitation with ethanol, which
denatures large protein complexes or by treatment
with dissociating agents as urea, SDS or formic
acid [31]. Whether the proteolytic cleavage of the
precursor is related to the activation of the neura-
lizing factor is, however, not known. The precur-
sor could be biologically inactive in the native state
and treatment with dissociating agents like formic
acid or SDS could lead to its activation. It is well
known that for instance in enzymes regulatory
domains can maintain a catalytic domain in an
inactive state within a single peptide chain. In such
molecules partial denaturation under dissociating
conditions can have a similar effect as limited
proteolysis (which eliminates the regulatory do-
main). Whether such an interaction between differ-
ent domains of a single peptide chain exists in the
large sized neural inducing factor is, however, not
known. It is on the other hand not excluded (and
may even be more likely) that other proteins which
are associated with the factor keep the factor in a
masked state and that the dissociation of the
complex leads to its activation. The physiological
process of demasking the factor(s) is unknown.

A small neural inducing activity has been found
in germinal vesicles and in nuclei from later stages

after activation with ethanol. Whether the factor
in the nuclei differs from the factor(s) in the

cytosolic fractions is not known [48]. Inducing
factors are not integral proteins of plasma mem-
branes [49].

Neural plates of Triturus alpestris induced by the
underlying mesoderm acquire in turn neural induc-
ing activity (homoiogenetic induction). This is
correlated with the activation (and release) of a

neuralizing factor from the neural plate and may
suggest an autocrine mechanism. It has, however,
to be proven whether the neuralizing factor from
mesoderm is identical with the releasing factor or
whether special releasing (and demasking) factors
exist [50].

Both the factors from RNP-particles and from
the cytosol remain fully active when they are
covalently bound to bromocyano-Sepharose or
bromocyano-Sephadex particles, which cannot be
taken up by the ectoderm cells. Control experi-
ments have shown that the inducing activity is not
due to a release of the bound factors [51]. This
suggested that a signal transduction mechanism is
involved in neural induction. We could show that
neural tissues are formed in isolated ectoderm of
Triturus alpestris [52] and to a lesser extent in
Xenopus ectoderm which in addition differentiates
also to mesodermal tissues [53], when phorbolester
(PMA =phorbolmyristate-acetate; TPA=tumor
promoting agent) is added [52]. Phorbolester
activates protein kinase C (PKC), which is
assumed to be involved in the transduction of
signals from the plasma membrane to the nucleus.
The activity of proteinkinases has therefore been
measured in isolated gastrula ectoderm induced
with a neuralizing factor. Davids [53] has shown
that the activities of proteinkinase C (or a related
enzyme), which was measured with an enzyme-
specific peptide substrate, as well as of c-AMP/c-
GMP dependent kinases increase after induction.
Addition of c-AMP or c-GMP or their mono- and
dibutyryl derivatives to ectoderm does, however,
not evoke neural differentiation [54]. It is there-
fore unlikely that the activation of c-AMP/c-GMP
dependent kinases is the primary event in neural
induction. Several proteins are more strongly
phosphorylated in homogenates of induced
ectoderm [53]. These proteins are also phosphory-
lated in homogenates of neural plates isolated
from early neurula stages. The phosphorylation of
31 kD and 15kD proteins seems to depend on
PKC or a related enzyme. These phosphoproteins
have first been detected 60 min after induction of
isolated ectoderm with neuralizing factor.
[Davids, unpublished experiments]. Their phos-
phorylation may not be the first event in neural
induction, but may rather be part of a phosphory-

176 H. TIEDEMANN

lation cascade. Otte et al. [55] have shown that
protein kinase C is translocated to the plasma
membrane after induction. Whereas phorbolesters
are artificial activators of PKC’s, the physiological
activators are diacylglycerols or unsaturated fatty
acids depending on the subspecies of the PKC’s
[review 56]. This may suggest that the breakdown
of membrane phospholipids is involved in signal
transduction after induction. Diacyglycerols and
Inosintriphosphate are generated by phosphoino-
sitide-specific phospholipase C (PLC), arachidonic
acid is generated by phospholipase Az. Experi-
ments on signal transduction by adenylate cyclase
have shown that the coupling of adenylate cyclase
to effector occupied receptors is mediated by G-
(GTP-binding) proteins. Adenylate cyclase activ-
ity is then terminated by GTP-ase activity intrinsic
to the G-proteins. Non-hydrolyzable GTP-
analogues as GTP,S (guanosine-5’-O-thiotripho-
sphate) has therefore an intensifying effect. The
finding that GIP,S stimulates PLC activity have
led to the assumption that the control of PLC
occurs in a way analogous to adenylate cyclase [57,
58 review 59].

GTP,S (1 ~M) can evoke neural differentiation
in gastrula ectoderm of Triturus alpestris (but not
of Xenopus laevis; Loppnow-Blinde and Tiede-
mann, unpublished experiments). Li* ions which
are known for many years [60, 61] to evoke neural
and mesodermal differentiation in amphibian gas-
trula ectoderm have also been shown to interfere
with the phosphoinositide cycle [62, 63]. This
could suggest that phospholipase C is involved in
the induction mechanism. The other enzyme of
the phospholipid metabolism which could be in-
volved, phospholipase A>, is easily activated by
disturbances of plasma membrane conformation.
It is possible that such processes could be related
to the so called “autoneuralization” effect.

It has to be stressed that absolutely no auto-
neuralization occurred under the conditions for the
test of neuralizing factors.

Phorbolester can activate the Na*/H~* antiport
system [64, 65].
stages of Triturus alpestris (but at the concentra-
tions employed not ectoderm of Xenopus) forms
neural N-(2-hydroxyethyl) pipe-
razine-N-ethansulphonic acid (Hepes) in its proto-

Ectoderm from early gastrula

tissues when

nated form is added to the medium as a buffer
substance [66]. These and other observations led
to the consideration that Hepes could lead to an
export of H* from the ectoderm cells by an
activation of the Na*/H* antiport system. Ethyl-
isopropyl-ameloride (100 ~M) a potent and spe-
cific inhibitor of the Na*/H* antiport [67], does,
however, not inhibit the induction of Triturus
alpestris ectoderm by the neuralizing factor. [Cra-
goe and Hildegard Tiedemann, unpublished ex-
periments]. Similarly the action of growth factors
is not inhibited by Ameloride derivatives in phy-
siological bicarbonate buffer. These and other
observations [reviewed in 68] suggest that a change
of pH is probably not an intracellular messenger
for neural induction or growth stimulation.

Concanavalin A, a lectin which binds especially
to mannose residues in glycoproteins and Con-
canavalin A coupled to Sepharose evoke neural
differentiation in gastrula ectoderm of Triturus
pyrrhogaster [69] as well of Xenopus laevis [70] and
of Rana temporaria {71, 71a]. The latter is only
weakly induced by Con A-Sepharose [71]. Con-
canavalin could either bind to a cell surface recep-
tor for the neuralizing factor or it could change the
conformation of the plasma membrane after bind-
ing to distinct sites. Other lectins lead to a loss of
neural competence [72]. Retinoic acid does not
induce neural differentiation in gastrula ectoderm
[Hildegard Tiedemann, unpublished experiments,
73]. The substance causes, however, microcepha-
ly. It has been suggested that retinoic acid specifies
regional differentiation of the central nervous sys-
tem in amphibians [73] and in chicken the anterior-
posterior axis during limb development [74]. The
identification of nuclear receptors for retinoic acid
in several tissues speaks strongly for its function as
a physiological regulator, but its many teratogenic
actions at a low concentration make it somewhat
difficult to discriminate between these two possibi-
lities.

A neural cell adhesion molecule [N-CAM, 75] is
expressed during early neurogenesis in Xenopus
[76]. Other neural specific proteins expressed after
induction are neurofilaments and tetanus-toxin
binding sites [77].

Embryonic Induction 177

Induction of mesoderm and endoderm
and the factors involved

The presumptive dorsal mesoderm has been
regarded as the “organizer” of embryonic develop-
ment. This should imply that this region is already
determined to its fate in the fertilized egg. But
when in 1962 Nakamura [78] isolated the presump-
tive mesoderm (the marginal zone) from different
developmental stages of Triturus pyrrhogaster, the
isolated mesoderm from very early stages did not
differentiate into mesodermal tissues, its prospec-
tive fate. The marginal zone acquires its dif-
ferentiation capacity in the morula stage. This
demonstrated the epigenetic development of the
“organizer” [79]. Hildegard Tiedemann [80] in
1965 observed that gastrula endoderm of Triturus
alpestris when implanted into the blastocoel of
early gastrula hosts induced in the ventral
ectoderm mesenchymatic tails in about 20% of the
cases. In 1967 Ogi [81, 82] combined isolated
endoderm and ectoderm and obtained the induc-
tion of mesodermal tissues. He explained the
formation of mesoderm as a result of regulation on
the basis of two opposite animal-vegetal and vege-
tal-animal physiological gradients. The induction
of mesodermal tissues in ectoderm explants which
were combined with endoderm has been investi-
gated in detail by Nieuwkoop and collaborators
[83, 84] and Nakamura and collaborators [85].
Nakamura emphasized the importance of an anim-
al-vegetal gradient, Nieuwkoop the induction of
mesodermal tissues in ectoderm by the endoderm
[86, 87]. Both views are certainly not mutually
exclusive. Grunz and Tacke [88] have shown that
the induction of mesoderm is not prevented by
placing a Nucleopore filter between endoderm and
ectoderm. Electronmicroscopy did rule out cell
processes traversing the filter. The inducing effect
is obviously mediated by diffusible factor(s).
Dawid et al. [89] came to a similar conclusion.
They observed that the appearance of a muscle
specific marker was prevented by completely dis-
sociating and dispersing Xenopus embryos during
the period from early cleavage to early gastrula, a
procedure that would dilute secreted inducing fac-
tors. Gurdon ef al. [90] have concluded from
dissection experiments that the “subequatorial”

zone of the fertilized Xenopus egg contains all
components for muscle gene activation. Because
the boundaries of the subequatorial zone are not
exactly defined, the zone could include some pre-
sumptive endoderm. It is, however, not excluded
that active factors which are needed for the dif-
ferentiation of mesodermal tissues are in the ferti-
lized egg already localized in the vegetal most part
of the marginal zone. Asashima [91] has investi-
gated the inducing capacity of endoderm from
different stages of Triturus alpestris. Endoderm
taken from uncleaved eggs induces mesothel and
blood cells in a low percentage, whereas endoderm
from later stages in addition induced muscle,
notochord and pronephric tubules. Blastula en-
doderm has the highest inducing activity. There-
after the inducing activity declines. The increase
may depend on the activation and release of
masked factor(s). Xenopus endoderm induces
mesoderm from the cleavage to the early gastrula
stage [92].

The inducing capacities of the dorsal and the
ventral endoderm differ. Boterenbrood and
Nieuwkoop [93] have shown that the dorsal en-
doderm induces dorsal mesodermal tissues
(notochord and somites) whereas the ventral en-
doderm induces more ventral mesodermal tissues
(absence of notochord, no well arranged somites,
blood cells). Experiments with cell lineage labels
and region specific markers confirmed that the
dorsovegetal material induces dorsal type
mesoderm and ventrovegetal material ventral type
mesoderm [94]. Yamada has already shown in
1940 [95] that organs which are formed from
different presumptive mesodermal regions change
to a more dorsal type (i.e. blood cells to nephric
tubules or nephric tubules to somites) when the
notochord anlage is added to the explants. This
suggests that within the presumptive mesoderm a
dorso-ventral gradient of (still unknown) regula-
tory factor(s) is established, which in addition to
factors from the ventral and dorsal endoderm is
involved in the subdivision of the mesoderm. Gur-
don et al. have shown that in embryos which have
just completed gastrulation a-skeletal and a-
cardiac actin genes start to be transcribed in the
somite region of the mesoderm and to a lesser
extent in the ventral mesoderm, which possibly

178 ; H. TIEDEMANN

gives rise to the heart [96]. Actin c-DNA probes
have been used as mesoderm markers. The ability
to react to inducing factors, the competence of the
ectoderm, is limited to certain stages. The reason
for this temporal limitation is not yet known. In
Triturus alpestris [97| and to a lesser extent in
Xenopus laevis [98], the loss of competence is
delayed when the protein synthesis in the
ectoderm is inhibited.

A factor which induces mesoderm and en-
doderm has been isolated from 9-11 days old
chicken embryos by Tiedermann et al. [99-102].
The factor is protein in nature. The most efficient
way for its separation from nucleic acids is the
extraction with phenol [103]. The phenol proce-
dure was developed because at that time it was
thought that the factor could be RNA in nature.
The phenol procedure has then been widely used
for the preparation of RNA. The RNA did,
however, not show inducing activity. The final
purification of the factor was achieved by size
exclusion and reversed phase HPLC. The acid
stable factor, which is enriched about 10° times,
has been called vegetalizing factor, because the
tissues which are induced constitute the vegetal
half of the embryo. On the basis of our earlier
investigations the factor has recently been isolated
in higher yield [Tiedemann ef al. unpublished
results]. The method employs extraction with acid
ethanol, the final purification is achieved by four
consecutive steps of reversed phase HPLC. The
factor induces at a concentration of 0.5—-1.0 ng/ml
in about 50% of the cases mesoderm, including
muscle. The apparent molecular mass of the factor
determined by SDS-polyacrylamide-electrophore-
sis is about 25,000 Dalton and that of the biologi-
cally inactive subunits after reduction of disulfide
bridges 13,000 Dalton, the isoelectric point about
pH 8,0. By size exclusion chromatography in 50%
formic acid an apparent molecular mass of 13,000
was found [102]. The dissociation into subunits
may be caused by reduction of interchain or in-
trachain disulfide bonds by formic acid and confor-
mational changes. The inducing activity is dimi-
nished after size exclusion HPLC in 50% formic
acid. It is only partially restored after the removal
of formic acid. The inducing activity is not dimi-
nished when the partially purified factor was incu-

‘
\
¥
\e
t

Fic. 3. A. Mesoderm induced in a Triturus alpestris
larva by the implantation method. Induced tissues:
N=notochord; M=muscle; PN=pronephric
tubules. AS=Axis system of the host larva. B.
Section through a Xenopus explant with induced
somites.

Embryonic Induction 179

bated with formic acid. A mesoderm inducing
factor (Mr 23,500) which was isolated by Smith et
al. from the XTC (fibroblast) cell line of Xenopus
laevis [105] has similar properties [104]. Another
factor which has similar properties as the factor
from chicken embryos has recently been isolated
from calf kidney (Plessow and Davids, unpub-
lished experiments). This suggests that the factor
is an evolutionary conserved protein which may
also have regulatory functions in later stages of
embryogenesis, in adult differentation processes
such as erythrocyte or cartilage differentiation or
in regeneration processes. Asashima and cowor-
kers have made the interesting observation that
activin A, which is identical with the erythroid
differentiation factor (EDF), has mesodem induc-
ing activity at a low concentration [106].

The vegetalizing factor induces, depending on
its concentration all kinds of mesodermal tissues.
Endoderm seems preferentially to be induced at a
very high concentration. At gradually lower con-
centrations pronephros, somites (Fig. 3), noto-
chord and mesothelia are induced [107]. In addi-
tion to mesodermal and endodermal tissues cells
with the typical appearance of primordial germ
cells were observed in explants which were cul-
tured for at least 20 days [108].

When tested at a very high concentration by the
implantation method the vegetalizing factor causes
an exovagination (not exogastrulation) of the gas-
trula (Fig. 4). Endoderm which had invaginated
during gastrulation, reappears in the blastopore
and spreads over the induced ectoderm. The
exovagination is caused by a change of cell affini-
ties [109] of the gastrula ectoderm induced to
endoderm and mesodem. A similar effect has
been observed after injection of XTC-cell factor
into the blastocoel of Xenopus embryos [110].

The vegetalizing factor is in contrast to the
neural inducing factor inactivated after covalent
coupling to BrCN-sepharose or BrCN-sephadex
[111]. The activity is completely recovered after
degradation of the sephadex matrix with dextra-
nase [112]. This suggests that the factor must be
taken up by the cells to become biologically active.
It does not exclude that cell surface receptors are
involved.

A factor from guinea pig bone marrow which
was partially purified by Yamada and Takata in-
duces as the vegetalizing factor besides

mesodermal also endodermal tissues [10, 113].
The histological identification of endodermal tis-
sues 1s, however, difficult because endoderm diffe-
rentiates late.

The availability of endodermal

Fic. 4. Exovagination of Triturus alpestris embryos produced by the implantation of vegetalizing factor in high

concentration into the blastocoel.
A. N=small neural plate.
the AAAS. B. =rudiment of epidermis.

The embryos are partially overspread by white migrating yolk-rich endoderm.
From Kocher-Becker and Tiedemann, Science, 147: 167 (1965).

Copyright 1965 by

A small rudimental neural plate was found in histological sections.

180 H. TIEDEMANN

markers will therefore facilitate the detection of
endoderm. Recently Rosa [114] has isolated
mRNA’s induced in Xenopus ectoderm by a par-
tially purified XTC-cell factor. One of these
RNA’s encoding a homeodomain containing pro-
tein (Mix 1), which is expressed 20° after the
addition of an inducing factor (XTC-cell factor or
bFGF and TGF-f in combination) to ectoderm, is
found in the embryo mostly in the future en-
doderm. Jones et al. [115] have prepared a mono-
clonal antibody which reacts with tail-bud en-
dodermal tissues to identify endoderm induced in
ectoderm explants.

Besides from guinea pig bone marrow meso-
derm inducing factors have been extracted and
partially purified from liver [116] and from the carp
swim bladder [117].

The vegetalizing factor is in vivo in part bound
to an acidic proteoglycan [118] and in vitro binds to
heparin-sepharose which was used by Born et al.
for affinity chromatography of the factor [119]. It
was therefore tempting to investigate whether
heparin binding growth factors of the FGF (fibro-
blast growth factor) protein family induce
mesodermal tissues. It could indeed be shown that
a-(acidic) as well as b-(basic) FGF induce the
formation of mesodermal tissues [120-123]. Like
the vegetalizing factor from chicken embryos both
FGF’s induce at a high concentration somites and
at lower concentrations endothelium lined vesicles
which contain, besides some pycnotic cells, single
cells with the typical appearance of immature
blood cells [123]. Recombinant human b-FGF
induces at higher concentrations besides skeletal
muscle also heart muscle with its typical honey-
comb like appearance, surrounded by a mesothe-
lium lined pericardial cavity [124].

In Xenopus the determination of heart
mesoderm occurs prior to the end of gastrulation.
The heart mesoderm is located in the gastrula in
the deep zone lateral to the head mesoderm and
migrates laterally and ventrally to fuse in the
ventral midline during the late neurula stage [re-
view 125]. The deep dorsal endoderm seems to
contribute to the specification of heart mesoderm,
whereas the superficial pharyngeal endoderm may
enhance heart morphogenesis during later stages

[126]. It is possible that in Xenopus ectoderm

explants endoderm, which is induced by b-FGF,
undergoes regional differentiation and specifies
the heart anlage. In urodeles (Triturus alpestris)
no heart is formed when the endoderm is removed
at the neural plate stage [127].

Notochord is not or very seldom induced by the
FGF’s in Xenopus ectoderm. The notochord
anlage is the dorsal most part of the mesoderm. It
has been suggested that FGF’s induce preferential-
ly ventral mesoderm [120]. The spectrum of
tissues, which are induced depends also on the
concentration of the factors, the species, and the
test methods which are used. Recombinant b-FGF
induces besides other mesodermal tissues also
notochord in ectoderm explants of Triturus alpes-
tris. Notochord is very rarely induced by the
vegetalizing factor from chicken embryos in Xeno-
pus ectoderm explants, but is induced at higher
frequency when tested by the implantation method
on Triturus alpestris gastrulae. Acidic and basic
fibroblast growth factors show an amino acid sequ-
ence homology of 57%. To the FGF protein
family belong also oncogene products and inter-
leukins [review 128]. The protein products of the
oncogenes int-2 and hst/ks (kfgf) have been shown
to induce mesoderm with different potencies [129].

The vegetalizing and the fibroblast growth fac-
tors share heparin affinity but differ in other
properties such as hydrophobicity, inactivation af-
ter reduction of disulfide bonds and molecular
mass. In these properties the vegetalizing factor
and the XTC-cell factor are more closely related to
the transforming growth factors 8. The transform-
ing growth factors ? stimulate phenotypic trans-
formation (anchorage independent growth) of two
cell lines, but their preferential action seems to be
growth inhibition. Whether TGF-f stimulates or
inhibits cell growth seems to depend on the entire
set of growth factors acting on a cell [130, 131].
The promotion of angiogenesis by TGF-f seems to
be mediated by monocytes which are attracted and
stimulated to synthesize interleukin 1 [132]. The
TGF-@ family comprises genes with regulatory
properties in embryogenesis, the ? subunits of
inhibin and the activins, substances which regulate
the release of the follicle stimulating hormone
[review 128]. The erythroid differentiation factor
[EDF; 133] is identical to activin A, a homodimer

Embryonic Induction 181

consisting of two Ba subunits [134]. The TGF-2
family includes also the Vgl gene, which was
discovered by Weeks and Melton [135]. The
m-RNA transcribed from this gene is uniformly
distributed in the cytoplasm of immature Xenopus
oocytes, but is then translocated to the vegetal half
where it is localized as a crescent at the vegetal
pole of mature oocytes [136].

Rosa et al. and Knochel et al. have shown that
transforming growth factors induce mesodermal
tissues [122, 137, 138]. TGF-f1 and £2 induce at a
concentration of 1 g/ml in Triturus alpestris
ectoderm in about 60% of the cases small endothe-
lium lined cavities which contain immature blood
cells as well mesenchyme and in elongated ex-
plants at one pole a dense blastema tissue and
metameric strands of cells like lateral plate
mesoderm, which in the distal part of the explant
form large masses of endothelial (mesothelial)
networks. The networks can form capillary like
structures. Muscle and notochord are induced in
Xenopus and Triturus ectoderm only by TGF-/).
Xenopus explants were not induced by TGF-/;
[137]. The TGF’s or closely related factors induce
in mammalian cell culture cartilage [139]. Asashi-
ma et al. [117] have recently shown that activin A
(EDF) induces mesoderm at a low concentration.
Activin A has a 40% sequence homology to TGF-
f. Activins and inhibins bind as the vegetalizing
factor to heparin-Sepharose [140]. The affinity of
these factors to heparin is, however, lower as
compared to the fibroblast growth factors. Binding
to heparin depends on the native protein structure.
Because the TGF’s are extracted under dissociat-
ing conditions which change their protein con-
formation, it is not known whether the TGF’s bind
also to heparin.

The transforming growth factors are probably
not identical with, but related to the vegetalizing
factor and the XTC-cell factor. TGF-f8 and a
mesoderm inducing factor in human _ blood
platelets can be separated by size exclusion chro-
matography [Dau et al. unpublished experiments].
The growth factors must be applied to gastrula
ectoderm in higher concentrations than the induc-
ing factors for mesoderm induction.

The factors which determine endoderm and
induce mesoderm in the embryo have not yet been

definitively identified. Kirschner et al. [141] have
found that a m-RNA which is present in Xenopus
oocytes and newly transcribed in the neurula stage,
codes for a protein that is 84% identical to human
b-FGF. The recombinant protein which was ex-
pressed from the c-DNA, induces at 20-50 ng/ml
muscle specific actin m-RNA. This protein may be
a natural inducer. b-FGF like proteins have been
enriched from Xenopus eggs and embryos by
heparin-Sepharose affinity chromatography [141,
142]. The factor is extracted in higher yield in the
presence of Chaps, a zwitterionic detergent
[Tiedemann er al., unpublished experiments] and
may in part be bound to particulate structures.
Slack et al. [143] have identified receptors for the
fibroblast growth factor in Xenopus Dblastula
ectoderm. Besides the 4,2 kb transcript coding for
the b-FGF like protein, a smaller transcript of 1 kb
has been found which represents an antisense
transcript to part of the FGF gene. It codes for an
evolutionary conserved protein with a hitherto
unknown function [144]. In addition to a b-FGF-
like factor mesoderm inducing factors which are
not bound to heparin-sepharose are present in
Xenopus embryos. So far we could not extract
with acid ethanol from the early stages of amphi-
bian embryos a mesoderm inducing factor with
properties similar to the vegetalizing or the XTC-
cell factor. This could be due to the low solubility
of the crude proteins, or sequences homologous to
these factors could be integrated into larger pro-
teins with other properties.

That different factors induce mesoderm is not
unexpected. The factors could either induce more
dorsal or more ventral regions of the mesoderm.
They may also interfere with different targets in
signal transduction chains from the cell surface to
the chromatin.

Gene activation and pattern formation
in early embryogenesis

In experiments with the vegetalizing factor from
chicken embryos Minuth and Grunz [145] have
shown that the differentiation of liver is enhanced
by preventing interactions between the induced
cells by dissociation of the induced Triturus

ectoderm for 20hr before reassociation.

182 H. TIEDEMANN

Mesodermal tissues were induced at a high percen-
tage if the ectoderm was not dissociated. This
suggests that not different threshold concentra-
tions of one factor, but cell interactions, in which
additional factors are involved, are needed for the
induction of different mesodermal tissues. Other
experiments support this view. A shift in the
quality of the induced tissues from mostly en-
doderm (induced at a high concentration of vege-
talizing factor) to muscle and notochord was
observed when a protein fraction, which was sepa-
rated during the purification of the factor, was
added to the highly purified factor. The added
protein fraction alone had no mesoderm inducing
activity [146]. Additional factors seem also to be
involved in the induction of mesoderm by TGF-f.
Medium which was conditioned by TGF-f induced
ectoderm enhances the inducing activity [138].
This suggests that additional factors are secreted,
which are either synthesized or activated in gastru-
la ectoderm treated with TGF ~.

This does, however, not imply that endoderm is
generally induced first and that factors generated
in the endoderm then induce mesodermal tissues.
A gene or genes activated by an inducer could
activate other genes in the same cell or in neigh-
boring cells, so that a network of genes would be
generated. In induced ectoderm explants a large
variety of interactions would be possible depend-
ing on inducer concentration, time of inducer
action and of geometry. This can explain that in
explants a variety of tissues in different propor-
tions are induced.

The factors for determination of the axis system
of the embryo and for the induction of the neural
anlage are at least in part of maternal origin. The
position of the factors which determine endoderm
and mesoderm in the oocyte depends on cytoplas-
mic movements after fertilization [147]. The
vegetal most blastomeres play an important part in
axis formation. Gimlich and Gerhart [149] have
shown that after UV-irradiation of the egg, which
impairs the formation of axial mesodermal and of
neural structures, one to three cells of the vegetal
most octet of blastomeres from non-irradiated
embryos of the 64 cell stage can partially or
completely reconstitute axis formation. The in-
ducing factors have probably their highest concen-

tration in these cells. Whether the maternal fac-
tor(s) which determine the endoderm act within
the cell in which they are located, or by an
autocrine mechanism on neighboring cells remains
to be shown.

The mesoderm inducing factors are located in
fertilized eggs and early embryos in a graded
distribution [review 80, 87]. The precise localiza-
tion of the factors and their mRNA’s will, how-
ever, only be known when the genes coding for the
factors have been isolated. It will then be possible
to synthesize c-DNA’s and after insertion into
expression vectors the proteins, so that the dis-
tribution of the factors and their mRNA’s can be
mesured by immunofluorescence or hybridization
methods. So far only a Xenopus b-FGF related
gene has been isolated [141].

The areas in which the factors are located in the
embryo are probably larger than the areas of the
tissues which are determined by these factors. A
small amount of a mesoderm inducer is found in
the animal (ectodermal) cap [150]. It is likely that
not only a vegetal-animal graded distribution of
factors, but also an animal-vegetal distribution of
so far unknown factors exists. Animal pole ex-
plants of Xenopus express epidermis specific anti-
gens which are not expressed in the vegetal half.
The information to express one of these antigens is
present in the animal half before cleavage [151,
152].

It should be borne in mind that the factors can
be masked so that their total concentration is not
equal to the concentration of the biologically ac-
tive factor(s) or that other substances could coun-
teract the inducing factors. Furthermore as in
Drosophila, factors which repress gene activities
[153] could be involved. Thus the ratio of two
factors could decide whether a gene is activated in
a certain position in the embryo. One factor in a
graded distribution could on the other hand acti-
vate more than one gene depending on different
threshold concentrations of the factor. A concen-
tration dependent activation of different genes has
been observed for the Drosophila biocoid protein
[154-157]. It is, however, unlikely that one and
the same factor directly induces different tissues at
different threshold concentrations. A number of
evolutionary conserved genes including homologs

Embryonic Induction 183

of Drosophila regulatory genes are transcribed in
Xenopus oocytes and embryos. Their differential
expression in the embryo is one of the earliest
events leading to tissue differentiation. The dis-
tribution of regulatory gene products seems not to
be confined to the borders of the germ layers which
in later stages reflect the tissue borders [114, 158].

These regulatory genes include genes which
specify proteins with homeotic domains [review
159; 24, 25, 114, 160, 161] as well proteins with
finger domains [162-164]. Both domains bind to
DNA sequences and are thought to act as trans-
cription factors.

Differential cell affinities which develop in the
embryo [165] and the differential distribution of
molecules of the extracellular matrix will then
guide the morphogenetic process.

ACKNOWLEDGMENTS

Our own investigations which are included in this
review were supported by the Deutsche Forschungs-
gemeinschaft and Fonds der Chemischen Industrie.

REFERENCES

1 Hadorn, E. (1965) Brookhaven Symposium on
Biology, 18: 148-161.

2 Grunz, H., Multier-Lajous, A.-M., Herbst, R. and
Arkenberg, G. (1975) Wilhelm Roux’s Arch. Dev.
Biol., 178: 277-284.

3 Spemann, H. and Mangold, H. (1924) Wilhelm
Roux’s Arch. Entwicklungsmech. Org., 100: 599-
638.

4 v. Woellwarth, C. (1952) Roux’s Arch. Dev. Biol.,
145: 582-668.

5 Eyal-Giladi, H. (1954) Arch. Biol., 65: 179-259.

6 Mangold, O. (1923) Wilhelm Roux’s Arch. Ent-
wicklungsmech., 100: 198-301.

7 Tiedemann, Hildegard (1986) In “Cellular Endo-
crinology: Hormonal Control of Embryonic and
Cellular Differentiation” Ed. by G. Serrero and J.
Hayashi, Alan R. Liss Inc., New York, pp. 25-34.

8 Grunz, H. (1987) Zool. Sci., 4: 579-591.

9 Becker, U., Tiedemann, H. and Tiedemann, H.
(1959) Z. Naturforsch., 14b: 608-609.

10 Yamada, T. and Takata, K. (1961) Dev. Biol., 3:
411-423.

11 Roberts, A. B., Anzano, M. A., Myers, Ch. A,
Wideman, J., Blacher, R., Pan Yu-Ch. E., Stein,
St., Lehrman, S. R., Smith, J. M., Lamb, L. C.
and Sporn, M. (1983) Biochemistry, 22: 5692-

12

13

14

32

33

34

35

36

37

38

5698.

Grunz, H. and Tacke, L. (1989) Cell Differ.
Develop., 28: 211-218.

Asashima, M. (1980) Roux’s Arch. Dey. Biol.,
188: 123-126.

Saxén, L. and Toivonen, S. (1962) Primary
Embryonic Induction. Logos Press-Academic
Press, London.

Saxén, L. and Toivonen, S. (1961) J.Embryol.
exp. Morphol., 9: 514-533.

Nieuwkoop, P. D. (1952) J. exp. Zool., 120: 1-
108.

Tiedemann, H., Becker, U. and Tiedemann, H.
(1963) Biochim. Biophys. Acta, 74: 557-560.
Tiedemann, H. and Tiedemann, H. (1964) Rev.
Suisse de Zool., 71: 117-137.

Saxén, L., Toivonen, S. and Vainio, T. (1964) J.
Embryol. exp. Morph., 12: 333-338.

Toivonen, S. and Saxén, L. (1968) Science, 159:
539-540.

Toivonen, S. (1967) Exp. Biol. Med., 1: 1-7.
Holtfreter, J. (1933) Roux’s Arch. Dev. Biol., 127:
591-618.

Holtfreter, J. (1933) Roux’s Arch. Dev. Biol., 127:
619-775.

Sharpe, C. R., Fritz, A., De Robertis, E. M., and
Gurdon, L. B. (1987) Cell, 50: 749-758.
Brivanlou, A. H. and Harland, R. M. (1989)
Development, 106: 611-617.

Saxén, L. (1989) Int. J. Dev. Biol., 33: 21-48.
Saxén, L. (1961) Dev. Biol., 3: 140-152.
Toivonen, S., Tarin, D., and Saxén, L. (1976)
Differentiation, 5: 49-55.

Tacke, L. and Grunz, H. (1988) Cell Differentia-
tion, 24: 33-44.

Grunz, H. and Staubach, J. (1979) Differentiation,
14: 59-65.

Born, J., Janeczek, J., Schwarz, W., Tiedemann,
H. and Tiedemann, H. (1989) Cell Differ. De-
velop., 27: 1-7.

Lane, Ch. D. (1983) Current Topics in Develop.
Biol., 18: 89-116.

John, M., Janeczek, J., Born, J., Hoppe, P.,
Tiedemann, H. and Tiedemann, H. (1983) Roux’s
Arch. Dev. Biol., 192: 45-47.

Duprat, A. M., and Gualandris, L. (1984) Cell
Differentation, 14: 105-112.

Suzuki, A. S., Nakatake, H. and Hidaka, T. (1984)
Differentiation, 28: 73-77.

Siegel, G., Grunz, H., Grundmann, U., Tiede-
mann, H. and Tiedemann, H. (1985) Cell Dif-
ferentiation, 17: 209-219.

Saxén, L. and Lehtonen, E. (1978) J. Embryol.
exp. Morph., 47: 97-109.

Ekblom, P., Nordling, S., Saxén, L., Rasilo, M.-L.

184

39
40

41

42

43

44

45

46

47

48

49

50

51

oy

53

54

55

56

57

58

59

and Reukonen, O. (1979) Cell Differentiation, 8:
347-352.
Engel, J. (1989) Febs Letters, 251: 1-7.

Mikhailov, A. T., Gorgolyuk, N. A., Virtanen, I.
and Lehto, V.-P. (1984) Ontogenez, 15: 137-145.
Janeczek, J., Born, J., Hoppe, P., Schwarz, W..,
Tiedemann, H. and Tiedemann, H. (1986) In
“Cellular Endocrinology: Hormonal Control of
Embryonic and Cellular Differentiation” Ed. by
G.Serrero and J. Hayashi, Alan R. Liss, New
York, pp. 11-24.

Janeczek, J., John, M., Born, J., Tiedemann, H.
and Tiedemann, H. (1984) Roux’s Arch. Dev.
Biol., 193: 1-12.

Janeczek, J., Born, J., Scharschmidt, M., Tieder-
mann, H. and Tiedemann, H. (1984) Eur. J.
Biochem., 140: 257-264.

Janeczek, J., Tiedemann, H. and Tiedemann, H.
(1986) Progress in Developmental Biology, B.
Alan R. Liss, New York, pp. 357-360.
Tiedemann, H. (1984) In “The Role of Cell
Interactions in Early Neurogenesis”. Ed. by A.-M.
Duprat, A. C. Kato and M. Weber, Plenum Press,
New York, pp. 89-105.

Malacinski, G. M., Chung, H.-M. and Asashima,
M. (1980) Dev. Biol., 77: 449-462.

Nakamura, O., Takasaki, H., Okumoto, T. and
lida, H. (1971) Proc. Japan Acad., 47: 203-208.
Bretzel, G. and Tiedemann, H. (1986) Roux’s
Arch. Dev. Biol., 195: 123-127.

Bretzel, G., Janeczek, J., Born, J., John, M.,
Tiedemann, H. and Tiedemann, H. (1986) Roux’s
Arch. Dev. Biol., 195: 117-122.

Grunz, H., Born, J., Tiedemann, H. and Tiede-
mann, H. (1986) Roux’s Arch. Dev. Biol., 195:
464-466.

Born, J., Hoppe, P., Janeczek, J., Tiedemann, H.
and Tiedemann, H. (1986) Cell Differ., 19: 97-
101.

Davids, M., Loppnow, B., Tiedemann, H. and
Tiedemann, H. (1987) Roux’s Arch. Dev. Biol.,
196: 137-140.

Davids, M. (1988) Roux’s Arch. Dev. Biol., 197:
339-344.

Grunz, H. and Tiedemann, Hildegard (1977)
Roux’s Arch. Dev. Biol., 181: 261-265.

Otte, P. A., Koster, C. H., Snoek, G. T. and
Durston, A. J. (1988) Nature, 334: 618-620.
Kikhawa, U., Kishimoto, A. and Nishizuka, Y.
(1989) Ann. Rev. Biochem., 58: 31-44.

Wallace, M. A. and Fain, J. N. (1985) J. biol.
Chem., 260: 9527-9530.

Cockroft, S. and Gomperts, B.D. (1985) Nature,
314: 534-536.

Majerus, Ph, W., Connolly, Th. M., Deckmyn, H.,

H. TIEDEMANN

67

68

69

70
71

Tla

72

73

74

75

81

Ross, Th. S., Bross, T. E., Ishii, H., Bansal, V. S.
and Wilson, D. B. (1986) Science, 234: 1519-1526.
Masui, Y. (1960) Mem. Konan Univ., 4: 79-102.
Grunz, H. (1968) Roux’s Arch. Dev. Biol., 160:
344-347.

Berridge, M. J., Downes C. P. and Hanley, M. R.
(1982) Biochem. J., 206: 587-595.

Busa, W. B. and Gimlich, R. L. (1989) Dev. Biol.,
132: 315-324.

Burns, C. P. and Rozengurt, E. (1983) Biochem.
Biophys. Res. Commun., 116: 931-938.

Rosoff, P. M., Stein, L., Cantley, L. C. (1984) J.
Biol. Chem., 259: 7056-7060.

Tiedemann, Hildegard (1986) Roux’s Arch. Dev.
Biol., 195: 399-402.

Vigne, P., Ferlin, C., Cragoe, E. J. Jr. and Laz-
dunski, M. (1984) Molecular Pharmacol., 25: 131-
136.

Thomas, R. C. (1989) Nature, 337: 601.

Takata, K., Yamamoto, K. Y. and Ozawa, R.
(1981) Roux’s Arch. Dev. Biol., 190: 92-96.
Grunz, H. (1985) Cell Differentiation, 16: 82—92.
Mikhailov, A. T., Gorgolyuk, N. A. and Bibikova
(1989) Ontogenez, 20: 507-515.

Mikhailov, A. T., and Gorgolyuk, N. A. (1987)
Cell Differentiation, 22: 145-154.

Duprat, A.-M., Gualandris, L., Kan, P, and Foul-
quier, F: (1984) In “The Role of Cell Interactions
in Early Neurogenesis”. Ed. by A.-M. Duprat, A.
C. Kato and M. Weber, Plenum Press, New York,
pp. 3-20.

Durston, A. J., Timmermans, J. P. M., Hage, W.
J., Hendriks, H. F. J., de Vries, N. J., Heidewald,
M. and Nieuwkoop, P. D. (1989) Nature, 340:
140-144.

Thalier, C. and Eichele, G. (1987) Nature, 327:
625-628.

Thiery, J. P., Dubaud, S. L., Rutishauser, V., and
Edelman, G. M. (1982) Proc. Natl. Acad. Sci.
USA, 79: 6737-6741.

Edelman, G. M. (1986) Ann. Rev. Cell. Biol., 2:
81-96.

Kintner, G. R. and Melton, D. A. (1987) Develop-
ment, 99: 311-325.

Duprat, A. M., Gualandris, L., Foulquier, F.,
Paulin, D. and Bizzini, B. (1986) Cell Differentia-
tion, 18: 57-64.

Nakamura, O. (1962) Japn. J. Exp. Morph., 15:
61-62.

Nakamura, O. (1967) Japn. J. Exp. Morph., 21:
256-275.

Tiedemann, H. (1975) In “The Biochemistry of
Animal Development”. Vol. III Ed. by R. Weber,
Academic Press, New York, p. 276.

Ogi, K. (1967) Sci. Rep. Tokoku Univ., Biol., 33:

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

Embryonic Induction

239-247.

Ogi, K. (1969) Res. Bull. Gen. Educ. Nagoya
Univ., 13: 31-40.

Nieuwkoop, P. D. (1969) Roux’s Arch. Dev.
Biol., 162: 341-373.

Nieuwkoop, P. D. and Ubbels, G. A. (1972)
Roux’s Arch. Dev. Biol., 169: 185-199.
Nakamura, O., Takasaki, H. and Ishihara, M.
(1971) Proc. Japan Acad., 47: 313-318.
Nakamura, O. and Takasaki, H. (1970) Proc.
Japan Acad., 46: 546-551.

Nakamura, O., Hayashi, J. and Asashima, M.
(1978) In “Organizer”. Ed. by O. Nakamura and
S. Toivonen, Elsevier-North Holland, Amsterdam,
New York, pp. 1-47.

Grunz, H. and Tacke, L. (1986) Roux’s Arch.
Dev. Biol., 195: 467-473.

Sargent, T. D., Jamrich, M. and Dawid, I. B.
(1986) Dev. Biol., 114: 238-246.

Gurdon, J. B., Mohun, T. J., Fairman, Sh. and
Brennan, S. (1985) Proc. Natl. Acad. Sci. USA,
82: 139-143.

Asashima, M. (1975) Roux’s Arch. Dev. Biol.,
177: 301-308.

Jones, E. A. and Woodland, H. R. (1987) De-
velopment, 101: 557-563.

Boterenbrood, E. C. and Nieuwkoop, P. D. (1973)
Roux’s Arch. Dev. Biol., 173: 319-332.

Smith, J. C., Dale, L. and Slack, J. M. W. (1985)
J. Embryol. Exp. Morphol. 89 (Suppl.),: 317-331.
Yamada, T. (1940) Okajimas Fol. anat. Jap., 19:
131-197.

Mohun, T. J., Brennan, S., Dathan, N., Fairman,
S. and Gurdon, J. B. (1983) Nature, 311: 716-721.
Grunz, H. (1970) Roux’s Arch. Dev. Biol., 165:
91-102.

Grainger, R. M. and Gurdon, J. B. (1989) Proc.
Natl. Acad. Sci. USA, 86: 1900-1904.
Tiedemann, H. and Tiedemann, H. (1959) Hoppe-
Seyler’s Z. physiol. Chem., 306: 7-32.

Born, J., Geithe, H.-P., Tiedemann, H., Tiede-
mann, H. and Kocher-Becker, U. (1972) Hoppe-
Seyler’s Z. physiol. Chem., 353: 1075-1084.
Geithe, H.-P., Asashima, M., Asahi, K.-I, Born,
J., Tiedemann, H. and Tiedemann, H. (1981)
Biochim. Biophys. Acta, 676: 350-356.

Born, J., Hoppe, P., Schwarz, W., Tiedemann, H.,
Tiedemann, H. and Wittmann-Leibold, B. (1985)
Biol. Chem. Hoppe-Seyler, 366: 729-735.
Tiedemann, H. and Tiedemann, H. (1956) Hoppe
Seyler’s Z. physiol. Chem., 306: 132-142.

Grunz, H., Born, J., Davids, M., Hoppe, P.,
Loppnow-Blinde, B., Tacke, L., Tiedemann, H.
and Tiedemann, H. (1989) Roux’s Arch. Dev.
Biol., 198: 8-13.

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

185

Smith, J. C., Yaqoob, M., and Symes, K. (1988)
Development, 103: 591-600.

Asashima, M., Nakano, H., Shimada, K., Kinoshi-
ta, K., Ishii, K., Shibai, H. and Ueno, N. (1990)
Roux’s Arch. Dev. Biol., 198: 330-335.

Grunz, H. (1983) Roux’s Arch. Dey. Biol., 192:
130-137.

Kocher-Becker, U. and Tiedemann, H. (1971)
Nature, 233: 65-66.

Kocher-Becker, U., Tiedemann, H. and Tiede-
mann, H. (1964) Science, 147: 167-169.

Cooke, J. and Smith, J.C. (1989) Dev. Biol., 131:
383-400.

Tiedemann, H. and Born, J. (1978) Roux’s Arch.
Dev. Biol., 184: 285-299.

Born, J., Grunz, H., Tiedemann, H. and Tiede-
mann, H. (1980) Roux’s Arch. Dev. Biol., 189:
47-56.

Takata, C. and Yamada, T. (1960) Embryologia,
5: 8-20.

Rosa, F. M. (1989) Cell, 57: 965-974.

Jones, E. A., Abel, M. H. and Woodland, H. R.
(1989) Abstract XI Congr. Int. Soc. Dev. Biol.,
Utrecht, The Nederlands.

Ya-Huei, W., Hui-yin, M. and Jie-Yi, S. (1963)
Acta Biol. Exp. Sinica, 8: 370-386.

Asashima, M., Nakano, H., Matsunaga, T., Sugi-
moto, M. and Takano, H. (1987) Develop.
Growth Differ., 29: 221-227.

Niebel, J., Tiedemann, H. and Tiedemann, H.
(1973) Eur. J. Biochem., 32: 242-246.

Born, J., Davids, M. and Tiedemann, H. (1987)
Cell Differentiation, 21: 131-136.

Slack, J. M. W., Darlington, B. G., Heath, J. K.
and Goodsave, S. F. (1987) Nature, 326: 197-200.
Kimelman, D. and Kirschner, M. (1987) Cell, 51:
869-877.

Knochel, W., Born, J., Loppnow-Blinde, B.,
Tiedemann, H., Tiedemann H., McKeehan, W. L.
and Grunz, H. (1987) Naturwissenschaften, 74:
604-606.

Grunz, H., McKeehan, W. L., Knochel, W., Born,
J., Tiedemann, H. and Tiedemann, H. (1988) Cell
Differ., 22: 183-190.

Knochel, W., Grunz, H., Loppnow-Blinde, B.,
Tiedemann, H. and Tiedemann, H. (1989) Blut,
59: 207-213.

Jacobson, A. G. and Sater, A. K. (1988) Develop-
ment, 104: 341-359.

Sater, A. K. and Jacobson, A. G. (1989) Develop-
ment, 105: 821-830.

Mangold, O. (1957) Naturwissenschaften, 44: 289-
290.

Knochel, W. and Tiedemann, H. (1989)
Differ. Develop., 26: 163-171.

Cell

186

129

130

131

132

133
134
135
136

137

138

139
140
141

142
143
144

145

Paterno, G. D., Gillespie, L. L., Dixon, M. S.,
Slack, S. M. W. and Heath, J. K. (1989) Develop-
ment, 106: 79-83.

Sporn, M. B., Roberts, A. B., Wakefield, L. M.,
and Assoian, R. K. (1986) Science, 233: 532-534.
Roberts, A. B. and Sporn, M. (1990) In “Peptide
Growth Factors and Their Receptors”. Handbook
of Experimental Pharmacology Vol. 95 I. Ed. by
M. B. Sporn and A. R. Roberts. Springer, Heidel-
berg pp. 419-472.

Wahl, Sh. M., Hunt, D. A., Wakefield, L. M.,
McCartney-Francis, N., Wahl, L. M. Roberts, A.
B. and Sporn, M. (1987) Proc. Natl. Acad. Sci.
USA, 84: 5788-5792.

Murata, M., Eto, Y., Shibai, H., Sakai, M. and
Muramatsu, M. (1988) Proc. Natl. Acad. Sci.
USA, 85: 2434-2438.

Vale, W., Rivier, J., Vaughan, J., McClintock, R.,
Corrigan, A., Woo, W., Karr, D. and Spiess, S.
(1986) Nature, 321: 776-779.

Weeks, D. L., and Melton, D. A. (1987) Cell, 51:
861-867.

Melton, D. A. (1987) Nature, 328: 80-82.

Rosa, F., Roberts, A. B., Danielpour, L. L.,
Sporn, M. B. and Dawid, I. (1988) Science, 239:
783-785.

Knoéchel, W., Tiedemann, H. and Tiedemann, H.
(1989) Naturwissenschaften, 76: 270-272.
Seyedin, S. M., Segarini, P. R., Rosen, D. M.,
Thompson, A. Y., Bentz, H. and Graycar, J.
(1987) J. Biol. Chem., 262: 1946-1949.

Ling, N., Ying, Sh.-Y., Ueno, N., Shimasaki, Sh.,
Esch, F., Hotta, M. and Guillemin, R. (1986)
Nature, 321: 779-782.

Kimelman, D., Abraham, J. A., Haaparanta, T.,
Palisi, Th. M. and Kirschner, M. W. (1988) Sci-
ence, 242: 1053-1056.

Slack, S. M. W. and Isaacs, H. V. (1989) Develop-
ment, 105: 147-153.

Gillespie, L. L., Paterno, G. D. and Slack, J. M.
W. (1989) Development, 106: 203-208.

Volk, R., Koster, M., Poting, A., Hartmann, L.
and Knochel, W. (1989) Embo J., 8: 2983-2988.
Minuth, M. and Grunz, H. (1980) Cell Differentia-
tion, 9: 229-238.

H. TIEDEMANN

146

147

148

149

150

151

152

153
154

155

156

157

158

159

160

161

162

163

164

Asahi, K.-i., Born, J., Tiedemann, H. and Tiede-
mann, H. (1979) Roux’s Arch. Dev. Biol., 187:
231-244.

Vincent, J.-P. and Gerhart, J. C. (1984) Dev. Biol.
123: 526-539.

Gerhart, J., Ubbels, G., Black, S., Hara, K. and
Kirschner, M. (1981) Nature (London), 292: 511-
516.

Gimlich, R. L. and Gerhart, J. C. (1984) Dev.
Biol., 104: 117-130.

Tiedemann, H., Becker, U., and Tiedemann, H.
(1961) Embryologia, 6: 204-218.

Jones, E. A. and Woodland, H. R. (1986) Cell, 44:
345-355.

London, Ch., Akers, R. and Philips, C. (1988)
Dev. Biol., 129: 380-389.

Gaul, U. and Jackle, H. (1987) Cell, 51: 549-555.
Struhl, G. Struhl, K. and Macdonald, P. M. (1989)
Cell, 57: 1259-1273.

Driever, W. and Niisslein-Volhard, C. (1988) Cell,
54: 95-104.

Driver, W. and Nisslein-Volhard, C. (1989) Na-
ture, 337: 138-143.

Schroder, C., Tautz, D., Seifert, E. and Jackle, H.
(1988) Embo. J., 7: 2882-2887.

Wright, C. V. E., Cho., K. W. Y., Hardwicke, J.,
Collins, R. H. and De Robertis, E. M. (1989) Cell,
59: 81-93.

Gehring, W. J. (1987) Science, 236: 1245-1252.
Miller, M. M., Carrasco, A. E. and De Robertis,
E. M. (1984) Cell, 39: 157-162.

Ruiz i Altaba, A. and Melton, D. A. (1989)
Development, 106: 173-183.

Tautz, D., Lehmann, R., Schnurch, H., Schulz,
R., Seifert, E., Kienlin, K. and Jackle, H. (1987)
Nature, 327: 383-389.

Koster, M., Pieler, T., P6ting, A. and Knéchel, W.
(1988) EMBO J., 7: 1735-1741.

Nietfeld, W., El Baradi, T., Mentzel, H., Pieler,
T., Koster, M., Poting, A. and Knéchel, W. (1989)
J. Mol. Biol., 208: 639-659.

Townes, P. I. and Holtfreter, J. (1955) J. exp.
Zool., 128: 53-120.

ZOOLOGICAL SCIENCE 7: 187-194 (1990)

© 1990 Zoological Society of Japan

Ammoniagenesis in the Mudskipper,
Periophthalmus chrysospilos

YUEN K. Ip, Soir F. CHEw and Rita W. L. Lim

Department of Zoology, National University of Singapore,
Kent Ridge, Singapore 0511

ABSTRACT—Enzymes associated with nitrogen metabolism were determined in the gill, liver and
muscle tissues of Periophthalmus chrysospilos. Glutamate dehydrogenase, alanine transaminase and
aspartate transaminase were detected in all tissues studied. A complete purine nucleotide cycle was not
present. Glutamine synthetase was not detectable, suggesting that this enzyme was not responsible for
detoxification of intra-mitochondrially generated ammonia. Glutamine was the most effective substrate

for in vitro ammoniagenesis in the isolated muscle mitochondria.

The possible role of glutamate

dehydrogenase in mitochondrial ammoniagenesis in this mudskipper is discussed.

INTRODUCTION

In teleostan fishes, ammonia is the major nit-
rogenous endproduct while urea and trimethyla-
mine oxide serve mainly for the maintenance of
Osmotic equilibria in species that synthesize them
in significant quantities [1, 2]. Available data
indicate that the sites of ammonia production in
fish occur mainly in the liver and kindey [3]. The
ammonia formed is then transported via the blood
and excreted through the gills. However, gills
without being surrounded by water are not
efficient excretory organs. It, therefore, intrigues
biologists to study the mechanisms by which
amphibious fishes solve their problem of terrestrial
nitrogenous waste excretion.

Mudskippers fascinate scientists for they enjoy
an amphibious mode of life that is unparalleled
among fishes. They resemble anuran amphibians
in many aspects of their behaviour and ecology [4].
Gregory [5] found both ammonia and urea but no
uric acid in the tissues of mudskippers. Except for
arginase and carbamoyl transferase, other en-
zymes invovled in the ornithine-urea cycle are
either not detectable in the mudskippers Perioph-
thalmus expeditionium and P. gracilis or have
activities too low to account for the possiblility of a

Accepted May 2, 1989
Received December 8, 1988

shift towards ureotelism when they are on land.
Several studies on ammonia and urea excretion by
mudskippers in aquatic and terrestrial environ-
ments have been performed [6-9]. Ammonia is
mainly accumulated in the body of the fish during
terrestrial exposure. Conversion of accumulated
ammonia to urea is hardly performed. For this to
be possible, the mudskipper must have a great
tolerance to ammonia. Indeed, P. cantonensis can
survive in 15 mM NH,Cl for more than 7 days [10].

Chew and Ip [11] reported that transdeamina-
tion is the major pathway for ammoniagenesis in
the mudskippers Boleophthalmus boddaerti and
Periophthalmodon schlosseri. | Aspartate was
found to be the the major substrate for in vivo
mitochondria ammonia production. However,
aspartate transaminase activities in the muscle and
hepatic mitochondria presented in the same report
cannot account for such a high rate of ammo-
niagenesis from this substrate. Moreover, if
ammonia production from aspartate involved
transdeamination and occurred at such high rate, it
is difficult to understand why then in vitro respira-
tion of the isolated mudskipper muscle mitochon-
dria in the presence of adenosine diphosphate
(ADP) was stimulated by externally added gluta-
mine and glutamate but not aspartate. In order to
shed light on this matter, the present studies were
undertaken to confirm the pathways of molecular
ammonia production in another local mudskipper

188 Y.K. Ip, S. F. CHEW AnD R. W. L. Liv

P. chrysospilos. Also, during the course of the
investigation, it was found necessary to re-examine
the in vitro production of ammonia by the
mitochondria of B. boddaerti and P. schlosseri in
the presence of glutamine, glutamate and aspar-
tate by the more advanced and selective ammonia
assay method adopted in the present studies.

MATERIALS AND METHODS

Field collection and fish maintenance

P. chrysospilos of both sexes, weighing 2 to 12 g,
were collected along the shores of West Coast and
Pasir Ris, Singapore. B. boddaerti (8-17 g body
weight) and P. schlosseri (70-80 g body weight) of
both sexes were caught at the estuary at Pasir Ris.
They were maintained in 50% seawater (18%0
salinity) in the laboratory for at least one day
before being sacrificed for the experiments.

Preparation of sample for enzyme activity deter-
minations

The lateral muscle, liver and gills of freshly
killed P. chrysospilos were excised, blotted dry
and weighed. Due to the samll size of the fish,
each gill sample contained tissues pooled from 7-8
fish while liver and muscle from 4-5 and 2 fish
respectively constituted a sample. Samples were
homogenized in ice-cold buffers (0-4°C) and frac-
tionated according to the procedure of Iwata et al.
[12]. The resulting mitochondrial and cytosolic
fractions were used for glutamate dehydrogenase
(GDH, EC 1.4.1.2), aspartate transaminase (EC
2.6.1.1), alanine transaminase (EC 2.6.1.2), gluta-
mine synthetase (EC 6.3.1.2) and phosphate-
dependent glutaminase (EC 3.5.1.2) activities de-
termination.

For the preparation of adenosine monophos-
phate deaminase (AMP deaminase, EC 3.5.4.6),
adenylosuccinate lyase (SAMP lyase, EC 4.3.2.2)
and adenylosuccinate synthetase (sAMP synth-
etase, EC 6.3.4.4), the homogenized samples were
centrifuged at 40,000 g for 1 hr. The resulting
supernatant fluid was used for enzyme assays.
Buffer used for AMP deaminase contained 50 mM
Tris cacodylate (pH 7.1) and 2 mM dithiothreitol.
For the extraction of sAMP lyase, buffer contain-

ing 0.25M sucrose, 5mM Hepes (pH 7.4) and 5
mM EDTA was used. The buffer used for sAMP
synthetase extraction contained 0.25 M sucrose, 5
mM Hepes (pH7.2), 1mM EDTA and 1mM
dithiothreitol.

Enzyme assays

Spectrophotometric measurements were made
at 25°C with a Shimadzu UV 260 double beam
spectrophotometer. Specific activities were ex-
pressed as enzyme activity per mg protein.

Glutamate dehydrogenase was determined in
the mitochondrial fraction according to Iwata et al.
[12]. For the reductive amination direction, one
unit of enzyme activity is defined as that amount
which oxidised one “mol of NADH per min. In
the oxidative deamination direction, one unit of
enzyme acivity is defined as one “mol of NAD
reduced per min.

Transaminases were measured in both the cyto-
solic and mitochondrial fractions. For alanine
transaminase, the assay procedure was based on
the method of Wroblewski and LaDue [13].
Aspartate transaminase was measured with a Sig-
ma aspartate aminotransferase assay kit no. 56-UV
(Sigma Chemical Co.). For both transaminases,
one unit of enzyme is defined as one «mol of
NADH oxidized per min.

Glutamine synthetase was determined col-
orimetrically by the method of Pamilijans ef al.
[14], using creatine phosphate and creatine kinase
as the ATP regenerating system. It was also
assayed according to Rowe ef al. [15] where ATP
was added to the reaction mixture. Phosphate-
dependent glutaminase was determined by the
method of Curthoys and Weiss [16] in both the
cytosolic and mitochondrial fractions. For gluta-
minase, one unit of enzyme is defined as one «mol
of NAD reduced per min.

AMP deaminase was assayed by the method of
Gibbs and Bishop [17]. One unit of enzyme is
defined as the amount that produces one «mol of
ammonia per min. sAMP lyase was determined
spectrophotometrically at 280 nm according to
Campbell and Vorhaben [18]. One unit of enzyme
is defined as the disappearance of one «mol of
adenylosuccinate per min. For sAMP synthetase,
both spectrophotometire and radiometric assays

Ammoniagenesis in Mudskipper 189

were also perfomred according to Campbell and
Vorhaben [18].

Preparation of muscle mitochondria for in vitro
ammonia production and oxidative phosphoryla-
tion studies

The lateral muscle of freshly killed P. chrsospi-
los, B. boddaerti and P. schlosseri were excised,
blotted dry and weighed. They were put into 10
volumes of ice-cold (0-4°C) extraction buffer con-
taining 305 mM sucrose, 3 mM Tris-HCl (pH 7.2),
3mM EDTA and 0.1% bovine serum albumin
(BSA, Sigma Chemical Co.) with a final osmolality
of 330 mosmo/kg. Homogenization was carried
out once by using an Ikawerk Staufen Ultra-
Turrax (Janke and Kunke Co.) homogenizer at
16,500 rpm for 10 sec. The homogenized samples
were centrifuged at 4°C and 1,200 g for 10 min
(Kokusan model H-103N). The supernatant fluid
was carefully drained and further centrifuged at
7,000 g for 15 min (Kokusan model H-251CS) to
obtain the mitochondria. The mitochondrial pel-
lets were pooled together in extraction buffer and
gently homogenized once with a Wheaton Teflon-
pestle homogenizer. This sample was centrifuged
at 1,200 g for 10 min again to remove any con-
taminating muscle tissue. Mitochondria were
obtained by resedimentation for 15 min at 7,000 x
g, washed one more time with extraction buffer
and centrifuged again at the same gravitainal force
for the same duration. The washed mitochondria
were finally resuspended in a small volume of
extraction buffer by gently homogenizing once as
stated before. Normally each gram of muscle
produced approximately 0.18mg of purified
mitochondrial protein with minimum contaimina-
tion of cytosolic lactate dehydrogenase.

In vitro ammonia production studies

To determine ammonia production from the
isolated mitochondria of P. chrysospilos, 0.1 ml of
the mitochondrial preparation (0.2—0.3 mg of mus-
cle mitochondrial protein) was added to 1 ml of
extraction buffer which was supplemented with 10
mM final concentration of the desired substrates
with or without inhibitors. An aliquot of the
sample was immediately assayed for ammonia by a
Tecator Aquatec 5200 Analyzer equipped with the

NRED cassette and interfaced with a Tecator FIA
Star 5032 Detector Controller. Ammonium chlor-
ide solution was used as standard. The principle of
this application is that the aqueous sample contain-
ing ammonium ions is mixed with sodium hydrox-
ide. The ammonia released is allowed to diffuse
through a gas permeable membrane to react with
the indicator. The resulting colour shift was then
measured spectrophotometrically. After 1 hr of
incubation at 25°C, the ammonia content of the
same sample was analyzed again. Such procedure
allowed for the differentiation of ammonia origi-
nally present in the mitochondrial preparation or
present in the amino acid substrate and that being
produced by the mitochondria during the course of
incubation. Normally, no attempt was made to
differentiate between intra- and _ extra-
mitochondrial ammonia. Ammonia production
was reported as “mol ammonia produced per mg
protein per hr.

In order to find out if the ammonia produced can
exit the muscle mitochondria of P. chrysospilos,
samples were incubatd in the presence of 10 mM of
glutamine for 1 hr at 25°C. The incubated samples
were then centrifuged at 10,000 x g for 5 min. The
ammonia content of the supernatant fluid was
analysed. Precipitated mitochondria were lysed by
the addition of 1 ml of 0.4M perchloric acid
followed by neutralization with 30% KOH. After
the removal of potassium perchlorate by centri-
fugation, mitochondrial ammonia content was de-
termined.

In vitro oxidative phosphorylation studies

Preliminary studies in our laboratory showed
that ADP: O ratios could not be used as a good
indicator of the satisfactory coupling condition of
the mudskipper mitochondria due to the presence
of high level of phosphatase activities. Therefore,
oxidative phosphorylation by isolated P. chrysos-
pilos muscle mitochondria was measured directly
by determining the incorporation of radioactive
inorganic phosphate (*Pi) into the organic fraction
according to the procedure of Grunberg-Manago
et al. [19]. Mitochondria obtained in extraction
buffer were suspended in respiratory medium
(0.16 mg mitochondrial protein/ml) of 330 mos-
molal consisting of 40 mM Hepes (pH 7.2), 70 mM

190 Y.K. Ip, S. F. CHEw AnD R. W. L. Lim

sucrose, 90 mM D-mannitol, 5mM MgCh, 3 mM
EDTA, 4mM KH >PO, containing 0.2 ~Ci/uzmol
of “Pi (NEN), 0.1% BSA, 30mM glucose, 2.5
mM ADP and 601U/ml of hexokinase (Sigma
Chemical Co.). Reactions were started by the
addition of 1 ml of the mitochondrial suspension
into test tubes containing 10 mM final concentra-
tion of substrate. After 20 min of incubation at
25°C, the reaction was terminated by the addition
of 0.1 ml of 40% trichloroacetic acid. Precipitated
protein was removed by centrifugation. Radioac-
tivity incorporated into glucose-6-phosphate was
separated from *’Pi by conversion of the latter to
ammonium phosphomolybdate which was ex-
tracted by isobutanol and ether [19, 20]. Radioac-
tivity was determined by using Aquasol II (NEN)
and a Kontron Betamatic scintillation spectro-
meter. Samples of the initial °"Pi were counted at
the same time as the incubated samples to circum-
vent corrections for isotopic decay. Quenching
effects were monitored and rectified by the sample
channel ratio method.

Protein assay

Protein was measured using the method of
Bradford [21] Bovine serum albumin (Sigma Che-
mical Co.) was used as standard for comparison.

RESULTS

Determination of enzymes associated with trans-
deamination and the purine nucleotide cycle in P.
chrysospilos

The enzymes associated with transdeamination
and the purine nucleotide cycle were assayed in the
liver, gill and muscle tissues of P. chrysospilos.
The activities of these enzymes were presented in
Table 1. GDH was detected in all tissues ex-
amined. Similar to reports for other teleosts [22,
23], the liver mitochondria of P. chrysospilos
showed the highest activities of GDH in both
directions; the oxidative deamination rate being
approximately 4.4% of that of reductive amina-
tion.

Alanine transaminase and aspartate transami-
nase were detected in all tissues studied with
greater activities in the cytosolic than mitochond-
rial fractions. The activity of the former was
higher than that of the latter in all tissues.

Glutamine synthetase was not detectable in P.
chrysospilos by the two methods used in the pre-
sent studies. Phosphate-dependent glutaminase,
however, was detected only in the mitochondrial
fractions of all the tissues.

Both sAMP lyase and AMP deaminase were
present in all the tissues studied. The activity of
sAMP layse in the muscle was approximately 5
times greater than those in the liver and the gills.

TABLE 1. Specific activities of enzymes involved in ammoniagenesis in the tissues of P. chrysospilos*
Gill Liver Muscle
Enzymes
Cytosol Mitochondria Cytosol Mitochondria Cytosol Mitochondria

Glutamate dehydro

genase

reductive amination — 0.288 + 0.087 — 1.913+0.049 — 0.036 + 0.003

oxidative deamination — 0.015 +0.003 — 0.084 + 0.005 _ 0.004 + 0.003
Aspartate transaminase 3.680+0.482 0.107+0.013 7.114+2.950 1.423+0.441 3.895+0.128 0.085+0.040
Alanine transaminase 0.516+0.012 0.007+0.002 1.562+0.677 0.033+0.008 0.597+0.087 0.004+0.002
AMP deaminase 0.227+0.126 as 0.050 +0.001 — 0.054 + 0.007 —
sAMP lyase 0.032 + 0.006 — 0.047 — 0.183 +0.053 a
Phosphate-dependent n.d. 0.079 + 0.034 n.d. 0.085 + 0.062 n.d. 0.016 +0.009

glutaminase

*The values given are the means+SD of enzyme specific activities (refer to Materials and Methods) from three
different samples except for sAMP lyase of the liver (n= 1) and sAMP lyase of the muscle and gill (n=2); —=not

determined; n.d.=not detectable.

Ammoniagenesis in Mudskipper 191

The greatest AMP deaminase activity was however
recorded in the gill tissue. sAMP synthetase was at
first assayed spectrophotometrically. No activity
was detected although the sample had been di-
alysed overnight to remove the interfering sAMP
lyase. When the more sensitive radiometric
method was employed, results were still negative.

In vitro ammonia production from the muscle
mitochondria of P. chrysospilos in the presence of
various L-amino acids

The muscle mitochondria of P. chrysospilos was
able to deaminate all amino acids tested except
serine (Table 2). The order of effectiveness in
stimulating ammonia production was glutamine >
glutamate > arginine > proline >lysine > alanine >
aspartate.

The effects of various inhibitors on mitochond-
rial ammonia production were also examined
(Table 2). Aminoxyacetate inhibited ammonia
production from aspartate completely and signi-
ficantly reduced ammonia production from gluta-
mine. When aminoxyacetate was included in the
incubation medium, it gave a higher than normal
blank absorbance value; 0.042+0.001 (n=5) as
compare to zero for distilled water. However, it
did not interfere with the ammonia assay process
as the corrected absorbance values for 0.5, 2 and 5
mg/l of ammonium chloride standard in 2mM
aminoxyacetate were 0.029+0.001 (n=5), 0.118+
0.003 (n=5) and 0.291 + 0.007 (n=5S) respectively,
which were not significantly different (P >0.05)
from the control values of 0.027+0.002 (n=5),
0.119+0.002 (n=5) and 0.287+0.002 (n=S) in

TABLE 3.

TABLE 2. Jn vitro ammonia production (mol
NH;/mg protein per hr+SD) from various
L-amino acids (10 mM) in the absence or
presence of either 12 mM bromofuroate or
2 mM aminoxyacetate at 25°C by isolated
muscle mitochondria of P. chrysospilos

Condition n Ammonia produced
Glutamine 12 0.119+0.013
Glutamine + bromofuroate 5 0.025 + 0.006
Glutamine+aminoxyacetate 3 0.038 + 0.002
Glutamate 9 0.039 + 0.017
Glutamate + bromofuroate 3 0
Arginine 3 0.027+0.010
Proline 30.018 +.0.004
Lysine 6 0.010 +0.004
Alanine 5 0.009 + 0.003
Aspartate 6 0.003 +0.001
Aspartate + bromofuroate 3 0
Aspartate + aminoxyacetate 3 0
Serine 3 0

the absence of the inhibitor.

Bromofuroate, an inhibitor of glutamate dehyd-
rogenase, decreased significantly (p<0.01) the
ammonia production from glutamine and stopped
ammonia release from glutamate and aspartate.

When glutamine was used as the substrate, 69.45
+7.31 (n=5) % of the ammonia produced was
found in the incubation medium and 32.29+6.79
(n=5) % was located within the mitochondria.

In vitro oxidative phosphorylation by muscle
mitochondria of P. chrysospilos

Mitochondria isolated from the muscle of P.

In vitro ammonia production (mol NH;/mg protein per hr+SD, n=4)

from 10 mM of glutamine or glutamate or aspartate in the absence or presence
of either 12 mM bromofuroate or 2mM aminoxyacetate at 25°C by isolated
muscle mitochondria of B. boddaerti and P. schlosseri

Ammonia produced

Condition

B. boddaerti P. schlosseri
Glutamine 0.225+0.011 0.106 + 0.060
Glutamate 0.026 +0.001 0.027 + 0.010
Aspartate 0.009 +0.003 0.007 + 0.003
Glutamine + bromofuroate 0.019 +0.007 0.040 + 0.009
Gluatmate + bromofuroate 0 0
Aspartate + bromofuroate 0 0
Aspartate + aminoxyacetate 0 0

192 Y.K. Ip, S. F. CHEW AND R. W. L. Lim

chrysospilos could readily undergo oxidative phos-
phorylation in the presence of externally added
substrate and ADP. Rate of phosphorylation
obtained in the presence of glutamine, glutamate
and aspartate, after correction for intrinsic phos-
phorylation in the presence of 2.5 mM ADP only,
were 2.432+0.254 (n=7), 0.520+0.059 (n=3)
and 0.056+0.011 ~mol *Pi incorporated per mg
mitochondrial protein per 20 min respectively.

In vitro ammonia production from the muscle
mitochondria of B. boddaerti and P. schlosseri in
the presence of either glutamine, glutamate or
aspartate

Mitochondria isolated from the muscle of B.
boddaerti and P. schlosseri were able to produce
ammonia in vitro in the presence of the three
substrates tested (Table 3). The order of effective-
ness in stimulating ammonia production was gluta-
mine >glutamate >aspartate. Bromofuroate also
decreased significantly the ammonia production
from glutamine and stopped ammonia release
from glutamate and aspartate. Aminoxyacetate
inhibited mitochondrial ammonia production from
aspartate totally.

DISCUSSION

The deamination of amino acids through the
purine nucleotide cycle proposed by Braunstein
[24] and Lowenstein [25] has been suggested to
occur in fish [26]. Similar to the other mudskip-
pers, B. boddaerti and P. schlosseri [11], AMP
deaminase and sAMP lyase were present in the
tissues of P. chrysospilos. However, sAMP synth-
etase was not detectable by the two methods
employed, thus rendering it improbable for this
cycle to play a significant role in ammoniagenesis
in this mudskipper. Janicki and Lingis [27] demon-
strated that the liberation of ammonia from aspar-
tate in teleost liver required both the mitochond-
rial and cytosolic fractions, a result which would
not be expected if purine nucleotide cycle were
solely responsible for ammoniagenesis. Casey ef
al. [22] also showed that the purine nucleotide
cycle was not responsible for ammoniagenesis in
the catfish, /. punctatus L. as the heavy nitrogen of
'SN-alanine was not incorporated into AMP in

isolated hepatocytes.

The activities of the mitochondiral GDH in P.
chrysospilos were slightly lower (10%) than those
obtained for the rat mitochondria [28] but much
higher than the value reported previously by Iwata
et al. [12] for another mudskipper P. cantonensis in
the direction of reductive amination. Thermo-
dynamically, GDH reaction favours reductive
amination of a-ketoglutarate rather than oxidative
deamination of glutamate [29]. In P. chrysospilos,
the activities of GDH determined in vitro in the
reductive amination direction were indeed greater
than those in the oxidative deamination direction.
However, the activities in vivo might in fact favour
glutamate oxidation and ammonia release owing to
factors such as relative levels of nucleotides and
removal of reaction products [29]. When the GDH
reaction is operated in conjunction with other
transaminases present, transdeamination would be
an avenue of ammonia production from various
amino acids in the mudskipper.

Similar to mammalian [30] and channel catfish
[31] hepatic mitochondria, the mechanism for
aspartate deamination in isolated P. chrysospilos
muscle mitochondria is transdeamination as it can
be inhibited totally by either aminoxyacetate, a
transaminase inhibitor, or bromofuroate, an in-
hibitor of GDH. However, contrary to the report
on B. boddaerti and P. schlosseri [11], aspartate
was not an effective substrate for in vitro
mitochondrial ammonia production in P. chrysos-
pilos. By re-examining the in vitro ammoniagenesis
in the muscle mitochondria of the former two
mudskippers by the ammonia assay methods de-
scribed herein, it was confirmed that aspartate was
not as efficient a substrate as compared to gluta-
mine and glutamate for ammonia production
(Table 3). Since aspartate normally is a non-
penetrant anion in the absence of glutamate [32,
33] and the aspartate transaminase activity in the
mitochondria of both B. boddaerti and P. schlos-
seri were not high enough to account for the fast
rate of ammonia production [11], it is possible that
the previous observations were results of interfer-
ence of the enzyme coupled ammonia assay proce-
dure used due to the presence of small amount of
contaminating cytosolic aspartate transaminase
and malate dehydrogenase. However, the results

Ammoniagenesis in Mudskipper 193

obtained in the present studies (Table 3) verified
the important role of GDH in mitochondrial
ammoniagenesis in B. boddaerti and P. schlosseri
as reported by Chew and Ip [11].

In agreement with the present studies, Campbell
et al. [31] also demonstrated that glutamine was
more effective as a substrate than glutamate for
ammonia production in isolated channel catfish
hepatic mitochondria. Such phenomenon can be
explained by a restricted permeability of the inner
mitochondria membrane to glutamate. When glu-
tamine is a substrate, the permeability of the
mitochondria membrane is not the rate-limiting
step, and glutamate is produced inside the
mitochondria. Glutaminase pesent inside the
mitochondria can release the amide nitrogen to
form ammonia and glutamate which can in turn be
deaminated. The fact that bromofuroate com-
pletely inhibited the release of ammonia from
glutamate and significantly reduced ammonia pro-
duction from glutamine verified the important role
of GDH in ammoniagenesis in the muscle
mitochondria of P. chrysospilos.

Aminoxyacetate, significantly reduced the in
vitro mitochondrial ammonia production from glu-
tamine in P. chrysospilos indicating that another
enzyme, glutamine transaminase [34] may also be
involved in the muscle mitochondrial glutamine
metabolism. Although the present investigation
did not examine the presence of this enzyme in P.
chrysospilos, it has been found in both the cytoso-
lic and mitochondrial compartments of the catfish
liver. The normal products of such enzymatic
reaction are a-ketoglutarate and ammonia. In
order to accommodate the involvement of GDH in
the metabolism of glutamine by the muscle
mitochondria of P. chrysospilos, glutamine trans-
aminase must therefore function in conjuction with
some other enzymes. A combination of glutamine-
phenylpyruvate transaminase, w-amidase and
phenylalanine-a-ketoglutarate transaminase
catalyzes a net phenylpyruvate-stimulated glutami-
nase reaction producing glutamate and ammonia.
Such pathway has indeed been discovered in the
rat kindey. In the presence of bromofuroate,
ammoniagenesis from glutamine was reduced to
such an extent as though deamidation was also
affected. Such observation might not be a direct

effect of bromofuroate on the deamidation pro-
cess, but the inhibitory effect of the accumulating
glutamate on both glutaminase and phenyalanine-
a-ketoglutarate transaminase.

The isolated mitochondria used in these studies
were in the coupling state as they could readily
undergo oxidative phosphorylation in the presence
of externally added substrate and ADP. In agree-
ment with the ammonia studies, glutamine was a
more effective substrate for oxidative phosphory-
lation than glutamate and aspartate. The fact that
the ammonia produced could exit the coupling
mitochondria indicates that exiting ammonia must
be accompanied by a proton so that the hydrogen
gradient generated by the electron transport sys-
tem was not disrupted. The absence of glutamine
synthetase suggested that this enzyme was not
responsible for the detoxification of intra-
mitochondrially generated ammonia in this muds-
kipper.

ACKNOWLEDGMENTS

This project was supported by grants RP70/85 from the
National University of Singapore and GR05690J from
the Singapore Turf Club.

REFERENCES

1 Goldstein, L. and Forster, R. P. (1970) Nitrogen
metabolism in fishes. In “Comparative Biochemistry
of Nitrogen Metabolism, vol. 2”. Ed. by J. W.
Campbell, Academic Press, New York, pp. 495-
518.

2 Wood, J. D. (1958) Nitrogen excretion in some
marine teleosts. Can. J. Biochem. Physiol., 36:
1237-1242.

3 Pequin, L. and Serfaty, A. (1963) L’Excretion
ammoniacale chez un teleosteen dulcicole: Cyprinus
carpis L. Comp. Biochem. Physiol., 10: 315-324.

4 Stebbins, R. C. and Kalk, M. (1961) Observations
on the natural history of the mudskipper Perioph-
thalmus sobrinus. Copeia, 1: 18-27.

5 Gregory, R. B. (1977) Synthesis and total excretion
of waste nitrogen by fish of the Periophthalmus
(mudskipper) and Scartelaos Families. Comp.
Biochem. Physiol., 57A: 33-36.

6 Gordon, M. S., Boetius, J., Boetius, I., Evan, D.
H., McCarthy, R. and Oglesby, L. C. (1965) Salin-
ity adaptation in the mudskipper fish
(Periophthalmus sobrinus). Hvalrad. Skr., 48: 85-
93.

10

11

12

13

14

15

194

Morii, H., Nishikata, K. and Tamura, O. (1978)
Nitrogen excretion of mudskipper fish Periophthal-
mus cantonensis and Boleophthalmus pectinirostris
in water and on land. Comp. Biochem. Physiol.,
60A: 189-193.

Morii, H., Nishikata, K. and Tamura, O. (1979)
Ammonia and urea excretion from mudskipper fish
Periophthalmus cantonensis and Boleophthalmus
pectinirostris transferred from land to water. Comp.
Biochem. Physiol., 63A: 23-28.

Morii, H., (1979) Changes with time of ammonia
and urea concentration in the blood and tissue of
mudskipper fish, Periophthalmus cantonensis and
Boleophthalmus pectinirostris kept in water and on
land. Comp. Biochem. Physiol., 64A: 235-243.
Iwata, K. (1984) A high ammonia tolerance in the
mudskipper Periophthalmus cantonensis. Zool. Sci.,
1: 877.

Chew, S. F. and Ip, Y. K. (1987) Ammoniagenesis
in mudskippers Boleophthalmus boddaerti and
Periophthalmodon schlosseri. Comp. Biochem. Phy-
siol., 87B: 941-948.

Iwata, K., Kakuta, I., Ikeda, M., Kimoto, S. and
Nada, N. (1981) Nitrogen metabolism in the muds-
kipper, Periophthalmus cantonensis: A role of free
amino acids in detoxification of ammonia produced
during its terrestrial life. Comp. Biochem. Physiol.,
68A: 589-596.

Wroblewski, F. and LaDue, J.S. (1956) Serum
glutamic pyruvate transaminase in cardiac and hepa-
tic disease. Proc. Soc. Exp. Biol. Med., 91: 569.
Pamiljans, V., Krishnaswamy, P. R., Dumville, G.
and Meister, A. (1962) Studies on the mechanism of
glutamine synthetase: isolation and properties of the
enzyme from sheep brain. Biochemistry, 1: 153-
158.

Rowe, W. B., Ronzio, R. A., Wellner, V. P. and
Meister, A. (1970) Glutamine Synthetase (sheep
brain). In “Method Enzymol., vol. XVIIA”. Ed. by
H. Tabor and C. W. Tabor, Academic Press, New
York, pp. 900-910.

Curthoys, N. P. and Weiss, R. F. (1974) Regulation
of renal ammoniagenesis. Subcellular localization of
rat kidney glutaminase isoenzymes. J. Biol. Chem.,
249: 3261-3266.

Gibbs, K. L. and Bishop, S. H. (1977) ATP-
activated adenylate deaminase from marine inverte-
brate animals: properties of the enzyme from lug-
worm (Arenicola cristata) body wall muscle.
Biochem. J., 163: 511-516.

Campbell, J. W. and Vorhaben, J. E. (1979) The
purine nucleotide cycle in helix hepatopancreas. J.
Comp. Physiol., 129: 137-144.

Grunberg-Manago, M., Ortiz, P. J. and Ochoa, S.
(1956) Enzyme synthesis of polynucleotides I. Poly-
nucleotide phosphorylase of Azotobacter virlandii.

20

21

22

23

24

25

26

27

28

29

30

31

34

Y.K. Ip, S. F. CHEw Anp R. W. L. Lim

Biochim. Biophys. Acta, 20: 269-285.

Berenblum, I. and Chain, E. (1938) An improved
method for the colorimetric determination of phos-
phate. Biochem. J., 32: 295-298.

Braford, M. M. (1976) A rapid and sensitive
method for the quantitation of microgram quantities
of protein utilizing the principle of protein-dye bind-
ing. Anal. Biochem., 72: 248-254.

Casey, C. A., Perlman, D. F., Vorhaben, J. E., and
Campbell, J. W. (1983) Hepatic ammoniagenesis in
the channel catfish, Jctalurus punctatus. Mol. Phy-
siol., 3: 107-126.

Waarde, A. V. (1980) Nitrogen metabolism in
goldfish, Carassius auratus L.: Activities of trans-
amination reactions, purine nucleotide cycle and
glutamate dehydrogenase in goldfish tissues. Comp.
Biochem. Physiol., 68B: 407-413.

Braunstein, A. E. (1957) Les voies principales de
assimilation et dissimilation de l°azote chez les
animaux. Adv. Enzymol., 19: 335-377.
Lowenstein J. M. (1972) Ammonia production in
muscle and other tissues: the purine nucleotide
cycle. Physiol. Rev., 52: 382-414.

Makarewicz, W. and Zydowo, M. (1962) Compara-
tive studies on some ammonia producing enzymes in
the excretory organs of vertebrates. Comp.
Biochem. Physiol., 6: 269-275.

Janicki, R. and Lingis, J. (1970) Mechanism of
ammonia production from aspartate in teleost liver.
Comp. Biochem. Physiol., 37: 101-105.

Arnold, H. and Maier, K. P. (1971) Crystallization
and some properties of glutamate dehydrogenase
from rat liver. Biochim. Biophys. Acta, 251: 133-
140.

Krebs, H. A. and Veech, R. L. (1969) Pyridine
nucleotide interrelationships. In “The Energy Level
and Metabolic Control in Mitochondria”. Ed. by S.
Papa, J. M. Tager, E. Quagliariello and E. C.
Slater, Adriatica Edice, Bari, pp. 329-382.

Hird, F. J. R. and Marginson, M. A. (1966) Oxida-
tive deamination of glutamate and transdeamination
through glutamate. Arch. Biochem. Biophys., 115:
247-256.

Campbell, J. W., Aster, P. L. and Vorhaben, J. E.
(1983) Mitochondrial ammoniagenesis in liver of the
channel catfish /ctalurus punctatus. Am. J. Physiol.,
244: R709-R717.30.

Chappell, J. B. (1968) Systems used for the trans-
port of substrates into mitochondria. Br. Med.
Bull., 24: 150 -157.

Klingenberg, M. (1979) ADP/ATP shuttle of
mitochondria. Trends Biochem. Sci., 5: 1-4.
Cooper, A. J. L. and Meister, A. (1972) Isolation
and properties of highly purified glutamine trans-
aminase. Biochemistry, 11: 661-671.

ZOOLOGICAL SCIENCE 7: 195-200 (1990)

Expression of Circular Plasmids Which Contain Bacterial
Chloramphenicol Acetyltransferase Gene Connected to
the Promotoer of Polypeptide IX Gene of Human
Adenovirus Type 12 in Oocytes, Eggs and
Embryos of Xenopus laevis

YUCHANG Fu, KENnzo Sato!, Ketcut Hosokawa’

and KoIcutro SHIOKAWA~”>

Department of Biology, Faculty of Science, Kyushu University, Fukuoka 812, and
"Department of biochemistry, Kawasaki Medical College,
Kurashiki 701-01, Japan

ABSTRACT—Polypeptide IX gene of adenovirus type 12 is unique in that it is expressed intermediately
inbetween early and late genes, but the structure and function of its 5’-upstream promoter region have
not been well characterized. In the present experiment, fertilized eggs as well as oocytes and
unfertilized eggs of Xenopus laevis were injected with circular plasmid, pAd12.IXCAT, which contains
bacterial chloramphenicol acetyltransferase (CAT) gene fused to the promoter of polypeptide IX gene
of adenovirus type 12, and the activity of this plasmid to promote CAT enzyme expression in Xenopus
embryonic cells was examined. For comparison, pAd12.ElaCAT which contains the promoter of Ela
protein of adnovirus type 12, pSV2CAT which contains SV40 early promoter, and pSVOCAT and
pA10CAT3m which do not contain promoter were also tested. In the oocyte nucleus, all these circular
plasmids were expressed similarly actively. In embryos and unfertilized eggs, however, while
pAd12.ElaCAT and pSV2CAT were strongly expressed, level of the expression of pAd12.IXCAT was
as low as those of pSVOCAT and pA10CAT3m. These results show that the promoter of polypeptide IX

© 1990 Zoological Society of Japan

gene of adenovirus type 12 is very weak as compared with that of Ela protein.

INTRODUCTION

The polypeptide IX is associated with the group
of nine hexons of adenovirus virion, and may play
a cementing role in the assemblage of virus particle
[1, 2]. Polypeptide IX is expressed intermediately
inbetween the early and late phases of adenovirus
infection, and the regulation of its synthesis
appears to differ from that of other structural
polypeptides of the virus particle [3]. Polypeptide
IX is unique in that it is encoded by a relatively
small mRNA of about 9S (ca. 485 b), and unlike

Accepted May 2, 1989
Received March 29, 1989

2 To whom all correspondence should be addressed.

3 Present address: Laboratory of Molecular Embryolo-
gy, Zoological Institute, Faculty of Science, University
of Tokyo, Tokyo 113, Japan.

other adenovirus mRNAs, the formation of
polypeptide IX mRNA does not involve splicing
[4]. To characterize the promoter function, Kruc-
zek and Doerfler [5] isolated polypeptide IX gene
from adenovirus type 12 genome, and after con-
necting its promoter to CAT gene, studied the
effect of methylation on the promoter function.
However, the expression of CAT enzyme activity
from the fusion gene has not yet been studied in
other eukaryotic cell system. In the present ex-
periment, fertilized eggs as well as oocyte nuclei
and unfertilized eggs of Xenopus laevis were in-
jected with pAd12.IXCAT, a plasmid which con-
tains the promoter of polypeptide IX of adenovirus
type 12 [5] and the activity of the plasmid to
promote CAT enzyme expression was compared
with those of other CAT-containing plasmids.

196 Y. Fu, K. Sato et al.

MATERIALS AND METHODS

Plasmid DNAs

Five different plasmids, pSVOCAT,
pAd12.IXCAT, pAd12.ElaCAT, pA10CAT3m,
and pSV2CAT were used throughout the experi-
ments. pSVOCAT contains CAT gene and has a
Hind III site in front of the CAT gene for ex-
perimental promoter insertion [6]. pAd12.IXCAT
was constructed by inserting the promoter region
(1.2 Kb) of polypeptide IX gene into Hind III site
of pSVOCAT (Fig. 1) [5]. pAd12.ElaCAT was
produced by inserting the promoter of Ela early
gene of adenovirus type 12 into pSVOCAT (Fig. 1)
[5]. pSV2CAT contains a CAT gene fused to a
relatively strong promoter of SV40 early gene (Fig.
1) [6]. pA10CAT3m is a derivative of pSVOCAT,
into which a polylinker was inserted [7].

why 4
ae, wh wt

N
RS ~
2 F
*» <> >
t t t
Fic. 1. Maps of pAd12.IXCAT as compared with

pAd12.ElaCAT and pSV2CAT. Thick lines indi-
cate promoter inserted. IX, E, and 2 corresponds
to pAdi2.IXCAT, pAd12.ElaCAT, and
pSV2CAT, respectively. Arrowheads outside cir-
cles denote Hind III sites, and those inside circles
indicate approximate TATA box positions. Small
arrows denote Hpa II sites. CAT denotes the site of
CAT gene. These figures were drawn according to
Kruczek and Doerfler (5).

Plasmids were propagated, and DNAs were
extracted as described previously by Tashiro et al.
[8, 9]. Agarose gel electrophoresis showed that all
the plasmid DNAs before injection consisted
mainly of closed circular (c.c.) DNA, with little
open circular (o0.c.) DNA (Fig. 2).

Microinjection

Ovaries were digested for 3-4 hr with 500 «g/ml
of collagenase in Barth’s solution (88 mM NaCl, 10
mM Hepes, pH7.4, 1.0mM_ KCl, 0.82 mM
MgSO,, 0.33mM Ca(NO3)., 0.41mM CaCl)
which contained 50 units/ml of penicillin and 50

Buch 1. ein: sa Oe

Fic. 2. Gel electrophoretic profiles of plasmid DNAs
before injection. The main band shown is c.c. form
for all the DNA preparations. Lane 1 (Marker Hind
IlI-digest of lambda DNA), lane 2 (pAd-
12.IXCAT), lane (pSV2CAT), lane 4 (pSVOCAT),
lane 5 (pA1O0CAT3m) and lane 6 (pAd12.ElaCAT).

g/ml of streptomycin. Defolliculated oocytes at
the stage 6 [10] were collected, and injected with
ca. 20nl of 100 ug/ml of DNA solutions. All
oocytes injections were aimed at the germinal
vesicle. Samples for CAT assay and DNA extrac-
tion were prepared from pools of 40 oocytes.
Fertilized eggs were obtained by artificial in-
semination [11], and after being dejellied in 2.5%
thioglycollate (pH 8.0) [12], injected into the cyto-
plasm with ca. 20 nl of 100 ug/ml of DNA solu-
tions in 1X MMR (0.1 M NaCl, 2mM KCl, 1 mM
MgsSO,, 2mM CaCl, 5mM Hepes, pH 7.4, 0.1
mM EDTA) that contained 5% Ficoll [8]. After
injection, embryos were left in 1x MMR with 5%
Ficoll, and transferred to 0.5 MMR when they
reached the stage 6 [13]. Embryos were incubated
until the blastula (stage 8.5), gastrual (stage 11)
and neurula (stage 18-19) stages at 20-
21°C. Ten embryos wer used for a smaple.
Unfertilized eggs were manually stripped out
from the gravid female, dejellied, then incubated
in modified Barths’ saline (88 mM NaCl, 1.0 mM
KCl, 0.83 mM MgSQ,), 0.34 mM Ca(NO3)>, 0.41
mM CaCl, 7.5mM Tris-HCl (pH 7.6), 2.4mM
NaHCOQ;). Eggs were injected into the cytoplasm
with DNA as above [14], and left in the modified

CAT Genes in Xenopus Embryos

Barths’ saline at 20-21°C for either 12 or 24 hr.
About 20 eggs were collected as a sample.

CAT enzyme assay

Oocytes, unfertilized eggs, and embryos were
homogenized in 0.25M Tris-HCl (pH 8.0), and
supernatants equivalent to 10 eggs were mixed
with 1 “Ci of (‘*C)chloramphenicol (Amersham
Corp.) and acetyl CoA. The mixture was incu-
bated at 37°C for 2hr, and was extracted twice
with ethyl acetate [15]. The extracts were spotted
onto a thin layer plate with silica gel, and chroma-
tographed for ca. 30 min in 95% chloroform-5%
methaol. Gels were dried and autoradiographed
usually for 3-5 days (for oocytes and embryos) or
1-2 months (for unfertilized eggs).

RESULTS AND DISCUSSION

Circular pAd12.IXCAT and other plasmids
(pAd12.ElaCAT, pSV2CAT, pSVOCAT,
pA10CAT3m) were injected into the oocyte nuc-
leus at 2 ng/oocyte, and CAT enzyme activity was
assayed after different periods of time. Southern
blot analysis carried out with pSV2CAT as a _probe
showed that injected plasmids were stably pre-
served after 24 hr of incubation (data not shown).

AC3—

197

CAT enzyme expression with pAd12.IXCAT was
not greatly different from that with other four
plasmids (pAd12.ElaCAT, pSV2CAT, pSV0O-
CAT, and pA10CAT3m) at 7 (lanes 1 to 3), 20
(lanes 4 to 9) and 24 hr (lanes 10 to 12) (Fig. 3).
Thus, as in the previous data which were obtained
with chicken ovalbumin genes [16], all the circular
genes were expressed equally actively.

Fertilized eggs were then injected with
pAd12.IXCAT and four other circular plasmids at
2 ng/egg, and CAT enzyme activity was tested at
different stages of development (Fig. 4). Under
the conditions used, CAT enzyme activity was not
detected at the blastula stage with plasmids that
carried adenovirus promoters (lanes 10, 13),
although pSV2CAT was expressed weakly as
Etkin and Balcells [15] recently showed (lane 7).
At the gastrula stage, however, all the plasmids
were expressed at widely differing extents. Thus,
pSV2CAT (lane 8) and pAd12.ElaCAT (lane 14)
were strongly expressed, whereas pAd12.IXCAT
was expressed only at a low level (lane 11), which
was close to those of pSVOCAT (lane 2) and
pA10CAT3m (lane 6). The results were essential-
ly the same also at the neurula stage (Fig. 4).

DNAs were extracted from the above DNA-
injected embryos at the blastula, gastrula and

-AC3
—-AC2
eae e*
Soe ger eC!
@ -C

Pees a6 8 OOF Il 2

Fic. 3. CAT enzyme assay in oocytes injected with circular plasmids. Oocytes were injected with 2 ng of circular
plasmids and harvested at 7 hr (lanes 1 to 3), 20 hr (lanes 4 to 9) and 24 hr (lanes 10 to 12). Lanes 1, 8 and 10
(pSVOCAT); lanes 2, 4, 11 (pAd12.IXCAT); lanes 3, 5, 9, and 12 (pSV2CAT); lane 6 (pAd10CAT3m); and lane
7 (pAd12.ElaCAT). C, AC1, AC2, and AC3 are for chloramphenicol, 1-acetylated chloramphenicol,
3-acetylated chloramphenicol, and 1,3-diacetylated chloramphenicol, respectively.

198 Y. Fu, K. Sato et al.

AGSiaa Ac3
- —_—
a
AC2 —
ACI ~ ~ ee “ - AC2
2o@ oe ea - AC!

C ~ @@@ Ceee--ecece-=— -<

eS 4 SG ie oar

Fic. 4.

lO 1) 12 13 14 15

CAT enzyme assay in developing embryos injected with circular plasmids. Fertilized eggs were injected with

2 ng of circular plasmids, and embryos were collected at the late blastula (lanes 1, 4, 7, 10 and 13), gastrula (lanes
2, 5, 8, 11, and 14), and neurula (lanes 3, 6, 9, 12, and 15) stages. Lanes 1 to 3 (pSVOCAT), lanes 4 to 6
(pA10CAT3m), lanes 7 to 9 (pSV2CAT), lane 10 to 12 (pAd12.IXCAT), and lanes 13 to 15 (pAd12.ElaCAT).

For C, AC1, AC2 and AC3, see the legent to Fig. 3.

neurula stages, and Southern blot analysis was
carried out using pSV2CAT as a probe. The
results showed that injected pAd12.IXCAT as well
as pSVOCAT and pA10CAT3m was not degraded,
but copy number of the injected DNAs increased
by several fold. Therefore, the low activity of
pAd12.IXCAT (Fig. 4) may not be due to the
instability of this plasmid after injection.

Expression of CAT enzyme activity from circu-
lar pAd12.IXCAT and other plasmids was also
tested after injection into unfertilized eggs at 2
ng/egg. As shown in Figure 5, very weak, but
distinct CAT enzyme activity was observed at 12
hr only with pAd1l2.ElaCAT (lane 5) and
pSV2CAT (lane 3), and the activity of other
plasmids was either very faint (pAd12.IXCAT)
(lane 2) or negligible (pSVOCAT, lane 1 and
pA1OCAT3m, lane 4). Thus, results obtained with
unfertilized eggs were quite similar to those
obtained with embryos, although the extent of the
expression was quite low.

Circular plasmids were injected into fertilized
eggs and RNAs were extracted from embryos at
the gastrula stage to compare the level of CAT
mRNA by Northern blot analysis [17] using CAT
antisense RNA [18] as a probe. The results
obtained showed that the level of the mRNA
which migrated at 1.6 Kb CAT antisense RNA was
roughly comparable to the CAT enzyme level

AG2Z ce
= @-
INC = oa
om — ACI
Gua ="
[Pave soe eae
Fic. 5. CAT enzyme assay in unfertilized eggs injected

with circular plasmids. Unfertilized eggs were in-
jected with 2 ng of circular plasmids and harvested at
12 hr after the injection. Lane 1 (pSVOCAT), lane 2
(pAd12.IXCAT), lane 3 (pSV2CAT), lane 4
(pALOCAT3m) and lane 5 (pAd12.ElaCAT). For
C, ACI, and AC2, see the legent to Fig. 3. Since
this tuoradiogram was obtained after autoradiog-
raphic exposure for 2 months, there appeared a faint
band between ACI and AC2 in all the lanes.
However, this is not due to the CAT enzyme
activity, becuase the same spot was obtained also
with sample which had not been injected with CAT
gene-comtaining plasmid (data not shown).

(data omitted). Therefore, we assume that CAT
enzyme activity may be roughly correlated to the

acticity of transcription of CAT genes in

CAT Genes in Xenopus Embryos

embryonic cells.

As shown in Figure 1, the size of the promoter
region of polypeptide IX gene used is almost twice
as large as those of Ela protein and SV40 early
genes. Nevertheless, the activity of CAT enzyme
expression with pAd12.IXCAT that carried the
promoter of polypeptide IX gene was as low as
those of pSVOCAT and pA10CAT3m which do
not contain the promoter.

pSV2CAT and pAd12.ElaCAT were known to
contain a relatively strong enhancer element with-
in their promoter region [5, 6]. However, the
5’-upstream region of the polypeptide IX gene of
adenovirus has not been shown to contain an
enhancer element [1, 2, 4], although the promoter
of protein IX gene of adenovirus type 12 is known
to contain TATA box and GC-rich region [4, 19].
The low activity of pAd12.IXCAT observed here
in Xenopus embryos and unfertilized eggs suggests
that the promoter of the protein IX gene of
adenovirus type 12 may not contain an enhancer
element.

ACKNOWLEDGMENTS

We thank Professor K. Yamana for his warm en-
couragement throughout the experiments. We also
thank Dr. W. Doerfler for kind supply of CAT gene-
containing plasmids. The present study was supported in
part by a Grant-in-Aid for Scientific Research to K. S.
(No. 61540523), a Grant-in-Aid for Cancer Research to
K. H. (No. 62010095) from the Ministry of Education,
Science and Culture of Japan, and grants from Takeda
Science Foundation (1986) and The Naito Foundation
(1987) to K. S.

REFERENCES

1 Everitt, E., Lutter, L. and Philipson, L. (1975)
Structural proeins of adenoviruses. XII. Location
and neighbor relationship among proteins of ade-
novirion type 2 as revealed by enzymatic lodination,
immunoprecipitation and chemical cross-liknking.
Virology, 67: 197-208.

2 Everitt, E., Sundquist, B., Pettersson, U. and Phi-
lipson, L. (1973) Structural proteins of adenoviruses
X. Isolation and topography of low molecular
weight antigens from the virion of adenovirus type 2.
Virology, 52: 130-147.

3 Persson, H., Pettersson, U. and Mathews, M. B.
(1978) Synthesis of a structural adenovirus
polypeptide in the absence of viral DNA replication.

10

12

13

15

16

199

Virology, 90: 67-79.

Alestrom, P., Akusjarvi, G., Perricaudet, M.,
Mathews, M. B., Klessig, D. F. and Pettersson, U.
(1980) THe gene for polypeptide IX of adenovirus
type 2 and its unspliced messenger RNA. Cell, 19:
671-681.

Kruczek, I. and Doerfler, W. (1983) Expression of
the chloramphenicol acetyltransferase gene in
mammalian cells under the control of adenovirus
type 12 promoters: Effect of promoter methylation
on gene expression. Proc. Natl. Acad. Sci. USA, 80:
7586-7590.

Gorman, C. M., Moffat, L. F., and Howard, B. H.
(1982) Recombinant genomes which express chlor-
amphenicol acetyltransferase in mammalian cells.
Mol. Cell. Biol., 2: 1044-1051.

Laimins, L. A., Gruss, P., Pozzatti R. and Khoury,
G. (1984) Characterization of enhancer elements in
the long terminal repeat of Moloney murine sarco-
ma virus. J. Virol., 49: 193-189.

Tashiro, K., Inoue, M., Sakaki, Y. and Shiokawa,
K. (1986) Preservation of Xenopus laevis rDNA-
containing plasmid, pXirl01A, injected into the
fertilized egg of Xenopus laevis. Cell Struct. Func.,
11: 109-114.

Tashiro, K., Shiokawa, K., Yamana K. and Sakaki,
Y. (1986) Structual analysis of ribosomal DNA
homologues in nucleolus-less mutant of Xenopus
laevis. Gene 44: 299-306.

Dumont, J. N. (1972) Oogenesis in Xenopus laevis
(Daudin) 1. Stages of oocyte development in labora-
tory maintained animals. J. Morph., 136: 153-180.
Tashiro, K., Misumi, Y., Shiokawa, K. and Yama-
na, K. (1983) Determination of the rate of rRNA
synthesis in Xenopus laevis triploid embryos pro-
ducted by low-temperature treatment. J. Exp.
Zool., 225: 489-495.

Shiokawa, K. and Yamana, K. (1976) Pattern of
RNA synthesis in isolated cells of Xenopus laevis
embryos. Dev. Biol., 16: 368-388.

Nieuwkoop, P. D. and Faber, J. (1956) Normal
Table of Xenopus laevis Daudin. Norht-Holland
Publ. Co.

Berg, C. A. and Gall, J. G. (1986) Microinjected
Tetrahymena rDNA ends are not recognized as
telomers in Xenopus eggs. J. Cell Biol., 103: 691-
698.

Etkin, L. D. and Balcells, S. (1985) Transformed
Xenopus embryos as a transient expression system
to analyze gene expression at the midblastula transi-
tion. Dev. Biol., 108: 173-178.

Wickens, M. P., Woo, S., O'Malley, B. W. and
Gurdon, J. B. (1980) Expression of a chicken
chromosomal ovalbumin gene injected into frog
oocyte nuclei. Nature, 285: 628-634.

Atsuchi, Y., Tashiro, K., Yamana, K. and Shioka-

18

200

wa, K. (1986) Level of histone H4 mRNA in
Xenopus laevis embryonic cells cultured in the abs-
ence of cell adhesion. J. Embryol. exp. Morph., 98:
175-185.

Melton, D. A., Krieg, P. A., Rebagliati, M. R.,
Maniatis, T., Zinn, K. and Green M. R. (1984)

19

Y. Fu, K. Sato et al.

Efficient in vitro synthesis of biologically active
RNA and RNA hybridization probes from plasmids
containing a bacteriophage SP6 promoter. Nucl.
Acids Res., 12: 7035-7056.

Berk, A. J. (1986) Adenovirus promoters and Ela
transactivation. Ann. Rev. Genet., 20: 45-79.

ZOOLOGICAL SCIENCE 7: 201-208 (1990)

© 1990 Zoological Society of Japan

Ultrastructural Studies of the Carotid Labyrinth
in the Newt Cynops pyrrhogaster

TATSUMI KUSAKABE

Department of Anatomy, Yokohama City University School
of Medicine, Yokohama 236, Japan

ABSTRACT—Fluorescence, light, and electron microscopic observations of the carotid labyrinth of the
newt Cynops pyrrhogaster showed the presence of glomus cells in the intervascular stroma of the

labyrinth.

These cells contain many dense-cored vesicles (60 nm—100nm in diameter) and their

ultrastructure was similar to that of glomus cells reported in many other animal species. In some glomus
cells, there were intranuclear inclusion bodies (0.1—-0.3 ~m in diameter), and long processes were in
close contact with endothelial cells or with pericytes, which have not been reported in the carotid
labyrinth so far. Two types of synapses (efferent and afferent) were found on the glomus cell surfaces.
So-called reciprocal synapses were also found. On the basis of these findings, I conclude that the newt
glomus cells may have a secretory as well as a chemoreceptor function.

INTRODUCTION

Amphibian carotid labyrinths have been mor-
phologically studied by many workers: (see
Adams)[1]. Ishii et al. [2] confirmed in physiolo-
gical experiments that they have an arterial che-
moreceptor function analogous to that of the
mammalian carotid body. The anuran carotid
labyrinth contains glomus cells similar to those in
mammalian carotid bodies and is innervated by
glossopharyngeal and sympathetic nerves [3, 4]. It
has been thought that the glomus cell functions as
an element of the arterial chemoreceptor. In
electron microscopy of the Xenopus carotid labyr-
inth, close contact of glomus cells with smooth
muscle cells was reported, suggesting that the
labyrinth might have another function other than
chemoreception [5]. Most of the ultrastructural
and physiological studies of the carotid labyrinth
have been performed on anurans, and there are
only a few morphological reports on the carotid
labyrinth in urodelans [6, 7]. In the present study,
some ultrastructural characteristics of the newt
glomus cell which are different from those pre-

Accepted July 2, 1989
Received March 31, 1988

viously described in other amphibia are reported.

MATERIALS AND METHODS

Fifteen Japanese newts, Cynops pyrrhogaster, of
both sexes weighing 8-10 g were used. The anim-
als were anesthetized with urethane (5 mg/g body
weight), and the region of the carotid labyrinth
was exposed on both sides. Through a thin nylon
tube inserted into the aortic trunk, the labyrinth
was washed with Ringer solution and perfused
with fixatives, then removed from the body. For
fluorescence microscopy, specimens were im-
mersed in Grillo’s fixative [8] at 4°C for 20 hr.
They were sectioned transversely at 60 um.
Observations were made with a fluorescence
microscope (Olympus BHF) equipped with a HBO
200 high pressure mercury lamp, UG-5 and BG
excitor filters, and a Y-475 barrier filter. For
electron microscopy, the labyrinth was immersed
in 2.5% glutaraldehyde in 0.1 M cacodylate buffer
(pH 7.3) at 4°C for 3hr. After being washed in
buffer solution, the specimens were divided into
small blocks and postfixed in 1% osmium tetroxide
buffered with 0.1M cacodylate for lhr. After
dehydration in a graded ethanol series, the speci-
mens were embedded in Epon-alardite mixture.
Sections of 1 ~m were stained with toluidine blue

202 T. KUSAKABE

for light microscopy. Ultrathin sections were cut
serially and double-stained with saturated uranyl
acetate and Reynold’s lead solution. electron
micrographs were taken with a JEM 200-CX elec-
tron microscope at 80 kv.

RESULTS

Light and fluorescence microscopy

The carotid labyrinth had a maze-like structure
consisting of the sinusoidal plexus and intervascu-
lar stroma. The stroma contained 6 main types of
cells: glomus cells, smooth muscle cells, pericytes,
fibroblasts, mast cells, and endothelial cells. The

Fic. 1. Light micrograph of a toluidine blue-stained section of the carotid labyrinth. A glomus cell (arrow) is located
near the lumen of a sinusoid (L). E, endothelial cells; Sm, smooth muscle cells. Scale bar, 20 «m.
Fic. 2. Fluorescence micrograph of the carotid labyrinth. Two greenish-yellow fluorescent cells (arrows) with long

processes (arrowheads). L, lumen of sinusoid. Scale bar, 50 um.

Ultrastructure of Newt Carotid Labyrinth 203

Fic. 3. A) Electron micrograph in low magnification of the glomus cell. The glomus cell (the boxed area in Figure 1)
contains numerous cytoplasmic granules. Cp, cytoplasmic projection. Inb, intranuclear inclusion bodies: L,
lumen of sinusoid; Nf, nerve fiber. Scale bar, 5 4m. B) High magnification of the intranuclear inclusion bodies.
Between bodies small dense materials lie in a uniform array at regular intervals (arrowheads). Scale bar, 0.2 pm.

204 T. KUSAKABE

glomus cells had pale cytoplasm and a large oval
nucleus, and were located singly or in clusters (S0-
60 ~m in diameter) of 3-4 cells between connec-
tive tissues and smooth muscle cells (Fig. 1). In
fluorescence microscopy, the glomus cells were
identified as small, intensely fluorescent (greenish
yellow) cells with long processes (Fig. 2).

Electron microscopy

The sinusoidal wall was composed of extremely
thin endothelial cells. In the subendothelial stro-
ma, several cellular elements and extracellular
components such as collagenous fibers and amor-
phous ground substance were located. Enveloped
by thin processes of supporting cells, the glomus
cells were located in the vascular stroma singly or
in clusters (Fig. 3A). Occasionally, a part of cell
surface was directly in contact with the connective

tissue, losing its covering of supporting cells. Be-
tween the clustered cells desmosome-like junctions
were often observed. The cells were usually oval.
A single oval nucleus contained dispersed clumps
of heterochromatin. Near the nucleolus, a group
of electron-dense inclusion bodies of various sizes
(0.1-0.3 um in dimeter) was sometimes observed
(Fig. 3A). At high magnification, many small
dense particles were attached to the surface of
intranuclear inclusion bodies (Fig. 3B).
Numerous dense-cored vesicles (60-100 nm in
diameter) were scattered throughout the cyto-
plasm. Coated pits were often found on the
surface of the cell body and its processes (Fig. 4).
A well-developed Golgi complex was found in the
perinuclear area, and in its vesicles small newly
synthesized low-density granules were seen. A
relatively small amount of granulated endoplasmic

Fic. 4. Two membrane invaginations of the cytoplasmic processes (CP) of the glomus cell. Extruded content has
almost dissolved. Arrows indicate membrane invaginations. Dev, dense-cored vesicles. Scale bar, 0.2 ~m.

Fic. 5.

Residual bodies (arrows) in the glomus cell. Myelinated inclusions, small vesicles and a small vacuole can be

seen. Dev, dense-cored vesicles; Mt, mitochondria. Ne, nerve ending. Scale bar, 0.5 «m

Ultrastructure of Newt Carotid Labyrinth 205

reticulum, numerous free ribosomes and oval elon-
gated mitrochondria were dispersed throughout
the cytoplasm. Centrioles were located near the
nucleus. Large residual bodies, a kind of lyso-
some, were also observed. They were round and
about 0.7 ~m in diameter with myelinated inclu-
sions and small vesicles (Fig.5). Losing their
covering of supporting cells, long thin cytoplasmic
projections of the glomus cells extended toward
the endothelium. Some of them were longer than
20 um and closely associated with the endothelial
cells (g-e connection) (Fig. 6A) or with the peri-
cytes (g-p connection) (Fig. 6B) without visible

intervening substances.

Enclosed by supporting cells, nerve endings lay
close to the glomus cells to make synapses. Two
types of synapses were distinguished on the glomus
cell. Most of the large synapses were morphologi-
cally efferent and were characterized by the accu-
mulation of clear vesicles (40-60 nm in diameter).
At the junction the membrane of the nerve ending
was thicker than the cell membrane (Fig. 7A).
Small afferent type synapses were sometimes
observed; specialized membrane thickenings were
conspicuous on the cell membrane where dense-
cored vesicles were aggregated (Fig. 7A). So-

Fic. 6. A) Three cytoplasmic processes extend into the intervascular stroma. The inner two processes lose the
covering of supporting cell cytoplasm (S), and luminal one makes g-e connection (arrow). Cf, collagenous fiber;
Dev, dense-cored vesicles; E, endothelial cell; L, lumen of sinusoid; Mt, mitochondria; Pv, plasmalemmal
vesicles. Scale bar, 1 zm. B) A pericyte (Pc) directly connect with the process of a glomus cell (arrow). E,
endothelial cell; f, filaments; L, lumen of sinusoid. Scale bar, 1 ~m.

206 T. KUSAKABE

Fic. 7. A) Synaptic junctions between the glomus cell and nerve ending. The arrow and arrowheads indicate
efferent and afferent type synapses, respectively. One nerve ending (Ne-1) which contains numerous clear
vesicles and some mitochondria synapses with the glomus cell (Gc). The synapse shows asymmetrical membrane
thickning on the Ne-1 side. Many synaptic vesicles accumulate at this junctional region (arrow). Another nerve
ending (Ne-2) without clear veiscles synapses with the same glomus cell (Gc). Asymmetrical membrane
thickenings are found on the glomus cell side. A few dense-cored vesicles aggregate at this synaptic region
(arrowheads). Scale bar, 0.2 ~m. B) So-called reciprocal synapses between the glomus cell and nerve ending.
The arrow and arrowhead show efferent and afferent type synapses, respectively. A and B are semi serial
sections. Scale bar, 0.2 «m.

called reciprocal synapses consisting of efferent
and afferent type synapses on the same nerve
ending were also observed (Fig. 7B). The carotid labyrinth of the urodeles consists of

a vascular maze [1, 9] and contains cells with many

DISCUSSION

Ultrastructure of Newt Carotid Labyrinth 207

dense-cored vesicles [6]. The present electron
microscopic study in the newt Cynops pyrrhogaster
showed that these cells (glomus cells) are situated
in the vascular stroma as isolated cells or in clusters
of 3 or 4 cells, and their ultrastructure is similar to
that of glomus cells observed in arterial che-
moreceptor regions of anurans [4, 10]. In the toad
Bufo vulgaris, the carotid labyrinth has been phy-
siologically confirmed to have an arterial che-
moreceptor function similar to that of the mamma-
lian carotid body [2]. The glomus cell has long
been known to be a chemoreceptor, but in recent
years it has been postulated to have a secretory
function as well [5, 11]. Although there is no direct
evidence, the structural characteristics of the newt
carotid labyrinth point to a chemoreceptor func-
tion, in which the glomus cells play the main role.

In the present paper, I have reported for the first
time the presence of intranuclear inclusion bodies
in the glomus cell nuclexs. Intranuclear inclusion
bodies have been observed only in clearly defined
secretory cells: in the intestine [12] and in the
colon [13] of the horse, in the epididymis of the
dog [14], and in the adenohypohysis of the rabbit
[15]. Although their origin and role are as yet
unknown, they may be considered a characteristic
feature of the secretory function. This finding may
indicate a secretory function of the glomus cell as
stated above, or it may show that intranuclear
inclusion bodies exist in more extensive cel! do-
mains than so far accepted.

In addition, the g-e connection was observed for
the first time in amphibian glomus cells. Kondo
[16] and Ookawara et al. [17] observed close
contact of granule-containing cells with endothelial
cells in the domestic fowl, and Kondo [16] prop-
osed a secretory function for the granule-
containing cells there. Since the vesicles in the
endothelial cells are commonly accepted to in-
volved in intracellular transport [18, 19], catecho-
lamines contained in dense-cored vesicles might be
released into the blood vessels. However these
might also be structures facilitating blood gas
uptake by glomus cells for chemoreception. On
the other hand, the g-p connection suggests that
the vascular tone might be modified by catechola-
mine in glomus cell granules, if catecholamine
releasing occurs in this region.

Two types of synapses, efferent and afferent,
were reported in toad carotid labyrinth [4] and in
tortoise carotid artery [20]. Also in this study,
afferent and efferent synapses were observed be-
tween the nerve endings and glomus cells as de-
scribed by Kobayashi [6]. Furthermore Yamauchi
[21] reported reciprocal synapses in the toad caro-
tid labyrinth. Serial sections suggested that all
nerve endings may be reciprocal in the newt caro-
tid labyrinth.

In conclusion, the newt glomus cells may have a
secretory function in addition to their chemorecep-
tor function.

ACKNOWLEDGMENTS

I wish to thank Prof. M. Asashima of the Department
of Biology for kindly supplying animals, Dr. R. C. Goris
of Department of Anatomy, Yokohama City University
for his help in preparing the manuscript and Drs. Kosei
Ishii and Kazuko Ishii for their critical reading of the
manuscript. This work was supported by Grants-in-Aid
(No. 63770024) from the Ministry of Education, Science
and Culture, Japan.

REFERENCES

1 Adams, W. E. (1958) The comparative morphology
of the carotid body and carotid sinus. Charles C.
Thomas, Springfield, pp. 202-214.

2 Ishii, K., Honda, K. and Ishii, K. (1966) The
function of the carotid labyrinth in the toad. Tohoku
J. exp. Med., 88: 103-116.

3 Rogers, D. C. (1963) Distinct cell types in the
carotid labyrinth. Nature, 200: 492-493.

4 Ishii, K. and Oosaki, T. (1969) Fine structure of the
chemoreceptor cell in the amphibian carotid labyr-
inth. J. Anat. (Lond), 104: 263-280.

5 Ishii, K. and Kusakabe, T. (1982) The glomus cell
of the carotid labyrinth of Xenopus laevis. Cell
Tissue Res., 224: 459-463.

6 Kobayashi, S. (1971) Comparative cytological stu-
dies of the carotid body. 2. Ultrastructure of the
synapse on the chief cell. Arch. histol. jap. , 33: 397-
420.

7 Wislang, M. R. (1965) The carotid labyrinth in two
species of urodeles. J. Anat. (Lond), 99: 949.

8 Grillo, M. A., Jacobs, L. and Comroe, Jr. J. H.
(1974) A combined fluorescence histochemical and
electron microscopic method for studying special
monoaminecontaining cells (SIF cells). J. Comp.
Neurol., 153: 1-14.

9 Ishida, S. (1954) So-called carotic body of the

10

11

208

amphibia. Igaku Kenkyu (Fukuoka), 24: 1024-1050.
Ishii, K. Ishii, K. and Kusakabe, T. (1985) Chemo-
and baroreceptor innervation of the aortic trunk of
the toad Bufo vulgaris. Respir. Physiol., 60: 365-
375.

Kusakabe, T., Ishii, K. and Ishii, K. (1987) A
possible role of the glomus cell in controlling vascu-
lar tone of the carotid labyrinth of Xenopus laevis.
Tohoku J. exp. Med., 151: 395-408.

Dual, D. E. (1980) The origin of nuclear bodies: A
study of the undifferentiated epithelial cells of the
equine small intestine. Am. J. Anat., 157: 61-70.
Pfeiffer, C. J., Marry, M. J. and Legends, L. (1987)
The equine colonic mucosal granular cell: Identifica-
tion and X-ray microanalysis of apical granules and
nuclear bodies. Anat. Rec., 219: 258-267.
Nicander, L. (1964) Fine structure and cytochemis-
try of nuclear inclusions in the dog epididymis. Exp.
Cell Res., 34: 533-541.

Foster, C. L., Young, B. A., Allanson, M. and
Cameron, E. (1965) Nuclear inclusions in the ade-
nohypophysis of the rabbit. J. Endocrinol., 33: 159-
160.

T. KUSAKABE

16

17

18

19

20

21

Kondo, H. (1974) On the granule-containing cells in
the aortic wall of the young chick. Anat. Rec., 178:
253-266.

Ookawara, S., Suzuki K., Yoshida, Y. and Ooneda,
G. (1974) Monoamine-storing cells in the media of
the thoracic aorta of Gullus domesticus. Cell Tiss.
Res., 151: 309-316.

Palade, G. E. (1953) Fine structure of blood capil-
laries. J. appl. Physics, 24: 1424.

Simionescu, N., Simionescu, M. and Palade, G. E.
(1975) Permeability of muscle capillaries to small
heme-peptides. Evidence for the existence of patent
transendothelial channels. J. Cell Biol., 64: 586—
607.

Kusakabe, T., Ishii, K. and Ishii, K. (1988) Dense
granule-containing cells in the arterial chemorecep-
tor area of the tortoise (Testudo hermanni). J.
Morphol., 197: 183-191.

Yamauchi, A. (1977) On the recepto-endocrine
property of granule-containing (GC) cells in the
autonomic nervous system. Arch. histol. jap., 40,
Suppl: 147-161.

ZOOLOGICAL SCIENCE 7: 209-215 (1990)

Abnormal Development of Preimplantation Embryos Derived
from Intersubspecific Hybrids between Mus musculus
molossinus and M. m. domesticus

MicHIko NrwA and NosBoru WAKASUGI

Laboratory of Animal Genetics, Faculty of Agriculture,
Nagoya University, Chikusa-ku, Nagoya 464, Japan

ABSTRACT—Japanese wild mice, Mus musculus molossinus, are genetically remote from laboratry
mice which are derived predominantry from European wild mice M. m. domesticus. Our previous study
demonstrated that F2 progeny between these two mouse subspecies show low ferility, even though F1
hybrids are fully fertile. In fact about half of the F2 females fail to become pregnant. We examined the
in vitro development of preimplantation embryos from F2 progeny between MOM (one of the inbred
strains derived from Japanese wild mice) and C57BL/6 (B6, an inbred strain of laboratory mice). We
found that the low pregnancy rate of F2 females results from a high embryonic mortality: only 58.4%
(66/113) of embryos developed to the blastocyst stage in the F2 x F2 cross, whereas in the B6 x B6 cross
the corresponding figure was 100.0% (23/23). The mortality was not due to defects in sperm from F2
males but rather to defects in the eggs from F2 females: the survival rate of embryos up to the blastocyst
stage was 52.2% (82/157) in the F2 2 xB6¢@ cross, whereas it was 94.9% (74/78) in the BO? XF2¢
cross. The factors responsible for this mortality are attributable to nucleus, not to maternally inherited
cytoplasm: more than 10% of N2 femals derived from backcrossing either F1 femals to B6 males or F1
males to B6 females showed the higher embryonic mortality. This finding suggests that intersubspecies
genic combinations, either intragenic or intergenic, give rise to some deleterious effects on the oocytes

© 1990 Zoological Society of Japan

during oogenesis.

INTRODUCTION

The species Mus musculus can be classified into
several groups (subspecies) from a genetic pers-
pective [1, 2]. It has been demonstrated that
common laboratory mice originate predominantly
from a European subspecies, Mus musculus
domesticus, by the analysis of mitochondrial and
nuclear genomes [3, 4]. Japanese wild mice, M. m.
molossinus, are very different from M. m. domesti-
cus and are closely related to M. m. musculus and
M. m. castaneus [2, 3]. The evolutionary diverg-
ence of M. m. molossinus from M. m. domesticus
is estimated to have oocurred about one million
years ago [3, 5, 6].

There appears to exist a severe restriction of the
gene flow in the contact zone between M. m.
domesticus and M. m. musculus in Europe [7]. It

Accepted July 6, 1989
Received March 20, 1989

can be inferred, therefore, that there is an incom-
patibility between the genomes of the two subspe-
Male sterility is suspected of being one
manifestation of the incompatibility that is re-
sponsible for restriction of the gene flow in this
area [7, 8].

Several inbred strains derived from Japanese
wild mice, Mus musculus molossinus, have been
established in our laboratory to provide us with a
wide variety of experimental material. When we
attempted to generate recombinant inbred strains
from F1 hybrids between molossinus and labora-
tory strains, a serious depression in fertility was
observed in the subsequent generations, despite
the fact that both parental strains had been bred
for more than 20 generations by brother-sister
matings. So far, no evidence has been reported
that there is a serious reproductive disturbance in
the F2 females derived from inter-strain crosses
among laboratory strains of inbred mice. We
previously reported details of the reproductive

cies.

210 M. NIEA AND N. WAKASUGI

preformance of Fl and F2 generations from cros-
ses between molossinus and laboratory strains [9].
Such F1 hybrids are fully fertile, but in the F2
generation, about half of the F2 females fail to
become pregnat. Since the F2 progeny copulated
normally, it appeared possible that preimplanta-
tion loss of embryos may be responsible for the
infertility observed in half of the F2 females.

In the present study, we examined the develop-
ment in vitro of preimplantation embryos obtained
from F2 and backross (N2) generations between
molossinus and loboratory mice.

MATERIALS AND METHODS

Animals

CS7BL/6 (B6) and MOM strains were used. The
MOM strain is derived from Japanese wild mice
(Mus musculus molossinus), whose ancestors were
captured at Nagoya in Japan. The number of
inbreeding generatons was 46 at the time when the
present study was undertaken. Fl and F2 genera-
tions were produced through reciprocal matings
between B6 and MOM. Since there were no
differences in the fertility between members of the
F2 generations derived from (B62 X MOM @ ) F1
and from (MOM ? XB6¢@) F1, they were pooled
in the present study. Backcross progeny (N2) were
produced from the crossing of either Fl ? x B64
or B6? XF1¢@.

TABLE 1.
and MOM strains and B6 females

Observation of embryos

Two- to twelve-month-old females were mated
with males and checked daily for vaginal plugs.
The day on which a plug was found to be present
was designated as Day 0 of pregnancy. The
embryos were recovered by flushing oviducts with
Medium 2 (M2) [10] from plug-positive females on
Day 1 or Day 2, and they were examined with
reference to developmental stage and morphology
under a dissecting microscope. Subsequently, the
embryos were cultured in _ pre-equilibrated
Medium 16 (M16) [11] under paraffin oil in an
atmosphere of 5% CO, in air at 37°C. Embryos
were examined at intervals of 24 hr and the num-
ber of embryos that developed to the blastocyst
stage was recorded.

RESULTS

Our of a total of 21 embryos obtained from three
MOM females mated with MOM males, 18
(85.7%) developed into expanded blastocysts.
Most embryos from Fl females which had been
mated with B6, MOM, or F1 males also developed
to the blastocyst stage: out of 103 morphologically
normal embryos collected from 13 females, 95
(92.2%) developed into blastocysts. In these cros-
ses, embryos were obtained on Day 1 and Day 2.
As summarized in Table 1, ovulation and feritiliza-
tion occurred normally in the F2 females.

All fertilized eggs were cultured in vitro and the
number of embryos that developed to the blasto-

Developmental ability of embryos obtained on Day 1 of pregnancy from F2 females between B6

LS

No. of eggs No. of embryos that
ria Me fobs collected No: uch ne ual developed to blastocysts
(mean+sem) ryOS during 4 days in culture
F2x F2 16 114 113 ( 99.1%) 66 ( 58.4)”
(71.+0.5)
F2 x B6 24 166 157 ( 94.6) 82 ( 52.2)
(6.9+0.4)
B6 x F2 9 80 78 ( 97.5) 74 (94.9)
(8.9+0.5)
B6 x B6 3 23 23 (100.0) 23 (100.0)
(7.7+0.3)

a
” Morphologically normal 2- to 6-cell embryos were counted.
t .
» Percentage was computed from the number of normal embryos.

Abnormal Development of Hybrid Embryos 211

Fic. 1. Morphological appearance after culture i in vitro of embryos from F2 ? X B6 cross, including degenerated or
retarded embryos. Embryos were collected on Day 1 of pregnancy and photographed after 72 hr in culture.
Arrows indicate the abnormal embryos that are developmentally retarded or have degenerated.

212 M. NigA AND N. WAKASUGI
TABLE 2. Proportion of normal embryos on Day 2 of pregnancy from F2 and B6 females mated with B6
males, and their developmental ability in culture
No. of No. of embryos that
Grosses NoveE ae i embryos Ne. sh nena) developed to blastocysts
obtained” y during 3 days in culture
F2 x B6 22 161 118 ( 73.3%) 95 (59.0%)
B6 x B6 ! 53 53 (100.0) 50 (94.3)
*) Embryos, and not unfertilized eggs, were counted.
>) Morphologically normal embryos at the 6-cell to the morula stage were counted.
2 No. of blastocysts/No. of embryos obtained.
(a)F2?x F2< jo} (c)BEe x Fae
5 N=9
8 2
© ow
E E
& o
— if 5
Oo xe)
ro) .
Z 2
0
Survival rate(%) of Survival rate(%) of embryos
jot (bIF2° x B6~ 10 (d)/B6e x B6~
N=10
n o)
® &
w o
= E
25 25 a
ro i Ke) ee
o A o 7 N
Zz a z
0 70 100
Survival rate(%) of embryos Survival rate(%) of embryos
(e)N2° x B6~
E(Fie x B6o~ )N2e
9 107 BRR(B6 2 x Fix)N2e
N=
E 28 Fic. 2. Distribution of females according to the
o rate of survival of their embryos*. (a) F2?
me XF22, (b) F22 XB6%, (c) B6? XF24,
Oo (d) B6& XB6%, (e) N2¢-**xB6e.
fe) *Number of embryos that developed to the
z blastocyst stage/total number of embryos

50
Survival rate(%) of embryos

100

obtained.
*“*Backcross generation from the Fl 29 x B6S
and B62 XF1 @ crosses.

Abnormal Development of Hybrid Embryos 213

cyst stage was counted. In the crosses of F2 x F2
and F2’ xB6¢@, the proportion of embryos that
developed to the blastocyst stage, including unex-
panded embryos, was very small (Table 1). Many
embryos from F2 females exhibited morphological
alterations or developmental arrest at various
stages from the 2- to the 6-cell stage as far as
blastulation. Figure 1 shows embryos from F2
females during culture. They were all morphologi-
cally normal, 2- to 4-cell embryos when collected
on Day 1. In the case of the B6 ? XF2¢ and B6 x
B6 crosses, almost all embryos developed into
normal, expanded blastocysts. These results sug-
gest that the abnormality in development can be
attributed to some disturbance in the eggs from F2
females.

Next we examined the embryos on Day 2 of
pregnancy from F2 and B6 frmales mated with B6
males. As shown in Table 2, more than 70% of
embryos collected from F2 ? x B6@ crosses were
morphologically normal at the 6-cell to the morula
stage while the others were abnormal or had
degenerated. During culture, about 60% of
embryos survived to form blastocysts. This result
confirmed the findings shown in Table 1: about
half of the embryos from F2 females died around
the morula stage both in vivo and in vitro. In the
control cross, B6B6, morphologically normal
embryos were obtained and most of them de-
veloped to form expanded blastocysts.

Figure 2 shows the distribution of females
according to the proportion of embryos that de-
veloped to the blastocyst stage. The results of
F22 xF2@ and F2? xB6% crosses (Fig. 2a, b)
demonstrated that there are two types of female:
one showing a higher rate of survival of embryos
and the other showing a lower rate. In the crosses
of B62 XF2 and B6? <XB6%@, the rate of sur-
vival of embryos that developed to the blastocyst
stage for each female was greater than 70% (Fig.
2c, d). This result demonstrates that the F2
sperm does not negatively influence the develop-
mental abnormality of the embryos. Figure 2e
shows the distribution of the survival rates of
embryos obtained from N2 females, progeny of the
Fl? xB62 and B62 XF1% crosses, when they
were mated with B6 males. The mean number of
embryos obtained on Day 2 was 8.2+0.3. Ex-

tremely poor rates of survival of embryos, lower
than 20%, were observed in the case of 5 N2
females out of 38 (13.2%); 3 of these females being
from the B62? XF1% cross, and 2 from the Fl x
B6% cross. The higher rates of embryonic mortal-
ity observed in the case of these N2 females
demonstrate that the abnormality is inherited from
both F1 females and F1 males.

DISCUSSION

In order to investigate the infertility of about
half of the F2 females between the two subspecies
Mus musculus domesticus and M. m. molossinus,
we examined the preimplantation development of
their embryos. Our data indicate that a higher
frequency of embryonic mortality during preim-
plantation development is responsible for the in-
fertility observed in the F2 females. We confirmed
our previous observation that there is segregation
of fertile and infertile F2 females. Some F2
females showed higher rates of survival of their
embryos and others showed lower rates. This
embryonic death can be attributed to defects in the
eggs, not in the sperm, from the F2 generation. F2
males are fertile when they mate with B6 females
[9]. The factors that cause the embryonic death
which occurs around the morula stage are inher-
ited from both F1 males and F1 females (Fig. 2e).
It has been reported that embryonic genes are
transcribed and expressed soon after fertilization
[12, 13]. Such a conclusion implies the possible
switching of the information for development from
maternal to embryonic [12]. The early expression
of embryonic genes is necessary for compaction of
8-cell mouse embryos [13]. Therefore, the mortal-
ity at the early developmental stage observed in
our study may possibly be due to some abnormal-
ity in the embryonic genome and not in the mater-
nally inherited cytoplasm which functions in the
early stages of development. In addition, the
factors that cause this abnormality appear to be
autosomal. Some of the N2 females that showed
lower rates of suvival of embryos were derived
from the B6 . x (B6? x MOM ¢ )F1 cross, hav-
ing two B6 X-chromosomes.

The reproductive abnormality in the F2 genera-
tion observed in this study is presumably related to

214

a genetic incompatibility that arises from the com-
bination of genes (chromosomes) from different
subspecies. Such an incompatibility is observed in
the contact zone between Mus musculus domesti-
cus and M. m. musculus in Europe, where a
restriction of the gene flow is observed [7, 14, 15].
Hybrid male sterility has been considered to be
responsible for this restriction. Recently Vanler-
berghe et al. [15] reported that the limitation of
Y-chromosome introgression in the contact zone is
not attributable to the sterility in Fl males between
the two subspecies. The developmental abnorma-
lities in the preimplantation embryos from F2
females of the intersubspecific crosses observed in
this study provide one possible explanation. From
an analysis of restriction patterns of mitochondrial
DNA, it has been shown that M. m. molossinus is
related to M. m. musculus and to M. m. castaneus,
both of which are quite remote from M. m.
domesticus [2, 4, 5, 16]. It has been proposed that
M. m. molossinus originated from the hybridiza-
tion of M. m. musculus and M. m. castaneus [17].
If this was indeed the case, it can be inferred that
hybrids between domesticus and musculus in the
contact zone in Europe could show reproductive
defects in the same way as the intersubspecific
hybrids described in this report. At investigation
of this interpretation should prove interesting.

ACKNOWLEDGMENTS

We thank Drs. K. Kondo, T. Tomita, T. Namikawa,
Y. Kawamoto, and other members of our laboratory for
their helpful discussions. We also thank Dr. K. Moriwa-
ki for his critical reading of the manuscript. We are also
grateful to Ms. S. Narita for her help in taking care of our
experimental animals.

REFERENCES

1 Bonhomme, F., Catalan, J., Britton-Davidian, J.,
Chapman, V. M., Moriwaki, K., Nevo, E. and
Thaler, L. (1984) Biochemical diversity and evolu-
tion in the genus Mus. Biochem. Genet., 22: 275-
303.

2 Moriwaki, K., Miyashita, N., Suzuki, H., Kurihara,
Y. and Yonekawa, H. (1986) Genetic features of
major geographical isolates of Mus musculus. In
“Current Topics in Microbiology and Immunology,
Vol. 127”. Ed. by M. Potter, J. H. Nadeau and M.
P. Cancro, Springer-Verlag, Berlin Heidelberg, pp.

10

12

M. NIEA AND N. WAKASUGI

55-61.

Moriwaki, K., Shiroishi, T., Minezawa, M., Aotsu-
ka, T. and Kondo, K. (1979) Frequency distribution
of histocompatibility-2 antigenic specificities in the
Japanese wild mouse genetically remote from the
European subspecies. J. immunogenet., 6, 99-113.
Yonekawa, H., Moriwaki, K., Gotoh, O., Miyashi-
ta, N., Migita, S., Bonhomme, F., Hjorth, J. P.,
Petras, M. L. and Tagashira, Y. (1982) Origins of
laboratory mice deduced from restriction patterns of
mitochondrial DNA. Differentiation 22, 222-226.
Ferris, S. D., Sage, R. D., Prager, E. M., Ritte, U.
and Wilson, A. C. (1983) Mitochondrial DNA
evolution in mice. Genetics, 105, 681-721.
Yonekawa, H., Moriwaki, K., Gotoh, O., Hayashi,
J.-I., Watanabe, J., Miyashita, N., Petras, M. L.
and Tagashira, Y. (1981) Evolutionary relationships
among five subspecies of Mus musculus based on
restriction enzyme cleavage patterns of mitochond-
rial DNA. Genetics, 98: 801-816.

Sage, R. D., Whitney, J. B. and Wilson, A. C.
(1986) Genetic analysis of a hybrid zone between
domesticus and musculus mice (Mus musculus com-
plex): hemoglobin polymorphisms. In “Current
Topics in Microbiology and Immunology, Vol. 127”.
Ed. by M. Potter, J. H. Nadeau and M. P. Cancro,
Springer-Verlag, Berlin Heidelberg, pp. 75-85.
Forejt, J. and Ivanyi, P. (1975) Genetic studies on
male sterility of hybrids between laboratory and wild
mice. Genet. Res., 24: 189-206.

Niwa, M. and Wakasugi, N. (1988) Fertility in
intersubspecific hybrids of laboratory mice (Mus
musculus domesticus) and molossinus mice (M. m.
molossinus). Exp. Anim., 37: 387-392.

Quinn, P., Barros, C. and Whittingham, D. G.
(1982) Preservation of hamster to assay the fertiliz-
ing capacity of human spermatozoa, J. Reprod.

. Fert., 66: 161-168.

Whittingham, D. G. (1971) Culture of mouse ova.
J. Reprod. Fert. (suppl.), 14: 7-21.

Petzoldt, U. and Muggleton-Harris, A. (1987) The
effect of the nucleocytoplasmic ratio on protein
synthesis and expression of a stage-specific antigen
in early cleaving mouse embryos. Development, 99:
481-491.

Smith, R. K. W. and Johnson, M. H. (1985) DNA
replication and compaction in the cleaving embryo
of the mouse. J. Embryol. exp. Morph., 89: 133-
148.

Hunt, W. G. and Selander, R. K. (1973) Bioche-
mical genetics of hybridisation in European house
mice. Heredity, 31, 11-33.

Vanlerberghe, F., Dod, B., Boursot, P., Ballis, M.
and Bonhomme, F. (1986) Absence of Y-chromo-
some introgression across the hybrid zone between
Mus musculus domesticus and Mus musculus muscu-

16

Abnormal Development of Hybrid Embryos 215

lus. Genet. Res., 48: 191-197.

Moriwaki, K., Yonekawa, H., Gotoh, O., Mineza-
wa, M., Winking, H. and Gropp, A. (1984) Implica-
tions of the genetic divergence between European
wild mice with Robertsonian translocations from the
viewpoint of mitochondrial DNA. Genet. Res.,
43:277-287.

17 Yonekawa, H., Moriwaki, K., Gotoh, O., Miyashi-
ta, N., Matsushima, Y., Shi, L., Cho, W. S., Zhen,
X. L. and Tagashira, Y. (1988) Hybrid origin of
Japanese mice “Mus musculus molossinus”: evi-
dence from restriction analysis of mitochondrial
DNA. Mol. Biol. Evol., 5: 63-78.

oo wih cenehd a
Aveyiht Anon Fraeipereh sasha poet .
. eae LY aqnlliagnehll yadeaibe pa:
e tn asa Bieduld. (ate) ’ wile T bina,
PAG tes ‘ att, waht 72 9
r Kapa font) eet OCT Pahs Heya (aon :
ee oe | id tole a -

tt, 48%

Manwelth

n. ©, Migs ra, Oe
re A Pottme MOM. ect Th
: bien, dng enetnte erie
~ avo aot DMA. CBN

vi ones i. td

- GE mm Wied» <<

; / ai3 Der We Ww wite Oo

rs a. View SVL: Westiw gly
vite”. OS Wasendbe, 3, Miya }
So ‘eit Fopedblja 19 ag

ee
LipfPetiet canpbe COQ \
2 oa of Ch, eit Si

i doe By Wie 2

f . Ya “in Ones a i vi
b : oy Jmivaiiows ute coe per’ |
dave hh =. PUTIN id ote

fie oe Micah crigy anus,

i “it hy we Panda t i Mas

f ; - “prrugre 2 4 Alag, Heri Hei
} ¥ i wad heey, FE (per

oat

sige dul by torpete te
eos 34
ee ad Aes Sorat)
= a wiiecsiae > Muha a4

reed uk ain! fen

; sisarte LES ys Lum 47 Sa
7) ays
hve *
F * ie ?
‘ f
7
er i
ed
-
Ps
> *
ai p
atte any
Miu
‘va
.s Me
i

wi Accept
ee al
rte

ZOOLOGICAL SCIENCE 7: 217-222 (1990)

© 1990 Zoological Society of Japan

Evidence for a Thy-1-Like Molecule Expressed
on Earthworm Leucocytes

ABDEL Hakim SAAp! and Epwin Cooper”

"Department of Zoology, Faculty of Science, Cairo University,
Cairo, Egypt, and *Department of Anatomy, University
of California, Los Angeles, CA, U.S.A.

ABSTRACT—In this study, we analyzed the presence of a Thy-1 homolog in the earthworm,
Lumbricus terrestris, using several monoclonal and xenoantisera in indirect immunofluorescence (IF)
assay. The reactivity of monoclonal antibodies Thy-1.1 and Thy-1.2 proved specificities to the Thy-1.1
determinant. A rabbit anti-rat Thy-1 (Thy-1.1) antiserum was further investigated in IF assay by in vitro

quantitative absorption.

In all assays, earthworm leucocytes inhibited reactivity of antiserum as

effectively as rat thymocytes in contrast to BALB/c (Thy-1.2) thymocytes. Thus, a Thy-1.1 cross-
reacting determinant is probably expressed by a Thy-1 homolog on leucocytes of earthworms. The
serological similarity between the earthworm Thy-1 homolog and the Thy-1 molecule in vertebrates will

be strengthened by future immunochemical data.

INTRODUCTION

Coelomic fluid of earthworms contains several
morphological categories of leucocytes which have
been shown to play prominent role in allogenetic
and xenogeneic graft rejection; the case of certain
leucocytes associated with class I antigen in mam-
mals [cf. in 1]. But until now, no attempts has
been made to elucidate the nature of leucocyte
membrane structures invovled in such phenomena.
Recently, Roch et al. [2] succeded in demonstrat-
ing serological evidence for a membrane structure
related to human /-microglobulin expressed by
certain earthworm leucocytes. In searching for the
origin of Thy-1, Shalev et al. [3] have demons-
trated Thy-1-like molecule in total extracts of
earthworms and several other invertebrates by
using radioimmunoassays. Their analysis did not
involve, however, a search for a Thy-1-like mole-
cule in association with certain earthworm leuco-
cyte membranes, which we have demonstrated in
the present study. The phylogeneic studies of a
molecule may contribute to increased understand-
ing of its function and significance. The function of

Accepted June 6, 1989
Received April 14, 1989

Thy-1 is still unknown but several lines of evidence
suggest that the molecule is involved in T cell
activation [4]. Recent reports strongly suggest that
Thy-1 shares amino acid homolog with the con-
stant and variable regions of immunoglobulins
(Ig), with beta-2-microglobulin and major histo-
compatibility complex (MHC) encoded antigens
[5, 6]. The hypothesis that either the Thy-1 or the
f>-microglobulin gene are representatives of the
ancestral gene must be supported by evidence of
high evolutionary conservation of these genes or
molecules [7]. Thus, it is of vital importance to
reveal more concerning the Thy-1 molecule in
invertabrates and to trace its evolutionary origin
and function.

In this study, we analyzed the presence of the
Thy-1 homolog on earthworms (Lumbricus terres-
tris) utilizing several monoclonal and xenoantisera
in indirect immnofluorescence assay. We observed
a strong cross-reaction in which both anti-rat Thy-
1.1 and anti-mouse Thy-2 monoclonal antibodies
detected a molecule, on the cell surface of ear-
thworm leucocytes suggesting the oocurrence of
cell surface-bound forms of a putative Thy-1
homolog at this level of evolution.

218 A-H. SAaD AND E. Cooper

MATERIALS AND METHODS

Animals

About 200 earthworms, Lumbricus terrestris
(Lumbricidae, Annelida) exhibiting secondary
sexual characteristics were purchased from Sure-
Live Meal Worm Co., Torrance, CA, and main-
tained at 12°C in natural soil in the laboratory.
Male and female, 5 to 7 weeks old, BALB/c mice,
Wistar rats and New Zealand white rebbits were
obtained from the United State Naval Medical
Research Unit No. 3, Cairo.

Cell suspensions

L. terrestris leucocytes were harvested by sub-
merging worms in 10% alcohol solution as de-
scribed previously [1]. After worms shed leuco-
cytes through integumentary pores at room
temperature (22°C) the leucocytes were decanted
into fresh Ca?*, Mg?* free buffered saline solu-
tion (BSS: 10 mM KH>PO,, 10 mM K,HPOx,, 0.11
M NaCl, 10 mM HEPES; pH 7.2) through a stain-
less steel screen to remove debris. Pools of leuco-
cytes from at least 30 worms were used, washed by
centrifugation at 450 xg for 5 min at 4°C, and their
viability assessed by trypan blue exclusion [8].
Suspensions of rat and BALB/c mouse thymocytes
were prepared by teasing the thymus in PBS, pH
7.2. After washing 3 times each for 5 min by
centrifugation at 600 x g, thymocytes were counted
and their viability assessed by trypan blue exclu-
sion.

Reagents

Rabbit anti-purified rat Thy-1 antiserum were
kindly provided by Dr. M. H. Mansour (Universi-
ty of California, Los Angeles). Previous findings
indicate that rabbit-anti-rat Thy-1 antiserum rec-
ognizes three antigenic determinants: the rat-
specific Thy-1 xenoantigen, the rat-mouse cross-
reacting xenoantigen, and the Thy-1.1 determinant
[9]. BALB/c mouse anti-rat Thy-1 (ascites fluid,
IgG MRCOX-7 anti-rat Thy-1.1 mAb) was purch-
ased from Accurate Chemical Scientific Corpora-
tion, Westbury, NY. BALB/c mouse anti-AKR/J
mouse thymocytes (ascites fluid, 7S IgG anti-
mouse Thy-1.1 mAb) and AKR/J mouse anti-C3H

thymocytes (19S IbM anti-mouse Thy-1.2 mAb)
were purchased from New England Nuclear, Bos-
ton, MA. Rabbit anti-BALB/c mouse brain serum
and rabbit anti-C3H mouse brain serum were
purchased from Bionetics Laboratory Products,
Kensington, MD and absorbed locally with crude
liver cell-membranes from BALB/c and C3H mice.
Normal rabbit and rat sera were collected in the
laboratory from unimmunized animals. Fluores-
cein isothiocyanate (FITC-) labelled goat Ig, anti-
rabbit globulins were purchased from Behring
Institue, Marburg, West Germany and FITC-
labelled rabbit anti-mouse Ig from GIBCO, Grand
Island, NY.

Indirect immunofluorescence (IF) assays

Analysis of different specificities in the anti-sera
against target earthworm leucocytes and/or rat and
BALB/c mouse thymocytes was measured in in-
direct immunofluorescence assay by quantitative
absorption. Preliminary experiments, performed
to define optimal labelling conditions for our sys-
tem, indicated that fluorescent antibodies must be
used at a dilution of 1 : 20 for the anti-mouse Ig and
1:25 for the anti-rabbit globulins. Apart from
appropriate changes in target cells, antisera and
anti-serum dilutions (details are further elaborated
in the Result Section), the standard assay involved
absorption of the first-step antibody (200 ul) with
increasing numbers of earthworm leucocytes and/
or rodent thymocytes for 16 hr at 4°C, followed by
incubation with 1-2 10° freshly-prepared target
earthworm leucocytes and/or rodent thymocytes
for 45min at 4°C. After 3-4 washings in PBS,
target cells were reincubated with 100 yl of fluores-
cent conjugate for an additonal 45 min at 4°C.
After 3-4 washings, cells were finally mounted on
microscope slides, scored alternately in phase con-
trast and fluorescence microscopy. Percentage of
positive labelled cells were determined by counting
a minimum of 200 cells. Positive as well as
negative controls, explained in the results, were
included in each expeirment.

Iodination of cell surface and indirect immunopre-
cipitation of Thy-1 homolog

Aliquots of 10’ viable earthworm leucocytes
were surface labelled with 300 ~Ci Na'™I by the

Thy-1-like Molecule in Earthworm 219

lactoperoxidase-catalyzed reaction [10]. Washed
labelled cells were solubilized with 200 ul of 0.5%
(w/v) Nonidet-P 40 in 0.15 M NaCl-0.02% NaN3
(w/v-0.01 M Tris-HCl, pH=7.4), by incubation on
ice for 1/2hr. The cell particulates insoluble in
NP-40 (nuclei and cellular debris) were removed
by centrifugation at 1200xg for 35 min at 4°C.
The supernatant containing solubilized cell surface
components was used for indirect immunopreci-
pitation. To 10’ labelled cells (200 sl), 50 pl of
monoclonal anti-mouse Thy-1 were added. After
incubation with antiserum for 45 min at 4°C, 100 pl
of a 10% protein A bound to Sepharose 4 B
(Pharmacia Fine Chemicals, Uppsala, Sweden)
were added and incubated for an additional 45 min
at room temperature. The precipitates were
washed thrice with 0.2M PBS, pH=7.2. Im-
munoprecipitates were dissolved in 2.2% SDS, 5%
2-mercaptoethanol, 0.1 mM EDTA in SDS-buffer

100

©
(e)

(0)
(eo)

L
(oe)

NO
(2)

=
fo)
(o)

80

% POSITIVE TARGET CELLS

‘Teo sa a 1
NUMBER OF -ABSORPING CELLS

Fic. 1.

2

and boiled for 3 min at 100°C. Sodium dodecyl
sulfate polyacrylamide gel electrophoresis (SDS-
PAGE) was performed on 10% polyacrylamide
gels. After electrophoresis, the gels were sliced (2
mm in length) and radioactivity counted in each
slice by gamma counter [12].

RESULTS

The reactivity of both anti-Thy-1.1 monoclonal
antibodies (mAbs Thy-1.1) and anti-Thy-1.2
monoclonal antibodies (mAbs Thy-1.2) was ti-
trated in IF against earthworm leucocytes. The
mAb Thy-1.1 showed a strong binding reactivity to
earthworm leucocytes in contrast to mAb Thy-1.2
which was completely negative (data not shown).
These results suggest that probably earthworm
leucocytes are bearers of the Thy-1.1 determinant
but lack the presence of the Thy-1.2 antigen.

(28)
h

3 4 1 2
(x 10® )

Demonstration of a Thy-1 homolog on earthworm leucocytes using rodent Thy-1 antibodies in quantitative

absorption assays. Numbers of absorbing cells are shown on the abscissae. Aliquots of earthworm leucocytes (@)
and rat (O) and BALB/c mouse (©) thymocytes were used as absorbents to assay the activity of anti-rat Thy-1.1
mAb (diluted 1: 1500 in the absorbing assay) to rat thymocytes (A) and earthworm leucocytes (B), the activity of
anti-mouse Thy-1.1 mAb (diluted 1 : 1800 in the absorbing assay) to rat thymocytes (C) and earthworm leucocytes
(D) and the reactivity of rabbit anti-rat Thy-1 antiserum (diluted 1: 100 in the absorbing assay) to earthworm
leucocytes (E) and rat thymocytes (F). Each point in the curves represents the mean value of two separate

experiments.

220 A-H. SaapD AND E. Cooper

Recent amino acid sequence data of murine
brain derived Thy-1 molecules have demonstrated
that protein encoded by the allels Thy-1.1 and
Thy-1.2 differ by one amino acid residue, there-
fore antibodies are directed exclusively to one
allelic Thy-1 product. Bearing this in mind, it was
necessary to perform a set of quantitative absorp-
tions to be certain that anti-Thy-1.1 mAb indeed
detects the earthworm putative Thy-1 homolog.
The absorptive capacity of worm leucocytes, rat
and BALB/c mouse thymocytes to reduce the
activity of BALB/c anti rat Thy-1 mAb towards rat
thymocytes and earthworm leucocytes is depicted
in Figure 1A and B, respectively. Within the given
range of absorbents, earthworm leucocytes re-
duced detectable reactivities of mAb Thy-1.1 de-
terminants as effectively as did the absorption of
rat thymocytes (Thy-1.1 strain). In contrast, the
binding capacity of mAb was not completely abo-
lished by BALB/c mouse thymocytes (Thy-1.2
strain). Significant absorptions were further con-
firmed in quatitative absorption assays using
BALB/c mouse anti-AKR/J mouse thymocyte
mAb with target rat thymocytes (Fig. 1C) and
earthworm leucocytes (Fig. 1D). This resulting
pattern of reactivity was similar to that observed
using anti-rat mAb, confirming the detectability of
a Thy-1.1 cross-reacting determinant, by a putative
Thy-1.1 homolog on earthworm leucocytes.

Although results from quantitative absorption
assays suggest that the Thy-1.1 epitope is shared
between rat Thy-1 molecules and a putative Thy-1
homolog on earthworm leucocytes, there is no
evidence that the molecule(s) are serologiacally
identical or not. In order to further map the
difference between rodent and earthworm Thy-1,
rabbit anti-rat Thy-1 antiserum was absorbed with
rat and BALB/c mouse thymocytes and assayed
with rat thymocytes (Fig. 1F) and earthworm
leucocytes (Fig. 1E) as target cells. While rat
thymocytes are capable of absorbing nearly all the
binding reactivity, BALB/c mouse thymocytes
absorbed only 30% and earthworm leucocytes
absorbed 70% of the reactivity. Specific absorp-
tion occurred using the same number of ear-
thworm leucocytes and rat thymocytes which indi-
cates two main antibody specificites: a specificity
directed to the antigenic determinants shared by

earthworm and rat (mouse), the earthworm-rat
(-mouse) cross reacting xenoantigenic determinant
and a specificity directed to an antigenic determi-
nant selectively absorbed by rat, the Thy-1.1 anti-
genic determinant. In contrast, reactivity of rabbit
anti-rat Thy-1 against BALB/c mouse thymocytes
as targets was completely diminished by BALB/c
mouse thymocytes and earthworm leucocytes,
(data not shown); rat thymocytes abosorbed only
about 28% of the antibody specificities.

Immunoprecipitates of earthworm '~I-labelled,
solubilized leucocytes were obtained by anti-
mouse Thy-1.1 mAb and subjected to SDS-PAGE
analysis using 10% gels under reducing conditions.
The results indicate an apparent molecular weight
of 28.2 KD for the Thy-1 homolog (Fig. 2).

125) bp X 102

1. 2°34 °'°5" 67 77 Saesene

MOBILITY RF

Fic. 2. Polyacrylamide gel electrophoresis (PAGE)
analysis of immunoprocipitated earthowrm leuco-
cytes. The position of protein markers that we run
simultaneously and stained with Coomassie Blue is
indicated by arrows (A, phosphorylase b (94.0); B,
BSA (67.0); C, ovalbumin (43.0); D, carbonic
anhydrase (30.0); E, Soybean trypsin inhibitor
(20.0) and F, Lactalbumin (14.0). This indicates an
apparent molecular weight of 28.2 KD for the Thy-1
homolog.

DISCUSSION

Although Thy-1 has been used as a T-cell mar-
ker, there is no difinite resolution as to its role in
T-cell functions. Recently, it has been postulated
that Thy-1 may be ancestral molecule from which

Thy-1-like Molecule in Earthworm 221

all members of the Ig superfamily might have
evolved [11-13]. To follow up this ancestral mole-
cule, studies in invertebrates, seemed to us, to be
highly warranted. Although, invertebrates leuco-
cytes have long been overlooked with respect to
cell markers, they might prove to be the most
suitable substrate for defining the roots of Thy-1
homolog from an evolutionary viewpoint.

In our approach to search for a Thy-1 homolog
on earthworm leucocytes, the reactivity of two
mAbs or proven specificities to the Thy-1.1 deter-
minant of rat and AKR/J mouse Thy-1 molecule
[14] and rabbit anti-rat Thy-1 antiserum [9] to-
wards earthworm leucocytes was investigated in IF
assays by quantitative absorption. Although,
neither reagents generated a response the putative
earthworm Thy-1 antigen, both were depleted of
anti-Thy-1.1 reactivity by absorption with earth-
worm leucocytes to the same degree as rat
thymocytes (Thy-1.1 strain). Due to the specificity
of the antisera and the sensitivity of our assay, the
membrane determinant revealed on earthworm
leucocytes seemed to be related to the Thy-1
molecule. Two antibody specifities in the antisera
were recognized and found to be directed towards
two antigenic determinants expressed on earth-
worm leucocytes. These were referred to as: the
earthworm-rat (mouse) cross reacting xenoan-
tigenic determinants and the Thy-1.1 qantigenic
determinant. This observation substantiates the
occurrence of a Thy-1 homolog on earthworm
leucocytes, which in terms of structure, might
share a common pattern with rodent Thy-1 mole-
cule, manifested by the expression of Thy-1.1
determinant.

The similarity of Thy-1.1 determinant in ear-
thworm and rat, in particular, may be streng-
thened by results from immunoprecipitation
assays. The estimated molecular weight was in
stricking agreement with values obtained for rat
[9], mouse [15], human [16], dog [16], frog [17] and
tunicate [18] Thy-1 glycoprotein. However, the
ultimate proof for this assumption would be
obtained by biochemical characterization, purifica-
tion and amino acid sequence studies on ear-
thworm Thy-1 epitope which are in progress. The
presence of another vertebrate-like molecule,
within the Ig superfamily such as Thy-1, among

earthworms may not seem surprising since our
results add to a growing list of such shared mole-
cules, including for example, />-microglobulin [4].
In terms of similarities in antigenicity, it is conceiv-
able that the earthworm leucocytes Thy-1 homolog
represents an ancestral Thy-1 molecule that under-
went diversification. Vertebrates are assumed to
have evolved from the chordate line represented
by tunicates which are deuterostomes. That a
putative Thy-1 homolog exists in earthworm which
belongs to the protostome line, supports the view
that Thy-1, a component of the Ig superfamily, is
present universally. Moreover, this argues for
later diversification of the terminal gene during
evolution of other members of the superfamily
such as Ig which is up to now not demonstrable in
invertebrates and only present in all vertebrates [6,
7]. Such information may provide essential clues
to our understanding of fundamental events con-
cerning the evolution of the immune system.

REFERENCES

1 Cooper, E. L. (1976) The earthworm coelomocytes.
A mediator of cellular immunity. In “Phylogeny of
Thymus Bone Marrow Cells”. Ed. by R. K. Wright
and E. L. Cooper, Elsevier, Amsterdam, pp. 9-42.

2 Roch, P. Cooper, E. L. and Eskinazi, D. P. (1983)
Surgical evidence for a membrane structure related
to human beta-2-microglobulin expressed by certain
earthworm leucocytes. Eur. J. Immunol., 13: 1037—
1046.

3 Shalev, A., Segal, S. and Bar Eli, M. (1985) Evolu-
tionary conservation of brain Thy-1 glycoprotein in
vertebrates and invertebrates. Dev. Comp. Im-
munol., 9: 497-506.

4 Growford, J. M. and Goldschreider, I. (1980) Thy-1
antigen and B lymphocyte differentiation in the rat.
J. Immunol., 124: 969-976.

5 Seki, T., Change, H. C., Moriuchi, T., Denome, R.
and Silver, J. (1985) Thy-1 a hydrophobic trans-
membrane segment at the carboxyl terminus. Fed.
Proc., 44: 2865-2869.

6 Williams, A. F. and Gagnon, J. (1982) Neuroral cell
Thy-1 glycoprotein: Homolog with immunoglobulin.
Science, 216: 696-703.

7 Williams, A. F. (1984) The immunoglobulin super-
family takes shape. Nature, 308: 12-18.

8 Cooper, E. L., McDonald, H. R. and Sordat, B.
(1979) Separation of earthworm coelomocytes by
velocity sedimentation. In “Function and Structure
of the Immune System”. Ed. by Plenum Press, New

10

11

12

13

222

York, pp. 101-104.

Barclay, A., Letarte-Muirhead, M. and Williams,
A. F. (1975) Purification of the Thy-1 molecule
from rat brain. Biochem. J., 151: 699-707.

Ades, E. W., Zwerner, R. K., Acton, R. T. and
Balch, C. C. (1980) Isolation and partial character-
ization of the human homolog of Thy-1. J. Exp.
Med., 151: 400-409.

Kaufman, J. E. and Strominger, J. L. (1982) HLA-
DR Light chain has a polymorphic N-terminal re-
gion and conserved immunoglobulin-like C-terminal
region. Nature, 279: 694-697.

Parnes, J. K. and Seidman, J. G. (1982) Structure of
wild-type and mutant mouse beta-2-microglobulin
genes. Cell, 29: 661-667.

Cushley, W. and Owen, M. J. (1983) Structural and
genetic similarities between immunoglobulins and
class-I histocompatibility antigens. Immunol. To-
day, 4: 87-92.

14

15

16

17

18

A-H. SAAD AND E. Cooper

Mason, D. W. and Williams, A. F. (1980) The
kinetics of antibody binding to membrane antigens
in solution and at the cell surface. Biochem. J., 187:
1-9.

Zwerner, R. K., Barstad, P. A. and Acton, R. T.
(1977) Isolation and characterization of murine cell
surface components. I. Purification of milligram
quantities of Thy-1.1. J. Exp. Med., 146: 986-995.
McKenzie, J. L., Allen, A. K. and Fabre, J. W.
(1981) Biochemical characterization including ami-
no acid and carbohydrate composition of canine and
human brain Thy-1 antigen. Biochem. J., 197: 629-
637.

Mansour, M. H. and Cooper, E. L. (1984) Purifica-
tion and characterization of Rana pipiens brain
Thy-1 glycoprotein. J. Immunol., 132: 2515-2525.
Mansour, M. H. and Cooper, E. L. (1984) Serolo-
gical and partial molecular characterization of Thy-1
homolog in tunicates. Eur. J. Immunol., 14: 1031-
1042.

ZOOLOGICAL SCIENCE 7: 223-228 (1990)

© 1990 Zoological Society of Japan

Neural Control of Flight Muscle Differentiation in the Fly,
Sarcophaga bullata

PAKKIRISAMY SIVASUBRAMANIAN and Dick R. NAssEL!

Department of Biology, University of New Brunswick, Fredericton,
New Brunswick, Canada E3B 6E1, and ‘Department of Zoology,
University of Stockholm, Svante Arrhenius vag 16,
S-10691 Stockholm, Sweden

ABSTRACT—Myoblasts derived from imaginal discs together with the degenerating larval muscles
contribute to the formation of flight muscles in adult flies. Severance of the mesothoraic larval nerve in
the freshly formed prepupa results in the absence of entire flight musculature on the operated side of the
adult fly. Histological observations reveal very early stages of muscle differentiation in the form of
association of myoblasts with degenerating larval muscle around day 3 of pupal development. Further
differentiation is inhibited by nerve transection leading to the eventual degeneration of all muscles on
the operated side. These results indicate that insect flight muscles are dependent on nerves from the

early stages of their differentiation.

INTRODUCTION

Although the importance of innervation in the
differentiation of vertebrate muscle has been
established by several investigations [1—6], insects
that undergo complete metamorphosis are better
suited for such studies because of precise timing in
the differentiation of adult muscles. When a
crawling maggot such as that of a fly, metamorpho-
ses into a flying insect there is a considerable
reorganization of its locomotor apparatus. The
larval muscles are histolysed and replaced by im-
aginal muscles. The nervous system too is remode-
lled accordingly. Such a metamorphosing system is
very convenient for the study of neural control of
muscle differention.

In one of his pioneering studies on the influence
of nerves on muscle differentiation Kope¢ [7]
removed the thoracic ganglia from gypsy moth
caterpillars. This resulted in the development of
adults without thoracic muscles. Similar results
were also obtained for silkmoths by Williams and
Schneiderman [8]. Niiesch extended these studies
to single nerves innervating specific muscles of the

Accepted May 9, 1989
Received March 29, 1989

moth, Antheraea pernyi and concluded that in-
nervation is essential for the completion of muscle
differntiation [9]. We have studied the influence of
nerves in the differentiation of muscles in a differ-
ent order of insects namely, Dipera. As in other
flies of Diptera, the adult Sarcophage bullata has
six pairs of dorsal longitudinal muscles (DLM), part
of the indirect flight musculature, that increase the
height of the thoracic box during flight and thus
depress the wings. These muscles are innervated
by the posterior dorsal mesothoracic nerve
(PDMN) from the thoracic ganglion [10]. In this
report we have examined the role of PDMN in the
differentiation of DLM.

METERIALS AND METHODS

The fleshfly Sarcophaga bullata was reared in the
laboratory under constant conditions of tempera-
ture (25°C) and photoperiod (16L: 8D). The adult
flies were fed with sugar and water ad libitum. The
larvae were raised in fresh beef liver. Post-feeding
mature third instar larvae were collected and used
for experiments within two hr after pupariation.

Denervation Yn the mature larva there are
groups of embryonic cells enclosed within non-
cellular peripodial membranes. These are called

224 P. SIVASUBRAMANIAN AND D. R. NASSEL

imaginal discs which differentiate into adult struc-
tures during metamorphosis. Each disc is con-
nected terminally to the larval epidermis via slen-
der epithelial stalk. Most of the discs also have a
basal stalk connecting them to the larvel central
nervous system. This stalk also contains a larval
nerve [11]. Thus the pro- and mesothoracic leg
discs are connected to the ventral side of the
ganglion by pro-and mesothoracic larval nerves
which also send off branches to innervate the larval
thoracic muscles [12]. Since it has been known that
(a) some larval thoracic muscles contribute to the
formation of adult DLM [13], (b) larval neurons
are remodelled into adult nerve cells [14], and (c)
the PDMN innervating the DLM of the adult
originates in the mesothoracic neuromere [15], we
hypothesised that the nerve branch of the larval

mesothoracic nerve is transformed into PDMN of
the adult and is essntial for the differentiation of
adult DLM. This larval nerve was transected as
follows: a triangular cut was made in the puparium
on the anterior ventral side (segments 4—5) of the
1-2 hr old prepupa. While lifting the puparial flap,
a fine iridectomy scissors was introduced inside
and the basal disc stalk along with its nerve branch
of the mesothoracic leg disc was severed (Fig. 1A,
top). The epithelial stalk was left intact. The
window was closed back with the triangular flap
and the drying of the hemolymph sealed the
wound. In the sham operated controls, the entire
mesothoracic leg disc was extirpated (Fig. 1A,
bottom) leaving the larvel nerve connection intact
as described by Nassel et al. [16]. A single oblique
cut was made on the anterior ventral (segments 3—

Fic. 1.

Nerve transection procedure and its effects on adult flies. Abreviations: DVM: Dorsoventral flight muscles;

DLM: Dorsal longitudinal muscles; EP: Epithelium; Segm. N: Segmental nerve. Scale bar in B, C and D is the
same=Ilmm. A. Top: Transection of mesothoracic leg disc stalk along with the larval mesothoracic nerve.
Bottom: Extirpation of mesothoracic leg disc (sham-operation). B. Cross section of a thorax from an unoperated

control fly. C. Cross section of a thorax from sham operated (mesothoracic leg disc extirpated) fly.

D: Cross

section of a thorax from nerve transected fly. Note the absence of flight muscles on the right side.

Muscle Differentiation in Fly 225

4) side of the freshly formed prepupa. Gentle
pressure on one side externalized the mesothoracic
leg disc which was then detached from its attach-
ments on both ends. The mesothoracic larval
nerve branch that innervates the larval muscles
was left intact. Upon completion of metamorpho-
sis (about 12 days at 25°C) the condition of their
flight muscles was examined by simply slicing the
thoraces as well as after histological staining of 8

EBS sbi: 18.

et 2

ym thick serial paraffin sections.

RESULTS

Control I (unoperated flies)

Figure 1B is a cross section of a thorax from an
unoperated adult fly. As in other cyclorrhaphous
diptera it contains six paris of DLM which are the
main depressors of the wings during flight. These

Ram Be PPADS ante tn Ties

Fic. 2. Chronological stages of muscle degeneration after nerve transection. Magnification in A, B and E is the
same. (scale bar=400 wm). C, D and F are of same magnification (scale bar=100 wm). A. 3 day old pupa
showing DLM on both sides (arrows). B. 3.5 to 4 day old pupa. Note degenerating DLM on left side (arrow
head). C. Enlarged view of degenerating DLM from 2B. D. Enlarged view of normal DLM from 2B. E. 7 day
old pupa operated on the right side showing the DLM on left side only (arrow). F. Enlarged view of normal DLM

from 2E.

226 P. SIVASUBRAMANIAN AND D. R. NASSEL

muscles extend from the anterior half of alinotum
to the post notum and second phragma.
Control II (sham operated flies)

The mesothoracic leg extirpated prepupae meta-
morphose into five-legged flies. Besides the
mesothoracic leg, these flies also lack certain scle-
rites on the operated side such as the mesoster-
num. Nevertheless, all six pairs of DLM develop
quite normally (Fig. 1C). However, the dorso-
ventral muscles (DVM) are absent on the operated
side.

Experimental flies

More than 75% of the pupae survive the opera-
tion and at least 35 flies that completed meta-
morphosis were used for examination. Upon
transection of the basal stalk along with the
attached larval nerve of the mesothoracic leg disc
the pupae metamorphosed quite normally with all
six legs and associated sclerites. However, they
invariably failed to eclose unaided. When the flies
were taken out of their puparia one could notice
some extermal color difference between the right
and left halves of the dorsal thorax. One half was
paler than the other half. This was due to the
complete absence of the entire set of fibrillar flight
muscles on the operated side which could be seen
through the as yet untanned cuticle. The muscle-
less half of the thoracic box was filled with fat body
(Fig. 1D). The thoracic ganglion too was slightly
abnormal in such flies and the PDMN was absent
on the left side.

At what time during the course of development
do muscles need innervation? In the experimental
flies do the muscles differentiate first and then
degenerate or they fail to differentiate altogether
due to lack of innervation? To answer these
questions, nerve transected pupae (10 per stage)
were histologically examined at different stages of
development. Three days after pupariation the
larval mesothoracic muscles had cleaved longitudi-
nally into six bundles on either (right and left) side
of the thorax with the myoblasts lined up around
them. In a cross section one can see these bundles
just beneath the dorsal epidermis (Fig. 2A). With-
in the next 12-24 hr (3-4 day old pupa) these
incipient muscle bundles start to degenerate on the
operated side (Fig. 2B and 2C) and by the next day
they have completely disappeared from the nerve

transected side. Figure 2E is a 7 day old pupa
showing the flight muscles on the unoperated side.

DISCUSSION

By selective transection of larval mesothoracic
nerve the present study confirms the earlier reports
on moths by Kope¢ [7] and Williams and
Schneiderman [8] and demonstrates the import-
ance of innervation for the differentiation of thor-
acic flight muscles in the fleshfly, Sarcophaga bulla-
ta. However, our results are slightly different from
those obtained by Niiesch [9] after denervation of
developing muscles in diapausing pupae of satur-
niid moth Antheraea pernyi where the nerve trans-
ection resulted in retardation of muscle differentia-
tion with fewer nuclei and thinner, shorter muscle
fibers. This difference in nerve influence may be
due to the timing of denervation. In the moth the
operation was performed in the pupal stage,
whereas in the fly it was done much earlier in the
freshly formed prepupa.

The indirect fligh muscles of holometabolous
insects develop from myoblasts derived from leg
imaginal discs [17, 18] in association with degener-
ating larval intersegmental muscles [19, 20]. The
residual larval intersegmental muscles of the
mesothorax from a scaffolding around which the
myoblasts line up in the form of compact columns
and subsequently differentiate into myofibrils [21].
Upon nerve transection in the prepupa the adult
thorax is completely devoid of flight muscles on the
denervated side. Our histological preparations
indicate that from the moment the myoblasts
associate with the longitudinally cleaved larval
muscles their further differentiation is dependent
on innervation because, on the nerve transected
side the muscles seem to start degenerating soon
after this stage (Fig. 2B). The precise nature of
degeneration is difficult to descern from the cur-
rent histological observations. Only future studies
using ultrastructural and immunocytochemical
techniques will shed more light in this regared.

The importance of motor innervation during
early stages of muscle differentiation is well
documented for vertebrate embryos [1]. In chick
embryos motor neuron growth cones associate
with muscle forming mesodermal cells even before

Muscle Differentiation in Fly 227

myotube formation [22]. Destruction of motor
neurons by bungarotoxin inhibits myotube forma-
tion in rat skeletal muscles [2]. It was suggested
that motor neuron terminals have some trophic or
inductive influence on myogenic cells [4, 5]. Accu-
mulation of myosin in developing limb bud muscu-
lature of quail embryos is nerve dependent [23]. In
grasshopper embryos too the motor neurons con-
tact muscle pioneers very early in development
[24] and may start infiuencing them from the very
begining.

The indirect flight muscles of the adult fleshfly
Sarcophaga bullata are innervated by PDMN [10].
It is intersting to note that this nerve is missing on
the operated side. The absence of muscles could
not have caused the degeneration of PDMN
because insect motor neurons are able to survive in
the absense of their targets [25] and innervate
inappropriate muscles [16, 26]. Therefore, it is
tempting to suggest that the larval nerve which
innervates the mesothoracic muscles of the larva
perhaps becomes transformed into PDMN of the
adult. Thus, the PDMN has a dual role; first,
during metamorphosis it influences the flight mus-
cle differentiation and then, in the adult fly it
controls the function of these muscles Transection
of the larval mesothoracic nerve results in the
absence of PDMN in the adult. Such a transforma-
tion of larval nerve innervating the dorsal muscula-
ture of the larva into adult nerve innervating the
DLM of the adult is well established for the
tobacco hornworm, Manduca sexta [14].

The absence of dorsoventral muscles (DVM) in
the sham-operated (mesothoracic leg disc extir-
pated) controls needs some explanation. In this
group of control flies there is an intact larval
mesothoracic nerve and therefore the DLM differ-
entiate. However, the DVM fail to form. Similar
results have also been reported earlier for Dro-
sophila melanogaster [27] and Sarcophaga bullata
[28]. We can suggest two possible explanations.
(a) Since mesothoracic leg discs also contribute to
the adult epidermis of ventral mesothoracic seg-
ment [29] the DVM of disc extirpated animals
would have degenerated secondarily due to lack of
ventral attachment sites (sclerites). Studies with
Sarcophaga bullata [30] and Drosophila melano-
gaster [31] support this possibility. (b) The source

of myoblasts for DVM and DLM may be different.
Investigations with Drosophila wing mutants sup-
port this view. In the mutant wingless develop-
ment of DVM is affected while DLM are normally
formed. In erect wing mutants, on the other hand,
DLM are completely absent while DVM are nor-
mal [31].

ACKNOWLEDGMENTS

This research was supported by grants from the Natu-
ral Sciences and Engineering Reseach Counci of Canada
and the Swedish Natural Sciences Research Council
(NFR 1820-103).

REFERENCES

1 Bonner, P. H. and Adams, T. R. (1982) Neural
induction of chick myoblast differentiation in cul-
ture. Develop. Biol., 90: 174-184.

2 Harris, A. J. (1981) Embryonic growth and innerva-
tion of rat skeletal muscles. I. Neural regulation of
muscle fiber numbers. Phil. Trans. Roy. Soc. Lond.,
Ser. B. 293: 257-277.

3 McLennan, I. S. (1983) Neural dependence and
independence of myotube production in chicken
hindlimb muscles. Develop. Biol., 98: 287-294.

4 Popiela, H. (1978) Trophic effects of adult peripher-
al nerve extract on muscle cell growth and
diferentiation in vitro. Exp, Neurol., 62: 405-416.

5 Popiela, H. and Ellis, S. (1981) Neurotrophic fac-
tor: Characterization and partial purification. De-
velop. Biol., 83: 266-277.

6 Sohla, G. S. and Holt, R. K. (1980) Role of
innervation on the embryonic development of
skeletal muscle. Cell Tiss. Res., 210: 383-393.

7 Kopec, S. (1923) The influence of the nervous
system on the development and regeneration of
muscles and integument in insects. J. Exp. Zool.,
37: 15-25.

8 Williams, C. M. and Schneiderman, H. A. (1952)
The necessity of motor innervation for the develop-
ment of insect muscles. Anat. Rec., 113: 560-561.

9 Nuesch, H. (1985) Control of muscle development.
In “Comprehensive Insect Physiology, Biochemistry
and Pharmacology” Vol. 2. Ed. by G. Kerkut, and
L. I. Gilbert, Pergamon Press, pp. 425-452.

10 Ikeda, K. (1977) Flight motor innervation of a
fleshfly. In “Identified Neurons and Behavior of
Arthropods” Ed. by G. Hoyle, Plenum press, New
York, pp. 357-368.

11 Auerbach, C. (1936) The development of the legs,
wings and halteres in the wild type and some mutant
strains of Drosophila melanogaster. Trans. Roy.

15

19

20

21

22

228

Soc. Edin., 58: 787-815.

Hertweck, H. (1931) Anatomie und Variabilitat des
Nervensystem und der Sinnesorgane von Drosophila
melanogaster. Zeits. wiss. Zool., 139: 559-563.
Beinbrech, G. (1968) Elektronenmikroskopische
Untersuchungen uber die Differenzierung von In-
sektenmuskeln wahrend der Metamorphose. Z.
Zellforsch., 90: 463-494.

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.
Power, M. E. (1948) The thoracico abdominal
nervous system of an adult insect, Drosophila mela-
nogaster. J. Comp. Neuro., 88: 347-40).

Nassel, D. R., Helgee, A. and Sivasubramanian, P.
(1986) Development of axon paths of motoneurons
after removal of target muscles in a holometabolous
insect. Devel. Brain Res., 26: 211-219.

Poodry, C. A. and Schneiderman, H. A. (1970)
Ultrastructure of the developing leg of Drosophila
melanogaster. Roux’s Arch. Dev. Biol., 166: 1-44.
Shatoury, H. H. El. (1956) Developmental inter-
actions in the development of imaginal muscles of
Drosophila. J. Embryol. exp. Morphol., 4: 228-239.
Reed, C. T., Murphy, C. and Fristrom, D. (1975)
The ultrastructure of the differentiating pupal leg of
Drosophila melanogaster. Roux’s Arch. Dev. Biol.,
178: 285-302.

Tiegs, O. W. (1955) The flight muscles of insects.
Phil. Trans. Roy. Soc. Lond. (Biol.), 238: 221-348.
Peristianis, G. C. and Gregory, D. W. (1971) Early
stages of flight muscle development in the blowfly
Lucilia cuprina: A light and electron microscopic
study. J. Insect Physiol., 17: 1005-1022.
Lance-Jones, C. and Landmesser, L. (1981) Path-
way selection by chick lumbosacral motoneurons
during normal development. Proc. Roy. Soc. Lond.

23

24

25

26

27

28

29

30

31

P. SIVASUBRAMANIAN AND D. R. NASSEL

(Biol.), 214: 1-18.

Merrifield, P. A. and Konigsberg, I. R. (1987)
Nerve dependent accumulation of myosin light chain
3 in developing limb musculature. Development,
101: 673-684.

Ball, E. E., Ho, R. K. and Goodman, C. S. (1985)
Development of neuromuscular specificity in the
grasshopper embryo: Guidance of motoneuron
growth cones by muscle pioneers. J. Neurosci., 5:
1808-1819.

Whitington, P. M., Bate, M., Seifert, E., Ridge, K.
and Goodman, C. S. (1982) Survival and dif-
ferentiation of identified embryonic neurons in the
absence of their target muscles. Science, 215: 973-
975.

Whitington, P. M. (1985) Functional connections
with foreign muscles made by a target deprived
insect motor neuron. Develop. Biol., 107: 537-540.
Zalokar, M. (1947) Anatomie du thorax de Dro-
sophila melanogaster. Rev. Suisse Zool. 54: 17-53.
Sivasubramanian, P. and Nassel, D. R. (1985) De-
velopment of thoracic musculature after leg imaginal
disc extirpation in the fly, Sarcophaga bullata (Par-
ker) (Diptera: Sarcophagidae). Int. J. Insect Mor-
phol. Embryol., 15: 463-464.

Bryant, P. J. (1978) Pattern formation in imaginal
discs. In “The genetics and Biology of Drosophila”
Vol. 2c, Ed. by M. Ashburner and T. R. F. Wright,
Acad. Press, New York/London. pp. 230-335.
Chiarodo, A. J. (1963) The effects of mesothoracic
leg disc extirpation on the nervous system of the
blowfly, Sarcophaga bullata. J. Exp. Zool., 153:
263-277.

Deak, I. I. (1978) Thoracic duplications in the
mutant wingless of Drosophila and their effect on
mucles and nerves. Develop. Biol., 66: 422-441.

ZOOLOGICAL SCIENCE 7: 229-234 (1990)

© 1990 Zoological Society of Japan

Effect of pH on the Participation of Calcium Ion in the
Cell Aggregation of Sea Urchin Embryos

Yasuto TonEGAWwa!, Encut Hosrro? and KAZUHIDE TAKAHASHI?

Department of Regulation Biology, Faculty of Science,
Saitama University, Urawa 338, Japan

ABSTRACT—The effect of pH on the cell aggregation of sea urchin embryos was investigated to
demonstrate the involvement of Ca’ in cell aggregation. The cell aggregation was maximum at pH 8,

decreased gradually at lower pHs and the aggregation did not occur below pH 4.

On further

examination, the pH profile of Ca? * -binding to the aggregation factor was found to be almost identical

with that of cell aggregation.

The similarity of these pH profiles suggested the involvement of

electrostatic interaction between Ca** and negatively charged groups of the aggregation factor in the

cell aggregation of sea urtin embryos.

INTRODUCTION

The success in immunochemistry stimulated
hypotheses involving antigen-antibody like reac-
tions and sugar-lectin type reactions as the specific
motive forces of cell aggregation. This line of
research has advanced so far to propose the
molecular mechanism for the interaction of
aggregation factor and cell surface receptor [1-4].
On the other hand, biophysical approach has sug-
gested the participation of electrostatic forces and
van der Waals forces. Ca bridge hypothesis is most
well known among them [5-7].

The cell aggregation factor of sea urchin
embryos was shown to be a highly negatively
charged sugar-protein complex extracted with
Ca**-g, Mg**-free antifical sea water (CMF-SW).
However, it has not yet been well characterized
because of its instability [8]. The surface charge of
dissociated cells is known to be negative at the pH
of sea water [9]. Ca ion is indispensable for cell

Accepted December 5, 1989
Received October 9, 1989

' Deceased on October 8, 1989.

* Present address: Chiyoda Dames and Moor Co,
LTD., 1-4-28 Mita, Minato-ku, Tokyo 108.

> Present address: Clinical Research Institute, Kanaga-
wa Cancer Center, 54-2 Nakao-cho, Asahi-ku, Yoko-
hama 241.

aggregation and the rate of cell aggregation is
dependent on Ca** concentrations [8]. Hence it is
reasonable to assume that the electrostatic forces
may play some substantial role in the cell aggrega-
tion of sea urchin embryos.

Following experiments were designed to ex-
amine this possibility. The cell aggregation was
significantly affected by pH with a maximum
aggregation at pH 8. The binding of *Ca’* to the
aggregation factor showed almost the same pH
profile as that of cell aggregation. These results
are discussed with special reference to the mode of
Ca** involvement in cell aggregation.

MATERIALS AND METHODS

A Japanese sea urchin, Hemicentrotus pulcherri-
mus, was mostly used for the following experi-
ments.

Preparation of aggregation factor and hyaline layer
substance

Aggregation factor was prepared from swim-
ming blastula embryos following the method pre-
viously described [8]. Hatched blastulae were
washed twice with Ca?*-, Mg’*-free artificial sea
water (CMF-SW) and gently stirred in ice-chilled
CMF-SW for 60 min until embryos were dissoci-
ated into constituent cells. After removing dissoci-

230 Y. TONEGAWA, E. Hosiro AND K. TAKAHASHI

ated cells with low-speed centrifugation (3000 x g,
5 min), the suernatant was subjected to high-speed
centrifugation (10,000xg, 20min) at 4°C. The
clear extract obtained was used as the cell aggrega-
tion factor.

To prepare [*S]-labeled aggregation factor
([?°S]-AF), fertilized eggs were raised in [°°S]-
MgSO,-containing artifical sea water (5 ~Ci/200
ml). At blastula stage, metabolically labeled
aggregation factor was extracted with CMF-SW
according to the same method as that to prepare
unlabeled eggregation factor.

Hyaline layer substance was extracted from fer-
tilized eggs with CMF-SW and purified by pre-
cipitation with Ca** as was described previously

[8].

Assay of cell aggregating activity

Dissociated blastula cells after the extraction of
the aggregation factor were washed twice with
CMF-SW,, filtered through nylon mesh (380 mesh)
and the cell numbers were counted with hemocyto-
meter. Cell aggregation assay was performed on a
zyratory shaker at 4°C. One ml of cell suspension
(10’ cells) and 1 ml of aggregation factor (100 yg
protein) were added to a 30 ml Erlenmeyer flask
containing 3 ml of Herbt’s artificial sea water buf-
fered with citrate (SmM) for pH3-6 or with
Na-barbiturate (5 mM) for pH 7-9. After rotation
(80rpm, 10mm rad.) for 60min, the average
number of cells in an aggregate was determined as
was described before [8].

To change the pH of the medium in the course
of experiment, 1 ml of 50 mM buffer of different
pH was added to the flask after 60 min of rotation
in the first medium. Rotation was continued for
further 60 min and the cell aggregation was scored.
To prepare the fixed cells, dissociated cells were
fixed with cold glutaraldehyde (1% in CMF-SW,
buffered to pH 8 with Tris-HCl for 3 hr and di-
alyzed thoroughly to remove excess glutaraldehy-
de. The fixed cells were filtered through nylon
mesh (380 mesh) to remove small aggregates
formed during fixation.

Preparation of cell surface glycopeptide

Dissociated blastula cells were treated with
0.1% trypsin in buffered CMF-SW (10 mM Tris-

HCl, pH 8.0) at 20°C for 1 hour. After removing
the cells by low speed centrifugation (1000 xg, 5
min), the extract was spun at 10,000 x g for 20 min
and the supernatant was concentrated by ultrafil-
tration (Amicon PM-30). The concentrated tryp-
sin extract was fractionated by gel filtration
through Sephadex G-50 column (2x 100 cm). The
first fraction containing most of the sugar was
pooled, lyophilized and used as cell surface gly-
copeptide.

Estimation of * Ca?* -binding

To estimate the binding of *Ca** to the cells, 1
ml of cell suspension (107 cells) and 1 ml of CMF-
SW or aggregation factor (100 g/ml) were added
to 3 ml of buffered artificial sea water containing 2
uCi of [*Ca**]-CaCl,. After rotation for 30 min
in the cold (80 rpm, 20 mm rad.), the cells were
separated from the supernatant by centrifugation
(1000 g, 5 min) and suspended in 5 ml of the
same buffer. After standing for 10 min, the cells
were spun down and transferred into scintillation
vials, and the radioactivity was counted with a
scintillation counter following addition of 10 ml of
Triton-toluene scinillator. Additive washing of the
cells did not affect the counting.

The binding of *Ca** to cell surface gly-
copeptide and to the aggregation factor was esti-
mated by gel filtration. Small columns of
Sephadex G-50 (6400 mm) were equilibrated
with buffered CMF-SW of different pHs. One
hundred microliters of sample solution (cell sur-
face glycopeptide, 100 ug; aggregation factor, 500
yg protein) were mixed with 100 «l of CMF-SW
containing 1 «Ci of *Ca**. After incubation for
15 min at 0°C, the mixture was applied on the top
of the column and eluted with the same buffer.
Ten drop fractions were collected in each scintilla-
tion vial and the radioactivity was counted after
addition of Triton-toluene scintillator. The peak
appeared in the void volume was separated from
the later peak retarded by the gel. The counts of
the former peak were summed and estimated as
bound *Ca** to macromolecular components.

Estimation of [*°S]-AF binding

The binding of [*°S]-AF to the cell was deter-
mined with the same procedure as that for See

Ca** Participation in Cell Aggregation 231

-binding, except for the use of [*°S]-AF (45,000
cpm/100 yg protein/ml) instead of *Ca**.

RESULTS

Effect of pH on cell aggregation

The cell aggregation induced by the aggregation
factor was examined in artificial sea water of
various pHs. As is shown in Figure 1, cell aggrega-
tion was dependent significantly on pH and was
maximal at pH8. The sizes of cell aggregates
decreased gradually at lower pHs and no aggre-
gates formed below pH 4. Cell aggregation did not
occur at any pHs in the absence of divalent cations
and also without the aggregation factor.

w
oO

nD
oO

[o)

Cell aggregation (cells/agg. )

Serene 4 Set 6 eee eee === =s=S=

3.0 7 On) OM OO QO 20 YG

7.
pH
Effect of pH on the cell aggregation of sea
urchin embryos. Cell aggregation in the presence of
the aggregation factor in artificial sea water (@—®),
the same in CMF-SW (O---O); cell aggregation in
the absence of the aggregation factor in artificial sea
water (4—4), the same in CMF-SW (4---4).

Fic. 1.

A similar experiment was done with fixed cells
to examine the possibility that the effect of pH on
cell aggregation might be due to its effect on cell
metabolism. The general pH profile of cell
aggregation with fixed cells was similar to that of
intact cells, although the extent of cell aggregation
was markedly reduced (Fig. 2).

When pH was changed in the course of experi-
ment (60 min) and the cells were kept at the
second pH for further 60 min, the size of cell
aggregates shifted close to that kept at the second

n(cells/agg.)
(oe)

(o)

fo)

Cell aggregatio

3.0 40 5.0 6.0 7.0 8.0 90
pH
Fic. 2. Effect of pH on the aggregation of fixed cells of
sea urchin embryos. Aggregation of fixed cells in
the presence of the aggregation factor in artificial sea
water (O---O) and aggregation of living cells under
the same conditions (@—®).

pH from the beginning. When pH was changed
from 4 to 8, the cells started to aggregate to reach
the similar size to that of the aggregates formed on
incubation at pH 8 (cells/aggregate, from 1.0 to
17.3). When pH was changed from 8 to 4, the cell
aggregates began to dissociate but they were not
completely dissociated after 60 min (cell/aggre-
gate, from 24.5 to 11.6).

Effect of pH on the binding of + §-labeled aggrega-
tion factor to the cells

The binding of *°S-labeled aggregation factor
({°°S]-AF) to the cells was examined in normal and
CMF-SW (Fig. 3). The binding of [*°S]-AF in
normal sea water was the highest at pH 8-9 and
decreased as pH was lowered to 6, but increased
again at pH 5-3. On the other hand, its binding in
CME-SW was low and did not change significantly
from pH 9 to 5 but increased at pH 4-3. When the
[°°S]-AF binding value in CMF-SW was subtracted
from its binding value in normal sea water, the
resulting pH profile (divalent cation-dependent
binding) turned out to be similar to that of cell
aggregation (Fig. 3).

Effect of pH on the binding of *®Ca’* to the cells
and to cell surface glycopeptide

The binding of *Ca** to dissociated cells was
examined in the absence and presence of the
aggregation factor. The binding of 45Ca’* to the
cells was the highest at pH 9 within the pH range

232 Y. ToNEGAWA, E. Hoyrro AND K. TAKAHASHI

30 40 50 60

7.0 8.0 9.0
pH

Fic. 3. Effect of pH on the binding of *S-labeled
aggregation factor ([*°S]-AF) to the cells in artificial
sea water (@—®) and in CMF-SW (0---O). Dotted
line indicates the difference between the values.

examined and decreased continuously to pH 6 and
retained half the maximal level at lower pHs (Fig.
4). The presence of aggregation factor at the
concentration enough to cause cell aggregation did
not alter this profile. In addition, the binding of
SCa** to cell surface glycopeptide showed a simi-
lar profile (Fig. 4).

200 iS

o
L
syed] "Ty aaa ee
U a
Ps aerials
5 a
°
(ea
3.0 40 5.0 6.0 7.0 8.0 9.0
pH
Fic. 4. Effect of pH on the binding of **Ca** to the sea
urchin embryo cells in CMF-SW in the presence

( ) and absence (@—@) of the aggregation
factor, and the effect of pH on the binding of **Ca**
to cell surface glycopeptide in CMF-SW (4-:--4).

Effect of pH on the binding of ®Ca’* to the
aggregation factor

The binding of *Ca’* to the aggregation factor
was analyzed by gel filtration. *Ca** was bound
to the aggregation factor maximally at pH 7-9,
and its binding was also dependent on pH. It
decreased continuously at lower pHs, and no signi-
ficant binding occurred below pH 4 (Fig. 5). This
pH profile of “Ca** binding to the aggregation
factor was almost identical to that of cell aggrega-
tion and to that of divalent cation-dependent [*°S]-
AF binding to the cells.

30 40 5.0 60 0 80 90

7.
pH
Fic. 5. Effect of pH on the binding of *Ca** to the

aggregation factor of sea urchin embryos in CMF-
SW.

Cell aggregation induced by the hyaline layer subst-
ance

The hyaline layer substance manifested cell
aggregating activity at all pHs tested in the pre-

Cell aggregation(cells/agg. )

30 40 50 .°60}.7.0 9e80iiau

Fic. 6. Effect of pH on the cell aggregation induced by
hyaline layer substance. Cell aggregation in the
presence of hyaline layer substance in artificial sea
water (@—@), the same in CMF-SW (©---O).

Ca** Participation in Cell Aggregation 233

sence of Ca**. The rate of cell aggregation did not
change significantly between pH 9 and 5. It was
reduced at pH 4 but large cell clumps were formed
at pH 3 (Fig. 6). Cell aggregating activity was also
observed in the absence of divalent cations at
hygher pHs. These pH profiles were quite diffe-
rent from those by the aggregation factor.

DISCUSSION

There has been a number of reports on the
effects of pH on cell aggregation [7, 10, 11]. These
authors showed pronounced cell aggregation at
physiological pHs and reduced aggregation at low-
er pHs. In the present study, cell aggregation,
induced by the aggregation factor of sea urchin
embryos, was examined at a wide range of pHs and
the extent of cell aggregation was shown to be
remarkably influenced by pH. This pH dependen-
cy suggested the participation of electrostatic
forces in cell aggregation. The change of pH
would influence the ionization of charged groups
of cell aggregation-related molecules and conse-
quently the electrostatic interaction among them.

When one considers the necessity of Ca’* in cell
aggregation, it would be natural to take account of
the electrostatic interactions among positively
charged Ca** and negatively charged groups of
the cell surface and those of the aggregation factor.
These negatively charged groups, when not io-
nized at lower pHs, would not interact with Chem
At higher pHs, on the contrary, they would be
ionized and accordingly be ready to interact elec-
trostatically with Ca**. Thus pH dependent
ionization of the negatively charged groups
appears to be the cause of pH dependent cell
aggregation.

There is a possibility that the pH dependency of
cell aggregation is due to indirect effect of pH
through cell metabolism. However, cell aggrega-
tion with fixed cells showed almost a similar pH
profile to that with live cells and the effect of pH
on cell aggregation was reversible. These results
favor the view that cell aggregation is influenced by
pH through reversible ionization of the charged
groups.

Previous experiments with labeled aggregation
factor [12], have shown that [*°S]-AF bound to the

cells and the binding was quantitatively prop-
ortional to the rate of cell aggregation. They
suggested actual involvement of the aggregation
factor in cell aggregation as an essential consti-
tuent. Accordingly, the effect of pH on [*°S]-AF
binding to the cells was examined in the present
experiments. Unexpectedly, its pH profile was not
similar to that of cell aggregation, and a consider-
able extent of binding was detected at lower pHs.
However, when the value of [*°S]-AF binding in
CMF-SW was subtracted from that in normal sea
water, the value of divalent cation-dependent
binding manifested the same pH profile as that of
cell aggregation.

When one assumes that the aggregation factor
and Ca’? constitute the intercellular bridges in cell
aggregation, two sites are possible to be influenced
by pH; one between Ca** and the negatively
charged groups of the cell surface and the other
between Ca’* and those of the aggregation factor.
The binding of “Ca’* to the cells was shown to be
independent of pH at lower pHs and the general
profiles was quite different from that of cell
aggregation. On the contrary, the pH profile of
*Ca’t-binding to the aggregation factor was
almost identical to that of cell aggregation. There-
fore, it would be reasonable to conclude that the
linkage between Ca** and the aggregation factor
constitutes the rate limiting step in the effect of pH
on cell aggregation. It is assumed that the binding
of Ca** and the aggregation factor forms an
essential link in the cell aggregation.

When one admit this assumption, the linkage of
aggregation factor to the cell surface still remains
to be elucidated. In our experiments, pH-
dependent binding of [?°S]-AF to the cells needed
Ca**, but it was also true that some significant
portion of [°°S]-AF bound to the cells without
Ca’*. As was reported in other systems [13-16], it
would be also possible that the aggregation factor
binds the cell surface by another mechanism with-
out the participation of Ca**. Calcium ion might
just enhance the cooperation among the apprega-
tion factors to stabilize the intercellular bridges.
Further studies to identify the mode of binding of
the aggregation factor and to elucidate the
mechanism of its binding to the cell surface are
needed to settle this problem.

234

The pH profile of cell aggregation induced by
the hyaline layer subtance turned out to be entirely
different from that by the aggregation factor.
Hyaline layer substance is a different cell aggregat-
ing agent from the so called aggregation factor [8].
Present result has added another proof to disting-
uish them.

ACKNOWLEDGMENTS

We thank to the staffs of Misaki Marine Biological
Station for their kind supply of sea urchins and the offer
of space and facilities. Thanks are also due to Dr. M.
Hozumi for his generous offer to use the liquid
scintillation spectrometer at the Saitama Cancer
Research Center. We are also inbebted to Prof. K.
Ishihara for his kind assistance in the preparation of the
manuscript. This work is partly supported by the grant in
aid from the Ministry of Education, Science and Culture,
Japan.

REFERENCES

1 Balsamo, J. and Lilien, J. (1975) The binding of
tissue-specific adhesive molecules to the cell surface.
A molecular basis for specificity. Biochemistry, 14:
167-171.

2 Edelman, G. M. (1983) Cell adhesion molecules.
Science, 219: 450-457.

3 Miller, W. E. G. and Miller, I. (1980) Sponge cell
aggregation. Mol. Cell. Biochem., 29: 131-143.

4 Turner, R. S. and Burger, M. M. (1973) Involve-
ment of carbohydrate group in the active site for
surface guided reassociation of animal cells. Nature,
244: 509-510.

5 Curtis, A. S. G. (1967) The Cell Surface: Its
Molecular Role in Morphogenesis. Logos Press,
London.

6

10

11

12

13

14

15

16

Y. ToneGawa, E. Hoyiro AND K. TAKAHASHI

Pethica, B. A. (1961) The physical chemistry of cell
adhesion. Exp. Cell Res. Suppl., 8: 123-140.
Steinberg, M. S. (1958) On the chemical bonds
between animal cells, A mechanism for type-specific
association. Amer. Nat., 92: 65-81.

Tonegawa, Y. (1973a) Isolation and characteriza-
tion of a particulate cell-aggregation factor from sea
urchin embryos. Dev. Growth Differ., 14: 337-352.
Sano, K. (1977) Changes in cell surface charges
during differentiation of isolated micromeres and
mesomeres from sea urchin embryos. Dev. Biol.,
60: 404-415.

Balsamo, J. and Lilien, J. (1974) Embryonic cell
aggregation: Kinetics and specificity of binding of
enhancing factors. Proc. Natl. Acad. Sci. USA, 71:
727-731.

Miller, W. E. G., Miller, I. and Zahn, R. K. (1974)
Two different aggregation principles in reaggrega-
tion process of dissociated sponge cells (Geodia
cydonium). Experientia, 30: 899-902.

Tonegawa, Y. (1973b) On the role of aggregation
factor in the reaggregation of dissociated cells. Zool.
Mag., 82: 254 (In Japanese).

Brackenbery, R., Rutishauser, U. and Edelman, G.
M. (1981) Distinct calcitum-independent and cal-
cium-dependent adhesion systems of chicken
embryo cells. Proc. Natl. Acad. Sci. USA, 78: 387-
391.

Jumblatt, J. E., Schlup, V. and Burger, M. M.
(1980) Cell-cel! recognition: Specific binding of
Microciona sponge aggregation factor to homotypic
cells and the role of calcium ions. Biochemistry, 19:
1038-1042.

Magnani, J. L., Thomas, W. A. and Steinberg, M.
S. (1981) Two distinct adhesion mechanisms in
embryonic neural retina cells 1. A kinetic analysis.
Dev. Biol., 81: 96-105.

Takeichi, M. (1977) Functonal correlation between
cell adhesive properties and some cell surface pro-
teins. J. Cell Biol., 75: 464-474.

ZOOLOGICAL SCIENCE 7: 235-247 (1990)

Mechanisms Underlying Regulation of Local Immune Responses
in the Uterus during Early Gestation of Eutherian
Mammals. III. Possible Functional Differentiation of

Macrophages Cultured Together with Blastocyst
in vitro, with Special Reference to the
Cellular Shape and Production of
Leukotriene C,

CHIKASHI TACHI and SumiE Tacur!

Zoological Institute, Faculty of Science, University of Tokyo, Tokyo 113, and
‘Department of Anatomy, Tokyo Women’s Medical
College, Tokyo 162, Japan

ABSTRACT— In the blastocyst-macrophage co-culture of the mouse, we found two major groups of
macrophages which were different in the morphology, i.e., the rounded cells, and the elongated cells.
The macrophages in direct contact with the embryonic cells, regardless of whether they were trophoblast
or ICM cells, assumed invariably rounded cellular shape. Colchicine, cytochalasin B and D induced
strong rounding of the macrophage. The rate of synthesis of leukotriene C, in those rounded
macrophages remained at a low level of the unstimulated cells. Therefore, it was tentatively proposed
that in the blastocyst-macrophage co-culture, the rate of production of LTC, in the rounded
macrophages, might remain at a low level, while in the elongated ones the rate might be enhanced [C.
Tachi and U. Zor, Zool. Sci., in press]. On the other hand, tuftsin, a naturally occurring tetrapeptide
known to augment phagocytosis as well as capability of antigen presentation in macrophages, raised
slightly the rate of LTC, synthesis around the concentration of 60nM or above. Recently Gupta
presented evidence indicating that LTC, mediates the initial estrogen-dependent phase of endometrial
modification during nidation in mice. While it is strongly suspected that the major source of LTC,
produced during that period, might be one of the functionally differentiated groups of macrophages in

© 1990 Zoological Society of Japan

the endometrium, further work is needed to corroborate the view.

INTRODUCTION

We proposed [1-3], on the basis of the observa-
tions made in the rat and the mouse, that mac-
rophages are probably involved in the local im-
mune responses elicited in the endometrium by
implanting blastocysts during early gestation of
eutherian mammals. We suggested [1-3], furth-
ermore, that the decidua might function to limit
the access of macrophages to the embryonic anti-
gens, regulating the afferent flow of immunological
information to the maternal immune system during
the initial phase of the recognition of the concep-

Accepted April 28, 1989
Received February 18, 1989

tuses. However, the precise role played by the
endometrial macrophages during implantation
and/or nidation that is one of most critical episodes
in the true viviparity of mammals, remains yet to
be clarified.

As an approach to the problem, we analyzed, at
ultrastructural levels, the mode of cell-to-cell in-
teractions between the blastocysts and mac-
rophages cultured together in vitro [4]. During the
course of experiments, it came to our notice that
the macrophages in the co-culture assumed, as will
be described in this report in detail, mainly two
distinctly different cellular shapes, i.e., they were
either rounded or elongated.

Macrophages have been known to change their
shape according to their functional states or to the

236 C. TACHI AND S. Tacut

changes in their environmental conditions [5-10].
In turn, it is possible to bring about functional
transformation of the macrophages cultured in
vitro by exposing them to compounds which affect
the cytoskeletal organization of the cells. Thus,
colchicine, cytochalasin B and vinblastin stimulate
the release of neutral peptidases, including elas-
tase [11, 12], collagenase [11], gelatinase [11],
azocaseinase [11] from the cultured peritoneal [11]
as well as alveolar [12] macrophages of the mouse.
Colchicine stimulates also the rate of production of
interleukin-1 [13], in the irradiated peritoneal mac-
rophages of the rat. Calcium ionophore A23187
added to the cultured peritoneal macrophages of
the mouse [Tachi and Zor, Zool. Sci., in press]
induced at alkaline pH’s of the medium, strong
elongation of the cellular shape, accompanied by
the highly accerelated rate of leukotriene C,
(LTC, in the following) production; stimulation of
the Ca**-dependent synthesis of LTC, [14] by
A23187 in macrophages has been reported earlier
in the literature [15-17].

Leukotrienes, as termed by Samuelsson et al.
[18], are a family of metabolites of arachidonic
acid, and produced via 5-lipoxygenase pathway.
Leukotriene C, is the first peptidolipid to be
synthesized in the family and held responsible for
mediating inflammation, asthma and many other
functions of leukocytes and/or macrophages [for
review see, 19-25] which are the major source of
LTC, in the mammalian body. Recently, in 1989,
Gupta et al. [26] proposed that LTC, might be one
of the mediators of the endometrial changes eli-
cited during the estrogen-dependent early phase of
implantation in the mouse, although the exact
source of LTC, produced during the period is yet
to be determined.

In order to examine the possibility that the
morphological difference of the macrophages in
the blastocyst-macrophage co-culture might reflect
the underlying functional difference which, in
turn, might be correlated with the rate of LTC,
synthesis in the phagocytes, we analyzed the dis-
tribution of macrophages according to their mor-
phology in the co-culture, and at the same time, we
assayed the rate of LTC, production in the mac-
rophages under the influence of the compounds
which affect the cytoskeletal organization of the

cells. It was hoped that such analyses will provide
us clues to understand the mechanisms underlying
the blastocyst-endometrial interactions during im-
plantation and/or nidation in eutherian mammals.

Part of the results has been presented in abstract
form [27-29].

MATERIALS AND METHODS

Animals

Specific pathogen free (SPF) mice of BALB/c
(H-2°) and C3H/HeJ (H-2*) strains were used
throughout the experiments. They were purchased
from a local dealer (Nippon Clea & Co., Ltd.,
Tokyo, Japan), and had been kept in the animal
room of our laboratory under regulated tempera-
ture (25°C) and illumination cycles (12 hr dark and
12 hr light per day), until they were used for the
experiments. Female BALB/c mice were mated
on the day of proestrus with fertile BALB/c males,
and if the vaginal plugs were found next morning,
the day was counted as Day 1 of pregnancy.

Drugs and reagents

Unlabelled LTC, was purchased from Wako
Pure Chemical Industries & Co., Ltd. (Osaka,
Japan). Calcium ionophore A23187, cytochalasin
B and D were obtained from Sigma Chemical &
Co., Ltd. (St. Louis, Mo, USA): each of the three
compounds was dissolved in dimethyl sulfoxide
(DMSO; Wako Pure Chemical Industries) at a
concentration of 500 ug/ml. Colchicine (Merck,
Darmstadt, West Germany) and Tuftsin (Wako
Pure Chemical Industries), a polypeptide known
to be an activator of macrophages, were dissolved
in glass distilled water at a concentration of 500
yvg/ml. Sterilized Ficoll-Hypaque solution (d=
1.090+0.001) was purchased from Otsuka Assay
Loboratories (Tokushima-shi, Japan).

Radioactive LTC,

Aqueous solution of tritium-labeled leukotriene
C, (specific activity, 39 Ci/mMol, Amersham
Japan & Co., Ltd., Tokyo, Japan), as supplied by
its manufacturer, was diluted with 50% ethanol to
give a final concentration of 1.0 #Ci/ml.

Blastocyst-Macrophage Co-Culture 237

Macrophages

The macrophages were collected from peri-
toneal exudate of C3H/HeJ mice by injecting 60
units of heparin dissolved in Earl’s minimum
essential medium (MEM) at a concentration of 30
units/ml under ether anesthesia. After 3—5 min,
the animals were killed by cervical dislocation and
the peritoneal fluid was gently aspirated into a
glass syringe.

For the co-culture experiments, the procedures
previously described were essentially followed [4].
The macrophages were collected by centrifugation
at approximately 1000 rpm for 5 min, and washed
three times with phosphate-buffered saline (PBS)
and added to the culture medium which is known
to favor the trophoblast spreading [30].

For the assay of LTC,, and the examination of
the effects of various compounds which affect the
cytoskeletal organization of the cells, highly
purified peritoneal macrophages were prepared as
described previously [28, C. Tachi and U. Zor,
Zool. Sci., in press].

Blastocysts

Blastocysts were collected from BALB/c mice as
described before [4] by flushing the uterine lumen
with the standard egg culture medium (SECM)
[31] on Day 4 of pregnancy. The collected
embryos were washed twice in SECM and intro-
duced into the culture of macrophages.

Co-culture of blastocysts and macrophages

The zona-encased blastocysts (BALB/c) col-
lected in the afternoon of Day 4 from the uterine

lumen were introduced to the culture of mac-
rophages (C3H/HeJ); usually 5 blastocysts were
added to a single Falcon dish (diameter of the dish,
34.5 mm) which contained macrophages allogeneic
to the embryos at an approximate concentration of
1-3X10° cells per dish. The co-cultures were
incubated at 37'C under the atmosphere of 5%
CO, and 95% air.

Determination of the rate of LTC production

The rate of LTC, production was determined, as
will be described elsewhere [C. Tachi and U. Zor,
Zool. Sci., in press], by assaying the total amount

of the cysteinyl leukotrienes released into the
medium from the macrophages during 1.5—2.0 hr
of the culture period. The radioimmunoassay of
the cysteinyl leukotrienes were done essentially
following the procedures described by Danzlinger
et al. [32] with minor modifications; the monoclon-
al antibodies against cysteinyl leukotrienes were
generous gift of Dr F. Kohen, Department of
Hormone Research, Weizmann Institute of Scien-
ce, Rehovot, Israel. The antibodies reacted with
LTC, (100%), LTD, (105%), and LTE, (77%) at
a 50% saturation level of binding [F. Kohen,
personal communication].

One tenth ml of the sample solution to be
assayed was mixed with an equal amount of the
antibody solution and incubated at 0°C for 30 min.
Then, 0.1 ml of 7H-LTC, solution containing 7 nCi
of the isotope, was added to the mixture, and had
been stood at 4°C overnight. Free LTC, was
removed by adding 0.2 ml of dextran-charcoal,
and by centrifugation at 15,000 rpm for 3 min. The
radioactivity was measured by liquid scintillation
counting.

Differential counting of macrophages according to
their cell shape

The photomicrographs of the co-cultures were
taken using a phase contrast microscope (Model
CK, Olympus & Co., Tokyo, Japan), approx-
imately 72 hr after the initiation of the culture. For
histological examination, the cells were fixed with
3.5% formaldehyde solution and stained with
Giemsa’s. The number of the cells of the elon-
gated, or the rounded cellular shape were dif-
ferentially counted on the photomicrographs, by
using a microcomputer-based graphic analysis sys-
tem (TACSYS/G) previously described by C.
Tachi [2].

RESULTS

1. Cellular shape of the macrophages cultured
together with blastocysts in vitro

Microscopic appearance of macrophages
adhered to the zona pellucida or the blastocysts
following the co-culture of the phagocytes with the
zona-encased embryos, is shown in Figure 1A.

238 C. TAcur AND S. Tacui

Fic. 1. Photomicrographs of blastocyst-macrophage co-culture. A) The blastocyst (BALB/c) is encased in zona
pellucida onto which numerous macrophages (C3H/HeJ) are firmly adhered. Approximately 6 hr after the
initiation of co-culture. B) The blastocyst is undergoing trophoblast spreading. Approximately 72 hr after the

initiation of co-culture.

The macrophages which were tightly bound to the
zona, assumed strongly rounded cellular shape
without exception; none of the zona-bound phago-
cytes observed was of elongated morphology.

In Figure 1B, representative photomicrograph
of the blastocyst-macrophage co-cultures approx-
imately 48 hr after the initiation, is presented. The

blastocyst was undergoing trophoblast spreading
and numerous macrophages were found in contact
with the periphery of the trophoblast cell layer.
Some of the macrophages were located within the
embryos (Fig. 1B).

In the co-culture, two groups of macrophages of
distinctly different morphology were discernible,

Blastocyst-Macrophage Co-Culture 239

i.e., those which were rounded and the others
which were either elongated or spread (Fig. 1B).
The macrophages with spread morphology,
however, were only rarely observed under the
conditions we employed. We counted the number
of macrophages according to their cellular shape,
and the distance from the periphery of the
trophoblast spread. The results are presented in
Table 1. Almost all the macrophages in contact
with the spreading trophoblast cells, were rounded
(Table 1); no elongated macrophages were found
inside the blastocysts. While the macrophages
lying in close vicinity of the embryos (less than 20
ym from the edge of the trophoblast spread)
tended to assume the rounded morphology, the
difference in the relative frequencies of the round-
ed cells, from those in other areas was not statisti-
cally significant.

2. Drug-induced rounding of macrophages and
production of LTC, in vitro

In order to understand the mechanisms under-
lying the rounding of the macrophages, induced
when they are in contact with the embryonic cells,
we tested the effect of various compounds which
are known to affect the cytoskeletal organization
of the cells. Furthermore, the rate of production
of LTC, in the phagocytes under the influence of
those compounds was examined.

TABLE 1.

In Figure 2A-D, the photomicrographs show the
effects of colchicine, cytochalashin B and D upon
the cellular shape of highly purified peritoneal
macrophages of the mouse. All the three com-
pounds induced strong rounding of the cells at the
concentrations indicated in the legends to the
figures. Tuftsin, an activator of macrophages, had
little effects upon the morphology of the cells (Fig.
2E).

In Figure 3A, the relative contents of the mac-
rophages of the rounded and the elongated mor-
phology in the culture, are presented according to
the concentrations of colchicine added. At the
concentrations above 5 g/ml, all the macrophages
were seen rounded. The rates of LTC, production
in the same cell populations are shown in Figure
3B. Colchicine affected little the rates which
remained unchanged after the addition of the
compound to the cells, and stayed at a low level
throughout the range of the colchicine concentra-
tions examined.

Cytochalasins, both B and D, at sufficiently high
concentrations, resulted in the complete rounding
of the phagocytes in the culture (Fig. 4A, 5A).
The rate of production of LTC,, however, re-
mained at the control level throughout the range of
concentrations of the compounds examined (4B,
SB).

Distribution of macrophages with either elongated or rounded cellular shape in blastocyst-

macrophage co-culture of allongeneic combinations”

2 No. of Rounded Macrophages Elongated Macrophages
Receoiacs Exper No. of Cells Me Eve No. of Cell ROEING
pose iments ; Frequency eee Frequency
cells/embryo” % cells/embryo %
On the Zona Pellucida 6 54.0+ 13.7 100.0 0.0+ 0.0 0.0
cells/embryo % cells/embryo %
Within 20 4m from 12 39.5+ 9.6 75.8 12.6+ 3.9 24.2
Edge of Trophoblast Spread
cells/0.1 mm? % cells/0.1 mm? %
Areas Surrounding the 12 78.8+25.0 61.5 49.0+11.2 38.5

Embryo (Further than
20 wm from the Edge)

) Blastocysts were obtained from BALB/c mice, and macrophages were collected from C3H/HeJ mice.
») Only the number of macrophages on a hemisphere of the zona facing toward the observed through the lenses,
was actually counted. The values obtained for the hemisphere were multiplied by a factor of 2, and presented as

number of cells per embryo.

240 C. TACHI AND S. TACHI

J ” 4 e 2 . e
¢ om my Fe x
P x a)
o Py , e .” a
° , s y »* ¢
\ ite é e g
e
sf » \ = a
. ’
asf ~ ‘ e e
a
at af oo oe a
A . 20pm B : Y *é 4
a e : 3
2 r bg e . 2 = ‘e as 4
& e e 2
e © e ee 4
e © se e
e - ‘, é : e
& ® Py) o
r be ® ee e
t e e
@ ; ~ e °

if * »
e a
y
&
@ oe
~ > ‘
e
. a ee
. J

ee

Fic. 2. Photomicrographs showing the effects of druge which affect the cytoskeletal organization of cells, upon the
morphology of macrophages in vitro. A) Control with no additions; B) colchicine (500 “ng/ml); C) cytochalasin
B (500 ng/ml); D) cytochalasin D (500 ng/ml); E) Tuftsin (60 nM/ml).

noticeable changes in the morphology of these
cells (Fig. 6A). It did not cause significant changes
in the rate of LTC, production, except at 60 nM

Tuftsin, a polypeptide known to be a stimulator where slight but significant elevation in the rate
of macrophage activities [33-41], did not induce —_ was observed (Fig. 6B).

3. Effects of tuftsin upon the cellular morphology
and the rate of production of LTC,

Blastocyst-Macrophage Co-Culture 241

»|A
100
®
C )
2)
o 50
(S)
Ld |
[|
ee
0 1.0 10.0 100.0

ng/10° cells /hr

po/ml

Colchicine

——————————————E——————

vz
(S}
® 10.0
e
®
‘<
-
°
x
s
o
5.0
e
0 1.0

Fic. 3.

10.0 100.0 acu

Effects of colchine upon the macrophages cultured in vitro. A) Morphology. @, rounded cells; m@, elongated

cells. B) The rate of production of leukotriene C, assayed as the total amount of cysteinyl leukotrienes released

into the medium (see Materials and Methods).

DISCUSSION

Present report demonstrated that in the blasto-
cyst-macrophage co-culture of the mouse, two
major morphologically distinct groups of mac-
rophages, i.e., those which were rounded, and the
others which were elongated, are present, and that
the phagocytes in direct contact with the

embryonic cells, regardless of whether they were
trophoblast or ICM cells, assumed invariably
rounded cellular shape. Evidence was presented,
furthermore, indicating that the rate of synthesis of
cysteinyl leukotrienes in the rounded macrophages
is probably low, and remains at the level of un-
stimulated macrophages.

Fauve et al. [42], as a part of their studies upon

242

Cells

C. TACHI AND S. TACHI

SS ee

ng/10°cells/hr

=z
(S)
10.0
o
e
o
i
a
CS)
x
3
r7)
JS 5
e
e
0.0

Fic. 4.

1.0 10.0 palmi
Cytochalasin B
5 e

1.0 10.0 pesmi

Cytochalasin B

Effects of cytochalasin B upon the macrophages cultured in vitro. A) Morphology. @, rounded cells; Ml,

elongated cells. B) The rate of production of leukotriene C, assayed.as the total amount of cysteinyl leukotrienes
released into the medium (see Materials and Methods).

the anti-inflammatory effects of murine malignant
teratocarcinoma cells, examined the interactions
between the trophoblast cells and the macrophages
cultured together in vitro. They noted [42] that in
the vicinity of the trophoblast cells, the mac-
rophages were unable to spread but became necro-
tic. They suggested that the trophoblast cells, like
the teratocarcinoma cells, might escape the im-
munological surveillance of the host by exerting a

direct cytotoxic effect on macrophages, and by
releasing a hypothetical inhibitor of inflammatory
reactions.

We could not convincingly observe, however,
the evidence for explicit cytotoxic influence of the
trophoblast cells upon the macrophages.

The peritoneal macrophages used by Fauve et al.
[42] were of considerably high purity [43]. Howev-
er, possibility cannot be entirely excluded that the

Blastocyst-Macrophage Co-Culture 243

100.0

Cells

ane See I ae

G.0 1.0

10.0 pe/mi

Cytochalasin D

ng/10° cells /hr

zs

(S)
10.0

o

(S

o

‘x

7

°

x

3

o

a 5.0
0.0

10.0 po/mi

Cytochalasin D

Fic. 5.

Effects of cytochalasin D upon the macrophages cultured in vitro. A) Morphology. @, rounded cells; m,

elongated cells. B) The rate of production of leukotriene C, assayed as the total amount of cysteiny] leukotrienes
released into the medium (see Materials and Methods).

necrotic cells were not macrophages but lympho-
cytes which adhered non-specifically to the culture
dishes, and contaminated the macrophage prepa-
ration.

We, too, occasionally observed in the blastocyst-
macrophage co-culture, cells which resembled
dead macrophages and were ingested by the
trophoblast cells [4]. Although we could not
definitely identify the type of the ingested cells, we

considered it rather unlikely that these cells repre-
sented the macrophages actively killed and ing-
ested by the trophoblast cells [4].

Under the experimental conditons we em-
ployed, colchicine, cytochalasin B and D induced
strong rounding of the cellular shape in the mac-
rophages.

According to the observations reported earlier
by White et al. [12], while colchicine caused round-

244 C. TACHI AND S. TAcuHi

|

%
100 A
2)
= 50
ao
oO
2
0 10.0

ng/10° cells /hr

=z
Oo

10.0
o
¢
o
<=
-_
°
x
Ss
o

— 5.0

eel |
0 10.0

100.0

Tuftsin

100.0 nM

Tuftsin

Fic. 6. Effects of tuftsin upon the macrophages cultured in vitro. A) Morphology. @, rounded cells; m, elongated
cells. B) The rate of production of leukotriene C, assayed as the total amount of cysteinyl leukotrienes released

into the medium (see Materials and Methods).

ing of the cultured macrophages, cytochalasin B
had little effect upon the morphology of the phago-
cytes; both of the drugs were added to the cells ata
concentration of 10° M. The cause for the discre-
pancy between our results and those described by
White ef al., is not clear.

Macrophages are one of the major sources of
leukotrienes in the mammalian body, and these

metabolites of arachidonic acid are known to
mediate variety of pathological conditions, includ-
ing inflammation, allergy, asthma etc. (for reviews
see [20-25]). While the production of LTC, is
dependent upon the increased intracellular levels
of calcium, macrophages contain a cytoskeletal
protein, gelsolin, the activity of which is regulated
by calcium [44-46]. Indeed, Ca ionophore added

Blastocyst-Macrophage Co-Culture 245

to macrophages at alkaline pH’s of the medium,
induced strong elongation of the macrophages,
while the rate of leukotriene C, production in-
creased approximately 100-fold [C. Tachi and Zor,
Zool. Sci., in press].

Since, as stated in Introduction, the disturbance
of cytoskeletal organization in the phagocytes by
drugs, results in the increased rate of release of
various molecules of biological activity from the
cells, it is pertinent to ask if the morphological
changes of macrophages induced by the contact
with the embryos, or by the drugs, might results in
the modified rate of synthesis of leukotrienes.

Our results presently described, however, clear-
ly demonstrated that the macrophages of the
rounded morphology induced by the compounds
which affect the cytoskeletal organization of the
cells, were inactive with regard to the release of
the cysteinyl leukotrienes.

On the basis of those findings, we would like to
tentatively propose, as a working hypothesis, that
in the blastocyst-macrophage co-culture, the rate
of production of LTC, in the rounded mac-
rophages, may remain at a low level, while in the
elongated ones the rate might be enhanced [C.
Tachi and U. Zor, Zool. Sci., in press].

However, the drug-induced transformations of
the cellular morphology may not be immediately
comparable to those caused in the co-culture,
under the influence of the embryos. Therefore, it
remains to be investigated, if the elongated and the
rounded macrophages in the co-culture, are in fact
synthesizing LTC, at different rates.

Tuftsin, a naturally occurring tetrapeptide de-
rived from Fe segment of IgG [33-36], has been
shown to specifically bind to macrophages [38],
augment phagocytosis [35, 36, 38] and triggers the
antigen-specific, macrophage-dependent educa-
tion of T-cells [37]. The peptide appeared to raise
slightly the rate of LTC, synthesis around the
concentration of 60 nM or above.

Gupta et al. [26] proposed that LTC, might
mediate the initial estrogen-dependent phase
(phase I) of peri-nidatory changes of the endomet-
rium in the mouse. Experimental evidence in
support of the hypothesis implicating the role of
prostaglandins [47-51] and/or leukotrienes [52-55]
in the process of decidualization, has been re-

ported in the literature.

While it is tempting to propose that the major
source of LTC, might be one of the functionally
differentiated groups of macrophages which abun-
dantly emerge, as described originally by ourselves
[1, 3], in the endometrium during the early phase
of implantation, further analytical work on the
functional as well as the morphological aspects of
the blastocyst-macrophage interactions in vitro is
necessary to corroborate the view.

ACKNOWLEDGMENTS

The authors wouls like to thank Prof. U. Zor and Dr
F. Kohen, Dept. Hormone Research, Weizmann Insti-
tute of Science, Rehovot, Israel, for valuable discussions
and for the generous gift of the monoclonal antibodies
against cysteinyl leukotrienes.

The work is supported in part by a grant-in aid (No.
63640004) to C. Tachi, for scientific research on priority
areas “Molecular Mechanisms Underlying the Mainte-
nance of Germ-Lines in Animals and Man”, from the
Ministry of Education, Science, Culture, Japan.

REFERENCES

1 Tachi, C., Tachi, S., Knyszynski, A. and Lindner,
H. R. (1981) Possible involvement of macrophages
in embryo-maternal relationships during ovum im-
plantation in the rat. J. Exp. Zool., 217: 81-92.

2 Tachi, C. (1985) Mechanisms underlying regulation
of local immune responses in the uterus during early
gestation of eutherian mammals. I. Distribution of
immuno-competent cells which bind anti-IgG anti-
bodies in the post-nidatory uterus of the mouse.
Zool. Sci., 2: 341-348.

3 Tachi, C. and Tachi, S. (1986) Macrophages and
implantation. Ann. New York Acad. Sci., 476: 158-
182.

4 Tachi, S., Tachi, C. and Knyszynski, A. (1985)
Mechanisms underlying regulation of local immune
responses in the uterus during early gestation of
eutherian mammals. II. Electron microscopic stu-
dies in the interactions between blastocysts and
macrophages cultured together in vitro. Zool. Sci.,
2: 671-680.

5 Carr, K. and Carr, I. (1970) How cells settle on
glass: a study by light and scanning electron micros-
copy of some properties of normal and stimulated
macrophages. Z. Zellforsch., 105: 234-241.

6 Carr, I. (1973) The Macrophage. A Review of
Ultrastructure and Function. Academic Press, New
York.

7 Rabinovitch, M. and DeStephano, M. J. (1973)

11

12

13

14

15

16

246

Macrophage spreading in vitro. I. Inducers of
spreading. Exptl Cell Res., 77: 323-334.
Rabinovitch, M. and DeStephano, M. J. (1973)
Macrophage spreading in vitro. II. Manganese and
other metals as inducers or as co-factors for induced
spreading. Exptl Cell. Res., 79: 423-430.

Cohn, Z. A. (1975) Macrophage physiology. Fed.
Proc., 34: 1725-1729.

Onozaki, K., Homma, Y., Miura, K. and Hashimo-
to, T. (1981) Scanning electron microscopic study of
changes in the surfaces of lymphokine-treated
guinea pig macrophages. J. Reticuloendoth. Soc.,
30: 471-481.

Gordon, S. and Werb, Z. (1976) Secretion of
macrophage neutral proteinase is enhanced by col-
chicine. Proc. Natl. Acad. Sci. U.S.A., 73: 872-876.
White, R. R., Leon, I. and Kuhn, C., III. (1981)
Effect of colchicine, vinblastin, D,O and cytochala-
sin B on elastase secretion, protein synthesis and
fine structure of mouse alveolar macrophages. J.
Reticuloendoth. Soc., 29: 295-305.

Stocic-Grujicic, S. and Simic, M. M. (1982) Mod-
ulation of interleukin 1 production by activated
macrophages: in vitro action of hydrocortisone, col-
chicine, and cytochalasin B. Cell. Immunol., 69:
235-247.

Zor, U., Her, E., Talmon, J., Kohen, F., Harell,
T., Moshonov, S. and Rivnay, B. (1987) Hydrocor-
tisone inhibits antigen-induced rise in intracellular
free calcium concentration and abolishes leuko-
triene C, production in leukemic basophils. Prostag-
landins, 34: 29-40.

Bach, M. K. and Brashler, J. R. (1978) Ionophore
A23187-induced production of slow reacting sub-
stnce of anaphylaxis (SRS-A) by rat peritoneal cells
in vitro: evidence for production by mononuclear
cells. J. Immunol., 120: 998-1005.

Bach, M. K., Brashler, J. R., Brooks, C. D. and
Neerken, A. J. (1979) Slow reacting substances:
comparison of some properties of human lung SRS-
A and two distinct fractions from ionophore-induced
rat mononuclear cell SRS. J. Immunol., 122: 160-
165.

Rankin, J. A., Hitchcock, M., Merrill, W., Bach,
M. K., Brashler, J. R. and Askenase, P. W. (1982)
IgE-dependent release of leukotriene C, from alveo-
lar macrophages. Nature, 297: 329-331.
Sammuelsson, B., Borgeat, P., Hammerstorm, S.
and Murphy, R. C. (1979) Introduction of a
nomenclature: leukotrienes. Prostaglandins, 17:
785-787.

Samuelsson, B. and Hammerstrom, S. (1982)
Leukotrienes: a novel group of biologically active
compounds. Vitam. Horm., 39: 1-30.

Piper, P. J. (1983) Pharmacology of leukotrienes.
Brit. Med. Bull., 39: 255-259.

21

22

23

24

25

27

28

29

30

31

32

33

C. TACHI AND S. TACHI

Piper, P. J. (1984) Formation and actions of leuko-
trienes. Physiol. Rev., 64: 744-761.

Lewis, R. A. and Austen, K. F. (1984) The biologi-
cally active leukotriences. J. Clin. Invest., 73: 889-
897.

Ford-Hutchinson, A. W. (1985) Leukotrienes: their
formation and role as inflammatory mediators. Fed.
Proc., 44: 25-29.

Salmon, J. A. and Higgs, G. A. (1987) Prostaglan-
dins and leukotrienes as inflammatory mediators.
Brit. Med. Bull., 43: 285-296.

Piper, P. J. and Samhoun, M. W. (1987) Leuko-
trienes. Brit. Med. Bull., 43: 445-459.

Gupta, A., Huet, Y. M. and Dey, S. K. (1989)
Evidence for prostaglandins and leukotrienes as
mediators of phase I of estrogen action in implanta-
tion in the mouse. Endocrinology, 124: 546-548.
Tachi, C. and Zor, U. (1986) Possible correlation
between the production of leukotriene C4 and cytos-
keletal organization in mouse peritoneal mac-
rophages cultured in vitro. Zool. Sci., 3: 1011.
Tachi, C. (1988) Cellular mechanisms of the interac-
tions between blastocysts and macrophages co-
cultured in vitro. Proc. 2nd Annual Meeting, Japan
Society for Basic Reproductive Immnology. Ed. by
O. Tanizawa, pp. 71-75.

Tachi, C., Tachi, S. and Zor, U. (1988) Role of
macrophages in local immune responses elicited by
implanting blastocysts during early gestation in the
endometrium of muridae rodents. In “Human Re-
production, Current Status/Future Prospect”. Ed.
by R. Iizuka and K. Semm, Elsevier Science Publ.
B. V., Amsterdam, pp. 519-521.

Spindle, A. I. and Pedersen, R. A. (1973) Hatch-
ing, attachment, and outgrowth of mouse blastocysts
in vitro: fixed nitrogen requirements. J. Exp. Zool.,
186: 305-318.

Biggers, J. D., Whitten, W. K. and Whittingham,
D. G. (1971) The culture of mouse embryos in vitro.
In “Methods in Mammalian Embryology”. Ed. by J.
C. Daniel, Jr, W. H. Freeman & Co., San Francis-
co, pp. 86-116.

Danzlinger, C., Guhlmann, A., Scheuber, P. H.,
Wilker, D., Hammer, D. K. and Keppler, D. (1986)
Metabolism and analysis of cysteinyl leukotrienes in
the monkey. J. Biol. Chem., 261: 15601-15606.
Fidalgo, B. V. and Najjar, V. A. (1967) The
physiological role of the lymphoid system. III.
leucophilic-globulin and the phagocytic activity of
the polymorphonuclear leucocytes. Proc. Natl.
Acad. Sci. U.S.A., 57: 957-964.

Fidalgo, B. V., Katayama, Y. and Najjar, V. A.
(1967) The physiological role of the lymphoid
system. V. The binding of autologous (erythrophi-
lic)-globulin to human red blood cells. Biochemis-
try, 6: 3378-3385.

35

36

37

38

39

40

41

42

43

44

45

46

Blastocyst-Macrophage Co-Culture

Najjar, V. A. and Nishioka, K. (1970) “Tuftsin’: a
natural phagocytosis stimulating peptides. Nature,
228: 672-673.

Nishioka, K., Constantopoulos, A., Satoh, P. S. and
Najjar, V. A. (1972) The characterization, isolation
and synthesis of the phagocytosis stimulating pep-
tide tuftsin. Biochem. Biophys. Res. Commun.., 47:
172-179.

Tzehoval, E., Segal, S., Stabinsky, Y., Fridkin, M.,
Spirer, Z. and Feldman, M. (1978) Tuftsin (and
Ig-associated tetrapeptide) triggers the im-
munogenic function of macrophages: implications
for activation of programmed cells. Proc. Natl.
Acad. Sci. U.S.A., 75: 3400-3404.

Bar-Shavitt, Z., Stabinsky, Y., Fridkin, M. and
Goldman, R. (1979) Tuftsin-macrophage interac-
tion: specific binding and augmentation of phagocy-
tosis. J. Cell. Physiol., 100: 55-62.

Nishioka, K., Amoscato, A. A. and Babcock, G. F.
(1981) Tuftsin: a hormone-like tetrapeptide with
antimicrobial and antitumor activities. Life Scien-
ces, 28: 1081-1090.

Fridkin, M. and Gottlieb, P. (1981) Tuftsin, Thr-
Lys-Pro-Arg. Mol. Cell. Biochem., 41: 73-97.
Gottlieb, P., Tzehoval, E., Feldman, M., Segal, S.
and Fridkin, M. (1983) Peptide fragments from the
tuftsin containing domain of immunoglobulin G
synthesis and biological activity. Biochem. Biophys.
Res. Commun., 115: 193-200. ;
Fauve, R. M., Hevin, B., Jacob, H., Gaillard, J. A.
and Jacob, F. (1974) Anti-inflammatory effects of
murine malignant cells. Proc. Natl. Acad. Sci.
U.S.A., 71: 4052-4056.

Fauve, R. M. (1964) Resistance cellulaire a l’infec-
tion bacterienne. Ann. Inst. Pasteur, 107: 472-483.
Stossel, T. P. and Hartwig, J. H. (1976) Interactions
of actin, myosin, and a new actin-binding protein of
rabbit pulmonary macrophage. J. Cell Biol., 68:
602-612.

Yin, H. L. and Stossel, T. P. (1979) Control of
cytoplasmic actin gel-sol transformation by gelsolin,
a calcium dependent regulatory protein. Nature
(Lond.), 281: 583-586.

Stossel, T. P., Hartwig, J. H., Yin, H. C. Davies,
W. A. (1979) Actin-binding protein. In “Cell Motil-
ity”. Ed. by S. Hatano, H. Ishikawa and H. Sato.
Univ. Tokyo Press, Tokyo, pp. 189-210.

47

48

49

50

51

52

53

54

55

247

Tachi, C. and Tachi, S. (1974) Cellular aspects of
ovum implantation and decidualization in the rat. In
“Physiology and Genetics of Reproduction”, Part B.
Ed. by E. M. Coutinho and F. Fuchs. Plenum Press,
New York, pp. 263-286.

Sananes, N., Baulieu, E. E. and Gascogne, C.
(1976) Prostaglandin(s) as inductive factor of de-
cidualization in the rat uterus. Mol. Cell. Endocri-
nol., 6: 153-158.

Rankin, J. C., Ledford, B. E., Jonsson, H. T., Jr.
and Baggett, B. (1979) Prostaglandins, indometha-
cin and the decidual cell reaction in the mouse
uterus. Biol. Reprod., 20: 399-404.

Kennedy, T. G. (1977) Evidence of a role for
prostaglandins in the initiation of blastocyst im-
plantation in the rat. Biol. Reprod., 16: 286-291.
Kennedy, T. G. (1986) Prostaglandins and uterine
sensitization for the decidual cell reaciton. Ann. N.
Y. Acad. Sci., 476: 43-48.

Pakrasi, P. L., Becka, R. and Dey, S. K. (1985)
Cyclooxygenase and lipoxygenase pathways in the
preimplantation rabbit uterus and blastocyst. Pros-
taglandins, 29: 481-495.

Holmes, P. V., Lindenberg, S., Hellberg, P. and
Janson, P. O. (1987) Evidence for the involvement
of lipoxygenase products during blastocyst implanta-
tion in the mouse. (Abstract). Human Reproduc-
tion, special issue, Abstracts from the 3rd Meeting
of the European Society of Human Reproduction
and Embryology, p. 87.

Malathy, P. V., Cheng, H. C. and Dey, S. K. (1986)
Production of leukotrienes and prostaglandins in the
rat uterus during peri-implantation period. Prosta-
glandins, 32: 605-604.

Tawfik, O. W., Sagrillo, C., Johnson, D. C. and
Dey, S. K. (1987) Decidualization in the rat: role of
leukotrienes and prostaglandins. Prostaglandins
Leukotrienes and Medicine, 29: 221-228.

Note added in proof. The paper cited in the text as C.
Tachi & U. Zor in press, has since been published as
follows;

Tachi, C. and Tachi, S. (1989) Effect of calcium
ionophore A23187 upon the rate of leukotriene C,
production and the cellular morphology in highly purified
mouse peritoneal macrophages cultured in vitro. Zool.
Sci., 6: 251-257.

‘
Es

0 ut (eee (ate ry

Mn fo" BAP al aditeyilantsis

- We tPF ek) a

1 Age. other
Teper (OTT) ee bh Lay

MAD LAGS fo we
a> : al w 4 Tet d
)Oaltitty Speedie Arlee fey !
wi) 7 4th aera ra oe ete by "
liwm(iee tL (ae ue rk td
1 SGN Neem? Ds 4 mera
viel gt $ vera s, ¥
i |. oad! t oat
simi bide ©
bie breed) | Mela
ao pe A | ' ‘
z a 7
xs iW ne
tel M4 A witht Fat “erat oh

WL ret jpde > ay
LAOS tid, Reanetithert pieve
fr Crea are yiebboedhy 'T

weet Be Ohya

Wile, - se AG Fee Y aoe
* U 4 woHtlined TW hs eonreau

1 NSD ATA ae ORE ary:

Preah icy? oot i 4) Wlrbotgtiuychic tf geal
whet és ; eels perma. Wiebe
in Hie wea . : ‘ mee
Bt vs: Te A Da? ere g |
4 cpaargt f <eyite! bn ion Fitthacricet FRA!
godess’ S° WAS kvP Oecd Se Fae
zi 1th LAMSON etal se eal eee ee
ree mY my LT TWP USA aeRO
st i f-t mer shpsind (ate) 4k til
. v cy Fgrihies aifab Won 6 Lows
Mel ee ee ra Sfanihinonth.¢
> Senwy Oh Listitie 'F aren or ae
iehvy (4 Golinetiidied Wea :

, aire rusteliy st Joeheiaeg

ts Hiteegae kag
1.) Ae
wilt) eel 1 f i tha vada

ae i a ots

cou Caevieee
4 ; - ‘Fee.

ZOOLOGICAL SCIENCE 7: 249-256 (1990)

Effects of Thyroxine on Locomotor Activity and Carbon
Dioxide Release in the Toad, Bufo japonicus

YoKo TASAKI and Susumu IsHiII

Department of Biology, School of Education, Waseda University,
Tokyo 169, Japan

ABSTRACT—To elucidate a role of thyroxine (T,4) in toad migration during the breeding season, we
studied effects of administration of T, on locomotor activity and CO, release using normal and
thyroidectomized adults of both sexes which were captured before hibernation. All the experiments
were performed during the breeding season under laboratory conditions. Locomotor activity was
estimated by passage of infrared beams in an activity box and CO, release was estimated comparing CO,
contents in inflow air to the activity box and outflow air from the activity box. Locomotor activity was
significantly suppressed (to 9%) by T, treatment (10 ~g/day for 2 weeks) in intact males. Ty, also
suppressed CO, release in intact males at moving states, while T, enhanced CO, release at resting states.
T, had no effect on either locomotor activity or CO, release in female toads. Thyroidectomy in males
resulted in a 3-fold increase in locomotor activity and T, administration suppressed activity in
dose-dependent fashion (1 to 100ng/g BW/day for 3 weeks). Neither thyroidectomy nor T,
administration had any effect in females. These results suggest that thyroxine can not be the factor

© 1990 Zoological Society of Japan

which induces breeding migration to the pond in the toad.

If the sedative effect of thyroxine is

physiological, it is probable that thyroxine initiates the post-breeding inactive stage.

INTRODUCTION

Thyroxine has been shown to play some role in
the migration of lower vertebrates as reported in
sticklebacks [1], young salmonids [1-4], tiger sala-
manders [5] and red-spotted newts [6]. Dent [6]
provided evidence suggesting that thyroxine initi-
ates the migration of post-breeding adult newts
from water to land. One of the other proposed
major roles of thyroxine is control of energy
metabolism. In adult anurans, thyroid hormone
administration stimulated the O, consumption of
the animal [7-10] and glycogen metabolism in the
liver [7, 11, 12].

We surveyed the annual cycles of plasma thyroid
hormone levels in the toad, Bufo japonicus, and
theorized two possible roles of thyroid hormone in
the toad in winter and early spring [13]. They are
initiation of migratory movement to and/or from
the pond and regulation of energy metabolism at
low temperatures during the breeding season.

Accepted April 28, 1989
Received February 16, 1989

In the present experiment, we observed effects
of administration of thyroxine on locomotor ac-
tivity and CO, release to examine a relation be-
tween circulating thyroxine and breeding migra-
tion.

MATERIALS AND METHODS

Material

Adult male and female toads (Bufo japonicus)
were captured in the suburbs of Tokyo in October
and November, 1984 (Experimen I) and 1986
(Experiment II). The mean of their body weights
and standard error was 203.8+9.5 g in Exp. I, and
127.4+6.2 g in Exp. II. Males and females were
put in separate plastic boxes (55 x 40 x 43 cm) with
loose fitting tops and kept outdoors. Wet pieces of
plastic sponge were put in the boxes with the toads
to maintain humidity. No feeding took place, since
toads abstain from food during winter and spring.

Design of Experiment I

Seven females and nine males were used. Ten
pg of L-thyroxine (SIGMA) suspended in saline

250 Y. TASAKI AND S. ISHII

was injected daily for two weeks into the dorsal
lymphsacs of four males and four females. The
remaining three females and five males received
injections of saline alone and served as controls.
All the injections were performed between 0900
and 1100 hr. Locomotor activity and CO) release
were measured for 18 hr from 1500 to 0900 hr the
next morning during the period between March
3rd and 28th, 1985.

Design of Experiment II

Twenty-nine female and twenty-four male toads
were used. Twenty-three females and nineteen
males were thyroidectomized under anestheti-
zation with MS-222 two weeks before the start of
thyroxine treatments. A part of the hyoid cartilage
was also removed with the thyroid. The remaining
six females and five males were sham-operated.
The thyroidectomized females were divided into
four groups of 5, 6, 6, and 6, and received daily
injections of 0, 0.001, 0.01, and 0.1 ug/g body
weight/day of L-thyroxine in 0.1 ml of saline, into
the dorsal lymphsac. The injections were per-
formed once a day between 1000 and 1200 hr for
three weeks until the day before the locomotor
activity measurement. Thyroidectomized males
were also divided into four groups of 5,5, 4, and 5,
and received the same injections as females. Loco-
motor activity of each toad was measured for 20 hr
from 1200 to 0800 hr the next morning during the
period between February 9th and March 2nd,
1987.

Recording of locomotor activity and CO; release in
Experiment I

A small plastic chamber, 42 cm long, 20 cm wide
and 15 cm deep (Fig. 1) was used to measure both
the locomotor activity and CO) release of a toad
simultaneously and automatically. Each toad was
kept in the chamber for 21 hr (1200 to 0900 hr the
next morning). After an initial three-hour accli-
mation, the locomotor activity and CO, release
were continuously recorded for 18hr. The
temperature of the chamber was regulated at 10+
1°C. The chamber was illuminated from 0600 to
1800 hr, and kept in darkness the remaining hours.
To quantify the locomotor activity of the toad in
the chamber, seven pairs of photosensor uints

Fic. 1. The chamber used in Experiment I in order to
record both the locomotor activity and the CO,
release of a toad simultaneously and automatically.
See text for details.

were mounted on the longitudinal side walls of the
chamber at 5cm intervals, 2cm from the floor
(Fig. 1). Each photosensor unit consisted of a
infrared LED lamp (TLN 110, Toshiba) and a
photodiode (TPS 703A, Toshiba). Interruption of
the infrared light was recorded for each photosen-
sor unit separately at 5 second intervals, and the
records were stored in the memory of an 8 bit
personal computer (NEC PC-8001, Fig. 2). Thus,
the longitudinal position of the toad was recorded
every 5 sec with the precision of +2.5cm. At the
end of each experiment, data in the memory were
transferred to a floppy disk. The total distance of
locomotion was calculated later by the same com-
puter.

Carbon dioxide released from a toad placed in
the chamber was quantified as follows. The inflow
air tube was divided into three parts and had 1.0
cm openings on the wall of one of the longitudinal
ends of the chamber. The outflow air tube was
connected similarly to the openings on the wall of
the other end. Inflow air, which had been col-
lected from the outdoors and stored in a balloon ,
was pumped into the chamber at the flow rate of
5.01 per min. It was humidified by being passed
through a water filter inserted between the balloon
and the chamber. Air in the chamber was circu-
lated by two small, slowly-rotating electric fans
(RF-510T, Mabuchi) which were installed on a
wall of the chamber (Fig. 1). The outflow air was
channeled into an open-flow infrared gas analyzer
(VIA-300, Horiba), and the CO, concentration
was determined (Fig. 2). At the same time, part of

Thyroxine Effect on Toad Locomotion 251

MICRO COMPUTER
(PC-8001)

1/0 PORT

CONVERTER

GAS ANALYZER

DATA FILE

(PC-803 1)

WATER FILTER

¢— Infrared LED lamp

Oo Photodiode

PRINTER

Fic. 2. Diagram showing the recording system in Experiment I.

the inflow air was introduced into the analyzer
through a bypass, and its CO, concentration was
also determined. From the difference in CO,
concentration between the inflow and outflow air
and the flow rate, the release of CO, from the toad
was calculated. The mean CO, release when the
toad stayed immobile at least one hour was refer-
red to as the basal CO) release (Fig. 5). The mean
difference between the active phase CO) release,
which is the CO; release when the toad is moving,
and the basal CO, release was referred to the
activated CO, release (Fig. 5). The activated CO,
release can be regarded as the rise in CO) release
caused by locomotion.

Recording of locomotor activity in Experiment II

In this experiment, only the locomotor activity
was measured. The chamber used had dimensions
of 303015 cm (Fig.3). The position of the
toad in the chamber was recorded two-dimension-
ally by eight photosensor units. Each of the four
walls was mounted with two infrared LED lamps
and two photodiodes which were arranged recipro-
cally at 6 cm intervals. Their height from the floor
was 2 cm on two opposing walls, and 4 cm on the
other two. The air temperature and humidily of
the chamber were regulated at 9.3+0.7°C and 54
+3%, respectively. The chamber was illuminated

252 Y. TASAKI AND S. Ison

Fic. 3. The chamber used for recording the locomotor
activity of a toad in Experiment II. The position of
the toad in the chamber is recorded two dimen-
sionally. See text for details.

40

30

20

Locomotion (m/18hr)

10

| 4 |
Saline T.
Males

Fic. 4. Total locomotion distances (open circles) of
thyroxine and saline-injected normal toads of both
sexes for 18 hr. The column and vertical bar indicate
the mean and standard error of each group, respec-
tively. The mean of the thyroxine injected male
group is significantly lower than that of the control
group. No significant difference was observed be-
tween the female groups (p>0.05) when given the
randomization test (*p=0.0317 by the randomiza-
tion test).

Saline T4
Females

from 0630 to 1730hr, and kept in darkness the
remaining hours. Data were recorded and ana-
lyzed as in Experiment I.

Statistical methods The significant difference
between the means of the two groups was deter-
mined by the randomization test in Experiment I.
The one-way matrix analysis of variance followed
by Duncan’s multiple range test was used in Ex-
periment II. For these tests, computer programs
[14] were employed.

RESULTS

Experiment I

The locomotor activity (total distance of

Locomotion (cm/5min)

CO2 release (ug/g/min)

22 24 02 04 06 os

Ad

Time

Fic. 5. Record of the locomotion distance (upper) and
CO, release (lower) of a thyroxine-injected female
toad (a typical case). Lights were turned on at 0600
hr (double arrow heads) and off at 1800 hr (single
arrow head). Note that the CO, release is synchro-
nized with the locomotion. In the lower figure, the
dotted area corresponds to the basal CQ; release
and the area above the dotted line corresponds to
the activated CO, release.

Thyroxine Effect on Toad Locomotion 253

locomotion) of toads varied individually over a
wide range (Fig. 4). Treatment with thyroxine
seemed to have no effect on the mean locomotor
activity of female toads, as the difference between
the means of the control and treated groups (7.22
+4,52 m and 10.36+8.28 m, respectively) was not
significant (p>0.05). However, in males, thy-
roxine suppressed activity, as the difference be-
tween the means of the control and treated groups
(13.16+6.24m and 1.14+1.61m, respectively)
was significant (p<0.05).

The change in CO) release faithfully coincided
with changes in locomotor activity (Fig. 5). In
females, there was no significant difference be-
tween activated CO, release of the control (377+
277 ng/g B.W./min) and treated (259+170 ng/g
B.W./min) groups (Fig. 6). In males, the activated

1000

800

fo)
oO
fo)

(ng/g B.W./min)

400

C02

200

14 Saline 14
Females

Fic. 6. The activated CO, release (open circles) of male

Saline
Males

and female toads. The column and vertical bar
indicate the mean and standard error of each group,
respectively. The mean of the thyroxine-injected
male group is significantly lower than that of the
control group (*p=0.0286 by the randomization
test).

CO, release in the control and treated groups was
532+175 ng/g B.W./min and 188+39 ng/g B.W./
min, respectively, and the difference was sig-
nificant (p<0.05, Fig. 6).

The basal CO release was higher in females
than in males. In females, it was not significantly
changed by thyroxine treatment (Fig. 7). In males,
however, the basal CO, release was significantly
increased by thyroxine treatment, up to or over the
levels of female toads (p<0.05, Fig. 7).

1000

800

600

400

CO2 (ng/g B.W./min)

200

14 Saline 14
Females

Saline
Males

Fic. 7. The basal CO, release of male and female
toads. The column and vertical bar indicate the
mean and standard error of each group, respective-
ly. The mean of the thyroxine-injected male group
was significantly higher than that of its control group
(*p=0.0286 by the randomization test).

Experiment II

In females, neither thyroidectomy nor thyroxine
administrations seemed to influence locomotor
activity, as the difference in mean activity among
the five groups was not significant when tested by
analysis of variance (p>0.05, Fig. 8). The group

254

Females
200 °

150

(m/20hr)

100

Locomotion

50

Sham Tx Tx Tx

Tx
10ng 100ng
14 /gB.W./day

Saline ing

Y. TASAKI AND S. Ison

Males

150

100

Locomotion (m/2Ohr)

50

Sham Tx
Saline

ing 10ng

T4 /9B.W./day

Fics. 8 and 9. Total locomotion distances (open circles) of sham-operated, thyroidectomized, and thyroidectomized
and thyroxine-treated female (Fig. 8, Left) and male (Fig. 9, Right) toads. The column and vertical bar indicate
the mean and standard error of each group, respectively. Thyroidectomized males showed a significantly higher

(p<0.01 when given Duncan’s multiple range test) locomotor activity than the shamoperated males.

Re-

placement therapy suppressed the activity significantly (p<0.01 for the highest dose and p<0.05 for the lowest
and middle doses when given Duncan’s multiple range test).

receiving the lowest does of thyroxine had a higher
mean activity level than the other groups, but this
could be within the range of random fluctuation.

In contrast to females, thyroidectomized males
showed significantly higher (p<0.01 by Duncan’s
multiple range test) locomotor activity than the
sham-operated males, increase being about three-
fold (Fig. 9). Replacement therapy suppressed the
activity to some extent or even to a subnormal
level depending upon the dose levels.

DISCUSSION

It is well known that prolactin is a factor which
induces migration of newts and salamanders from

land to water for breeding [15-20]. Recently
however, Yoneyama et al. [21], Ishii et al. [22] and
Yamamoto et al. [23] presented evidence showing
that prolactin can not be the factor inducing migra-
tion to the breeding pond, at least in Bufo. Our
survey of the annual cycle of plasma thyroid
hormone levels in the toad, Bufo japonicus, re-
vealed that the plasma thyroxine level increased
gradually during the inactive winter period and
reached a relatively high level at the commence-
ment of the breeding migration. From this
observation, we previously postulated that thyrox-
ine, instead of prolactin, is the factor which in-
duces breeding migration. However, in the pre-
sent study, we found that both endogenous and

Thyroxine Effect on Toad Locomotion 255

exogenous thyroxine suppressed the locomotor
activity of male toads in spring, but we failed to
show that effect in female toads. In either case, it
is difficult to suggest that thyroxine is a suitable
candidate for the migration inducing factor in the
toad.

Dent [6] proposed the hypothesis that thyroxine
causes the movement of terrestrial species of
amphibians from water to land after breeding.
Our recent finding [13] that the plasma thyroid
hormone level in the toad is remarkably elevated
when they arrive at the breeding pond strongly
supports Dent’s hypothesis. However, our present
finding showing the sedative effect of thyroxine on
locomotor activity is neutral to or may contradict
Dent’s hypothesis. This effect of thyroxine can
however, explain the commencement of the post-
breeding inactive period of the toad which lasts
until May or June. Recently, Kubokawa and Ishii
[24], surveying the annual cycle of various endo-
crine and metabolic parameters of the toad,
pointed out that among various hormones, only
thyroxine is secreted in the post-breeding inactive
period. Further study is needed to elucidate the
hormonal mechanism controlling the migration of
toads to and from the breeding pond.

From many years past, it has been repeatedly
reported that thyroxine stimulates O2 consumption
in whole animals [7, 8, 25] or liver slices in
amphibians [9, 10, 26] as well as in higher verte-
brates. In the present study, we observed that the
basal CO, release in the male toad was elevated by
thyroxine injection. This result coincides well with
previous reports on Oz consumption [7-9, 24, 25].
In contrast, the activated CO, release in the thy-
roxine-treated male toad was lower than in the
normal male toad. This may be due to decreased
intensity of locomotor activity caused by thy-
roxine.

The basal CO, release reflects the basal me-
tabolism. Accordingly, our results on basal CO,
release suggest that the enhancement of basal
metabolism by thyroxine is accompanied by a
decrease of muscular activity in male toads. This
reminds us of the old work that thyroxine leads to
the uncoupling of oxidative phosphorylation [27],
although this effect was shown to be nonphysio-
logical [28].

ACKNOWLEDGMENTS

The authors are grateful to Prof. Yasuyuki Oshima for
his valuable advice and suggestions and to Mr. Taroh
Ishii, Mitsubishi Kasei Co. Ltd for his assistance in
designing the recording system. This study was sup-
ported by a Grant-in-Aid from the Japanese Ministry of
Education, Science and Culture.

REFERENCES

1 Baggerman, B. (1962) Some endocrine aspects of
fish migration. Gen. Comp. Endocrinol. Suppl., 1:
188-205.

2 Baggerman, B. (1963) The effect of TSH and
antithyroid substances on salinity preference and
thyroid activity in juvenile Pacific salmone. Can. J.
Zool., 41: 307-319.

3 Nishioka, R. S., Young, G., Bern, H. A.,
Jochimsen, W., and Hiser, C. (1985) Attempts to
intensify the thyroxin surge in coho and king salmon
by chemical stimulation. Aquaculture, 45: 215-225.

4 Yamauchi, K., Ban, M., Kasahara, N., Izumi, T.,
Kojima, H., and Harako, T. (1985) Physiological
and behavioral changes occurring during smoltifica-
tion in the masu salmon, Oncorhynchus masou.
Aquaculture, 45: 227-235.

5 Norris, D. O., Duvall, D., Greendale, K., and
Gern, W. A. (1977) Thyroid function in pre- and
postspawning neotenic tiger § salamanders
(Ambystoma tigrinum). Gen. Comp. Endocrinol.,
33: 512-517.

6 Dent, J. N. (1985) Hormonal interaction in the
reguration of migratory movements in urodele
amphibians. “The endocrine system and the en-
vironment”, ed. B.K. Follet, S. Ishii, and A.
Chandola, Japan Sci. Soc. Press. Tokyo/Springer-
Verlag, Berlin. pp. 79-84.

7 Warren, M. R. (1940) Studies on the effect of
experimental hyperthyroidism on the adult frog,
Rana pipiens, Schreber. J. Exp. Zool., 83: 127-159.

8 Maher, M. J. (1967) Response to thyroxine as a
function of environmental temperature in the toad,
Bufo woodhousii, and the frog, Rana pipiens.
Copeia, 2: 361-365.

9 Packard, G. C., Packard, M. J., and Stiverson, R.
K. (1974) The influence of thyroxine on oxygen
consumption of tissues from the frog Rana pipiens.
Gen. Comp. Endocrinol., 22: 195-198.

10 Packard, G. C. and Packard, M. J. (1975) The
influence of acclimaion temperature on the meta-
bolic response of frog tissue to thyroxine adminis-
tered in vivo. Gen Comp. Endocrinol., 27: 162-168.

11 Lagerspetz, K. Y. H., Harri, M. N. E., and
Okslahti, R. (1974) The role of the thyroid in the
temperature acclimation of the oxidative metabo-

12

14

15

18

19

20

256

lism in the frog, Rana temporaria. Gen. Comp.
Endocrinol., 22: 169-176.

Kasprzyk, A. and Obuchowicz, L. (1980) The effect
of thyroxine and triiodothyronine on glucose-6-
phosphate and 6-phosphogluconate dehydrogenase
activity in liver and fat body of the frog, Rana
esculenta. Gen. Comp. Endocrinol., 42: 384-388.
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.

Ishii, S. (1983) “Programs of statistical methods for
biologists by N88-BASIC.” Baifukan, Tokyo.
Reinke, E. E. and Chadwick, C. S. (1940) The
origin of the water drive in Triturus viridescens. J.
Exp. Zool., 83: 223-233.

Chadwick, C. S. (1944) Further observations on the
water drive in Triturus viridescens. J. Exp. Zool.,
86: 175-187.

Crim, J. W. (1975) Prolactin-induced modification
of visual pigments in the eastern red-spotted newt,
Notophthalmus viridescens. Gen. Comp. Endocri-
nol., 26: 233-242.

Duvall, D. and Norris, D. O. (1977) Prolactin and
substrate stimulation of locomotor activity in adult
tiger salamanders (Ambystoma tigrinum). J. Exp.
Zool., 200: 103-106.

Duvall, D. and Norris, D. O. (1980) Stimulation of
terrestrial substrate preferences and locomotor
activity in newly transformed tiger salamanders
(Ambystoma tigrinum) by exogenous or endogenous
tyroxine. Anim. Behav., 28: 116-123.

Moriya, T. (1982) Prolactin induces increase in the
specific gravity of salamander, Hynobius retardatus,
that raises adaptability to water. J. Exp. Zool., 223:

Y. TASAKI AND S.

21

22

23

24

25

26

27

28

IsHI

83-88.

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.

Ishii, S., Yoneyama, H., Inoue, M., Yamamoto, K.,
and Kikuyama, S. (1989) Changes in plasma and
pituitary levels of prolactin in the toad, Bufo
japonicus, throughout the year with special refer-
ence to the breeding migration. Gen. Comp. Endo-
crinol., 74: 365-372.

Yamamoto, K., Kikuyama, S., and Ishii, S. (1989)
Homologous radioimmunoassay for plasma and
pituitary prolactin in the toad, Bufo japonicus Gen.
Comp. Endocrinol., 74: 373-376.

Kubokawa, K. and Ishii, S. (1989) Annual cycles in
various hormones in the toad, Bufo japonicus. Pro-
ceedings of the Japan Society for Comparative
Endocrinology, No. 3 (In press)

May, T. W. and Packer, R. K. (1976) Thyroid
hormones stimulate in vivo oxygen consumption of
adult Rana pipiens berlandieri at high environmental
temperatures. Gen Comp. Endocrinol., 30: 525-
527.

Packard, G. C. and Packard, M. J. (1973) Prelimi-
nary study of the influence of thyroxine, tempera-
ture, and sex on oxygen uptake by tissues from the
spadefoot toad Scaphiopus bombifrons. Gen.
Comp. Endocrinol., 20: 530-533.

Martius, C. and Hess, B. (1951) The mode of action
of thyroxin. Arch. Biochem. Biophys., 33: 486-487.
Tata, J. R., Ernster, L., Lindberg, O., Arrhenius,
E., Pedersen, S., and Hedman, R. (1963) The
action of thyoid hormones at the cell level.
Biochem. J., 86: 408-428.

ZOOLOGICAL SCIENCE 7: 257-263 (1990)

Intragranular Colocalization of Arginine Vasopressin- and
Angiotensin II-Like Immunoreactivity in the
Hypothalamo-Neurohypophysial System
of the Goldfish, Carassius auratus

Cuirumi YAMADA!, SHINoBU Nosi?*, SEvI SHIODA‘,

Yasumitsu Nakar’ and Hipesui KopayaAsmi

Research Laboratory, Zenyaku Kogyo Co., Ltd., Tokyo 178, *Department of
Biology, Faculty of Science, Toho University, Chiba 274, “Department of
Anatomy, Showa University School of Medicine,

Tokyo 142, Japan

ABSTRACT—In the preoptic nucleus (PON) of the goldfish, Carassius auratus, four types of cells were
observed under a light microscope: cells showing colocalization of immunoreactive angiotensin IT (ANG
II) and immunoreactive agrinine vasopressin (AVP), cells with only ANG II-like immunoreactivity,
cells with only AVP-like immunoreactivity and cells with neither immunoreactivity. Under an electron
microscope, only two types of nerve terminals were found in the neurohypophysis: those showing
immunoreactivity of both antisera and those with neither immunoreactivity. No terminals showing
alternative immunoreactivity could be found. The discrepancy in these findings obtained by light and
electron microscopes is discussed. In nerve terminals reactive to both antisera, an immunogold
technique indicated the presence of neurosecretory granules with colocalization of immunoreactive
ANG II and AVP, granules with only ANG II-like immunoreactivity and granules showing only
AVP-like immunoreactivity. The AVP-like immunoreactivity observed in the PON and the neurohy-

© 1990 Zoological Society of Japan

pophysis is considered due to arginine vasotocin.

INTRODUCTION

A renin-angiotensin system in the brain has been
demonstrated biochemically and pharmacological-
ly in mammals [1]. Immunohistochemically,
angiotensin II (ANG II) and arginine vasopressin
(AVP) have been shown to be present in the same
neurons of the supraoptic, paraventicular and sup-
rachiasmatic nuclei of the rat [2-4]. However,
intragranular colocalization of these peptides in
these neurons has not been studied.

In the present study, the colocalization of im-
munoreactive ANG II and immunoreactive AVP
was examined in neurosecretory cells of the preop-

Accepted May 11, 1989
Received April 27, 1989

' Present address: Department of Physiology, National
Defense Medical College, Tokorozawa 359, Japan

3 Present address: Department of Clinical Laborator-
ies, Keio University Hospital, Tokyo 160, Japan

tic nucleus (PON) of the goldfish, Carassius anra-
tus. Further, intragranular colocalization of these
substances was electron microscopically examined
in axon terminals in the neurohypophysis.

MATERIALS AND METHODS

Antisera

The following antisera were used for light mic-
roscopy. Antiserum to ANG II was raised in
rabbit against synthetic Asp!-Ileuw°-ANG II (Pro-
tein Research Foundation, Osaka) by Yamaguchi
[5]. The complete cross-reactivities of this anti-
serum with Asp!-Val’-ANG II [5] and Asn'-Val-
ANG II (unpublished data) were demonstrated by
radioimmunoassay. Antiserum to arginine AVP
was raised in rabbit against synthetic AVP (Protein
Research Foundation, Osaka) and cross-reacted
with arginine vasotocin (AVT) at 42% and with

258 C. YAMADA, S. Nos et al.

oxytocin (OXT) at only 3.5% [6].

For electron microscopy, the ANG II antiserum
was the same as that used for the light microscopic
experiment. AVP antiserum raised against synthe-
tic AVP in rabbit (UCB Bioproducts, Belgium)
and having complete cross-reactivity with AVT
and less than 0.003% cross reactivity with oxytocin
or mesotocin was used.

Light microscopy

Twenty five goldfish, Carassius auratus (about
10 cm in total length) were obtained commercially.
After decapitation, brains with the pituitary or
brains alone were quickly removed and fixed in
Bouin’s solution overnight. Tissue was dehydrated
through a series of ethanol, cleared in xylol and
embedded in paraffin. Four sm thick sagittal
sections were made and mounted on slides. To
examine the colocalization of immunoreactive
ANG II and AVP, two consecutive sections were
immunostained, one with ANG II and the other
with AVP antiserum.

Deparaffinized preparations were immuno-
stained by the peroxidase-anti-peroxidase (PAP)
method of Sternberger ef al. [7]. Incubation was
performed as follows: (1) in 0.3% H>O> for 30 min
at room temperature (RT), (2) in ANG II anti-
serum (1: 1000) or AVP antiserum (1 : 2000) over-
night at 4°C, (3) in goat anti-rabbit IgG (GAR;
Polysciences Inc., Warrington, Pennsylvania; 1:
200) for 90min at RT, (4) in peroxidase-anti-
peroxidase (PAP; Dako Corp., Copenhagen or
Cappel Laboratories, West Chester, Pennsylvania;
1: 200) for 90 min at RT, and (5) in 0.02% 3,3’-
diaminobenzidine in 0.05 M Tris buffer (pH 7.6)
containing 0.006% H>O> for 10-15 min at RT. To
rinse the preparations and dilute the antisera,
0.1 M phosphate buffer saline (pH 7.2) containing
0.3% Triton X-100 was used.

For the control, immunostaining was conducted
using the following sera instead of the primary
antisera: normal rabbit serum (NRS; Polysciences
Inc., Warrington, Pennsylvania; 1 : 1600), ANG II
antiserum preabsorbed with Asn'-Val-ANG II
(Hypertensin, Ciba; 20, 100 g/ml diluted anti-
serum), ANG II antiserum preabsorbed with AVT
(Protein Reserarch Foundation, Osaka; 20 g/ml
diluted antiserum), AVP antiserum preabsorbed

with Asn'-Val?-ANG II (20 g/ml diluted anti-
serum), the primary antisera preincubated with
1% bovine serum albumin(BSA) and AVP anti-
serum preabsorbed with AVT (20, 100 «g/ml di-
luted antiserum). AVP antiserum was preab-
sorbed with AVT but not AVP, since, as is well
known, neurosecretory cells produce AVT but not
AVP in teleosts.

Electron microscopy

Ten goldfish (each about 8 cm in total length)
were obtained from a commercial source. They
were anesthetized with 0.01% ethyl m-
aminobenzoate methanesulfonate (MS222) and
perfused with a mixture of paraformaldehyde
(4%) and glutaraldehyde (0.4%) in 0.05 M phos-
phate buffer (PB; pH 7.2). The pituitary of each
specimen was removed and fixed in the same
fixative for 2-3 hr at 4°C. This was followed by
rinsing in 0.1 M Millonig PB and postfixation in
2% OsO, in 0.1 M Millonig PB for 1.5 hr at 4°C.
All tissue was subsequently dehydrated through a
series of ethanol, transferred to propylene oxide
and embedded in an Epon-Araldite mixture.
Ultrathin sections were cut and mounted on 200-
mesh nickel grids.

Ultrathin sections were stained by a double
immunogold technique. First, one face of a section
was incubated in (1) saturated sodium
metaperiodate for 30min at RT, (2) 1% egg
albumin in 0.01 M PBS (pH 7.2) for 10 min at RT,
(3) AVP antiserum (1: 16000) overnight at 4°C,
and (4) colloidal gold labeled GAR (1:20; gold
particles of about 5 nm in diameter) for 90 min at
RT. Next, another face was incubated in the same
way, but immunostained with ANG II antiserum
(1:1000) overnight at 4°C, and colloidal gold
labeled GAR (1:10; gold particles of about 15 nm
in diameter) for 90min at RT. Immunostained
sections were stained further with both uranyl
acetate and lead citrate, and examined with
Hitachi HS-9 and HU-12A electron microscopes.

For the control, ANG II antiserum preabsorbed
with either Asn'-Val°-ANG II or AVT (each 10
vg/diluted antiserum), and AVP antiserum preab-
sorbed with AVT, ANG II or isotocin (Protein
Research Foundation, Osaka) (each 10 «g/diluted
antiserum) were used as primary antisera.

Fic.

Fic.

Colocalization of AVP and ANG II 259

ile

A: ANG II-like immunoreactive cells in the PON (arrow-heads) and their fibers (arrows) extending to the
neurohypophysis in the hypothalamus of the goldfish, Carassius auratus. Bar=100 ym. B: ANG II-like
immunoreactive cells in magnocellular (M) and parvocellular groups (P) of the PON. In both groups, many
non-immunoreactive cells (NC) were found. Bar=50 ~m. C: ANG II-like immunoreactive cells (arrowheads) in
the pars distalis and immunoreactive fibers (arrow) extending to the pars distalis from the PON. Reaction of
fibers to ANG II antiserum was abolished by preabsorption of the serum with ANG II, but not that of the cells.
Bar=20 pm.

2. Two consecutive sections of a PON region of the goldfish. Immunostaining with ANG II (A) and AVP (B)

antisera. Cells a, c and d were reactive to both antisera; cells b, e, f and g were reactive only to AVP antiserum.
Bar=20 pm.

260 C. YAMADA, S. Nogsi et al.

RESULTS

Light microscopy

ANG II and AVP antisera, preabsorbed with
synthetic ANG II and AVT, respectively, showed
no indication of immunoreaction. However, ANG
II immunoreaction in the cells of the pars distalis
was not abolished by ANG II antiserum preab-
sorbed with ANG II. Immunoreaction of these
cells is considered nonspecific. Immunostaining by
NRS also failed to indicate immunoreaction.
Other control sera used did not abolish im-
munoreactions. Immunoreaction to ANG II anti-
serum in the brain may thus be considered specific

to ANG II and the immunoreaction to AVP
antiserum observed in the brain is considered due
not to AVP but to AVT. That teleostean
neurosecretory neurons produce AVT but not
AVP supports these considerations.
Immunoreactivity to ANG II antiserum was
observed in the cells of the magnocellular and
parvocellular groups of the preoptic nucleus
(PON) (Figs. 1A, B, 2A) as well as to AVP
antiserum (Fig. 2B). The colocalization of ANG
II- and AVP-like immunoreactivity was evident in
many neurons (Fig. 2). Certain neurons possessed
only AVP-like immunoreactivity (Fig. 2), while
others, only ANG II-like immunoreactivity; the
number of the latter was very small. Neurons

Fic. 3.

Ultrastructural localization of ANG II- and AVP-like immunoreactivity in the neurohypophysis of a goldfish.

Large colloidal gold particles (diameter, about 15 nm) and small colloidal gold particles (diameter, about 5 nm)
demonstrate ANG II- and AVP-like immunoreactivity, respectively. A nerve terminal (A) contained both
immunoreactivities while another, (B), neither. Bar=100 nm.

Fic. 4.

Intragranular localization of ANG II- and AVP-like immunoreactivity in the same nerve terminal of the

neurohypophysis of a goldfish. Large colloidal gold particles (diameter, about 15 nm) and small colloidal gold
particles (diameter, about 5 nm) demonstrate ANG II- and AVP-like immunoreactivity, respectively. In some
neurosectretory granules, both immunoreactivities could be detected (arrows). Some granules show only ANG
II-like immunoreactivity (large arrowheads) while others, only AVP-like immunoreactivitiy (small arrowheads).

Bar=100 nm.

Colocalization of AVP and ANG II 261

showing no immunoreaction to either antiserum
were also present. The fibers with either ANG II-
or AVP-like immunoreactivity extended as far as
to the neurohypophysis (Fig. 1A) and pars distalis
(Fig. 1C). In the pars distalis, immunoreaction to
ANG II antiserum was observed in the cells and
fibers (Fig. 1C). The reaction of the fibers in the
pars distalis was abolished by ANG II antiserum
preabsorbed with ANG II, but that of the cells was
not abolished because of its nonspecificity. The
cells of the pars intermedia were not stained by
either antiserum.

Electron microscopy

In control experiments, immunoreaction was
abolished by preabsorption of ANG II and AVP
antisera with Asn'-Val°-ANG II and AVT, respec-
tively. Immunoreaction to ANG II antiserum was
not abolished by preabsorption of ANG II anti-
serum with AVT, nor was that to AVP antiserum
preabsorbed with ANG II or isotocin.

Electron microscopy indicated a number of gra-
nules and synaptic vesicle-like structures to be
present in nerve terminals in the neurohypophysis.
Two types of axon terminals were observed, those
with both ANG II- and AVP-like immunoreactiv-
ity and those with neither immunoreactivity (Fig.
3). In the former terminals, the three following
kinds of granules (about 80 nm in diameter) were
detected: 1) granules showing both ANG II- and
AVP-like immunoreactivity, indicated by large
and small gold particles, respectively, 2) granules
showing only ANG II-like immunoreactivity and
3) granules showing only AVP-like immunoreac-
tivity (Fig. 4). The ratio of these kinds of granules
differed for each terminal. Some granules in the
terminals with both immunoreactivities showed
greater ANG II-like immunoreactivity than AVP-
like immunoreactivity or visa versa.

DISCUSSION

In the present study, immunoreaction to AVP
antiserum was frequently observed in the cells of
the magnocellular and parvocellular groups of the
PON and in the neurohypophysis. This reaction
may possibly be due to AVT, since the AVP
antiserum used in the present experiment was

demonstrated to cross-react with AVT, and also, it
is known that one of the neurohypophysial hor-
mones in teleosts is AVT but not AVP. Thus, in
the following description, AVT was used instead
of AVP.

Most fibers of ANG II-like immunoreactive
neurons in the magnocellular and parvocellular
groups extended as far as to the neurohypophysis,
as in the case of the rat [8, 9]. Some invaded the
pars distalis of the adenohypophysis, as well as
AVT fibers. It would thus seem that the hypotha-
lamo-hypophysial nervous system of ANG II is
present in the goldfish, although its function has
yet to be clarified.

The colocalization of ANG II- and AVT-like
immunoreactivity was observed in the perikarya of
many neurosecretory neurons of the PON. It has
also been demonstrated that immunoreactive
ANG II and AVP colocalized in the neurons of the
paraventricular, supraoptic and suprachiasmatic
nuclei in the rat [2, 4]. Further, the intragranular
colocalization of immunoreactive ANG II and
immunoreactive AVT in some axon terminals of
the neurohypophysis was found in the present
study. It would appear that both peptides are
simultaneously released from these terminals into
the capillaries. In the rat, ANG II has been shown
to stimulate ACTH release from the adenohy-
pophysis as well as AVP [10] and ANG II and
AVP to potentiate the ACTH-releasing activity of
corticotropin-releasing factor (CRF) [11 for ANG
II, 12 for AVP]. In the goldfish, ANG II and AVT
stimulate the release of ACTH [13]. It should thus
be reasonable to conclude that ANG II and AVT,
following their simultaneous release from the same
terminal, may potentiate the ACTH-releasing
activity of CRF in this fish.

The colocalization of AVP and CRF in the
neurons of the paraventricular nucleus and in the
fibers in the median eminence has been reported in
mammals [14-17]. In teleosts, this has also been
shown in some neurons of the PON [13, 18, 19].
The present authors noted ANG II-, AVT- and
CRF-like immunoreactivity in the same neurons in
the PON of the goldfish (unpublished data). Thus,
ANG II, AVT and CRF may be released simul-
taneously from the same neurons. These peptides
possibly exert a synergistical effect on the release

262

of ACTH from the adenohypophysis or ANG II
and AVT may modulate the release of CRF.

By light microscope, four types of the nerve cells
were observed in the PON: 1) cells with both
immunoreactive ANG II and AVT, 2) cells with
only immunoreactive AVT, 3) cells with only
immunoreactive ANG II, and 4) cells without any
immunoreaction. By electron microscope, howey-
er, only two types of nerve terminals were found in
the neurohypophysis: 1) terminals showing im-
munoreactive ANG II and AVT, and 2) terminals
showing no immunoreactivity. The discrepancy
with respect to cell type number as determined
using these different microscopes may be due to
variation in the amount of storage of these pep-
tides in the cell bodies and in the terminals.
Immunoreactivity in cells containing the peptides
in very small amounts would not be detected by
light microscopy, leading to the erroneous conclu-
sion that there are four cell types. However, both
light and electron microscopy also indicate the
presence of cells and nerve terminals containing
neither ANG II- nor AVT-like immunoreactivity.
These neurons may contain neuropeptides other
than ANG II or AVT.

ACKNOWLEDGMENTS

We are grateful to Dr. Ken-ichi Yamaguchi, Depart-
ment of Physiology, Niigata University School of Medi-
cine, for kindly providing the ANG II antiserum, and to
Professor Seiichiro Kawashima, Zoological Institute,
Faculty of Science, University of Tokyo, and Dr. Keiichi
Kawamoto, Zoological Institute, Faculty of Science,
Hiroshima University, for giving us the AVP antiserum.

REFERENCES

1 Printz, P., Ganten, D., Unger, T. and Phillips, M. I.
(1982) The brain renin angiotensin system. In
“Experimental Brain Research Suppl. 4, The Renin-
Angiotensin System in the Brain”. Ed. by D. Gan-
ten, M. Printz, M. I. Phillips, and B. A. Scholkens,
Springer-Verlag, Berlin Heidelberg New York, pp.
5-52.

2 Kilcoyne, M. M., Hoffman, D. L. and Zimmerman,
E. A. (1980) Immunocytochemical localization of
angiotensin II and vasopressin in rat hypothalamus:
evidence for production in the same neuron. Clinical
Sci., 59: 57s—60s.

3 Fuxe, K., Ganten, D., Anderson, K., Calza, L.,

10

C. Yamapa, S. Nos et al.

Agnati, L. F., Lang, R. E., Poulsen, K., Hékfelt, T.
and Bernardi, P. (1982) Immunocytochemical de-
monstration of angiotensin II- and renin-like im-
munoreactive nerve cells in the hypothalamus.
Angiotensin peptides as comodulators in vasopres-
sin and oxytocin neurons and their regulation of
various types of central catecholamine nerve termin-
al systems. Exp. Brain Res., Suppl., 4: 208-232.
Hoffman, D. L., Krupp, L., Schrag, D., Nilaver,
G., Valiquette, G., Kilcoyne, M. M. and Zimmer-
man, E. A. (1982) Angiotensin immunoreactivity in
vasopressin cells in rat hypothalamus and its relative
deficiency in homozygous Brattleboro rats. Ann.
New York Acad. Sci., 394: 135-141.

Yamaguchi, K. (1981) Effect of water deprivation
on immunoreactive angiotensin II levels in plasma,
cerebroventricular perfusate and hypothalamus of
the rat. Acta Endocrinol., 97: 137-144.
Kawamoto, K. (1985) Immunohistochemical study
of vasopressin and oxytocin in the neurosecretory
system during reorganization of the neural lobe in
mice. Zool. Sci., 2: 371-380.

Sternberger, L. A., Hardy, P. H. Jr., Cuclis, J. J. Jr.
and Meyer, H. G. (1970) The unlabeled antibody
enzyme method of immunohistochemistry: prepara-
tion and properties of soluble antigen-antibody com-
plex (horseradish peroxidase-anti-horseradish pero-
xidase) and its use in identification of spirochetes. J.
Histochem.-Cytochem., 18: 315-333.

Brownfield, M. S., Reid, I. A., Ganten, D. and
Ganong, W. F. (1982) Differential distribution of
immunoreactive angiotensin and angiotensin-
converting enzyme in rat brain. Neurosci., 7:
1759-1769.

Lind, R. W., Swanson, L. W. and Ganten, D.
(1985) Organization of angiotensin IJ immunoreac-
tive cells and fibers in the rat central nervous system.
An immunohistochemical study. Neuroendocrinolo-
gy., 40: 2-24.

Spinedi, E. and Negro-Vilar, A. (1983) Angiotensin
II and ACTH release: site of action and potency
relative to corticotropin releasing factor and vasop-
ressin. Neuroendocrinology., 37: 446-453.
Schoenberg, P., Kehre, P., Muller, A. F. and Gail-
lard, R. C. (1987) Angiotensin II potentiates corti-
cotropin-releasing activity of CRF41 in rat anterior
pituitary cells: mechanism of action. Neuroendocri-
nology., 45: 86-90.

Gillies, G. E., Linton, E. A. and Lowry, P. J. (1982)
Corticotropin releasing activity of the new CRF is
potentiated several times by vasopressin. Nature,
229: 355-357.

Fryer, J. N. (1988) Neuropeptides regulating the
secretory activity of goldfish corticotropes and mela-
notropes. Program and Abstracts, Ist Intern. Symp.
on Fish Endocrinology, Univ. of Alberta, Edmon-
ton, Canada, p. 19.

14

15

16

17

Colocalization of AVP and ANG II

Tramu, G. Croix, C. and Pillez, C. A. (1983)
Ability of the CRF immunoreactive neurons of the
paraventricular nucleus to produce a vasopressin-
like material. Neuroendocrinology., 37: 467-469.
Kiss, J. Z., Mezey, E. and Skirboll, L. (1984)
Corticotropin-releasing factor-immunoreactive neu-
rons of the paraventricular nucleus become vasop-
ressin positive after adrenalectomy. Proc. Natl.
Acad. Sci. USA, 81: 1854-1858.

Whitnall, M. H., Mezey, E. and Gainer, H. (1985)
Co-localization of corticotropin-releasing factor and
vasopressin in median eminence neurosecretory
vesicles. Nature, 317: 248-250.

Hisano, S., Tsuruo, Y., Katoh, S., Daikoku, S.,
Yanaihara, N. and Shibasaki, T. (1987) Intragranu-

18

19

263

lar colocalization of arginine vasopressin and
methionine-enkephalin-octapeptide in CRF axons in
the rat median eminence. Cell Tissue Res., 249:
497-507.

Olivereau, M., Moons, L., Olivereau, J. and Van-
desande, F. (1988) Coexistence of corticotropin-
releasing factor-like immunoreactivity and vasotocin
in perikarya of the preoptic nucleus in the eel. Gen.
Com. Endocrinol., 70: 41-48.

Yulis, C. R. and Lederis, K. (1987) Co-localization
of the immunoreactivities of corticotropin-releasing
factor and arginine vasotocin in the brain and pituit-
ary system of the teleost Catostomus commersoni.
Cell Tissue Res., 247: 267-273.

te

oie We ANNA tie’ tel
nyt otal ayy fea Buedisliw

‘yoy ogling de adh ‘i

i Toe Vp es

4} teal A sy bi nage

eens (eat I Peper:
i Nx aah real Whit

Ah gt wit Wwe Gira

q D Te |

* ire) trie Sb - Dye

‘aoe wostrvanngngl tal Dag
efi bd lang we ta)
fea WA WH) ieee nm

ai
f

o—

v1

0 Samadi pan

7

=
>

alae Wire 3 Libt

eee
eA" ‘a
cet a8 a hi

Luni? agence
in awa gee farlietid ¥

s eH ‘ » PaKle site ae.

“nam. © "ai, lw

| aeh peeled eh! ayn ae
Ts ¢ wags <—_
\ . rts nei
Ae Handy Lee
#3} i -.
war evita
& fi et
ek Vee i
ne
mee) = Vie
E . :
ae ‘
<j
eave '.%
ai
va
\
- %

ai:
f ‘ aay

<r peat oA
sy tee omnes
‘«< Sored, lat ineie

: hin hg, I

ZOOLOGICAL SCIENCE 7: 265-271 (1990)

Ovarian Development and Sex Steroid Hormones during the
Reproductive Cycle of Rana esculenta Complex

ALBERTA MARIA POLZONETTI-MAGNI, ROBERTA CurINt!, OLIANA CARNEVALI,
CLAUDIO NovARA, MASSIMO ZERANI and ANNA GOBBETTI

Dipartimento di Biologia Cellulare, Universita di Camerino,
Via Camerini 1, 62032 Camerino, and ‘Dipartimento
di Scienze Chimiche, Universita di Camerino,
Via S. Agostino 1, 62032 Camerino, Italy.

ABSTRACT—The aim of this work is to investigate the ovarian development of Rana esculenta living in
Colfiorito pond (820m a. s.l.). This process depends on a complex interplay between the estradiol-
induced hepatic vitellogenin and the uptake of vitellogenin by the ovary. In order to establish whether
the ovarian growth may depend on hydratation process as well, the ovaries have been analyzed— during
the annual cycle—by thermo-analysis. As far as we know, this method has been employed for the first
time to lower vertebrate tissues with the present work. Vitellogenin titre varies according to a temporal
pattern which is characteristic of this mountain population and is well correlated with plasma estradiol
concentration. Samples of ovarian tissues, monitored by thermal analysis, show that, in the
pre-spawning period three different kinds of water are bonded to the tissues with different energies of
interaction. In the ovulatory period, the water is released in one process only. Samples of the
post-ovulatory period give the curves, which can be plotted in-between the pre-spawning and ovulatory
ones. In conclusion, the present results describe how two different kinds of water are released during
the ovarian development. The low weight ovaries, with small oocytes, show a high water percentage
which is lost in only one step; the high weight ovaries show a low water percentage, that is lost in three
steps. Therefore, in the recovery phase, the ovarian growth, seems mainly due to yolk storage and only
partly to the hydratation process which, moreover, is closely related with plasma sex steroid

© 1990 Zoological Society of Japan

concentration.

INTRODUCTION

The ovarian development in anuran amphibian
Rana esculenta is cyclical: the growth starts in
September (recovery phase), stops temporarily
during winter stasis and resumes just before ovula-
tion (April-May). This development is regulated
by several environmental cues (temperature and
photoperiod) and endocrine mechanisms.

Previous studies [1] have shown that the length
of the recovery phase depends mainly on tempera-
ture. In fact the recovery phase in the population
of green frogs living in the mountains is shorter
than that in the population living at sea level. Such
a difference depends on an early drop of mountain

Accepted June 2, 1989
Received January 6, 1989

temperature in autumn. The ovarian growth is
essentially due to the storage of vitellogenin,
synthesized by the liver and then released in the
blood. Moreover, also hydratation process—not
yet well known—plays a part in ovary weight
increase [2].

The vitellogenesis occurs under hormonal con-
trol: in fact, as in other amphibian species [3-5],
also in Rana esculenta [6, 7], estradiol is the
hormone most responsible for the vitellogenin
synthesis, while the gonadotropins support the
ovarian uptake. Nevertheless, Gobbetti et al. [8]
pointed out a significant intervention of gonado-
tropins in inducing hepatic vitellogenin synthesis.
It is certain, however, which is the role of hydrata-
tion in ovarian growth and how it is regulated by
sex steroid hormones. Therefore it seemed a good
opportunity to investigate these aspects by thermal

266 A. M. PoLzonetTTI-MAGNI, R. CurINI et al.

analysis; as far as we know, this was the first
application to lower vertebrate tissues. At the
same time, the plasma vitellogenin and sex hor-
more titre were monitored during the reproductive
cycle of Rana esculenta. All samples were taken in
the field to avoid the stress of captivity on plasma
steroid concentration [9].

MATERIALS AND METHODS

Animals and tissues

Ten female frogs, Rana esculenta complex, were
monthly captured in the moutain poud of Colfiori-
to (820 m above sea level) and anesthetized with
MS 222 (Sandoz). Previous data indicate that MS
222 does not affect the hormone profiles in Rana
esculenta [10, 11]. Blood was collected into hepari-
nized centrifuge tubes by a heparinized glass capil-
lary which was inserted into the conus arteriosus.
After centrifugation, individual plasma samples
were stored at —20°C until hormonal determina-
tions. Ovaries-taken from these animals- were
removed and weighed. One of these was kept at
—20°C until thermal analysis and the other one
was placed in amphibian Krebbs-Ringer solution.
The latter was opened carefully and dissected in
order to separate follicular oocytes, as previously
described [1]. The dissected oocytes were
screened by granulometric sieves into three clas-
ses: black atretic follicles (4: 0.4-0.69 mm), pre-
vitellogenic follicles (¢: 0.70-1.2 mm), and vitel-
logenic follicles (¢: 1.21-2.0 mm).

Thermal analysis

The water interactions, in the biological sys-
tems, are in function of the hydrogen bonds, of the
Van der Waals forces and the London forces etc.;
so the water in the biolobical systems is bound to
the matrix by different energies, also in relation to
the different kinds of molecules present in the
matrix. When a biological system is heated, the
water is released in different successive steps in
function of the necessary activation energy to be
obtained in the break of each different bond from
its water-matrix. Each step will thus represent a
particular onertype of water characterized by one
particular value of the interaction energy with the

matrix.

The different kinds of water and their percen-
tage were determined by thermal analysi (TG) as
described by Wendlandt [12], using a Perkin Elmer
TG-S2 thermobalance equipped with a data sta-
tion. The operational atmosphere was air or
oxygen at a flow of 150ml min~'. An inert
atmosphere was also used in the preliminary phase
to check whether oxidation phenomena may an-
ticipate the decomposition of the organic matter
which would interfere with the water percentage
determination. The temperature program chosen
was 10°C min ' and the sample weight ranged
between 20 and 50 mg.

Vitellogenin assay

Male frogs were estrogenized with estradiol
silastic tubes implanted into the ventral cavity and
were kept in water tanks on a diet of mealworms.
After 20 days, the estrogenized males were anaes-
thetized on MS 222 and blood samples were col-
lected as previously described. Vitellogenin was
purified from estrogenized serum by the EDTA-
MgCl, method [13] and then chromatographized
on DEAE cellulose (Whatman).

Using a protocol developed in our laboratory,
purified vitellogenin has been used to raise anti-
bodies. When antiserum was tested by immuno-
blotting, strong reactivity was observed vs. isolated
vitellogenin. Control experiments were made with
preimmune serum. The antibodies recognized a
protein present in estrogenized frog serum but
failed to react against control serum. The data
lead us to conclude that i) the antiserum is specific
for vitellogenin and ii) the antigen was purified to
a high extent. So the a-VTG Abs were used to
determine the plasma vitellogenin titre by enzyme-
linked assay (ELISA). The serum of ten female
frog, used as antigen, was absorbed on polystyrene
microplates. Rabbit vitellogenin antiserum was
diluted 1: 1000 and was incubated for 2 hr at room
temperature. The conjugate was _ alkaline-
phosphatase goat antirabbit immunoglobulin,
diluted 1/1000 and incubated for 2hr at room
temperature. P-nitropheyl-phosphate was em-
ployed as enzyme substrate for 60 min, the results
were expressed in absorbance units at 405 nm.
The calibration curve shown in Figure 1 enables

Ovarian Development in Rana esculenta 267

is)

Q e

c

©

2

°o

1)

2

©

1 e
01 0.3 0.5 07 09 11 13
VTG (yg)
Fic. 1. Calibration curve of Rana esculenta plasma

vitellogenin as determined by enzyme-linked im-
munosorbent assay. The absorbance was plotted
versus the relative amount of antigen placed in the
hole. Each point represents the mean of three
independent measurements.

the antigen titre to be measured in all the sera
collected. :
The sensitivity of the ELISA method was 3 ng
(intraassay 6%; interassay 11%).
Protein concentration was determined by the
procedure which uses bovine serum albumin as
standard.

Plasma hormone determinations

Plasma samples, taken monthly from ten
females, were extracted with ether; subsequently
radioimmunological analyses (RIA) of estradiol-
178 (E), progesterone (P) and androgens (A),
were carried out as previously described [15].

The following sensitivities were observed: tes-
tosterone 5 pg (intraassay, 7%; interassay, 13%;
progesterone, 7 pg (intraassay, 4%; interassay,
8%). Testosterone was not separated from dihydro-
testosterone and therefore, since its antibody
reacts with dihydrotestosterone, the data are ex-
pressed as “androgen”. The antisera were pro-
vided by Dr G. Bolelli, Centro di Fisiopatologia
della Riproduzione, C. N. R., Bologna.

All numerical data of the present experiments
were analyzed using an ANOVA method.

RESULTS

The ovarian growth, during the reproductive
period, has been evaluated considering the oocytes
diameter. During the ovulatory period (May), the
ovary comprises mainly full-growth oocytes (1.21-
2.0mm). In the post-reproductive period (July),
the number of full-growth oocytes decreases and
the smallest oocytes look atretic as in the previtel-
logenic stage. During the recovery phase (Septem-
ber), the number of oocytes entering the vitel-
logenic cycle increases and the cells continue their
growth and gradually accumulate in the later
stages.

The plasma vitellogenin profile is correlated
with ovarian composition. The ovary weight pre-
sents its main peak in winter (stasis period), before
ovulation (April). In the post-reproductive
period, because the oocytes are ovulated, the
ovary weight is very low, and begins to grow
during the recovely phase (autumn), to reach its
maximum the next winter, as can been seen in
Figure 2.

These modifications, when submitted to analysis
of variance, were statistically significant (p<
0.001).

Also the variations of plasma vitellogenin titre
were Statistically significant (p<0.001). The high
vitellogenin levels both in stasis and in the recov-
ery period are correlated with the ovary weight.
On the contrary, in the reproductive and post-
reproductive period, the ovary weight is low be-
cause the ovulation and vitellogenin accumulates
in the plasma. The estradiol profile agrees with
our previous dat [15]. The significant increase (p<
0.001) of plasma estradiol levels in July is responsi-
ble for the induction of hepatic vitellogeinin syn-
thesis which in turn allows recovery of the ovarian
weight [8].

Samples of ovarian tissues, corresponding to the
different periods of the ovarian cycle, were analy-
sed by TG to assay the hydratation process. The
curves obtained show three types of behavior, each
characteristic of a specific period of the cycle.
Figure 3 shows a characteristic trend of water
percentage in the fullgrowth oocytes (pre-
reproductive period). In fact the water is released
through three different processes occurring within

268 A. M. PoLzonettiI-MaAGNI, R. CurRInI et al.

e estradiol (ng/ml plasma)

months

Fic. 2.

1.2 4
0.9 13
o
oc o~
E ro)
w 4
© ~
6 oe 2
O60. =
£ ®
> >
Do
£ >
WIZ =
~ >
> °o
< a

Time course of vitellogenin and estradiol titre in the plasma of Rana esculenta cycle includes four periods:

repreductive period (re), post-reproductive (pr), recovery (rc) and stasis (st). Number of frogs used is shown in

the text.

—-- = DERIVATIVE
188. 28

% WEIGHT
ia
8

TEMPERATURE (C)

RATE: 18.88 deg/min

WT: 18.6226 mg

TG

Fic. 3. Thermogravimetric analysis (TG) of Rana esculenta ovaries during the pre-ovulatory period.

the temperature intervals of 25-105°C (1st pro-
cess), 105-155°C (2nd process) and 155-210°C
(3rd process). They ssuggest that three different
kinds of water are bonded to the tissues with
different energies of interaction.

The thermal behavior is surprisingly constant
during a very long interval of time, especially in

consideration of the fact that such samples were
obtained from 10 different animals. The percen-
tage of total water contained in the examined
tissues ranged between 45 and 55%. The curves
corresponding to the ovulatory period samples
(low weight ovaries) show that the water is lost in
one process only, occurring in the temperature

Ovarian Development in Rana esculenta 269

— --- = DERIVATIVE
188. 88

% WEIGHT
ff
8

WTs 28.8422 mg RATEs

10.88 deg/min

TEMPERATURE <C) TG

Fic. 4. Thermogravimetric analysis (TG) of Rana esculenta ovaries during the ovulatory period.

—---- = DERIVATIVE
188. BB

WTs 27.5519 mg RATEs 18.88 deg/min |

TEMPERATURE (C) TG

Fic. 5. Thermogravimetric analysis (1G) of Rana esculenta ovaries during the post-reproduction period.

interval of 25-180°C with a functuation as de-
scribed above. The thermal behavior is quite
constant and only rarely is the peak split in two
different tips. The percentage of total water
contained in the examined tissues ranged between
65 and 75% (Fig.4). Samples of the post-
ovulatory period originate TG curves, that are
plotted in between those for the ovulatory and

pre-spawning period. Figure 5 depicts an example
of typical intermediate TG behavior.

DISCUSSION

Ovarian development in non-mammalian verte-
brates entails a complex interplay between the
ovary and various other organs [16]. A high

270

molecular weigh yolk precursor (vitellogenin) is
synthesized and secreted by the liver and subse-
quently transferred to the ovary by endocytic
uptake. Since synthesis and uptake of vitellogenin
are both hormone-dependent, the turn-over of
vitellogenin in the blood appears to be a function
of the hormonal balance influencing both proces-
ses. In Rana esculenta, Gobbetti et al. [17] found
that estradiol-178 induces vitellogenin synthesis,
as in other amphibian species. Recent results also
suggest that the availability of specific liver recep-
tors is strongly related to vitellogenin hormonal
regulation [18].

The present data show a strict relationship be-
tween vitellogenin titre and ovarian behavior as
witnessed by ovary weight during stasis and the
recovery phase. On the contrary, when the ovary
weight is very low in May (full-growth oocytes are
ovulated), high concentrations of vitellogenin
accumulate in the plasma.

The results obtained on ovarian tissue checked
by TG at the different periods of the cycle support
the hypothesis that the determination of water
content could be important in the explanation of
ovarian growth. In fact, our results indicate that
two different kinds of water are released during
ovarian development. The low weight ovaries with
small oocytes show a high water parcentage, which

4 H20 (% ovary weight)

J F M A M J
months

Fic. 6.

A. M. PoLzONETTI-MAGNI, R. CurIni et al.

is lost in only one step. In contrast, high weight
ovaries (full-growth oocytes) show a low water
percentage, that is lost in three steps. In the high
weight ovaries, therefore, the water percentage is
lower than in small ovaries. The water of the full
growth oocytes, released at high temperature
(210°C), is probably connected with yolk macro-
molecules. On the contrary, the high water per-
centage of small oocytes is released at low temper-
ature. Therefore, the increase of ovarian weight
during the recovery period of Rana esculenta
seems mainly due to the storage of yolk and only
partly to hydratation processes [19].

The water percentage is well related with the
plasmatic sex steroid concentrations. In Rana
esculenta the main peak of plasma androgen levels
was found in the reproductive period [20]. Also in
these experiments the androgen titre is higher
when the water percentage is about 50% (Fig. 6).
During the post-reproductive period the high titre
of plasmatic estradiol agrees with the high percen-
tage of water in low weight ovaries. It is already
known that steroid hormones in fish and in mam-
mals are involved in tissues hydratation process
[21, 22]; while, as far as amphibians are concerned,
other hormones, such as prolactin in urodeles, may
play an important role. It is therefore worthwhile
studying more in depth in anurans the implication

© ovary weight (gr)

JA S$ O N OD

Relationship between ovary weight, ovary water percentage and plasma sex steroid levels (A =androgens; E

=estradiol-17/), in the different phases of female Rana esculenta reproductive cycle.

Ovarian Development in Rana esculenta

of other hormones in the hydratation process.

ACKNOWLEDGMENTS

Financially aided by the Italian Ministry of Education

(40 and 60%) and CNR.

10

REFERENCES

Polzonetti-Magni, A., Pagano, M., Gobbetti, A.,
Bellini-Cardellini, L. and Botte, V. (1984) Accres-
cimento annuale del-l’ovario in popolazioni di Rana
esculenta, viventi in localita con differente
andamento climatico. Riv. Biol., 77: 97-103.
Pierantoni, R., Varriale, B., Simeoli, C., Di Mat-
teo, L., Milone, M., Rastogi, R. K. and Chieffi, G.
(1983) Fat body and autumn recrudescence of the
ovary in Rana esculenta. Comp. Biochem. Physiol.,
76A: 31-35.

Wallace, R. A. and Jared, D. W. (1969) Studies on
amphibian yolk. VIII. The estrogen-induced hepatic
synthesis of a serum lipophosphoprotein and its
selective uptake by the ovary and transformation
into yolk platelet proteins in Xenopus laevis. Dev.
Biol., 19: 498-526.

Wallace, R. A. (1978) Oocyte growth: non mamma-
lian vertebrates. In “Evolution of the Vertebrate
Ovary”. Ed. by R. E. Jones, Plenum Press, New
York, pp. 469-502.
Searle, P. F. and Tata, J. R. (1981) Vitellogenin
gene expression in male Xenopus hepatocytes dur-
ing primary and secondary stimulation with estrogen
in cell cultures. Cell, 23: 741-746.

Giorgi, F., Gobbetti, A. and Polzonetti-Magni, A.
(1982) Variations in the vitellogenin titre during the
reproductive cycle of Rana esculenta L. Comp.
Biochem. Physiol., 72B: 501-506.

Varriale, B., Pierantoni, R., Di Matteo, L., Minuc-
ci, S., Milone, M. and Chieffi, G. (1988) Rela-
tionship between estradiol-17 seasonal profile and
annual vitellogenin content of liver, fat body,
plasma, and ovary in the frog (Rana esculenta). Gen.
Comp. Endocrinol., 69: 328-334.

Gobbetti, A., Polzonetti-Magni, A., Zerani, M.,
Carnevali, O. and Botte, V. (1985) Vitellogenin
hormonal control in the green frog, Rana esculenta.
Interplay between estradiol and pituitary hormones.
Comp. Biochem. Physiol., 82A: 855-858.

Licht, P., Mc Creery, 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.

D’Istria, M., Delrio, G., Botte, V. and Chieffi, G.

11

12

13

14

15

16

18

19

20

Al

22

271

(1974) Radioimmunoassay of testosterone, 17/-
oestradiol and oestrone in the male and female
plasma of Rana esculenta during sexual cycle.
Steroids Lipids Res., 5: 42-48.

Pierantoni, R., Iela, L., Delrio, G. and Rastogi, R.
K. (1984) Seasonal plasma sex steroid levels in the
female Rana esculenta. Gen. Comp. Endocrinol.,
53: 126-134.

Wendlandt, W. W. (1986) Thermal analysis. Anal.
Chem., 58: 1R-6R.

Wiley, H. S. and Dumont, J. N. (1978) Stimulation
of vitellogenin uptake in stage. 1V. Xenopus oocytes
by treatment with chorionic gonadotropin in vivo.
Bio. Reprod., 18: 762-771.

Lowry, O. H., Rosebrough, N. J., Farr, A. L. and
Randall, R. J. (1951) Protein measurement with
Folin phenol reagent. J. Biol. Chem., 193: 265-275.
Polzonetti-Magni, A., Botte, V., Bellini-Cardellini,
L., Gobbetti, A., Crasto, A. (1984) Plasma sex
hormones and post-reproductive period in the green
frog, Rana esculenta complex. Gen. Comp. Endo-
crinol., 54: 372-377.

Follett, B. K., Redshaw, M. R. (1974) The physio-
logy of vitellogenesis. In “Physiology of Amphibia”.
vol. 2, Ed. by B. Lofts, Academic Press, New York,
pp. 219-308.

Gobbetti, A., Polzonetti-Magni, A., Zrani, M. and
Paolucci, M. (1983) Vitellogenin synthesis induced
by estradiol-178 in the green frog, Rana esculenta
complex. Rend. Acc. Sci. Fis. Mat., 4: 60.
Paolucci, M. and Botte, V. (1988) Estradiol-binding
molecules in the hepatocytes of the female water
frog, Rana esculenta, and plasma estradiol and vitel-
logenin levels during the reproductive cycle. Gen.
Comp. Endocrinol., 70: 466-476.

Rastogi, R. K., Izzo-Vitiello, I., Di Meglio, M., Di
Matteo, L., Franzese, R., Di Costanzo, M. G.,
Minucci, S., lela, L. and Chieffi, G. (1983) Ovarian
activity and reproduction in the frog, Rana
esculenta. J. Zool. Lond., 200: 233-247.

Lupo, C., Zerani, M., Carnevali, O., Gobbetti, A.
and Polzonetti-Magni, A. (1988) Testosterone bind-
ing protein in the encephalon and plasma sex hor-
mones during the annual cycle in Rana esculenta
complex (Amphibia Ranidae). Monitore Zool. Ital.
(N. S.), 22: 133-144.

Babiker, M. M. and Ibrahim, H. (1979) Studies on
the biology of reproduction in the cichlid Tilapia
nilotica (L.): effects of steroid and trophic hormones
on ovulation and ovarian hydration. J. Fish Biol.,
15: 21-30.

Warren, J. C. and Crist, R. D. (1973) Effects of
ovarian steroids on uterine metabolism. In “Hand-
book of Physiology”, Ed. by R. O. Greep and E. B.
Aswood, publ. pp. 49-67.

*6 2.4,

rw
JHB 1

iiupsteray ty
ninelgots. ane, tp waren.
The aah de abeoate

7 | iPlniie
alry Wiha ke Cuber a
ean Ghat’ ai gel
hl ber Gf 1 Aw
‘ wibere df
e.h1) 82. ot
Phaece Pb aaa
Me iciiemi
ve ' Wy li BA A aawidaH fe
' 4 Aut five ay Dl poet,

Pre. A j yn .Fi yar si) a

cab

Mery

A *aywereret enn) | Laedie eee hias }

{

a?

\

— Leary. ott Sas

Pe Ree REO ON IGOR IM

nl tees neglendemestogtt ath o a

eves «Sali. ey walt) <xvey = mw
oe epipgt that (e lost ty yA F int
Ceti cme, THEN vat

nares ei

oll Gaicytex

3 he) , 4

win (sow heap pay it
MSI Ne ~anlibdbe ate
(Of vt hirSt.. doin
UDA Wee) lead yi Het
fs Heed Ups 4,3 douan
el Jee side 2 de ibd Wy uty
they ase t
ni i ite 2 ri five
miter Ww aS bee
- Diba (i besnutair cgensorralis la

APY Se aes
weasel Genre, oe :
4 saiet in to ne oe
on

ith Abst Wis CORNY SEED PEPE g
eit Upto hi
wees howl weer een icity
{imei) :
LANE) vat
ts ) Slat Gt
ae

i , le er 00

ise y

20 Ly

= + och bee
ratte matin, atailictes’
ae ~th'

alt UpetitOF Wag, ./

OT Sa Fishtail’ ahini ae
a ea » ork olay 4
up. (ie ET lieved?
} 4 1. # ott ‘
tid ton). tants) ia
niles boddow 7} 5 Age
Ya nesieuy if Gita

- * _#
Sorgen ) 209 ML yn Ue
‘

ar? Sa. Tee a

inaelt bt sind |

pyle v pate th ang Os
sofhnosiy atin (ouhay’
nibeies ras tte
ein abo tf Hts :
line ls rs
da cf das te teghe lenganad i

r’ vighe F nigusibheogs «
ed ali ae
Leb] Oe toniobal
yl rar bani ahh

iol] ere

‘Wis ti 1

*

Cet) bos ¥

ZOOLOGICAL SCIENCE 7: 273-279 (1990)

Long-Term Effects of Hypophysectomy on the Growth and
Calcification of Otoliths and Scales in the Goldfish,
Carassius auratus

YASUO MuGIyvA

Department of Biology and Aquaculture, Faculty of Fisheries,
Hokkaido University, Hakodate 041, Japan

ABSTRACT— Goldfish (Carassius auratus) were given tetracycline for time-marking, followed by
hypophysectomy. They were kept at 0.25% NaCl-tap water for 32 weeks and then sacrificed to examine
long-term effects of hypophysectomy on the accretive growth and calcification of otoliths and scales.
Plasma was also analysed for calcium and sodium concentrations.

Hypophysectomy resulted in cessation of somatic growth in length. Hypercalcemia and hyponatremia
were induced by hypophysectomy. Tetracycline-induced bands indicated that scale growth almost
ceased after operation, while otoliths continued to grow at a rate reduced to approximately half. The
marginal osteoid of scales was poorly developed in hypophysectomized fish. X-ray energy spectra
revealed no difference in elemental constituens of the newly formed regions of these tissues between the
experimental and control groups: major elements were phosphorus and/or calcium, suggesting that
these regions are fully calcified. These results were discussed with the allometric growth of otoliths to

© 1990 Zoological Society of Japan

somatic growth.

INTRODUCTION

Teleost otoliths and scales are mainly composed
of calcium salts, and are used for age determina-
tion. They grow by the deposition of calcium
carbonate or phosphate on the non-collagenous or
collagenous matrix. Their growth has been consi-
dered to be isometric to somatic growth, at least
before the somatic growth becomes asymptotic.
However, Secor and Dean [1] have recently indi-
cated the presence of two components in otolith
growth: one is an endogenoulsy defined in-
cremental growth, which can occur during periods
of decreased or arrested somatic growth, and the
other is a continuous growth within a day in some
proportion to daily somatic growth. They have
proposed a daily increment packing (DIP) model
for such a mode of otolith growth. The DIP model
suggests that somatic and otolith growth is not
necessarily isometric in detail.

Somatic growth is under endocrine control, and

Accepted June 28, 1989
Received June 1, 1989

it has repeatedly been reported that the removal of
the pituitary gland results in cessation of fish
growth in length [2-4]. Pickford [2] examined
otoliths and scales from hypophysectomized kil-
lifish and found no new circulus formation in the
scales and no new deposition on the otoliths. Her
findings came from a morphological comparison of
the outermost margin of these tissue between
hypophysectomized and control fish. Therefore, it
remains uncertain how, quantitatively, hypophy-
sectomy affects the growth rate of otoliths and
scales.

The present study was undertaken to examine
quantitatively long-term effects of hypophysecto-
my on the growth and calcification of otoliths and
scales using tetracycline-labeled goldfish.

MATERIALS AND METHODS

The goldfish, Carassius auratus, weighing aprox-
imately 21.7 g were obtained from a commercial
dealer and acclimated to experimental conditions
for at least 2 weeks before use. They were
maintained at 23+1°C and fed carp pellets once a

274 Y. MUGIYA

day throughout the acclimation and experimentai
periods.

Twenty fish were given a single intraperitoneal
injection of tetracycline (Takeda Chemical Ind.) at
a dose of 0.1 mg/g body weight and allowed to
recover for 4 days. Then, they were randomly
divided into 2 groups and one group was
hypophysectomized by the opercular approach
technique [5]. The other group was subjected to
drilling of the prootic bone only, and used as
sham-operated control.

Hypophysectomized and sham-operated fish
were separately placed into one of the two com-
partments halved by a wire-netting in a 1201 tank
containing aerated and filtrated 0.25% NaCl tap
water (Na conc. 43.1 mM and Ca conc. 0.2 mM).
They were taken out for measurement of body size
twice a month and returned to the other compart-
ment each time. This replacement made their
environmental conditions as even as possible. No
individual discrimination was made. The dividing
netting was removed after about 3 months and
both groups were maintained in the same compart-
ment from then on. Nevertheless, it was easy to
distinguish the experimental fish from the control,
because the red body color had faded to almost
white in the hypophysectomized fish. They were
sacrificed 32 weeks after operation.

Blood was collected from the caudal vessels by
cutting the tail of the fish and draining it into
heparinized (as Li salt) capillary tubes. After
centrifugation, the separated plasma was stored at
—40°C for 1 day and analysed for calcium and
sodium concentrations by emission spectropho-
tometry (Hitachi, 518).

Somatic growth was expressed as a relative
growth rate (RGR) after Ricker [6].

B-A 1000

haba daca bao
where A and B represent initial and final sizes,
respectively, and T-t is the change in time (32

weeks). RGR in scales and otoliths was calculated
in the same manner.

After blood collection, 6 scales were removed
with forceps from the antero-dorsal part of each
fish. Special attention was paid not to include

regenerating scales. They were carefully rinsed

several times with distillated water and examined
for tetracycline-induced fluorescence under ultra-
violet illumination (Nikon, EFD). The distance of
accretive growth during the experiment was mea-
sured with a calibrated eye-piece to the nearest ~m
in the anterior part of scales and RGR was calcu-
lated. After measurement, scales were stained
with 1% silver nitrate [7] for calcium detection.

Otoliths (asterisci and lapilli) were dissected,
rinsed with distillated water and dried. After
measurements of weight and length in the dorso-
ventral axis, asterisci were embedded within a
mass of methacrylate of suitable hardness and
ground down to approximately 100 “m in thick-
ness so that the polished plane was parallel to the
axis accross the center. Ground otoliths were
subjected to microscopic examination for fluoresc-
ence. The distance of accretive growth was mea-
sured in the dorsal and ventral directions. Only
weight and length in the antero-posterior axis were
measured in lapilli.

Student’s t-test for unpaired observations was
applied to assess statistical significance of differ-
ences between mean values. Significance was
accepted at a P-value of <0.05.

The degree of calcification of the newly formed
regions of scales and otoliths was estimated by
X-ray microanalyses. Scales and ground otoliths
were ultrasonically cleaned in xylene and/or distil-
lated water and mounted on a carbon holder with a
carbon tape. After being coated with carbon, they
were examined for elemental constituents at 20 kV
by spot analyses using a Horiba EMAX-2000 ener-
gy-dispersive X-ray microanalysis system. Spec-
trum collection time was 100 sec.

RESULTS

Hypophysectomy resulted in a marked reduc-
tion in somatic growth (Table 1). Relative growth
in weight gain was reduced by 70% in hypophysec-
tomized fish compared to the sham-operated con-
trol. No increase in length was found in the
experimental fish in spite of their active feeding.

Plasma sodium and calcium were significantly
affected by hypophysectomy (Table 2). Although
fish were maintained in 0.25% NaCl throughout
the course of the experiment, sodium concentra-

Hypophysectomized Effects in Goldfish 275

TABLE 1.
goldfish

Somatic growth in hypophysectomized and sham-operated

Sham-operated

Hyophysectomized

Body weight (g)

Initial 21.0+1.0 (8)* 22.4+1.0 (9)

Final 33.6+3.0 (7) 26.2+1.6 (8)

RGR** 18.8 5.3
Standard length (cm)

Initial 8.3+0.1 (8) 8.4+0.2 (9)

Final 9.3+0.3 (7) 8.3+0.1 (8)

RGR 3.77 —0.38

* Mean+SE (No. of fish examined).

** Relative growth rate.

TaBLE2. Plasma calcium and sodium concentrations (mEq/l) in hypophysecto-

mized and sham-operated goldfish

Sham-operated Hypophysectomized Significance
Ca 4.7+0.1 (7)* 5.2+0.2 (6) P<0.05
Na 135.2+1.1 (7) 129.9+0.7 (6) P<0.01

* Mean+SE (No. of fish examined).

TaBLeE 3. Otolith sizes in hypophysectomized and sham-operated goldfish
Sham-operated Hypophysectomized Significance
Length (um)
Asteriscus 2988 +64 (10)* 2605+22 (12) P<0.001
Lapillus 1957+31 (12) 1879+22 (12)
Weight (mg)
Asteriscus 7.9+0.3 (10) 6.7+0.3 (12) P<0.01
Lapillus** 8.0+0.4 (6) 6.4+0.2 (6) P<0.01

* Mean+SE (No. of otoliths examined).

** A pair of lapilli were pooled for weighing.

tions decreased in hypophysectomized fish (p<
0.01). On the other hand, hypophysectomy re-
sulted in an increase in plasma calcium concentra-
tions at the end of the experiment (p< 0.05).
Measurements of whole otoliths showed that
otolith growth was inhibited by hypophysectomy
(Table 3). Asterisci of hypophysectomized fish
gained significantly less (p<0.01), 13% and 15%
less in length and weight respectively, than those
of the control, while a significant (p<0.01) effect
of hypophysectomy on lapillus growth was found

only in weight, showing a 20%-reduced gain. Tet-
racycline-induced bands visually confirmed the in-
hibitory effects of hypophysectomy on otolith
growth in length (Fig. 1). Otolith growth toward
the dorso-ventral directions is presented as a rela-
tive growth rate (Table 4). Hypophysectomy re-
duced the rates by half in both directions (p<
0.001). The pattern of X-ray energy spectra was
almost the same between the experimental and
control otoliths. The heavy presence of calcium
with a trace of phosphorus was confirmed in the

276 Y. MuGIya

Hypophysectomized Effects in Goldfish

277

TaBLE4. Relative growth rates in otoliths and scales in hypophysectomized and

sham-operated goldfish

Sham-operated Hypophysectomized Significance
Otolith (asteriscus)
Dorsal 5.91+0.38 (10)* 3.05+0.19 (14) P<0.001
Ventral 4.93+0.33 (10) 2.44+0.20 (14) P<0.001
Scale 6.43+0.76 (18) 0.62+0.06 (20) P<0.001

* Mean+SE (No. of samples examined).

newly formed region following hypophysectomy
(Fig. 2). Therefore this region should be calcified.

Scale growth was markedly reduced by
hypophysectomy. Only 1 or 2 ridges were added to
the hypophysectomized fish scales during the ex-
periment, while 16-30 ridges were newly formed
in the control (Fig. 1). Relative growth rates
showed that hypophysectomy reduced scale
growth to about one-tenth of the control (Table 4).
Therefore, it is evident that scale growth was more
severely affected by hypophysectomy than otolith
growth. The newly formed region of scales follow-
ing hypophysectomy was positively stained by sil-

4K rere TI

CA

fie estnestis bot

0-00 5.12

10-23

Fic. 2. X-ray energy spectrum from the newly formed
region (cf. Fig. 1b) of an otolith after hypophy-
sectomy. The major elemental constituent is Ca.

ver nitrate. The marginal osteoid was poorly
developed in the scales (Fig. 1). X-ray microanaly-
ses revealed no essential difference in the energy
spectra between experimental and control scales.
Major elemental constituents in the newly formed
region were calcium and phosphorus (Fig. 3),
which are probably present in apatite. No spec-
trum difference was found between the ridge and
inter-ridge regions.

4K T

0-00 5.12 10-23

Fic. 3. X-ray energy spectrum from the newly formed
region (cf. Fig. 1d) of a scale after hypophysectomy.
The major elemental constituents are Ca and P.

DISCUSSION

The pituitary secretes various kinds of hor-

Fic. 1.
tetracycline-induced fluorescent bands.
performed (Figs. 2 and 3).

from sham-operated (a) and hypophysectomized (b) fish.

Otoliths (asterisci) and scales of goldfish at the dorsal and anterior margins, respectively. Arrows represent
White asterisks indicate the spots when X-ray microanalyses were
Scale bars are 0.1 mm. (a) and (b) Fluorecsent images of vertically ground otoliths

(c) and (d) Fluorescent images of scales from

sham-operated (c) and hypophysectomized (d) fish. (e) and (f) Scales stained with silver nitrate. The osteoid (0)
is well developed in sham-operated fish (e), but hypophysectomized fish (f) lack the osteoid.

278 Y. Muaiyva

mones. Therefore, the removal of the gland
severely affects the fish: cessation in somatic
growth [2], disturbances in ion and water balance
[8], and arrest in sexual development [9]. The
present results of the removal of the pituitary
induced cessation in somatic growth in the length
of goldfish, which agrees with earlier reports in
other fish species [2, 4]. Scale growth was marked-
ly inhibited by hypophysectomy, while otoliths
continuted to grow at a reduced rate after
hypophysectomy, showing an allometric rela-
tionship to somatic growth. Outgrowth and cal-
cification in scales are preceded by the formation
of marginal osteoid [10], which is collagenous in
nature. In hypophysectomized fish, little or no
osteoid was found at the scale margin. Therefore,
hypophysectomy appears to affect scale growth by
retarding osteoid formation. Retardation in
osteoid formation probably resulted from the lack
of somatotropin, which stimulates the synthesis of
proteins and other matrix-related macromolecules
via somatomedin [11]. Replacement therapy with
somatotropin was effective in _ restoring
hypophysectomy-induced cessation in somatic
growth in killifish [3] and bullhead [12].

The above-mentioned interpretation is fun-
damentally applicable to the effect on otolith
growth. However, since otoliths are a highly
calcified tissue and the amount of organic matrix is
much less than that of scales [13], the relative
contribution of the matrix to accretive growth may
be not so critical as in scales. According to Degens
et al. [14], otoliths grow by two processes: the new
deposition of calctum as carbonate on non-
collagenous matrix and successive growth of the
crystals by epitaxy. The latter is a physicochemical
process and therefore may be less affected by
hypophysectomy. Even if the former phase of
matrix formation is reduced to some extent, oto-
liths could continue to grow at a reduced rate
through the process of crystal growth, because the
endolymph bathing otoliths are considered to be
highly supersaturated to the otolith mineral, cal-
cium carbonate, in goldfish [15].

Allometric growth of otoliths to somatic growth
has been reported especially in subopitimal en-
vironmental conditions such as low temperatures
and food deprivation [16, 17]. Secor and Dean [1]

were the first to explain the allometry by daily
increment formation in otoliths. They deduced
that a range of somatic growth effects on otolith
allometry can be explained by two associated
processes: incremental growth of otoliths during
periods of arrested somatic growth and continuous
growth in some proportion to somatic growth.
They have proposed to call such otolith growth
pattern the daily increment packing model.
Although the DIP model cannot be directly ap-
plied to the hypophysectomy-induced asynchro-
nous relationship, it will be necessary to examine
incremental microstructures in the newly formed
region of otoliths in hypophysectomized fish.

Hypophysectomy induced hypocalcemia in kil-
lifish [18, 19] and goldfish [5]. One possible cause
for this is a reduction in branchial uptake of
calcium [5]. However, hypercalcemia was the case
with the present study which focused on long-term
effects of hypophysectomy. Since the ex-
perimental fish showed no gain in length during the
experiment, an increase in bone mass is not ex-
pected. Since most calcium (more than 80%)
taken up by goldfish is distributed in bone and
scales [13], cessation in new skeletal formation
would result in an excess of plasma calcium even at
a reduced rate of branchial calcium uptake. There-
fore, the present increase in plasma calcium con-
centrations following hypophysectomy can be
attributed to the secondary effect of reduced
skeletal tissue growth.

Then, why did hypophysectomy induce hypocal-
cemia and hypercalcemia? Mugiya and Odawara
[5] showed that calcium incorporation into scales
was not affected by hypophysectomy 2 weeks after
operation. This is supported by the present result
that 1 or 2 ridges were newly added to scales after
hypophysectomy. Therefore, it is reasonable to
suppose that inhibitory effects of hypophysectomy
on skeletal tissue growth become effective with a
lagged period of a few weeks after operation. This
would partially explain the conflicting effects of
hypophysectomy on plasma calcium concentra-
tions in goldfish in the short-term and long-term
experiments.

Hypophysectomized Effects in Goldfish 279

ACKNOWLEDGMENTS

The author is grateful to Professor N. Watabe, Uni-
versity of South Carolina for a critical reading of the
manuscript. Thanks are also due to Messrs. F. Odawara
and T. Hakomori for their techniqual assistance during
the experiment.

REFERENCES

1 Secor, D. H. and Dean, J. M. (1989) Somatic
growth effects on the otolith-fish size relationship in
young pond-reared striped bass, Morone saxatilis.
Can. J. Fish. Aquat. Sci., 46: 113-121.

2 Pickford, G. E. (1953) A study of the hypophysec-
tomized male Killifish, Fundulus heteroclitus
(Linn.). Bull. Bingham Oceanogra. Coll., 14: 5-41.

3 Pickford, G. E. (1954) The response of hypophysec-
tomized male killifish to purified fish growth hor-
mone, aS compared with the response to purified
beef growth hormone. Endocrinology, 55: 274-287.

4 Ball, J. N. (1965) Partial hypophysectomy in the
teleost Poecillia: Separate identities of teleostean
growth hormone and teleostean prolactin-like hor-
mone. Gen. Comp. Endocrinol., 5: 654-661.

5 Mugiya, Y. and Odawara, F. (1988) Effects of
hypophysectomy and replacement therapy with
Ovine prolactin on serum calcium levels, and cal-
cification in otoliths and scales in goldfish. Nippon
Suisan Gakkaishi, 54: 2079-2083.

6 Ricker, W. E. (1979) Growth rates and models. In
“Fish Physiology, Vol. 8”. Ed. by W. S. Hoar, D. J.
Randall and J. R. Brett, Academic Press, New
York, pp. 677-743.

7 McManus, J. F. A. and Mowry, R. W. (1960)
Staining Methods. Harper & Row Pub., New York,
pp. 191-222.

8 Griffith, R. W. (1794) Pituitary control of adapta-
tion to fresh water in the teleost genus Fundulus.
Biol. Bull., 146: 357-376.

9 Yamazaki, F. (1965) Endocrinological studies on

11

12

13

14

15

16

17

18

19

the reproduction of the female goldfish, Carassius
auratus L., with special reference to the function of
the pituitary gland. Mem. Fac. Fish. Hokkido
Univ., 13: 1-64.

Kobayashi, S., Yamada, J., Maekawa, K. and
Ouchi, K. (1972) Calcification and nucleation in
fish-scales. Biomineralisation, 6: 84-90.

Nasu, N., Kato, Y. and Suzuki, F. (1980) Bioassay
of somatomedin activity by cultured rabbit chondro-
cytes. Bone Metab., 13: 294-299, (In Japanese).
Kayes, T. (1977) Effects of hypophysectomy, beef
growth hormone replacement therapy, pituitary
autotransplantation, and environmental salinity on
growth in the black bullhead (Ictalurus melas). Gen.
Comp. Endocrinol., 33: 371-381.

Ichii, T. and Mugiya, Y. (1983) Comparative
aspects of calctum dynamics in calcified tissues in the
goldfish, Carassius auratus. Bull. Japan. Soc. Sci.
Fish., 49: 1039-1044.

Degens, E. T., Deuser, W. G. and Haedrich, R. L.
(1969) Molecular structure and composition of fish
otoliths. Mar. Biol., 2: 105-113.

Shichiri, T. (1985) Growth of statolith. J. Japan.
Assoc. Cryst. Growth, 12: 42-56.

Marshall, S. L. and Parker, S. S. (1982) Pattern
identification in the microstructure of sockeye sal-
mon (Oncorhynchus nerka) otoliths. Can. J. Fish.
Aquat. Sci., 39: 542-547.

Campana, S. E. and Neilson, J. D. (1985) Micros-
tructure of fish otoliths. Can. J. Fish. Aquat. Sci.,
42: 1014-1032.

Pang, P. K. T., Schreibman, M. P. and Griffith, R.
W. (1973) Pituitary regulation of serum calcium
levels in the killifish, Fundulus heteroclitus L. Gen.
Comp. Endocrinol., 21: 536-542.

Pang, P. K. T., Schreibman, M. P., Balbontin, F.
and Pang, R. K. (1978) Prolactin and pituitary
control of calcium regulation in the killifish, Fundu-
lus heteroclitus. Gen. Comp. Endocrinol, 36: 306-
316.

.}
f Thier wal at GPa ape |
7 Tart. wor Sbweaue te
nue tt by ey pA (as
Aa joel. yt,
fava F oe int ae
| me Nig emery
, } eM WATE eae TET ete i
} : au hen TT a ea n
(GPM Titieiite nthe da OS
woe, Abel, andl aster
1yt y4 Mia) T, wep Mj
Nn, Anos , Bee,
nai fm auwloaaigeacnivne
’ tiki déalt (iy ht Woy
OE Anti brel ste
\ Sai Gos | SeY OlideP 2a
yy red yin Ohad ut; REP
4 a eT fo et ioe 4
ba} oi Rittn |
; M xsore (i Tiago 1 A
bs . ay wih Vay
¥ wit bid ¢
‘0 ¢ F | a Lisits ee
ar d beve el mee ly 2
i nn i ayia o
{ 5
a f y
4 | s
1 =~ beat
2
Lay aye

asthe i dita

tet
Pa ae eer nite

—_

— se ni, veg

~ CA
=e af

hie VRP be

ian

meet “tok

, bestangice’ Sey SubtheteY g mre
(ees eer silty ccstaOee teat lant

=) lta? aed Sap
io sana py Dingoes ;

sae Ws ‘4 fs i is
alebinal Cute aceentaelen V
HOM. ¥ eds lb AO,
Wy ftir MERE MT oth
a) tale. tins

} i is ie nape

‘al

ae 4

spelt :

iw y

Lipo Ws ave pe Clea
{ re" airem wv ef a yay
NF i: -
rT “opt? OS ire uh ae (rae
vi
3 oh
ew | |
11ey7
vt iis
J a-) Ait 7.
ue oe

ZOOLOGICAL SCIENCE 7: 281-286 (1990)

Urea Stimulation of Pituitary Pars Intermedia Cells of
Suckling Mice under Copious Drinking

YASUO KOBAYASHI and MANABU OKADA

Department of Biology, Faculty of Science, Okayama University,
Okayama 700, Japan

ABSTRACT— Ultrastructural alterations in cells of the pars intermedia (PI) of the pituitary gland were
investigated morphometrically in suckling mice being fed with the pelleted food (Group J), in suckling
dams given a liquid diet containing 11.5% milk powder (Group II) and in those received a 11.5% liquid
milk diet supplemented with 1% urea (Group III) for 5 days from day 13 to day 17 of lactation. The
percent area of the cytoplasm occupied by the rough endoplasmic reticulum increased from 11.9+1.0%
(Group I) through 17.9+1.0% (Group II) to 36.9+1.1% (Group III). The numerical density of
immature Golgi granules also increased from 0.9 +0.06 (Group I) through 1.2 +0.09 (Group II) to 1.5+
0.09 (Group III). Whereas the numerical density of secretory granules decreased from 2.3+0.11
(Group I) through 1.9+0.07 (Group II) to 1.0+0.05 (Group III). These results indicated that the
release of secretory granules and the protein synthesis increased significantly in PI cells of suckling mice
fed a liquid milk diet and, further, the urea supplement to the liquid milk diet elicited the secretion

© 1990 Zoological Society of Japan

activity of PI cells more significantly than those of animals received the liquid milk diet only.

INTRODUCTION

Our previous electron microscopic and mor-
phometric studies have demonstrated marked
hypersecretion in cells of the pars intermedia (PI)
of the mouse pituitary gland in response to dietary
sodium restriction [1-3] suggesting a new role for
the PI hormone(s) as a major pituitary factor in the
regulation of aldosterone secretion by the adrenals
[4, 5]. This new proposition has now been sup-
ported and confirmed by in vitro studies indicating
that pro-opiomelanocortin-derived peptides such
as alpha-MSH, beta-MSH and gamma-MSH are
involved in the control of aldosterone secretion by
the adrenals from rats and human beings with
idiopathic hyperaldosteronism [6, see also refer-
ences]. However, several lines of evidence have
shown that the renin-angiotensin system is a prime
regulator of aldosterone secretion [7, 8]. Thus, the
physiological interrelationship between the pro-
opiomelanocortin-derived PI prptides and the re-
nin-angiotensin system in the regulation of alsos-
terone secretion has not yet been settled.

Accepted July 18, 1989
Received May 30, 1989

Our recent morphometric studies have also de-
monstrated that experimental copious drinking has
an effect to cause marked hypersecretion in PI
cells of the mouse pituitary [9] and that the PI cells
of new-born mice indicate structural hypersecre-
tion when, after Shr of separation, pups are
reunited with their mother and allowed to suck
their mother’s breast for 1 hr [10]. These results
suggest another new role of PI peptides in the
regulation of hydro-mineral metabolism in mice.

On the other hand, one of the interesting be-
haviours of mice observed during lactation is a
kind of coprophagy [11]. The suckling mice used
to lick pups’ urine and stools off their rump during
nursing. In addition, there is a parallelism be-
tween a high blood urea concentration and the
enlarged pars intermedia in mice with hereditary
nephrogenic diabets inspidus [12, 13]. Then the
present study was designed to investigate the
effects of experimental polydipsia [9] and of com-
bined urea and polydipsia on pituitary PI cells of
suckling dams whose water consumption is about 4
times as much as that of age matched nulliparous
mice. Plasma osmolality and hematocrit were also
evaulated since the specificity of the effect of
osmolality on aldosterone secretion has been re-

282 Y. KoBAYASHI AND M. OKADA

cently reported [14, 15].

MATERIALS AND METHODS

Albino mice of the ICR/Jcl strain were housed at
a temperature of 22+1°C and 12hr-periods of
light and darkness (7.00-19.00 light on). Primigra-
vid mice were isolated single in a cage provided
with a pelleted mouse food and tap water. One
day after the parturition, the number of siblings
was reduced and adjusted to 6 pups. On day 12 of
lactation, suckling dams were divided into three
groups, and the following experiments were under-
taken during the period from day 13 to day 17 of
lactation. The first group served as controls receiv-
ing a pelleted food and distilled water. The second
group was subjected to copious drinking; animals
were deprived of a pellted food and water and ,
instead, were fed with a liquid diet consisting of
powdered milk (Yuki-zirushi Neomilk, Yukizir-
ushi, Tokyo) at the concentration of 11.5%, the
same concentration used in nursing a human baby.
The third group also received a liquid diet of
11.5% powdered milk supplemented with 1%
urea. The final concentration of urea used was
determined in a way to keep the consumption of a
urea plus milk liquid diet on a level with that of a
milk liquid diet. Oral intake of a liquid diet was
measured daily and the body weight of both suck-
ling dams and pups was weighed. On day 17 of
lactation, the suckling mice were killed by de-
capitation and the blood was collected from trunk
vessels into heparinized capillary tubes. The
hematocrit value was obtained after centrifugation
of the blood samples for 10 min at a speed of 9,500

TABLE 1.

rpm. Plasma osmolality (mOsm/kg) was measured
using Shimazu OSM-1 osmometer. The pituitaries
were removed immediately after the sacrifice and
small pieces of parasagittal slices of specimens
were fixed in the solution of 1% glutaraldehyde
and 1% paraformaldehyde with phosphate fuffer
at pH 7.4 for 1 hr followed by 1% OsO, (pH 7.4)
for another lhr. After dehydration specimens
were embedded in Quetol 812. Ultrathin sections
were stained with uranyl acetate and lead citrate
and examined with an H-11E Hitachi Electron
Microscope. Electron micrographs were taken at
original magnification of 2,500 and enlarged opti-
cally at 10,000 times.

In each group approximately 50 PI cells of
electron micrographs were chosen at random for
morphometric analysis. The area of the cyto-
plasm, nucleus and Golgi region was measured

TaBLeE2. Hematocrit and osmolality of suck-
ling mice given a pelleted food and distilled
water (Group I), a 11.5% milk diet (Group
II) and a 11.5% milk diet added with 1%
urea (Group III) on day 17 of lactation

Hematocrit (%) (ibe aiee)
con re 48.7+0.6 305+2.2 (3)
(Geeta) 57.4+0.6* 309+42.2 (6)
Milk+ Urea 56.5+0.5* 308+1.3 (8)

(Group III)

*) Mean+S.E.M., * Group I vs. Group II and
Group III, p<0.01. No. of animals is shown in
parentheses.

Water (D.W.) and liquid diet consumption (ml/10 g b.w./day) by suckling mice being fed with a

pelleted food (Group I) , with a 11.5% milk diet (Group IT) and with a 11.5% milk diet added with 1%
urea (Group III) from day 13 to day 17 of lactation

Eee 13 14 15 16 17

| ene 1540.89 7.8+0.7 7.540.5 6.440.3 8.5+0.6 (3)
(Group. Il) a tn 13.8+ 1.0" 14340" 17.0+1.5* 16.441.2* (6)
Milk + Urea ya8t0.6" mo, 7 16.6+1.0* er a. 1.3* (8)

(Group III)

———————— ———————

“”) Mean+S.E.M., * Group I v.s Group II and Group III, p<0.01 No. of animals is shown in parentheses.

Urea and Pituitary PI Cells 283

Fic. 1. Part of pituitary pars intermedia cells of the suckling mouse received pelleted food and distilled water. Note
numerous secretory granules (SG), scanty rough endoplasmic reticulum (ER) and rod shaped mitochondria (M)
in the cytoplasm. The nucleus (N) shows marked indentation with a prominent nucleolus.

Fic. 2. Part of pipuitary pars intermedia cells of the suckling dam fed a liquid diet containing 11.5% milk powder and
1% urea. Note well developed rough endoplasmic reticulum (ER), extensive Golgi apparatus (G), sparse
secretory granules (SG) and scattering mitochondria (M). The nucleus (N) is irregular in shape.

TaBLE 3. The percent area of the rough endoplastic reticulum (% r-ER), the numerical density of
secretory granules (Secretory G./um7) and of immature Golgi granules (Golgi G./ym7*) in the
Golgi region, and the total number of Golgi granules in the median of Golgi area (Golgi G./Med.)
in suckling dams fed with a pelleted food and distilled water (Group I), with a 11.5% milk diet

(Group II) and with a 11.5% milk diet supplemented with 1% urea (Group III) from day 13 to day
17 of lactation

% TER ee Giym? G/Med.”
(Gas 11.9+1.0 2.3+0.11 0.9+0.06 9.0 (3)
(Goon 17.9+1.0** 1.9+0.07* 1.2+0.09° 12.6 (6)
Pe 36.941.1**tt 1.0+0.05**** 1.5+0.09*** 25.5 (6)

* Total number of Golgi granules in the median of the Golgi area was obtained multiplying the number of
Golgi granules per unit area by the median of the Golgi area. °) Mean+S.E.M., Group I vs. Group II and

Group III; * p<0.01 ** p<0.001. Group II vs. Group III; ‘p<0.01 “p<0.001. No. of animals is shown in
parentheses.

284 Y. KOBAYASHI AND M. OKADA

with a planimeter (X-PLAN 360, Ushikata,
Tokyo). The number of secretory granules and of
Golgi immature granules was counted in each cell.
The numerical density of secretory granules and of
Golgi immature granules was calculated, in which
the cytoplasmic area for secretory granules was
estimated subtracting the Golgi area from the
whole cytoplasm of the cell. The percent area of
the cytoplasm occupied by the r-ER was obtained
with a Planimex 25 (Nihon Regulator, Japan).
Statistical significance was assessed with Student’s
t-test.

RESULTS

1. Oral intake of distilled water and of liquid milk
diet supplemented with or without urea

During the experimental period of lactation
from day 13 to day 17, suckling dams (Group I)
drank distilled water in the mean of 7.5+1.5 ml/10
g b.w./day (Table 1). This water consumption of
lactating mice was about 4 times as much as that of
age matched unlliparous female mice (1.8+0.3
ml/10g b.w./day, data not presented). When
suckling mice were deprived of a pelleted food
and, instead, given a 11.5% milk diet (Group II)
their daily liquid consumption increased to 181%,
177%, 191%, 266% and 193% of the each corres-
ponding control value (Group I), respectively,
from day 13 to day 17 of lactation (Table 1).
Likewise, lactating mice given a 11.5% milk diet
plus 1% urea (Group III) showed copious drinking
in a range of 171%, 179%, 221% 266% and 216%
of the each control value (Group I) during the
same period from day 13 to day 17 of lactlation
(Table 1).

2. Hematocrit and plasma osmolality

Hematocrit of the blood increased to 118% in
suckling mice given a liquid milk diet (Group II)
and to 116% in those fed a milk plus urea diet
(Group III) when compared to that of the control
(Table 2). Plasma osmolality in the Group II and
Group III was not significantly different from that
of the control Group I (Table 2).

3. Electron microscopy and morphometry of PI
cells of the pituitary gland

In the control suckling mice (Group I) PI cells of
the pituitary appeared to contain numerous secre-
tory granules in the cytoplasm. The rough endo-
plasmic reticulum (r-ER) was pooly developed
(Fig. 1). Whereas PI cells of the suckling dams
given milk plus urea liquid regimen showed a
cellular sign of hypersecretion; well developed
r-ER, extensive Golgi apparatus and sparse secre-
tory granules in the cytoplasm (Fig. 2).

The percent area of the cytoplasm occupied by
the r-ER in suckling mice given liquid milk diet
(Group II) and in those fed milk plus urea diet
(Group III) increased to 150% and 310% of the
control value, respectively (Table 3). The numer-
ical density of immature Golgi granules in the
Group II and Group III increased to 133% and
167% of the control value, respectively. In this
case, however, the Golgi apparatus showed
marked hypertrophy especially in those of the
Group III. Thus the theoretical number of Golgi
granules per the median of Golgi area was culcu-
lated. The total Golgi granules of the median
Golgi area increased to 140% and 283% of the
control value in the Group II and Group III,
respectively (Table 3). On the contrary, the
numerical density of secretory granules in the
cytoplasm decreased to 83% and to 43% of the
control value in the Group II and Group III,
respectively.

DISCUSSION

The present study demonstrated an enhanced
activity in’ cells of the pars intermedia (PI) of the
pituitary in suckling mice given either a 11.5%
liquid milk diet or a 11.5% milk diet supplemented
with 1% urea for 5 days from day 13 to day 17 of
lactation. It is of interest to describe here that oral
intake of a urea plus milk diet resulted in marked
hypersecretion in PI cells of suckling dams more
significantly than that of a milk diet only.

The development of polydipsia and polyurea in
old male rats of the Wistar/Tw strain at the age of
16 months has been reported [16-18], but the
studies have not focused on the cytology of pituit-

Urea and Pituitary PI Cells 285

ary PI cells. Polydipsia (137% v/w/day) in mice
with hereditary nephrogenic diabetes inspidus
(DI) associated with marked hypertrophic and
hyperplastic PI has been well documented [12, 13,
19]. Further, electron microscopic observations
have revealed that predominant light PI cells con-
tain many electron dense core granules and the
sparse endoplasmic reticulum [20]. Thus, polydip-
sia in DI mice exerts also little effects on the
synthesis and release of hormones of PI cells in fine
structure. Therefore, it seems reasonable to
assume that these polydipsic animals including DI
mice as well as normal suckling dams (ca. 75%
v/w/day) are in a condition under which hydro-
mineral metabolism is regulated within a physiolo-
gical range. What is the exact meaning of a
physiological range in polydipsia remains to be
disclosed.

When the pelleted food is absent, suckling dams
drink a large amount of liquid milk (164% v/w/
day) during the test period, and their pituitary PI
cells show small but significant hypersecretion in
fine structure (Table 3). In this case animals are
considered to be in a nutritionally balanced condi-
tion since they are fed with a 11.5% milk diet that
is the same concentration used in a human baby.
Thus this small change in the PI cell morphology
despite copions drinking may be due to either a
supply of minerals from a milk diet or less amount
of water intake than that of the case of a 5%
glucose regimen (oral intake ca. 200% v/w/day)
[9].

An interesting new finding in the present study is
unexpected hypersecretion in PI cells of suckling
dams given a 11.5% milk plus 1% urea diet.
Morphometric data in two dimensional analysis
clearly indicate the sigificance of hypersecretion of
PI cells. The percent area of the cytoplasm occu-
pied by r-ER, a parameter of protein synthesis, is
2.06 times; and the number of Golgi immature
granules, a capacity of granule formation, is 2.02
times; whereas the numerical density of secretory
granules reflecting the release of secretory gra-
nules is 0.53 times, respectivety, when compared
to those of suckling mice given a milk diet only
(Table 3). These results imply that the rate of
degranulation exceeded the rate of both increased
protein synthesis and granule formation in PI cells

by twice in magnitude in suckling dams fed a urea
plus milk diet. This urea-induced activation of PI
cells in mice during lactation is the first report in
the present study.

An abnormally high blood urea concentration
and increased serum osmolality are apparently
associated with the hypertrophy (but not with the
hypersecretion) of the pars intermedia of mice
with hereditary nephrogenic diabetes inspidus [12,
24]. In their studies, however, causative rela-
tionships between the hypertrophy of the PI and
the high blood urea concentration in DI mice are
not explained. Although it is generally assumed
that the cell membranes are freely permeable to
urea this assumption has met the conflicting data
[21-23]. A high blood urea nitrogen (BUN)
concentration is associated with a shift of water
from the cells into the extracellular water and
blood. Therefore, when the blood urea nitrogen is
high, there is variable lowering of the serum
sodium concentration due to dilution of the ex-
tracellular water [23]. It is uncertain in the present
study whether the same mechanism of lowering
sodium concentration by a high blood urea nit-
rogen is involved. Alternatilvely, PI peptids(s)
being released by oral intake of excessive urea plus
milk diet may have some pivotal functions other
than the stimulation of aldosterone secretion by
the adrenals as reported before [4, 5]. Further
studies on urea mediated activation of PI cells in
mice are in progress.

ACKNOWLEDGMENTS

This work was supported in part by a Grant-in-Aid for
Scientific Research to Y. K. from the Ministry of Educa-
tion, Science and Culture of Japan (No. 62540568).

REFERENCES

1 Kobayashi, Y. (1974) Quantitative and microscopic
studies on the pars intermedia of the hypophysis. I.
Dietary effect of brown rice on the kindey, adrenal
and pituitary of CS7BL/6 mouse. J. Electron Micros-
copy, 23: 107-115.

2 Kobayashi, Y. (1974) Quantitative and microscopic
studies on the pars intermedia of the hypophysis. II.
Alteration of the pars intermedia and adrenal zona
glomerulosa of albino mice following sodium restric-
tion. Annot. Zool. Japon., 47: 221-231.

286

Kobayashi, Y. (1974) Quantitative and microscopic
studies on the pars intermedia of the hypophysis.
Ill. Effect of short-term administration of a sodium
deficient diet on the pars intermedia of mice. Cell
Tissue Res., 154: 321-327.

Kobayashi, Y. and Takema, M. (1976) A mor-
phometric study on the pars intermedia of the
hypophysis during impairment of the renin-
angiontensin-aldosterone system in sodium depleted
mice. Cell Tissue Res., 168: 153-159.

Kobayashi, Y. (1977) Morphological evidence for a
pituitary factor, the pars intermedia hormone, that
may stimulate alsosterone secretion by adrenals in
sodium depleted mice. In “Hormonal Regulation of
Body Fluid.” Ed. by H. Oide and Y. Kondo, Proc.
14th Gunma Symp. Endocrinol., pp. 61-74.
Brownie, A. C. and Pedersen, R. C. (1986) Control
of aldosterone secretion by pituitary hormones. J.
Hypertension, 4(suppl. 5): s72—-s75.

Davis, J. O. (1962) The control of aldosterone
secretion. The Physiologist, 5: 65-86.

Swartz, S. L., William, G. H. Hollenberg, N. K.
Dluhy, R. G. and Moor, T. J. (1980) Primacy of the
renin-angiotensin system in mediating the aldoster-
one response to sodium restriction. J. Clin. Endocri-
nol., 50: 1071-1074.

Kobayashi, Y., Kumazawa, T. and Takeuchi, M.
(1984) A new method for inducing copious drinking
and the accompanying stimulation of the pars inter-
media of the mouse pituitary gland. Arch. Histol.
Japon., 47: 71-77.

Kobayashi, Y. and Takeuchi, M. (1985) Postnatal
development of inhibitory control and suckling-
induced activation of pituitary pars intermedia in
infact mice. Proc. Japan Acad., 61(B): 37-40.
Ebino, K. Y., Yoshinaga, K., Saito, T. R. and
Takahashi, K. W. (1988) Coprophagy as an innate
behavior in the mouse. Zool. Sci., 5: 863-868.
Naik, D. V. and Sokol, H. W. (1970) The hypotha-
lamo-hypophyseal neurosecretory system in mice
with vasopressin-resistant urinary concentrating de-
fects. Gen. Comp. Endocrinol., 15: 59-69.

13

14

15

16

17

18

20

22

23

24

Y. KOBAYASHI AND M. OKADA

Naik, D. V. (1970) Pituitary-adrenal relationships
in mice with hereditary nephrogenic diabetes inspi-
dus, with special emphasis on the neurohypophysis
and pars intermedia. Z. Zellforsch., 107: 317-342.
Schneider, E. G., Radke, K. J., Ulderich, D. A.
and Taylor, Jr., R. E. (1985) Effect of osmolality on
aldosterone secretion. Endocrinology, 166: 1621-
1626.

Taylor, Jr., R. E., Glass, J. T., Radke, K. J. and
Schneider, E. G. (1987) Specificity of effect of
osmolality on aldosterone secretion. Am. J. Phy-
siol., 252: E118-E123.

Kobayashi, Y. and Kawashima, S. (1980) Polydipsia
and polyurea in aged male rats of the Wistar/Tw
strain. Proc. Japan. Acad., Ser. B., 56: 643-648.
Kobayashi, Y. and Kawashima, S. (1984) Age-
related changes in the water and electrolyte meta-
bolism in male rats of the Wistar/Tw strain. Exp.
Gerontol., 129: 107-113.

Kawashima, S., Kawamoto, K. and Kobayashi, Y.
(1986) Aging of the hypothalamo-neurohypophysial
system and water metabolism in rats. Zool. Sci., 3:
227-244.

Naik, D. V. (1972) Salt and water metabolism and
neurohypophyseal vasopressor activity in mice with
hereditary nephrogenic diabetes inspidus. Acta En-
docrinol., 69: 227-244.

Naik, D. V. (1972) Electron microscopic studies on
the pars intermedia in normal and in mice with
hereditary nephrogenic diabetes inspidus. Z. Zell-
forsch., 133: 415-434.

Bozler, E. (1961) Distribution of non-electrolytes in
muscle. Am. J. Physiol., 200: 651-655.

Bradbury, M. W. B. and Coxon, R. V. (1962) The
penetration of urea into the central nervous system
at high blood levels. J. Physiol., 163: 423-435.
Goldberger, E. (1980) A primer of water, electro-
lyte and acid-base syndromes. Les and Febiger,
Philadelphia, 6th ed., pp. 460-462.

Naik, D. V. and Valtin, H. (1969) Hereditary
vasopressin resistant urinary concentration defects
in mice. Am. J. Physiol., 217: 1183-1189.

ZOOLOGICAL SCIENCE 7: 287-295 (1990)

Formation of Ganglions and Stomodaeum in Normal and Separate
Embryos of Horseshoe Crab, Tachypleus tridentatus

Tomio Irow, TatsuyA Masupa and Korcni Sexicucut!

Department of Biology, Faculty of Education, Shizuoka University,
Shizuoka 422, and 'Jobu University, Shin-Machi,
Gunma 370-13, Japan

ABSTRACT—The process of formation of ganglions and stomodaeum was examined in normal
embryos and separate embryos whose stomodaea did not pass through the middle of the nervous system.
The following facts were established. 1) The structure of the stomodaeum is recognizable aftre stage 14.
2) The cell masses of ganglions are observed clearly after stage 16; the commissures are stained by eosin
after stage 19. 3) The final number of ganglions is settled by stage 19; this means that the formation of
segment primordia is settled by stage 19. 4) The stomodaeum passes through the area between the
commissure of the first prosomatic ganglions (chelicera segment) and that of the second ganglions. 5)
The crossing of the stomodaeum and the nervous system is thought to be formed at stage 18-19. 6) The
“crossing is constructed during the clustering of the cells composing the brain; the stomodaeum does not
determine the pathway for the clustering of these cells. 7) The position of the mouth is not appropriate
as an indicator for determination of homologous segments among arthropodan species. 8) The crossing
does not constitute a valid reason for rejecting the idea that Deuterostomia have originated from

© 1990 Zoological Society of Japan

Protostomia.

INTRODUCTION

The morphology and embryology of the
horseshoe crab have been described by many
authors [Tachypleus tridentatus: 1-3; Limulus
polyphemus: 4-7]. However, not all structures
and processes have yet been clarified. Study of the
formation of ganglions and stomodaeum, especial-
ly the crossing between the nervous system and the
stomodaeum, remains incomplete. The present
study tries to make this process clear for the
Japanese horseshoe crab, Tachypleus tridentatus.
For this purpose, we examined the formation of
ganglions and the stomodaeum in normal
embryos. We also examined the process in sepa-
rate embryos whose stomodaea did not pass
through the middle of the nervous system. Such
embryos are formed by treatment with calcium-
free seawater or DNA synthesis inhibitors [8, 9].

The cause of release of the crossing of the
nervous system and stomodaeum is discussed in

Accepted April 28, 1989
Received February 22, 1989

the light of cell construction during the process of
embryonic development. As the crossing is a
remarkable characteristic of Protostomia, the pos-
sibility for change of form and structure in macro-
evolution is also considered.

MATERIALS AND METHODS

Adult Japanese horseshoe crabs, Tachypleus
tridentatus (Chelicerata, Arthropoda), collected in
north Kyushu, Japan, were brought to Shizuoka
University, where eggs were inseminated artificial-
ly. The developmental stages of the embryos were
determined from Sekiguchi’s normal table [2].

The separate embryos were obtained by 24-hr
treatment with calcium-free seawater, 10 'M
NaHCO; or inhibitors of DNA synthesis (10? and
5x10-7M__ hydroxyurea and 10-25 g/ml
azaserine). The treatment stages are described in
the results.

Normal embryos and ones given these chemical
treatments were stained vitally with neutral red
and observed under a stereomicroscope. Normal
and treated embryos were also fixed in Bouin’s,

288 T. Irow, T. MASUDA AND K. SEKIGUCHI

Carnoy’s, or FAA (formalin-70% ethanol-acetic
acid, 5:15:1) solutions, embedded in celloidin
and paraffin, and sectioned at 5-20 um. The
sections were stained with Mayer’s hematoxylin
and eosin. Some normal embryos and larvae were
dissected for an examination of the nervous system
and alimentary canal.

RESULTS

Formation of ganglions and stomodaeum in normal
embryos

In this paper, the ganglions in the front area of
the 1st prosomatic segment (except for ganglions
of the 1st segment) are called the brain.

Enlargement of the germ disc of the horseshoe
crab finished at stage 10 (stage of completion of
germ disc). Obvious morphogenetic movement
started at stage 10. Observation with time-lapse
cinemicrography in the previous study [10] had
shown that the formation of the stomodaeum
begins at the stage of morphogenetic movement
(stage 11). The stomodaeum appeared at the
anterior margin of the embryonic area. The posi-
tion of the stomodaeum differed from that of the
blastopore. In this process, two narrow bands are
formed along the median body axis; these narrow
bands may be early nervous systems.

At the stage of the appearance of prosomatic
appendages (stage 14), the stomodaeum was
observed as a tubular structure. The existence of
neuroblasts which would become the brain was

recognized, but construction of the brain was not
yet complete (Fig. 1).

Ganglions could be observed in embryos fixed at
the stage of development of prosomatic append-
ages (stage 16). The opening of the stomodaeum
(mouth) could be observed clearly at the area in
front of the lst prosomatic segment (segment with
chelicera= 1st prosomatic appendages). However
the development of the brain and other ganglions
was incomplete, the commissures of ganglions in
particular being under-developed. The formation
of the epithelium of the stomodaeum proceeded
further at stage 16.

When embryos at the stage after the Ist
embryonic moulting (stage 18) were stained with
neutral red, nervous systems were stained clearly.
A brain and 9 pairs of ganglions could be
observed. The commissures of the ganglions were
not clear in the stained embryos or sectioned
specimens, although they were recognized in fixed
embryos. The mouth began to migrate posteriorly
at stage 18; it was situated in the region of the Ist
prosomatic segment at this stage.

At the stage after the 2nd embryonic moulting
(stage 19), the commissures became clear. The
neurofibers developed and were stained easily with
eosin. The mouth migrated to the region of the 3rd
prosomatic segment. The stomodaeum passed
through the area between the commissure of the
lst prosomatic segment and that of the 2nd one.

All the ganglions were formed by stage 19; that
is, the formation of all segment primordia was
complete at this stage. The mouth was situated in

Fic. 1.

The histological features in the neighboring region of the stomodaeum. A. Stage 14.
b; prospective region of the brain, s; stomodaea. The bar shows 0.05 mm.

B. Stage 16.

Ganglions & Stomodaeum in Horseshoe Crab 289

Fic. 2. The horizontal section of a 2nd instar larva. a; alimentary canal, b; brain, 1-6; each cephalothoracic
ganglion, Al-A3; each abdominal ganglion, B; border between prosome and opisthosome. The number of pairs
of prosomatic ganglions except for the brain is eight. The bar shows 0.5 mm.

290 T. Irow, T. MASUDA AND K. SEKIGUCHI

the region between the 3rd and 4th prosomatic
segments. The position was the same as that in the
adults.

The embryos at stage 21 (the stage after the 4th
embryonic moulting) have the same form and
structure as the Ist instar larvae. At this stage the

circumbuccal part (the structure of the peristome)
was formed completely. The stomodaeum was

well developed and differentiated. As a result, the
proventriculus (fore-gut) and intestine (mid-gut)
were also differentiated. However, there was yolk
intestine,

in the and the formation of the

Fic. 3. The opisthosomatic nervous system of the 2nd instar larva. 1-8; each opisthosomatic ganglion. The Ist and
2nd abdominal ganglions belong to the prosome. The Ist opisthosomatic ganglion is equal to the 3rd abdominal

one. The bar shows 0.5 mm.

Ganglions & Stomodaeum in Horseshoe Crab 291

alimentary canal was not complete.

The condition and position of the brain and
other ganglions of the 1st instar larvae were similar
to those of the 2nd instar larvae. In these larvae all
the prosomatic ganglions had commissures. In the
border area between the prosome and opistho-
some there were no ganglions without appendages
(Fig. 2). The prosome had a brain and 8 pairs of
prosomatic ganglions. The prosomatic ganglions
consisted of 6 pairs of cephalothoracic ganglions
and 2 pairs of abdominal ganglions (ganglions of
chilaria and operculum). The shape of the cepha-
lothoracic ganglions differed from that of the
abdominal ones. The opisthosome had 8 pairs of
ganglions (the rest were abdominal ganglions)
(Fig. 3). The shapes of the opisthosomatic gan-
glions were similar to those of the abdominal
ganglions in the prosome.

The alimentary canal of the 2nd instar larva was
complete. At this stage, formation of intestine was
finished and the larvae began to eat.

Release of the crossing of the nervous system and
stomodaeum in separate embryos

Calcium-free seawater, NaHCO; and inhibitors
of DNA synthesis induced the separate embryo
(Table 1, Fig. 4). The ventral plate of the separate
embryo was divided into an anterior region and a
posterior region.

The conditions of induction were as follows.
When embryos were treated for 24 hr with cal-
cium-free seawater or NaHCO; at stages 7, 8 and 9
(stage of enlargement of the germ disc, the gastru-

TABLE 1.

The frequency of formation of separate embryos.

la stage), they developed into the separate
embryos whose ventral plates were separated
mainly at the region between the 3rd and Sth
prosomatic segments. When treated for 24 hr at
stages 10 and 11 (stages of obvious morphogenetic
movement), the ventral plates of the treated
embryos were separated mainly at the 2nd and 3rd
prosomatic segments. Following treatment for 24
hr with an inhibitor of DNA synthesis at the stage
of enlargement of germ disc, the treated embryos
developed into separate embryos, whose ventral
plates were separated mainly at the 2nd and 3rd
prosomatic segments.

When embryos at the stage of enlargement of
the germ disc were treated with calcium-free sea-
water or NaHCO;, the connection between cells
composing the germ disc was weakened. Their
ventral plates were separated mainly at the region
between the 3rd and 5th segments of the cepha-
lothorax, which was formed in the process of
enlargement of the germ disc. When embryos at
the stage of enlargement of the germ disc were
treated by inhibitors of DNA synthesis, cell prolif-
eration of germ disc became incomplete and the
cell density of the germ disc became low in spite of
normal spreading of the germ disc. The incom-
plete germ disc was separated mainly at the 2nd
and 3rd prosomatic segments in the process of
obvious morphogenetic movement. During the
movement, the embryonic area elongated anter-
iorly and posteriorly at the region where the 2nd
and 3rd prosomatic segments had recently been
formed. Calcium-free seawater and NaHCO,

The embryos were

treated for 24 hr either at the stage of enlargement of the germ disc (I) or at the
stage of obvious morphogenetic movement (II)

Separate Developed
embryo embryo
Number (%) Number (%)
Hydroxyurea 5X10°*M 134 (32.1) 318 (74.2)
(1)
Azaserine 25 pg/ml 73 (16.7) 436 (68.4)
NaHCO; 107-'M 20 (19.2) 104 (13.4)
(11)
Ca**-free seawater 57 (29.2) 192 (43.3)

Normal seawater

6 ( 0.1) 6,829 (100.0)

292 T. Irow, T. MASUDA AND K. SEKIGUCHI

Fic. 4.

Examples of a normal embryo and separate embryos. A: A normal embryo (stage 20). B: An incomplete

separate embryo (stage 20) whose ventral plate is separated at the point between the 1st prosomatic segment and
the 2nd one. C: A complete separate embryo (stage 21) whose ventral plate is separated at the point between the
3rd prosomatic segment and the 4th one. D: A complete separate embryo (stage 21) whose ventral plate is
separated at the 2nd prosomatic segment. The 2nd appendages are lost. The bar shows 0.5 mm.

directly affected the central region of elongation at
the stage of obvious morphogenetic movement.
The ventral plates of the treated embryos were
separated mainly at the 2nd and 3rd prosomatic
segments.

The form and structure of the separate embryos
depended on the position of separation. It was
impossible to recognize any difference between
embryos induced by different reagents, if the posi-
tion of separation was the same. In addition the
separate embryos obtained in normal seawater had
the same characteristics as those of the separate
embryos obtained by treatment with chemical rea-
gents.

The separate embryos could be classified into
the complete and incomplete separate ones. The
ventral plate of the complete separate embryo was
separated completely, but that of the incomplete
one was not. The central nervous systems of both
types of separate embryos were separated at the
position of separation of the ventral plate,
although light microscopy did not reveal clearly
the very fine nervous fibers. On the other hand,
alimentary canals were not separated in any type
of embryo.

In some of the separate embryos, the stomodaea
passed through the middle of the brain. The
crossing of the nervous system and stomodaeum

Ganglions & Stomodaeum in Horseshoe Crab 293

Fic. 5. The brains and stomodaea in a normal embryo and in separate ones at stage 20. A: A horizontal section of a

normal embryo.

B: A horizontal section of a separate embryo whose ventral plate is separated at the 2nd

prosomatic segment. The stomodaeum passes through the middle of the brain. C: A horizontal section of a
separate embryo whose ventral plate is separated at the region between the 1st prosomatic segment and the 2nd
one. The stomodaeum and the nervous system do not cross each other. D: A longitudinal section of a separate
embryo whose ventral plate is separated at the same region as in “C”. “B” and “C” were taken at the same

magnification. The bars indicte 0.1 mm.

was released in some of the separate embryos (Fig.
5). Release of the crossing was observed in both
types of separate embryo, the complete separate
embryos and the incomplete ones. This means that
the release of the crossing was not related directly
to the degree of separation, but it was related
closely with the position of separation, that is, the
extent of the anterior region of the separate ven-
tral plate (Fig. 6). When the anterior region had
more than three pairs of appendages, the stomo-
daeum passed behind the brain and both structures
crossed each other. When it was moderate, the
stomodaeum passed the middle of the brain.
When the anterior ventral plate of the separate

embryos had no or only one pair of cephalothor-
acic appendages, the crossing was released. The
size of the posterior region of the separate ventral
plate was not related directly to the release of the
crossing of the nervous system and stomodaeum.
In embryos which had lost the anterior ventral
plates (no-anterior embryos), there was no brain
or stomodaeum. In the separate embryos whose
ventral plates were separated at the 1st prosomatic
segment, the neuroblasts of the brain were clus-
tered behind the stomodaeum and formed the
brain at this position. Crossign of the nervous
system and stomodaeum did not occur (Fig. 7).

294 T. Irow, T. MASUDA AND K. SEKIGUCHI

behind 5th behind 4th behind 3rd behind 2nd
i prosomatic prosomatic prosomatic Pprosomatic
Posterior segment segment segment segment

piece

Anterior
piece

_
Nothing

Only

embryonic
area (No

appendages )

Only 1st
appendages

Isic & Drel
appendages
3} jQEULIES Ope
appendages

Fic. 6. Release of the crossing of the stomodaeum and nervous system. The schematic diagrams show the features of
each piece of the separate ventral plates. Each mark indicates an embryo. X: Embryos having no brains or
stomodaea. @: Nervous systems and stomodaea do not cross each other. 4: Embryos whose stomodaea pass
through the middle of the brain. ©: Embryos whose stomodaea pass behind the brain (normal position). *:
Embryos without anterior pieces of separated ventral plates (=no-anterior embryos).

A B C

Fic. 7. Diagrammatic representations of development of normal embryos and separate embryos. A: A normal
embryo. B: A separate embryo whose ventral plate is separated at the 2nd and the 3rd prosomatic segment. C: A
separate embryo whose embryonic area is separated at the Ist prosomatic segment.

Ganglions & Stomodaeum in Horseshoe Crab 295

DISCUSSION

Several facts about normal embryos are re-
corded for the first time here. We consider the
crossing of the nervous system and stomodaeum.

The tubular structure of the stomodaeum was
formed at stage 14. The commissures of the
ganglions were not formed at this stage. Remark-
ably developed commissures were recognizable at
stage 19. The opening of the stomodaeum (mouth)
first appeared in front of the 1st prosomatic seg-
ment. The mouth began to migrate posteriorly at
stage 18. These results and the fact that the
stomodaeum passed through the area between the
commissure of the ganglions of the Ist and 2nd
prosomatic segment indicate that the crossing of
the nervous, system and stomodaeum occurs at
stage 18-19.

Prospective cells of the stomodaea and brains
were determined before the time of separation of
the ventral plates, because regulation did not occur
after separation. The prospetive cells of the brain
were moved by the separation (Fig. 7). When the
anterior part of the separate ventral plates was
very small, the prospective cells of the brain were
situated behind the stomodaeum. As a result the
crossing of the nervous system and stomodaeum
did not occur.

The release of crossing reveals the following
four points. (1) The crossing is constructed
through the process of clustering of brain cells. (2)
It is not determined that the cells of the brain
cluster anterior to the stomodaeum; that is, the
stomodaeum does not determine the clustering
point of the brain. Further the brain does not
determine the pathway of elongation of the stomo-
daeum. (3) As the position of the mouth is
changed easily, it is not appropriate as an indicator
of determination of homologous segments among
arthropod species [11, 12]. (4) The crossing is
susceptible to modification in at least one repre-

sentative Protostomia, the horseshoe crab.
Attempts to derive Deuterostomia from Protosto-
mia cannot be rejected because of the crossing of
the alimentary canal and central nervous system.

REFERENCES

1 Shoji, K. (1929) The horseshoe crab. In “The
Anatomy of the Animal III” Ed. by Y. Okada,
Kyoritsu, Tokyo, pp. 109-160. (in Japanese)

2 Sekiguchi, K. (1973) A normal plate of the develop-
ment of the horseshoe crab, Tachypleus tridentatus.
Sci. Rep. Tokyo Kyoiku Daigaku B., 15: 1253-162.

3 Sekiguchi, K. (Ed.) (1984) The Biology of
Horseshoe Crabs. Science House, Tokyo. (in
Japanese)

4 Kigsley, J. S. (1893) Embryology of Limulus (part

- IJ). J. Morphol., 8: 195-268.

5 Packard, A. S. (1893) Further studies on the brain
of Limulus polyphemus, with notes on its embryolo-
gy. Mem. Nat. Acad. Sci., 6: 289-331.

6 Patten, W. (1896) Variations in the development of
Limulus polyphemus. J. Morphol., 12: 17-148.

7 Scholl, G. (1977) Beitrage zur Embryonalentwick-
lung von Limulus polyphemus L. (Chelicerata,
Xiphosura). Zoomorphologie, 86: 99-154.

8 Itow, T. (1979) Experimentally induced separation
of embryonic area in the eggs of the horseshoe crab.
Bull. Edu. Shizuoka Univ., Nat. Sci., 30: 5-14.

9 Itow, T. (1982) Effect of the glutamine-analogue
“azaserine” on embryonic development of the
horseshoe crab. Develop. Growth and Differ., 24:
295-303.

10 Itow, T. and Sekiguchi, K. (1980) Morphogenic
movement and experimentally induced decrease in
number of embryonic segments in the Japanese
horseshoe crab, Tachypleus tridentatus. Biol. Bull.,
158: 324-338.

11 Savory, T. (1977) Arachnida, Academic press,
London, 2nd ed., pp 3-6.

12 Weycoldt, P. (1979) Significance of later embryonic
stages and head development in arthropod phy-
logeny. In “Arthropod Phylogeny”. Ed. by A. P.
Gupta, Van Nostrand Reinhold Co., New York, pp.
107-135.

ay :
ae
i
4
jas TU (a 1G, «¢ WON GIN Ge De
6
- wets * ie has heer phe Go) eran.
f 4 n> ot leat? herouioy Gt Jonites iin
if in Ln) fried (ite Soaly emi} lly F festa
* - ——— sie! is
my 2 Metts
oo 5 ies Peete apt
eeield an We states
eae ee D | or pat wee SS
tig oat ay tuesern meee: OAT
, j ‘3 ad AG) , : a Gin ae i hin bearer bi
1 ae wise gt! Wieohtinthan ye
MATNKIN arn Faw. #5
i iw] 74 Wh) .<Pree. ann We | } nde
' ce hystroon (EV) 0) Aue 8 Donnas) pselannntat watt 2Oajeth
i j mu vbnserews wl We lieben , mani 2 aeevig ? 1 wt ce ine
1-H Wdonink doh elles Teed Pod! “i, aM ean uy Gl need
bay! GN 4 ater 0 SP eel oA beck. anh
j fer \ Sakina 7 .
ieee “1 whet aarrtih * sed aay if) ritetirvazlt el
peers ft.
; “ . tery Ani tel, au Wy. noilanegey
Cat Pee Wily ‘ M- rs
af ydbeamp iv sy i ite) sear
f ¢ > |i i ni x9 2 how a
aa PY ae ‘vey i. j ‘ay af teres
é' wed gy 1 a te he Tell aed
4 cry uy Wetat ws Tt via - ,
8 1 < ; bb i ips Haat oye Rt
1. tes h ; om in? ss a
c ALM it 3 if
5 lt §
% y vitey rel ED
j \ y #Siy
f j : mani:
EG FO TT dif roves eA oD reat a
i it ' ( , ghd v2 reg are bare re yer
on "a Hi Svete , c 1 iv? i erang
fs Cores ari ' ¥ Lrtrot pes inkl
ne ‘Fer v7? Sonu, Sse ale i¥a1 ener aa
Vi used ei 0c Gees wiriaereere | fe
i " ath f + rise Get prironils ln vane
{ ty , i) late hey
. iT, rmusabartwiy sdt £
: A
io i Sonirypeiehs soit ;
oe 4 ir.
Pas i nner! gat Ort ONS
1 t i. “
pede *y iii tegils lt yew ie
3 og? glfive vie ee Le Sem ply es

shut y ¢ alow wo mage dy tz is
pirienery ST “Heqet nut teen
: - * Nt
Wl (Oe. (ST LD) erga
, ' >
41 tf OF NOTH aio

. ¥
ot v
he - i>
“
a ty
fy 34, Fe
on
rai «

ZOOLOGICAL SCIENCE 7: 297-302 (1990)

© 1990 Zoological Society of Japan

The Genus Phorticella Duda (Diptera: Drosophilidae)
from Burma and Southern China’

Sor Wynn2, Masanori J. Topa® and Tone Xu PEnG*

Laboratory of Genetics, Department of Biology, Tokyo Metropolitan University,
Setagaya, Tokyo 158, Japan, 3Institute of Low Temperature
Science, Hokkaido University, Sapporo 060, Japan, and
*Guangdong Institute
of Entomology, Guangzhou, China

ABSTRACT— Three new species of the genus Phorticella Duda are reported from Burma and southern
China, along with collection records of two known species from Burma.

INTRODUCTION

Okada and Carson [1] resolved the taxonomical
confusion among the genera Phorticella Duda,
Zaprionus Coquillett, and the Drosophila lineosa
subgroup of the D. immigrans species-group, all of
which are characterized by silvery or chalky white
longitudinal stripes on frons and mesoscutum.
They included six Oriental, New Guinean and/or
Australian species in the genus Phorticella, and
classified those species into two subgenera, Phor-
ticella Duda and Xenophorticella Okada et Carson.
On the other hand, Okada and Carson [1] and
Bock [2] pointed out some affinities between the
subgenus Phorticella and the subgenus Scaptodro-
sophila Duda of the genus Drosophila Fallén.
However, taxonomical revision of these subgenera
awaits comprehensive phylogenetical analyses in-
cluding the whole subgenus Scaptodrosophila,
which is a quite large subgenus including a total of
229 species.

The present paper deals with 3 new species of
the genus Phorticella from Burma and southern
China, along with collection records of 2 known
species from Burma.

Accepted April 28, 1989
Received March 31, 1989

' Drosophilidae (Diptera) in Burma, V.

2 Present address: 554 (c), Nei-bain-da road, Nan-tha-
gone quarter, Insein township, Insein post office, Yan-
gon Division (Rangoon), Union of Myanmar (Burma).

Genus Phorticella Duda

Phorticella Duda, 1923, Ann. Hist.-nat. Mus. Natn.
Hung., 20: 36.—Okada and Carson [1], 540. Type
species: Drosophila bistriata de Meijere, 1911.

Diagnosis. Anterior reclinate orbital usually
fine, posterior reclinate near to proclinate than to
inner vertical. Epandrium truncate below. Nova-
sternum with 2 or 3 pairs of long submedian spines.

Subgenus Phorticella Duda
Phorticella: Okada and Carson [1], 540.

Diagnosis. Frons laterally with broad, silvery
white, longitudinal stripes, but without median
stripe. Third antennal joint white, except for
Australian Ph. albostriata (Malloch, 1924) [3].
Tarsi of mid and hind legs without minute
cuneiform bristles. Prescutellars absent. Mid
sternopleural minute. Male genitalia closely re-
semble those of D. (Scaptodrosophila) species [2].

Other characters commonly seen in the follow-
ing 3 new species are first described below.

Head: Eye with pile. Second antennal joint
brown. Arista with ca. 4 upper and ca. 2 lower
branches in addition to terminal fork. Frons
slightly broader than long, with some frontal hairs.
Carina low, narrow. Cheek narrow.

Legs: Preapicals on all tibiae; apicals on fore
and mid tibiae.

298 S. Wynn, M. J. TopA AND T. X. PENG

Wing: R>,3 straight; R45 and M nearly para-
llel. Haltere white.

Periphallic organs: Epandrial heel prominent-
ly protruded; toe round. Cercus separated from
epandrium, nearly entirely pubescent. Surstylus
somewhat semicircular in caudal view.

Phallic organs: Posterior parameres(?) medial-
ly fused to each other.

Phorticella (Phorticella) tortia Wynn
et Toda, sp. nov.

(Fig. 1)

*, £. Body length, ¢ ca. 2.2mm, ? ca. 2.3
mm. Thorax length (including scutellum) ca. 1.1
mm in both sexes.

Head: Eye brownish red. Frons yellowish
brown; periorbit dark brown, narrow, restricted to
upper half of frons; ocellar triangle dark brown.
Anterior reclinate orbital ca. 1/3 length of post-
erior reclinate; proclinate slightly longer than post-
erior reclinate. Face and cheek yellowish brown;
carina brown. Clypeus brown. Cheek ca. 1/10 as
broad as maximum diameter of eye. Second oral
weak, ca. 1/2 length of vibrissa. Palpus yellowish
white, with ca. 5 bristles aligned laterally; terminal
bristle longest.

Thorax: Mesoscutum brownish yellow,
medially with broad, dark brown, longitudinal
stripe which is laterally bordered by slightly silvery
yellow stripe from anterior margin to level of
anterior dorsocentrals and is broadened between
dorsocentrals. Scutellum dark brown, antero-
laterally yellowish, apically white. Thoracic pleura
brown. Humerals 2. Acrostichal hairs in 6 rows.
Anterior dorsocentral ca. 3/5 length of posterior;
cross distance of dorsocentrals ca. 3 times length
distance. Anterior/posterior scutellar ca. 7/10.
Distance from posterior scutellar to anterior
almost equal to distance between posteriors. Sterno-
index ca. 0.8.

Legs brownish yellow; coxae and fore femur
darker. Fore metatarsus slightly longer than 2
succeeding tarsal joints together; mid and hind
metatarsi as long as 3 succeedings together.

Wing hyaline. Veins brownish yellow; costa
slightly clouded at 2nd break; crossveins clear.
Cl-bristles 2; ventral one weak. Wing indices: C

ca. 1.8, 4V ca. 2.6, 4C ca. 1.7, 5x ca. 2.0, Ac ca.
2.4, C3-fringe ca. 0.6.

Abdomen: First tergite entirely pale yellow;
2nd laterally with 1 pair of dark brown patches; 3rd
and 4th in @ and 3rd to Sth in with broad, dark
brown, caudal band medially and laterally pro-
truded; 5th and 6th in ¢ and 6th in ? entirely
dark brown. Sternites yellowish white.

Periphallic organs (Figs. 1A, D): Epandrium
pubescent except anterior and ventral marginal
portions, with ca. 16-20 bristles; caudoventral part
lobular. Surstylus with ca. 13 primary teeth on
concave distal margin, several small bristles on
caudoventral portion and many long spines on
inner surface. Cercus narrow, with ca. 25 bristles.
Decasternum rhomboidal, medially sparsely
pubescent (Fig. 1D).

Phallic organs (Figs. 1B, C): Aedeagus lateral-
ly hirsute, shaped like torch in lateral view (thus
the species name), dorsally bilobed and with 1 pair
of small, triangular, marginally serrate flaps;
apodeme broad, slighty longer than aedeagus.
Anterior paramere long, curved ventrad, apically
slightly expanded and round, dorsomedially
sparsely hirsute, with ca. 9 sensilla aligned along
nearly entire length on outer surface. Novaster-
num somewhat quadrate, broader than long, con-
cave on anterior margin, with 3 pairs of submedian
spines on caudal margin; base of submedian spines
expanded, forming small lobe.

2 reproductive organs: Ovipositor (Fig. 1E)
apically blunt, with ca. 1 bristle-like discal, ca. 5
somewhat long apical and ca. 16 marginal teeth,
and 1 long subterminal and 3 small terminal hairs.
Spermatheca (Fig. 1F) broader than long, some-
what quadrangular in lateral view; duct slightly
constricted medially in introvert.

Holotype @, Burma: Pyin Oo Lwin,
30. XII.1981-6.1.1982, ex trap (Toda); deposited
in Entomological Institute, Hokkaido University,
Sapporo, Japan (EHU). Paratypes, Burma: 7%,
17., same data as holotype; in EHU and the
collection of senior author (S.W.).

Distribution. Burma (Pyin Oo Lwin).

Relationship. This species is somewhat similar
to Ph. singularis (Duda, 1924) in having 6 rows of
acrostichal hairs, but clearly distinguishable from
the latter by color patterns on thorax and legs [1,

The Genus Phorticella 299

po

\
IN

Fic. 1.

Phorticella (Phorticella) tortia Wynn et Toda, sp. nov.

A: Periphallic organs, B: phallic organs (ventral

view), C: ditto (lateral view), D: decasternum, E: ovipositor, F: spermatheca. (Scale-line=0.1 mm.)

Fig. 1E] and having 3 pairs of submedian spines on
novasternum (2 pairs in Ph. singularis).

Phorticella (Phorticella) htunmaungi sp. nov.
(Fig. 2)

This species is very close to the foregoing spe-
cies, Ph. tortia. The following description is made
referring only to the differences from the latter.

&, £. Boby length, $ ca. 2.2-2.3mm, ca.
2.3-2.8mm. Thorax length, $ ca. 1.0mm, $ ca.
1.1mm.

Head: Anterior reclinate orbital ca. 1/4-1/3
length of posterior reclinate; proclinate as long as
posterior reclinate. Carina pale yellow. Cheek
width ca. 1/10 maximum diameter of eye. Second
oral ca. 1/3—1/2 length of vibrissa.

Thorax: Mesoscutum dark brown, patterned

as follows: 1 pair of broad, slightly silvery shining
pale brown stripes appearing to continue from
frontal white stripes and extending posteriorly to
level of anterior dorsocentrals; narrow yellowish
stripes present medially and along dorsocentral
lines; yellowish patches present laterally.
Scutellum dark brown, laterally black. Thoracic
pleura dark brown. Anterior dorsocentral ca. 7/10
length of posterior; cross distance of dorsocentrals
ca. 2.5-3.3 times length distance. Anterior/post-
erior scutellar ca. 7/10—4/5; posteriors slightly
more distant from each other than from anterior.
Sterno-index ca. 0.7.

Wing indices: C ca. 1.7-2.0, 4V ca. 2.4-2.9,
4C ca. 1.4-1.8, 5x ca. 1.9-2.3, Ac ca. 2.0-2.3,
C3-fringe ca. 0.5-0.6.

Abdomen: Second to 4th tergites pale yellow,
sublaterally with 1 pair of large or small, dark

300 S. Wynn, M. J. ToDA AND T. X. PENG

brown patches; 5th in ? with 3 dark brown
patches.

Periphallic organs (Figs. 2A, D): Epandrium
with ca. 18-20 bristles. Cercus with ca. 20 bristles.
Decasternum medially densely haired (Fig. 2D).

Phallic organs (Figs. 2B, C): Aedeagus dorsal-
ly with small, marginally serrate flap somewhat
variable in shape; apodeme as long as aedeagus.
Anterior paramere heavily hirsute medially to
subapically on dorsal margin, with ca. 12 sensilla.
Novasternum medially slightly notched and sub-
laterally with 3 pairs of submedian spines on
caudal margin; base of submedian spines not so
expanded.

9 reproductive organs: Ovipositor (Fig. 2E)
with ca. 5—6 apical and ca. 15-18 marginal teeth.
Spermatheca (Fig. 2F) hemispherical; introvert
slightly annulate.

Holotype ¢, Burma: Pyin Oo _ Lwin,
30. XII.1981-6.1.1982, ex trap (Toda); in EHU.
Paratypes, Burma: 9%, 42, same data as holoty-

pe; in EHU and S.W. China: 1%, 12, Conghua,
Guangdong Province, 27.1.1987, by sweeping on
tree trunks and forest floor (Toda); Dinghushan,
Guangdong Province, 17, 5-13.VII.1986, 12,
21-27.VII.1986, 12, 20-25.11.1987, 22, 14-
23.V.1987, ex traps (Peng); in the Guandong
Institute of Entomology, Guangzhou, China
(GIE) and EHU.

Distribution. Burma (Pyin Oo Lwin), China
(Guangdong).

Relationship. As mentioned above, this species
is closely related to the foregoing species, Ph.
tortia, but can be distinguished from the latter by
color patterns on thorax and abdomen, denser
hairs on anterior paramere and decasternum, and
unexpanded base of submedian spines.

Remarks. This species is named in honor of Dr.
Htun Maung, the Emeritus Professor of Zoologic-
al Department and the Rector of Mandalay Uni-
versity.

Fic. 2.

Phorticella (Phorticella) htunmaungi sp. nov. A: Periphallic organs, B: phallic organs (ventral view), C: ditto

(lateral view), D: decasternum, E: ovipositor, F: spermatheca. (Scale-line=0.1 mm.)

The Genus Phorticella 301

Phorticella (Phorticella) nullistriata sp. nov.
(Fig. 3)

é, %. Body length, ¢ ca. 2.25mm, ? ca. 2.5
mm.

Head: Eye dark red. Frons brown. Carina
brown. Clypeus black. Cheek ca. 1/20 as broad as
maximum diameter of eye. Second oral ca. 1/3
length of vibrissa. Palpus grayish white, with ca. 3
bristles aligned laterally; terminal bristle longest.

Thorax black, shiny. Mesoscutum without whit-
ish longitudinal stripes (thus the species name).
Scutellum laterally dark, apically milky white.
Humerals 2, unequal. Acrostichal hairs in 8 rows.
Cross distance of dorsocentrals ca. 2.4—-3.7 times
length distance. Anterior/posterior scutellar ca.
9/10; posteriors slightly more distant from each
other than from anterior. Sterno-index ca. 0.8.

Legs: All femora blackish brown, tibiae and
tarsi whitish yellow.

Wing somewhat fuscous. Cl-bristle 1. Wing
indices: C ca. 1.6—2.0, 4V ca. 2.7-3.0, 4C ca. 1.5-
1.7, 5x ca. 2.5-3.5, Ac ca. 2.1-2.5, C3-fringe ca.
OS,

Abdomen:

First tergite yellow; 2nd and 3rd in

¢ and 2nd to Sth in ? medially yellow, laterally
black; 4th to 6th in @ entirely black and subshin-
ing. First to 3rd sternites yellowish white, 4th to
6th gray or black in @; 1st to 6th yellowish white,
7th black in 2.

Periphallic organs (Figs. 3A, D): Epandrium
pubescent except anteroventral portion, with ca. 3
bristles on upper part and ca. 14 on lower part;
anterior and posterior margins nearly parallel.
Surstylus with ca. 12 primary teeth (upper ca. 5
blunt and slightly longer than rest), ca. 3 minute
bristles at ventral corner and ca. 4 minute ones on
inner surface. Cercus somewhat oblong, with ca.
13 bristles. Decasternum consisting of 2 parts;
ventral plate rectangular; dorsal part shirt-like in
ventral view (Fig. 3D).

Phallic organs (Figs. 3B, C): Aedeagus hairy,
pointed in both ventral and lateral views. Anterior
paramere finger like, with a few sensilla aligned in
oblique row basally on outer surface. Novaster-
num hexagonal, with 2 pairs of submedian spines;
inner pair much longer than outer pair.

9 reproductive organs: Ovipositor (Fig. 3E)
grayish, apically round, with ca. 1 discal, ca. 6
apical (ultimate one especially long), ca. 10 mar-

Fic. 3. Phorticella (Phorticella) nullistriata sp. nov. A: Periphallic organs, B: phallic organs (ventral view), C: ditto
(lateral view), D: decasternum, E: ovipositor, F: spermatheca. (Scale-line=0.1 mm.)

302 S. Wynn, M. J. ToDA AND T. X. PENG

ginal teeth, and 1 long subterminal and ca. 2 small
terminal hairs. Spermatheca (Fig. 3F) black,
umbrella-shaped, apically slightly indented; intro-
vert deep, annulate.

Holotype @, China: Guanzhou, Guangdong
Province, 23.1X.1986 (Peng); in GIE. Paratypes,
China: 3 , 72, same data as holotype; 1 ? , same
data as holotype except 21-29. X1.1985; 1? , Ding-
hushan, Guangdong Province, 24.XI-1.XII.1986
(Peng); in GIE and EHU.

Distribution. China (Guangdong).

Relationship. This species certainly belongs to
the subgenus Phorticella, because of having white
3rd antennal joint and lateral silvery white longitu-
dinal stripes on frons, but is unique in having no
whitish longitudinal stripes on mesoscutum.

Phorticella (Phorticella) bistriata (de Meijere)

Drosophila bistriata de Meijere, 1911, Tijdschr. Ent., 54:
397 (Java).

Phorticella bistriata: Duda, 1924, Arch. Naturg., 90(A):
182 (Java).

Phorticella (Phorticella) bistriata: Okada and Carson [1],
540 (Sumatra, Burma).

Zaprionus albicornis Enderlein, 1922, Deutsch. ent.
Zeitschr., 1922: 295 (syn. by Duda, 1926, Suppl. Ent.,
14: 45) (Taiwan).

Drosophila albicornis: Lin and Tseng, 1973, Bull. Inst.
Zool. Acad. Sinica, 12: 22 (Taiwan).

Phorticella fenestrata Duda, 1923, Ann. Hist.-nat. Mus.
Natn. Hung., 20: 36 (as var. of bistriata) (Taiwan).

Specimens examined. Burma: 12, Pyin Oo
Lwin, 30.XII.1981-6.1.1982, ex trap (Toda); 5%,
22, Mandalay, 26.XII.1981—4.1.1982, ex traps
(Toda); 18, 1%, Mandalay, 31.XII.1981, by
sweeping on tree trunks (Toda); 2¢, Shwebo, 2,
3.1.1982, by sweeping on tree trunks (Toda); 134,
82, Rangoon, 18, 22.XII.1981, 9, 10, 13,
14.1.1982, by sweeping at ditches and on tree
trunks (Toda).

Distribution. China (Taiwan, Guangdong),
Java, Sumatra, Burma (Pyin Oo Lwin, Mandalay,
Shwebo, Rangoon).

Subgenus Xenophorticella Okada et Carson

Xenophorticella Okada et Carson [1], 542. Type spe-
cies: Zaprions flavipennis Duda, 1929.

Diagnosis. Frons with median, longitudinal,
whitish stripe in addition to lateral ones. Third
antennal joint gray. Tarsi of mid and hind legs
with minute cuneiform bristles.

Phorticella (Xenophorticella) flavipennis (Duda)

Zaprionus flavipennis Duda, 1929, Treubia, 7: 416 (Buru
Is.).

Phorticella flavipennis: Wheeler, 1981, Genetics and
Biology of Drosophila, 3a: 73.

Phorticella (Xenophorticella) flavipennis: Okada and
Carson [1], 543 (Ryukyu Is., India, Singapore, New
Guinea).

Drosophila (Hirtodrosophila) bicolovittata Singh, 1974,
Zool. J. Linn. Soc., 54: 162 (India).

Phorticella striata Sajjan et Krishnamurthy, 1975, Orient.
Ins., 9: 118 (India).

Phorticella carinata Takada, in Takada and Maki-
no, 1981, J. Fac. General Educ. Sapporo Univ.,
(19): 31 (Ryukyu Is.).

Specimens examined. Burma: 1%, 1, Pyin Oo
Lwin, 30.XII.1981-6.1.1982, ex trap (Toda); 8,
12%, Mandalay, 26.XII.1981-6.1.1982, ex traps
(Toda).

Distribution. Ryukyu Is., China (Taiwan,
Guangdong), Singapore, Burma (Pyin Oo Lwin,
Mandalay), India, New Guinea, Moluccas (Buru
Is.).

ACKNOWLEDGMENTS

This work was supported by a Grant-in-Aid for Over-
seas Scientific Survey from the Ministry of Education,
Science and Culture, Japan (Nos. 56041049, 57043044,
60041061, 61043056, 62041085).

REFERENCES

1 Okada, T. and Carson, H. L. (1983) The genera

Phorticella Duda and Zaprionus Coquillett (Diptera,

Drosophilidae) of the Oriental Region and New

Guinea. Kontyu, Tokyo, 51: 539-553.

Bock, I. R. (1982) Drosophilidae of Australia. V.

Remaining genera and synopsis (Insecta: Diptera).

Aust. J. Zool., Suppl. 89: 1-164.

3 Bock, I. R. (1976) Drosophilidae of Australia, I.
Drosophila (Insecta: Diptera). Aust. J. Zool., Suppl.
40: 1-105.

in)

ZOOLOGICAL SCIENCE 7: 303-309 (1990)

© 1990 Zoological Society of Japan

Early Larval and Postlarval Morphology of the Soldier Crab,
Mictyris brevidactylus Stimpson (Crustacea:
Brachyura: Mictyridae)

YASUSHI FUKUDA

Biological Lobaratory, Faculty of Education, Kumamoto University,
Kumamoto 860, Japan

ABSTRACT—The first zoea and megalopa of Mictyris brevidactylus are described and illustrated.
Their relationships with those of M. longicarpus are discussed. The megalopa is unique in having a
whorl-like arrangement of setae on the top of the antenna which possibly represents one of the
diagnostic characters of either Mictyris or Mictyridae.

INTRODUCTION

The soldier crabs, familiar to us especially be-
cause of their habits of aggregation on sandy mud
flats during low tide, belong to the genus Mictyris
which is now known to accommodate four species
[1, 2]. A Japanese species, Mictyris brevidactylus,
was first described by Stimpson [3] based: upon
materials collected from the Ryukyu Islands and
Hong Kong, but it has long been considered
identical with M. longicarpus from Australia [4].
Yamaguchi [5] suggested that there may be ecolo-
gical, and therefore taxonomic differences be-
tween Australian and Japanese populations. Most
recently, in his revisionary work of the genus
Mictyris, Takeda [2] concluded that the Japanese
and the Australian forms were specifically diffe-
rent.

The purpose of this study is to provide detailed
descriptions of larval and postlarval morphology of
M. brevidactylus, and to discuss the taxonomic
status of the Japanese soldier crab from the view-
points of larval and postlarval morphology.

MATERIALS AND METHODS

Ovigerous females of M. brevidactylus, collected
on Amamin-oshima of the Ryukyu Islands, Febru-

Accepted june 7, 1989
Received January 8, 1989

ary 22, 1978, were transported to Kumamoto and
reared under laboratory conditions. First zoeas
hatched on March 11; but all of them died during
the preparation of food supply; the first zoeas were
fixed with 50% ethylene glycol for examination.
The megalopas here used were collected from the
same locality by Dr. T. Yamaguchi on March 1,
1973, and by myself on February 22, 1978. The
collected samples of the megalopas are referable to
M. brevidactylus because of their habitat, and
morphology, particularly of the whorl-like
arrangement of setae on the top of the antenna as
displayed by the Australian M. longicarpus [7].
All the specimens were preserved in 75% (ethyl)
alcohol. The setation of each appendage is pre-
sented from proximal to distal.

Description of first zoea

Size. —Carapace length (distance between tip of
rostral spine and posterior margin of carapace)
0.71-0.73 mm (average 0.72 mm): 10 specimens
examined.

Carapace (Fig. 1A, B).—Smooth, inflated and
globose; lateral and dorsal spines absent; rostral
spine overreaching antenna, bearing row of fine
setae distolaterally; eyes immovable.

Abdomen (Fig. 1A).—Five somites and telson;
somite 1 completely concealed beneath carapace;
somite 2 with anteriorly directed lateral process;
remaining somites each with small posterolateral

304 Y. FUKUDA

Fic. 1. First zoea of Mictyris brevidactylus Stimpson. A, lateral view; B, frontal view; C, anternnule; D, antenna; E,
mandible: F, maxillue; G, maxilla; H, first amxilliped; I, second maxilliped; J, telson. Scales, 0.1 mm.

Larvae of Japanese Soldier Crab 305

spine.

Telson (Fig. 1A, J).—Elongate, medially con-
stricted; furcae slightly curving dorsal, bearing
numerous tiny spines on mesial margin and fine
setae on lateral margin; posterior margin bearing 6
simple elongate setae.

Antennle (Fig. 1C).—Rod-like, somewhat inflated
basally; 2 long aesthetascs and 1 short seta.
Antenna (Fig. 1D).—Protopodal process elon-
gate, bearing a row of spinules on mesial and
lateral margin; exopod falling short of end of
protopodal process, with 1 spinule at 1/3 of length
from proximal end; plus numerous setae mesially
and laterally.

Mandible (Fig. 1E).—Molar process short and
subcylindrical, with denticles of irregular size; in-
cisor process bluntly bidentate.

Maxillule (Fig. 1F).—Endopod 2-segmented, pro-
ximal segment short, 1/4 as long as distal segment,
with 1 distolateral seta; distal segment with 1
midlateral, 2 subterminal, 2 terminal setae; coxal
and basal endites bearing 5 setose spines each;
other pubescence as illustrated.

Maxilla (Fig. 1G).—Endopod distally bifurcate
with 2 terminal, 2 subterminal setae; distal and
proximal lobes of basal and coxal endites, bearing
4, 5 and 2, 5 setae respectively; scaphognathite
with 4 soft plumose setae plus 1 stout, apical
plumose projection; fine hairs on margins of en-
dopod, coxal endite and scaphognathite, as illus-
trated.

First maxilliped (Fig. 1H).—Endopod _ 5-seg-
mented, with setation of 2, 2, 1, 2, 4+1; exopod
2-segmented, with 4 natatory setae; basis elongate,
subcylindrical, bearing 4 groups of setae (2, 2, 3,
3), progressing distally 10 in all as illustrated; coxa
with 1 seta.

Second maxilliped (Fig. 11).—Three-segmented
endopod relatively short, about half as long as
exopod, with setation of 0, 1, 2+3-+1; exopod
2-segmented, with 4 natatory setae; basis elongate
and subcylindrical with 4 equidistant setae; coxa
naked.

Chromatophores (Fig. 1A, B).—Located at base
of rostrum, between eyes, carapacial center, car-
diac and postcardiac regions, labrum, mandible,
basis of first maxilliped, second through fifth abdo-
minal somites and telson. Color not noted.

Description of megalopa

Size.—Carapace length 1.72-1.81mm (average
1.77 mm); carapace width 1.41-1.48 mm (average
1.45 mm): 5 specimens examined.

Carapace (Fig. 2A-B).—Posteriorly widened,
dorsally convex; 2 tubercular processes on gastric
region, each fringed with setae posteriorly; median
rostral process very short and strongly deflexed,
lateral processes small, directed upward, each with
one terminal seta; postero-lateral margin of cara-
pace setiferous; orbit rather shallow; buccal cavern
large, nearly trapezoidal.

Thorax and pereopods (Fig. 2A, C, E, F).—Sixth
and seventh sternal segments with tubercular pro-
cess near lateral margin. Pereopods sparsely
setose; chelipeds relatively slender, stouter but
distinctly shorter than the three walking legs;
walking legs somewhat compressed laterally, first
leg longest, fourth leg reduced in size, with 5 long
branchyuran feelers.

Abdomen (Fig. 2B, D).—Six somites and telson;
dorsal surface sparsely setose as illustarted; so-
mites 2—4 with 2 small posterolateral spines on
each pleuron; pleuron of somite 5 with single
large, acute, posteriorly directed spine; somite 6
unarmed, about half as long as telson.

Pleopods (Fig. 2B, D).—Decreasing in size on
somites 2-6; expods of somites 2, 3, 4 and 5
bearing 18, 18, 18, and 15 natatory setae, respec-
tively; endopods each with 3 small terminal hooks;
uropod with 5 long plumose setae.

Antennule (Fig. 3A).—Four-segmented; proximal
segment markedly inflated, with 2 lateral spine-
like setae; second segment with 1 short lateral seta;
third segment with 2 setae at articulation; 6 aesthe-
tascs, 1 subterminal seta on fourth segment.
Antenna (Fig. 3B).—Five-segmented; _penulti-
mate segment bearing whorl of 16-18 long setae as
illustrted.

Mandible (Fig. 3C).—Incisor process with dentate
cutting edge; molar process smooth, not toothed;
mandibular palp 3-segmented, ultimate segment
with 9 short setose spines on distal half.
Maxillule (Fig.3D).—Endopod unsegmented
with 1 terminal and 2 lateral setae; basal and coxal
endites with numerous spines and setae; a long
plumose seta at base of endopod.

306 Y. FUKUDA

Fic. 2. Megalopa of Mictyris brevidactylus Stimpson. A, dorsal view; B, lateral view; C, thoracic sternum; D,
posterior half of abdomen and tail fan; E, right chela, interior view; F, right second walking leg, interior view.

Scales, 0.5 mm.

Maxilla (Fig. 3E).—Endopod unsegmented, bear-
ing 3 proximal marginal setae; scaphognathite well
developed, fringed with 58-62 plumose setae; bas-
al and coxal endites bilobed, distal and proximal
lobes bearing 12, 9 and 9, 23-24 spine-like setae
respectively.

First maxilliped (Fig. 3F).—Exopos and endopod
apparently unsegmented: exopod with 2 long plu-
mose setae at 1/3 of length from distal end, termin-
al margin with 1 setal nub; endopod without setae
but with 2 setal nubs at distal end; epipod roughly
triangular, bearing a total of 5 plumodenticulate

Larvae of Japanese Soldier Crab 307

Fic. 3.

Megalopa of Mictyris brevidactylus Stimpson. A, antennule; B, antenna; C, mandible; D, maxiliule; E,

maxilla; F, first maxilliped; G, second maxilliped; H, third maxilliped. Scales, 0.2 mm.

setae at corners, as illustrated.

Second maxilliped (Fig. 3G).—Exopod unseg-
mented with 1 setal nub at distal end; endopod
4-segmented, segment 1 (proximal) very broad and
foliaceous naked; segments 2, 3, 4 bearing 1, 8,
15-17 spine-like setae respectively; epipod absent.

Third maxilliped (Fig. 3H).—Exopod reduced in
size, terminal margin with 1 setal nub; endopod
5-segmented, with setae as shown; narrowed dis-
tally, ischial segment elongate, fully twice as long
as meral segment; epipod well developed with
several long setae on distal 1/3 of length.

308 Y. FUKUDA

DISCUSSION

Based upon adult morphology, Takeda [2] re-
vised the genus Mictyris and concluded that the
Japanese and Australian soldier crabs, both pre-
viously merged with Mictyris longicarpus, are spe-
cifically distinct and that the Japanese species
should be called M. brevidactylus Stimpson. These
two species, however, are not remote in distribu-
tion. The southern limits of their ranges are rather
close, the boundary being placed roughly between
the vicinity of the Sulu Sea and the south of the
Philippines [1]. Notwithstanding their close range,
ecological differences, especially in tunnel feeding
habits and wandering activity patterns, have been
noted [5]. In addition to these data, the differ-
ences between M. longicarpus and M. brevidacty-
lus are also distinct from the viewpoint of larval
and postlarval morphology.

Cameron [6] briefly described the first zoea of
M. longicarpus obtained from females collected
from Moreton Bay, Southern QueenslInad. Fielder
et al. [7] provided definitive descriptions of the 5
zoeal stages of that species obtained by rearing,
and of megalopas collected on the same shore of
Moreton Bay, and thus modified Cameron’s defini-
tion of the larvae of the species in several respects.

The present data are compared with those of
Fielder et al. Listed below are the morphological
differences in the first zoeas and megalopas be-
tween M. longicarpus and M. brevidactylus.

M. longi- M. brevidac-
carpus tylus

FIRST ZOEA

Exopod of 1 short seta without seta

antenna at midlength

Coxal endite 7 spines 6 spines

of maxilla

Basis of first 9 spines 10 spines

maxilliped

MEGALOPA

Molar process absent present

of mandible

Distal segment 8 spines 9 spines

of mandibular

palp

Endopod of without seta 3 proximal
maxilla marginal setae
Exopod 1 plumoden- _— without seta
of third ticulate seta

maxilliped

Hair formula 0-6 0-5

of uropod

These differences are sufficient to warrant the
Australian and Japanese populations to be re-
garded as distinct species. Thus, zoeal and mega-
lopal morphology fully complements the work by
Takeda [2] on adults.

The oval, dorsally rounded carapace, the strong-
ly deflexed rostrum, and the very wide buccal
cavern containing largely foliaceous endopods of
the second and third maxillipeds and slenderly
reduced exopods of three pairs of maxillipeds, are
very characteristic of megalopa in the genus Mic-
tyris. The last two of these seem to be related to
the feeding habits, i.e. the making of sand pellets
during feeding. A similar situation is seen in the
megalopa of the ocypodid crab Scopimera globosa
[8].

The exopod of the third maxilliped in M. brevi-
dactylus iacks a flagellum, whereas that in M.
longicarpus has a distinct plumodenticulate seta
which, however, is possibly lost in the next juve-
nile stage, for the absence of the flagellum repre-
sents one of the familial characteristics of the
Micytridae [4, 8].

The antennal whorl of setae, which is shared by
the megalopas of both the Japanese and the Aus-
tralian species of Mictyris, has not been recorded
in postlarvae of other brachyuran crabs. Very
possibly it represents either a generic or a familial
characteristic, though no information is available
on the two other known species of this genus.

ACKNOWLEDGMENTS

I thank Dr. K. Baba of Kumamoto University and Dr.
R. H. Gore, Bio-Econ, Incorporated, Naples, Florida,
for reviewing a draft of the manuscript. Thanks are
also due to Dr. M. Takeda of the National Science
Museum, Tokyo, for his advice regarding systema-

Larvae of Japanese

tic treatment of Mictyris. I am obliged to Dr. T.
Yamaguchi of the Aitsu Marine Biological Station,
Kumamoto University, for making specimens of M.
breavidactylus available for study. I acknowledge Dr. Y.
Miya of Nagasaki University and Dr. Y. Nakasone of the
University of the Ryukyus, for preparing copies of
references at my request. I am also grateful to the
anonymous reviewers for making helpful comments on
the manuscript.

This study was supported in part by a grant-in-aid from
the Ministry of Education, Science and Culture of Japan,
No. 57340036.

Contribution No. 52 from the Aitsu Marine Biological
Station, Kumamoto University.

REFERENCES

1 McNeill, F. A. (1926) Studies in Australian carcinol-
ogy. No. 2. Rec. Aust. Mus., 15: 100-131, pls. 9-10.

2 Takeda, M. (1978) Soldier crabs from Australia and
Japan. Bull. Natn. Sci. Mus., Ser. A (Zool.), 4: 31-
38.

3 Stimpson, W. (1858) Prodromus descriptionis anima-
lium evertebratorum, quae in Expeditione ad

Soldier Crab 309

Oceanum Pacificum Septentrionalem, a Republica
Federata missa, Cadwaladaro Ringgold et Johanne
Rodgers Ducibus, observavit et descripsit. Pars V.
Crustacea Ocypodidea. Proc. Acad. Natl. Sci. Phila.,
10: 93-110.

Sakai, T. (1976) Crabs of Japan and the adjacent
seas. Kodansha, Tokyo. (In 3 volumes: (1) English
text, xxix+773 pp. (2) Plates volume, 16 pp., 251 pls.
(3) Japanese text, 461 pp.)

Yamaguchi, T. (1976) A preliminary report on the
ecology of the sand bubbler crab, Mictyris longicar-
pus Latreille. Benthos Res. Jap., 11/12: 1-13. (In
Japanese with English abstract)

Cameron, A. M. (1965) The first zoea of the soldier
crab Mictyris longicarpus (Grapsoidea: Mictyridae).
Proc. Linn. Soc. N. S. W., 90: 222 —224.

Fielder, D. R., Greenwood, J. G. and Quinn, R. H.
(1984) Zoeal stages, reared in the laboratory, and
megalopa of the soldier crab Mictyris longicarpus
Latreille, 1806 (Decapoda, Mictyridae). Bull. Mar.
Sci., 35: 20-31.

Ono, Y. (1965) On the ecological distribution of
ocypodid crabs in the estuary. Mem. Fac. Sci.,
Kyushu Univ., Ser. E (Biol.), 4: 1-60.

ae a
da) ee ers saat = oe ea .
nae rousitiat gpomodie ; ba sh nntinos |

AED yivtin uterstre rorya sacha a

tania qwrere > Taree — Mir spefos -yaniaeny aa RE lk
(1). Senko Teronmintradgelit sage ee aah GUNG els Pm zs

Ae nt a eee Fass Via “ \c wey! iAast bisedk t
| Mabiliepye i) iswer? | ¥ Seaiervorica | ‘ ' is
thd a) nat Wy > pat ete copes 0 —¥ v1 bos ¢ a ee: og

am

TORTAT ARG Te eee ee iettig hal rr
Gy Lic haart A ; a er
OT soageuet — F pont bps Me lan tg « vl Piety fl
Na ae Se sr
iSP> aheepeey ed? Baal i a . oe
oe Lito AE Ate Se 7 i : iy ae i weet walt A.
i. 3D We ited a i i
}) Cat eames aegis, -A20) fia ee. bet wie?
. _ a ae one { é gorphs tice . ue
! A = bane a eaitui, ta 4
may On 4 at lpm rei fae aah Feu,

ul Tas mite, MVE eda whit aes
2 Bev Ra Be re pers iec

ee aoa ; ; nove. rye ly AME
MEP ere thin cahaotaaead neers ben Te '

insr gl MRL, A PEND BG 768
' herder atte oh one te at
rir The jan nae ot ea
: j e ‘the a sig
ocd 2 . uv +15 Lee ryan
me de pioall ete eae
wher
: ; Psion olucwg

Ae, lal
i hed
‘ bY

oMeNie es

: ire 1 ae
} m - ie WANS OU, rae
" retr<h at avin igi
hea cite Sane

}

iuisb ew.” 15

ZOOLOGICAL SCIENCE 7: 311-318 (1990)

© 1990 Zoological Society of Japan

Some Nesothrips (Insecta, Thysanoptera,
Phlaeothripidae) from East Asia

SHOJI OKAJIMA

Laboratory of Entomology, Tokyo University of Agriculture,
Setagaya, Tokyo 156, Japan

ABSTRACT—Four East Asian species of the genus Nesothrips Kirkaldy are reported. N. brevicollis
sensu Mound (1974) is divided into two species, N. brevicollis (Bagnall) and N. minor (Bagnall). N.
malaccae Mound is additionally recorded from West Malaysia, Indonesia and the Philippines. An
additional new species, N. yasumatsui Okajima, is described from Thailand. These thrips are usually
found on grass tussocks, dead branches or leaves, and may be fungus-feeders.

The genus Nesothrips Kirkaldy belongs to the
tribe Pygothripini of the subfamily Idolothripinae,
and was revised by Mound in 1974 [1]. In 1983, the
related genus Rhaebothrips Karny was synony-
mized with this genus by Mound and Palmer [2].
According to these publications, 22 Nesothrips
species have been known from the world, and
three of them have been known from eastern Asia
(exclusive of Oceanic islands). N. brevicollis (Bag-
nall) and N. /ativentris (Karny) are widespread in
the tropical and subtropical regions of eastern Asia,
and another species, N. malaccae Moun4, is distrib-
uted to West Malaysia and Sumatra.

Two widespread species, brevicollis sensu
Mound, 1974, and Jativentris, show intraspecific
variation between local populations in colour and
structure. It is therefore very difficult to know if
two or more species are involved under the respec-
tive names. Recently, I had an opportunity to
examine a long series of brevicollis sensu Mound,
1974, from many localities. After a careful ex-
amination, it became clear that two distinct species
were involved in the lot. The other species,
lativentris, is excluded from the present paper,
since the material at hand is still insufficient to
clarify its status.

In this paper, four Nesothrips species from east-
ern Asia are dealt with; they are: N. brevicollis, N.

Accepted June 9, 1989
Received May 10, 1989

malaccae, N. minor (Bagnall) which is revived
from a synonym of brevicollis, and an additional
new species from Thailand.

The holotype and most of the paratypes of the
new species described hereinafter will be pre-
served in the Laboratory of Entomology, Tokyo
University of Agriculture.

The following abbreviations are used for the five
pairs of prothoracic setae: aa, antero-angulars; am
antero-marginals, ml, mid-laterals; pa, postero-
angulars; epim, epimerals.

Nesothrips brevicollis (Bagnall, 1914)
(Figs. 1-6, 11-13, 16-17, 19 and 20)

Oedemothrips (?) brevicollis Bagnall, 1914 [3], 29-
30.

Neosmerinthothrips formosensis Priesner, 1935 [4],
368-372. [Synonymized by Mound, 1974 [1], 162].
Nesothrips brevicollis: Mound, 1974 [1], 162-163
(in part).

[Nesothrips fulviceps: Kudo, 1974 [5], 115; mis-
identification].

Bagnall [3] described this species only from the
apterous form. Recently, Mound [1] treated
Nesothrips minor (Bagnall), which was described
only from the macropterous form, as a synonym of
this species. He considered that the two are
different morphs of the same species. More re-
cently, however, I studied more than 200 recently
collected specimens of brevicollis sensu Mound,

312 S. OKAJIMA

1974, from many localities in East Asia and the
Hawaiian Islands, and after carefully examining
them, came to the conclusion that they included
two distinct species. In the present paper, I regard
one of them as brevicollis (Bagnall) and the other
as minor (Bagnall).

There are some differences between them,
which are probably not related to the presence or
absence of wings. Relative length of the tube is
shorter than the head with rounded sides in brevi-
collis, but longer than the head with rather straight
sides in minor. The pelta of brevicollis is some-
times eroded in median portion of the posterior
margin and with somewhat wide median lobe
(Figs. 1-6), though that of minor is not eroded and

with narrower median lobe (Figs. 7-10), in both
macropterae and micropterae. Moreover, the
wing retaining setae of brevicollis are short and
straight (Figs. 11-13), while those of minor are
long and curved (Figs. 14-15) also in both macrop-
terae and micropterae. In macropterae, only two
subbasal wing setae are developed in brevicollis
(Fig. 16), but three of them are developed in
minor (Fig. 18). Finally, these two forms show
sympatric distribution in the Ogasawara (Bonin)
Islands, northwestern end of Micronesia. Judging
from these characters and distribution, I consider
brevicollis and minor to be specifically different.
Thus redefined, N. brevicollis is distributed to
Japan, including the Ryukyu and the Ogasawara

LEE

Fics. 1-10.

Female pelta of Nesothrips species—1-6, N.

brevicollis, macroptera from Honshu, Japan (1),

macroptera from Ishigaki Is., Ryukyus (2), macroptera from Taiwan (3), microptera from Honshu, Japan (4),
aptera from the Ogasawara Isls. (5) and microptera from Okinawa Is., the Ryukyus (6); 7-10, N. minor,
macroptera from the Ogasawara Isls. (7), macroptera from Hawaii (8), macroptera from Thailand (9) and

microptera from Java (10).

Nesothrips from East Asia 313

Fics. 11-15.

Left half of female abdominal tergite VI of Nesothrips species—11-13, N. brevicollis, aptera from the

Ogasawara Isls. (11), macroptera from Ishigaki Is., the Ryukyus (12) and microptera from Yonaguni Is., the
Ryukyus (13); 14-15, N. minor, macroptera from the Ogasawara Isls. (14) and microptera from Java (15). Figs.
16-18. Subbasal wing setae of Nesothrips species—16-17, N. brevicollis from Honshu, Japan (16) and from

Taiwan (17); 18, N. minor from the Ogasawara Isls.

Islands, and Taiwan. It is common at certain
localities, and is found mainly on grass tussocks.
Micropterous (rarely apterous) females and males
are usually found but macropterous females are
collected only occasionally.

There are some differences between local
populations in colour and structure. Specimens
from the Ryukyus and Taiwan have somewhat
smaller body and paler head, in contrast with
specimens from Honshu, mainland of Japan,
which have larger body and darker head. Moreo-
ver, macropterous females from Taiwan have re-
duced subbasal wing setae (Fig. 17), and two
apterous females from the Ogasawara Islands have
somewhat small compound eyes. However, there
are some intermediate specimens from _ the
Ryukyus, and there is no evidence that more than
one species is involved.

The macropterous female (Figs. 19 and 20) and
micropterous male are briefly described below.

Female (macroptera). General structure almost
as in micropterous female: Forewings each with 6—
11 duplicated cilia; subbasal wing setae B, re-
duced, B> and B; distinct, B; the longest, but they

are reduced in material from Taiwan; wing retain-
ing setae short and rather straight.

Male (microptera). Some of the structures show
extreme allometry: Prominent setae of head and
prothorax well developed in large male; prothorax
and forefemora well developed in large male;
foretarsus with a tooth which is variable in shape
and size; prothorax with a strong median longitu-
dinal line in large male, without in small male.

Measurements of large (small) males in pm.
Total distended body length 1900 (1500). Head
length 175 (150), maximum width across cheeks
192 (185); eye length 65 (55). Pronotum median
length 175 (110), width 260 (210); forefemur
length 230 (130). Abdominal tergites median
length/width as follows: II 65 (55)/ 370 (330); IV 80
(65)/ 400 (350); VI 110 (90)/ 405 (340); VIII 95
(85)/ 280 (255); IX 75 (65)/ 170 (160). Tube length
145 (125), basal width 90 (78), apical width 44 (40).
Antennal segments I to VIII length/width as fol-
lows: 38 (37)/ 40 (35); 50 (48)/ 34 (31); 63 (57)/ 30
(30); 65 (55)/ 31 (30); 68 (58)/ 32 (30); 57 (48)/ 33
(30); 43 (38)/ 26 (25); 30 (25)/ 13 (13).

Length of setae. Postocellars 60-65 (32-40);

314 S. OKAJIMA

postoculars 98-102 (68-70). Prothoracic aa 35-45
(less than 20), am 28-35 (25), ml 85-105 (43-45),
pa 85-98 (48-50), epim 75-77 (60-65). By, on
tergite IX 80-82 (75-80), Bz 100-105 (85-88);
anals 110-120 (90-95).

Material examined. Taiwan: Makoo, Hookotoo
Is., lectotype $ (mic.) and 1 paralectotype ?
(mic.) of Neosmerinthothrips formosensis, 5-vi-
1930, S. Minowa (in coll. Senckenberg Museum,
Frankfurt); Pintung Hsien, Kenting National Park,
on grass, 1 (mic.) 25-v-1972, 271 (mic.) and
1 (mac.) 18-11-1984, 293% (mic.) and 2?
(mac.) 19-iii-1984, S. Okajima; Kaohsiung Hsien,
Liukuei, 22 (mic.) on bush, 22-iii-1984, S. Oka-
jima; Nantou Hsien, Nanshanchi, on grass 32.1
(mic.) 25-11-1984, 2 (mic.) 30-iii-1984, S. Oka-
jima; Nantou Hsien, foot of Mt. Nonkao, nr.
Wanta, | ? (mic.) on grass, 1-iv-1984, S. Okajima;
Taipei Hsien, Mt. Tatung-shan, 22.1 f (mic.) on
grass, 4-iv-1984, S. Okajima; Lan-Yu, 2 (mic.)
6-vi-1980, H. Makihara. Japan, the Ryukyus (in-
cluding Satsunan Isls.): Yonaguni Is., Mt. Urabu,
22 (mic.), 19-11-1977, W. Suzuki: Yonaguni Is..
Sonai, 1? (mic.) on dead oak, 29-iii-1975, S.
Saito; Ishigaki Is., 1? (mac.), 6-vi-1971, S. Oka-
jima; Ishigaki Is., Mt. Omoto, 1 (mic.) 12-vi-
1972, 12 (mic.) 14-vi-1972, S. Okajima; Ishigaki
Is., Hirakubo, 32 (mic.) on Miscanthus, 15-vi-
1972, S. Okajima; Iriomote Is., Sonai, 2 (mic.)
on dry twigs, 19-vi-1972, S. Okajima; Iriomote Is.,
Mt. Tedou, 12 (mic.) on dead Palmae, 19-vi-
1972, S. Okajima; Okinawa Is., Nago, 2?1¢/
(mic.) on dead leaves, 13-v-1972, S. Okajima;
Okinawa Is., Katsuyama, 2? (mic.) on dead
leaves, 2-vii-1973, S. Okajima; Yuku Is.,
Onoaida, 97-2 { (mic.) on Miscanthus, 6-1-1972,
K. Haga. Japan, Honshu: Kyoto Pref., nr. Shizu-
hara, 1" (mic.) and 1 (mac.) on grass, 6-vilii-
1980, S. Okajima; Aichi Pref., Nisshin Nokata,
12 (mic.) in soil, 13-i-1986, T. Kato; Fukui Pref.,
Tsuruga, Shiraki, 4 (mic.) on dead branches,
7-vil-1978, S. Okajima; Kanagawa Pref., nr. Tsu-
kui, 221 (mic.) on grass, 10-x-1982, S. Oka-
jima; Yamanashi Pref., nr. Fujiyoshida, 1°14
(mic.) on grass, 30-vii-1981, S. Okajima; Chiba
Pref., Ichikawa-shi, lower reaches of Riv. Tone-
gawa 1021 (mic.) and 5% (mac.) on grass,
vi-1983, R. Terakoshi; Akita Pref., Honjoh-shi,

lower reaches of Riv. Koyoshi-gawa, 1528 7%
(mic.) and 4 (mac.) on grass, 9-vii-1988, S.
Okajima. Japan, Ogasawara Islands: Chichi-jima
Is., Mt. Asahi-yama, 1 ? (apt.) on dead branches,
11-11-1988, S. Okajima; Haha-jima Is., nr. Mina-
mi-zaki, 1 (apt.) on grass, 5-1ii-1988, S. Oka-
jima.

Nesothrips malaccae Mound, 1974
(Fig. 21)

Nesothrips malaccae Mound, 1974 [1], 164-166.

This species was described by Mound [1] based
on three females from West Malaysia and Sumat-
ra. Recently, additional females and males were
collected from West Malaysia, Indonesia (Sulawesi
and Bali) and the Philippines (Luzon and Minda-
nao). These records suggest its wide distribution in
Southeast Asia.

All females and males of this species are mac-
ropterae, micropterae still having been unknown.
There are some differences between local popula-
tions in colour of the legs and antennae. Speci-
mens from West Malaysia have the third antennal
segment largely brown and all femora brown with
distal third yellow, but specimens from southern
Sulawesi have the third antennal segment yellow to
brownish yellow and all femora brown with distal
half yellow. However, those from central
Sulawesi, Bali and the Philippines are something
intermediate between them.

The males are recorded for the first time, and a
brief description is given below based on a male
from West Malaysia.

Male (macroptera). Colour almost as in female.
Head (Fig. 21) 0.88 times as long as broad; post-
ocular setae longer than half the length of head,
postocular and middorsal (vertexal) setae well
developed; prothorax well developed, pronotum
with a strong median longitudinal line, ml well
developed, much longer than pa; forefemora en-
larged, foretarsus with a strong tooth; B> and B; of
subbasal wing setae long; tube 1.17 times as long as
head.

Measurements of male in um. Total distended
body length 2220. Head length 210, width across
cheeks 240; eye length 70. Pronotum median
length 184, width 290; forewing length 890. Pelta

Nesothrips from East Asia 315

median length 100, width 305. Tube length 245,
basal width 107, apical width 49. Antennal seg-
ments I to VIII length (width) as follows: 55 (46);
56 (34); 87 (32); 79 (36.5); 76 (34.5); 69 (31); 45
(26); 20 (16.5).

Length of setae. Postocellars 40-61, postoculars
137-140, middorsals (vertexals) 24-50. Prothor-
acic aa 60-70, am 25-55, ml 170, pa 122, epim
140-150. Forewing subbasals B, 45-103, Bz 175-
200, B3 217-227. B, on tergite IX 180-185, Bz on
IX 121-148: anals 172-190.

Material examined. West Malaysia: Cameron
Highlands, Tanah Rata, 291 (mac.) on dead
leaves, 2-iii-1976, 1 (mac.) 5-11-1976, W. Suzuki,
1 (mac.) on dead branches, 24-vii-1976, S. Oka-
jima; Genting Highlands, 30 ml E of Kuala Lum-
pur, 4,500’, 1 2 (mac.) on dead wood with leaves,

Fics. 19-23.

28-ix-1973, L. A. Mound. Indonesia: South
Sulawesi, Malino, alt. about 900 m, 3% (mac.) on
dead Palmae, 3-vili-1984, S. Okajima; South
Sulawesi, Karaenta Forest Res., Maros to Camba,
alt. about 400m, on dead branches, 3? (mac.)
5-vill-1984, 422 ¥ (mac.) 6-vill-1984, S. Okajima;
Central Sulawesi, nr. Rantepao, Pedamaran, alt.
about 1,000 m, on dead leaves and branches, 1 ?
(mac.) 8-viti-1984, 1? (mac.) 12-viil-1984, 1919
(mac.) 13-vili-1984, 19 (mac.) 14-vili-1984, S.
Okajima; Central Sulawesi, 31 km W from Palo-
po, Puncak, alt. about 1,300 m, 221 (mac.) on
dead leaves and branches, 19-vii-1984, S. Oka-
jima; Bali Is., Candi Kuning, alt. about 1,200 m,
4 (mac.) on dead branches, 26-vii-1984, S. Oka-
jima. Philippines: Luzon Is., Quezon National
Forest Park, 12 (mac.) on dead Palmae, 20-vii-

Head and tube of Nesothrips species—19-20, N. brevicollis, female, head (19) and tube (20); 21, N.
malaccae, male, head; 22-23, N. minor, female, head (22) and tube (23).

Figs. 24-28. N. yasumatsui sp.

nov. —24, female, head; 25, female, tube; 26, female, pelta; 27, small male, foretarsus; 28, large male, foretarsus.

316 S. OKAJIMA

1979, S. Okajima; Mindanao Is., Mt. Apo, Agko,
alt. about 1,300m, 1 (mac.) on dead leaves,
3-viil-1979, S. Okajima; Mindanao Is., Mt. Apo,
Agko, alt. about 1,100 m, 12 (mac.) 5-vii-1979,
W. Suzuki.

Nesothrips minor (Bagnall, 1921), sp. rev.
(Figs. 7-10, 14-15, 18, 22 and 23)

Coenurothrips minor Bagnall, 1921 [6], 287-288.
Neosmerinthothrips formosensis var. karnyi Pries-
ner, 1935 [4], 368-370.

Nesothrips minor: Mound, 1968 [7], 141.
Nesothrips brevicollis: Mound, 1974 [1], 162-163
(in part).

This species was described by Bagnall [6] based
on a macropterous female from Rodrigues. Re-
cently, Mound [1] treated it as a junior synonym of
brevicollis. However, the specific name minor is
now revived from a synonym of brevicollis, the
reason for which is given under brevicollis. Mound
[1] also treated N. formosensis var. karnyi, de-
scribed on macropterous females from Java, as a
synonym of brevicollis. However, it has the long
wing retaining setae and tube, which are typical of
minor but not found in brevicollis. Moreover,
recently collected material from India, West
Malaysia, Thailand, Indonesia (Java and Bali),
Ogasawara Isls. and Hawaiian Isls. could well be
determined as minor. These suggest its wide-
spread distribution in tropical and subtropical
Asia.

Contrary to the fact that the micropterous form
is common in brevicollis, the macropterous form is
prevalent in this species and the micropterous form
is only rarely collected. However, I have ex-
amined some micropterous females and males
from Thailand and Indonesia (Java and Sulawesi),
which have well developed, long and curved wing
retaining setae as in macropterae without excep-
tion.

Mound [1] regarded the Australian species rhi-
zophorae possibly as a mere local colour variant of
At present, rhizophorae is possibly
only a colour variant of minor.

Material examined. Rodrigues: Holotype &
(mac.) of Coenurothrips minor, vii to xi-1918, H.
J. Snell and H. P. Thomasset (in coll. British

brevicollis.

Museum (Natural History), London). Indonesia:
Java, Tjibodas, 1,400 m, lectotype ? (mac.) of
Neosmerinthothrips formosensis var. karnyi, 1923,
H. Karny (in coll. Senckenberg Museum, Frank-
furt); Java, Mt. Tengger, 52 (mic.) on dead
leaves, 14-iv-1981, T. Senoh; Bali Is., Kuta, Sea
level, 12 (mac.) on dead leaves and branches,
30-vill-1984, S. Okajima; South Sulawesi, Malino,
alt. about 900 m, 1 (mac.) on grass, 3-viii-1984,
S. Okajima; South Sulawesi, 11 km E from Mali-
no, Kanreapia, alt. about 1,500 m; 121% (mic.)
on grass, 2-vili-1984, S. Okajima. India: Aryank-
avu, 12 (mac.) on dry twing, 19-vii-1969, T. N.
Ananthakrishnan. West Malaysia: Genting High-
lands, 5,000’, 42 (mac.) and 4 (hemimac.),
8-x-1973, L. A. Mound. Thailand: Phuket Is.,
5?5f% (mac.) on dead leaves, 20-viii-1976,
209-15 § (mac.) on dead Banana leaves, 17-viii-
1976, S. Okajima; Bangkok, Bangkhen, Campus
of Kasetsart University, NBCRC, on dead bran-
ches of Casuarina equisetifolia, 2 { (hemimac.)
13-1-1988, 1 f (hemimac.) 18-i-1988, S. Okajima;
same locality as above, 1 (mac.) and 1 (mic.)
on Bougainvillea ? glabra, 28-xii-1987, S. Oka-
jima; nr. Chiang Mai, Mt. Doi Suthep, alt. about
500m, 16°10 (mac.) on grass, 8-viii-1976, S.
Okajima. Hawaiian Isls.: Hawaii, Pupukea, 12
(mac.) on Leucaena glauca, 16-x1i-1969, F. Andre;
Hawaii, Barber’s Point, 16-xii-1969, 12 (mac.) by
sweeping, F. Andre, 1 { (hemimac.) on Desman-
thus, K. Sakimura; Oahu, Honolulu, 12 (hemi-
mac.) on Cyperus rotundus, 11-xii-1960, K. Saki-
mura. Japan, Ogasawara Isls.: Haha-jima Is.,
Okimura, 3? (mac.) on dead leaves, 13-vi-1972,
Y. Watanabe; Haha-jima Is., 1 # (mac.) on dead
leaves and branches, 18-v-1984, M. Hasegawa;
Haha-jima Is., nr. Okimura, Mt. Chibusa-yama,
1 (mac.) on dead leaves, 9-iii-1988, S. Okajima;
Chichi-jima Is., Mt. Asahi-yama, 1 (mac.) on
bamboo (Pseudosasa japonica), 11-iii-1988, S.
Okajima; Chichi-jima Is., nr. Tokoyonotaki, 2 $
(mac.) and 1 (hemimac.) on dead branches,
11-iii-1988, S. Okajima.

Nesothrips yasumatsui sp. nov.
(Figs. 24-28)

Female (macroptera). Head yellow to brownish

Nesothrips from East Asia 317

yellow; thorax brown, in contrast with yellowish
head; abdomen brown to dark brown, segments II
and IX somewhat paler than intermediate seg-
ments; tube blackish brown. Antennal segments I
to VI yellow, almost concolorous with head or a
little paler, sometimes apex of segment VI shaded
with brown, segment VII yellowish brown, seg-
ment VIII brown. Legs yellow to brownish yellow,
almost concolorous with head. Postocular and
postocellar setae yellowish, all other major setae
on thorax and abdomen brownish.

Head (Fig.24) much broader than long,
broadest across cheeks, very weakly projecting in
front of eyes, dorsal surface weakly sculptured
posteriorly; cheeks rounded; postocular setae
acute; postocellar setae usually a little longer than
half the length of postocular setae, situated just
inside posterior ocelli. Eyes prolonged ventrally.
Ocelli small, about 13 ~m in diameter. Antennae
about 2.7 times as long as head; segments III and
IV with two and four sense cones respectively.
Maxillary stylets typical of the genus.

Pronotum weakly sculptured posteriorly, with a
weak short median line; major setae acute, am
reduced to a short and slender hair. Metanotum
sculptured with polygonal reticulation, but weak
medially; median pair of setae slender, about 35
ym long. Forefemora not enlarged, foretarsi un-
armed. Forewings each with 7 duplicated cilia;
only two subbasal wing setae developed, By, re-
duced.

Pelta (Fig. 26) with a pair of micro-pores; me-
dian lobe broad, lateral wings not separated from
median lobe, but somewhat constricted at base;
posterior margin not eroded. Tergal wing retain-
ing setae well developed, long and curved, but
reduced on tergite II; B; and B> setae on tergite IX
acute, much shorter than tube, B> longer than Bj.
Tube (Fig. 25) about 1.2 times as long as head
(excluding preocular part), sides almost straight;
anal setae much shorter than tube.

Measurements of holotype female (mac.) in um.
Total distended body length 1940. Head length
from anterior margin of eyes to base at middle 153,
width across cheeks 220; eye dorsal length 66,
ventral length 87. Pronotum median length 132,
width 244; forewing length about 800. Pelta me-
dian length 81, width 271. Abdominal tergites

median length (width) as follows: II 66 (406); IV
84 (464); VI 127 (472); VIII 107 (362); [X 78 (214).
Tube length 184, basal width 91, apical width 46.
Antenna total length 413, segments I to VIII
length (width) as follows: 47.5 (46); 50 (36); 63
(32); 60 (32); 58 (33); 56 (31.5); 39.5 (24); 29 (15).

Length of setae. Postocellars about ?40, post-
oculars 70. Prothoracic aa 37—40, am 32-40, ml
40-44, pa 55-58, epim 80-85. Forewing subbasals
B, 71-72, B3 82-84. B, on tergite IX 76-80, B> on
IX 97-100; anals 105-116.

Female (microptera). Colour and general struc-
ture almost as in macropterous female. Antenna
about 2.7 times as long as head; metanotum a little
shorter than that of macroptera; tube 1.05-1.16
times as long as head (excluding preocular projec-
tion); size of ocelli and shape of pelta very similar
to those of macroptera.

Measurements of female (mic.) in um. Total
distended body length 2020. Head length from
anterior margin of eyes to base at middle 158,
width across cheeks 220; eye dorsal length 71,
ventral length 84-87. Pronotum median length
143, width 256. Pelta median length 82, width 277.
Abdominal tergites median length (width) as fol-
lows: II 71 (372); IV 92 (470); VI 131 (484); VUI
111 (344); IX 78 (209). Tube length 183, basal
width 91, apical width 48. Antenna total length
402, segments I to VIII length (width) as follows:
47 (46.5); 52 (36); 60 (31); 59.5 (33.5); 55 (32.5);
52 (31); 38 (20); 28.5 (14).

Length of setae. Postocellars 42—45, postoculars
74-76. Prothoracic aa 38—40, am 25-28, ml 48-52,
pa 50-55, epim 95-105. B; on tergite 1X 82-85, B>
on IX 110-116; anals 92-116.

Male (microptera). Colour almost as in female.
Large male: prothorax well developed, pronotum
with a strong median longitudinal line, forefemora
enlarged, foretarsi each with a strong tooth (Fig.
28). Small male: Prothorax almost as in female or
smaller, pronotum with or without a weak median
longitudinal line, forefemora not enlarged, fore-
tarsi each with a small tooth (Fig. 27).

Measurements of large (small) males in pum.
Total distended body length 1500 (1200). Head
length from anterior margin of eyes to base at
middle 146 (128), width across cheeks 179 (163);
eye dorsal length 60 (51), ventral length 62 (62).

318 S. OKAJIMA

Pronotum median length 152 (96), width 219
(168); forefemur length 220 (118). Pelta median
length 51 (44), width 214 (143). Abdominal ter-
gites median length/ width as follows: II 46 (40)/
311 (229); IV 66 (50)/ 335 (253); VI 96 (76)/ 331
(252); VIII 81 (68)/ 228 (190); IX 66 (61)/ 146
(128). Tube length 148 (120), basal width 81 (66),
apical width 40 (38). Antennal segments I to VIII
length/ width as follows: 42 (34.5)/ 39.5 (32.5); 45
(39)/ 31.5 (30); 55 (46)/ 27 (24); 53 (43)/ 29 (27); 50
(42)/ 28 (28); 47.5 (43)/ 29 (28); 36 (29)/ 23 (20.5);
25 (24)/ 13 (13).

Length of setae. Postocellars 32-36 (26-29),
postoculars 70-80 (45-54). Prothoracic aa 44-48
(20-30), am 30-34 (20-25), ml more than 70
(240), pa 72 (37-40), epim 87 (58-60). Metanotal
medians 40—45 (22-24). B, on tergite IX 74 (61-
66), Bz on IX 88 (about 80); anals 90-95 (75-80).

Holotype $ (mac.). Thailand: Kamphaeng
Saen Campus of Kasetsart University, on grass,
22-xi1-1987, S. Okajima.

Paratypes (427122 in total). Thailand: 34
(mic.), data almost the same as for holotype, but
12-i-1988; Chiang Mai, 262.8 # (mic.) on grass in
rice field, 4-v-1978, K. Yasumatsu; 132.4 / (mic.),
data very similar to above, but 7-v-1978.

Comments. This species is somewhat similar in
appearance to propinquus (Bagnall) and fodinae
Mound. The pelta of propinquus is eroded post-
eriorly and the lateral wings are separated from the
median lobe, but the former is not eroded in this
species and the lateral wings are widely fused to
the median lobe. Moreover, the pedicel of the
seventh antennal segment of this species is more or
less broader than that of propinquus. From fodi-
nae, it can easily be distinguished by the paler head
and longer tube.

ACKNOWLEDGMENTS

I wish to express my gratitude to Prof. T. N. Anan-
thakrishnan, Loyola College, Madras, Dr. L. A. Mound,
Keeper of Entomology, British Museum (Natural His-
tory), London, and Dr. R. zur Strassen, Senckenberg
Museum, Frankfurt, for loan of specimens and other
ways. I am much indebted to Dr. Shun-Ichi Uéno,
National Science Museum, Tokyo, for kindly reading the
original manuscript. My thanks are also due to the late
Prof. K. Yasumatsu, Drs. K. Haga, T. Senoh and W.
Suzuki, Messrs. M. Hasegawa, T. Kato, H. Makihara,
K. Sakimura, R. Terakoshi and Y. Watanabe for their
kindness in offering specimens.

REFERENCES

1 Mound, L. A. (1974) The Nesothrips complex of
spore-feeding Thysanoptera (Phlaeothripidae: Ido-
lothripinae). Bull. Br. Mus. nat. Hist., (Ent.), 31:
110-188.

2 Mound, L. A. and J. M. Palmer (1983) The generic
and tribal’ classification of spore-feeding Thysano-
ptera (Phlaeothripidae: Idolothripinae). Bull. Br.
Mus. nat. Hist., (Ent.), 46: 1-174.

3 Bagnall, R. S. (1914) Brief descriptions of new
Thysanoptera III. Annls. Mag. nat. Hist., (8), 13:
287-297.

4 Priesner, H. (1935) New or little known oriental
Thysanoptera. Philip. J. Sci., 57: 351-375.

5 Kudo, I. (1974) Some graminivorous and gall form-
ing Thysanoptera of Taiwan. Kontya, Tokyo, 42:
110-116.

6 Bagnall, R. S. (1921) On Thysanoptera from the
Seychelles Islands and Rodrigues. Annls. Mag. nat.
Hist., (9), 7: 257-293.

7 Mound, L. A. (1968) A revision of R. S. Bagnall’s
Thysanoptera collection. Bull. Br. Mus. nat. Hist.,
(Ent.), Suppl. 11: 1-181.

ZOOLOGICAL SCIENCE 7: 319-325 (1990)

© 1990 Zoological Society of Japan

Nautiliniellid Polychaetes Collected from the Hatsushima
Cold-Seep Site in Sagami Bay, with Descriptions
of New Genera and Species

Tomoyuki Miura! and Lucien LAuBIER?

Faculty of Fisheries, Kagoshima University, Kagoshima 890, Japan, and
?IFREMER, 66 avenue d’léna, 75116 Paris, France

ABSTRACT—Polychaete species belonging to the family Nautiliniellidae were found in the mantle
cavities of two bivalve species collected from the Hatsushima cold-seep site in Sagami Bay at a depth of
1170m. As a result of the comparison with Nautiliniella calyptogenicola, new combination, two new
genera and two new species are described. The new genus Shinkai differs from the type genus of the
family in having only one pair of prostomial antennae instead of two pairs. Shinkai sagamiensis new
species is parasitic on Calyptogena soyoae. Natsushima bifurcata, new genus and species, differs from all
other species of the family by the presence of additional numerous bifurcate setae on each parapodium
instead of the exclusive presence of simple hooks in tne others. This species is parasitic in the mantle
cavity of an undescribed bivalve species of the genus Solemya. The position of the family Nautilinielli-
dae is discussed, after a reexamination of the type specimens of Antonbruunia viridis.

The polychaete family Nautiliniellidae characte-
rized by simple ventral hooks, was proposed with
the description of a single representative species,
Nautilina calyptogenicola Miura & Laubier, 1989
[1]. However, the classification of the family has
not been deeply discussed because of the scarcity
of the knowledge on the Japanese cold-seep com-
munity and on the parasitic polychaetes on
bivalves. In the course of the serial dives of the
deep-sea submersible “Shinkai 2000” of the Japan
Marine Science and Technology Center, the first
author (T.M.) could dive at the Calyptogena
soyae-dominant community of the Hatsushima
cold-seep site [2] which may be comparable to the
Calyptogena phaseoliformis-dominant sites of the
Japan Trench [3, 4]. During Dives 315, 316 and
381 of the submersible “Shinkai 2000” at depths of
1100 to 1200m, many specimens of cold-seep
bivalves and vestimentiferans were collected. The
parasites and associated invertebrates living on
these cold-seep animals were removed from their
hosts on the mother ship, “Natsushima” and two

Accepted June 28, 1989
Received May 29, 1989
' To whom all correspondence should be addressed.

polychaete species of the family Nautiliniellidae, a
species of the family Phyllodocidae and a species
of poecilostomatoid copepod were found [5, 6]. In
this paper, these two new species and new genera
of the nautiliniellid polychaetes parasitic in the
mantle cavity of cold-seep bivalves are described
and a new name is also proposed for the previously
described genus for reason of the preoccupation.
The types are deposited in the National Science
Museum, Tokyo (NSMT) and the Japan Marine
Science and Technology Center (JAMSTEC).

Nautiliniellidae

Type genus: Nautiliniella, new genus with the
type species Nautilina calyptogenicola Miura and
Laubier, 1989, by monotype.

Nautiliniella new genus

Type species: Nautilina calyptogenicola Miura
and Laubier, 1989, by monotypy.

Remarks: The generic name Nautilina is preoc-
cupied in a molluscan species and in a protozoan
species (after Nomenclator Zoologicus). The new
generic name Nautiliniella is proposed here. A

320 T. Miura AnD L. LAuBIER

newly combined name, Nailiniella calyptogenicola
is also proposed for the previously described spe-
cies, Nautilina calyptogenicola.

Shinkai new genus

Type species: Shinkai sagamiensis, new species,
by monotypy. Gender feminine.

Diagnosis: Body long, vermiform, tapering post-
eriorly with numerous setigerous segments; body
in cross-section flattened ventrally and more or less
arched dorsally. Prostomium short with a pair of
antennae, without eyes. Muscular proventriculus
present. Achaetous periostomial ring absent.
First setiger more or less fused with prostomium.
Parapodia subbiramous with dorsal and ventral
cirri; dorsal cirri well developed; ventral cirri very
short; neuropodia with a single embedded acicula
and a few simple stout hooks. Pygidium cylindrical
without appendage.

Etymology: The genus is named in the honor of
the submersible “Shinkai 2000” of the JAMSTEC
with which the host bivalves of the parasitic
polychaetes were collected during Dives 315 and
381.

Remarks: Species of the new genus Shinkai
resemble Nautiliniella calyptogenicola in having a
dorsally arched body in cross-section, a muscular
proventriculus and ventral simple hooks. However
they differ from the latter in having only one pair
of prostomial antennae instead of two pairs.

Shinkai sagamiensis, new species
(Fig. 1)

Materials: Holotype (NSMT-Pol. H-293), com-
plete with regenerated posterior segments, off
Hatsushima, Sagami Bay 34°00.0°N, 139°13.8’E,
1170 meters, 19 November 1987, deep-sea sub-
mersible “Shinkai-2000” Dive 315, collected from
the mantle cavity of Calyptogena soyoae. Para-
types (JAMSTEC), one anterior fragment, same
station as the holotype, from washings and sievings
of sediment with C. soyoae collected by a power-
driven grab; one complete, same site, 5 Novem-
ber, 1988, Dive 381, washings.

Measurements: Holotype, 14 mm long, 1.0 mm
wide including parapodia, with 65 setigers (38

anterior segments and 27 regenerated segements).
Larger fragmental paratype, 8.0 mm long, 1.2 mm
wide, with 32 anterior setigers.

Description: Body vermiform, flattened ven-
trally and slightly arched dorsally. Integument
smooth. Specimens preserved in alcohol pale or
colorless.

Prostomium very short, anteriorly incised, with
a pair of very short cirriform antennae, without
eyes or other appendages (Fig. la, b). Achaetous
periostomial ring absent. Mouth opening situated
ventrally, between prostomium and first setiger.
Ventral cirri of first setiger larger than followings,
inserted in front of neuropodia (Fig. 1f). Foregut
with well-developed muscular part (may be prove-
ntriculus).

Pygidium cylindrical, without anal cirri (Fig.
Ic).

Parapodia subbiramous, with well-developed
dorsal cirri and much reduced ventral cirri; first
dorsal cirrus greatly reduced; first ventral cirrus
located in front of neuropodial fascicle; bases of
dorsal cirri swollen, forming globular pads with
embedded slender notoacicula; dorsal cirri five
times as long as ventral ones; neuropodia globular
(Fig. 1d).

Setae consisting of simple ventral hooks only;
several hooks projected from each neuropodium
on anterior parapodia, e.g. 3-4 on parapodium 1,
5-8 on parapodia 2-6, 1-3 on parapodia 7-20, and
1 on posterior parapodia; several developing
hooks embedded around acicula. Hooks simple,
stout and strongly curved on very short distal end
with remarkable knob (Fig. le).

Etymology: The specific name is derived from
the type locality, Sagami Bay.

Natsushima new genus

Type species: Natsushima bifurcata, new spe-
cies, by monotypy. Gender feminine.

Diagnosis: Body long, vermiform, tapering
posteriorly with numerous setigerous segemnts;
body in cross-section flattened ventrally and slight-
ly arched dorsally. Prostomium short with a pair of
antennae, without eyes. Muscular proventriculus
present. Achaetous peristomial ring absent. Para-
podia subbiramous with dorsal and ventral cirri;

New Nautiliniellid Polychaetes 321

Fic. 1. Shinkai sagamiensis g. sp. n.: a, Anterior end, dorsal view (holotype); b, Same, ventral view; c, Pygidium,
dorsal view; d, Parapodium 19, anterior view; e, Hook; f, Anterior end, lateral view (paratype).

dorsal cirri longer than ventral ones; neuropodium
with a single embedded acicula, a few simple stout
hooks and many bifurcate simple setae. Pygidium
simple without appendage.

Etymology: The genus is named in the honor of
the research vessel “Natsushima” of the JAM-
STEC, the mother ship of the submersible “Shink-

ai 2000”.

Remarks: The new genus belongs to the family
Nautiliniellidae in having subbiramous parapodia
with characteristic simple ventral hooks, however
the single species of this genus differs from those of
other genera in having additional numerous smal-
ler bifurcate setae on each parapodium instead of

322 T. Miura AND L. LAUBIER

exclusive presence of simple hooks.

Natsushima bifurcata, new species
(Fig. 2)

Material: Holotype (NSMT-Pol. H-294), com-
plete, off Hatsushima, Sagami Bay, 34°00.0'N,
139°13.8E, 1170 meters, 19 November 1987,
deep-sea submersible “Shinkai-2000” Dive 315,
collected from the mantle cavity of Solemya sp.

Measurements: Holotype 5.0 mm long, 0.6 mm
wide including parapodia, with 47 setigers.

Description: Body vermiform, flattened ven-
trally and slightly arched dorsally. Integument

Fic. 2.

smooth. Specimens preserved in alcohol pale or
colorless.

Prostomium very short, anteriorly slightly in-
cised, with a pair of short cirriform antennae,
without eyes or other appendages (Fig. 2a, b).
Achaetous peristomial ring absent. Mouth open-
ing situated between prostomium and first setiger,
without jaws or paragnaths. First segment partial-
ly fused with prostomium. Foregut with well-
developed muscular part, without tubiform
pharynx.

Pygidium simple, rounded, without anal cirri
(Fig. 2c).

Parapodia subbiramous throughout body, with

ana 0.2 mm
0.04 mm
e f 0.004 mm

Natsushima bifurcata g. sp. n. (holotype): a, Anterior end, dorsal view; b, Same, ventral view; c, Pygidium;

d, Parapodium 16, anterior view; e, Hook; f, Bifurcate seta.

New Nautiliniellid Polychaetes 323

Fic. 3. Antonbruunia viridis Hartman and Boss, 1965 (Paratypes: USNM 56718): a, Anterior end, lateral view; b,
Posterior end, lateral view; c, Posterior parapodium, anterior view.

well-developed dorsal and short ventral cirri; bases
of dorsal cirri swollen, with very fine embedded
notoacicula; dorsal and ventral cirri of first setiger
reduced; neuropodia cylindrical (Fig. 2d).

Setae consisting of stout simple hooks and smal-
ler bifurcate simple setae; a few hooks (2-4)
projected from each setal lobe; several developing
hooks embedded around acicula. Hooks simple,
stout and slightly curved on distal end (Fig. 2e).
Numerous smaller bifurcate setae situated below
stout hooks. Distal teeth of bifurcate setae sepa-
rated each other, strongly curved (Fig. 2f).

Etymology: The specific name is derived from
the presence of bifurcate setae.

DISCUSSION

The single representative species of the family
Antonbruuniidae Fauchald, 1977 [7], Anton-
bruunia viridis Hartman and Boss, 1965, is known
as living in the mantle cavity of the bivalve Lucina

fosteri Hartman and Boss, 1965 [8]. This species
resembles the nautiliniellid species in having sim-
ple body with subbiramous parapodia and com-
mensal life. However, several important morpho-
logical differences between these two families were
found from the original description of A. viridis
and after the reexamination of four paratypes
(USNM 56718) of this species. A. viridis has five
occipital antennae including central unpaired one,
while nautiliniellid species have one or two pairs of
lateral antennae and lack the central one. There
are a distinct achaetous periostomial ring with tow
pairs of cirri and a pair of cylindrical anal cirri in
the former (Fig. 3a, b), while those are absent in
the latters. In A. viridis, each parapodium is
supported by three similar acicula with bases ad-
joining or contacted one another. Two neuropo-
dial acicula are directed to the setal fascicle and a
single notopodial one to the base of dorsal cirrus
but not penetrating inside the cirrus (Fig. 3c). The
appearance of acicula in nautiliniellied species

324 T. Miura AND L. LAUBIER

completely differs from that of A. viridis. In
nautiliniellid species, there are a single slender
notoaciculum and a single thick neuroaciculum.
The notoacicula of nautiliniellids are always situ-
ated apart from the neuroacicula and often embed-
ded in the dorsal cirri. They are very thin com-
pared with the neuroacicula. The setal composi-
tion is also different in these two families. Simple
stout hooks are present in all nautiliniellid species
but do not occur in A. viridis. A. viridis also
exhibits clear sexual dimorphism, with dwarf
males. This biological adaptation was not encoun-
tered in nautiliniellid species, which could be due
to the small number of specimens collected.

The discovery of two new species enables to
establish the morphological features of the family
Nautiliniellidae on a more definitive basis. The
major characters of the family are the presence of
a muscular proventriculus, subbiramous parapodia
and vental simple hooks, and the absence of
achaetous periostomial ring. Some pilargid
polychaetes resemble the species of the family
Nautiliniellidae in the characters mentioned
above, with the exception of the typical pilargid
notopodial acicular spines. The species of the
genus Litocorsa have stout neuropodial spines [9-—
11]. Although the simplified nautiliniellid body
recalls the species of the families Calamyzidae and
Levidoridae as mentioned in our previous paper
[1], the compound setal structure in these families
was not found in the nautiliniellid species. The
simple setae of the family Levidoridae are thought
to be derived from compound setae by fusion of
shafts and blades [12], while the typical vental
simple hooks of nautiliniellids may have originated
from simple ventral spines or setae, like the
neuropodial spines of Litocorsa. The simple struc-
ture of typical nautiliniellid ventral hooks may be
an important character supporting a hypothetical
relation of the family Nautiliniellidae with the
Pilargidae.

It was clarified above that the nautiliniellid
polychaetes may have another type of setae beside
the ventral simple hooks. The simple bifurcate
setae of Natsushima bifurcata rather recall those of
the family Oweniidae than the others. Nilsen and
Holthe discussed the phylogenetical development
of the typical oweniid uncini with two equal teeth

and considered that they might have derived from
the typical long-shafted uncini found in several
sedentary families [13]. The shafts of these uncini
are long and curved with distinct narrow parts
called “neak” [13, 14], and completely separated
from the nautiliniellid straight setae without neck.
The nautiliniellid bifurcate setae may have inde-
pendently evolved in the pathway to the parasitic
life.

As a conclusion, even if there may still be some
doubt, the family Nautiliniellidae should be placed
in the order Phyllodocida, near the family Pilar-
gidae.

ACKNOWLEDGMENTS

The authors wish to thank Dr. Suguru Ohta of Ocean
Research Institute, University of Tokyo for his identifica-
tion of the host vibalves and the staff of JAMSTEC for
their assistance in sampling the materials available for
study at “Shinkai 2000” Dives 315 and 381. We also
express our thanks to Dr. Kristian Fauchald and Dr.
Linda Ward of the Smithsonian Institution for their kind
arrangement of the loan of type material examined here.
Part of this study was supported by the grant-in-aid from
Itoh Science Foundation.

REFERENCES

1 Miura, T. and L. Laubier (1989) Nautilina calyp-
togenicola, a new genus and species of parasitic
polychaete on a vesicomyid bivalve from the Japan
Trench, representative of a new family Nautilinidae.
Zool. Sci., 6: 387-390.

2 Okutani, T. and K. Egawa (1985) The first under-
water observation on living habit and thanato-
coenoses of Calyptogena soyoae in bathyal depth of
Sagami Bay. Venus (Japan. J. Malacol.), 44: 285-
289.

3 Meétivier, B., T. Okutani, and S. Ohta. (1986)
Calyptogena (Ectenagena) phaseoliformis n. sp., an
unusual vesicomyid bivalve collected by the sub-
mersible Nautile from abyssal depths of the Japan
and Kurile Trenches. Venus (Japan. J. Malacol.),
45: 161-168.

4 Ohta, S. and L. Laubier (1987) Deep biological
communities in the subduction zone of Japan from
bottom photographs taken during “nautile” dives in
the Kaiko project. Earth Planet. Sci. Let., 83: 329-
342.

5 Miura, T. (1988) Parasitic animals collected in a
Calyptogena-dominant community developing off
Hatsushima, Sagami Bay. JAMSTECTR Deepsea

New Nautiliniellid Polychaetes

Res., 4: 239-244. (In Japanese).

Miura, T. (1988) A new species of the genus
Protomystides (Annelida, Polychaeta) associated
with a vestimentiferan worm from the Hatsushima
cold-seep site. Proc. Japan. Soc. Syst. Zool., 38: 10-
14.

Fauchald, K. (1977) The polychaete worms. Defini-
tions and keys to the orders, families and genera.
Nat. Hist. Mus. Los Angeles Cty., Sci. Ser., 28: 1-
190.

Hartman, O. and K. J. Boss (1965) Antonbruunia
viridis, a new inquiline annelid with dwarf males,
inhibiting a new species of pelecypod, Lucina fos-
teri, in the Mozambique Channel. Ann. Mag. Nat.
Hist., ser. 13, 8: 177-186.

Pearson, T. H. (1970) Litocorsa stremma a new
genus and species of pilargid (Polychaeta: Annelida)
from the west coast of Scotland, with notes on two
other pilargid species. J. Nat. Hist., 4: 69-77.

10

11

12

13

14

325

Wolf, P. S. (1986) Three new species of Pilargidae
(Annelida: Polychaeta) from the east coast of Flor-
ida, Puerto Rico, and the Gulf of Mexico. Proc.
Biol. Soc. Wash., 99: 464-471.

Imajima, M. (1987) Pilargidae (Annelida,
Polychaeta) from Japan (Part 1). Bull. Natn. Sci.
Mus., Tokyo, Ser. A, 13: 151-164.

Perkins, T. H. (1987) Levidoridae (polychaeta),
new family, with descriptions of two new species of
Levidorum from Florida. Bull. Biol. Soc. Wash., 7:
162-168.

Nilsen, R. and T. Holthe (1985) Arctic and Scan-
dinavian Oweniidae (polychaeta) with a description
of Myriochele fragilis sp. n., and comments on the
phylogeny of the family. Sarsia, 70: 17-32.
Imajima, M. and Y. Morita (1987) Owentidae
(Annelida, Polychaeta) from Japan. Bull. Natn. Sci.
Mus., Tokyo, Ser. A, 13: 85-102.

tabacum nine =

r

; fo eat deideyp cee ret Ty ON Rh SR IT aa warmers
: . ii esemiond ff “alba As ihe :
. ern ee Le ee any wheel etsy
; Lan OE Fs Se wid
i i ait te shear W as
_ ; “iid eee FS
mo 7 parse miter’ tr iit hae” See

Z 23:0 ok ge j eystaoriai > Pa vow usa
y a pvinetebiltven te hota VCs she SA arom rserwe
1 dep areca teopaees ty L pate. ae | | ene + Wie 208, pia

, f f Hed . sbookt pin) dee t.
tees, | dex hawel) aod
_ iL) ~- MTA FF pen f el = a
| PERT a oT tote ea et Soop ea) ee Bitsank’ aie
in They ERO Ae. all ‘heaps Sipe
4 ie MANS nt: Oe Gh Oy pad Meyers

( eres. lari be anegtiv Oe, A = E gate se
. PA. 1c Xo Nedttin a be - Geel aern wi want i, (ANAT erica (OTOL)

i letsadiv er puede” (ellie aout phen i

re a al Lath j . , yan Ws sapstenr’ &

ee + One Ai, te ke
q : rama Lay aa c=
| ae: pos ae

i elke

a Pra)

te

> ne | doy |
u bia tS
; med

‘ ‘
ey |
‘
vie | ¥
| ari.
, z & i ht hl

nr ppayt¥

\ANMBITOT

ZOOLOGICAL SCIENCE 7: 327-330 (1990)

[COMMUNICATION]

© 1990 Zoological Society of Japan

Tectal Visual Afferents from Fish
Dorsolateral Tegmental Cells

AKIYOSHI NiDA and TAKASHI OHONO

Department of Biology, Faculty of Science, Okayama University,
Okayama 700, Japan

ABSTRACT—The cells of the nucleus dorsolateralis
tegmenti (NDT) in the crucian carp were physiologically
identified and marked with Lucifer dye. The Lucifer
dye filled axons projected into the tectum, where
their main axons extended into the deep tectal
layer. All the identified NDT cells responded to
both optic nerve and rhombencephalic electrical
stimulation. Out of 40 such NDT cells, 24 cells
were visual and/or tactile. The remaining cells
were unresponsive. However, some of the unre-
sponsive cells were visually driven in conjunction
with rhombencephalic electrical stimulation.

INTRODUCTION

The nucleus dorsolateralis tegmenti (NDT) in
fish is located ventrolaterally to the torus semicir-
cularis, which is considered to be a recipient of
visual, auditory and lateral line inputs [1-3]. The
dorsolateral tegmental area, including the NDT or
deep tegmentum has been found to be reciprocally
connected with the optic tectum by degeneration
and HRP-labelling studies [4-9]. In a previous
study [10] by means of intra-axonal dye marking
and intracellular recordings we obtained the fol-
lowing results: (1) wide distribution of axonal
branching of the NDT cells in the deep layer of the
ipsilateral tectum; (2) further projection of the
axon described in (1) to the contralateral tectum
via the tectal commissure; (3) responses of the
NDT cells to three set of electrical stimulation,
optic nerve, rhombencephalon and tectal commis-
sure. These results strongly indicate that there

Accepted May 24, 1989
Received March 17, 1989

exist, in the tectum, afferents with various kinds of
response modalities from the tegmentum. The
goal of this study is thus to show the presence of
visual or other sensory related responses of the
NDT cells, which were identified by physiological
criteria combined with cell morphology.

MATERIALS AND METHODS

Experiments were performed on 50 crusian carp
(Carassius carassius), 15-20 cm in overall length.
The surgical procedure, methods for electrical and
sensory stimulation, the recording apparatus and
histological procedures have been described in
detail elsewhere [10, 11]. The fish were initially
anesthetized with MS-222 and immobilized with an
intraperitonial injection of Flexedil. The gills were
kept out of water, and perfused with aerated water
through a tube inserted into the oral cavity.
Beveled glass micropipettes, filled with 4% Lucifer
Yellow CH (Sigma) in distilled water [12], were
used for potential recording and markings. A
hyperpolarizing DC current of 2nA for 2-5 min
gave good marking of the cells. The brain was
removed 3-5 hr after the injection of dye and fixed
for 13-15 hr with formalion acidified to pH 4.0
with acetate buffer.

The criteria for physiological identification of
the NDT have been established in a previous study
[10] by a combination of intra-axonal recording
and Lucifer dye marking. The criteria were: 1)
antidromic response of the axon, running through
the stratum album centrale (SAC), to electrical
stimulation (300 Hz) of the tectal commissure and

328 A. NDA AND T. OHNO

2) orthodromic response of the same axon de- _ [10], the following latency values were used for the
scribed in (1) to electrical stimulation of both the identification of the NDT cell: 0.4—0.8 ms for the
rhombencephalon and the optic nerve. In addition _ tectal commissure, 5-10 ms for the optic nerve and
to such criteria, based on a previous experiment 1.2—1.6 ms for the rhombencephalon.

Fic. 1. Fluorescence photomicrograph of a NDT cell filled with Lucifer dye. This montage photograph was prepared
by two serial coronal sections. The NDT cell is located ventrolaterally to the torus semicircularis. Dendrites
(arrow heads) extend toward the tectobulbar pathway and the axon (thin arrows) arising from the soma (asterisk)
ascends toward the ipsilateral tectum. The axon (thick arrows) running through SAC, the deep tectal layer could
be further traced to the contralateral tectum by observing serial sections. The overall morphology of this cell is
shown in Figure 2Bb. Lines of squares lateral and medial to the midline show a lower boundary of SAC and a
part of the wall of the optic ventricle, respectively. The tectobulbar pathway courses downward along the line of
squares lateral to the midline. Abbreviation: O.T., optic tectum; SAC, stratum album centrale; T.S., torus
semicircularis; V.C., valvula cerebelli.

Tectal Afferents from Fish Tegmentum 329

RESULTS AND DISCUSSION

Using above-mentioned criteria, we observed
the responses of 40 NDT cells. Twenty-one of
these cells, in which Lucifer dye had been injected,
were well-stained (as seen from Fig. 1) and permit-
ted tracing of the axon to the contralateral tectum.
Seven of the 40 cells identified as NDT were
visuo-tactile, 16 cells were visual, 1 cell was tactile
and 16 cells were unresponsive.

Visuo-tactile cells

Figure 2Aa and 2Ab show one example of
recordings from the bimodal NDT cells. These
responses were obtained from the cell illustrated in
Figure 2Ba: A spot of light (0.5 subtense angle)
induced a transient response and subsequent sti-
muli gave a reduced number of spikes (Fig. 2Aa),
indicating remarkable habituation. Simul-
taneously, this cell also responded to the tactile
stimuli delivered by touching the facial part (stip-
pled area in the inserted drawing) with a writing
brush (Fig.2Ab). Another example from the
visuo-tactile cell following morphological iden-
tification is shown in Figure 2Ac, where normally
occurring spontaneous discharges notably in-
creased by touching the facial part. This cell
responded to moving objects as well (not shown
here). Bimodal units obtained by extracellular
recordings have been reported by Page and Sutter-
lin [1] in the dorsolateral tegmentum of goldfish
that are closely related to the present matrial.
Unlike our results, they were all acoustico-visual
units, This discrepancy is possibly due to differ-
ences in their topographycal positions where visuo-
acoustic and visuo-tactile cells are located: the
recording sites shown by Page and Sutterlin [1] lie
more anterior to those of our cells.

Visual cells

Besides visual NDT cells coupled with tactile
input, we encountered visual NDT cells with
trhombencephalic inputs, which were not ascer-
tained in response modality. They were mostly
on-transient or sensitive to moving objects. As
seen from Figure 2Ad, the exemplified NDT cell
was directionally selective: the leading edge of a
black rectangular stripe (subtense angle: 8°), mov-

‘
= =e ——_—— }40mv
T N N T 500ms

jy Mae Cea im Ca oe
a en |

Fic. 2. A: responses of NDT cells to visual and/or
tactile stimuli. (Aa) responses of the bimodal cell to
a 0.5° spot of light, and (Ab) responses of the same
cell to touching of the facial part (inserted drawing).
Dot pattern representation of spike discharges,
together with (Ac). These responses (Aa and Ab)
were recorded from the cell shown in Figure 2Ba. In
both responses (Aa and Ab), remarkable habitua-
tion occurs, whereas another bimodal cell (Ac)
shows much less habituation with the responses
induced by tactile stimuli. Their tactile receptive
fields were on the facial part (stippled area in the
inserted drawing). Upward deflection in each trace
shows light-on for Aa and Ae, and touch for Ab and
Ac. In Ab and Ac, time scale represents 1 sec. (Ad)
responses of NDT cell to moving edge. When a
leading edge of a black rectangular stripe moved in
the temporal to nasal direction (as seen from the
inserted drawing), a response was vigorously in-
duced. A stationary spot of light to this cell was
almost ineffective (Ae). These responses were
recorded from the cell shown in Figure 2Bb.
Calibration in (Ad) also serves for (Ae). B: compo-
site drawings of NDT cells marked with Lucifer dye.
Explanation of each cells’s morphology is given in
the text. Each arrow indicates axon and asterisks
position of soma. Calibration bars: 50 ~m. Abbre-
viation: D, dorsal; L, lateral; M, medial; N, nasal;
T, temporal; V, ventral.

ing (40°/sec) in the temporal to nasal direction
through the receptive field, produced spike dis-
charges, whereas motion in the reverse direction
(nasal to temporal direction) gave a much weak
response. In this example, the slightly deviated

330 A. NupA AND T. OHNO

orientation of the edge from naso-temporal axis, as
seen in the inserted drawing, was the most
effective in the initiation of spike discharges. A
moving spot of light gave no response (not shown),
and a stationary spot of light was also not effective
in this cell (Fig. 2Ae). The response characteristics
mentioned above were obtained from the cell in
Figure 2Bb, where the axon (thin arrow) filled with
Lucifer dye projected into the ipsilateral optic
tectum, and the dendrite field of this cell expanded
in a fan-like manner toward the tectobulbar
pathway (see also Fig. 1). Unlike the cell illus-
trated in Figure 2Ba and 2Bb, in Figure 2Bc the
ventral dendrite of the NDT cell, which was
sensitive to moving objects, extended toward the
F.L.L. (fasciculus longitudinalis lateralis). All the
identified visual cells, except for the bimodal cells,
were unresponsive to acoustic and/or tactile stimu-
li. However, they responded to rhombencephalic
electrical stimulation, indicating that these cells
receive rhombencephalic inputs, such as lateral
line, vestibular, and possibly proprioceptive in-
formations.

Unresponsive cells

Among NDT cells there were some unrespon-
sive cells in a slightly greater frequency (about
40%). They did not respond to visual, acoustic or
tactile stimuli, although these cells were responsive
to both optic nerve and rhombencephalic electrical
stimulation. Some of these cells, however, like
some of tectal efferent cells [11], which normally
failed to response to acoustic, tactile, or visual
stimuli, responded to light simuli under specific
conditions: when continuing the rhombencephalic
electrical stimulation, simultaneously applied
visual stimuli induced visual responses. One of the
possible interpretations for this, is that summation
of visual input and heterosynaptic sensory inputs
might activate responsiveness of the NDT cell. As
yet neural mechanisms responsible for the unre-
sponsiveness of these cells remain to be studied in

detail. Morphological features of the unresponsive
cells were substantially similar to those of visual or
bimodal cells: the wide distribution of the axonal
branching in the tectum, the dendritic profile
extending toward the tectobulbar pathway, and
the cell locations similar to those of responsive cell
of the NDT.

The present study raised a crucial problem that
the responses, in the deep tectum, derived from
the NDT cells may be erroneously identified as
intrinsic tectal unitary reponses, unless we use
well-defined criteria for determining whether the
responses recorded in the tectum originate from
the intrinsic tectal cells or from the tegmental cells.
The same appears to be the case in tectal afferents
from the pretectal area and the nucleus isthmi.
This situation will be overcome by a combination
of more sophisticated electrical stimulation and
recording techniques by which cell identification
can be made.

REFERENCES

1 Page, C. H. and Sutterlin, A. M. (1970) J.
Neurophysiol., 33: 129-136.

2 Knudson, E. I. (1977) J. comp. Neurol., 173: 417-
432.

3 Schellart, N. A. M. (1983) Neurosci. Lett., 42: 39-
44,

4 Ebbesson, S. O. E. and Vanegas, H. (1976) , 184:
435-454.

6 Grover, B. G. and Sharma, S. C. (1981) J. comp.

Neurol., 196: 471-488.

Luiten, P. G. M. (1981) Brain Res., 220: 51-65.

Wolf, F. A. De., Schellart, N. A. M. and Hoogland,

P. V. (1983) Neurosci Lett., 38: 209-213.

9 Echteler, S. M. (1984) J. comp. Neurol., 230: 536-
Spl.

10 Niida, A. and Ohono, T. (1984) Neurosci. Lett., 48:
261-266.

11 Niida, A., Ohono, T. and Iwata, K. S. (1989) Brain
Res. Bull., 22: 389-398.

12 Stewart, W. W. (1978) Cell, 14: 741-759.

on

ZOOLOGICAL SCIENCE 7: 331-334 (1990)

[COMMUNICATION]

© 1990 Zoological Society of Japan

Morphological and Physiological Characterization of the
Para-Ocellar Nerve of the Cockroach,
Periplaneta americana

AKIKO MIZUTANI and YOSHIHIRO ToH

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

ABSTRACT—The structure and physiological prop-
erties of the para-ocellar nerve have been examined in
American cockroaches. The para-ocellar nerve contains
120-200 fibers, originating from several bundles around
the ocellus. The fibers project into the deutocerebrum
and descent to terminate in the suboesophageal ganglion.
Spike discharges were recorded in the para-ocellar nerve
in response to air puffs directed towards the ocellar
cornea, indicating that the para-ocellar nerve contains
axons of mechanosensory sensilla surrounding the ocel-
lus. It was previously assumed that the para-ocellar
nerve coupled ocellar photoreception with neurosecre-
tion of the corpora cardiaca. However, our data could
not confirm this assumption, since we could not find
responses of the para-ocellar nerve to ocellar illumina-
tion.

INTRODUCTION

The dorsal ocellus of insects is usually connected
with the protocerebrum by a single ocellar nerve
[1]. In cockroaches, however, the ocellus is
connected with the brain by two nerves: the ocellar
nerve and the para-ocellar nerve. Structure of the
ocellus and ocellar nerve has been well
documented in cockroaches [1-6], but only little is
known about the para-ocellar nerve [7]. Silver
impregnation showed two courses of para-ocellar
nerve fibers in the deutocerebrum. Some fibers
terminated with arborizations in the superior
internal cortical area of olfactory lobe, where they
keep close proximity to the neurons projecting

Accepted June 29, 1989
Received March 10, 1989

their axons into para-cardiac nerve. Other fibers
further descend to the circumoesophageal connec-
tives [7]. Based upon this morphology the para-
ocellar nerve was assumed to couple ocellar
photoreception with the neurosecretion of the
corpora cardiaca [7]. However, this assumption
has not been examined electrophysiolgically or
anatomically using electron microscopy.

In the present study, the para-ocellar nerve
fibers of American cockroaches Periplaneta amer-
icana have been examined using cobalt backfills,
electron microscopy and electrophysiological tech-
niques.

MATERIALS AND METHODS

Morphology

The para-ocellar nerves of the cockroaches P.
americana were prepared for scanning and trans-
mission electron microscopy through the same
procedures previously reported for the ocellus and
ocellar nerve of the same species [4, 5, 8]. The
projections of the para-ocellar nerve fibers in the
brain were studied by backfilling the fibers from
the ocellus with cobalt chloride. Backfilled fibers
were examined by light microscopy using the same
procedure previously described for the ocellar
nerve fibers in the same species [4, 5, 8].

Physiology

The cockroach was fixed on an acrylic platform

332 A. MIZUTANI AND Y. ToH

1. A surface view of the ocellar region. Chaetic hairs occur around the ocellus. C, ocellus. X68.
Fic. 2. A bihalved brain viewed from the median plane. The para-ocellar nerve (PON) enters the deutocerebrum

Fic.
(DC). CC, circumoesophageal connective; ON, ocellar nerve; PC, protocerebrum; SG, suboesophageal

ganglion. 65

Para-Ocellar Nerve of Cockroach 333

with bees wax, and the para-ocellar nerve was
exposed by partial removal of the cuticular integu-
ment. A suction electrode filled with physiological
saline [9] was attached to the mid-region of the
para-ocellar nerve, an indifferent electrode being
inserted into the apex of the head. Two different
stimuli were presented. Light from a tungsten
lamp was focused upon the ocellus, its intensity
being high enough to produce maximal ocellar
ERG. Air puffs derived from an air compressor
were directed towards the ocellar cornea through
silicon tubes (2mm in diameter). The outlet of the
tube was 10mm apart from the head, and the
velocity of the air stream was about 4 m/sec.
Responses of the para-ocellar nerve were AC-
amplified and photographed. Moreover, re-
sponses of chaetic sensilla located around the
ocellar cornea to air puffs were extracellularly
recorded by a sharpened tungsten electrode in-
serted into the base of the sensillum.

RESULTS AND DISCUSSION

The cockroach P. americana possesses a pair of
ocelli: each ocellus (400 x 500 «m) is located near
the base of the antenna. The ocellar cornea is
externally depressed, and sensory hairs occur
around and on the cornea (Fig. 1). The ocellus is
connected to the brain by two nerves. The ocellar
nerve (50m in thickness, 500 um in length)
originates in the posterior part of the ocellus, and
enters the protocerebrum, whereas the para-
ocellar nerve (25 um in thickness, 250 um in
length) originates in the anterior part of the ocellus
and enters the deutocerebrum (Fig. 2). The para-
ocellar nerve, viewed in cross sections through the
mid-region, contains 130-200 fibers, about ten of
them ranging between 1-5 um (Fig. 4). The fibers
contain many microbutules and mitochondria and
only few vesicles. Adjacent fibers are separated
from each other by thin glial cell envelopes, but no
synaptic specializations occur among them. The
para-ocellar nerve appears to be divided distally

into several bundles near the ocellus (Fig. 3).

The distribution of the para-ocellar nerve fibers
within the CNS can be seen with cobalt backfills
from the ocellus. The cobalt backfills show quite
different courses of the ocellar nerve fibers and
para-ocellar nerve fibers in the brain (Fig. 5).
Usually several para-ocellar nerve fibers were
stained. They descend through the ipsilateral
circumoesophageal connective, and terminate in
the suboesophageal ganglion. On their descending
way they project lateral branches (up to 50-80 um
in length) in the mid-region of the circu-
moesophageal connective and into the anterior
part of the suboesophageal ganglion (Fig. 6). They
change their course in the suboesophageal gang-
lion towards the median plane, and terminate
there with many branches (Fig. 7). Unlike the data
obtained by silver impregnation [7], fibers, which
were arborized and terminated around the olfac-
tory lobe could not been cobalt-backfilled in the
present study.

The distribution of the para-ocellar nerve fibers
in the CNS, as was shown in the present study, is
similar to previously reported mechanoreceptor
axons: some examples are antennal mecha-
noreceptors of the locust which also terminate in
the suboesophageal ganglion [10, 11]. In accord-
ance with these anatomical findings it is assumed
that the para-ocellar nerve may consist of primary
axons of mechanoreceptive sensilla around the
ocellar cornea: these axons must be backfilled by
cobalt chloride which diffused out from the ocellus
to the surrounding region. This assumption was
physiologically confirmed. We found that the
sensilla surrounding the ocellus responded to air
puffs, and spike discharges were also recorded
from the para-ocellar nerve in response to air puffs
directed towards to ocellar cornea (Fig. 8).

These data suggest that the para-ocellar nerve
contains axons transmitting mechanosensory in-
formation to the CNS. Whether the para-ocellar
nerve is also involved in transmitting ocellar
information remains unresolved, since we could

Fic. 3.
para-ocellar nerve shown in Fig. 4. 2,200.

Bundles of axons near the ocellar capsule. These bundles come together on their way to the brain to form the

Fic. 4. A cross section of the para-ocellar nerve near the deutocerebrum. Of about 200 fibers included three (as

terisks) are more than 5 wm thick. 5,200.

334 A. MIZUTANI AND Y. Tou

Fic. 5. Cobalt backfilled para-ocellar nerve fibers (PON) and ocellar nerve fibers (ON) in a lateral view of the

bihalved brain.

Arrow, a cell body of the ocellar nerve fiber; CC, circumoesophageal connective; DC,

duetocerebrum; PC, protocerebrum; TC, tritocerebrum; SG, suboesophageal ganglion. X30.
Fics. 6 and 7. Cobalt backfilled para-ocellar nerve fibers (PON) in frontal view of the brain. Note side branches

(arrows) in Fig. 6, and terminal arborizations (arrow) in Fig. 7.

CC, circumoesophageal connective; SG,

suboesophageal ganglion. X75 in Fig. 6; 94 in Fig. 7.

Fic. 8. Responses of the chaetic sensillum (A) and
para-ocellar nerve fibers (B, C) to air puffs to the
ocellar cornea. The response of the para-ocellar
nerve fibers is phasic (B) or tonic (C). Multiple units
were usually recorded (B). Bottom bar indicates the
stimulus duration.

not find any responses to ocellar illumination.
More elaborate studies are required to answer this
question, because only several, probably thick,
fibers were stained and their activities were re-
corded in the present study, and morphology and
physiology of remaining fibers are still prob-
lematic.

ACKNOWLEDGMENTS

This work was supported in part by a grant-in-aid for

Special Project Research on Molecular Mechanism of

Bioelectrical Response 60123002 from Japanese Ministry
of Education, Science and Culture. The authors express
their thanks to Dr. J. M. Ramirez, Department of
Physiology, University of Alberta, Edmonton, Canada
for critical reading of the manuscript.

REFERENCES

1 Goodman, L. J. (1981) In “Handbook of Sensory
Physiology”. Ed. by H. Autrum, Springer, Berlin,
Vol. VII/6C, pp. 201-286.

2 Cooter, R. J. (1975) Int. J.
Embryol., 4: 273-288.

3 Weber, G. and Renner, M. (1976) Cell Tissue Res. ,
168: 209-222.

4 Toh, Y. and Hara, S (1984) J. Ultrastruct. Res., 86:
135-148.

5 Toh, Y. and Sagara, H. (1984) J. Ultrastruct. Res.,
86: 119-134.

6 Koontz, M. A. and Edwards, J. S. (1984) Cell
Tissue Res., 236: 133-146.

7 Brousse-Gaury, P. (1964) C. R. Acad. Sci. (Paris),
267: 649-650.

8 Toh, Y. and Yokohari, F. (1988) J. Comp. Neurol.,
269: 157-167.

9 Yamasaki, T. and Narahashi, T. (1959) J. Insect
Physiol., 3: 146-158.

10 Aubele, E. and Klemm, N. (1977) Cell Tissue Res.,
178: 199-219.
11 Gewecke, M. (1979) Entomol Gen., 5: 317-320.

Insect Morphol.

D | nt Published Bimonthly by the Japanese Society of
eve opm e Developmental Biologists
Distributed by Business Center for Academic

Growth & Differentiation Societies Japan, Academic Press, Inc.

Papers in Vol. 32, No. 2. (April 1990)

J.D. Ebert: Yoshihiro Kato (1924-1988)—A celebration of a life in science

13. L.E. Stephens, G. W. Shiflet and F. Wilt: Gene expression, DNA synthesis and protein synthesis in cells
from dissociated sea urchin embryos

14. H. Urushihara, T. Saigo and K. Yanagisawa: Cell fusion promoting factor common to homothallic and
heterothallic mating system in Dictyostelium discoideum

15. M.R.Diaz, T.C. Takahashi and K. Takata: Concanavalin A acts as a factor in establishing the
dorso-ventral gradiant in the ventral mesoderm of newt gastrula embryo

16. H.Kondoh: The mechanism of o1-crystallin gene regulation: cooperation of lense-specific and non-specific
elements

17. Y.-C. Hsu: Heterogenous macromolecular contributions to early mouse embryo development

18. M. Ito, T. Kaneko-Ishino, F. Ishino, M. Mitsuhashi, M. Yokoyama and M. Katsuki: Developmental
potential of haploid-derived parthenogenic cells in mouse chimeric embryos

19. S. Tanaka and K. Dan: Study of leneage and cell cycle of small micromeres in embryos of the sea urchin,
Hemicentrotus pulcherrimus

20. K. Kitamura, M. Sezaki and M. Yanazawa: Analysis of embryonic chick periderm by monoclonal antibody
ageinst periderm

21. H. Nakano, K. Kin Shita, K. Ishii, H. Shibai and M. Asashima: Activities of mesoderm-inducing factors
secreted by mammalian cells in calture

22. T. Mizuno, K. Kitamura, M. Saito and S. Tanemura: Epidermal metaplasia induced on amniotic ectoderm
by the dermis of chicken embryos

23. Y. Suzuki, T. Obara, S. Takiya, C.-C. Hui, K. Matsuno, T. Suzuki, E. Suzuki, M. Ohkubo, and T. Tamura:
Differential transcription of the fibroin and sericin-1 genes in cell-free extracts

24. R.-P. Huang, H. Muramatsu, and T. Muramatsu: Effect of different conditions of retinoic acid treatment
on expression of MK genes, which is transienthy activated during differentiatin of embryonal carcinoma cells

25. N. Hashimoto, S. Iwashita, Y. Shoji-Kasai, T. Kishimoto and K. Imahori: Thiol protease inhibitor, E-64-
d, prevents spindle formation during mouse oocyte maturation

26. Papaconstantinou, J., J.P. Rabek and Dong-er Zhang: Molecular mechanisms of liver-specific albumin
and a-fetoprotein gene regulation: A review

27. Sawyer,R.H.: Avian scale development XV: A study of cell proliferation in the epidermis of the scaleless,
sc/sc, mutant

28. H.Asaka, T. Inoue, and K. Mikoshiba: Two dimentional polypeptide mapping of the cerebella from
neuropathological mutant mice, weaver, nervous and staggerer

29. DeHaan,R.L., S. Fujii and J. Satin: Cell interactions in cardiac development

Development, Growth and Differentiation (ISSN 0012-1592) is published bimonthly by The Japanese Society of
Developmental Biologists, Department of Developmental Biology, Mitsubishi Kasei Institute of life Science,
Minami-ootani 11, Machida, Tokyo 194, Japan. 1989: Volume 31. Annual subscription for Vol. 32, 1990: U. S.$
148,00, U.S. and Canada: U.S. $ 163,00, all other countries except Japan. All prices include postage, handling
and air speed delivery except Japan. Second class postage paid at Jamaica, N.Y. 11431, U.S.A.

Outside Japan: Send subscription orders and notices of change of address to Academic Press, Inc., Journal
Subscription Fulfillment Department, 1 East First Street, Duluth, MN 55802, U.S.A. Send notices of change of
address at least 6-8 weeks in advance. Please include both old and new addresses. U.S.A. POSTMASTER:
Send changes of address to Development, Growth and Differentiation, Academic Press. Inc., Journal Subscription
Fulfillment Department, 1 East First Street, Duluth, MN 55802, U.S. A.

In Japan: Send nonmember subscription orders and notices of change of address to Business Center for
Academic Societies Japan, 16-3, Hongo 6-chome, Bunkyo-ku, Tokyo 113, Japan. Send inquiries about
membership to Business Center for Academic Societies Japan,4—16, Yayoi 2-chome, Bunkyo-ku, Tokyo 113, Japan.

Air freight and mailing in the U.S. A. by Publications Expediting, Inc., 200 Meacham Avenue, Elmont, NY
11003, U.S.A.

NARISHIGE

THE ULTIMATE NAME IN MICROMANIPULATION

OUR NEW MODELS WR-88 and MO-102M
MAKE PRECISION MICROMANIPULATION SO EASY!

SOME FEATURES of THE WR-88 WATER ROBOT MICROMANIPULATOR (3-DIMENSIONAL)
* Drift-free, the new WR-88 has a DRIFT movement of less than 2 microns.
* The new WR-88 has a SMOOTH MICRODRIVE MECHANISM.
* An Aqua Purificate remote control ensures totally vibration-free operation.

(WARISHIGE) NARISHIGE SCIENTIFIC INSTRUMENT LAB.
AEE “oa OE Su gee

CABLE : NARISHIGE LABO, TELEX, NARISHIGE J27781

Sophisticated Balance between
Safety and Centrifugation
Capability without Compromise.

Centrifuge in
Integrated with A
Refrigerator

Extra-Quiet
Operation

Ease of Loading/
Unloading
The Rotors

Quick Start/
Quick Stop

High Quality

Triple Safety

Design
Corrosion
Resistance
‘ : HIGH SPEED
| REFRIGERATED
MICRO CENTRIFUGE
mMope. MR=-150
TOMY CORPORATION SOLE AGENT

1002 SOLEIL NARIMASU BLDG., 31-8. NARIMASU 1-CHOME,
ITABASHI-KU, TOKYO 175 JAPAN

TEL:(03)976-341 1 TLX: 02723111 TOMYCO J
CABLE: TOMYSHO TOKYO FAX:(GIII GII)(03)930-7010
TOMY SEIKO CO., LTD. MANUFACTURER

2-2-12, ASAHICHO NERIMA-KU. TOKYO 176 JAPAN
TEL:(03)976-3111

(Contents continued from back cover)

Ovarian development and sex steroid hor-
mones during the reproductive cycle of Rana
esculenta complex
Mugiya, Y.: Long-term effects of hypophys-
ectomy on the growth and calcification of
otoliths and scales in the goldfish, Carassius
auratus
Kobayashi, Y. and M. Okada: Urea stimula-
tion of pituitary pars intermedia cells of
suckling mice under copious drinking ....281

Morphology
Itow, T., T. Masuda and K. Sekiguchi: Forma-
tion of ganglions and stomodaeum in normal
and separate embryos of horseshoe crab,
Tachypleus tridentatus

Taxonomy
Wynn, S., M. J. Toda and T. X. Peng: The
genus Phorticella Duda (Diptera: Drosophi-
lidae) from Burma and southern China. . 297
Fukuda, Y.:
phology of the soldier crab, Mictyris brevi-
dactylus Stimpson (Crustacea: Brachyura:
Mictyridae)

Early larval and postlarval mor-

Okajima, S.: Some Nesothrips (Insecta, Thy-

sanoptera, Phlaeothripidae) from east Asia
Oc Ore et CRC ae ELEC Lee Ce te 311

Miura, T. and L. Laubier: Nautiliniellid

polychaetes collected from the Hatsushima
cold-seep site in Sagami Bay, with descrip-
tions of new genera and species

ZOOLOGICAL SCIENCE

VOLUME 7 NUMBER 2 APRIL 1990
CONTENTS
REVIEWS I I
Yasugi, S. and T. Mizuno: Mesenchymal- TE ie

epithelial interactions in the organogenesis

of digestive tract
Tiedemann, H.: Cellular and molecular

aspects of embryonic induction

ORIGINAL PAPERS

Physiology

Ip, Y. K., S. F. Chew and R. W. L. Lim:
Ammoniagenesis in the mudskipper,
Periophthalmus chrysospilos
Niida, A. and T. Ohono: Tectal visual affer-
ents from fish dorsolateral tegmental cells
(COMMUNICATION)
Mizutani, A. and Y. Toh: Morphological and
physiological characterization of the para-
ocellar nerve of the cockroach, Periplaneta
americana (COMMUNICATION)

Cell Biology
Fu, Y., S. Sato, K. Hosokawa and K. Shioka-
wa: Expression of circular plasmids which
contain bacterial chloramphenicol acetyl-
transferase gene connected to the promoter
of polypeptide IX gene of human adenovirus
type 12 in oocytes, eggs and embryos of
Xenopus laevis
Kusakabe, T.:
carotid labyrinth in the newt, Cynops pyr-
rhogaster

Ultrastructural studies of the

Genetics
Niwa, M. and N. Wakasugi:
development of preimplantation embryos
derived from intersubspecific hybrids be-

Abnormal

tween Mus musculus molossinus and M. m.
ee LRP Fhe loiain 3 209

domesticus

Saad, A. H. and E. Cooper: Evidence for a
Thy-1-like molecule expressed on earthworm
leucocytes

Developmental Biology
Sivasubramanian, P. and D. R. Nassel: Neu-
ral control of flight muscle differentiation in
the fly, Sarcophaga bullata
Tonegawa, Y., E. Hojiro and K. Takahashi:
Effect of pH on the participation of calcium
ion in the cell aggregation of sea urchin
embryos

Reproductive Biology

Tachi, C. and S. Tachi:
lying regulation of local immune responses in
the uterus during early gestation of eutherian
mammals. III. Possible functional dif-
macrophages cultured
together with blastocyst in vitro, with special

Mechanisms under-

ferentiation of

refrence to the cellular shape and production
of leukotriene C4

Endocrinology
Tasaki, Y. and S. Ishii: Effects of thyroxine
on locomotor activity and carbon dioxide
release in the toad, Bufo japonicus
Yamada, C., S. Noji, S. Shioda, Y. Nakai and
H. Kobayashi: Intragranular colocalization
of arginine vasopressin- and angiotensin II-
like immunoreactivity in the hypothalamo-
neurohypophysial system of the goldfish,
Carassius auratus
Polzonetti-Magni, A. M., R. Curini, O. Carne-
vali, C. Novara, M. Zerani and A. Gobbetti:

(Contents continued on inside back cover)

INDEXED IN:
Current Contents/LS and AB & ES,
Science Citation Index,
ISI Online Database,
CABS Database, INFOBIB

Issued on April 15
Printed by Daigaku Letterpress Co., Ltd.,
Hiroshima, Japan

————————

|» ZOOLOGICAL
SC] ane:

An International Jou

ZOOLOGICAL SCIENCE
The Official Journal of the Zoological Society of Japan

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

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

Chitaru Oguro (Toyama)

President: Hiromichi Morita (Fukuoka)
Assistant Editors:

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

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

Editorial Board:

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

Hiroshi Watanabe (Tokyo) Mayumi Yamada (Sapporo) Ryuzo Yanagimachi (Honolulu)

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

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

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

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

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

ZOOLOGICAL SCIENCE 7: 335-353 (1990)

© 1990 Zoological Society of Japan

REVIEW

Recent Studies of Fish Pancreatic Hormones: Selected Topics

ERIKA M. PLISETSKAYA

School of Fisheries, HF-15, University of Washingto
Seattle, Washington 98195, U.S.A.

Contents

Introduction

Immunocytochemical and structural studies of pancreatic hormones
Circulating levels of pancreatic hormones and their biological activity

Hormone-receptor interactions

Pancreatic hormones in fish under various physiological conditions

a) Smoltification
b) Spawning migration

c) Experimental fasting versus feeding in teleost fishes

Conclusion
References

INTRODUCTION

Interest in fish pancreatic hormones has grown
substantially during recent years. This interest
stems not only from the fact that Brockmann
bodies of fish provide a useful model for the
processing and secretion of peptide hormones, but
also from the conviction among biologists that fish
research promises both theoretical and practical
benefits [1].

Various aspects of progress in the field of fish
pancreatic hormones and their roles in regulation
of metabolism in cyclostomes and fishes have been
reviewed recently [2-7]. Epple and Brinn’s [8]
book “The Comparative Physiology of the Pan-
creatic Islets”, an encyclopedic source of informa-
tion, deals with most aspects of the structure of the
vertebrate pancreas and the functions of its active
peptides. Specific details of the amino acid se-
quences of pancreatic hormones and their bio-
synthetic pathways in fish, as compared to other
vertebrates, have been thoroughly covered by

Received January 22, 1990

Conlon [9]. Nevertheless, new information in this
field is accumulating so rapidly that it seems worth-
while at this time to summarize the latest trends
and findings in the studies of the fish endocrine
pancreas. Most references to the actions of
mammalian pancreatic hormones on fish were de-
liberately omitted because these topics have been
repeatedly and adequately discussed in the litera-
ture [3, 8, 10].

List of abbreviations used
aPY or YG-anglerfish pancreatic peptide Y
CCK-cholecystokinin
EGF-epidermal growth factor
ELISA-enzyme linked immunoassay for soluble anti-
gens
GLP-glucagon-like peptide
GLU-glucagon
GH-growth tiormone
IGF-1, IGF-2-insulin-like growth factors 1 and 2
INS-insulin
NPY-neuropeptide Y
PP-pancreatic polypeptide
RIA-radioimmunoassay
sPP-salmon pancreatic polypeptide
SST-somatostatin
YY-peptide YY

336 E. M. PLISETSKAYA

IMMUNOCYTOCHEMICAL AND
STRUCTURAL STUDIES OF
PANCREATIC HORMONES

Reports of basic immunocytochemical investiga-
tions, as well as correlative immunocytochemical
and electron microscopical studies, continue to be
numerous [11-20]. A typical pattern of recent
research of this type is the use of several mono- or
polyclonal antisera (instead of one), raised against
the same antigen, but recognizing different oligo-
peptide or polypeptide fragments of hormone
molecule.

Employing this approach McDonald et al. [21],
discovered that two peptides of the somatostatin
(SST) family, SST-14 and SST-28, which in the
anglerfish (Lophius americanus) are the products
of two separate genes (gene I and gene II), are
expressed in different types of pancreatic cells.
Moreover, the islet cells that process the product
of gene II, (so called SST-28-II) were localized in
close association with glucagon-immunopositive
(GLU-immunopositive) cells. This observation
was extended by Nozaki et al. [17] who found that,
in salmon and trout Brockmann bodies, the cells
producing SST-25-II were in close topographical
association with GLU-immunopositive cells, while
the cells producing SST-14-I, were located more
centrally, in association with insulin-immuno-
positive (INS-immunopositive) cells (Fig. 1). This
finding suggests that, in teleostean fish, such as
catfish ([ctalurus sp.), eel (Anguilla anguilla), scul-
pin (Cottus scorpius) and probably many others,
which, in contrast to mammals, possess two sepa-
rate sets of genes for SSTs [3, 4, 9, 22], these SSTs
will also be found in different types of cells.
Indeed, Abad et al. [13] reported recently that, in
the Brockmann body of gilthead sea bream (Sparus
auratus), from which the gene II SST has not been
yet isolated, immunostaining with antibodies
against SST-25-II from salmon and_ against
mammalian SST-14-I, follows a pattern similar to
that in the anglerfish, salmon and trout. These
observations lead, naturally, to several questions
that need to be addressed. First, why is there such
a specific separation of anatomical sites of produc-
tion of the two SSTs, since some other peptides,
even if they belong to different families, such as

GLU and pancreatic polypeptide (PP), often co-
exist in the same cells and, moreover, in the same
secretory granules [reviewed in 12, 18]? Second,
does the close topographical association between
cells that produce SST-25-II with GLU cells and
between cells that produce SST-14-I with INS
cells, imply an as yet unknown functional rela-
tionship between these pancreatic peptides in fish?
Yet another enigma to be resolved is the distribu-
tion in the brain of some “big” SSTs, which are
either the products of gene II (e.g., in salmonids,
catfish, sculpin, anglerfish and eel), or truncated
products of gene I [lamprey, 23; hagfish, 24].
Morel et al. [25], came to the conclusion that
processing of two distinct precursors of SSTs in the
teleostean fish operates “in a fixed pattern rather
than in a tissue-specific manner”, however, they
found that in the pancreas of anglerfish the product
of gene II (SST-28-II) prevails while in the brain
the level of SST-28-II is very low. In the gut cells
only traces of SST-28-II could be detected. Nozaki
et al. [17] failed to find any cells that were immuno-
positive for SST-25-II in the neurohypophysis or
hypothalamus of either Pacific salmon or trout,
while cells positive for SST-14-I were abundant. It
is noteworthy that Marchant ef al. [26] and Mar-
chant and Peter [27] found that neither catfish
SST-22-II, nor salmon SST-25-II inhibited the re-
lease of growth hormone (GH), while SST-14-I
retained its full inhibitory potency. Therefore, the
abbreviation SRIF (somatotropin release inhibit-
ing factor) does not seem to be applicable to the
SSTs encoded by the gene II family. SST-14-I was
the only SST that has been found in the pancreas
and in the gut of cartilaginous fishes [28]. In these
fishes, the distribution of SST and GLU-
immunoreactivities suggests a possible regulatory
role of both peptides in gastric secretion and/or cell
proliferation. | Moreover, paracrine interrela-
tionships between GLU and SST-14-I have been
suggested both in the pancreas and in the gut [29,
30].

In dogs, only SST-28-I (the amino-terminal ex-
tension of SST-14-I) from the stomach and intes-
tine seems to respond to physiological stimuli. The
pancreatic SST-14-I, by contrast, is believed to
have mainly a nonhormonal, local paracrine func-
tion [31-35]. What then is the situation in lam-

Fish Pancreatic Hormones 337

d

Fic. 1. Four successive sections of a rainbow trout pancreas stained differentially with antibodies against salmon
SST-25-II, preabsorbed with SST-14-I (a); antibodies against mammalian-type SST-14-I (b); antibodies against
salmon insulin (c) and antibodies against salmon glucagon (d). Note the topographical differences in the
distribution of cells that contain SST-25-II-like (a) and SST-14-I-like (b) immunoreactivities. Note also
topographical associations between the cells containing sSST-25-I-like and glucagon-like immunoreactivities (a,
d) and SST-14-I-like and insulin-like immunoreactivites (b, c) respectively. From reference [17] (Gen. Comp.
Endocrinol., with permission).

338 E. M. PLISETSKAYA

preys and teleost fish, in which the pancreatic cells
that express gene I SST-34 (lamprey) or one of
gene II SSTs (teleosts) are either present in equal
numbers or are more abundant than any other
types of cells, and “big” SSTs are the major
peptides processed in the islet organ [16, 17, 23,
36, 37]? Is the role of these pancreatic SSTs
confined to paracrine effects on adjacent GLU/
GLP cells (in teleost fish) or upon INS-secreting
cells (in lampreys)? Do these SSTs also have
metabolic or still other potencies, as has been
reported by Sheridan et al. [38]? Is there any
functional difference between SSTs of the gut and
those of pancreatic origin in lampreys, which have
abundant SST-immunopositive cells located both
in the gut and in the pancreatic islets [16, 36]?

In 1988 much progress was made in the elucida-
tion of the primary structures of pancreatic hor-
mones of agnathans (hagfish and lamprey), the
only two current representatives of the most primi-
tive vertebrates. Pancreatic SSTs of hagfish
(Myxine glutinosa) and lamprey (Petromyzon
marinus) were isolated and their amino acids sequ-
enced [23, 24]. SSTs of lamprey are peptides of
34-37 amino acids with SST-34-I, as the predomi-
nant form. Hagfish SST is also a peptide of 34
amino acids and it is strikingly similar to lamprey
SST at the carboxyl end, where 17 of the amino
acids are identical. By contrast, 16 other amino
acids in the hagfish and lamprey SST-34 are com-
pletely different, and only 2 of them are in identic-
al positions [23] (Fig. 2). Agnathan SSTs are the
result of significant differences in the processing of
proSSTs (precursors to SSTs) as compared to the
processing of SSTs in either teleost fish or mam-
mals: in both hagfish and lamprey, a series of

Somatostatins

5) 10 15

amino-terminally truncated peptides is processed
proteolytically at a single Arg residue, as well as at
adjacent basic residues [23, 24].

Lamprey INS, also isolated in 1988 [39], differs
from both teleostean and mammalian INSs to the
same extent as the latter differ from one another.
For example, lamprey INS has 14 amino acid
substitutions at the variant positions when com-
pared to porcine INS, and the same number of
substitutions when compared to salmon INS. The
primary structures of agnathan insulins seem to
confirm that hagfishes and lampreys, have fol-
lowed markedly independent routes of evolution
[40]. Lamprey (Petromyzon marinus), as com-
pared to hagfish (Myxine glutinosa), has 17 sub-
stitutions among 52 amino acids in the A- and
B-chains of INS [39, 41]. These differences con-
trast with the known similarities in the structures
of INS from related species of fish. For example,
three species of Pacific salmon (Oncorhynchus
kisutch, O. keta and O. gorbuscha) and three
species of holocephalan fish, belonging to each of
three existing families, namely, Hydrolagus colliei,
Chimaera monstrosa and Callorhynchus milii, are
96-100% identical in therms of their INS struc-
tures [42-48]. Even more striking is the result,
that the ray (Torpedo marmorata) and the shark
(Squalus acanthias), despite the divergence in their
evolution about 200 million years ago, still retain
more than 90% homology in their INS structures
[49, 50]. It would be worthwhile to determine
whether the INS structures of various species of
lamprey (or hagfish) show as much similarity with-
in the respective groups as do the structures of
salmonid or holocephalan insulins.

The amino acid sequences of several C-peptides

20 25 30 35

Lamprey ALRIAAAVAGSPQQLLPLGQRERKIAGCKNEFWKTFSSC

Hagfish

Fic. 2.

-VERPRQDG-VHEPPG----|----------- T- -

Comparison of amino acid sequence of somatostatins from the lamprey (Petromyzon marinus)

and the hagfish (Myxine glutinosa). From references [23 and 24]. The vertical lines indicate putative
sites of processing for production of SST-14-I and SST-34-I.

Fish Pancreatic Hormones 339

from fish have been deduced from nucleotide
sequences of clones of cDNA for preproinsulin
[reviewed in 51]. However, only one C-peptide of
fish (European eel, Anguilla anguilla) has actually
been isolated [51]. A comparison of its structure
with the predicted structures of C-peptides from
hagfish, ray, anglerfish, salmon and carp has re-
vealed that, unlike the insulins, the structural
similarity among C-peptides is weak, with the
exception of several amino acids in the central
region of the polypeptide chain.

Information of considerable interest has
appeared concerning another family of pancreatic
hormones, namely glucagon (GLU) and its related
peptides. This information followed the discovery
of the so-called glucagon-like peptides (GLPs), the
sequences of 31-34 amino acids located at the
carboxyl end of the preproglucagon molecule. In
contrast to the mammalian species, in which two
GLPs, organized in tandem, are encoded in the
same preproglucagon sequence [52], only one GLP
has been isolated from teleostean fishes
[anglerfish, catfish, salmon and sculpin, 3]. The
same is true for the primitive holostean garfish,
Lepisosteus spatula [53], and for a holocephalan
fish, Hydrolagus colliei [54]. To date no GLP has
been found in agnathans. However, the above
mentioned studies do not exclude the possibility
that piscine proglucagons may still contain more
than one GLP sequence. Thus far, both GLPs
(GLP-1 and GLP-2) have been found only at the
evolutionary stage of amphibia: two GLPs are
expressed, in the endocrine pancreas of the bull-
frog, Rana catesbeiana just as they are in mammals
[55].

Multiple, often truncated, molecular forms of
SST, INS, GLU and GLP, each encoded in the
same preprohormone, seem to be the rule rather
than the exception in fish. Recent examples have
been provided by Andrews et al. [23] and Conlon
et al. [24, 54] who found multiple molecular forms
of SST in the lamprey, and multiple forms of INS
and GLP in the ratfish. Each of these groups of
peptides probably contains the products of the
same gene. By contrast, salmon may contain two
preproinsulin genes that encode for two different
preproinsulins, one of which is present at much
higher levels [42] than the other [48, 56]. Two

different insulins were discovered about thirty
years ago in bonito fish [57] and quite recently in
an amphibia, Xenopus laevis [58, 59].

The pancreatic polypeptides (PP) and their ex-
pression in fish have also been the focus of substan-
tial attention during the recent years. Anglerfish
(Lophius americanus) PP, named YG (glycine-
extended form) or aPY, resembles neuropeptide Y
(NPY) from porcine brain and the peptide YY
from intestine more closely than it resembles any
mammalian PP [60]. The same is true for PP from
salmon [61], sculpin [22, 62] and garfish [63]. It is
remarkable that such similarities seem to be con-
fined to fish, while amphibian (bullfrog) PP is a
typical bird- or mammalian-type peptide, being
similar to the human PP sequence [55].

More detailed studies on anglerfish have re-
vealed that there is more than one molecular form
of NPY-like peptide in their islet organ. The
majority of NPY-like peptides appear to be the
YG-peptide that is expressed in a subset of islet
cells, while the minor form of NPY-like peptide,
closely resembling porcine or human NPY, is
localized in the neurons of anglerfish brain and in
islet nerves [64, 65]. It has been suggested that
peptide YG is a precursor of biologically active
peptide [Des*’-Gly]-aP Y-amide [65a].

As far as we know, there are no reports of the
identification of a novel peptide, pancreastatin, in
fish endocrine islets. This peptide of 49 amino
acids was recently purified from extracts of the
porcine pancreas by Tatemoto et al. [66] and was
demonstrated to be important in the regulation of
pancreatic exocrine and endocrine secretions in
mammals [67, 68].

CIRCULATING LEVELS OF PANCREATIC
HORMONES AND THEIR BIOLOGICAL
ACTIVITIES

The major technique used for the measurement
of circulating levels of pancreatic hormones in fish
has been the radioimmunoassay (RIA), although it
is now evident that the enzyme-linked immuno-
sorbent assay (ELISA) should be considered
seriously as a future substitute for RIAs. It was
anticipated in 1979 that non-radiometric, ELISA
methods could be developed that would be as

340

sensitive as or even more sensitive than existing
RIAs [69, 70]. Such methods are less hazardous
since they do not involve the use of radiolabeled
hormones and they eliminate the problem of
radioactive-waste disposal.

Assays for specific measurements of pancreatic
hormones in fish systems are still not common.
However, assessment of INS by RIAs using piscine
components are already performed in scientific
laboratories in Canada, Israel, Japan, Norway,
Spain, the United Kingdom, the USA and the
USSR. The main obstacle for much of the poten-
tially important fish-related research remains the
lack of homologous fish hormones and antisera,
although heterologous antibodies raised against
insulins from scorpion fish, bonito, cod and
anglerfish and their respective ['*°I] derivatives as
tracers, have been used satisfactorily in RIAs of
insulins from other species of teleosts [71-78]. The
need to develop more assays for teleost INS has
increased since the initiation of projects directed
towards the transplantation of the Brockmann
bodies of fish into diabetic mammals [79] which
made necessary measurements of the hormones
released from the transplants.

A fully homologous RIA for fish (salmon) INS
[80] has been used extensively. The results of
assays of plasma INS in juvenile salmonids, in a
wild population of pink salmon (O. gorbuscha)
during their spawning migration, and in domesti-
cated fish starved or fed specially designed diets,
were reported recently [3—-5, 80-86]. These results
are described in more detail below.

As is now the case in mammalian studies, the
regulatory effects of the novel peptides galanin and
pancreastatin [66, 87] on the secretion of INS in
fish and the mechanisms of their actions will prob-
ably become the focus of numerous studies as soon
as these peptides are isolated from fish gut and
pancreas.

Another breakthrough can be expected in the
measurement of circulating levels of the second
most abundant fish pancreatic hormone, a gene II
SST. A fully homologous assay system for coho
salmon (O. kisutch) SST-25-II, which has proved
to be suitable for measurements in a variety of
other salmonid species, was recently developed by
Sheridan et al. [88]. Since the structure of second

E. M. PLISETSKAYA

SST from fish, SST-14-I, is identical in fish and
homeothermal vertebrates, the mammalian RIA
systems should suffice for the measurement of this
peptide in fish.

Unlike INS and SST-25-II, circulating levels of
GLU can be assayed by mammalian RIA [77],
although such assays are still rarely used. Only two
groups of researchers have employed either catfish
[89] or salmon [90] homologous RIA systems to
measure GLU in the Brockmann body and plasma
of the respective species of fish.

RIAs for mammalian glucagon-like peptide
(GLP) have been described by @rskov and Holst
[91] and Oshima et al. [92] but we not know of any
application of these RIAs to fish. It seems that
antibodies raised against piscine GLP do not cross-
react with mammalian GLP (Plisetskaya, unpubl.).
The only published data on levels of circulating
GLP in fish have been obtained by homologous
salmon RIAs [85, 86, 90, 93]. Under non of the
experimental conditions studied were the titres of
GLU and GLP in plasma of salmonids higher than
the INS titres [90]. Plasma levels of GLP were
usually higher than plasma levels of GLU. The
same pattern was observed after extraction of the
principal islets of the fish [94, 95]. Several hypoth-
eses have been suggested to explain the discrepan-
cy in the yields of the two peptides that are part of
the same prohormone. None has been proven.
However, the differences in circulating levels of
GLU and GLP have, seemingly, been provided
with a logical explanation: Oshima ef al. [92]
reported that, in mammals, GLP in vivo is de-
graded more slowly than GLU. Our preliminary
results from studies of incubation of isolated salm-
on hepatocytes in the presence of salmon GLP and
GLU (Mommsen and Plisetskaya, unpubl.) sug-
gest the same trend in fish.

While the data concerning the biological activi-
ties of INS and SST in fish continue to accumulate
steadily [reviewed in 90, 96, 97], explorations into
the role of GLU and, in particular, GLP in both
mammals and fish made the very rapid progress
during the last three years [90, 93, 96, 98-100].
The most unexpected finding was the apparently
strong glycogenolytic, gluconeogenic and lipolytic
effects of teleost GLP in fish [90, 101], while all
attempts of find similar effects in the mammalian

Fish Pancreatic Hormones 341

liver have failed [102-105]. It was even suggested
that GLP, although a member of the GLU-family,
has no metabolic activity [102]. Part of the solu-
tion to this problem may have been found recently,
when several research groups [106-108] reported
simultaneously that the biologically active form of
mammalian GLP-1 consists, not of 37 amino acids,
but of 31 (sequence 7-37) which correspond to the
amino acid sequence of GLP from salmon and
anglerfish [94, 95].

The biological activities of GLP 7-37 (or GLP
7-36-amide) in mammals were tested primarily to
determine the relation of these peptide to other
pancreatic hormones, and GLP emerged as the
most potent stimulator of the release of INS [106-
110]. In addition GLP 7-36-amide enhances the
release of SST and inhibits the release of GLU
[110-111]. To reveal any insulinotropic effect of
salmon GLP on perifused Brockmann bodies from
fish of the same species, this peptide should be
applied at concentrations at least 100-fold higher
than those reported for mammals [97]. In experi-
ments in vivo, the insulinotropic action of the GLP
is barely detectable [90]. Teleost fishes still remain
the only vertebrate group in which glycolytic and
gluconeogenic fluxes are influenced by GLP [100].
By contrast, a fragment of GLU (GLU 19-29),
which exhibits a potent metabolic action that is
mediated by calcium ions in the mammalian liver
[99, 112], seems to be without effect in piscine
liver, supposedly because of the absence of analo-
gous calcium-dependent systems [100].

Although GLPs from salmon, anglerfish and
catfish all activate the production of glucose in
teleosts, notable differences exist between fish
species. Moreover, the actions of GLP, as is also
true for GLU, are strongly dependent on the
season and, probably, on the stage of the fish life
cycle [100]. Both GLU and GLP seem to affect
identical targets in the liver. However, the
mechanisms of their action may differ: Mommsen
and Moon [100] reported that no direct rela-
tionship exists between the amount of intracellular
cAMP and either the metabolic action of GLU, or,
in particular, of GLP. (Fig. 3). Once again, the
differences between fish species are substantial
[93, 101].

In this field of research more questions remain

(%) %
A A Goes B Yoo Gtu 100 nM
20005 4
3007 a GLU 100 nM / os
Aa
4
ae Ere GLP
1004 a a OOS SS i
SS tS
0 min 30 Co) min 30
Fic. 3. Time course of intrahepatic accumulation of

cAMP after application of GLU and GLP to hepato-
cytes from (A) trout (O. mykiss) and (B) eel
(Anguilla rostrata). Values are expressed in terms of
the percentage increase over vehicle-treated con-
trols. Control levels of cAMP in salmon and eel,
respectively, were 386 and 533 pmoles/g fresh weight
of cells. Both bovine glucagon and salmon GLP
were applied at concentrations of 10’ M.

A glucagon; 4 glucagon-like peptide. From refer-
ence [100] (Fish. Biochem. Physiol., with permis-
sion).

to be addressed than have been answered. The
first among them is: why does the metabolic action
of GLP seem to be confined to fish? Does this
unique property of GLP have any special physiolog-
ical meaning? Can the preproglucagon be express-
ed in piscine organs, other than the pancreas and
gut, for example, in the brain, as it is in mammals
[113]?

Diurnal oscillations in piscine plasma levels of
INS have been reported [83, 114] but it remains to
be determined whether INS, GLU and SSTs in fish
are secreted steadily or in pulses, as they are in
mammals [115].

The study of the fourth group of fish pancreatic
hormones, the pancreatic polypeptides (PP) has
also progressed substantially during the recent
years, mostly due to efforts of B. D. Noe, P. C.
Andrews, A. Balasubramanian and their associ-
ates. Radioimmunoassays for anglerfish peptide
YG (glycin-extended form) and for NPY-amide
have been developed and aPY-like peptides in the
anglerfish brain and pancreas were characterized
[65, 115a]. The next logical step will probably be
an attempt to assess levels of these peptides in
plasma, followed by an evaluation of changes in
these levels under various physiological condi-
tions.

Anglerfish peptides belonging to the PP-family,

342 E. M. PLISETSKAYA

namely, YG and APY-NHb, and salmon PP (NPY-
NH)) have been synthetized and tested for their
biological activity in mammals. Bolus doses of
natural and synthetic salmon PP and synthetic
anglerfish aPY-NH)p increased the blood pressure
and decreased the heart rate in anesthetized rats in
a dose-dependent manner; aPY-NH> diminished
the volume and bicarbonate content of pancreatic
juice during secretin-stimulated, exocrine pancreat-
ic secretion in conscious dogs [116-117].

These results demonstrate that teleost PPs are
not only structurally similar to mammalian NPY
and PYY, but also that they can mimic NPY- and
PYY-like activities in mammals [116, 117]. Of
additional interest are the pilot data (Balasubra-
maniam, personal communication) from the direct
injection of fish PP into the rat hypothalamus.
This treatment enhanced the feeding behavior of
the experimental animals, as does mammalian
NPY [118, 119]. Consequently, we can hypothe-
size that PP plays a similar role in teleost fish
during naturally occuring periods of either fasting
or restoration of feeding. Moreover, these data
appear to support the hypothesis [61] that a very
low number of PP (NPY) cells and a low content of
the peptide in the Brockmann body of spawning
coho salmon are the correlates of a particular
period in the life cycle, when the fish is naturally
fasting.

In comparison to efforts in previous years, less
time is now being devoted to experiments in vitro
and in situ on fish Brockmann bodies and/or prin-
cipal islets. Ronner and Scarpa [89], using their
experimental model, which consists of a perifused
isolated Brockmann body from the channel catfish
([ctalurus punctatus), reported a striking similarity
in the responses of the catfish and higher verte-
brates to hexoses. However, the catfish Brock-
mann body seemed to be sensitive to fewer of the
common stimuli for the release of pancreatic hor-
mones than is the mammalian pancreas.

One observation which clearly deserves more
attention is that the D-cells of the catfish Brock-
mann body, which produce SST-14-I, are more
sensitive to glucose (but not to amino acids) than
INS cells [89]. This finding was confirmed by
Sheridan et al. [88] in experiments in vivo. We can
speculate that this particular pattern may be re-

sponsible for the comparatively low, and some-
times delayed, glucose-stimulated release of INS in
teleost fish, as compared to the amino acid-
stimulated release of INS. If, in addition, it is
confirmed, in the future, that GLP in fish is not as
potent a stimulator of the secretion of INS as it is
in mammals [5] we will be closer to an understand-
ing of the intolerance to carbohydrates of
apparently INS-nondeficient fish.

No reports of studies in vitro of agnathans’
pancreatic islets have appeared during the past few
years.

The average plasma levels of INS, SST, GLU
and GLP in the peripheral blood of teleosts are
usually higher than those in mammals (cf., for
example, Table 1 and 4). It is known, however,
that the high levels of immunoreactive INS,
observed in mammals under certain conditions,
such as in cases of INS-resistant diabetes mellitus,
are in fact caused by an increase in levels of
proinsulin in the immunoreactive pool, measured
as INS [120]. Conlon and Thim [51] recently
isolated the first teleostean (eel) C-peptide of
proinsulin. It is to be hoped that this accomplish-
ment will provide researchers with a tool for the
assessment of levels of proinsulin in fish, so that
the question of the contribution by proinsulin to
plasma levels of INS in fish and to binding to
receptors can finally be addressed.

Our knowledge of plasma levels of pancreatic
hormones in fish has resulted in a clear change in
the attitudes of comparative physiologists and
biochemists: most of them are turning now to
usage of “physiological doses” of hormones for
experiments both in vivo and in vitro. However,
the question remains as to whether the levels of
hormones that are routinely measured in the
peripheral blood correspond to those that the liver
cells actually “see”.

As in mammals, the fish liver is the major organ
for glucose homeostasis, the primary target for
pancreatic peptides and an important site for their
degradation [121]. Consequently, it is logical to
expect that in fish, as in mammals, the liver is
exposed, through the hepatic portal vein, to much
higher levels of pancreatic peptides than is any
other organ. However, untill 1989 no actual values
for levels of pancreatic peptides circulating in

Fish Pancreatic Hormones 343

TaBLE 1. Titres of some pancreatic hormones in the peripheral blood of various salmonids
(ng/ml)*

Species Insulin Glucagon GLP SST-25-II References
O. kisutch 0.9-15.0 0.01—2.00 0.10-2.30 0.15-1.20 INS, GLU, GLP
O. tshawytscha 1.5-9.1 0.05—0.06 0.20—2.80 0.15-1.20 [3, 4, 80, 83]
O. gorbuscha 0.2-3.0 [85, 86, 90, 17]
(wild population) SST-25-IT
O. mykiss 1.7—48.0 0.01-2.00 0.10-2.10 0.15—-1.20 [88]

S. salar 2.0—40.0 0.01-1.60 0.05-1.90
S. gairdneri** 0.1-4.2

(steelhead,
wild population)

* The values represent a summary of results of many different assays of fish of a particular species (60-200 fish
per group), but of different age, sex and feeding conditions, maintained on various diets and at various water

temperatures.

For variations in levels of hormones in uniform groups of fish see the original publications.

** New species name for Salmo gairdneri is Oncorhynchus mykiss.

various blood vessels of the same fish could be
found in the literature. Such values were obtained
from an adult trout and published recently [122].
As could have been expected, levels of INS, GLU
and GLP in peripheral blood constituted only.
20-30% of those in the hepatic portal vein (Table
1). There is no doubt that differential blood
sampling from various vessels, if continued in
studies on fish, will provide us with new insights
into the uptake and processing of biologically
active peptides by their target tissues.

Precise information about the concentrations of
pancreatic hormones at the “entrance” to and
“exit” from the liver will be of benefit to several
groups of researchers who are experimenting with
slices of fish liver or isolated hepatocytes [93, 100,
121, 123]. These studies have become even more
challenging since the publication of the new con-
cepts that have been introduced into investigations
on mammalian liver. The idea [124] that the liver
acinus is a unit of hepatic microcirculation seems
to be generally accepted by mammalian physiolog-
ists [125]. Three distinct metabolic zones in liver
acini, namely, the periportal, perivenous (or
pericentral) and intermediate zones have been
proposed. Morphological, histochemical and
biochemical differences between these zones [125,
126] and some differences in the hormonal regula-
tion of gluconeogenesis and ketogenesis in peri-
portal and perivenous rat hepatocytes have been

reported [127]. Methods have been developed for
the separation of so-called periportal and
perivenous liver cells from mammals [128-131].
Whether this idea of metabolic zonation can be
applied to the piscine liver, microstructure of
which differs in many ways from that of the
mammalian liver [132-134], remains controversial.
The first studies on two populations of hepatocytes
separated from the same liver of trout and catfish
by pulses of digitonin that were followed by diges-
tion with collagenase and Percoll-gradient centri-
fugation, were undertaken by Mommsen et al.
[135] and Ottolenghi et al. [136]. Although some
metabolic differences between “periportal” and
“perivenous” pols of cells were found, both studies
failed to reveal either a real “metabolic zonation”
or different responsiveness to glucagon, in terms of
carbohydrate fluxes, in these cells. Improved
methods for the separation of cells may be needed
for future research in this field.

HORMONE-RECEPTOR INTERACTIONS

In sharp contrast to extensive investigations in
mammalian systems the receptors for pancreatic
hormones in fish tissues remain almost completely
unexplored [2], although the cyclostome and fish
hormones, in particular INS, have been tested for
their binding to mammalian plasma membranes.
These studies have concentrated mostly on evalua-

344 E. M. PLISETSKAYA

tions of the potency of newly isolated piscine
hormones, as compared to their mammalian
counterparts. Fish pancreatic peptides either do
not bind to mammalian receptors or have much
lower binding affinities than do the mammalian
peptides [42, 54, 105, 137-140]. The same pattern
(lower binding affinity as compared to mammalian
INS) has been found when agnathan or piscine
insulins are bound to the receptors in homologous
tissues [141-144]. Report on the structure of the
INS receptor in the plasma membranes of the
hagfish (Myxine glutinosa) liver [145] supports an
earlier conclusion that the receptor protein has
been much better conserved in the course of
vertebrate evolution than the ligand [141, 142,
146]. The partial amino acid sequence of the INS
receptor from coho salmon (S. Chan, personal
communication) lends further support to this
hypothesis. However, an INS receptor from the
stingray (Dasyathis americana) liver [147] seems to
display some peculiar features in its structure. It
has been reported to be a dimer, consisting of two
identical subunits, each with a binding (alpha) and
a tyrosine kinase (beta) domain. Stuart [147]
suggested that the stingray receptor is not com-
pletely cleaved. This stands in marked contrast to
all other (mammalian, avian, reptilian and amphi-
bian) receptors for INS, which are tetramers, built
from two extracellular alpha subunits and two
transmembrane/intracellular beta subunits. The
alpha subunits are connected to one another and to
the beta subunits by disulfide bonds [148-151].
The same description seems to be applicable to the
INS receptor from Drosophila melanogaster [152]
which has a subunit structure similar to that of the
mammalian INS receptor. By contrast, the pro-
posed structure of the stingray INS receptor [147]
more closely resembles that of receptors for IGF II
and for EGF.

Competitive binding of INS and IGF to the INS
receptor of the stingray led Stuart [147] to the
conclusion that, at the phylogenetic stage of the
elasmobranch fish, both INS and IGF may trans-
duce their signals through the same receptor.
Some overlap in structure, specificity and function
between receptors for IGF-I, IGF-II and INS has
been reported recently in the early chick embryo
and in the lizard brain [151, 153]. Nevertheless,

recent studies of Gutiérrez and Plisetskaya [154]
on liver plasma membranes from salmon demon-
strated binding with much higher affinity for INS
than for either mammalian IGF-I or IGF-II. Simi-
lar studies should be undertaken with a wider
range of lower vertebrates and lower chordates.
Such investigations appear especially relevant
since the report by Chan et al. [155] that the overall
organization of the preproinsulin gene of an
amphioxus (Branchiostoma lanceolatum) indicates
close relation to both INS and IGF-I.

It is widely accepted, that, in mammalian INS-
sensitive tissues, the numbers of receptors for INS
on the plasma membrane are regulated by the
circulating levels of INS [156-158], so that eleva-
tion in levels of INS causes a decline in the number
of receptors. This assumption has been extrapo-
lated to the properties of the INS receptors in fish.
Therefore, when Ablett et al. [159] found that
isolated hepatocyte from trout reared on a high-
carbohydrate diet had more specific binding sites
for INS than those from control fish, it was con-
cluded that the plasma INS levels in this fish were
low. Developing this idea still further, Christiansen
and Klungs@yr [160] suggested that, because of the
strictly reciprocal relationship between the concen-
tration of receptors and plasma levels of INS, the
assessment of receptor binding can provide an
estimate of INS levels in the course of the nutri-
tional studies on fish. Since the high-carbohydrate
diet resulted in an increase of the numbers of
specific binding sites for INS in trout liver [159], it
was concluded that glucose does not stimulate the
secretion of INS in the rainbow trout.

Such a conclusion from the abovementioned
experiment is difficult to accept. First of all,
glucose does indeed stimulate the secretion of INS
in agnathans and in teleost fishes [71, 143, 161],
including the rainbow trout [81], though it is not as
potent in this respect as amino acids. Secondly,
the question should be addressed as to whether the
reciprocity between plasma levels of INS and
number of receptors, as reported in mammals,
should be applied to fish. Two recent publications
[162, 163] provide some evidence that, at least in
the Baltic lamprey (Lampetra fluviatilis), neither
natural INS deficiency, caused by prolonged pre-
spawning anorexia, nor hyperinsulinemia induced

Fish Pancreatic Hormones 345

by injection of insulin, changes the binding para-
meters of INS in the liver, heart muscle or brain.
Moreover, the data obtained from mammals [164]
indicate that, at least, receptors for INS from
brains of adult animals are unaffected by altera-
tions in the levels of circulating INS. In primary
cultures of cortical cells from fetal mice, elevated
concentrations of INS in the media cause “up-
regulation” of INS receptors [165].

Leibush [144] commented that since the “down-
regulation” phenomenon is related, most prob-
ably, to the internalization of the receptors for
INS, a process that does not take place at low
temperatures, this phenomenon could hardly be
expected to occur in either agnathans or fish that
are maintained at low water temperatures. It is
likely that the problem is even more complicated,
since Gutierrez et al. [166] and Gutierrez and
Plisetskaya [154] found either the presence or the
absence of down-regulation, dependent not only
on environmental temperature but on some un-
identified physiological conditions of fish. Whatever
may be the mechanism that governs the number of
the receptors for INS in fish, it is clear that the time
has come to examine it in more detail. The same is
true for other pancreatic peptides, such as GLU
and GLP. While the receptors for GLU and the
transduction of signals from the cell surface to the
intracellular targets have been analyzed in great
detail in mammals [99, 112, 167, 168] no similar
studies have been undertaken in fish. However,
some indirect information about receptors and
post-receptor events is available. Clear differences
are already apparent at the level of binding of
GLU to its receptor. For example, catfish gluca-
gon, although very close in terms of structure to
mammalian GLU, does not bind to GLU receptors
in the liver, pituitary or hypothalamus of the rat.
It does not activate adenylate cyclase in any of
these tissues, while its porcine counterpart strongly
activates this enzyme. By contrast, catfish GLP
does stimulate the activity of hypothalamic and
pituitary adenylate cyclase in rats [105]. Postre-
ceptor transduction of signals also seems to differ
between fish and mammals, especially in regard to
the stimulation of adenylyl cyclase activity, which
can be achieved only after treatment with pharma-
cological concentrations of GLU (see above and

Fig. 3).

As far as we know, there is at present no
available information about hormone-receptor
interaction of peptides that belong to the gene II
SST- or to PP families in fish.

PANCREATIC HORMONES IN FISH UNDER
VARIOUS PHYSIOLOGICAL CONDITIONS

a) Smoltification

Smoltification or the parr-to smolt-transforma-
tion in salmonids involves profound morphological
and physiological changes. This crucial period
prepares juvenile salmon for downstream migra-
tion as well as for survival at sea. Smoltification
includes changes, mostly increases, in the titres of
many hormones, such as thyroxin, GH, prolactin,
catecholamines, sex steroids, cortisol and some
others [169-171]. It is surprising that pancreatic
hormones have received only scant attention in
this regard [169]. The first pancreatic hormone
measured during the entire period of the parr-to-
smolt transformation was INS [83]. In experiments
continued from 1984 to 1989, the plasma profiles of
this hormone revealed an annual peak at a specific
time, namely at the very beginning of the trans-
formation of parr to the transitional stage (Table
3). The annual peak was followed by a rapid or
gradual decline in levels of INS. It is noteworthy

TABLE 2. Circulating levels of insulin in plasma of
juvenile coho salmon, Oncorhynchus kisutch, in
1985; According to Plisetskaya et al. [83]

Date Insulin (ng/ml) Stage
February 1 2.0+0.2 (9)* Parr

February 15 1.1+0.1 Q) Parr

March 15 1.0+0.1 (8) Parr

March 29 7.7£1.7 (6) Parr-transitional
April 6 1.9+0.4 (7) Parr-transitional
April 12 0.8+0.1 (7) Transitional-smolt
May 3 0.8+0.1 (8) Smolt

May 17 0.6+0.1 (7) Smolt

May 24 1.1+0.2 (7) Smolt

June 21 2.4+1.0 (8) Smolt transferred

to seawater

* In parentheses: number of samples.

346

TABLE 3.

E. M. PLISETSKAYA

Circulating levels of glucagon and glucagon-like peptide in plasma of juvenile

coho salmon, Oncorhynchus kisutch, in 1989

Date NaC ae ee Stage
February 22 0.40+0.08 (4)* 1.60+0.20 (4) Parr
March 8 1.00+0.04 (4) 1.10+0.04 (4) Parr
March 21 0.50+0.06 (7) 1.00+0.10 (7) Parr-transitional**
April 3 0.60+0.06 (8) 1.70+0.26 (8) Parr-transitional
April 17 0.50+0.02 (5) 0.80+0.17 (5) Smolt
April 26 0.70+0.06 (4) 2.00+0.32 (4) Smolt
May 5 2.30+0.40 (6) 1.10+0.18 (6) Smolt

* Each sample of plasma represents a pool from 2-3 fish.
** Insulin levels rise to a maximum of 7-10 ng/ml; data for insulin are not shown but the profiles

are similar to those presented in Table 2.

TABLE 4. Titres of pancreatic hormones (ng/ml+S.E.M.) in salmonids that were either fed or fasted

Hormone Fed Fasted Species Details Reference”
Insulin 6.8+0.60 (10)' 1.6 +0.30 (10) S. gairdneri Fasted 1 week [73*]
4.2+0.30 (10) 2.2 +0.30 (10) Fasted 1 week
4.5+0.80 (10) 1.4 +0.20 (10) O. kisutch Fasted 1 week [80**]
4.3+0.80 (9) 0.9 +£0.10 (9) Fasted 2 weeks
12.1+1.10 (10) 2.0 +0.10 (10) O. mykiss 1986 __Fasted 6 weeks [85**]
13.0+1.70 (15) 2.2 +0.10 (15) O. mykiss 1988 Fasted 6 weeks [86**]
10.9+0.70 (30) 3.0 +0.10 (26) O. mykiss 1989 Fasted 6 weeks [191**]
Glucagon 0.80+0.10 (10) 0.20+0.10 (10) O. mykiss Fasted 6 weeks [85**]
0.12+0.05 (15) 0.08+0.03 (15) O. mykiss Fasted 6 weeks [86**]
1.60+0.10 (29) 1.20+0.20 (25) O. mykiss Fasted 6 weeks _— Plisetskaya,
(unpublished) **
GLP 0.60+0.10 (10) 0.30+0.04 (10) O. mykiss Fasted 6 weeks [85**]
1.10+0.04 (10) 0.80+0.01 (10) O. mykiss Fasted 6 weeks [86**]
1.90+0.30 (28) 0.40+0.02 (25) O. mykiss Fasted 6 weeks _ Plisetskaya
(unpublished) **

' In parentheses: numbers of fish assayed. * Radioimmunoassay with * cod

components

that neither the maximum number of the receptors
for INS in the liver nor the maximum binding of
INS to the liver plasma membrane, were coinci-
dent with this surge in levels of INS [154]. The
titres of members of the glucagon family peptides,
namely GLU and GLP, assessed during smoltifica-
tion in 1989, fluctuated without any particular
periods of significant elevation or decline (Table
4).

The surge of levels of INS seems to be consistent
with a switch in metabolic conditions, as the clearly
anabolic parrs become catabolic smolts [172, 173].

**

components; coho salmon

Changes in the secretion of INS evidently play a
major role in these metabolic shifts, as was demon-
strated in model experiments that included either
inactivation of INS in parr by injection of INS-
specific anti-serum or administration of INS to
smolts. After these experimental treatments, parrs
acquired some metabolic features of smolts and
vice versa.

Within several weeks after entering the sea
water, smolts of Pacific salmon usually regain high
plasma levels of INS which are favourable for a
rebuilding of stores of lipid and glycogen. In

Fish Pancreatic Hormones 347

juvenile fish that reach the sea prematurely and
cease their normal growth (“stunting” phe-
nomenon), levels of INS remain low [83, 169].

b) Spawning migration

Profiles of plasma INS were assessed in wild
population of Pacific salmon (O. gorbuscha) along
their way to the spawning grounds [82, 174] and in
reproductively maturing Atlantic salmon, S. salar
[175]. Fish of both species, although anorexic
during the spawning period, maintain relatively
high levels of plasma INS. Metabolic studies on
the liver cells isolated from anorexic O. keta, at
four sampling sites along their 1150-km migration
route, led French et al. [176] to the conclusion that
spawning of sockeye salmon is apparently sup-
ported by the catabolism of carbohydrates. The
final depletion of carbohydrate reserves from both
muscles and liver of Pacific salmon occurs during
or immediately after spawning. In maturing salm-
on, INS may participate in the preservation of
these carbohydrate reserves, accumulated mostly
via gluconeogenesis, from premature exhaustion.
It is remarkable, that relatively high levels of INS
during spawning migration in anorexic fish coin-
cide with enhanced gluconeogenesis while, in nor-
mally feeding fish, INS is believed to act to reduce
gluconeogenesis [177]. Therefore, the interaction
between INS and gluconeogenic fluxes in anorexic
fish should be reevaluated and studied in greater
detail.

In Atlantic salmon, the elevation in levels of
INS, GLU and GLP is observed in spring prior to
the onset of maturation, when the fish continue to
feed extensively [175, 178]. Similar changes have
previously been described for the scorpion fish,
Scorpaena porcus [161], and for the sea bass,
Dicentrarchus labrax [179]. Although some in-
volvement of INS in the regulation of the uptake of
vitellogenin has been reported [180] and binding
sites for INS have been found on the plasma
membrane of the ovaries of S. salar (Gutiérrez,
personal communication), details of the role of
INS in the maturation process await further inves-
tigation. Another possibility that deserves to be
explored is the direct transfer of INS from the
blood of females into the eggs prior to spawning.

The spring elevation in plasma levles of INS in

Atlantic salmon seems to coincide with an increase
in levels of GLU and GLP, an observation that
again contradicts the relationship between these
hormones in mammals. The surge in levels of INS
in the spring may enable the maturing fish to
increase their stores of protein, lipids and carbo-
hydrates in somatic tissues, in anticipation of
maturation. It is possible that, when these meta-
bolic stores reach some particular level, they may
signal the readiness of fish for sexual maturation
and for the switch from somatic to gonadal growth.
After feeding has ceased, the gonads, and in
particular the ovaries, continue to incorporate
proteins and lipids stored in the somatic tissues
[175].

In both Pacific and Atlantic salmon at the time
of spawning, males have higher plasma levles of
INS than do females [82, 178]. This difference may
be caused by the necessity of meeting an increased
demand for energy, since the males arrive at the
spawning grounds before the females so that they
can defend their territory, and the males remain
there for a longer time to engage in multiple
spawnings [175]. An alternative explanation for
the difference in INS levels is that, in females,
some INS in transferred from the circulation to the
eggs.

Is the hormonal regulation of metabolism in
fasting fish similar (albeit slow because of their
natural heterothermic conditions) to the regulation
of metabolic fluxes in other vertebrates? Such a
comparison is difficult to make because the fish
that have been studied are mostly carnivorous,
while common laboratory mammals and birds are
omnivorous. Fortunately, in this regard, a group
of French scientists has conducted a thorough
study of the physiology and biochemistry of the
long-term, natural fasting in penguins. Their study
included measurements of metabolic indices and
the regulation of these indices by pancreatic hor-
mones [181-183]. The experimental subjects,
chicks of the king penguin (Aptenodytes patagoni-
ca), experience a natural 4-to-6 month fast during
the subantarctic winter. These birds are strictly
carnivorous. Therefore, the results accumulated
by the French group should be suitable for com-
parisons with the data obtained from fish. The
researches distinguish two phases of the natural

348 E. M. PLISETSKAYA

fast: the first is accompanied by the early metabolic
adjustments that are characterized by a decline in
the plasma levels of hormones (including INS) and
a decline in the utilization of energy sources.
During the next, long-term phase of the fast, the
levels of INS and GLU remain stable [182, 183], a
situation that resemles conditions in migrating fish.

c) Experimental fasting versus feeding in teleost
fishes

Because of their remarkable ability to tolerate
long non-feeding periods, fish are becoming valu-
able experimental models for studies of the meta-
bolic patterns of fasting conditions. The metabolic
strategies of previously actively feeding fish, which
are then experimentally deprived of food, seem to
differ from those described above for upstream-
migrating anorexic fish. At the same time they
bear some similarity to the early period of fasting
in penguins [182].

The elevation of plasma levels of GLU observed
in the sea bass by Gutierrez et al. [184] and in the
juvenile Atlantic salmon by Sundby (personal
communication), is not easily detectable. Even if
such an elevation occurs it lasts for only a short
period of time (usually during fourth and fifth days
after withdrawal of food). Moreover, in contrast
with the situation in mammals [85], in fish plasma
the molar ratios of GLU to INS never reach, not to
mention never exceed, 1.0, with levels of GLU
always remaining lower than those of INS. After
this initial period, experimental fasting leads to a
decline in plasma levels of INS, GLU and GLP
(Table 4). This decline is, however, not uniform:
levels of INS drop more rapidly than levels of
GLU and GLP. As a result, the molar ratios of
GLU and, especially, of GLP to INS tend to
increase during the course of starvation, and this
increase favours an enhancement of the activities
of gluconeogenic enzymes and, consequently, the
gluconeogenic potential of the liver [85, 86]. At
the same time, as has been demonstrated in experi-
ments on chinook salmon in vitro, fasting for three
weeks sharply diminishes the responsiveness of the
liver slices to the glycolytic action of GLU [185].

We have not discussed the role of hormones in
fish growth in this survey. The subject was well
presented in both 1986 and 1987 [186, 187]. De-

ciphering of the structure of the first identified
piscine IGF [188], as well as the findings that the
expression of IGF in piscine liver is regulated by
the GH [188] and that an injection of GH elevates
levels of immunoreactive IGF in fish plasma [189]
open new and fertile areas for exploration of the
mutual interrelationships between GH, IGF and
INS in the regulation of fish growth and metabo-
lism. There in no doubt that stunted fish, which
possess high levels of plasma GH [187, 190],
coincidental with low levels of INS [83], may be
considered an ideal model for these studies.

CONCLUSION

The divergent life cycles and feeding habits of
fish, their consistent growth, and amazing toler-
ance of long periods of food deprivation are char-
acteristics that keep fishes in a central position in
studies of the adaptive evolution of metabolism
and its regulatory mechanisms in vertebrates.

Pancreatic hormones play the pivotal role in
these regulatory mechanisms, and an overview of
the recent literature demonstrates the growing
importance of fish as an experimental model for
studies of the structure and functions of these
hormones. There is no doubt that, if the pace of
studies of pancreatic hormones in fish remains the
same as it has been during the past several years,
answers to the majority of the questions addressed
in this review will soon become available.

ACKNOWLEDGMENTS

The author’s research was supported by grants from
the U. S. National Science Foundation (DCB 8615551
and DCB 8915935) and a Washington Sea Grant (R/A-
54). The author is indebted to Dr. Aubrey Gorbman and
Steven Duguay for critical reading of the manuscript.

REFERENCES

1 Powers, D. A. (1989) Science, 246: 352-358.

2 Plisetskaya, E. M. (1985) In “Evolutionary Biolo-
gy of Primitive Fishes”. Ed. by R. E. Foreman, A.
Gorbman, J. M. Dodd and R. Olsson, Plenum
Publ. Corp., New York, pp. 339-361.

3 Plisetskaya, E. M. (1989) In “Comparative Phys-
iology of Regulatory Peptides”. Ed. byS. Holmgren.
Chapman and Hall, London/New York, pp. 174-

10

11

12

13

14

15

16

il7/

18

19

20

21

22

23

Fish Pancreatic Hormones

202.

Plisetskaya, E. M. (1989) Fish. Physiol. Biochem.,
7: 39-48.

Plisetskaya, E. M. (1990) In “Progress in Compar-
ative Endocrinology”. Ed. by A. Epple, C. S.
Scanes and M. H. Stetson, Wiley-Liss, Inc., New
York, pp. 67-72.

Conlon, J. M., Reinecke, M., Thorndyke, M. C.
and Falkmer, S. (1988) Horm. Metab. Res., 20:
406-410.

Steiner, D. F. and Chan, S. J. (1988) Horm.
Metab. Res., 20: 443-444.

Epple, A. and Brinn, J. E. (1987) The Compara-
tive Physiology of the Pancreatic Islets. Zoophys-
iology, vol. 21, Springer Verlag, Berlin.

Conlon, J. M. (1989) In “The Comparative Phys-
iology of Regulatory Peptides”. Ed. by S. Holm-
gren, Chapman and Hall, London/New York, pp.
344-369.

Plisetskaya, E. M. (1975) Hormonal Regulation of
Carbohydrate Metabolism in Lower Vertebrates.
Nauka, Leningrad (In Russian), 215 pp.

Abad, M. E., Peeze Binkhorst, F. M., Elbal, M. T.
and Rombout, J. H. W. M. (1987) Gen. Comp.
Endocrinol., 66: 123-136.

Abad, M. E., Taverne-Thiele, J. J. and Rombout,
J. H. W. M. (1988) Gen. Comp. Endocrinol., 70:
9-19.

Abad, M. E., Lozano, M. T., Taverne-Thiele, J. J.
and Rombout, J. H. W. M. (1990) Gen. Comp.
Endocrinol., 77: 1-8.

Avala, A. G., Lozano, M. T. and Agulleiro, B.
(1987) Gen. Comp. Endocrinol., 68: 235-248.
Elliott, W. M. and Youson, J. H. (1987) Cell Tiss.
Res., 247: 351-357.

Elliott, W. M. and Youson, J. H. (1988) Amer. J.
Anat., 182: 73-83.

Nozaki, M., Miyata, K., Oota, Y., Gorbman, A.
and Plisetskaya, E. M. (1988) Gen. Comp. Endo-
crinol., 69: 271-280.

Nozaki, M., Miyata, K., Oota, Y., Gorbman, A.
and Plisetskaya, E. M. (1988) Cell Tiss. Res., 253:
371-375.

Yui, R., Nagata, Y. and Fujita, T. (1988) Arch.
Histol. Cytol., 51: 109-119.

Cheung, R., Andrews, P. C., Plisetskaya, E. M.
and Youson, J. H. (1990) Gen. Comp. Endocri-
nol., In press.

McDonald, J. K., Greiner, F., Bauer, G. E., Elde,
R. P. and Noe, B. D. (1987) J. Histochem.
Cytochem., 35: 155-162.

Cutfield, S. M., Carne, A. and Cutfield, J. F.
(1987) FEBS Lett., 214: 57-61.

Andrews, P. C., Pollock, H. G., Elliott, W. M.,
Youson, J. H. and Plisetskaya, E. M. (1988) J.

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

349

Biol. Chem., 263: 15809-15814.

Conlon, J. M., Askensten, U., Falkmer, S. and
Thim, L. (1988) Endocrinology, 122: 1855-1859.
Morel, A., Kuks, P. F. M., Bourdais, J. and
Cohen, P. (1988) Biochem. Biophys. Res. Comm..,
151: 347-354.

Marchant, T. A., Fraser, R. A., Andrews, P. C.
and Peter, R. E. (1987) Reg. Pept., 17: 41-52.
Marchant, T. A. and Peter, R. E. (1989) Fish
Physiol. Biochem., 7: 133-139.

Conlon, J. M., Agoston, D. V. and Thim, L.
(1985) Gen. Comp. Endocrinol., 60: 406-413.
Tagliafierro, G., Faraldi, G. and Pestarino, M.
(1985) Cell. Mol. Biol., 31: 201-207.
Tagliafierro, G., Farina, L., Faraldi, G., Rossi, G.
G. and Vacchi, M. (1989) Gen. Comp. Endocri-
nol., 75: 1-9.

Klaff, L. J. and Taborsky, G. J., Jr. (1987) Amer.
J. Physiol., 252 (Endocrinol. Metab. 15): E 751-E
IS).

Klaff, L. J. and Taborsky, G. J., Jr. (1987) Di-
abetes, 36: 592-596.

Klaff, L. J., Dunnung, B. E. and Taborsky, G. J.
Jr. (1988) Endocrinology, 123: 2668-2674.
D’Alessio, D. A., Sieber, C., Beglinger, C. and
Ensinck, J. W. (1989) J. Clin. Invest., 84: 857-862.
Ensinck, J. W., Laschansky, E. C., Vogel, R. E.,
Simonowitz, D. A., Roos, B. A. and Francis, B.
H. (1989) J. Clin. Invest., 83: 1580-1589.

Elliott, W. M. and Youson, J. H. (1986) Cell Tiss.
Res., 243: 629-634.

Plisetskaya, E. M., Pollock, H. G., Rouse, J. B.,
Hamilton, J. W., Kimmel, J. R., Andrews, P. C.
and Gorbman, A. (1986) Gen. Comp. Endocri-
nol., 63: 252-263.

Sheridan, M. A., Plisetskaya, E. M., Bern, H. A.
and Gorbman, A. (1987) Gen. Comp. Endocri-
nol., 66: 405-414.

Plisetskaya, E. M., Pollock, H. G., Elliott, W. M.,
Youson, J. H. and Andrews, P. C. (1988) Gen.
Comp. Endocrinol., 69: 46-55.

Hardisty, M. W. (1982) In “The Biology of Lam-
preys”. Ed. by M. W. Hardisty and I. C. Potter
Vol. 4B, Academic Press, London, pp. 165-259.
Peterson, J. D., Steiner, D. F., Emdin, S. O. and
Falkmer, S. (1974) J. Biol. Chem., 250: 5183-
5191.

Plisetskaya, E. M., Pollock, H. G., Rouse, J. B.,
Hamilton, J. W., Kimmel, J. R. and Gorbman, A.
(1985) Reg. Peptides, 11: 105-116.

Conlon, J. M. and Thim, L. (1987) In “Protein
Purification: Micro to Macro”. Ed. by R. Burgess,
UCLA Symposia on Molecular and Cellular Biolo-
gy. Vol. 68, New Series, A. R. Liss, New York, pp.
255-261.

44

45

46

47

48

49

50

51

52

53

54

55

56

SI

58

59

60

61

62

350

Rusacov, Yu. I., Karasev, V. S., Pertseva, M. N.
and Pankov, Yu. A. (1987) Biokhimia, 52: 211-
217. Plenum Press. (In Russian, English transla-
tion)

Rusacov, Yu. I., Karasev, V. S., Pertseva, M. N.
and Pankov, Yu. A. (1987) Zh. Evol. Biokhim.
Fiziol., 23: 473-480. (In Russian)

Conlon, J. M., Andrews, P. C., Falkmer, S. and
Thim, L. (1988) Gen. Comp. Endocrinol., 72:
154-160.

Berks, B. C., Marshall, C. J., Carne, A., Gallo-
way, S. M. and Cutfield, J. F. (1989) Biochem. J.,
263: 261-266.

Koval, A. P., Petrenko, A. I., Dmitrenko, V. V.
and Kavsan, V. M. (1989) Mol. Biol., 23: 473-480.
Bajaj, M., Blundell, T. L., Pitts, J. E., Wood, C.
P., Tatnell, M. A., Falkmer, S., Emdin, S. O.,
Gowan, L. K., Crow, H., Schwabe, S., Wollmer,
A. and Strassburger, W. (1983) Eur. J. Biochem.,
135: 535-542.

Conlon, J. M. and Thim, L. (1986) Gen. Comp.
Endocrinol., 64: 199-205.

Conlon, J. M. and Thim, L. (1989)
Biochem. Physiol., 93B: 359-362.

Bell, G. I., Santerre, R. F. and Mullenbach, G. T.
(1983) Nature, 302: 716-718.

Pollock, H. G., Kimmel, J. R., Ebner, K. E.,
Hamilton, J. W., Rouse, J. B., Lance, V. and
Rawitch, A. B. (1988) Gen. Comp. Endocrinol.,
69: 133-140.

Conlon, M. J., Goke, R., Andrews, P. C. and
Thim, L. (1989) Gen. Comp. Endocrinol., 73:
136-146.

Pollock, H. G., Hamilton, J. W., Rouse, J. B.,
Ebner, K. E. and Rawitch, A. B. (1988) J. Biol.
Chem., 263: 9746-9751.

Sorokin, A. V., Petrenko, O. I., Kavsan, V. M.,
Koslov, Y. I., Debabov, V. G. and Zlochevskij,
M. L. (1982) Gene, 20: 367-376.

Yamamoto, M., Kotaki, A., Okuyama, T. and
Satake, K. (1960) J. Biochem. (Tokyo), 48: 84-92.
Shuldiner, A. R., Bennett, C., Robinson, E. A.
and Roth, J. (1989) Endocrinology, 125: 469-477.
Shuldiner, A. R., Phillips, S., Roberts, C. T., Jr.,
LeRoith, D. and Roth, J. (1989) J. Biol. Chem.,
264: 9428-9432.

Andrews, P. C., Hawke, D., Shively, J. E. and
Dixon, J. E. (1985) Endocrinology, 116: 2677-
2681.

Kimmel, J. R., Plisetskaya, E. M., Pollock, H. G.,
Hamilton, J. W., Rouse, J. B., Ebner, K. E. and
Rawitch, A. B. (1986) Biochem. Biophys. Res.
Comm., 141: 1084-1091.

Conlon, J. M., Schmidt, W. E., Gallwitz, B.,
Falkmer, S. and Thim, L. (1986) Regul. Pept., 16:

Comp.

63

64

65

65a

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

E. M. PLISETSKAYA

261-268.

Pollock, H. G., Kimmel, J. R., Hamilton, J. W.,
Rouse, J. B., Ebner, K. E., Lance, V. and
Rawitch, A. B. (1987) Gen. Comp. Endocrinol.,
67: 375-382.

Noe, B. D., McDonald, J. K., Greiner, F., Wood,
J. G. and Andrews, P. C. (1986) Peptides, 7: 147-
154.

Noe, B. D., Milgram, S. L., Balasubramaniam,
A., Andrews, P. C., Calka, J. and McDonald, J.
K. (1989) Cell Tiss. Res., 257: 303-311.
Balasubramaniam, A., Andrews, P. C., Renugo-
palakrishnan, V. and Rigel, D. F. (1989) Peptides,
10: 581-585.

Tatemoto, K., Efendic, S., Mutt, G., Makk, G.,
Feistner, G. J. and Barchas, J. D. (1986) Nature,
324: 476-478.

Ahrén, B., Lindskog, S., Tatemoto, K. and Efen-
dic, S. (1988) Diabetes, 37: 281-285.

Ishizuka, J., Asada, I., Poston, G. J., Lluis, F.,
Tatemoto, K., Greeley, G. H., Jr. and Thompson,
J. C. (1989) Pancreas, 4: 277-281.

Voller, A., Bidwell, D. E. and Bartlett, A. (1979)
The Enzyme-Linked Immunosorbent Assay (EL
ISA). Dynatech Laboratories, Inc. pp. 3-67.
Albertson, B. D. and Haseltine, F. P.(Eds.) (1988)
Non-Radiometric Assays. Technology and Ap-
plication in Polypeptide and Steroid Hormone De-
tection. Alan R. Liss, New York.

Patent, G. J. and Foa, P. P. (1971) Gen. Comp.
Endocrinol., 6: 41-46.

Plisetskaya, E. M. and Leibush, B. N. (1974) Z.
evol. Biokhim. Physiol., 10: 623-625. (In Russian)
Thorpe, A. and Ince, B. W. (1976) Gen. Comp.
Endocrinol., 30: 332-339.

Furuichi, M., Nakamura, Y. and Yone, Y. (1980)
Bull. Jap. Soc. Sci. Fish., 46: 1177-1181.
Wagner, G. F. and McKeown, B. A. (1982) Can.
J. Zool., 60: 2079-2084.

Tilzey, J. F., Waights, V. and Holmes, R. (1985)
Comp. Biochem. Physiol., 81A: 821-825.
Gutiérrez, J., Fernandez, J., Blasco, J., Gesse, J.
M. and Planas, J. (1986) Gen. Comp. Endocrinol.,
63: 328-333.

Hertz, Y., Epstein, N., Abraham, M., Madar, Z.,
Hepher, B. and Gertler, A. (1989) Aquaculture,
80: 175-187.

Schrezenmeir, J., Laue, Ch., Faust, P., Kirschgess-
ner, J., Boddin, J., Liao, E., Lassak, D., Stall-
mach, Th., Muntefering, H. and Beyer, J. (1989)
Acta Endocrinol. Suppl., 1: 1. (Abstract)
Plisetskaya, E. M., Dickhoff, W. W., Paquette, T.
L. and Gorbman, A. (1986) Fish. Physiol.
Biochem., 1: 37-43.

Hilton, J. W., Leatherland, J. F. and Plisetskaya,

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96
97

98
99

100

101

102

103

104

Fish Pancreatic Hormones

E. M. (1987) Physiol. Biochem. Fish., 4: 113-120.
Plisetskaya, E. M., Donaldson, E. M. and Due, H.
M. (1987) J. Fish. Biol., 30: 21-26.

Plisetskaya, E. M., Swanson, P., Bernard, M. G.
and Dickhoff, W. W. (1988) Aquaculture, 72:
251-264.

Sheridan, M. A. and Plisetskaya, E. M. (1988)
Amer. Zool., 28: 56A. (Abstract)

Moon, T. W., Foster, G. D. and Plisetskaya, E. M.
(1989) Can. J. Zool., 67: 2189-2193.
Storebakken, T., Hung, S. S. O., Calvert, C. C.
and Plesetskaya, E. M. (1990) J. Nutrition. (In
preparation)

Tatemoto, K., Rékaeus A., Jérnvall, H., McDo-
nald, T. J. and Mutt, V. (1986) FEBS Letters, 164:
124-128.

Sheridan, M. A., Eilertson, C. D. and Plisetskaya,
E. M. (1990) Gen. Comp. Endocrinol. In Press.
Ronner, P. and Scarpa, A. (1987) Gen. Comp.
Endocrinol., 65: 354-362.

Plisetskaya, E. M., Ottolenghi, C., Sheridan, M.
A., Mommsen, T. P. and Gorbman, A. (1989)
Gen. Comp. Endocrino!., 73: 205-216.

@rskov, C. and Holst, J. J. (1987) Scand. J. Clin.
Invest., 47: 165-174.

Oshima, I., Hirota, M., Ohboshi, C. and Shima,
K. (1988) Reg. Pept., 21: 85-93.

Ottolenghi, C., Puviani, A. C., Gaviolli, M. E.,
Fabbri, E., Brighenti, L. and Plisetskaya, E. M.
(1989) Fish. Physiol. Biochem., 6: 387-394.
Andrews, P. C. and Ronner, P. (1985) J. Biol.
Chem., 260: 3910-3914.

Plisetskaya, E. M., Pollock, H. G., Rouse, J. B.,
Hamilton, J. W., Kimmel, J. R. and Gorbman, A.
(1986) Reg. Pept., 14: 57-67.

Plisetskaya, E. M., Sheridan, M. A. and Momm-
sen, T. P. (1989) J. Exp. Zool., 249: 158-164.
Plisetskaya, E. M. (1990) J. Exp. Zool. (In press)
Conlon, J. M. (1988) Diabetologia, 31: 563-566.
Blache, P., Kervran, A., Dufour, M., Martinez, J.,
Le-Nguyen, D., Lotersztajn, S., Pavoine, C.,
Pecker, F. and Bataille, D, (1989) C. R. Acad. Sci.
Paris, 308, Ser. III: 467-472.

Mommsen, T. P. and Moon, T. W. (1989) Fish
Physiol. Biochem., 7: 279-288.

Mommsen, T. P., Andrews, P. C. and Plisetskaya,
E. M. (1987) FEBS Letters, 219: 227-232.
Ghiglione, M., Blazquez, E., Uttenthal, L. O., de
Diego, J. G., Alvarez, E., George, S. K. and
Bloom, S. R. (1985) Diabetologia, 28: 920-921.
Ghiglione, M., Uttenthal, L. O., George, S. K.
and Bloom, S. R. (1984) Diabetologia, 27: 599-
600.

Shimizu, I., Hirota, M., Ohboshi, C. and Shima,
K. (1986) Biomed. Res., 7: 431-436.

105

106

107

108

109

110

111

112

116

117

118

119

120

121

122

123

124

125

126
127

128

351

Hoosein, N. M., Marenholz, A. M., Andrews, P.
C. and Gurd, R. S. (1987) Bioch. Biophys. Res.
Comm., 143: 87-92.

Holst, J. J., Orskov, C., Vagn-Nielsen, O. and
Schwartz, T. W. (1987) FEBS Letters, 211: 169-
174.

Mojsov, S., Weir, G. C. and Habener, J. F. (1987)
J. Clin. Invest., 79: 616-619.

Shima, K., Hirota, M. and Ohboshi, C. (1988)
Regulatory Peptides, 22: 245-252.

Weir, G. C., Mojsov, S., Hendrick, G. K. and
Habener, J. F. (1989) Diabetes, 38: 338-342.
Komatsu, R., Matsuyama, T., Namba, M., Wata-
nabe, N., Itoh, H., Kono, N. and Tarui, S. (1989)
Diabetes, 38: 902-905.

Orskov, C., Holst, J. J. and Nielsen, O. V. (1988)
Endocrinology, 123: 2009-2013.

Mallat, A., Pavione, C., Dufour, M., Lotersztajn,
S., Bataille, D. and Pecker, F. (1987) Nature, 325:
620-622.

Drucker, D. J. and Asa, S. (1988) J. Biol. Chem.,
263: 13475-13478.

Perez, J., Zanuy, S. and Carillo, M. (1988) Fish
Physiol. Biochem., 5: 191-197.

Weigle, D. S. (1987) Diabetes, 36: 764-775.
Milgram, S. L., Balasubramaniam, A., Andrews,
P. C., McDonald, J. K. and Noe, V. D. (1989)
Peptides, 10: 1013-1017.

Rudnicki, M., Balasubramaniam, A., McFadden,
D. W., Rigel, D. F., Nussbaum, M., Andrews, P.
C. and Fischer, J. E. (1989) American Endocrine
Society Meeting. Seattle. (Abstract)
Balasubramaniam, A., Rigel, D. F., Chance, W.
T., Stein, M., Fischer, J. E., King, D. and
Plisetskaya, E. M. (1990) Peptides, In press.
Clark, J. T., Kalra, P. S. and Kalra, S. P. (1985)
Endocrinology, 117: 2435-2442.

Morley, J. E. (1987) Endocrinol. Rev., 8: 256-
287.

Porte, D., Jr. and Kahn, S. E. (1989) Diabetes, 38:
1333-1336.

Suarez, R. K. and Mommsen, T. P. (1987) Can. J.
Zool., 65: 1869-1882.

Plisetskaya, E. M. and Sullivan, C. W. (1989)
Gen. Comp. Endocrinol., 75: 310-315.

Janssens, P. A. and Waterman, J. (1988) Comp.
Biochem. Physiol., 91B: 451-456.

Rappaport, A. M. (1958) Anat. Rec., 130: 637-
686.

Jungermann, K. and Katz, N. (1989) Physiol. Rev.,
69: 708-764.

Gumucio, J. J. (1989) Hepatology, 9: 154-160.
Tosh, D., George, K., Alberti, M. M. and Agius,
L. (1988) Biochem. J., 256: 197-204.

Lindros, K. O. and Pentilla, K. E. (1985)

352

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

Biochem. J., 228: 757-760.

Quistorff, B. and Grunnet, N. (1985) Biochem. J.,
226: 289-297.

Quistorff, B. and Grunnet, N. (1987) Biochem. J.,
243: 87-95.

Misra, U. K., Yamanaka, H., Kizaki, Z., Kauff-
man, F. C. and Thurman, R. G. (1988) Biochem.
Biophys. Res. Comm., 155: 455-462.

Hampton, J. A., McCuskey, P. A., McCuskey, R.
S. and Hinton, D. E. (1985) Anat. Rec., 213: 166-
175.

Hinton, D. E., Walker, E. R., Pinkstaff, C. A. and
Zuchelkowski, E. M. (1984) In “Use of Small Fish
Species in Carcinogenicity Testing”. National Can-
cer Institute Monograph #65. Ed. by G. L. Hoov-
er. NCI, Bethesda.

Diaz, J. P. and Connes, R. (1988) Ann. Sci. Nat.
Zool., 9: 123-141.

Mommsen, T. P., Danulat, E., Gavioli, M. E.,
Foster, G. D. and Moon, T. W. (1990) Can. J.
Zool., In press.

Ottolenghi, C., Ricci, D., Gaviolli, M. E., Puviani,
A. C., Fabbri, E., Capuzzo, A., Brighenti, L. and
Plisetskaya, E. M. (1990) Can. J. Zool., In press.
Emdin, S. O., Gammeltoft, S. and Gliemann, J.
(1977) J. Biol. Chem., 252: 602-608.

Emdin, S. O., Steiner, D. F., Chan, S. J. and
Falkmer (1985) In “Evolutionary Biology of Primi-
tive Fishes”. Ed. by R. E. Foreman, A. Gorbman,
J. M. Dodd and R. Olsson. Plenum Press, New
York, pp. 363-378.

Makower, A., Dettmer, R., Rapoport, T. A.,
Knospe, S., Behlke, J., Prehn, S., Franke, P.,
Etzoldt, J. and Rosenthal, S. (1982) Eur. J.
Biochem., 122: 339-345.

Leibush, B. N. (1981) Z. evol. Biokhim. Fiziol.,
17: 141-147. (In Russian)

Muggeo, M., Van Obberghen, E., Kahn, C. R.,
Roth, J., Ginsberg, B. H., De Meyts, P., Emdin,
S. O. and Falkmer, S. (1979) Diabetes, 28: 175-
181.

Muggeo, M., Ginsberg, B. H., Roth, J., Neville,
D. M., De Meyts, P. and Kahn, C. R. (1979)
Endocrinology, 104: 1393-1402.

Emdin, S. O. (1981) Myxine insulin. Amino acid
sequence, three dimensional structure, biosynthe-
sis, release, physiological role, receptor binding
affinity, and biological activity. Umea University
Medical Dissertations, New Series #66, Umea,
pp. 1-158.

Leibush, B. N. (1989) Insulin receptor interactions
in vertebrate evolution. D. Sc. Thesis, Leningrad,
USSR. (In Russian)

Czech, M. P. and Massague, J. (1982) Fed. Proc.,
41: 2719-2723.

146

147

148

149

150

151

152
153

154

155

156

157

158

159

160

161

162

163

164

165

166

E. M. PLISETSKAYA

Le Roith, D., Lowe, W. L., Jr. and Roberts, C. T.,
Jr. (1989) In “Molecular and Cellular Biology of
Diabetes Mellitus”. Ed. by B. Draznin, S. Melmed
and D. LeRoith, Vol. 2, pp. 1-9.

Stuart, C. A. (1988) J. Biol. Chem., 263: 7881-
7886.

Foty, R. A. and Liversage, R. A. (1989) Amer.
Zool., 29: 20A.

Simon, J. and LeRoith, D. (1986)
Biochem., 158: 125-132.

Simon, J., Rosebrough, R. W., McMurtry, J. P.,
Steele, N. C., Roth, J., Adamo, M. and LeRoith,
D. (1986) J. Biol. Chem., 261: 17081-17088.
Shemer, J., Perrotti, N., Roth, J. and LeRoith, D.
(1987) J. Biol. Chem., 262: 3436-3439.

Rosen, O. M. (1987) Science, 237: 1452-1458.
Bassas, L. L., De Pablo, F., Lesniak, M. A. and
Roth, J. (1987) Endocrinology, 121: 1468-1476.
Gutiérrez, J. and Plisetskaya, E. M. (1989) Insulin
binding to the liver membranes: teleost model. XI
Intern. Symp. Comp. Endocrinol., Malaga, Spain,
Thomas Jefferson University Printing Center, Phi-
ladelphia, p. 146.

Chan, S. J., Cao, Q-P. and Steiner, D. F. (1989)
Conference on Insulin. Structure, Chemistry and
Biology, University of York, England, p. 48. (Ab-
stract)

Gavin, J. R., II, Roth, J., Neville, D. M., Jr., De
Meyts, P. and Buell, D. N. (1974) Proc. Natl.
Acad. Sci. U.S.A., 71: 84-88.

Bar, R. S., Gorden, P., Roth, J., Kahn, C. R. and
de Meyts, P. (1976) J. Clin. Invest., 58: 1123-
1135.

Kahn, C. R. (1979) Proc. Soc. Exper. Biol. Med.,
162: 13-21.

Ablett, R. F., Taylor, M. J. and Selivonchick, D.
P. (1983) Br. J. Nutr., 50: 129-140.
Christiansen, D. S. and Klingsgyr, L. (1987)
Comp. Biochem. Physiol., 88B: 701-711.
Plisetskaya, E. M., Leibush, B. N. and Bondareva,
V. M. (1976) In “The Evolution of Pancreatic
Islets”. Ed. by T. A. I. Grillo, L. Leibson and A.
Epple, Pergamon Press, Oxford, pp. 251-269.
Leibush, B. N. and Bondareva, V. M. (1987) Z.
Evol. Biokhim. Fiziol., 23: 193-198. (In Russian)
Leibush, B. N., Soltitskaya, L. M. and Ukhanova,
M. V. (1987) Z. Evol. Biokhim. Fiziol., 23: 468-
472. (In Russian)

LeRoith, D., Lowe, W.L.,Jr.,Shemer, J., Raizada,
M. K. and Ota, A. (1988) Int. J. Biochem., 20:
225-230.

Van Schravendijk, F. H., Hooghe-Peters, E. L.,
De Meyts, P. and Pipeleer, D. G. (1984) Biochem.
J., 220: 165-172.
Gutiérrez, J.,

Bure:

Asgard, T., Fabbri, E. and

167

168

169

170

171

172

173

174

175

176

177

178

Fish Pancreatic Hormones

Plisetskaya, E. M. (1989) Amer. Zool., 29: 19A.
(Abstract)

Shimizu, I., Hirota, M., Ohboshi, C. and Shima,
K. (1987) Endocrinology, 121: 1076-1082.
Horwitz, E. M. and Gurd, R. S. (1988) Arch.
Biochem. Biophys., 267: 758-769.

Bern, H. A. (1978) In “Comparative Endocrinolo-
gy’. Ed. by P. J. Gaillard and H. H. Boer.,
Elsevier/North Holland Biomedical Press, Amster-
dam, pp. 97-100.

Barron, M. G. (1986) J. Endocrinol., 108: 313-
319.

Hoar, W. S. (1988) In “Fish Physiology”. Ed. by
W. S. Hoar and D. J. Randall. Acad. Press. San
Diego, vol. 11B, pp. 273-343.

Woo, N. Y. S., Bern, H. A. and Nishioka, R. S.
(1978) J. Fish. Biol., 23: 625-639.

Sheridan, M. A., Woo, N. Y. S. and Bern, H. A.
(1985) J. Exper. Zool., 236: 35-44.

Bondareva, V. M., Soltitskaya, L. P. and Rusacov,
Ju. I. (1980) Z. Evol. Biokhim. Fiziol., 16: 518-
521. (In Russian)

Dickhoff, W. W., Yan, L., Plisetskaya, E. M.,
Sullivan, C. V., Swanson, P., Hara, A. and Ber-
nard, M. G. (1989) Fish Physiol. Biochem., 7:
147-155.

French, C. J., Hochachka, P. W. and Mommsen,
T. P. (1983) Am. J. Physiol., 245, Reg. Integr.
Comp. Physiol., 14: R827—R830.

Cowey, C. B., Knox, C., Walton, M. J. and
Andron, J. W. (1977) Brit. J. Nutr., 38: 463-470.
Yan, L. (1988) Profiles of seasonal change in
plasma sex steroids, thyroid hormones, pancreatic
hormones and vitellogenin in Atlantic salmon (Sal-
mo salar). Ph. D. Thesis, University of Washing-
ton, Seattle.

179

180

181

182

183

184

185

186

187

188

189

190

191

353

Gutierrez, J., Fernandez, J., Carillo, M., Zanuy,
S. and Planas, J. (1987) Fish Physiol. Biochem., 4:
137-141.

Tyler, C., Sumpter, J. and Bromage, N. (1987)
Third Intern. Symp. Reprod. Physiol. of Fish., p.
142. St. John’s, Canada. (Abstract)

Cherel, Y., Robin, J. P., Walch, O., Karmann, H.,
Netchitailo, P., and LeMaho, Y. (1988) Amer. J.
Physiol., 254, Regul. Integr. Comp. Physiol., 23:
R170-R177.

LeNinan, F., Cherel, Y., Robin, J.-P. Leloup, J.
and LeMaho, Y. (1988) J. Comp. Physiol., B. 158:
395-401.

LeNinan, F., Cherel, Y., Sardet, C. and LeMaho,
Y. (1988) Gen. Comp. Endocrinol., 71: 331-337.
Gutierrez, J., Perez, J., Navarro, I., Zanuy, S. and
Carillo, M. (1989) XI Intern. Symp. Comp. En-
docrinol., Malaga, Spain, Thomas Jefferson Uni-
versity Printing Center, Philadelphia, p. 147.
Klee, M., Eilertson, C. and Sheridan, M. A. (1990)
J. Exp. Zool., 254: 202-206.

Matty, A. J. (1986) Fish Physiol. Biochem., 2:
141-150.

Bolton, J. P., Young, G., Nishioka, R. S., Hirano,
T. and Bern, H. A. (1987) J. Exp. Zool., 242: 379-
382.

Cao, Q-P., Duguay, S. J., Plisetskaya, E. M.,
Steiner, D. and Chan, S. J. (1989) Molec. Endocri-
nol., 3: 2005-2010.

Funkenstein, B., Sibergeld, A., Cavari, B. and
Laron, Z. (1989) J. Endocrinol., 120: R19-R21.
Gray, E. S., Young, G. and Bern, H. (1989)
Amer. Zool., 29: 46A. (Abstract)

Gutiérrez, J. and Plisetskaya, E. M. (1990) In
preparation.

oF
Ja
aie

¥ - a
, rae -
rat! no ibedtlis’ 20 pelle OTL
frm i Fd OL (PRR) ft OUT Btpobeen t
Thlath! :
1) Wey ticeter A die oo leeg erie ORG} fiehy t |
he tetheenty banged syd ev sag with :
) TE) aati D1 eRe BAT ies
rdw OC) hho yh i te art reo
peree CPE | POM, Corte | Ba Cor eat a
oad do tone ieee, . dant on
Taas wah ee | : eh ara
diet eb dos) AY pera ate Ga age mite eM
fb ioral Liseeiay vuole Mr eon - tz © fot ae
d bite DOE CMS Me ae
io) te nile Victoealiate 2° niiiviek! 2g (itd show), ln, Nebrwdt
! ocr Scaigull woiect J. vial) (ABGT) 2 Y ot ey olsen s: "
rule. Serie 1 soto 1 senotten.” Beb- CS tie? Ge (MAR,
a3: yeh Zein aK TORY nile ip? i a tia Aho oe
comtjat shiiad® otuye .eqelel’ , tose, a Poel boar eT YD Ae
ber Gehan . sete gett) eiased Wh. ad : 2k dattiade
se > sl ead rit .. Wyn A eiRt veeqived tog beoely at

oa

oo 2 ara 1 bl 2 eden vont ert, Ang
ty ghd! Ved At) th Ad Milo RE he i femeote Ay a
; (a? Tt f Pal .' Wawa 1 ogi] m | net, we
/ a eae 5% wiley. 9. Eel ™) — —s1utt my. ol 7 "4 an
nes, 123) eee at Pt ore Chaat trate ae
- any 8

edsieot . Acdarveliaytt Oe ae cmnintina WM wettool 9i
4 ‘RO } . ttt ken 6). tonto i wet ea Pe. eros oer:
Sushi’ oh. (Tie : (aS), LSR ST OY

fel ded od ede: We ee eee el 12 oh

: ey “onkiihba , S nen OP. nb snub acne TR, by
seeet, % ee r eid tal as <qulinty YWwreaete bty emit
ih 7 Chute tired rey davenrtod tera, thi si

: : ‘ TS vt at) etn: SOT AB Cn Galidings§ a

“oy pralt Je eabhovind keels .O ig
” c 4 Ab tind
ig mm, ia

oF

my

‘
' 2 oo
i \
‘ - -
Cu
'
ha ae
‘fi
*
i
t
ot, A Pa ee

peOhay Tint

i ’ Oy
ast Moet, 1 Ub es

Aare! tT .
~_

ZOOLOGICAL SCIENCE 7: 355-370 (1990)

REVIEW

Structure and Function of Sea Urchin Egg Jelly Molecules

Norio SUZUKI

Noto Marine Laboratory, Kanazawa University, Ogi, Uchiura,
Ishikawa 927-05, Japan

ABSTRACT— Sea urchin egg jelly consists of a fucose sulfate glycoconjugate, a sialoglycoprotein and
sperm-activiting peptides. The fucose sulfate glycoconjugate isolated from the solubilized jelly layer
of Hemicentrotus pulcherrimus is composed of several proteins (a sugar-containing core protein, a 258
kDa protein, a 237 kDa protein and a 120 kDa protein); the substance plays a major role in the
inducetion of the acrosome reaction in Hemicentrotus pulcherrimus spermatozoa. The protein
components of the fucose sulfate glycoconjugate seem to be associated with one another through
disulfide bonds. Sixty-three sperm-activating peptides have been isolated from the solubilized jelly
layers of fifteen sea urchin species distributed over four taxonomic orders. Sperm-activating peptides
stimulate sperm respiraton and motility, increase the levels of both cAMP and cGMP in sea urchin
sperm cells, induce net H* efflux from sea urchin spermatozoa and cause a shift in the mobility of
guanylate cyclase, a major sperm plasma protein, on an SDS-polyacrylamide gel. These peptides are
specific at the ordinal level and are classified into four groups, i.e., sperm-activating peptide I (SAP-I),
sperm-activating peptide II (SAP-II), sperm-activating peptide III (SAP-III) and sperm-activating
peptide IV (SAP-IV). SAP-I (Gly-Phe-Asp-Leu-Asn-Gly-Gly-Gly-Val-Gly) promotes the induction
of the acrosome reaction in Hemicentrotus pulcherrimus spermatozoa by acting as a specific co-factor

© 1990 Zoological Society of Japan

of the fucose sulfate glycoconjugate.

INTRODUCTION

Sea urchin eggs are surrounded by a transpa-
rent, gelatinous matrix called the jelly layer,
through which spermatozoa must pass before
reaching the plasma membrane of the egg during
fertilization. The jelly layer has been shown to
induce the acrosome reaction, which was first
described in detail by Dan [1-14]. The jelly layer
of sea urchins mainly consists of polysaccharide-
protein complexes which can be separated into a
fucose sulfate glycoconjugate and a sialoglycopro-
tein [13, 15-22]. The latter is considered to cause
sperm-isoagglutination and the former to be re-
sponsible for the induction of the acrosome reac-
tion. The acrosome reaction in spermatozoa is an
essential requirement for fertilization of eggs in
many animals [23-43]. In sea urchins, the acro-
some reaction occurs within seconds after sperma-

Received March 26, 1990

tozoa have come into contact with the egg jelly
layer [4, 9-14, 43] or the egg surface component, a
sperm-binding factor on the vitelline membrane
[45].

Considering the large number of acidic groups
belonging to the ester-linked sulfate in the fucose
sulfate glycoconjugate and carboxyl residues in the
bound sialic acids of the sialoglycoprotein, it fol-
lows that the pH values within the jelly layer would
be lower than the pH of normal sea water [46-47].
At low pH values, the respiration and motility of
sea urchin spermatozoa are markedly reduced and
their movements in the jelly layer, therefore, are
slower than those that prevail in sea water [48-51].
Thus, one can surmise that the egg jelly is provided
with a system of controlling motility and respira-
tion, because the spermatozoa need to maintain
their activity at the initial high level attained in sea
water during their passage through the acidic en-
vironment of the jelly layer to reach the egg
surface.

356

It has been known for over 70 years that the jelly
layers of certain species of sea urchins, dissolved in
sea water, are able to cause a transient activation
of sperm motility and sperm agglutination [52].
Since then many investigators have shown that
soluble factors associated with sea urchin eggs
stimulate the motility and respiration of sea urchin
spermatozoa [53-58]. However, such stimulation
has only been seen on occasion and has been
difficult to reproduce. In 1976, Ohtake demons-
trated that the respiration of sea urchin spermato-
zoa could be reproducibly stimulated by jelly layer
factors obtained from the sea urchin Pseudocentro-
tus depressus if the extracellular pH was main-
tained at acidic values [50-51]. Thereafter, Kopf
et al. and Hansbrough and Garbers reported that
the jelly layer of the sea urchin Strongylocentrotus
purpuratus contains a substance which elevates the
respiration rates and cyclic nucleotide concentra-
tions in Strongylocentrotus purpuratus spermato-
zoa [59-61]. Hansbrough and Garbers named the
substance speract which stands for sperm-
activating peptide. In 1981, my collaborators and I
first purified a substance from the solubilized jelly

.5SM KC
(as |

Centrifugation
at 600xg

for 10 min

—_—_————————

See

and

at 10,000xg

for 30 min

FSW * Egg
pH 8,02 Suspension

N. SuZUKI

layer of the sea urchin Hemicentrotus pulcherrimus
and demonstrated that the substance is a decapep-
tide whose structure is Gly-Phe-Asp-Leu-Asn-Gly-
Gly-Gly-Val-Gly and stimulates the respiration
and motility of Hemicentrotus pulcherrimus sper-
matozoa [62]. The same peptide has also been
purified from the solubilized jelly layer of Strongy-
locentrotus purpuratus [63].

PURIFICATION OF EGG JELLY
MOLECULES

Since the jelly layer of sea urchin mainly consists
of highly viscous large molecular weight polysac-
charide-protein complexes, it is quite difficult to
deal with an intact jelly layer using common
biochemical methods. In preparation for analysis
of the jelly layer, it must first be dissolved, which is
accomplished by treating eggs with sea water ad-
justed to pH. 5.0 or 5.5 (Fig. 1). The centrifuged
supernatant (solubilized jelly layer) is used for the
purification of jelly molecules.

To purify sperm-activating peptides, the solubil-
ized jelly layer is mixed with a 2-fold volume of

Supernatant
(sialic acid-rich
material)
Crude
Jelly >

Solution Freezing Centrifugation

and at 10,000xg
Thawing for 30 min

Pellet
(fucose-rich
material)

Supernatant
»

Concentration by Ultra-Filtration

9

Chromatography on Sepharose 2B
and DEAE-Sephadex A-25 columns
and Density gradient centrifugation

(Amicon YM-10 membrane)

Yd

Filtrate

¥

HPLC using
reverse-phase columns

Pellet

¥

Gel filtration
on Sepharose 4B, 2B
and CL-2B columns

¥

Fucose Sulfate
Glycoconjugate (FSG)

Sialoglycoprotein

Sperm-—Activating Peptide

(SAP-I : Gly-Phe—Asp—Leu-Asn-Gly—Gly—Gly-Val-Gly)

Fic. 1. Schematic drawing of the purification procedure for sea urchin egg jelly molecules.

Egg Jelly Molecules 357

99% ethanol and then centrifuged. The resulting
supernatant is concentrated under reduced press-
ure at 50°C and lipids are removed by chloroform
extraction. The water layer is then lyophilized and
the residue, dissolved in deionized and distilled
water (DDW), is used for peptide purification.
When my collaborators and I first isolated sperm-
activating peptides from the solubilized jelly layer
of Hemicentrotus pulcherrimus, we used sequential
chromatographies on Sephadex G-25, DEAE-
Sephadex A-25, Sephadex G-15 and Avicell SF
thin layer plate [62]. At that time we used about 20
liters of solubilized jelly layer prepared from 5,000
female sea urchins. This procedure, however, was
time-consuming and required a large amount of
solubilized jelly layer to obtain pure peptides. We
then modified the purification procedure and we
now use sequential high performance liquid chro-
matography (HPLC) on a reverse-phase column to
purify the peptides [64]. In general, separations
are carried out using a combination of the fllowing
programs. Program I: Flow rate is 9.9 ml/min, the
column (Shimpack C-8 Prep, particle size 5 ~m, 20
250mm) is equilibrated with 5% acetonitrile
(ACN) in 0.1% trifluroacetic acid (TFA) in DDW
and eluted 15 min with equilibration solvent, fol-
lowed by elution with 60% ACN in TFA in DDW
for the next 15 min. Program II: Flow rate is 1.0
ml/min, the column (Unisil C-8, particle size 5 um,
4.6250 mm) is equilibrated with 10% ACN in
0.1% TFA in DDW and eluted for 10 min with
equilibration solvent, followed by a linear gradient
of ACN from 10% to 50% in 0.1% TFA in DDW
over a 50 min time period. Program III: Flow rate
is 1.0 ml/min, the same column used in Program II
is equilibrated with 5% ACN in 5 mM sodium
phosphate (pH 5.7) and eluted for 20 min with the
equilibration solvent, followed by a linear gradient
of ACN from 5% to 30% in 5 mM sodium phos-
phate (pH 5.7) over a 40 min time period. By
using this modified procedure, we purified three
new sperm-activating peptides from solubilized
jelly layer obtained from 150 female Hemicentro-
tus pulcherrimus sea urchins.

For purification of the large molecular weight
components of the jelly layer, solubilized jelly
layer is dialyzed against 0.1M NaCl and then
applied to a Sepharose 2B column equilibrated

with 0.1M NaCl at 4°C [17-18, 65]. Fucose-
containing material is eluted earlier than the mate-
rial containing sialic acid. To purify a fucose
sulfate glycoconjugate, fractions containing fucose
are subjected to chromatography on a Sepharose
CL-2B column with 7M urea containing 10 mM
HEPES (pH7.0) [18, 65]. The purified fucose
sulfate glycoconjugate which contains one mol
sulfate/mol fucose possesses 2.0 times the amount
of protein to fucose by weight. When further
purification is needed, a fucose sulfate glycoconju-
gate containing-fraction obtained from Sepharose
CL-2B gel filtration is subjected to HPLC on a
TSK G-6000 PW column equilibrated with 0.1 M
sodium phosphate (pH 6.8) containing 0.1% SDS.

Fractions containing sialic acid obtained from
chromatography of solubilized jelly layer on a
Sepharose 2B column are pooled and then di-
alyzed against 10mM sodium acetate (pH 5.0)
containing 0.1 M NaCl and chromatographed on a
DEAE-Sephadex A-25 column with a linear gra-
dient of NaCl from 0.1M to 1.2M in sodium
acetate (pH 5.0) [65]. Fraction containing sialic
acid are pooled and dialyzed against DDW. Solid
guanidine hydrochloride, CsCl and 1M sodium
acetate (pH 5.0) are added to the dialysate at final
concentrations of 4M, 0.47 g/ml and 10 mM, re-
spectively. The solution is centrifuged at 100,000
xg for 48 hr in a Hitachi RPS 50 rotor at 4°C. A
sialoglycoprotein is obtained from the clear, bot-
tom fraction with a density of 1.47 g/ml. The
sialoglycoprotein consists of sialic acid (90%, w/w)
and protein (10%, w/w).

It should be noted that in the case of
Hemicentrotus pulcherrimus, the fucose-containing
material becomes insoluble in a 1 M NaCl solution
upon freeze-thawing. About 80% of the fucose in
the original solubilized jelly layer is recovered in
the precipitate fraction by centrifugation at 10,000
xg for 30 min (Fig. 1) [65]. The precipitate thus
obtained dissolves easily in DDW but not in arti-
ficial sea water (ASW) or even in 4M guanidine
hydrochloride solution. The supernatant fraction
contains about 76% of the sialic acid present in the
original solubilized jelly layer. Thus, we some-
times use the precipitate or the supernatant frac-
tion for the purification of a fucose sulfate glyco-
conjugate or a sialoglycoprotein. However, this

358 N. Suzuki

works only with solubilized jelly layer prepared _ sispina, Strongylocentrotus nudus, Pseudocentrotus
from Hemicentrotus pulcherrimus eggs. The solu- depressus or Clypeaster japonicus does not precipi-
bilized jelly layer obtained from Anthocidaris cras- _ tate upon freeze-thawing [66].

TABLE 1. Sperm-activating peptides from the egg jelly of various species of sea urchins*

Subclass Regularia
Order Diadematoida
Suborder Diademina
Family Diadematoidae
Diadema setosum
GCPWGGAVC (SAP-IV, Mw 847)

Order Arbacioida
Suborder Phymosomatina
Family Phymosomatidae

Glyptocidaris crenularis
SAKLCPGGNCV (Ser, Ala-SAP-IIB, Mw 1048)
KLCPGGNCV (SAP-IIB, Mw 890)
LCPGGNCV (Des-Lys'-SAP-IIB, Mw 762)
SFKLCPGGQCV (ser, Phe-[Gln’]-SAP-IIB, Mw 1138)
KLCPGGQCV (]Gin’]-SAP-IIB, Mw 904)
LCPGGQCV (Des-Lys'-[GIn’]-SAP-IIB, Mw 776)

Suborder Arbacina
Family Arbaciidae
CVTGAPGCVGGGRL-NH, (SAP-ITA, Mw 1245)

Order Echinoida
Suborder Temnopleurina
Family Toxopneustidae
Lytechinus pictus
GFDLTGGGVQ_ ([Thr°, Gln'°]-SAP-I, Mw 950)
FDLTGGGVQ (Des-Gly?-[Thr°, Gln’°]-SAP-I, Mw 893)

Tripneustes gratilla

GFDLNGGGVG (SAP-I, 892)

GFNLNGGGVG_ (Asn°]-SAP-I, Mw 891)
GFSIGGGGVG _ ({Ser®, Ile*, Gly°]-SAP-I, Mw 807)
GFDLGGGGVG _ ([Gly°]-SAP-I. Mw 835)
GFSLGGGGVG _ ([Ser*, Gly*]-SAP-I, Mw 807)
GFGLGGGGVG ([Gly*?]-SAP-I, Mw 777)
G(Br-F)NLNGGGVG _ ([Br-Phe?, Asn*]-SAP-I, Mw 971)
G(Br-F)DLNGGGVG _([Br-Phe*]-SAP-I, Mw 972)

Pseudoboletia maculata
GFALDGVN (Des-Gly®’-[Ala*, Asp°,’ Asn'°]-SAP-I, Mw 792)
GFALDGVG (Des-Gly®”-[Ala®, Asp°]-SAP-I, Mw 735)

Pseudocentrotus depressus

GFDLNGGGVG (SAP-I, Mw 892)
GFDLTGGGVG _ (Thr°]-SAP-I, Mw 879)
GFALGGGGVG (Ala*, Gly°]-SAP-I, Mw 791)

Suborder Echinina
Family Strongylocentrotidae

Strongylocentrotus nudus
GFDLNGGGVG (SAP-I, Mw 892)
GFSLSGGGVG ____({Ser*°]-SAP-I, Mw 837)
GFALGGGGVG ({Ala*, Gly°]-SAP-I, Mw 791)
GFSLGGGGVG __ ({Ser*, Gly*]-SAP-I, Mw 807)
GFDLTGGGVG __({Thr°]-SAP-I, Mw 879)

Egg Jelly Molecules

(continued Table 1)

Strongylocentrotus purpuratus

GFDLNGGGVG
GFALGGGGVG
GFSLTGGGVG

(SAP-I, Mw 892)
([Ala*, Gly>]-SAP-I, Mw 791)
({Ser*, Thr?]-SAP-I, Mw 851)

Hemicentrotus pulcherrimus

GFDLNGGGVG
GFDLTGGGVG
GFSLNGGGVS
SFALGGGGVG
GFSLSGSGVD

Family Echinometridae

(SAP-I, Mw 892)

([Thr°?]-SAP-I, Mw 879)
([Ser?'°]-SAP-I, Mw 894)

({Ser’, Ala*, Gly°]-SAP-I, Mw 821)
({Ser?°’, Asp’°]-SAP-I, Mw 925)

Echinometra mathaei (type A)

GYSLSGGAVD
GFALSGGGVG
GFSLSGGGVG

GFDLTGGGVG

([Tyr?, Ser?°, Ala’, Asp'°]-SAP-I, Mw 925)
({Ala’, Ser°]-SAP-I, Mw 821)
({Ser?]-SAP-I, Mw 837)

({Thr3]-SAP-I, Mw 879)

Echinometra mathaei (type B)

GYSLSGGAVD
GYNLNGDRID
GFSLSGGGVG
GFDLTGGGVG

([Tyr?, Ser?, Ala’, Asp'°]-SAP-I, Mw 925)

({(Try?, Asn?, Asp’"°, Arg®, Ile’]-SAP-I, Mw 1136)

([Ser?”]-SAP-I, Mw 837)
({(Thr°]-SAP-I, Mw 879)

Anthocidaris crassispina

GFDLTGGGVG
GFDLSGGGVG
GFSLSGSGVG

({(Thr°]-SAP-I, Mw 879)
([Ser°?]-SAP-I, Mw 865)
(Ser*>]-SAP-I, 867)

Heterocentrotus mammillatus

GTLPTGSGVS
GFEMGGTGVG
GYNLGGGGID
GFGLSGGGIG

Subclass Irregularia
Order Clypeasteroida
Suborder Clypeasterina
Family Clypeasteroidae

((Thr?”, Leu®, Pro*, Ser”’'°]-SAP-I, Mw 875)
([Glu*, Met*, Gly°, Thr’]-SAP-I, Mw 911)

({Tyr?, Asn’, Gly, Ile’, Asp’°]-SAP-I, Mw 922)

([Gly*, Ser°, Ile?]-SAP-I, Mw 821)

Clypeaster japonicus

DSDSAONLIG
GTDSAQNLIG
SDSAQNLIG
DSDSAHLIG
DTDSAHLIG

NTDSAHLIG
GTDSAHLIG
SDSAHLIG
DSDSAFLIG

Suborder Laganina
Family Astriclypeidae
Astriclypeus manni
DSDSAHLIG
DTDSAHLIG
TDSAHLIG

* Amino acids in the sequences are given a one-letter abbreviation: asparagine (N), aspartic
acid (D), alanine (A), arginine (R), isoleucine (I), glycine (G), glutamine (Q), glutamic acid
(E), cystine (C C), cysteine (C), serine (S), tyrosine (Y), tryptophan (W), valine (V), histidine

(SAP-III, Mw 1019)

({Gly!, Thr?]-SAP-III, Mw 975)
(Des-Asp'-SAP-III, Mw 904)
(Des-Asn’-[His°]-SAP-III, Mw 914)
(Des-Asn’-[Thr*, His°]-SAP-III, Mw 928)

(Des-Asn’-[Asn’, Thr’, His°]-SAP-III, Mw 928)
(Des-Asn’-[Gly!, Thr?, His°]-SAP-IIIl, Mw 870)
(Des-Asp', Asn’-[His®]-SAP-III, Mw 799)
(Des-Asn7-[Phe®]-SAP-III, Mw 924)

(Des-Asn/-[His°]-SAP-III, Mw 914)
(Des-Asn’-[Thr*, His®]-SAP-III, Mw 928)
(Des-Asp’, Asn’-[Thr*, His°]-SAP-III, Mw 813)

(H), phenylalanine (F), proline (P), methionine (M), lysine (K) and leucine (L).

359

360 N. Suzuki
TaBLE2. Respiratory stimulating effect of sperm-activating peptides on sea urchin sperma-
tozoa
Spermatozoa used

Diadema Glyptocidaris Hemicentrotus Clypeaster

setosum crenularis pulcherrimus japonicus
SAP-IV =F = = =
SAP-IIB = ar = =
SAP-ITA — + = =
SAP-I — = oF =
SAP-III = = = se

The respiratory-stimulating activity of a sperm-activating peptide is expressed as a plus sign (+)
when the peptide stimulated sperm respiration one half-maximally less than 5nM. When the
half-maximal respiratory stimulation was induced by a peptide concentration between 5 and 500
nM, the plus-minus sign (+) was used. Practically no respiratory stimulation was depicted by a

minus sign (—).

SPERM-ACTIVATING PEPTIDES

During the last ten years, my collaborators and I
purified sixty-three sperm-activating peptides from
the solubilized jelly layer obtained from eggs of
fifteen sea urchin species distributed over four
taxonomic orders (Table 1) [62, 67-75]. These
peptides essentially demonstrate the same biolo-
gical activity toward a given sea urchin spermato-
zoa although the biological activities of the pep-
tides are specific at the ordinal level (Table 2) [76-
77]. Considering structure and biological specific-
ity, the peptides can be classified into four groups,
i.e., sperm-activating peptide I from the species in
the order Echinoida, sperm-activating peptide II
(subclass A and B) from the species in the order
Arbacioida, sperm-activating peptide III from the
species in the order Clypeasteroida and sperm-
activating peptide IV from the species in the order
Diadematoida. These groups of peptides may be
abbreviated as SAP-I, SAP-II, SAP-III and SAP-
IV [64].

The peptides stimulate the decreased respiration
rates of sea urchin spermatozoa due to the aci-
dification of sea water, back to the level of respira-
tion rates in normal sea water [68]. The stimulated
respiration rate does not exceed that of spermato-
zoa in normal sea water even if large amounts of
peptides are added to the sperm suspension
medium. The respiratory stimulation induced by
the peptide continues for a few minutes and then
sperm respiration rates usually decline to the basal

rate normally observed at that particular pH.
After reaching the basal state, spermatozoa can
again be stimulated by the addition of peptide.
This is also true when respiratory stimulation is
induced by solubilized jelly layer. When a limited
concentration of peptide is used for respiratory
stimulation of spermatozoa and the sperm sus-
pending medium is centrifuged before the second
addition of peptide, the resultant supernatant
fluid, upon examination for respiration-stimulating
activity toward spermatozoa, does not stimulate
sperm respiration [78]. This suggests that the
peptide binds tightly to the spermatozoa. Smith
and Garbers reported that Strongylocentrotus pur-
puratus spermatozoa had approximately 6,000-
8,000 binding sites (receptors)/cell specific for
SAP-I [79]. We suggested that in Hemicentrotus
pulcherrimus spermatozoa, the receptors exclu-
sively localize on the sperm tail [78].

Monensin, an ionophore that catalyzes an elec-
tro-neutral Nat/H* exchange across the cell mem-
brane stimulates sea urchin sperm respiration and
motility one half-maximally at about 1-10 uM,
and induces a Na*-dependent net proton efflux as
well as an increased flux of Na* in both directions
across the cell membrane [61, 68, 80-82]. Sperm-
activating peptide stimulate sperm respiration one
half-maximally at about 10-100 pM. The respira-
tory stimulation induced by SAP-I or monensin is
dependent on the concentration of external Na*
[61, 68, 80, 82]. Approximately 50mM Na® is
required for half-maximal respiratory responses to

Egg Jelly Molecules

peptides or monensin. There seem to be similar-
ities between the effects of SAP-I and monensin.
It is well known that many types of metabolic
activation in cells are induced by intracellular
alkalinization [83]. Thus, induction of respiratory
stimulation by sperm-activating peptides may be
explained by the hypothesis that the peptides
trigger Na*/H* exchange across sperm plasma
membrane and raise the intracellular pH [43, 84].

It has been known that spermatozoa from va-
rious species including sea urchin spermatozoa
possess enzymes such as adenylate cyclase, guany-
late cyclase, cyclic nucleotide phosphodiesterase,
cyclic nucleotide-dependent protein kinases and
phosphoprotein phosphatases which are involved
in cyclic nucleotide metabolism [84-97]. In many
instances, these enzymes possess higher specific
activity in spermatozoa than in other tissues.

Sperm-activating peptides cause transient in-
creases in sea urchin sperm cGMP concentrations
as well as cAMP concentrations in both acidic and
normal sea water [63, 69, 73]. Half-maximal
elevations of cGMP are 2X10~-°M and _ half-
maximal elevations of cAMP are 2x 10° M pep-
tide (Fig. 2). The increases in cGMP concentra-
tions are explained by transient activation of the
membrane form of guanylate cyclase, which is a
major protein of sperm tail plasma membrane [43,
88, 98-102].

250F pH6.6 S)
§ 60 g 4
Q 5
© &
= nee |,
eal 402
ope o|Na/” 03
D 'o))
E E |,
oO ~~
2 =
O S
To) on
= e
0 0 0

o1514131211109 8 76 5
—Log4ol Concentration] [M]

nmol O2/mg spermatozoa

361

Ward and Vacquier reported that within seconds
after additon of solubilized jelly layer prepared
from Arbacia punctulata to an Arbacia punctulata
sperm suspension, a 160 kDa sperm protein dis-
appeared and a new protein appeared at 150 kDa
onan SDS-polyacrylamide gel [103-105]. The
extent fo the change was a function of jelly concen-
tration but, at a given jelly concentration, was
independent of incubation time. The change is
specific for Arbacia punctulata jelly. Thereafter,
we demonstrated that the factor responsible for
the change is a sperm-activating peptide whose
sequence is Cys-Val-Thr-Gly-Pro-Gly-Gys-Val-
Gly-Gly-Gly-Arg-Leu-NH> (SAP-ITA) [69]. Simi-
lar changes are commonly observed in sea urchin
spermatozoa of many species treated with a spe-
cific sperm-activating peptide [70-76].
Hemicentrotus pulcherrimus spermatozoa possess
several major proteins and one of them, which was
recently identified as guanylate cyclase, changes its
relative mobility on SDS-polyacrylamide gels from
131 kDa to 128 kDa upon treatment with SAP-I
(Fig. 3) [106]. The 160 kDa protein of Arbacia
punctulata spermatozoa can be labeled with ip.
orthophosphate when intact Arbacia punctulata
spermatozoa are incubated in sea water containing
?P_orthophosphate. The label which is in the form
[°*P]-phosphoserine, disappears completely after
exposure by the spermatozoa to the solubilized

wo oO
ea ee :
on Male s
40 ©
E OOF 3 :
D
Sa cli ee
0 5 10 15
o Time (sec.) oo
o
S 500 =
<x 0)
) O
Te) Te)
E patti E
= Pl BOOS =

o1514131211109 8765
—Log;9[Concentration] [M]

Fic. 2. Concentration-dependent effect of SAP-I on cAMP and cGMP concentrations and respiration of
Hemicentrotus pulcherrimus spermatozoa. See [ref. 73] for experimental details.

362 N. SuzuKI

©
{4
0)
At)
Sh
=)
@)

'* Guanylate ¥
, cyclase

<S
a
0.0
kinase &
ied irene
= = |iSilk
* - We =
’ nw
ae

Whole sperm, sperm tails and
sperm heads were incubated
with or without SAP-I(1.0nM)
in ASW (pH8.2)

Fic. 3.

SAP-| concentrations (nM)
10 10 100

Whole sperm were incubated with
or without various concentrations
of SAP-I in ASW (pH8.2)

1000 atrees
by Whole sperm

17.

Sperm tails ;

Effect of SAP-I on the electrophoretic mobility of a high molecular weight protein of Hemicentrotus

pulcherrimus spermatozoa. See [ref. 78] for experimental details.

jelly layer of Arbacia punctulata or SAP-IIA,
without the corresponding appearance of radioac-
tivity at 150 kDa [103, 107-108]. The 160 kDa and
150 kDa proteins have proven to be two forms of
the same protein and identified as guanylate cyc-
lase [104]. Loss of phosphate from guanylate
cyclase and change in the molecular form which is
dependent upon external Na* and Ca** concen-
trations result in a large decrease in the specific
activity of the enzyme [95-96]. Therefore, sperm-
activating peptides appear to cause an initial tran-
sient activation and subsequent inactivation of
guanylate cyclase [102]. The mechanism in detail,
however, remains to be elucidated.

Guanylate cyclase of Arbacia punctulata sper-
matozoa has been proven to be the receptor for
SAP-IIA but not for SAP-I [98-101, 109]. Dan-
gott and Garbers reported that the receptor for
SAP-I on Strongylocentrotus purpuratus spermato-
zoa is a 77 kDa protein which has no known
enzyme activity [110-112]. Receptors for SAP-
IIB, SAP-III and SAP-IV have yet to be identified.

FUCOSE SULFATE GLYCOCONJUGATE

It has long been believed that a fucose sulfate
polysaccharide or a fucose-sulfate-rich complex in
the jelly layer of sea urchins is the only substance
responsible for the induction of the acrosome
reaction in sea urchin spermatozoa [9, 11-16].
Recently, my collaborators and I demonstrated
that full induction of the acrosome reaction in
Hemicentrotus pulcherrimus spermatozoa with a
fucose sulfate glycoconjugate requires a specific
co-factor, i.e., sperm-activating peptide I (Gly-
Phe-Asp-Leu-Asn-Gly-Gly-Gly-Val-Gly) [17-18,
65]. The acrosome reaction of sea urchin sperma-
tozoa is accompanied by Na‘ and Ca** influx, and
H+ and K* efflux [44, 113-128]. Activation of sea
urchin spermatozoa by sperm-activating peptides
is also accompanied by these ionic changes which
cause an increased intracellular pH and trigger
sperm plasma membrane depolarization [82, 129-
130].

In 1952, Dan first discovered using sea urchin
gametes that the acrosome reaction is essential for
fertilization and that the factor responsible for the

Egg Jelly Molecules

acrosome reaction exists in an extracellular matrix
called the jelly layer [1]. Since then many investi-
gators have made tremendous efforts to identify
this factor and to clarify the mechanism of the
reaction [2-16]. Ishihara and Dan reported that
treatment of the solubilized jelly layer of
Hemicentrotus pulcherrimus with trypsin and pro-
nase results in a reduction of the acrosome reac-
tion-inducing ability to about half of that with the
untreated jelly [5]. SeGall and Lennarz isolated a
fucose sulfate-containing substance containing less
than 30% protein from the solubilized jelly layer of
four species of sea urchin by gel filtration and
ion-exchange chromatography [13]. They demons-
trated that the acrosome reaction-inducing capac-
ity of the substance resides solely in the fucose
sulfate polysaccharide, although the inducing
activity decreases during purification. They con-
cluded that this decrease in activity is due to the
loss of Ca?* during purification. Garbers et al.
reported that a fucose-sulfate-rich complex
purified from the solubilized jelly layer of Strong-
lyocentrotus purpuratus is capable of elevating
sperm cAMP concentrations and inducing the

TABLE 3.

363

acrosome reaction [16]. They also demonstrated
that NaOH or pronase treatment of the complex
results in significant reductions in both acrosome
reaction-inducing potency and maximal ability of
the complexes to elevate cAMP concentrations.
The fucose sulfate glycoconjugate purified from
the solubilized jelly layer of Hemicentrotus pul-
cherrimus induces the acrosome reaction in
Hemicentrotus pulcherrimus spermatozoa, but the
potency is about half that of the unfractionated
jelly (Table 3) [18, 65]. When a sperm-activating
peptide-free macromolecular fraction is prepared
from the solubilized jelly layer of Hemicentrotus
pulcherrimus which is composed of a sialogly-
coprotein and a fucose sulfate glycoconjugate, the
rates of the acrosome reaction induced by the
fraction are also about half that of the original
unfractionated jelly [17]. The additon of synthetic
SAP-I to the fraction or the fucose sulfate glyco-
conjugate increases the rate of the acrosome reac-
tion to a level comparable with that of the unfrac-
tionated jelly at all pHs tested. SAP-I promotes
the acrosome reaction with fucose sulfate glyco-
conjugate in Hemicentrotus pulcherrimus sperma-

The rate of acrosome reaction in Hemicentrotus pulcherrimus spermatozoa

in the presence or absence of SAP-I treated with crude jelly, sialoglycoprotein,
FSG, pronase-digested FSG and carboxymethylated FSG

Percentage of Reacting Spermatozoa

Exp. 1 Exp. 2
Crude jelly 90% 87%
FSG + none 32 45
+SAP-I 75 81
Pronase-digested FSG
+ none 15
+SAP-I 49
Carboxymethylated FSG
+ none 24
+SAP-I 54
Sialoglycoprotein
+ none 3
+SAP-I 5
ASW alone 3 2

Spermatozoa were incubated in 0.5 ml of ASW or ASW containing crude jelly [62.5 (Exp.
1) or 100 (Exp. 2) nmol fucose/ml], sialoglycoprotein (100 nmol sialic acid/ml), fucose
sulfate gltcoconjugate (FSG) [62.5 (Exp. 1) or 100 (Exp. 2) nmol fucose/ml], pronase-
digested FSG (62.5 nmol fucose/ml), and carboxymethylated FSG (100 nmol fucose/ml)

with or without SAP-I (0.5 4M).

364 N. Suz

tozoa in a concentration-dependent manner with a
half-maximal rate of 4x10~'°M which is almost
the same concentration of peptide needed to cause
half-maximal stimulation of sperm respiration [18].
Since SAP-I alone does not induce the acrosome
reaction and sialoglycoprotein, another large
molecular weight component of the jelly layer,
also does not show any appreciable potency for
induction of the acrosome reaction, we think that

UKI

the major substance responsible for induction of
the acrosome reaction is the fucose sulfate glyco-
conjugate and SAP-I promotes the induction of
the acrosome reaction by acting as a co-factor [65].

The fucose sulfate glycoconjugate isolated from
the solubilized jelly layer of Hemicentrotus pul-
cherrimus contains a large amount of protein [65].
The proteins associated with fucose sulfate can not
be removed even when the glycoconjugate is dis-

| | WW
4 74 7
Intact FSG — FUCOSE 46 ge
\ °—~ PROTEIN : a me
— A280
150
—
oS)
3
2ou;
ke
S| =
= a;-019
25| :
5 g
= 10} ©
x 10 2
Q a
© 5+ es
P i: 2
0 2a Loo~
5 50 7.5 10.0 125 15.0
ELUTION VOLUME (ml)
| Vt
Carboxymethylated | bis
arboxymethylate
FSG ie
ale
-— FUCOSE | = ©
e— PROTEIN 203 re
— A280 | ZZ). S
Ww Zz
uF) FE sa
iS (oat faa
g Alene
S70); 110 1p
Ww x
O
© oF
=)
Le
o! 0 §.00

B40) 7) 00) 2G)
ELUTION VOLUME (ml)

Fic. 4. HPLC and SDS-PAGE of intact an
gate isolated from the solubilized jelly
sulfate glycoconjugate is abbreviated
perimental details.

15.0

d carboxymethylated fucose sulfate glycoconju-
layer of Hemicentrotus pulcherrimus. Fucose
as FSG in the figure. See [ref. 65] for ex-

Egg Jelly Molecules 365

solved in 7 M urea or 0.1% SDS and chromatog-
raphed on a Sepharose CL-2B or a TSK G-6000
PW column (Fig. 4). Furthermore, no protein
band is detectable on the electrophoretogram of
SDS-polyacrylamide gel electrophoresis (SDS-
PAGE) of the intact fucose sulfate glycoconjugate
without a reducing agent such as _ 2-
mercaptoethanol. These results suggest that the
protein may be covalently bound to the polysac-
charide portions of the glycoconjugate. When the
glycoconjugate is dissolved in an SDS solution
containing 2-mercaptoethanol and subjected to
SDS-PAGE in the presence of 2-mercaptoethanol,
two major (258 kDa and 238 kDa) and one minor
(120 kDa) protein bands appeard. A reduced and
carboxymethylated fucose sulfate glycoconjugate
(CmFSG) separated into two major (260 kDa and
240 kDa) and two minor (140 kDa and 135 kDa)
protein bands on an SDS-gel despite the absence
of 2-mercaptoethanol (Fig. 4). When CmFSG was
subjected to HPLC on a TSK G-6000 PW column,
a fucose-containing material, which possesses ab-
out 30% of the protein originally associated with
the fucose sulfate glycoconjugate, was separated
from the rest of the proteins. These proteins
separated into 260 kDa, 240 kDa, 140 kDa and 135
kDa components by SDS-PAGE. This indicates
that the fucose sulfate glycoconjugate consists of
fucose sulfate polymers bound to a core protein
and several other protein components associated
with each other through disulfide bounds. In order
to confirm the presence of disulfide bonds in the
fucose sulfate glycoconjugate, free sulfhydryl
groups in the fucose sulfate glycoconjugate are first
treated with non-radioactive iodoacetic acid,
which results in the carboxymethylation of about
half of the cysteine residues present in the original
glycoconjugate, and they are then reduced and
carboxymethylated with ['*C] iodoacetic acid. The
resulting labeled carboxymethylated glyconconju-
gate was analyzed by SDS-PAGE, and showed
that protein bands corresponding to 140 kDa and
135kDa _ proteins have strong radioactivity
although weaker radioactivity was detected in the
260 kDa, 240kDa, 110kDa and 87kDa protein
bands. In amino acid analyses of protein compo-
nents isolated from CmFSG, a 140 kDa and 135
kDa protein fraction contained the largest mol%

of carboxymethylated cysteine. These results sug-
gest that 140 kDa and 135 kDa proteins, which are
probably the same protein(s) detected in a 120
kDa protein band in SDS-PAGE of the intact
fucose sulfate glycoconjugate with 2-
mercaptoethanol have more sulfhydryl groups
than other protein components available for dis-
ulfide bond formation and may serve to make
bridge between the polysaccharide bound core
protein and other protein components.

Complete carboxymethylation of cysteine re-
sidues in the major acrosome reaction-inducing
jelly molecules, the fucose sulfate glycoconjugate,
results in the release of most of the constituent
proteins from the glycoconjugate and about a 50%
decrease in the acrosome reaction-inducing capac-
ity of the glycoconjugate [65]. When about 70% of
the constituent proteins is removed from the fu-
cose sulfate glycoconjugate by pronase digestion,
the resulting fucose-rich glycoconjugate loses more
than 50% of the acrosome reaction-inducing activ-
ity of the original untreated glycoconjugate (Table
3) [18]. In both cases, the decreased potency could
be partially restored by the addtion of SAP-I,
monensin, 3-isobutyl-1-methylxanthine (IBMX)
which is a inhibitor for cyclic nucleotide phospho-
diesterase or a Ca**-ionophore, A23187 [18].
Fucoidan, a polymer of L-fucose-4-sulfate which is
a major sugar component in the fucan sulfate
isolated from the solbilized jelly layer of
Hemicentrotus pulcherrimus, does not appreciably
induce the acrosome reaction of Hemicentrotus
pulcherrimus spermatozoa [65]. These results sug-
gest that the protein moiety of fucose sulfate
glycoconjugate is involved in the acrosome reac-
tion although the mechanism remains unclear.

FUTURE PROSPECTS

It is known that the acrosome reaction is depen-
dent on increased intracellular pH [29-30]. SAP-I
increases an intracellular Ca** which is coupled to
Na? entry and induces a rise in intracellular pH in
Strongylocentrotus purpuratus spermatozoa [129-
130]. Thus, we may presume that SAP-I promotes
an acrosome reaction with fucose sulfate glyco-
conjugate by increasing intracellular pH in sperm
cells.

366 N. Suzuki

A fucose-sulfate-rich complex obtained from the
solubilized jelly layer of Strongylocentrotus pur-
puratus causes Ca**-accumulation, elevates
cAMP and induces the acrosome reaction [16, 44,
114, 131-134]. In connection with this, it should be
noted that monoclonal antibodies against a 210
kDa glycoprotein purified from Strongylocentrotus
purpuratus sperm plasma membrane cause an in-
crease in intracellular Ca** of Strongylocentrotus
purpuratus spermatozoa and induce the acrosome
reaction when intracellular pH is increased with
NH,Cl [118, 135-139]. Antibodies induce the
cAMP-dependent phosphorylation of sperm his-
tone H1 as occurs upon treatment of spermatozoa
with the solubilized jelly layer of Strongylocentro-
tus purpuratus [91-93]. Similar results are pre-
sented by Sendai ef al. using sea urchin gametes of
Strongylocentrotus intermedius [38-39]. However,
an ionophore A23187 which is known to induce the
acrosome reaction and acts as a co-factor of a
pronase-digested fucose sulfate glycoconjugate for
induction of the acrosome reaction neither induces
histone H1 phosphorylation nor increases cAMP
concentrations in sea urchin spermatozoa [9, 18,
92, 140]. This supports the idea which suggests
that cAMP is not directly involved in the acrosome
reaction. Since a fucose-sulfate-rich complex
obtained from Strongylocentrotus purpuratus jelly
induces the elevation of inositol trisphosphate and
the elevation is dependent on external Ca’*, the
products generated from stimulated PI turnover in
sea urchin spermatozoa may be potential second
messengers in induction of the acrosome reaction
[141].

Sea urchin gametes have long been used for
studying fertilization and development [3, 29, 43,
64, 84, 127, 142-152]. It is not an exaggeration to
say that sea urchin are one of the most extensively
used experimental animals in basic biology.
However, we must realize that we do not know
much about the substances involved in sea urchin
gamete interaction. An understanding of this
would not only lead to an understanding of fert-
lization in mammals and other animals, but it
would serve as a valuable indicator of mechanisms
which can be anticipated in cells other than ga-
metes. Sperm-activating peptides cause many
biochemical responses in sea urchin spermatozoa

such as increases in cGMP and cAMP concentra-
tions, induction of ionic exchanges across the
plasma membrane and increases in intracellular
pH. The mechanism of action of the peptides
resembles in many aspects that of the atrial nat-
riuretic peptides, although the mechanism has yet
to be described in detail [153-156]. Since de Bold
et al. suggested that a factor released from the atria
of the heart evokes a variety of physiological
responses affecting cardiovascular homeostasis, in-
tensive studies have been carried out in an attempt
to identify the factor and a receptor(s) for the
factor [153-159]. Recently, the structure of a
receptor for atrial natriuretic peptides was deter-
mined by Chinkers et al. as a result of advanced
research on a sperm-activating peptide receptor
[160]. This serves as an excellent example for
denonstrating that the results of research in one
field can be applied to solving the problems in
other fields.

The receptor for the fucose sulfate glycoconju-
gate, which is the major substance responsible for
induction of the acrosome reaction has yet to be
isolated. Study of the receptor protein appears to
have great significance because questions have
arisen as to whether or not this protein could also
represent a calcium channel. Parallel to the isola-
tion and characterization of the receptor, the func-
tion of primary and secondary messengers will be
an important future area of research, much as in
other cells.

ACKNOWLEDGMENTS

The research described in this article was conducted at
Teikyo University School of Medicine and Note Marine
Laboratory, Kanazawa University with the financial sup-
port of the Ministry of Education, Science and Culture of
Japan (Nos. 57540425, 58540474, 59540471, 60480022,
62540458 and 63480021). The author would like to thank
Drs. Kohji Nomura, Masaaki Yamaguchi, Saburo Isaka,
Toshifumi Takao and Yasutsugu Shimonishi for their
kind and helpful collaboration. He wishes to express his
appreciation to graduate students in his laboratory for
their hard work and assistance, and to Miss Hiroko
Kajiura of the National Institute for Basic Biology who
performed the sequencing with an automated sequencer.
He is also indebted to Prof. Chitaru Oguro of Toyama
University for help with scanning electron microscopy,
Prof. Yoshitaka Nagahama of the National Institute for
Basic Biology and the staff of the Radioisotope Center,
Kanazawa University, for facilitating the use of equip-

Egg Jelly Molecules

ment for determining cyclic nucleotide concentrations.
He is grateful to the staff members of the Sesoko Marine
Science Center of the University of Ryukyus and of the
Asamushi Marine Biological Station of Tohoku Universi-
ty for the time and energy they spent collecting sea
urchins for this research.

BWN Re

23
24

REFERENCES

Dan, J. C. (1952) Biol. Bull., 103: 43-66.

Dan, J. C. (1954) Biol. Bull., 107: 335-349.
Dan, J. C. (1956) Int. Rev. Cytol., 5: 365-393.
Dan, J. C., Ohori, Y. and Kushida, H. (1964) J.
Ultrastruct. Res., 11: 508-524.

Ishihara, K. and Dan, J. C. (1970) Develop.
Growth and Differ., 12: 179-188.

Tilney, L. G., Hatano, S., Ishikawa, H. and
Mooseker, M. S. (1973) J. Cell Biol. , 59: 109-126.
Summers, R. G. and Hylander, B. L. (1975) Exp.
Cell Res., 93: 63-68.

Summers, R. G., Hylander, B. L., Colwin, L. H.
and Colwin, A. L. (1975) Am. Zool., 15: 523-551.
Decker, G. L., Joseph, D. B. and Lennarz, W. J.
(1976) Develop. Biol., 53: 115-125.

Collins, F. and Epel, D. (1977) Exp. Cell Res.,
106: 211-222.

Kinsey, W. H., SeGall, G. K. and Lennarz, W. J.
(1979) Develop. Biol., 71: 49-59.

Kinsey, W. H., Rubin, J. A. and Lennarz, W. J.
(1980) Develop. Biol., 74: 245-250.

SeGall, G. K. and Lennarz, W. J. (1979) Develop.
Biol., 71: 33-49.

SeGall, G. K. and Lennarz, W. J. (1981) Develop.
Biol., 86: 87-93.

Ishihara, K., Oguri, K. and Taniguchi, H. (1973)
Biochim. Biophys. Acta, 320: 628-634.

Garbers, D. L., Kopf, G. S., Tubb, D. J. and
Olson, G. (1983) Biol. Reprod., 29: 1211-1220.
Yamaguchi, M., Niwa, T., Kurita, M. and Suzuki,
N. (1988) Develop. Growth and Differ., 30: 159-
167.

Yamaguchi, M., Kurita, M. and Suzuki, N. (1989)
Develop. Growth and Differ., 31: 233-239.

Isaka, S., Akino, M., Hotta, K. and Kurokawa, M.
(1968) Exp. Cell Res., 54: 247-249.

Hotta, K., Hamazaki, H., Kurokawa, M. and
Isaka, S. (1970) J. Biol. Chem., 245: 5434-5440.
Hotta, K., Kurokawa, M. and Isaka, S. (1970) J.
Biol. Chem., 245: 6307-6311.

Isaka, S., Hotta, K. and Kurokawa, M. (1970)
Exp. Cell Res., 59: 37-42.

Bedford, J. M. (1976) Adv. Biosci., 4: 35-50.
Yanagimachi, R. and Usui, N. (1974) Exp. Cell
Res., 89: 161-174.

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41
42

43

44

45

46

47

48

49

367

Baccitti, B. and Afzelius, B. A. (1976) The Biolo-
gy of the Sperm Cell. In “Monographs in Develop-
mental Biology”. Ed. by A. Wolsky, vol. 10, S.
Karger, Basel and New York.

Cornett, L. E. and Meizel, S. (1978) Proc. Natl.
Acad. Sci. USA, 75: 4954-5958.

Saling, P. M., Sowinski, J. and Storey, B. T. (1979)
J. Exp. Zool., 209: 229-238.

Uno, Y. and Hoshi, M. (1978) Science, 200: 58-
59.

Shapiro, B. M. and Eddy, E. M. (1980) Int. Rev.
Cytol., 66: 257-302.

Shapiro, B. M., Shackmann, R. W. and Gabel, C.
A. (1981) Ann. Rev. Biochem., 50: 815-843.
Nishiyama, I., Sasaki, H., Matsui, T. and Hoshi,
M. (1985) Develop. Growth and Differ., 27: 461-
468.

Matsui, T., Nishiyama, I., Hino, A. and Hoshi, M.
(1986) Develop. Growth and Differ. 28: 339-348.
Matsui, T., Nishiyama, I., Hino, A. and Hoshi, M.
(1986) Develop. Growth and Differ., 28: 349-357.
Matsui, T., Nishiyama, I., Hino, A. and Hoshi, M.
(1986) Develop. Growth and Differ., 28: 359-368.
Nishiyama, I., Matsui, T., Yasumoto, T., Oshio, S.
and Hoshi, M. (1986) Develop. Growth and Dif-
fer., 28: 443-448.

Hoshi, M., Matsui, T., Nishiyama, I., Fujimoto,
Y. and Ikekawa, N. (1986) Adv. Invertebrate
Reprod., 4: 275-282.

Nishiyama, I., Matsui, T. and Hoshi, M. (1987)
Develop. Growth and Differ., 29: 161-169.
Nishiyama, I., Matsui, T., Fujimoto, Y., Ikekawa,
N. and Hoshi, M. (1987) Develop. Growth and
Differ., 29: 171-176.

Fujimoto, Y., Yamada, T., Ikekawa, N.,
Nishiyama, I., Matsui, T. and Hoshi, M. (1987)
Chem. Pharm. Bull., 35: 1829-1832.

Wasserman, P. M. (1987) Ann. Rev. Cell Biol., 3:
109-142.

Wasserman, P. M. (1987) Science, 235: 553-560.
Wasserman, P. M. (1988) Ann. Rev. Biochem.,
57: 415-442.

Garbers, D. L. (1989) Ann. Rev. Biochem., 58:
719-742.

Kopf, G. S. and Garbers, D. L. (1980) Biol.
Reprod., 22: 1118-1126.

Aketa, K., Yoshida, M., Miyazaki, S. and Ohta, T.
(1979) Exp. Cell Res., 123: 281-284.

Levin, Y., Pecht, M., Goldstein, L. and Katch-
alski, E. (1964) Biochemistry, 3: 1905-1913.
Goldstein, L., Levin, Y. and Katchalski, E. (1964)
Biochemistry, 3: 1913-1919.

Vasseur, E. and Hagstrom, B. (1946) Ark. Zool.,
37A, 17: 1-17.

Rothschild, Lord (1958) J. Exp. Biol., 33: 155-

368

50
51
52
53
54
55
56
57
58
59
60
61

62

63

64

65

66
67

68

69

70

71

72

73

74

173.

Ohtake, H. (1976) J. Exp. Zool., 198: 303-312.
Ohtake, H. (1976) J. Exp. Zool., 198: 313-322.
Lillie, F. R. (1913) J. Exp. Zool., 14: 515-574.
Carter, G. S. (1931) J. Exp. Biol., 8: 176-193.
Cohn, E. J. (1918) Biol. Bull., 34: 167-218.
Gray, J. (1928) J. Exp. Biol., 5: 337-344.

Gray, J. (1928) J. Exp. Biol., 5: 362-365.
Rothschild, Lord (1956) Vie Milieu, 7: 405-412.
Hathaway, R. R. (1963) Biol. Bull., 125: 486-498.
Kopf, G. S., Tubb, D. J. and Garbers, D. L. (1979)
J. Biol. Chem., 254: 8554-8560.

Hansbrough, J. R. and Garbers, D. L. (1981) J.
Biol. Chem., 256: 1447-1452.

Hansbrough, J. R. and Garbers, D. L. (1981) J.
Biol. Chem., 256: 2235-2241.

Suzuki, N., Nomura, K., Ohtake, H. and Isaka, S.
(1981) Biochem. Biophys. Res. Commun., 99:
1238-1244.

Garbers, D. L., Watkins, H. D., Tubb, D. J.,
Smith, A. C. and Misono, K. S. (1982) J. Biol.
Chem., 257: 2734-2737.

Suzuki, N. (1989) In “Bioorganic Marine Chemis-
try”. vol. 3, Ed. by P. J. Scheuer, Springer-Verlag,
Berlin, pp. 47-70.

Shimizu, T., Kinoh, H., Yamaguchi, M. and Suzu-
ki, N. (1990) Develop. Growth and Differ., in
press.

Shimizu, T. and Suzuki, N. unpublished data.
Nomura, K., Suzuki, N.,Ohtake, H. and Isaka, S.
(1983) Biochem. Biophys. Res. Commun., 117:
147-153.

Suzuki, N., Ohizumi, Y., Yasumasu, I. and Isaka,
S. (1984) Develop. Growth and Differ., 26: 17-24.
Suzuki, N., Shimonura, H., Radany, E. W.,
Ramarao, C. S., Bentley, J. K. and Garbers, D. L.
(1984) J. Biol. Chem., 259: 14874-14879.
Shimomura, H., Suzuki, N. and Garbers, D. L.
(1986) Peptides, 7: 491-495.

Suzuki, N., Kurita, M., Yoshino, K., Nomura, K.
and Yamaguchi, M. (1987) Zool. Sci., 4: 649-657.
Suzuki, N., Kajiura, H., Nomura, K., Garbers, D.
L., Yoshino, K., Kurita, M., Tanaka, H. and
Yamaguchi, M. (1988) Comp. Biochem. Physiol.,
89B: 687-693.

Suzuki, N., Yoshino, K., Kurita, M., Nomura, K.
and Yamaguchi, M. (1988) Comp. Biochem. Phy-
siol., 90C: 305-311.

Yoshino, K., Kajiura, H., Nomura, K., Takao, T.,
Shimonishi, Y., Kurita, M., Yamaguchi, M. and
Suzuki, N. (1989) Comp. Biochem. Physiol., 94B:
739-751.

Yoshino, K., Kurita, M., Yamaguchi, M.,
Nomura, K., Takao, T., Shimonishi, Y. and Suzu-
ki, N. (1990) Comp. Biochem. Physiol., 95B: 423-

N. SuzuKI

76

77

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

429.

Suzuki, N., Hoshi, M., Nomura, K. and Isaka, N.
(1982) Comp. Biochem. Physiol., 72A: 489-495.
Suzuki, N., Yoshino, K., Kurita, M., Yamaguchi,
M. and Amemiya, S. (1988) In “Echinoderm
Biology”. Ed. by R. D. Burke, P. V. Mladenov, P.
Lambert and R. D. Parsley, A. A. Balkema,
Rotterdam, pp. 213-218.

Suzuki, N., Kurita, M., Yoshino, K. and Yama-
guchi, M. (1987) Zool. Sci., 4: 641-648.

Smith, A. C. and Garbers, D. L. (1984) In
“Biochemistry of Metabolic Processes”. Ed. by D.
L. F. Lennon, F. W. Stratman and R. N. Zhalten,
Elsevier/North-Holland, New York, pp. 15-28.
Hansbrough, J. R. and Garbers, D. L. (1981) In
“Adv. Enzyme Regulation”. vol. 19, Ed. by G.
Weber, Pergamon Press, New York, pp. 351-376.
Ovchinnikov, Yu. A., Ivanov, V. T. and Shkrob,
A. M. (1974) In “Membrane-Active Complex-
ones”. B. B. A. Library, vol. 12, Elsevier Scientific
Publ. Co., Amsterdam/Oxford/New York, pp.
331-405.

Repaske, D. L. and Garbers, D. L. (1983) J. Biol.
Chem., 258: 6025-6029.

Nuccitelli, R. and Deamer, D. W. (eds.) (1982)
Intracellular pH: Its Measurement, Regulation,
and Utilization in Cellular Functions. Alan R. Liss,
Inc., New York.

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

Garbers, D. L. and Hardman, J. G. (1975) Nature,
257: 677-678.

Garbers, D. L., Hardman, J. G. and Rudolph, F.
G. (1974) Biochemistry, 13: 4166-4171.

Garbers, D. L., Watkins, H. D., Tubb, D. J. and
Kopf, G. S. (1978) Adv. Cyclic Nucleotide Res. , 9:
583-595.

Sano, M. (1976) Biochim. Biophys. Acta, 172: 20-
30.

Swarup, G. and Garbers, D. L. (1982) Biol.
Reprod., 2: 953-960.

Wells, J. N. and Garbers, D. L. (1976) Biol.
Reprod., 15: 46-53.

Green, G. R. and Poccia, D. L. (1985) Develop.
Biol., 108: 235-245.

Porter, D. C. and Vacquier, V. D. (1986) De-
velop. Biol., 116: 203-212.

Porter, D. C., Moy, G. W. and Vacquier, V. D.
(1988) J. Biol. Chem., 263: 2750-2755.

Radany, E. W., Gerzer, R. and Garbers, D. L.
(1983) J. Biol. Chem., 258: 8346-8351.
Ramarao, C. S. and Garbers, D. L. (1985) J. Biol.
Chem., 260: 8390-8396.

Ramarao, C. S. and Garbers, D. L. (1988) J. Biol.
Chem., 263: 1524-1529.

97

98

99

100

101

102

103

104

105

106

107

108
109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

Egg Jelly Molecules

Thorpe, D. S. and Garbers, D. L. (1989) J. Biol.
Chem., 264: 6545-6549.

Bentley, J. K. and Garbers, D. L. (1986) Biol.
Reprod., 34: 413-421.

Bentley, J. K. and Garbers, D. L. (1986) Biol.
Reprod., 35: 1249-1259.

Bentley, J. K., Shimonura, H. and Garbers, D. L.
(1986) Cell, 45: 281-288.

Bentley, J. K., Tubb, D. J. and Garbers, D. L.
(1986) J. Biol. Chem., 262: 14859-14862.
Garbers, D. L. (1988) In “ISI Atlas of Science:
Biochemistry/1988”. pp. 120-126.

Ward, G. E. and Vacquier, V. D. (1983) Proc.
Natl. Acad. Sci. USA., 80: 5578-5582.

Ward, G. E., Garbers, D. L. and Vacquier, V. D.
(1985) Science, 227: 768-770.

Ward, G. E., Moy, G. W. and Vacquier, V. D.
(1986) J. Cell Biol., 103: 95-101.

Harumi, T., Kurita, M., Yamaguchi, M. and Suzu-
ki, N. unpublished data.

Vacquier, V. D. and Moy, G. W. (1986) Biochem.
Biophys. Res. Commun., 137: 1148-1152.
Vacquier, V. D. (1986) TIBS., 11: 77-81.
Shimomura, H., Dangott, L. J. and Garbers, D. L.
(1986) J. Biol. Chem., 261: 8719-8728.

Dangott, L. J. and Garbers, D. L. (1984) J. Biol.
Chem., 259: 13712-13716.

Dangott, L. J. and Garbers, D. L. (1987) Ann. N.
Y. Acad. Sci., 513: 274-283.
Dangott, L. J., Jordan, J. E., Bellet, R. A. and
Garbers, D. L. (1989) Proc. Natl. Acad. Sci.
USA., 86: 2128-2132.

Schackmann, R. W., Eddy, E. M. and Shapiro, B.
M. (1978) Develop. Biol., 65: 483-495.

Garbers, D. L. and Kopf, G. S. (1980) Adv. Cyclic
Nucleotide Res., 13: 251-306.

Schackmann, R. W. and Shapiro, B. M. (1981)
Develop. Biol., 81: 145-154.

Schackmann, R. W., Christen, R. and Shapiro, B.
M. (1981) Proc. Natl. Acad. Sci. USA., 78: 6066-
6070.

Garcia-Soto, J. and Darszon, A. (1985) Develop.
Biol., 110: 338-345.

Trimmer, J. S. Trowbridge, I. S. and Vacquier, V.
D. (1985) Cell, 40: 697-703.

Garcia-Soto, J., Gonzalez-Marinez, N., Torre, L.
de De la and Darszon, A. Develop. Biol., 120:
112-120.

Towicharanont, P. and Shapiro, B. M. (1988) J.
Biol. Chem., 263: 6877-6883.

Sendai, Y., Ohta, T. and Aketa, K. (1989) De-
velop. Growth and Differ., 31: 459-466.

Sendai, Y. and Aketa, K. (1989) Develop. Growth
and Differ., 31: 467-473.

Gurrero, A. and Darszon, A. (1989) Biochim.

124

125

126

27)

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143
144

145

146

369

Biophys. Acta, 980: 109-116.

Lee, H. C., Johnson, C. and Epel, D. (1983)
Develop. Biol., 72: 126-137.

Schackmann, R. W., Christen, R. and Shapiro, B.
M. (1984) J. Biol. Chem., 259: 13914-13922.
Darszon, A., Garcia-Soto, J., Lievano, A., San-
chez, J. A. and Islas-Trejo, A. D. (1986) In “Ionic
Channels in Cells and Model Syatem”. Ed. by R.
Latorre, Plenum Press, New York, pp. 291-305.
Epel, D. (1978) Curr. Top. Develop. Biol., 12:
185-246.

Christen, R., Schackmann, R. W. and Shapiro, B.
M. (1983) Develop. Biol., 98: 1-14.
Schackmann, R. W. and Chock, P. B. (1986) J.
Biol. Chem., 261: 8719-8728.

Lee, H. C. and Garbers, D. L. (1986) J. Biol.
Chem., 261: 16026-16032.

Tubb, D. J., Kopf, G. S. and Garbers, D. L. (1978)
Biol. Reprod., 18: 181-185.

Watkins, H. D., Kopf, G. S. and Garbers, D. L.
(1978) Biol. Reprod., 18: 890-894.

Garbers, D. L., Tubb, D. J. and Kopf, G. S. (1980)
Biol. Reprod., 22: 526-532.

Garbers, D. L. (1981) J. Biol. Chem., 256: 620-
624.

Lopo, A. C. and Vacquier, V. D. (1980) Develop.
Biol., 79: 325-333.

Podell, S. B. and Vacquier, V. D. (1984) Exp. Cell
Res., 155: 467-476.

Podell, S. B. and Vacquier, V. D. (1985) J. Biol.
Chem., 260: 2715-2718.

Trimmer, J. S., Schackmann, R. W. and Vacquier,
V. D. (1986) Proc. Natl. Acad. Sci. USA., 83:
9055-9059.

Trimmer, J. S., Ebina, Y., Schackmann, R. W.,
Meinhof, C-G. and Vacquier, V. D. (1987) J. Cell
Biol., 105: 1121-1128.

Summers, R. G., Talbot, P., Keough, E. M.,
Hylander, B. L. and Franklin, L. E. (1976) J. Exp.
Zool., 196: 381-385.

Domino, S. E. and Garbers, D. L. (1988) J. Biol.
Chem., 263: 690-695.

Metze, C. B. (1967) In “Fertlization: Comparative
Morphology, Biochemistry and Immunology”. vol.
1, Ed. by C. B. Metze and A. Monroy, Academic
Press, New York, pp. 163-236.

Longo, F. G. (1973) Biol. Reprod., 9: 149-215.
Guidice, G. (1973) Developmental Biology of the
Sea Urchin Embryo. Academic Press, New York.
Horstadius, S. (1973) Experimental Embryology
of Echinoderms. Oxford Univ. Press, London/New
York.

Stearns, L. W. (1974) Sea Urchin Development,
Cellular and Molecular Aspects. Dowden, Hutch-
inson & Ross, Stroudsville.

370

147

148

149

150

151
152

153

154

Czihak, G. (1975) The Sea Urchin Embryo:
Biochemistry and Morphogenesis. Springer-
Verlag, Berlin.

Shapiro, B. M. (1977) Horizons Biochem. Bio-
phys., 4: 201-243.

Poste, G. and Nicolson, G. L. (eds.) (1978) Cell
Surface Rev., vol. 5, North Holland, Amsterdam.
Dirksen, E. R., Prescott, D. and Fox, C. H. (eds.)
(1978) Cell Reproduction. Academic Press, New
York.

Schatten, G. (1982) Int. Rev. Cytol., 79: 59-163.
Guidice, G. (1985) The Sea Urchin Embryo. A
Developmental Biology System. Springer-Verlag,
Berlin.

Kuno, T., Andresen, J. W., Kamisaki, Y., Wald-
man, S. A., Chang, L. Y., Saheki, S., Leitman, D.
C., Nakane, M. and Murad, F. (1986) J. Biol.
Chem., 261: 5817-5823.

Paul, A. K., Marala, R. B., Jaiswal, P. K. and
Sharna, P. K. (1987) Science, 235: 1224-1226.

N. SUZUKI

155

156

157

158

159

160

Takayanagi, R., Snajdar, R. M., Imada, T.,
Tamura, M., Pandey, K. N., Misono, K. S. and
Inagami, T. (1987) Biochem. Biophys. Res. Com-
mun., 144: 244-250.

Hamet, P., Trembly, J., Pang, S. C., Garcia, R.,
Thibault, G., Gutkowska, J., Cantin, M. and
Genest, J. (1984) Biochem: Biophys. Res. Com-
mun., 123: 515-527.

de Bold, A. J., Borenstein, H. B., Veress, A. T.
and Sonnenberg, H. (1981) Life Sci., 28: 89-94.
Mittal, C. K. and Murad, F. (1982) In “Handbook
of Experimental Pharmacology”. Ed. by J. A.
Nathanson and J. W. Kebabian, vol. 58, pp. 225-
260, Springer-Verlag, Berlin.

Kangawa, K. and Matsuo, H. (1984) Biochem.
Biophys. Res. Commun., 118: 131-139.
Chinkers, M., Garbers, D. L., Chang, M.-S.,
Lowe, D. G., Chin, H., Goeddel, D. V. and
Shultz, S. (1989) Nature, 338: 78-83.

ZOOLOGICAL SCIENCE 7: 371-376 (1990)

Regulation of Water Permeability of the Skin of
the Treefrog, Hyla arborea japonica

Hirosui NAKASHIMA and YosHIHISA KAMISHIMA!

Department of Biology, Faculty of Science, Okayama University,
Okayama 700, Japan

ABSTRACT— Water permeabaility of excised skin was investigated in Japanese treefrogs during the
non-breeding season. In normal frogs, water flowed inward across the ventral skin at a rate of 0.30
l/cm?/min, and outward at a rate of 0.0015 pl/cem?/min across the dorsal skin. The water flux across the
ventral skin was greatly stimulated when frogs were submitted to dehydration. The sympathomimetic
agent noradrenaline and the beta-adrenergic receptor agonist isoproterenol (IP) greatly increased the
the water flux across the ventral skin, but these had no effect on the dorsal skin. The beta-receptor
antagonist propranolol (PP) suppressed the enhanced water influx by dehydration or IP treatment.
However, PP had no effect on normal water permeation across the ventral skin. Ouabain, a specific
inhibitor of Na*, K*-ATPase, also suppressed the stimulatory effect of IP, whereas this inhibitor again
had no effect in normal water flux of the ventral skin. The adrenergic alpha-receptor agonist
phenylephrine had no effects on water permeation in either ventral or dorsal skin. Prolactin
administration for one week prior to the experimentation decreased water permeability significantly
across the ventral skin in both normal and stimulated preparations. These results all argue for the
presence of two types of water transport system in the ventral skin of the Japanese treefrog. The first is
mediated by the adrenergic beta-receptor and by Nat, K*-ATPase activity and may be activated by
dehydration. The second is not mediated by the adrenergic receptor and may function as the normal,

© 1990 Zoological Society of Japan

basal water permeation.

INTRODUCTION

Anuran amphibians are known not to take water
orally. They take the most of their water through
the skin [1-6]. Water permeation through the skin
is mainly confined to the pelvic patch [7-9] and can
be modified by various factors, including sym-
pathomimetic agents [10-15] and neurohypophy-
sial hormones [1, 2, 5, 7, 15]. Water permeation in
the skin is stimulated by dehydration [4, 11].
Water permeability differs according to both spe-
cies and season. Generally, terrestrially-adopted
species have more permeable skin than do aquatic
species [8, 9]. Water also evaporates readily from
normal frog skin [1, 3]. This evaporation helps to
cool the animal, and is known to be less in
terrestrially-adopted anurans such as toads and
treefrogs [16, 17]. These findings raise the para-

Accepted September 8, 1989
Received May 8, 1989
' To whom all correspondence should be addressed.

doxical question of how these anurans manage to
survive in hot and dry environments, where they
cool themseves at the cost of water which is
already scarce in their habitats.

The Japanese treefrogs are known to have sur-
vived long periods of drought, and have been
reported to accumulate glycosaminoglycans in
their epidermis following dehydration treatment.
The glycosaminoglycans might cause retardation
of water movement across the skin [18]. Prolactin
administration brought similar morphological
changes in the skin of these frogs. In the present
experiment we intended to find out regulating
facotors of water permeation in the Japanese treef-
rogs, especially in terms of neural and hormonal
control under the dehydrated condition.

MATERIALS AND METHODS

Japanese treefrogs, Hyla arborea japonica, were
collected from fields and kept for at least one week
in a terrartum in the laboratory prior to the

372 H. NAKASHIMA AND H. KAMISHIMA

experiments. The terrarium was maintained at
24°C under 12L-12D photoperiod. Frogs were fed
periodically with crickets and allowed access to
water adlibitum. Young adults weighing 1+0.3 g
were used. In these specimens no sexual differ-
ence was observed in water permeation between
the period November through February during
which time the present experiment was performed.

Water permeation of the skin was measured by
recording water movement between two small
chambers partitioned with an experimental skin.
Both chambers were connected with a filter holder
(Swinnex sx00 013 00, Millipore Corp.), on which
excised skin was mounted in the position of a filter.
The chamber facing the dermal side (the inside) of
the skin was filled with an airated, modified Ringer
solution (230mOsM) [19] and attached with a
graded pipette. The chamber facing the epidermal
side was filled with distilled water (DW). Net
water movement across the partitioned skin was
measured by recording changes of the meniscus in
the pipette. Dilute Ringer solution (20 mOsM)
was substituted for the DW in the outside bath, but
significant difference in water movement was not
detected. Thus, DW was used in all experiments
for the outside bath. The osmotic difference
between the two chambers was maintained at
approximately 230 mOsM throughout the experi-
ments.

To dehydrate frogs, they were kept in a chamber
in which relative humidity was maintained at 90%.
The formation of water droplets in the chamber
was carefully prevented. Frogs were neither fed
nor allowed to access to water during this treat-
ment. Dehydration treatment ended after approx-
imately 40 hr when the animal lost 30% of its
initial weight.

In the prolactin treatment, frogs were injected
daily with ovine prolactin (Sigma, 0.25 IU (8.1
pg)/gbw/day) in 2 ml of Ringer solution intraperi-
toneally for 7 days. During this treatment they
were kept under the normal conditions and fed as
per usual. As a control, another group of frogs
were injected with egg albumin (Nakarai Chem.
LTD) in the same protein concentration as pro-
lactin.

Five animals were used in each experiment and
the experiment was repeated three times. Since

the available filtrating area of the apparatus was
0.75 cm’, direct reading of the pipette was cali-
brated as a net water movement in l/cm* and
plotted in 10 min intervals for 100 min. In order to
compare water permeability of the experiments,
an average water flux (Jw) was calculated and
expressed in pl/cm?/min. Positive value in Jw
shows water movement from the epidermal side to
the dermal side and a negative value indicates the
reverse.

RESULTS

Hyla arborea japonica exhibited a fundamental
difference between water permeation of the dorsal
and ventral sides of their isolated skin (Fig. 1).
Under normal conditions, that is, immersing the
dermal side of the excised skin in Ringer solution
(230 mOsM) and the epidermal side in distilled
water, water flowed inwardly across the ventral
skin at a constant rate of 0.29 1 per cm* per min,
or Jw=+0.29. In the dorsal skin, however, water
permeation took place at a lower rate of Jw=
—0.015, which meant water movement from the
inside to the outside. The movement was against
the osmotic gradient.

When frogs were kept in the dehydration cham-
ber (24°C, 90% relative humidity) for approx-
imately 40 hr and had lost nearly 30% of their
initial weight, the ventral skin enhanced water
permeability (Fig. 1). The water flux was not
constant throughout the experiment. It lessened
over time, but the average rate over 100 min was
as high as Jw=+0.91, four times greater than that
in the control group. In dorsal skin, however, no
such change in water permeability was observed
during dehydration treatment.

When noradrenalin, one of the sympathomime-
tic agents, was added to the inside bath of the
normal preparation, the water flux increased to Jw
=1.52 in ventral skin, five times the control rate
(Fig. 2). Daily administration of prolactin for 7
days caused a significant decrease in water permea-
tion across the ventral skin (Fig. 2). The water
flux, Jw=-+0.08, was only one-fourth the control
value. Dehydration had a less pronounced effect
in the frogs pre-treated with prolactin. The rate of
water flux (Jw=+0.51) was less than 60% of that

Water Permeability of the Frog Skin 373

150
E
U
=
= 100
>
(e)
rm
2
lo}
>

50

-10

0 50 100
time in min

Fic. 1. Rates of the water permeation across the ex-

cised skin of the treefrog. Water flowed from
mucousa to serosa at a rate of 29 1/100 min across
the normal ventral skin (filled circle) and at a rate of
—1.5 41/100 min across the dorsal skin, which
showed a net water movement from the inside to the
outside (open circle). Dehydration of the frogs
caused a marked increase in the water flow across
the ventral skin, to a rate of 112 1/100 min (filled
triangle), but dehydration had no effects on the
dorsal skin (open triangle). The vertical bar shows
the standard margin of error.

of dehydrated normal frogs (Fig. 2). Stimulation
of water flux by noradrenalin was also significantly
depressed in the prolactin-treated frogs. The wa-
ter flux (Jw=+0.78) was one-half of the normal
skin treated with noradrenalin. When egg albumin
at the same protein concentration as the prolactin
was injected for the same term, no significant
change was observed (Fig. 2).

An adrenergic beta-receptor agonist, isoprotere-
nol (IP, 10~° M), enhanced the water flux (Jw= +
1.6) similarly to the noradrenalin treatment. Prop-
ranolol (PP, 10-° M), an adrenergic beta-receptor
antagonist, not only counteracted isoproterenol,
but also depressed the water flux induced by the

‘e
x
= 100
3
o
®
(eo)
3
50

50 100
time in min

Fic. 2. Hormonal effects on water permeation across
the excised ventral skin of the treefrog. The sym-
pathomimetic agent noradrenalin (10°-° M) added
to the inner experimental medium caused an en-
hancement on the water flow (open star). Daily
administration of prolactin (8.1 ug/day) for a week
decreased the water flow (filled circle). Prolactin
pretreatment also suppressed the water flow which
had previously been stimulated by dehydration
(filled triangle) and by noradrenalin administration
(filled star). Albumin administration serving as the
control for prolactin-treated group, caused no
effects on water flow (open circle). The vertical bar
shows the standard margin of error.

dehydration significantly to Jw= +0.61, two-thirds
the rate of dehydrated frogs (Fig. 3). However,
propranolol did not affect the water flux in normal
frogs at all. Isoproterenol administration to the
normal preparation caused an instantaneous rise of
the water flux to Jw=+1.6 (Fig. 3). However,
replacement of this medium by one containing
propranolol caused slow and gradual depression of
the water flux, which finally reached the normal
rate, Jw=+0.35, in 40min (Fig.3). Alpha-
adrenergic agents such as phenylephrine (alpha-
receptor agonist) or dibenamine (alpha-receptor

374

150

E

U

= 100

=

>

ie)

=

o

5

> 50

0 50 - 100
time in min
Fic. 3. Effects of adrenergic agents on water permea-

tion across the ventral skin of the treefrog. When
the adrenergic beta-receptor agonist isoproterenol
(10~° M) was added to the experimental medium,
the water flux increased markedly (filled star), while
the addition of the alpha-agonist phenynephlin
(10~°M) had no effect at all (open circle). An
adrenergic beta-antagonist propranolol (PP, 10°
M) couteracted the isoproterenol stimulation (IP).
Propranolol suppressed the enhancement induced
by dehydration (filled triangle). Propranolol,
however, had no effect on normal skin (open triang-
le). The vertical bar indicates the standard margin
of error.

antagonist) had no effect on the water flow across
the skin of either dehydrated or normal frogs.

Stimulation of the water flux by isoproterenol
was very strong. It persisted continuouslly for 60
min or more even after the agent had been washed
out from the medium (Fig. 4). However, pretreat-
ment of the skin with ouabain, a specific Na* , K*-
ATPase inhibitor, obstructed the stimulating effect
of isoproterenol completely. Ouabain, however,
did not affect the basal water flux in the normal
frogs.

H. NAKASHIMA AND H. KaMISHIMA

150

E
1°)
~
= 00
>
2
o
Lo)
3
50

in min

Fic. 4. Effect of ouabain on the isoproterenol-
stimulated water flow across the ventral skin of the
treefrog. Stimulation by isoproterenol for 20 min
caused a prolonged increase in the water flow (filled
star) even after the agent had been removed (—IP).
However, pretreatment with ouabain (107° M) for
40 min completely counteracted the stimulatory
effect of isoproterenol (filled triangle). Ouabain
treatment, however, did not have any effect on
normal skin (open circle).

DISCUSSION

It has previously been reported in American
treefrogs that water flux in the ventral skin is 10 to
20 times greater than that of the dorsal skin [8].
The present study shows an even more remarkable
dorso-ventral difference in water permeation
across the skin of Japanese treefrogs. When
excised skin was examined, water normally flowed
inwardly in great quantities across the ventral skin,
while it flowed outwardly in a small amount across
the dorsal skin. When the treefrogs were kept in
dehydrated conditions, water flux increased
markedly in the ventral skin, but no significant
change was observed in the dorsal skin. Similar
increase in water flux can be induced by adrenergic
beta-stimulation in various anurans [10-15].

Water Permeability of the Frog Skin 375

When dehydrated treefrogs, which had lost 30%
of their body weight, were allowed to take water,
they recovered 90% of their previous weight with-
in 10 min by absorbing the water across the ventral
skin (unpublished data). This rapid water absorp-
tion across the ventral skin seems to be regulated
by the adrenergic beta-receptor, since proprano-
lol, a beta-receptor antagonist, depressed en-
hancement of the water flux caused by dehydra-
tion. In the present experiment, the propranolol
depressed only the stimulated water flux across
ventral skin, and did not affect normal water flux
across either side of the skin. Thus, it is probable
that there are two types of water pathways in the
ventral side of the treefrog skin. The first pathway
is mediated by beta-receptor stimulation that be-
gins to function in cases of urgent water requir-
ment, such as dehydration, and is ouabain sensi-
tive. The second pathway is the basal one which
funtions under the normal condition and is not
affected by beta-adrenoceptor stimulation. Simil-
ary, two water transport systems have also been
reported in toads [10, 14]. In one of these studies
[14] alpha-receptors have been reported to inhibit
water flux. However, no effects of alpha-agents on
the water flow across the ventral skin of Japanese
treefrogs were observed in this study.

De Sousa et al. reported that ouabain did not
have any effect on the isoproterenol-stimulated
water flow in toad skin [10]. In the present
experiment, concomitant administration of oua-
bain and isoproterenol also failed to produce any
clear suppressive effect. This may be due to the
difference in modes of action of the two agents. As
mentioned above, the stimulating effect of isop-
roterenol appears instantaneously and persists for
a certain period even after it has been washed out,
while ouabain takes 30 to 40 min to produce the
suppressive effect. Therefore, it is likely that the
Na*, K*-ATPase activity is necessary for the
attainment of the beta-action.

Pretreatment with prolactin for one week re-
duced the water flux approximately 50% in the
ventral skin. Suppression by prolactin occurred
not only in the stimulated water flow but also with
the basal flow in the normal frogs. This prolactin
suppression seems to be caused by a mechanism
different from that of the suppressive agents men-

tioned above. Prolactin is generally known to
prevent osmotic water permeation in adult uro-
dales [5, 20], as well as in larval anurans [20] and
fresh water fish [19]. This is primarily due to the
mucous secretion on the integument under the
hormonal stimulation. The mucous not only func-
tions as the water-resistant coating of the skin, but
may also produce certain osmotic effects itself by
retaining electrolites in the coating, resulting in a
lesser osmotic gradient between inside and outside
of the animal body [18]. Indeed, frogs have been
reported to have increasing amount of prolactin
receptor in the skin during the breeding season
when they become aquatic [21]. Suppression of
water flux by prolactin pretreatment in the present
experiment also seems to be related to the mucous
secretion, which has previously been observed
histochemically in those frogs (not published). On
the other hand, extirpation of pars distalis or
removal of the whole pituitary in the toad Bufo
bufo depressed water flux which had been stimu-
lated by dehydration [22]. This, in a sense, con-
flicts with the present findings that administration
of prolactin depressed the stimulated water flow.
However, the hypophysectomy does not indicate
disappearance of prolactin alone in the animals.

In summary, the following results are character-
istic of this species: marked difference in dorsal
and ventral water permeation, particularly, re-
verse-directional water movement in ventral and
dorsal skin under normal conditions, and two
water pathways in the ventral skin, including a
regulatory system mediated by adrenergic beta-
receptor and Na‘, K*-ATPase, and a basal sys-
tem indifferent to these mediators. Both pathways
are regulated by prolactin.

REFERENCES

1 Deyrup, I. J. (1964) Water balance and kidney. In
“Physiology of the Amphibia”. Ed. by J. A. Moor,
Academic Press, New York, pp. 251-328.

2 Shoemaker, V. H. and Nagy, K. A. (1977)
Osmoregulation in amphibians and reptiles. Ann.
Rev. Physiol., 39: 449-471.

3 Deullman, W. E. and Trueb, L. (1986) Biology of
Amphibians. McGraw-Hill, New York, pp. 197-
228.

4 Bentley, P. J. and Yorio, T. (1979) Do frogs drink?

10

11

12

13

376

J. Exp. Biol., 79: 41-46.

Lorenz, C. A. and Bern, H. A. (1982) Prolactin and
osmoregulation in vertebrates. Neuroendocrinolo-
gy, 35: 292-304.

Bentley, P. J. (1982) Comparative Vertebrate En-
docrinology. 2nd. Ed., Cambridge Univ. Press,
Cambridge, pp. 303-337.

Hillyard, S. D. (1976) Variation in the effects of
antidiuretic hormone on the isolated skin of the
toad, Scaphiopus couchi. J. Exp. Zool., 195: 199-
206.

Yorio, T. and Bentley, P. J. (1977) Asymmetrical
permeability of the integument of treefrogs (Hyli-
dae). J. Exp. Biol., 67: 197-204.

McClanahan, L. and Baldwin, R. (1969) Rate of
water uptake through the integument of the desert
toad, Bufo punctatus. Comp. Biochem. Physiol., 28:
381-389.

De Sousa, R. C. and Grosso, A. (1982) Osmotic
water flow across the abdominal skin of the toad
Bufo marinus: effect of vasopressin and isoprena-
line. J. Physiol., 329: 281-296.

Yokota, S. D. and Hillman, S. S. (1984) Adrenergic
control of the anuran cutaneous hydroosmotic re-
sponse. Gen. Comp. Endocrinol., 53: 309-314.
Brown, D., Grosso, A. and de Sousa, R. C. (1980)
Isoproterenol-induced intramembrane particle
aggregation and water flux in toad epidermis.
Biochem. Biophys. Acta, 595: 158-164.

Hyllard, S. D. (1979) The effect of isoproterenol on
the anuran water balance response. Comp.
Biochem. Physiol., 62C: 93-95.

14

15

19

20

21

22

H. NAKASHIMA AND H. KAMISHIMA

Gamundi, S. S., Scheucher, A. and Coviello, A.
(1986) Alpha-2 adrenergic agonists inhibit basal and
stimulated osmotic water permeability in toad skin.
Comp. Biochem. Physiol., 84C: 199-203.

Elliot, A. B. (1968) Effect of adrenaline on water
uptake in Bufo. J. Physiol., 197: 87-88.

Wygoda, M. L. (1984) Low cutaneous evaporative
water loss in arboreal frogs. Physiol. Zool., 57: 329
337.

Withers, P. C., Hillman, S. S., Drewes, and Sokol,
O. M. (1982) Nitrogen excretion in sharp-nosed
reed frogs (Hyperolius nasutus: Anura, Hyperoli-
dae). J. Exp. Biol., 97: 335-343.

Shephard, K. I. (1981) The influence of mucus on
the diffusion of water across fish epidermis. Physiol.
Zool., 54: 224-229.

Johnsen, A. H. and Nielsen, R. (1984) Correlation
between cAMP in isolated frog skin epithelium and
stimulation of sodium transport and osmotic water
flow by antidiuretic hormone and phosphodiesterase
inhibitor. Gen. Comp. Endocrinol., 54: 144-153.
Loudi, B., Biciotti, M. and Viotto, B. (1982)
Cutaneous osmoregulation in Tritrus cristatus car-
nifex (Laur.) (Urodela). Gen. Comp. Endocrinol.,
46: 452-457.

D’Istria, Fasano, S. and Delrio, G. (1987) Prolactin
receptors in the male Rana esculenta. Gen. Comp.
Endocrinol., 68: 6-11.

Cristensen, C. U. and Jorgensen, C. B. (1972) Role
of pars distalis and pars nervosa of the hypophysis in
the water economy of the toad. Gen. Comp. Endoc-
rinol., 18: 169-174.

ZOOLOGICAL SCIENCE 7: 377-384 (1990)

Regulatory Actions of 5-Hydroxytryptamine and Some
Neuropeptides on the Heart of the African
Giant Snail, Achatina fulica Férussac

KatsuHIko Hori’, YASuo FuRUKAWA~ and Maxoto Kopayasur

Physiological Laboratory, Faculty of Integrated Arts and Sciences,
Hiroshima University, Hiroshima 730, Japan

ABSTRACT— Effects of putative neurotransmitters and modulators on the atrium preparation of the
African giant snail, Achatina fulica were observed to investigate the regulatory mechanisms of these
substances on the heart beat. Application of 5-hydroxytryptamine (5-HT) as well as stimulation of a
heart excitatory neuron, PON, resulted in the enhancement of the heart beat. The enhancement of the
beat was blocked by a 5-HT blocker, methysergide. A neuropeptide FMRFamide potentiated the
excitatory responses of the atrium to both PON stimulation and 5-HT application, whereas small
cardioactive peptide B (SCP ) depressed them.

A burst of impulses in the cerebral neuron, d-RCDN or d-LCDN, evoked excitatory response in the
heart excitatory neurons, TAN, TAN-2 and TAN-3, as well as in PON. It depolarized the membrane
and increased the frequency and duration of spikes in three TANs. Similar responses were elicited by
5-HT application. On the contrary, FARFamide hyperpolarized the membrane and shortened the spike
duration in three TANs and PON. The experiment of current measurement showed that 5-HT and

© 1990 Zoological Society of Japan

FMRFamide act antagonistically in the heart excitatory neurons.

INTRODUCTION

It has been known that several neurotransmit-
ters, such as acetylcholine (ACh) and 5-
hydroxytryptamine (5-HT), are involved in the
control of the molluscan heart beat, and that some
invertebrate neuropeptides show powerful mod-
ulatory actions in the synaptic transmission [1, 2].
The modes of action of the transmitters as well as
the modulators are not uniform in general but
quite variable depending on species and organs
[3-5].

In the African giant snail, Achatine fulica, sever-
al heart regulatory neurons have been identified in
the central nervous system [6, 7]. Among them,

Accepted September 11, 1989
Received August 11, 1989

" Present address: Biological Science Research Institute
Mitsui-Seiyaku, Mobara 297, Japan.

* Present address: Center for Neurobiology and Be-
havior, Columbia University, 722 West 168th Street,
New York, NY 10032, U.S.A.

° To whom all correspondence should be addressed.

>

two cerebral ganglion cells, the dorsal right cere-
bral distinct neuron (d-RCDN) and the dorsal left
cerebral distinct neuron (d-LCDN), produce a
slow depolarization and prolong the duration of
action potential in the periodically oscillating
neuron (PON), which is the most effective heart
excitor. The transmitter of the two cerebral
neurons is suggested to be 5-HT [8]. The direct
action of 5-HT on the heart has also been ex-
amined by using the isolated ventricle and dual
effects, enhancing and arresting, on the beat have
been demonstrated [9]. However, in Achatina
heart, responses to putative neurotransmitters and
modulators are not always similar between the
ventricle and atrium [10].

In the present study, to demonstrate the regula-
tory mechanisms on the heart beat, the effects of
application of several putative neurotransmitters
including neuropeptides and stimulation of heart
excitatory neurons on the isolated atrium prepara-
tion were observed. In addition, the mode of
action of the cerebral neurons (d-RCDN and D-
LCDN) on the heart excitors were also investi-
gated.

378 K. Hori, Y. FURUKAWA AND M. KoBAyYASHI

MATERIALS AND METHODS

The African giant snail, Achatina fulica Férus-
sac, which was captured in Okinawa, transported
by air to Hiroshima and bred in our laboratory at
24°C, was used. Circumoesophageal ganglia and
heart connected with the intestinal nerve were
dissected from the animal. The connective capsule
and the inner sheath covering the dorsal surface of
the cerebral ganglia and the right parietal ganglion
were completely removed by dissection. Most of
the ventricle was cut off leaving the atrium intact.
The preparation was pinned to the bottom of an
experimental chamber coated with silicone resin.
The chamber consisted of two compartments
(ganglia compartment and heart compartment)
which could be perfused separately [6].

The composition of normal physiological solu-
tion was as follows (mM/l): NaCl, 61.0; KCl, 3.3;
CaCl, 10.7; MgCl, 13.0; glucose, 5.0; and Hepes,
10.0 (pH adjusted to 7.5 by titration with NaOH).
High magnesium solution was prepared by merely
adding extra MgCl, to normal saline.

For the application of 5-HT to the preparation a
definite volume of the drug was introduced into the
solution in the chamber through a small pipette,
which was rapidly spread throughout the solution
by means of air bubbles released from the bottom
of the chamber. The concentration of the drug was
stated as final concentration in the solution. Dura-
tion of 5-HT application was usually less than 2
min, except for cases stated otherwise, during
which perfusion of the solution was stopped, and
interval of at least 20 min was allowed between the
two applications. The applicaton of 5-HT antagon-
ist, methysergide, and several neuropeptides for a
longer period was made by perfusing the solution
cntaining the chemical at a given concentration.

The following drugs were used: 5-hydroxy-
tryptamine creatinine sulfate (5-HT, Sigma),
methysergide-hydrogenmaleinate (methysergide,
Sandoz), 3-hydroxytyramine hydrochloride (dopa-
mine, Katayama), DL-octopamine hydrochloride
(octopamine, Nakarai), FMRFamide, YGGFMRF-
amide, pOQDPFLRFamide, small cardioactive pep-
tide A (SCP,) and SCPx, (Peninsula Laboratories
Inc.), and FLRFamide (Cambridge Research
Biochemicals).

Intracellular recording and stimulation of
neurons were carried out using glass microelec-
trodes filled with 3 M potassium acetate, having a
resistance of 5-10 MQ. Heart beat was recorded
by a strain gauge. In a few experiments, the
intestinal nerve was cut off from the ganglia and it
was stimulated by an Ag-AgCl bipolar electrode at
the point just before entering the pericardium.

When interconnections between two neurons or
effects of chemicals on neurons were investigated,
only the preparation of circumoesophageal ganglia
was used. Membrane current was measured using
a voltage-clamping method by two microelectrodes
as described previously [8].

The data were stored in an FM tape recorder
(Sony, DFR3515) for later analysis and redisplay-
ed on an ink-writing pen-recorder (Nihon Kohden,
RJG 4024).

All the experiments were carried out at room
temperature of 23-25°C.

RESULTS

Effects of 5-HT application and nerve stimulation
on heart beat

Isolated atrium usually repeats regular beating
in the experimental chamber. When 5-HT at
concentrations above 10-°M was applied to the
atrium, the frequency and amplitude of heart beat
were enhanced. Stimulation of the intestinal nerve
produced potentiation of heart beat similar to that
obtained by 5-HT application (Fig. 1). The poten-
tiation by nerve stimulation was abolished when
the preparation was perfused with high magnesium
(3 Mg**) solution which may block the neuro-
muscular synapses. The enhancement of heart
beat by 5-HT application, however, was not block-
ed in 3X Mg** solution, suggesting that 5-HT acts
postsynaptically. Dopamine exhibited effects simi-
lar to 5-HT on the atrium, but the threshold
concentration was 10°-°M. Octopamine showed
no significant effects at concentrations up to 10~°
M.

The action of heart excitatory neurons

A neuron in the right parietal ganglion, PON,
has been shown to be the most potent heart excitor

5-HT and Neuropeptides on Snail Heart 379

1 Control

—_———_ 105M 5-HT

2 HighMg* 20min

after 15min
ee | 10M 5-H
h 4

3 ae ain 20min
t@))
19
(—)

ae ORM) GHET 1min

Fic. 1. Effects of high-Mg** in solution on the excita-
tory responses of the atrium to nerve stimulation
(st.) and 5-HT application. The intestinal nerve was
stimulated with electrical pulses of 3 V, 1 msec at 5
Hz for 1 sec. 5-HT was applied during the period
shown by the horizontal line under each record.

1 Control

2 10°M UML

extending axons directly to the heart [6]. In the
experiments illustrated in Figure 2, the effects of
methysergide, a potent blocker of 5-HT receptor
in the gastropod heart muscles, on the action of
PON stimulation and 5-HT application were ex-
amined. The intracellular stimulation of PON
evoked impulses in PON, which, in turn, enhanced
the frequency and amplitude of heart beat. This
enhancement of the beat was almost completely
blocked by perfusing the atrium preparation with
methysergide (Fig. 2A). Similarly, the potentia-
tion of heart beat by 5S-HT application was blocked
by methysergide, although methysergide showed
no significant direct effects on the beat (Fig. 2B).
These results suggest that the potentiation of heart
beat by PON stimulation may be mediated by
5-HT.

To obtain a better understanding on the mode of
action of PON, modulatory effects of several
neuropeptides on the heart excitatory action of this
neuron as well as direct effects of those substances
on the heart beat were examined. Of six
neuropeptides (FMRFamide, FLRFamide,
YGGFMRFamide, pOQDPFLRFamide, SCP, and

3 Wash

after 40min

>
Se |
: (—)
=z
st! St. st:

1 Control

10 °M 5-HT

2 10°M
after 30min
LA
w
(—)
10°M5-HT 1min

Fic. 2. Effects of a 5-HT blocker, methysergide (UML), on responses of the atrium to PON-
stimulation (A) and 5-HT application (B). A and B are records from different preparations. In A,
the heart beat (upper tracings) induced by the spikes of PON (lower tracings) are recorded
simultaneously. PON was made to fire by current injection (st.) at 5 Hz for 10 sec. In B, 5-HT was
applied during the period shown by the horizontal line under each record.

380 K. Hort, Y. FURUKAWA AND M. KosayYASHI

2 3x10°°m FMRFa

old

A1 Control

C1 Control

3x10°"'m 5-HT
2 3x10°°m FMRFa

3x107'm 5-HT
Fic. 3.

B1 Control

2 105m SCPz

ui iy

st st

D1 Control

oH
n
i [—}
10°m 5-HT
2 10°m SCP,
10°6m 5-HT 41min

Effects of neuropeptides on the enhancement of heart beat induced by the spikes of PON (A

and B) and 5-HT application (C and D). A, B, C and D are records from different preparations. In
A and B, PON was stimulated (st) at 5 Hz for 20 sec and 10 sec, respectively. A> was recorded 20
min after application of FMRFamide (FMRFa), and B, was 30 min after application of SCPy. In C
and D, 5-HT was applied during the period shown by the horizontal line under each record. C> and
D, were recorded 20 min after application of FMRFa and SCPg, respectively.

SCPg) tested, FMRFamide and FILRFamide
showed slight enhancing effects on the heart beat
with the threshold concentration at 10°. °~3x 10°
M. SCPx had no effects on the most preparations
but exhibited potentiation in a few. YGGFMRFa-
mide, pOQDPFLRFamide and SCP, showed neith-
er direct effects on the heart beat nor modulatory
effects on the action of PON. Thus, the modula-
tory effects of FMRFamide and SCP on the heart
excitatory action of PON were further investi-
gated. The preparation used in the experiment
with the results being illustrated in Figure 3
showed no significant direct responses to FMRFa-
mide at 3X10°-°M nor to SCP, at 10°°M.
However, when the atrium preparation was per-
fused with FMRFamide for more than 20 min, the
excitatory action of PON was enhanced and the
response to 5-HT was also potentiated (Fig. 3A,
C). On the contrary, perfusion with SCP x resulted
in depression of the excitatory responses to both
PON stimulation and 5-HT application (Fig. 3B,

D). The effects of FMRFamide and SCP, were
reversible (not shown).

The mode of action of the other heart excitors
named tonically autoactive neurons, TAN, TAN-2
and TAN-3 [6, 11] was also examined. However,
the heart excitatory action of these neurons was
not modulated by perfusing the atrium preparation
with any of the foregoing six kinds of
neuropeptides.

The action of d-RCDN and d-LCDN on heart
excitors

When the cerebral neuron, d-RCDN or d-
LCDN, was stimulated intracellularly to evoke a
burst of impulses, excitatory responses were pro-
duced in TAN, TAN-2 and TAN-3. These three
TANSs behaved similarly with no different prop-
erties. Figure 4A shows an example in the case
between d-RCDN and TAN-2. By the stimulation
of d-RCDN at 10 Hz, TAN-2 which had previously
been hyperpolarized to stop firing was depolarized

5-HT and Neuropeptides on Snail Heart 381

A Control

Wash

a
N-2

B Control UML

10°M 5-HT

10°M 5-HT

Wash

10°M 5-HT

1min

Fic. 4. Blocking action of methysergide (UML) on the depolarizing responses of TAN induced by a
burst of impulses in d7-RCDN (A) and 5-HT application (B). A. TAN-2 was hyperpolarized by 10
mV. d-RCDN was stimulated at 10 Hz for 50 sec. Middle record (UML) was obtained 60 min after
application of methysergide. B. TAN was hyperpolarized by 30 mV. The top of action potentials
was cut off. 5-HT was applied during the period shown by the horizontal line under each record.
Middle record (UML) was obtained 30 min after application of methysergide.

and began to fire. These excitatory responses were
found to be depressed reversibly by 5S-HT antagon-
ist, methysergide. Similarly, as shown in Figure
4B, application of 5-HT produced spikes superim-
posed on a slow depolarization in TAN, which
were also depressed by methysergide. These re-
sults are essentially the same as those obtained in
PON [8], suggesting that the neurotransmitter of

Control

5 msec

Fic. 5.

the two cerebral neurons is 5-HT.

In the experiments shown in Figure 5, the effects
of a burst of impulses in the cerebral neuron or the
application of putative neurotransmitters on the
activities of three TANs were examined. A burst
of impulses in d-LCDN increased the spike fre-
quency in TAN-2 and produced a broadening of
the spikes (Fig. SA). The application of 5-HT also

Control

105M FMRFamide _

10sec 5 msec

Control

40mV

105M FMRFamide _

10sec SS
5 msec

Change in spike duration of TAN produced by a burst of impulses in d-LCDN (A), application

of 5-HT (B) and FMRFamide (C and D). In A~C, spontaneous activities of TAN-2 (A), TAN-3
(B) and TAN (C) were recorded. In D, TAN was driven to fire by a depolarizing current injection
at 2 Hz. In A, d-LCDN was stimulated at 10 Hz for 20 sec. In B~D, 5-HT (B) or FMRFamide (C
and D) was applied during the period shown by the horizontal line under each record. Arrows in
A,, B;, C; and D, indicate selected spikes which are displayed at expanded time scale in A>, Bz, Co

and D,.

382 K. Hort, Y. FURUKAWA AND M. KosayASHI

A

Control

40mV

x
/;
a A
we a
|
A
Vv

a

\

4—_

RF
a/
Sg
vy / Control

-150nA

Fia. 6.

FMRFamide

Wash

200nA

50 msec

2000 nA A

/,
/

x

{7
ae
yy

// FMRFa
w
Vi.

60mV

Effects of FMRFaminde on the membrane currents of TAN. A. Membrane currents with and

without 1.25 10~° M FMRFamide. Holding potential was —40 mV. The command pulse was 50
msec in duration and depolarized to OmV. B. I-V relationships of peak inward currents with

(closed triangles) and without (open triangles) FMRFamide (FMRFa).

Open upright (4) and

upside-down (v) triangles denote values before application of FMRFamide (Control) and after
wash (Wash), respectively. C. I-V relationships of oupward currents measured at the end of the
pulse with and without FMRFamide. Symbols mean the same with B.

elicited similar responses (Fig. 5B). On the con-
trary, by the application of FMRFamide to TAN a
tentative cessation of spontaneous firings and re-
markable shortening of recovered spikes were
demonstrated (Fig. 5C). Even when TAN was
driven to fire by current injection at 2 Hz, FMRFa-
mide produced a slight hyperpolarization and shor-
tened the spike duration (Fig. 5D). These inhibi-
tory actions of FMRFamide were also demon-
strated in PON, results of which were reported in
part previously [5]. Further, FMRFamide-related
peptides such as FLRFamide and pQDPFLRFa-
mide were found to cause similar responses in
TANs.

Finally, the effects of FMRFamide on the mem-
brane currents were examined in TAN, which was
axotomized to get better conditions for space-
clamp. Holding potential was set at —40 mV. The
membrane currents measured using depolarizing

command pulses consisted of a transient inward
current and slowly developing outward current
(Fig.6A). Application of FMRFamide remark-
ably reduced both the peak inward current and
delayed outward current. In Figures 6B and C, I-V
relationships with and without FMRFamide are
illustrated. These results are in contrast to those
obtained by applying 5-HT to PON [8].

It is concluded that 5-HT and FMRFamide act
antagonistically in the heart excitatory neurons,
PON and TANs.

DISCUSSION

The present study demonstrated that application
of 5-HT produced potentiation of the beat in the
atrium preparation like stimulation of a heart
excitatory neuron, PON. The potentiation of
heart beat by both 5-HT application and PON

5-HT and Neuropeptides on Snail Heart 383

stimulation was blocked by a 5-HT blocker,
methysergide, suggesting that potentiation by
PON stimulation may be mediated by 5-HT.

A neuropeptide FMRFamide usually showed
direct enhancing effects slightly on the beat of
Achatina atrium. However, since the threshold
was quite high and the effects were variable de-
pending on preparations, it may be difficult to
consider this peptide acts physiologically directly
to Achatina atrium. On the other hand, the
modulatory action of FMRFamide on the effects of
PON stimulation or 5-HT application was effective
at relatively low concentrations with little variabil-
ity. MRFamide is known to show powerful
modulatory effects on the synaptic transmission in
molluscs [12, 13]. In the cerebral and sub-
oesophageal ganglia of Achatina there have been
shown a number of FMRFamide immunoreactive
neurons [14]. By using a immunohistochemical
method, we have also observed FMRFamide-
containing nerve terminals in the atrium as well as
FMRFaminergic neurons in the ganglia (unpub-
lished data). Thus, the excitatory modulation by
FMRFamide (or FMRFamide-related peptide) at
the synapse from PON to the heart seems to be
probable physiologically.

In the present experiment, the activity of two
cerebral neurons, d-RCDN and d-LCDN, pro-
duced excitatory responses in three TANs, which
were depressed by a 5-HT blocker like those to
5-HT application. Moreover, both the activity of
the cerebral neurons and 5-HT application pro-
duced the spike broadening in TANs. These
results are consistent with our previous results [8]
that d-RCDN and d-LCDN may be serotonergic
neurons. However, the results conflict with those
by Croll [15], who showed using histochemical
methods that these cerebral neurons do not con-
tain significant amounts of 5-HT. One possible
explanation for this disagreement would be, as
Croll has suggested (personal communication),
that d-RCDN and d-LCDN may exert their effects
upon PON and TANs via a polysynaptic pathway
with the last cell in the chain being serotonergic.
However, we have considered that the pathway
could be monosynaptic from our results of phy-
siological experiments [7]. The second possibility
would be that the specificity of methysergide to the

receptor of Achatina neurons might not be so strict
and it may block the receptor of the other trans-
mitters rather than serotonin. The third possibility
would be that the cell bodies of d-RCDN and
d-LCDN do not contain or less synthesize 5-HT,
which may be synthesized during axonal transport
and will be released from the axonal terminals.
This seems likely to us but it remains to be
examined further.

5-HT and FMRFamide showed antagonistic ac-
tions to the heart excitatory neurons, PON and
three TANs. 5-HT depolarized the membrane of
PON, closed 5-HT-sensitive K channels, increased
the voltage-dependent Ca** current and produced
spike broadening [8, 16]. The present experiment
showed that spike broadening by 5-HT also occur-
red in TANs. Contrary to these, FMRFamide
hyperpolarized TAN membrane, produced spike
shortening and decreased inward current possibly
by increasing background K* current. These
antagonistic actions between 5-HT and FMRFa-
mide are similar to those found in Aplysia sensory
neurons [17-20].

It is well known that actions of FARFamide are
variable on the same organ in different species as
well as on different organs in one species [5, 12].
Thus, it may not be surprising that in Achatina
FMRFamide inhibited the action of heart excita-
tory neurons in the ganglia and enhanced the effect
of excitatory substances released from the neurons
at the peripheries. It is postulated that FMRFa-
mide causes spike shortening in PON and TANs,
possibly resulting in the decrease of the transmitter
release, and promotes the efficacy of the substance
to the heart, 1.e. FMRFamide may contribute to
the efficient use of the transmitter.

ACKNOWLEDGMENTS

The authors are very grateful to Dr. Yojiro Muneoka
for his helpful suggestions and discussion. The authors
also thank Sandoz Ltd. for a sample of methysergide.
This research was supported in part by Grant-in-Aid
(No. 63540575) from the Ministry of Education, Science
and Culture, Japan.

REFERENCES
1 Leake, L. D. and Walker, R. J. (1980) “Inverte-

11

384

brate Neuropharmacology”. Blackie & Son Ltd.,
Glasgow, p. 358.

Jones, H. D. (1983) The circulatory systems of
gastropods and bivalves. In “The Mollusca”. Vol. 5,
Physiology, Part 2. Ed. by A. S. M. Saleuddin and
K. M. Wilbur, Academic Press, New York, pp. 189-
238.

Painter, S. D. and Greenberg, M. J. (1982) A
survey of the responses of bivalve hearts to the
molluscan neuropeptide FMRFamide and to 5-
hydroxytryptamine. Biol. Bull., 162: 311-332.
Walker, R. J. (1986) Transmitters and modulators.
In “The Mollusca”. Vol. 9, Neurobiology and Be-
havior, Part 2. Ed. by A. O. D. Willows, Academic
Press, New York, pp. 279-485.

Kobayashi, M. and Muneoka, Y. (1989) Functions,
receptors, and mechanisms of the FMRFamide-
related peptides. Biol. Bull., 177: 206-209.
Furukawa, Y. and Kobayashi, M. (1987) Neural
control of heart beat in the African giant snail,
Achatina fulica Férussac. 1. Identification of the
heart regulatory neurones. J. exp. Biol., 129: 279-
293.

Furukawa, Y. and Kobayashi, M. (1987) Neural
control of heart beat in the African giant snail,
Achatina fulica Férussac. II. Interconnections
among the heart regulatory neurones. J. exp. Biol.,
129: 295-307.

Furukawa, Y. and Kobayashi, M. (1988) Modula-
tion of ionic currents by synaptic action and 5-HT
application in the identified heart excitatory neurone
of the Africa giant snail, Achatina fulica Férussac. J.
exp. Biol., 137: 319-339.

Akagawa, M., Furukawa, Y. and Kobayashi, M.
(1988) Dual effects of 5-hydroxytryptamine on the
heart of a pulmonate, Achatina fulica Férussac.
Comp. Biochem. Physiol., 89C: 327-331.

Hori, K., Akagawa, M., Furukawa, Y. and
Kobayashi, M. (1987) Actions of putative neuro-
transmitters on the heart beat of a mollusc, Achatina
fulica Férussac. Dobutsu seiri, 4: 148.

Matsuoka, T., Goto, T., Watanabe, K. and
Takeuchi, H. (1986) Presence of TAN (tonically
autoactive neuron) and its two analogous neurons,
located in the right parietal ganglion of the sub-

12

13

14

15

16

17

19

20

K. Hori, Y. FURUKAWA AND M. KoBAYASHI

oesophageal ganglia of an African giant snail (Acha-
tina fulica Férussac). Morphological and electrophy-
siological studies. Comp. Biochem. Physiol., 83C:
345-351.

Greenberg, M. J., Payza, K., Nachman, R. J.,
Holman, G. M. and Price, D. A. (1988) Rela-
tionships between the FMRFamide-related peptides
and other peptide families. Peptides, 9 (Suppl. 1):
125-135.

Bulloch, A. G. M., Price, D. A., Murphy, A. D.,

Lee, T. D. and Bowes, H. N. (1988) FMRFamide
peptides in Helisoma: Identification and physiologic-
al actions at a peripheral synapse. J. Neurosci., 8:
3459-3469.

Takayanagi, H. and Takeda, N. (1987) FMRFa-
mide immunoreactive neurons in the central nervous
system of the snail, Achatina fulica. Comp.
Biochem. Physiol., 88A: 263-268.

Croll, R. P. (1988) Distribution of monoamines
within the central nervous system of the juvenile
pulmonate snail, Achatina fulica. Brain Res., 460:
29-49.

Furukawa, Y. and Kobayashi, M. (1988) Two
serotonin-sensitive potassium channels in the iden-
tified heart excitatory neurone of the African giant
snail, Achatina fulica Férussac. Experientia, 44:
738-740.

Siegelbaum, S. A., Belardetti, F., Camardo, J. S.
and Shuster, M. J. (1986) Modulation of the seroto-
nin-sensitive potassium channel in Aplysia sensory
neurone cell body and growth cone. J. exp. Biol.,
124: 287-306.

Siegelbaum, S. A., Camardo, J. S. and Kandel, E.
R. (1982) Serotonin and cyclic AMP close single K*
channels in Aplysia sensory neurones. Nature, 299:
413-417.

Belardetti, F., Kandel, E. R. and Siegelbaum, S. A.
(1987) Neuronal inhibition by the peptide FMRFa-
mide involves opening of S K* channels. Nature,
325: 153-156.

Brezina, V., Eckert, R. and Erxleben, C. (1987)
Modulation of potassium conductances by an en-
dogenous neuropeptide in neurones of Aplysia cali-
fornica. J. Physiol., Lond., 382: 267-290.

ZOOLOGICAL SCIENCE 7: 385-393 (1990)

© 1990 Zoological Society of Japan

Evidence for the Phagocytotic Removal of Photoreceptive
Membrane by Pigment Cells in the Eye of
the Planarian, Dugesia japonica

NosBuAKkI TAMAMAKI_2

Department of Anatomy, Fukui Medical School, Matsuoka, Fukui 910-11 Japan

ABSTRACT—Fine structural changes induced by daily cycles and dark adaptation were investigated in

the eye of Dugesia japonica.

During the daily cycle, an irregular arrangement of microvilli and an

accumulation of vesicles in the microvillar area were often observed in animals fixed before dawn. The
changes before dawn were enhanced as the period of dark adaptation was prolonged. Pigment cells,
which surround the microvillar area, are shown to serve much the same function as the vertebrate
pigment epithelium. The pigment cells phagocytose the accumulated vesicles and ingested debris is
degraded further into granules and membrane whorls. Internalization by the pigment cells is regarded
as one of the mechanisms for the photoreceptive membrane removal in the planarian, Dugesia japonica.

INTRODUCTION

The eye of the planarian Dugesia is composed of
a pigment cup formed by pigment granule-
containing cells (pigment cells) and photoreceptor
cells whose apical microvilli-bearing processes are
enclosed in the pigment cup [1-3]. The fine
structure of other planarian eyes has also been well
described [4-7]. According to the results of the
following studies [4, 8-10], the eye structures are
thought not to be stationary, but to change con-
stantly during the normal daily cycle and in condi-
tions of abnormal light- or dark-adaptation. The
structural changes may be due to the turnover of
the photoreceptive membrane, i.e. addition and
removal, as reviewed by Schwemer [11].

Concerning the mechanism of the removal of the
photoreceptive membrane, it is well known that
the rhabdomeric membrane in the compound eye
of many arthropods is internalized by photorecep-
tor cells via the formation of coated vesicles [12-
15]. Also in the planarian Dalyellia, Bedini et al.

Accepted July 5, 1989
Received March 6, 1989
' Present address: Department of Neurobiology and
Behavior, State University of New York at Stony
Brook, Stony Brook, New York 11794, U.S.A.
Reprint requests should be addressed to the address in
Japan.

[10] postulated an internalization of receptive
membrane into the photoreceptor cells, in addition
to the drastic changes in the eye structure in daily
cycles.

On the other hand, Carpenter et al. [3] discussed
the possibility that cytoplasmic extensions of the
pigment cells, which cover the pupilary opening of
the pigment cup, may phagocytose photoreceptive
membrane in the eye of the planarian Dugesia
dorotocephala. Glial cells such as pigment epithe-
lium are known to serve as a removal system in
vertebrate eyes [16-18]. Removal of the photore-
ceptive membrane by surrounding glial cells has
also been reported in annelid eyes [19] and in
arthropod eyes [20, 21], but generally such the
removal is unusual in invertebrate eyes. There-
fore, it is interesting to investigate how the pig-
ment cells in the Dugesia eye may participate in
the membrane turnover system.

I report in this paper that the pigment cells
phagocytose remnants of shed photoreceptive
membrane in the eye of the Dugesia japonica.

MATERIALS AND METHODS

Specimens of Dugesia japonica were collected at
Takeda river in the vicinity of Fukui in May and
June. The animals live on the underside of rocks
which are half buried in the sand. Since it was

386 N. TAMAMAKI

Fic. 1. A: Section through an eye of an animal fixed at 4:00 showing the pigmented eye cup surrounding the
microvillar area. (M) microvilli; (PC) pigment cells; (P) microvilli-bearing process of receptor cell. Arrow head
indicates the accumulation of vesicles between rhabdomeres and pigment cells. Bar=10 «m B: Accumulation of
vesicles between rhabdomeres fixed at 4:00. The tips of some microvilli were swollen. (M) microvilli; (V)
vesicles; Bar=1 «~m. C: Higher magnification electron micrograph of the boundary between the microvillar area
and the pigment cells fixed at 4:00. The pigment cells seemed to phagocytose the vesicles. Bar=0.1 «m.

Phagocytotic Removal by Pigment Cells 387

impossible to perform an accurate measurement of
luminous intensity at the places where the animals
live, the animals were not maintained in the
laboratory but collected at each time they were
required for use. Morphological changes depend-
ing on daily cycles were investigated in groups of 5
animals, collected under a dim red light and fixed
immediately, at 0:00, 4:00, 8:00, 12:00, 16:00
and 20:00hr. Further groups of five animals
collected at 20:00 were dark-adapted for 1 day, 2
days, 4 days, 6 days, 8 days, and 10 days at the
same temperature as that of river water, and fixed.
All the animals were fixed in 2.5% glutaraldehyde
and 0.1 M phosphate buffer (pH 7.4) for 2 hr at
4°C. After trimming the specimens, tissue blocks
containing the eyes were postfixed in 1% OsO,
and 0.1M phosphate buffer (pH 7.4) for 2 hr at
4°C. The tissue blocks were dehydrated with
alcohol series and embedded in Epon 812. Silver
thin sections were contrasted with uranyl acetate
and lead citrate, and observed with an electron
microscope (Hitachi H600).

TABLE 1.

Eye volume changes induced by dark adaptation.

RESULTS

Morphological changes in the normal daily cycle

The morphology of Dugesia eyes has already
been described well by MacRae [1], Kishida [2]
and Carpenter et al. [3]. The eye of Dugesia
Japonica is composed of rhabdomeric photorecep-
tor cells and pigment cells (Fig. 1A). The pigment
cells form a pigmented eye cup surrounding photo-
receptive parts of the photoreceptor cells (micro-
villar area). Microvilli of the photoreceptor cells
present an orderly appearance. The microvilli
belonging to one cell form a rhabdomere. In the
section shown in Figure 1A about 14 rhabdomeres
were observed. The microvilli-bearing process
contains much smooth endoplasmic reticulum and
many mitochondria. Multivesicular bodies smaller
than 1 ~m in diameter were sometimes observed in
the stalk region [3] of the photoreceptor cells. The
pigment cup is formed by a single layer of pigment
cells. Inner concave surface of the pigment cup
possesses few microvilli and is rather flat. Most of
spaces in the pigment cell are occupied by mem-
brane bound pigment granules and nucleus.

Area occupied by pigment cells and

microvillar area, and their ratios to the total area (pigment cells+microvillar area) were
measured on photographs of sections which appropriately contain the optical axis of eyes fixed

at 16:00 and eyes dark adapted for 10 days.

Volume changes were estimated from these

values
Eyes at 16:00

Total Microvillar area (/total) Pigment _ cells (/total)

(ym*) (ym*) (ym*)
1 3,430 2,270 66.2% 1,160 33.8%
2 3,230 2,180 67.3% 1,060 32.7%
3 2,550 1,590 62.4% 726 37.6%
4 2,740 1,780 64.9% 961 35.1%
Mean 2,990 1,950 65.4% 1,030 34.6%

Eyes of 10 days dark adaptation

Total Microvillar area (/total) Pigment. cells (/total)

(ym) (um”) (ym*)
1 2,310 989 42.8% 1,320 57.2%
2 1,700 839 49.3% 862 50.7%
3 2,460 1,390 56.4% 1,070 43.6%
4 2,110 1,075 50.9% 1,035 49.1%
Mean 2,150 1,070 50.0% 1,070 50.0%

388

Investigation of eye structures fixed at 4 hr inter-
vals revealed that some structural changes occur in
daily cycles. The rhabdomeric microvilli common-
ly present an orderly appearance. However, in the
eye fixed at 4:00 the microvilli were often irregu-
larly arranged, especially in the area facing the
inner concave surface of the pigment cup (Fig. 1A,
B). In addition, the tips of some microvilli were
swollen (Fig. 1B). Accumulations of vesicles were
sometimes observed in extracellular spaces, i.e.
between rhabdomeres, in the eyes fixed at 4:00

adie By TR " i
ee Bl

£6
a a

Fic. 2.
after 10 days’ dark adaptation.

N. TAMAMAKI

(arrow head in Fig. 1A, B). The inner concave
surface of nearby pigment cells possessed some
cytoplasmic extensions extending into the accu-
mulation of vesicles, and the pigment cells seemed
to phagocytose the vesicles (Fig. 1C). These
structural changes were sometimes observed in the
eyes fixed at times other than 4:00, but on a
smaller scale.

Morphological changes in dark-adapted condition
After 10 days of dark adaptation, the diameters

chi

> ye.

A: Cross-section showing an eye of an animal fixed at 4:00. Bar=S0 “m B: Cross-section showing an eye

Bar=50 ~m C: Electron micrograph of pigment cells of an eye fixed after 10

days’ dark adaptation. (M) microvilli; (NP) nucleus of pigment cell. Bar=1 um.

Phagocytotic Removal by Pigment Cells 389

of most eyes became smaller than those of animals
maintained in the normal daily cycle (Fig. 2A, B).
The dark-adapted eyes had a smaller microvillar
area, while the pigment cells were thicker than in
normal eyes. In order to make a comparison
between the 10 days dark-adapted eyes and the
eyes fixed at 16:00, semithin sections containing
optical axes of these eyes were made carefully, and
the microvillar area and the area occupied by
pigment cells were measured (Table 1). During

Fic. 3. A: Accumulation of vesicles between rhabdomeres fixed after 8 days’ dark adaptation.

the dark adaptation, microvillar area decreased
more than 40%, but the area of pigment cells
seemed to increase by a few percent. The pigment
cells of dark-adapted eyes contain areas devoid of
pigment granules (Fig. 2B), in addition to the
perinuclear region. The cellular regions lacking
pigment granules contain many vacuoles, which in
turn contain membrane debris (Fig. 2C). In nor-
mal eyes, pigment granules are lacking only in the
perinuclear region (Fig. 2A).

Bar=1 ~m B:

Vacuole containing tubular membrane debris observed in the apical portion of a pigment cell fixed after 6 days’
dark adaptation. Bar=0.5 ~m C: Boundary region between the accumulated vesicles and the pigment cells fixed
after 6 days’ dark adaptation. Arrow heads indicate the cytoplasmic extensions surrounding the accumulated
vesicles. (M) microvilli; (VA) vacuole containing tubular membrane debris. Bar=1 um.

390 N. TAMAMAKI

After longer dark-adaptation, many more mic-
rovilli appeared to be irregularly arranged and a
considerable number of vesicles were accumulated
between the rhabdomeres and pigment cells (Fig.
2C) or between adjacent rhabdomeres (Fig. 3A).
The inner surface of the pigment cell, which is
ordinarily concave, became convex in the region
facing the accumulation of vesicles and possessed
cytoplasmic extensions extending into the accu-
mulation of vesicles (arrow heads in Fig. 3C).
Vacuoles containing vesicles were often observed
in a part of the pigment cells closest to the accu-
mulation of vesicles. Tubular membrane debris,
which seem similar to the microvilli of the photore-
ceptor cells, were also observed in these vacuoles
(Fig. 3B and VA in Fig.3C). The vacuoles
observed in that region of the pigment cells closest

Fic. 4. A: Vacuoles containing vesicles, granules
and whorls observed in the basal portion of a
pigment cell fixed after 10 days’ dark adapta-

tion. (G) granule; (V) vesicles; (W) whorl.
Bar=1 um. B: Golgi apparatus in a pigment
cell fixed after 8 days’ dark adaptation. (GA)
Golgi apparatus. Bar=0.5 um C: Multivesicu-
lar bodies in a microvilli-bearing process of a
photoreceptor cell in the eye fixed after 10
days’ dark adaptation. (M) microvilli; (MV)
multivesicular body. Bar=1 um.

to the outer surface of the pigment cup contained
mainly vesicles, granules and membrane whorls
(Fig. 4A). The pigment cells of dark-adapted eyes
had Golgi apparatus as shown in figure 4B (Fig.
4B). As one more additional feature of dark-
adapted eyes, photoreceptor cells often contained
multivesicular bodies larger than 1 «m in dia-
meter, in their microvilli-bearing processes as well
as in their stalk region (Fig. 4C).

DISCUSSION

In the eyes of the planarian Dugesia japonica,
morphological changes in daily cycles and in dark
adaptation were investigated with special attention
to the pigment cup and microvillar area. An
irregular arrangement of photoreceptive microvilli

Phagocytotic Removal by Pigment Cells 391

and an accumulation of vesicles in the perimicro-
villar extracellular space were often found in anim-
als fixed before dawn. The eye morphology of the
animals used in this study did not show rhythmic
circadian changes in total darkness. The morpho-
logical changes observed before dawn were en-
hanced as the period of dark adaptation was
prolonged. Therefore, as reported by others [4, 8,
10], light conditions actually have significant
effects on photoreceptive membrane structures in
the planarian eyes.

Formation of vesicles in the perimicrovillar ex-
tracellular space was sometimes regarded as an
artifact caused by an inadequate fixation. Howev-
er, the formation of vesicles is now known to be
prevalent in some animals [19, 21, 22], and is
regarded as one mechanism of the photoreceptive
membrane turnover [11]. In the planarian eye, it is
believed that dark adaptation caused a swelling [4]
and an irregular wavy appearance in the microvilli
[9] as the result of decreased membrane stability
and eventual photoreceptor atrophy [8]. There-
fore, in the eyes of the planarian Dugesia japonica,
the accumulation of vesicles which follows the
decrease in microvilli is unlikely to be an artifact
rather than the result of a biological mechanism.
Shedding of microvilli may be initiated by a swell-
ing of the apical edge, leading to an accumulation
of vesicles in the perimicrovillar extracellular
space.

When the animals are dark-adapted, the de-
crease in the size of the microvillar area was
accompanied by a certain increase in the volume of
the pigment cells (Table 1). Such a correlation
between the decrease in the microvillar area and
the increase in the volume of pigment cells may
imply that some amounts of substances are trans-
ferred from the microvillar area to the pigment
cells. Up until now, the possible phagocytotic
activity of pigment cells has been discussed once
with respect to the Dugesia eyes [3]. The accumu-
lated vesicles in the microvillar area of the dark-
adapted eyes seem to be taken up into the pigment
cells by phagocytosis (Fig. 3C). Sometimes micro-
villi may also be taken up into the pigment cells by
direct phagocytosis (Fig. 3B). The cellular regions
lacking pigment granules contain many vacuoles
containing membrane debris (Fig. 2C). These

ultrastructural observation undoubtedly support
the notion that the substances transferred from the
microvillar area to the pigment cells are the vesi-
cles produced by the shedding of microvilli and the
microvilli themselves.

Granules and membrane whorls are observed in
the vacuoles of the basal half of the pigment cells.
Similar prticles are observed in the accessory eye
of a giant snail Achatina fulica an are thought to be
made from shed microvilli membrane [23]. The
work of Chamberlain and Barlow [24] supports the
idea that membrane whorls are normal breakdown
products within the Limulus retina. The granules
and membrane whorls in the Dugesia eye may also
be changed from the phagocytosed vesicles.

The process of degradation of the vesicles taken
into the pigment cells may be carried out by the
action of lysosomal enzymes, such as acid phos-
phatase (AcPh) [25, 26]. Up to now, I have not
seen precisely localized AcPh-deposits within the
phagocytic vacuoles and the Golgi apparatus.
However, Golgi apparatus may produce this kind
of enzyme to degrade the debris of photoreceptive
membrane [20]. The reuse of photoreceptive
membrane taken into pigment cells has also been
discussed by Brandenburger and Eakin [25-27].

After the dark adaptation, multivesicular bodies
in the photoreceptor cells increase in number and
size. Multivesicular bodies larger than 1 ~m were
often observed not only in the stalk region but also
in the microvilli-bearing processes of the dark
adapted photoreceptor cells. The increase in num-
ber and size of the multivesicular bodies is also
correlated with the decrease in microvillar area.
Multivesicular bodies can contribute to that de-
crease by a resorbence of microvilli membrane,
and the multivesicular bodies will be transported
from the microvillar area to the perinuclear region
of sensory cell. This correlation may imply that the
formation of multivesicular bodies is one of the
mechanisms of the photoreceptive membrane re-
moval in the eyes of Dugesia japonica.

Bedini et al. [10] attributed the drastic decrease
in microvilli induced by darkness in the Dalyellia
eye to the resorbence of microvilli into the sensory
cells. The resorbence of microvilli into the sensory
cells in the planarian Dalyellia reminded us of the
resorbance of microvilli by the formation of mul-

392

tivesicular bodies in arthropoda eyes [12-15]. On
the other hand, the phagocytosis of pigment cells
to remove shed photoreceptive membrane in
Dugesia japonica reminded us of the phagocytosis
of pigment epithelium to remove tipes of rod and
cone outer segments in vertebrate eyes [16-18].
There may be two mechanisms by which the
photoreceptive membrane is removed from the
eye of Dugesia japonica. One is the phagocytosis
of the pigment cell, i.e. removal by phagocytosis of
surrounding glial cells, and the other is the forma-
tion of multivesicular bodies in the sensory cells,
i.e. removal by resorbence of sensory cells. Swell-
ing of the pigment cells after dark adaptation is
induced by the phagocytosis of shed photorecep-
tive membrane in the microvillar area. Although
the increase in the volume of pigment cells was
small and did not completely compensate the
decrease in the microvillar area, the greater
volume of membrane debris would be phagocy-
tosed and digested by the pigment cells. There-
fore, even if the formation of multivesicular bodies
may be a potential explanation for the decrease of
the microvillar area, the phagocytosis by pigment
cells must be also regarded as one of the mechan-
isms for the photoreceptive membrane removal in
the planarian Dugesia japonica. It is very interest-
ing to know that the two mechanisms for photore-
ceptive membrane removal prevailing in verte-
brate and invertebrate coexist in planarian eyes.

ACKNOWLEDGMENTS

The author thank Mrs. Yayoi Asamoto for her skilled
technical help.

REFERENCES

1 MacRae, E. K. (1964) Observations on the fine
structure of photoreceptor cells in the planarian
Dugesia tigrina. J. Ultrastruct. Res., 10: 334-349.

2 Kishida, Y. (1965) The ultrastructure of the eyes in
Dugesia japonica. 1. The distal portion of the visual
cell. Zool. Mag. (Tokyo), 74: 149-155.

3 Carpenter, K. S., Morita, M. and Best, J. B. (1974)
Ultrastructure of the photoreceptor of the planarian
Dugesia dorotocephala. 1. Normal eye. Cell Tissue
Res., 148: 143-158.

4 MacRae, E. K. (1966) The fine structure of photo-
receptors in a marine flatworm. Z. Zellforsch., 75:

10

12

13

14

15

19

N. TAMAMAKI

469-484.

Eakin, R. M. and Brandenburger, J. L. (1981) Fine
structure of the eye of Pseudoceros canadensis (Tur-
bellaris, Polycladida). Zoomorphology, 98: 1-16.
Lanfranchi, A. and Bedini, C. (1982) The ultras-
tructure of the sense organs of some Turbellaria
thabdocoela. I. The eyes of Polycystis naegelii Kol-
liker (Eukalyptorhynchia Polycystididae). Zoomor-
phology, 101: 95-102.

Fournier, A. (1984) Photoreceptors and photosensi-
tivity in platyhelminthes. In “Photoreception and
Vision in Invertebrates”, Plenum, Ed. by Al: Ma,
Plenum, New York, pp. 217-239.

RGhlich, P. and Tar, E. (1968) The effect of
prolonged light-deprivation on the fine structure of
planarian photoreceptors. Z. Zellforsch., 90: 507-
518.

Carpenter, K. S., Morita, M. and Best, J. B. (1974)
Ultrastructure of the photoreceptor of the planarian
Dugesia dorotocephala. 11. Changes induced by
darkness and light. Cytobiologie, 8: 320-338.
Bedini, C., Ferrero, E. and Lanfranchi, A. (1977)
Fine structural changes induced by circadian light-
dark cycles in photoreceptors of Dalyelliidae (Tur-
bellaria: Rhabdocoela). J. Ultrastr. Res., 58: 66-77.
Schwemer, J. (1985) Turnover of photoreceptive
membrane and Visual Pigment in Invertebrantes. In
“The molecular mechanism of photoreception”. Ed.
by H. Stieve, Springer, pp. 303-326.

White, R. H.(1964) The effect of light upon the
ultrastructure of the mosquito eye. Am Zool., 4:
433.

White, R. H. (1967) The effect of light and light
deprivation upon the structure of the larval mos-
quito eye, II. The rhabdom. J. Exp. Zool., 166:
405-425.

White, R. H. (1968) The effect of light and light
deprivation upon the structure of the larval mos-
quito eye, III. Multivesicular bodies and protein
uptake. J. Exp. Zool, 169: 261-268.

Eguchi, E. and Waterman, T. H. (1967) Changes in
retinal fine structure induced in the crab Libinia by
light and dark adaptation. Z. Zellforsch., 79: 209-
229.

Young, R. W. and Bok, D. (1969) Participation of
the retinal pigment epithelium in the rod outer
segment renewal process. J. Cell Biol., 42: 392-403.
Young, R. W. (1971) Shedding of discs from rod
outer segments in the rhesus monkey. J. Ultrastruct.
Res., 34: 190-203.

Young, R. W. (1977) The daily rhythm of shedding
and degradation of cone outer segment membranes
in the lizard retina. J. Ultrastruct. Res., 61: 172-
185.

Eakin, R. M. and Brandenburger, J. L. (1985)
Effects of light and dark on photoreceptors in the

20

Pl

22,

23

Phagocytotic Removal by Pigment Cells

polychaete annelid Nereis limnicola. Cell Tissue
Res., 242: 613-622.

Piekos, W. B. (1986) The role of reflecting pigment
cells in the turnover of crayfish photoreceptors. Cell
Tissue Res., 244: 645-654.

Blest, A. D. and Maples, J. (1979) Exocytotic
shedding and glial uptake of photoreceptor mem-
brane by a salticid spider. Proc. Roy. Soc. Lond. B.,
204: 105-112.

Williams, D. S. and Blest, A. D. (1980) Extracellu-
lar shedding of photoreceptor membrane in
the open rhabdom of a tipulid fly. Cell Tissue
Res., 205: 423-438.

Tamamaki, N. and Kawai, K. (1983) Ultrastructure
of the accessory eye of the giant snail, Achatina

24

25

26

393

fulica (Gastropoda, Pulmonata). Zoomorphology,
102: 205-213.

Chamberlain, S. C. and Barlow, R. B. Jr. (1984)
Transient membrane shedding in Limulus photore-
ceptors: Control mechanisms under natural lighting.
J. Neurosci., 4: 2794-2810.

Brandenburger, J. L. and Eakin, R. M. (1980)
Cytochemical localization of acid phosphatase in
ocelli of the seastar Patiria miniata during recycling
of photoreceptoral membranes. J. Exp. Zool., 214:
127-140.

Brandenburger, J. L. and Eakin, R. M. (1985)
Cytochemical localization of acid phosphatase in
light- and dark-adapted eyes of a polychaete worm,
Nereis limnicola. Cell Tissue Res., 242: 623-628.

ie) a et ry “sa ememnast “beret -

<

siemrerclie hk: chil iperriighal | es ban ne “ee mod dincinaeih aa f
evbomsios stamina 7 ‘ew Redbrbpor
bo wokieet hate dt ie eee ey reinalalapacilibatael ae
L at snibbeade Souactnren pasty sD : esp -mecinantanser he
POT NMT Thi dibet | Piney oie . 5 aoe A ipl ed U lf
(AS) BONE Wey ee ‘sue UTR yschess elgg i ba
‘ t vee titted Re | atts vores hee sat
fAovileoa “aoiniedbity oo go Naya} 08) ren j .

fiwew) 4 * o ied :

F HN reg ae ; Jian einer, Aly
, e163 co. Saneiken pia rite tigi
‘ RGaeo “st visert, Sed Leh Taney its

feo fide AT) Led ae ae "= r, if ts

one.

fey Tei eOonmreReoag ‘nya? (beet) De ‘a
es) tlds toes ete wank. diel evechay vont ve mi
ray" j ey raps eT ut 2

: ‘er ; ite, Stee
pri i tpg wish

. io. Coreen,

ra viel haw

;
Ofer cheap.
i Mice?
‘
se ; i) “Tyran

(a ervent a
i] ibd ah oy 1]

ees’
tat cftyes ’
Fh ye jens j

T et adet
=)
ly
{ fo
iw
‘
i
-
.
=i
i
i
i
‘i
i bee
-

ZOOLOGICAL SCIENCE 7: 395-399 (1990)

Eurythermic Growth and Synthesis of Heat Shock
Proteins of Primary Cultured Goldfish Cells

MAKI SATO, HirRosHI MITANI and AKIHIRO SHIMA

Zoological Institute, Faculty of Science, University of Tokyo,
Bunkyo-ku, Tokyo 113, Japan

ABSTRACT—Cells derived from the tail fin of goldfish (Carassius auratus) are cultured at 37°C (GTF
e-2) and 27°C (GTF e-3). It is observed that the doubling times of GTF e-2 cells and that of GTF e-3
cells depend solely on the incubation temperature, irrespective of difference in the temperature at which
the primary culture is started. It has been reported previously that some goldfish cell lines cultured in
vitro for a long time grew stenothermically. The present study indicated that cells derived from the
primary culture retained their ability to grow eurythermically at least in the early passages.

The relationship of protein synthesis as to incubation temperature was studied empirically. When
GTF e-3 cells were exposed to 37°C, four major heat shock proteins were induced. Their molecular
weights were 90 K dalton (hsp 90), 70 K dalton (hsp 70), 42 K dalton (hsp 42) and 30 K dalton (hsp 30).
In GTF e-2 cells these proteins were synthesized constitutively at 37°C. The levels of synthesis of these
proteins were much higher than those observed when the cells were incubated at 27°C. At 40°C, hsp 70

© 1990 Zoological Society of Japan

and hsp 30 were the dominantly synthesized proteins of GTF e-2 and e-3 cells.

INTRODUCTION

Cells derived from fish and those derived from
mammals can be cultured using the same medium
and the same serum, but the incubation tempera-
ture is usually different. For the cultured mamma-
lian cells, the range of permissive growth tempera-
ture is 36-39°C with an optimal at 37°C [1]. For
most fish cell lines the optimal growth temperature
is 20-25°C and upper limit temperature for growth
is about 30°C [2]. In general, the suitable tempera-
ture for cultured fish cells correlates with the
temperature of the animals natural habitat; the
cultured cells derived from cold-water fish grow
most rapidly at low temperature (about 20°C),
while those from some tropical fish could be kept
at 37°C [2 and 3 for review]. Cell lines derived
from different species of fish have different optim-
al growth temperatures [4-6]. The RBCF-1 cells
derived from the caudal fin of the goldfish grow
most rapidly at 37°C. They also can grow con-
tinuously at 20°C [6]. Recently, we isolated from
RBCF-1 cells two cell clones which grow only in a

Accepted July 11, 1989
Received May 12, 1989

narrow range of temperatures (stenothermic
growth), the optimal growth temperature being
37°C and 27°C. It has been shown that cell hybrids
of two stenothermic clones with different optimal
growth temperature could continue to grow in a
wide range of temperatures (the eurythermic
growth) [7]. These results suggest that the
eurythermic or stenothermic characteristics of
growth of goldfish cells in culture may be modified
during a prolonged in vitro cultivation. In this
report, we investigated the growth temperature for
primary cultured goldfish cells, and the effects of
incubation temperature on the protein synthesis of
the cells.

MATERIALS AND METHODS

Primary culture

The tail fin about 1 cm? of an adult goldfish with
about 5 cm body length was cut off. It was soaked
in 0.4% NaClO, washed in Ca?*- and Mg’*-free
phosphate-buffered saline (PBS(—)), and cut into
pieces in trypsin-EDTA (0.1% trypsin and 0.02%
EDTA in PBS(—)). The small tissue pieces were
gently stirred in the presence of trypsin-EDTA at

396 M. Sato, H. MITANI AND A. SHIMA

room temperature for about 60 min. The tissue
pieces and dispersed cells were collected by centri-
fugation and seeded into two 25 cm’ plastic flasks
(Corning Glass Works, Corning, N.Y.). Each
flask was incubated at 37°C (GTF e-2) or 27°C
(GTF e-3). The medium used was Leibovitz’s L-15
medium (GIBCO, Grand Island, N.Y.), contain-
ing 15% fetal bovine serum (Hyclone Laborator-
ies, Logan) and the antibiotics (50 “g/ml strep-
tomycin and kanamycin, 60 ~g/ml). A half volume
of the medium was renewed every 3 days. The
primary culture was harvested 18 days after the
inoculation, and 1x 10° cells were inoculated into
the fresh dishes with 10 ml of the medium. Both
cell lines were subcultured every 3 days. The
population doubling number (PDN) was calcu-
lated from the day of first subcultivation by log
(cumulative growth ratio)/log 2.

Growth curve

Cells were seeded in culture petri dishes at 2.0 x
10° cells/dish. For each cell line half the dishes
were incubated at 27°C, half at 37°C. Each 3 days
3 dishes were counted for the number of cells. This
was continued for 12 days. The population doubl-
ing number (PDN) of the GTF e-2 was 3.2. The
PDN of the GTF e-3 was 1.3.

Protein analysis

To analyze the protein synthesis of GTF cells at
various temperatures, each flask was inoculated
with 5 x 10° cells (GTF e-2 at PDN 10.8, and GTF
e-3 at PDN 4.2) and incubated for 10 hr at 37°C or

1000 t (a)

100

number of cells (x104)

Fic. 1.

27°C, respectively. Before starting labeling, cells
were cultured for 2 hr in methionine-free Dulbec-
co’s modified Eagle medium with 10% FBS. Then,
Tran *°S-label™, E. coli hydrolysate labeling rea-
gent, containing **S-methionine (ICN Biomedic-
als, Irvine: specific activity >1000 Ci/mmol) was
added to the medium to a final concentration of 10
yuCi/ml and transferred to desired temperatures
and incubated for additional 2 hr. Subsequently,
the medium was removed and the cells were
washed with PBS(—), harvested by a small rubber
policeman. The cells were suspended in Laemmli’s
buffer [8], and boiled for 3 min.

The protein was analyzed on 10% polyacryla-
mide-SDS slab gels with 2.5% stucking gel using
the discontinuous buffer system of Laemmli [8].
Protein samples with approximately the same *°S
counts (about 30,000 cpm) were used for analysis.
Slab gels (5 cmX8.5 cm) were run at 15 mA for
about 110 min, and stained with silver stain (2D-
Silver Stain Kit ‘DPC’; Daiichi Pure Chemical,
Tokyo). The gels were dried and autoradiog-
raphed using Kodak X-Omat RS film.

RESULTS

The cells dispersed from caudal fin of the
goldfish and the remaining tissue pieces attached
to the plastic substratum. Many cells migrated
from tissue pieces and continued to proliferate
both at 37°C and 27°C. After 18 days of incuba-
tion, GTF e-2 and GTF e-3 cells reached confluen-
cy. At that time the total cell numbers were 5x 10°

0 100 200 300

hours

Growth curve of GTF e-2 and e-3 cells at 27°C (a) and 37°C (b). The ordinate is the average

number of cells recovered from a dish, and the abscissa is time in hours after inoculation. @—®:
GTF e-2, primary culture was started at 37°C. O—O: GTF e-3, primary culture was started at HG:

Eurythermic Growth of Fish Cells 397

ai oo

(b)

iN oi © » [2 G

Fic. 2. The autoradiograph of heat shock proteins in (a) GTF e-3, and (b) GIF e-2. The temperature

was shifted as shown below:

lane A: 27°C (12 hr)-27°C (labeled for 2 hr)

lane B: 37°C (12 hr)=27°C (labeled for 2 hr)

lane C: 27°C (12 hr)-37°C (labeled for 2 hr)

lane D: 37°C (12 hr)-37°C (labeled for 2 hr)

lane E: 27°C (12 hr)-40°C (labeled for 2 hr)

lane F: 37°C (12 hr)-40°C (labeled for 2 hr)
The cells were labeled with *°S-methionine at the corresponding times. The arrows indicate hsp 90,
hsp 70, hsp 42 and hsp 30 (from top to botoom), respectively.

for GTF e-2 and 4x10° for GTF e-3, and the
morphology of the cells did not show any observ-
able difference between the two cell strains. Both
cell strains continued to grow without any sign of
crisis of growth. This is one of the notable
characteristics of cultured fish cells reported pre-
viously [4, 9, 10]. Figure 1 shows growth curves of
GTF cells at early passages (PDN<11). The
doubling time of GTF e-2 cells calculated from the
slope of the initial straight line portion of the
growth curve was 35 hr at 27°C, and 25 hr at 37°C.
The doubling time for GTF e-3 cells was 36 hr at
27°C, and 25 hr at 37°C. Thus, the doubling time
of cultured goldfish cells at early passages did not
depend on the temperature at which the primary
culture was started, but depended only on the
incubation temperature.

Figure 2 shows the autoradiographs of newly
synthesized proteins. The four major proteins
were identified when the cells were transferred to

higher incubation temperature. The molecular
weights of hsps observed in the present study were
90, 70, 42 and 30-kD, and would correspond
respectively to hsp 90, hsp 70, hsp 42 and hsp 30 of
RBCF-1 cells [7]. The synthesis of these proteins
was markedly increased when the GIF e-2 and
GTF e-3 cells were incubated first at 27°C for 12 hr
and then transferred to 37°C or after a long
incubation at 37°C. At higher temperature (40°C)
the hsp 70 and hsp 30 became dominant newly
synthesized protein, and relative amount of hsp 90
and hsp 42 synthesis decreased in both cell strains.

DISCUSSION

It is generally accepted that cultured cells grow
optimally when the incubation temperature is
slightly higher than that preferred by the intact
animal. For example, in case of cold-water fish
like rainbow trout (Salmo gairdnerii), the cultured

398 M. Sato, H. MITANI AND A. SHIMA

cells grow most rapidly at temperature from 8 to
12°C [3]. As to goldfish cell lines, the optimal
growth temperature reported has not been consis-
tent. Rio et al. [4] reported 20°C for SJU-1 cell
line, while 33°C was reported for CAF cell line by
Etoh and Suyama [5]. Shima et al. established the
RBCF-1 cell line which was initially cultured at
37°C and could grow at a wide range of tempera-
tures from 20°C to 37°C (eurythermic growth) [6].
Recently, after a long term cultivation at 27°C or
37°C, clones of RBCF-1 line which could grow
only at a narrow range of temperatures (stenother-
mic growth) were isolated [7].

In this study, we found that both GTF e-2 and
e-3 cells which were derived from a goldfish tail fin,
retained eurythermic growth properties at early
passages (PDN< 11), in spite of 10°C difference in
primary culture. A probable cause for difference
between eurythermic and stenothermic growth
may be the difference in the length of their subcul-
tivation time. SJU-1 cell line was at 110th passage
after 39 months of subcultivation during which
periods the cells were cultured at 20°C. RBCF-1
cells have been subcultured for more than 10
years. GTF cells were at only 2nd passage (the
PDN is 3.2 for GTF e-2, and 1.3 for GTF e-3) after
4 days of subcultivation when they were used for
the experiments. The intact goldfish as individuals
can survive both at 27°C and 37°C, and this fact
seems to correlate with the growth properties of
GTF cells. So the cells in vivo may have the
eurythermic growth properties. SJU-1 and RBCF-
1 cells, which have been cultured for a long time at
a constant temperature, may have lost their ability
to grow eurythermically.

The molecular weights of major heat shock
proteins synthesized by GTF cells were almost the
same as those synthesized by cells of Drosophila
[11, 12], mammals [13] and rainbow trout [14].
The hasp 90 of GTF cells may correspond to hsp 83
of Drosophila in the manner of response to the
change of temperature; hsp 83 was not induced at
a higher temperature (38°C) in Drosophila [11],
and in GTF cells hsp 90 was not induced at 40°C.

The hsp 42 was induced in GTF cells when the
cells were transferred from 27°C to 37°C, and also
when they had been kept at 37°C. This response to
temperature shift was similar to that of hsp 90.

The cells of Drosophila do not synthesize hsp 42
[11, 12]. Rainbow trout cells (hsp 42) [14], chicken
embryo fibroblasts (hsp 47) [15] as well as HeLa
cells (hsp 43) [13] synthesize this class of hsp, all of
which may correspond to hsp 42 of GTF cells.
Therefore, hsp 42 may be the common heat shock
protein in the cells of vertebrates so far examined.

The cells of Drosophila synthesize hsp 70 and
hsp 26 which correspond to has 70 and hsp 30 of
GTF cells. The hasps 70 and 26 are reported to be
induced at a higher temperature than the tempera-
ture which induced hsp 90 [12]. This was also
observed in GTF cells.

The relationships between the range of growth
temperatures of cells and the temperature which
can induce heat shock proteins in the cells have
been reported for only a few species [11-14].
However, it may generally be said that hsps synth-
eses are induced when cells are transferred to a
temperature which is a few degrees higher than
their optimal growth temperature. Furthermore,
the heat shock proteins may be synthesized as a
consequence of environmental stress. In this
study, we found that GTF cells do not follow this
pattern, at 37°C hsps seem to be induced con-
tinuously, although the cells could grow actively at
that temperature. This is quite different from
RBCF-1 cells. When RBCF-1 cells were transfer-
red from 26°C to 37°C, four major heat shock
proteins were induced, but after 12 hr incubation
synthesis of heat shock proteins decreased [7].

To summarize, in this study we found that
cultured fish cells in the very early passages (PDN
<3.2) can grow actively at both 27°C and 37°C.
This eurythermic growth may reflect a growth
characteristic of goldfish cells in vivo. It was also
observed that in the cultured goldfish cells at early
passages, four major hsps were induced at 37°C in
spite of continued growth at this temperature for a
long time. These results indicate that primary
cultured goldfish cells are quite useful for investi-
gating factors that determine the optimal growth
temperature of cells, and the function of hsps in
eurythermic animals.

ACKNOWLEDGMENTS

This research was supported by a grant from Fisheries
Agency, Japan to A. Shima.

Eurythermic Growth of Fish Cells

REFERENCES

Sisken, J. K., Morasca, L. and Kibby, S. (1965)
Effects of temperature on the kinetics of the mitotic
cycle of mamalian cells in culture. Exp. Cell Res.,
39: 103-116.

Wolf, K. and Ahne, W. (1982) Fish cell culture.
Advances in Cell Culture, 2: 305-328.

Wolf, K. (1979) Cold-blooded vertebrate cell and
tissue culture. Methods in Enzymology, 58: 466-
477.

Rio, G. J., Magnavita, F. J., Rubin, J. A. and
Beckert, Wm. H. (1973) Characteristics of an
established goldfish Carassius auratus (L.) cell line.
J. Fish Biol., 5: 315-321.

Suyama, I. and Etoh, H. (1977) A cell line derived
from the fin of the goldfish, Carassius auratus. Zool.
Mag., 88: 321-324.

Shima, A., Nikaido, O., Shinohara, S. and Egami,
N. (1980) Continued in vitro growth of fibroblast-
like cells (RBCF-1) derived from caudal fin of the
fish, Carassius auratus. Exp. Gerontol., 15: 305-
314.

Mitani, H., Naruse, K. and Shima, A. (1989)
Eurythermic and stenothermic growth of cultured
fish cells and their. J. Cell Sci., 93: 731-737.
Laemmli, U. K. (1970) Cleavage of structure pro-
teins during the assembly of the head of bacter-

10

11

12

13

14

15

399

iophage T4. Nature, 227: 680-685.

Komura, J., Mitani, H. and Shima, A. (1988) Fish
cell culture: Establishment of two fibroblast-like cell
lines (OL-17 and OL-32) from fins of the medaka,
Oryzias latipes. In Vitro, 24: 294-298.

Shima, A. and Setlow, R. B. (1985) Establishment
of a cell line (PF line) from a gynogenetic teleost,
Poecilia formosa (Girard) and characterization of its
repair ability of UV-induced DNA damage. Zool.
Sci., 2: 477-483.

Lindquist, S. (1986) The heat shock response. Ann.
Rev. Biochem., 55: 1151-1191.

Lindquist, S. (1980) Varying patterns of protein
synthesis in Drosophila during heat shock: Implica-
tion for regulation. Develop. Biol., 77: 463-479.
Hickey, E. D. and Weber, L. A. (1982) Modulation
of heat-shock polypeptide synthesis in HeLa cell
during hyperthermia and recovery. Biochemistry,
21: 1513-1521.

Kothary, R. K. and Candid, E. P. M. (1981) Induc-
tion of a normal set of polypeptides by heat shock or
sodium arsenite in cultured cells of rainbow trout,
Salmo gairdnerii. Can. J. Biochem., 60: 347-355.
Nagata, K., Hirayoshi, K., Obara, M., Suga, S. and
Yamada, K. M. (1988) Biosynthesis of a novel
transformation-sensitive heat shock protein that
binds to collagen. J. Biol. Chem. 263: 8344-8349.

A eam:
if j
'
ea VINA
} #) 4
if ‘
yf i
{ AS
a ut)
‘ 4
™ f
i ¥
’ zi ) H ‘
rf
a
M4 ty
4 Ved
4 way
i T
i a '
j
Le i
; } }
}
U *
4
i mm
Ai
N
by Bide g
Lie
Lye ?
j

we gh
(be

riboeaste

‘A

rete uee vy ah ra) five Leal " iran
h Me lich iy LR AR ;
Hifetacd compte Mes

te toe bie ie | BY

Vp tte) gait ogenrs -
i) eee llnds wre bin is saetealanial =e

wet yeve assed bl

‘a a dt a) a sul Wa 2 1 i fai Fite a iv
ty : ' niin ieee” " ruil hee (ol yal sada) are

tvlay Dormia alt tes Jian” a mi ath nn ang

rgb it ADL oni ve

ot . eae

bay

gma bah lal ert
39) ca Wore wm

Libel tis of .guliedie alah @ eer m0

isnaeid ic re “| ie yy La daner pak en Sy

jam 40d) vara) et oh wiea it , Te

tyes «| Lorine aadlitadird ea

xc ita tap) ;
aay t ks
fat iti + '

hap ae neat * aT Kae Leib tis het ah
an pieedieraye TF UM og ohare?) Oy bandon
wt FEyry ot gta) ; y TT Wwitie Bid : bs

j itinrmney a TEESE ew re yar: HOTS, Ste
fi vyre f ‘f ey #

is ‘ iAG ta i coe
late? Sri ier WH rey sy" Ove

Meet ts HATO hg 9 Whetrineah gt

iy i ‘ag || taf

Hie eclf oti

i 7 m Dirt
5
t
m Bec '
m *
y
-
2
\
»
swe
w
HT
’
Yet,

ZOOLOGICAL SCIENCE 7: 401-407 (1990)

© 1990 Zoological Society of Japan

Motility of Cultured Iridophores from the Freshwater
Goby Odontobutis obscura

TETSURO IGA, JiIRO KINUTANI and Naomi MAENO

Department of Biology, Faculty of Science, Shimane University, Matsue 690, Japan

ABSTRACT—Iridophores in the integument of the freshwater goby, Odontobutis obscura, are motile.
Iridophores isolated from scales of the goby were cultured in L-15 medium. The primary cultured cells
were motile. Their movements involved aggregation and dispersion of platelets within the cells, which
were not caused by a reversible retraction of cellular processes. Alpha-MSH induced aggregation of
platelets, while melatonin and norepinephrine, separately, induced dispersion of the platelets. These
responses of the cultured iridophores to drugs were the same with as those of iridophores in preparations
of isolated scales. The speed of migration of platelets in cultured iridophores was very slow, and it

appeared to be the same as that in iridophores in intact scales.

Most of the cultured iridophores

exhibited sensitivity to light; they assumed a dispersed state in the light and an aggregated state in

darkness.

INTRODUCTION

Iridophores are light-reflecting chromatophores
commonly found in the dermis of many poiki-
lothermal vertebrates, and they are known to play
a predominant role in the generation of skin
coloration [1]. Electron-microscopic observations
have revealed that iridophores in fishes contain a
large number of platelets, which run parallel to
each other and form stacks [2-5]. These platelets
are mainly composed of guanine and have a very
high reflective index, so that, when stacked, they
generate various colors. The phenomenon is cal-
led physical or structural coloration [6, 7]. Until
recently, these iridophores in fishes were not
thought to play an active part in changes of color
via phenomena that involved motility.

Quite recently, we found that iridophores in the
integument of the freshwater goby, Odontobutis
obscura, respond to neural and hormonal stimula-
tion via changes in the reflective surface of the cells
[8]. Light- and electron-microscopic observation
suggested that the motility of the iridophores in-
volved the translocation of reflecting platelets
within the cells [8, 9]. At present, however,
information on the movements of the reflecting

Accepted July 27, 1989
Received May 1, 1989

platelets within the iridophores is very scanty.
Studies with cultured iridophores may provide us
with much useful information about such move-
ments.

The purpose of the present experiments was to
present the motility of cultured iridophores from
the freshwater goby, Odontobutis obscura.

MATERIALS AND METHODS

Culture of iridophores

Scales isolated from the dorso-lateral side of the
freshwater goby, Odontobutis obscura, were im-
mersed in a solution composed of a mixture of
equal volumes of physiological saline (128 mM
NaCl, 2.6 mM KCl, 1.8 mM CaCl, 5 mM HEPES-
NaOH buffer, pH 7.2) and an isotonic solution of
KCl for 40 min. This solution, with in its high level
of K*, induced dispersion of platelets in the
iridophores [8]. The epidermis was removed with
fine forceps from the scales, after they had been
immersed in physiological saline supplemented
with 2.5 mg/ml collagenase (Type II, Worthington
Bochemical Co., Freehold) for 20min. The
epidermis-free scales were then transferred into a
vial filled with a dissociation medium which con-
sisted of 2.5 mg/ml collagenase and 1.5 mg/ml
trypsin (Sigma Chemical Co., St Louis). The vials

402 T. Ica, J. KINUTANI AND N. MAENO

were gently stirred for 40-60 min at room temper- Nitta Gelatin, Osaka). The culture medium was
ature. Dissociated cells were collected with a fine Leibovitz L-15 medium (Gibco Lab., New York)
pipette under a dissecting microscope and cultured supplemented with 10% fetal calf serum, 100 IU/
in plastic dishes coated with collagen (Type A, ml penicillin, 100 g/ml streptomycin (Gibco

Fic. 1. Micrographs showing responses of cultured iridophores from Odontobutis obscura to alpha-MSH and
melatonin. A, In culture medium. B, In saline. C, D, E, F and G, 10, 30, 60, 90 and 120 min, respectively, after
treatment with 100 nM alpha-MSH. H, I, J, K and L, 20, 40, 60, 90 and 180 min, respectively, after treatment
with 1 ~M melatonin.

Motility of Cultured Iridophores

Lab.), and 10% DW. The dishes were incubated cellular motility.

at 26°C in air.
Drugs

Recording of responses of cultured iridophores Alpha-MSH, melatonin and norepinephrine
Cultures were observed under an inverted hydrochloride were obtained from Sigma Chemic-

phase-contrast microscope (Olympus CK-2) and al Co. These drugs were dissolved in physiological
their responses were photographed for analysis of _ saline.

Fic. 1. —Cont.

404 T. Ica, J. KINUTANI AND N. MAENo

All experiments were performed at room
temperature (22.0-25.0°C).

RESULTS

Cultured iridophores

After inoculation, the iridophores attached to
the substratum and began to spread within a day.
After 2 to 3 days in culture most iridophores were
fully spread and platelets were evenly dispersed
throghout the cells. The shapes of iridophores
were variable, with rod-like, dendritic, and dis-
coidal iridophores being observed. The diameters
ranged from 20 to 50 um, like those of iridophores
in intact scales.

The platelets in these iridophores assumed a
dispersed state within the cells in the light. After
being kept overnight in darkness, aggregation of
platelets in the cell centers occurred in most of the
iridophores, while some remained in the dispersed
state. If the cultures were transferred to the light,
iridophores with aggregated platelets returned to
the dispersed state within 3 hr. Thus, the cultured
iridophores appear to be sensitive to light.

Responses of cultured iridophores

Iridophores fully spread after 2 to 3 days in
culture were used for experiments. Alpha-MSH
(100 nM) induced aggregation of platelets into the
central regions of the iridophores. The platelets
began simultaneously to move centripetally and
became aggregated in the central region of each
cell after 90 to 120 min of the treatment (Fig.
1A-G).

During this time, the cell membranes attaching
to the substratum remained in their original state
without retraction. When melatonin (1 “#M) was
applied to iridophores with platelets in an aggre-
gated state, the platelets began to disperse centri-
fugally from the aggregates and returned to their
original dispersed state after 90 to 180 min (Fig.
1H-L). Norepinephrine (1 “M) also induced dis-
persion of platelets with the same time course as
that observed with melatonin. If the solution of
alpha-MSH was changed to physiological saline,
the platelets remained aggregated for at least 60
min without any sign of dispersion.

Requirement for Ca*~ in the action of MSH

The action of MSH was inhibited in Ca* *-free
saline that contained 1 mM EDTA. If the vehicle
for the peptide was changed to the standard saline,
aggregation of the platelets was induced in stan-
dard fashion. It is noteworthy that a similar
aggregation of platelets occurred when iridophores
that had been exposed to a prolonged treatment
with MSH, in an absence of Ca* *, were immersed
in the standard saline.

Migration of reflecting platelets

For analysis of the migration of reflecting
platelets within the iridophores during the course
of the aggregation response, the results of the
application of 100 nM alpha-MSH to cultured iri-
dophores were followed by photographing them at
intervals of 10 min, and the migration of platelets
was traced from the micrographs.

A typical example, indicating the migration of
platelets within a cell, is shown in Figure 2. The
platelets did not always appear to move linearly.
The profile of their velocity was also not linear.
Among the bulk of platelets that were moving

10 um

Fic. 2. A typical example of recordings that show the
centripetal migration of platelets in a cultured iri-
dophore from Odontobutis obscura. Each point
with a number shows, in order, the position of an
individual platelet at intervals of 10 min.

Motility of Cultured Iridophores 405

centripetally, there were some that were stationary
and other that even moved in the reversal direc-
tion. From the distance of platelet migration
measured every 10 min with 35 of platelets in 7
cells, the velocity of platelet migration was calcu-
lated. The maximum velocity of platelet migration
was calculated as 0.5 ~m/min, with a mean value
of 0.1 «m/min.

DISCUSSION

Iridophores with platelets in a dispersed state
were used for isolation of cells, because they gave
well-spread cultured cells. Iridophores with
platelets in the dispersed state may be more resis-
tant to the isolation treatment than those with
platelets in the aggregated state, which did not give
such well-spread cells in culture. Iridophores of
Odontobutis obscura appeared dendritic in vivo.
In culture, however, most of the cells assumed a
discoidal shape. Iwata et al. [10] reported that the
shapes of cultured cells may be influenced by
various factors, especially by the properties of the
substratum in the case of melanophores from the
medaka, Olyzias latipes, and that melanophores
cultured on a collagen-coated substratum had com-
plex shapes. In the present case, better adhesion
and spreading of the cells were obtained on a
collagen-coated substratum than on uncoated plas-
tic. There were no observable differences in the
shapes of cells in the well-spread state between
cells on the two substrata.

The cultured iridophores were responsive to
alpha-MSH, melatonin and norepinephrine. The
responses were same as those of iridophores in the
preparations of intact scales. Properties of hor-
mone receptors appeared to be unchanged by
culture in vitro.

The present experiments clearly showed that
motility of the iridophores involves centripetal or
centrifugal migration of the platelets within the
cells, but no retraction or elongation of cellular
processes. Most of the cultured cells did not
change their contours during their responses to the
agents applied. Some cells did change their shapes
during their responses, perhapes as a result of a
weakness of adhesion of the cells to the substra-
tum. Responses of such cells were not so clear-cut.

In iridophores in preparations of isolated scales,
the shapes of cells were unchanged after repeated
responses. _Electron-microscopic observations
have suggested that dendrites of the iridophores
are firmly anchored in the connective tissues and
that the platelets move within the cells [9].

The iridophores of the integument of some
species of amphibian undergo conspicuous changes
in shape, from a dendritic to a punctate appear-
ance, and such changes are regulated by MSH
from the hypophysis [11, 12]. However, whether
movement of platelets results passively from de-
ndritic retraction, or whether it is a selective
movement independent of changes in dendritic
morphology remains to be determined [13]. Cul-
tured iridophores from the tail skin of tadpoles of
the bullfrog, Rane catesbeiana, have been shown to
attach to dishes and form dendritic structures; such
iridophores respond to ACTH with contraction of
cells [14]. Recently, Butman et al. [15] studied
responses to hormones of cultured iridophores
from the integument of the Mexican leaf frog,
Pachymedusa dacnicolor. These iridophores were
of two distinct types which differed with respect to
both morphological and physiological features.
One type (type I) of iridophore responded to some
hormones by a reversible retraction of cellular
processes and rounding up of cells.

Physiologically active iridophores are also
known to be present in the integument of some
species of fish, namely the neon tetra, Paracheiro-
don innesi [16], and the blue damselfish, Chrysip-
tera cyanea [17]. In iridophores of the damselfish,
the motility is assumed to involve simultaneous
changes in the distance between contiguous
platelets in all the piles of platelets within the cells,
and such changes cause a shift in the spectral
reflectance from the cells [18]. The situation may
be similar to that in iridophores of the neon tetra.

Thus, motile iridophores in fishes can be clas-
sified into two types, in terms of motility: the
damselfish type, where changes occur in the dis-
tance between adjoining platelets; and the goby
type, where the intracellular migration of platelets
occurs. Quite recently, we found that iridophores
in the integument of some species of Gobiidae are
motile. The motility appears to be of the goby type
[19, Honma and Iga, unpubl.].

406

In the centripetal migration of platelets, the
course and the velocity of the migration are not
always linear. The movement is very slow, as in
iridophores in preparations of scales [8, 20]. As far
as we know, the movements may be the slowest of
those examined in fish chromatophores [21-25].
The present findings are important for elucidation
of the mechanisms of movement of the iri-
dophores.

MSH possesses strong pigment-dispersing action
in melanophores of poikilothermal vertebrates [1,
26], and Ca** is indispensable for its action [27-
29]. Such a dependence of the action of MSH on
Ca** was also shown in some non-melanophoral,
motile chromatophores of fishes, namely in
platyfish erythrophores, and in xanthophores as
well as leucophores of the medaka, Olyzias latipes
[30]. The involvement of Ca** may be important
for signal transduction from the receptor to the
catalytic unit of adenylate cyclase and/or for the
activation of the catalytic unit [31-33]. MSH acts
to induce “aggregation” of platelets in the iri-
dophores of Odontobutis obscura, as it does in
preparations of isolated scales [8], and Ca** is
indispensable for its action. Upon subsequent
application of standard saline, in the absence of
the peptide, after prolonged treatment with MSH
in a Cat *-free medium, the platelets in the iri-
dophores began to aggregate. This result suggests
arole for Ca** in transduction of signal generated
by MSH.

Iridophores in the integument of the neon tetra
were light-sensitive and changed in color from
deep violet to blue-green in response to illumina-
tion [16]. Most of the cultured iridophores from
the integument of the Odontobutis goby appear to
be light-sensitive. Studies on the light sensitivity of
chromatophores in fishes are now in progress.

REFERENCES

1 Bagnara, J. T. and Hadley, M. E. (1973) Chroma-
tophores and Color Change. Prentice-Hall, New
Jersey.

2 Kawaguti, S. and Kamishima, Y. (1964) Electron
microscopic study on the iridophore of the Japanese
porgy. Biol. J. Okayama Univ., 10: 75-81.

3 Kawaguti, S. (1965) Electron microscopy on iri-
dophores in the scale of the blue wrasse. Proc. Japn.

10

11

12

13

14

15

16

17

T. Ica, J. KINUTANI AND N. MAENO

Acad., 41: 610-613.

Harris, J. E. and Hunt, S. (1973) The fine structure
of iridophores in the skin of the Atlantic salmon
(Salmo salar L.) Tiss. Cell, 5: 479-488.

Hawkes, J. W. (1974) The structure of fish skin IT.
The chromatophore unit. Cell Tiss. Res., 149: 159-
172.

Fox, D. L. (1976) Animal Biochromes and Structu-
ral Colours. Univ. of California Press, Berkeley,
Los Angels.

Land, M. F. (1972) The physics and biology of
animal reflectors. Prog. Biophys. Molec. Biol., 24:
75-106.

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

Matsuno, A. and Iga, T. (1989) Ultrastructural
observations of motile iridophores from the freshwa-
ter goby, Odontobutis obscura. Pigment Cell Res.,
2: 431-438.

Iwata, A., Iwata, M. and Nakano, E. (1986)
Changes in the shape of melanophores of the meda-
ka, Oryzias latipes, cultured on collagen and
fibronectin-coated substrata. Zool. Sci., 3: 73-81.
Bagnara, J. T., Hadley, M. E. and Taylor, J. D.
(1969) Regulation of bright-colored pigmentation of
amphibians. Gen. Comp. Endocrinol., suppl. 2:
425-438.

Hadley, M. E. and Goldman, J. M. (1972) The
physiological regulation of the amphibian iri-
dophores. In “Pigmentation: Its Genesis and Biolo-
gic Control”. Ed. by V. Riley, Appleton-Century-
Crofts, New York, pp. 225-245.

Taylor, J. D. and Bagnara, J. T. (1972) Dermal
chromatophores. Am. Zool. 12: 43-62.

Ide, H. (1973) Effects of ACTH on melanophores
and iridophores isolated from bullfrog tadpoles.
Gen. Comp. Endocrinol., 21: 390-397.

Butman, B. T., Obika, M., Tchen, T. T. and
Taylor, J. D. (1979) Hormone-induced pigment
translocations in amphibian dermal iridophores, in
vitro: Changes in cell shape. J. Exp. Zool., 208: 17-
34.

Lythgoe, J. N. and Shand, J. (1982) Changes in
spectral reflexions from the iridophores of the neon
tetra. J. Physiol., 325: 23-34.

Oshima, N., Sato, M., Kumazawa, T., Okeda, N.,
Kasukawa, H. and Fujii, R. (1985) Motile iri-
dophores play the leading role in damselfish colora-
tion. In “Pigment Cell 1985: Biological, Molecular
and Clinical Aspects of Pigmentation”. Ed. by J.
Bagnara, S. N. Klaus, E. Paul, and M. Schartl,
Univ. of Tokyo Press, Tokyo, pp. 241-246.
Kasukawa, H., Oshima, N. and Fujii, R. (1987)
Mechanism of light reflection in blue damselfish
motile iridophores. Zool. Sci., 4: 243-257.

19

20

21

22

23

24

25

26

27

Motility of Cultured Iridophores

Iga, T., Maeno, N. and Asari, T. (1988) Receptor
mechanisms of motile iridophores of the floating
goby, Chaenogobius annularis. Zool. Sci., 5: 1211.
Iga, T., Takabatake, I. and Watanabe, S. (1987)
Nervous regulation of motile iridophores of a fresh-
water goby, Odontobutis obscura. Comp. Biochem.
Physiol., 88C: 319-324.

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

Iga, T. (1978) The mode of action of potassium ions
on the leukophores of a freshwater teleost, Oryzias
latipes. J. Exp. Zool., 205: 413-422.

Schliwa, M. and Euteneuer, U. (1978) Quantitative
analysis of the microtubule system in isolated fish
melanophores. J. Supramol. Struct., 8: 177-190.
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
Ill”, Ed. by A. M. F. O’Hare, SEM Inc., Chicago,
pp. 1279-1288.

Obika, M. (1986) Intracellular transport of pigment
granules in fish chromatophores. Zool. Sci., 3: 1-11.
Fujii, R. and Oshima, N. (1986) Control of chroma-
tophore movements in teleost fishes. Zool. Sci., 3:
13-47.

Vesely, D. L. and Hadley, M. E. (1979) Ionic
requirements for melanophore stimulating hormone

28

30

31

32

33

407

(MSH) action on melanophores. Comp. Biochem.
Physiol., 62A: 501-508.

Van de Veerdonk, F. C. G., Worm, R. A. A.,
Seldenrijik, R. and Heussen, A. A. (1979) The role
of calcium in hormone-controlled pigment migration
in Xenopus laevis. In “Pigment Cell”. Ed. by S. N.
Klaus, Karger, Basel, pp. 72-78.

Iga, T. and Takabatake, I. (1982) Action of mela-
nophore-stimulating hormone on melanophores of
the cyprinid fish Zacco temmincki. Comp. Biochem.
Physiol., 73C: 51-55.

Oshima, N. and Fujii, R. (1985) Calcium require-
ment for MSH action on non-melanophoral chroma-
tophores of some teleosts. Zool. Sci., 2: 127-129.
De Graan, P. N. E., Eberle, A. N. and van de
Veerdonk, F. C. G. (1982) Calcium sites in MSH
stimulation of Xenopus melanophores: studies with
photoreactive a-MSH. Mol. Cell. Endocrinol., 26:
327-339.

Sawyer, T. K., Hruby, V. J., Hadley, M. E. and
Engel, M. H. (1983) a-melanocyte stimulating hor-
mone: chemical nature and mechanism of actions.
Amer. Zool., 23: 529-540.

De Graan, P. N. E., van de Kamp, A. J., Hup, D.
R. W., Gispen, W. H. and van de Veerdonk, F. C.
G. (1984) Calcium requirement for e-MSH action
on melanophores: studies with forskolin. J. Recep-
tor Res., 4: 521-536.

oisetplael dnweaates ‘ Hutheges i

net ec rhe EI) Pe ee a | (RRO 140 G ieee A ee a
Neh ht Pal es dirname eorataetelhroet.) 14 eenmelieth jaan valle
Fuygiapgy @ioei WV, AAP T3tS Buel cheo Sretelentees vase

42), Li A yp iagphtee», TRE aelarnte WT help SOAlp
orem th ity meresaS WO Mindi a fens yeanla weeORt
A West algaaeey, en © tating eeqntcns abnoey Gustein eh
ae yearn a WAR ‘ oye ROEDER
( ctparleswy tine i dations » @eenoderny Tat beet
Ly Spray hodgrare? hecervgiige ue estilo . ot eG beth :

ae { ” bhi TE 2a Ad o2%
PN (rs c etlenaepeee, Te: HE Bo Steep)
it be 0 Jape so aibnihamarrrg eas wut
PG WAR SSE at Hie meat lon
4 a ee Oran Pe 1) Ab pekodsii ;
eeyer fl OM (ash drat) Coe cars unten gaan ‘

ae PTT a ens lage!
" vdatlayerad hf Strate Anwod (edie

t ip derpenyiey wtetl Nes

sag 7? Ay OS P2oip ene
R noerork Uru | ieee aia ter

: rene. “eeputerine di Bite! 35/4 qramwea2 al a
togriecs Oy Satigh . onl MildaawRtC) hash
aie. 208k ss SRN kT “ae ncalaan 7
y 2g Wisse elpliooatinged a
ff i [Deeded don dt erage
ba! i iitutsoeer boda «) (ePID 3
¢ dr imoN cated tactsied se
NN ul wes CG Kien
, pie eA wold hoe ee

et patt een fad (tle peda ae
“ zy: A »4

ath

ZOOLOGICAL SCIENCE 7: 409-417 (1990)

Changes in Lipid Composition in the Tail of Rana
catesbeiana Larvae during Metamorphosis

Masasui RyuzaAki! and Makoto OonukI

Department of Biology and Department of Physiology, Kitasato
University School of Medicine, Sagamihara 228, Japan

ABSTRACT—The composition of simple lipids and phospholipids obtained from the larval tails in Rane
catesbeiana at Taylor and Kollros (TK) stage V, X, XX, XXI, and XXII-XXIII were analyzed to clarify
their relationship with the regressive process of tails during metamorphosis in this frog. The weight
percentage of total simple lipid to total lipid (TL) was about 33% at TK stages V and X and increased
gradually to about 70% at TK stages XXII-XXIII. That of total phospholipid to TL was about 66% at
TK stages V and X decreased gradually to about 28% at TK stages XXII-XXIII. The weight percentage
of free fatty acid (FFA) and triglyceride (TG) to TL were about 12-15% and 4% at TK stages V and X,
and increased gradually to about 38% and 13% at TK stages XXII-XXIII, respectively. Those of the
other simple lipids, i.e., cholesterol and cholesterol ester did not change much during metamophosis.
The weight percentage of phospholipid classes, i.e., phosphatidylethanolamine, phosphatidylcholine,
phosphatidylserine, phosphatidylinositol, sphingomyelin and lysophosphatidylcholine to total phospho-
lipid also did not significantly change throughout metamorphosis. The resulting increase in the weight
percentage of total simple lipid to TL was due primarily to increase in FFA and TG during the

© 1990 Zoological Society of Japan

metamorphic climax.

INTRODUCTION

Biochemical changes occurring during anuran
metamorphosis have been reviewed by Bennett
and Frieden [1], Brown [2], Frieden [3], Weber
[4], Fox [5] and Yoshizato [6]. A critical review of
these studies shows that although remarkable
changes have been described in the ornithine-urea
cycle enzymes [7], haemoglobin [8, 9], serum
protein [9, 10], visual pigments [11], nucleic acids
especially of liver [12] and lysosomal acid hydro-
lases such as acid phosphatase [13-15], compara-
tively very little is known of the alternations in the
lipids of the anuran tadpoles during metamorph-
Osis.

Whatever information is available on the lipids
during anuran metamorphosis comes from the
work of Camerio, Italian group, reviewed by
Urbani [16], regarding changes during meta-
morphosis of Bufo vulgaris and Rana esculenta.

Accepted September 20, 1989
Received January 20, 1989
" To whom reprints request should be addressed.

Urbani [17] found an approximate fourfold de-
crease in the total body lipid of Bufo vulgaris
during the metamorphosis period. Light and Was-
chek [18] studied the liver fatty acids of the tadpole
and adult Rana grylio and reported no appreciable
difference in them. Sawant and Varute [19] have
reported on the lipid changes; that is, total lipids,
neutral lipids and phospholipids, in the tadpole
Rana tigrina during development. Recently Oka-
mura and Kishimoto [20] have reported on qualita-
tive and quantitative changes in the nervous sys-
tem glycolipids during metamorphosis of Xenopus
laevis, and Yates et al. [21] studied the patterns of
brain gangliosides of Rana catesbeiana during
metamorphosis and in the adult frog.

Lipids have the potential of serving as a source
of biochemical intermediates for the tricarboxylic
acid cycle. Their rapid utilization during a period
of starvation usually follows when the glycogen
reserves are almost depleted. Lipids have also
long been recognized as important membrane con-
stituents playing a significant role in various cellu-
lar phenomena [22-24]. Striking differences have
also been reported in the fatty acids of aquatic and

410 M. RyuZAkI AND M. OoNUKI

terrestrial animals [25]. Metamorphosis is a pic-
turesque even in the life cycle of anurans involving
biochemical, physiological, and anatomical
alternations, which have an adaptive value in the
transition from one environment to another. It
follows therefore that an insight into the alterna-
tions, if any, in the lipids during metamorphosis
will be interesting and hence desirable. As a first
step with this view, the present author analyzed
two kinds of lipids; that is, neutral lipids and
phospholipids, contained in the larval tail of Rane
catesbeiana in order to clarify the relationship
between these lipid compositions and the regres-
sion of the larval tail during metamorphosis.

MATERIALS AND METHODS

Animals

Rana catesbeiana tadpoles were collected from
their natural habitats in the suburbs of Ryugasaki
City, Ibaragi Prefecture. The animals fed on
boiled spinach in tubs in the laboratory at ca. 18—
22°C. Extraction and analysis of lipids from the
larval tails were made at five developmental stages
of Taylor and Kollros (TK) [26]; that is, V (the
length of the limb bud is twice its diameter), X (the
margin of the foot paddle is indented between all
five toes), XX (one or both fore-legs have prot-
ruded), XXI (the angle of the mouth has reached a
point midway between the nostril and the anterior
margin of the eye) and XXII-XXIII (XXII: the
angle of the mouth has reached the level of the
middle of the eye, XXIII: the angle of the mouth
has reached the level of the posterior margin of the

eyeball). Extraction and analysis of lipids from the
fat bodies were made at four developmental stages
of TK; that is, XV (the proximal toe pads appear),
XX, XXI and XXII-XXIII. Extraction and analy-
sis of lipids from the fat body were also mede on
the adult female frog. The fat bodies of tadpoles
and adult frogs, as well as the whole larval tails,
were subjected to chemical analysis. After the
animals were pithed, samples were removed from
the body, washed in saline, and analyzed im-
mediately.

Lipid extraction

The total lipids were extracted by our modifica-
tion [27] of the Folch method [28]. The samples
were homogenized with 20 volumes of chloroform/
methanol (2:1 or 1:2, v/v) by the use of an
ultrahomogenizer, UH-1 type, Nissei, Tokyo. The
extracts were filtered through a Buchner funnel.
The residues were again homogenized with 20
volumes of chloroform/methanol (2:1 or 1:2,
v/v). The extracts were combined and concen-
trated to dryness in a rotary evaporator under
nitrogen. The residue obtained was dissolved in
chloroform/methanol (2:1, v/v) and was filtered
through a glass filter (GA-100, Toyoroshi, Tokyo).
The soluble material was concentrated to dryness
in a rotary evaporator under nitrogen and then the
dry lipid samples were weighed. The total lipids in
the samples were determined gravimetrically. The
dry lipid samples were dissolved in 25 ml of chlor-
oform/methanol (2:1, v/v), flushed with nitrogen
and stored at —20°C for lipid analysis.

Separation of simple lipids and phospholipids
Thin-layer plates of Silica Gel HR (Merk,
Darmstadt, West Germany) were prepared
according to routine procedure. Sodium carbonate
impregnated plates were made by the method of
Skipski et al. [29]. The lipid samples dissolved in
chloroform/methanol (2: 1, v/v) were applied with
a Terumo microsyringe (MS-N25) on the activated
plates. Simple lipids were separated by one-
dimensional thin-layer chromatography (TLC) on
layers of Silica Gel HR using hexane/diethyl ether/
acetic acid (85: 15:2, v/v/v) as developing solvent.
Phospholipids were separated by one-dimensional
TLC on layers of Silica Gel HR impregnated with
sodium carbonate using chloroform/methanol/ace-
tic acid/water (50:25: 8:4 or 25: 15:4:2, v/v/v/v)
[29] as developing solvent. Authentic standards of
the simple lipids and phospholipids (Sigma) were
cochromatographed in each respective run.
Identification of simple lipids and phospholipids
on the dried plates was made by exposing the plate
to iodine vapour. The phospholipid spots were
further identified by employing the following
sprays: Dittmer and Lester’s reagent [30] and
Vaskovsky’s modified spray [31] for general phos-

Lipids in the Tadpole Tail

pholipids, ninhydrin (0.2% in butanol) for phos-
pholipids containing free amino groups,
Dragendroff reagent [32] for choline phospholi-
pids, p-benzoquinone for ethanolamine, ammo-
nium silver nitrate for inositol and mercuric oxide
barium acetate for inositol. Details of these sprays
and their diagnostic importance in thin-layer chro-
matography are critically described by Marinetti
[33]. The simple lipids were identified by em-
ploying a dichromate sulfuric acid spray [34]. The
detection of cholesterol and cholesterol ester was
further confirmed by employing antimony trichlor-
ide spray [35].

Infrared spectra of the simple lipids and phos-
pholipids were measured by pressing a film be-
tween NaCl plates.

Separation of phospholipid spots and estimation of
phospholipids

Phospholipids were fractionated by one-
dimensional TLC as described above. Segments of
the plate containing each phospholipid were
scraped off and the amount of each phospholipid
was determined by measuring the phosphorus con-
tent, according to the method of Bartlett [36].

Separation of simple lipid spots and estimation of
simple lipids

Simple lipids were separated by one-
dimensional TLC as described above. The seg-
ment of the Silica Gel HR layer containing each
simple lipid was scraped off and the amount was
measured. The amount of triglyceride was esti-
mated according to the method of Snyder and
Stephens [37]. The free fatty acid isolated by the
preparative Silica Gel HR layer was esterified with
3% hydrogen chloride-methanol at 100°C for three

411

hours. The fatty acid methyl esters were analyzed
and estimated at 160°C by a Shimadzu GC-5A unit
equipped with a 1.5 m3 mm glass column packed
with 15% ethylene glycol succinate on Celite 545
HMDS. Methyl heptadecanoate (Applied Science
Laboratories Inc., Lot 1933, Penna.) was used as
the internal standard. The amount of cholesterol
and cholesterol ester was estimated according to
the method of Zak [38].

For confirmation of results, the thin-layer chro-
matographic separations and the assays of phos-
pholipids and simple lipids were carried out in
triplicate sets of tadpoles in batches as described
above.

RESULTS

Lipid content in larval tail

The lipid content from the whole tail of each
specimen at various stages of metamorphosis (TK
stage XX to XXII-XXIII) of tadpoles is recorded
in Table 1. The total lipid content at TK stage XX
was about 5.44 of wet tissue and 52.05 mg/g of
tissue residue after lipid extraction, and gradually
increased during metamorphosis. The content was
found to be about 11.13 mg/g wet tissue and 77.02
mg/g residue at TK stages XXII-XXIII.

Thin-layer chromatograms of total lipid from larval
tial

Thin-layer chromatographic patterns of total
lipid from the whole tails of each specimens at the
various stages of premetamorphic stage (TK stages
V and X) and metamorphic climax stage (TK stage
XX to XXII-XXIII) of tadpoles, as shown in
Figure 1, quantitatively indicate that the total

TABLE 1. The content of total lipids from the metamorphosing tadpole tail of Rana catesbeiana
Developmental stages of Taylor and Kollros
XXI XXII
Total lipid
mg/g of wet tissue weight 5.44+0.10 6.88+ 1.22 11.13+0.46
mg/g of residue of the tissue
after lipid extraction 52.05+1.11 67.20+7.65 77.02 +5.60

The data are expressed as mg of lipid per g of wet tissue and per g of residue of the tissue after lipid

extraction.

The values are the mean+SD of triplicate specimens.

412 M. RyuUZAKI AND M. OoNUKI

=-ss =

——aSapar s
PE- 7 <— Gum aah aoe Ol
Bil =
owe
Sh:

1 2 3 4 5

Fic. 1. Thin-layer chromatograms of total lipids from
the tadpole tail of Rana catesbeiana at the following
five developmental stages of Taylor and Kollros
(TK); i.e. 1(V), 2(X), 3(KX), 4(XXI) and 5(XXII-
XXIII). Total lipids were separated by one-
dimensional thin-layer chromatography (TLC) on a
layer of Silica Gel HR using chloroform/methanol/
water (65:25: 4, v/v/v) as developing solvent. The
bands are the lipids, visualized by spray application
of anthrone-sulfuric acid. SL, simple lipid; PE,
phosphatidylethanolamine; PC, phosphatidylcho-
line; Sph, sphingomyelin; GL, glycolipid.

lipids consist mainly of simple lipids and phospho-
lipids as the major component, and glycolipids
which were positively visualized by spraying with
anthrone-sulfuric acid. By comparing these chro-
matographic patterns, it was found that the simple
lipid fraction which contains free fatty acid (FFA)
and triglyceride (TG) was quantitatively different
at TK stage XX to XXII-XXIII than those at TK
stages V and X.

Thin-layer chromatograms of phospholipids from
larval tail

Thin-layer chromatographic separation of the
phospholipids at the various stages of development
of tadpoles, as shown in Figure 2, quantitatively
indicates that the total phospholipids consist main-
ly of phosphatidylethanolamine (PE), phospha-
tidylcholin (PC), phosphatidylserine (PS), phos-
phatidylinositol (PI), sphingomyelin (Sph), as the
major components and_ lysophosphatidylcholin
(LysoPC). A quantitative comparison shows that
PC and PE were predominant, whereas PS, Sph
and PI were present in lesser concentrations and
LysoPC occurred in trace amount. Exept for the
minor quantitative changes, the gross patterns of

“bebe

Sph — a

—— LysoPC

Fic. 2. Thin-layer chromatograms of total phospholi-
pids from the tadpole tail of Rana catesbeiana at the
following five developmental stages of TK; i.e. 1(V),
2(X), 3(XX), 4(XXI) and 5(XXII-XXIII). Total
phospholipids were separated by one-dimensional
TLC on a layer of Silica Gel HR impregnated with
sodium carbonate using chloroform/methanol/acetic
acid/water (50:25:8:4 or 25:15:4:2, vw/v/v/v) [29]
as developing solvent. The phospholipids are visual-
ized by spray application of 50% aqueous sulfuric
acid. PE, phosphatidylethanolamine; PS, phospha-
tidylserine; PI, phosphatidylinositol; PC, phospha-
tidylcholine; Sph, sphingomyelin; LysoPC,
lysophosphatidylcholine.

the phospholipids did not show any significant
changes during the various developmental stages
of the tadpoles.

Thin-layer chromatograms of simple lipids from
larval tail

Thin-layer chromatographic separation of the
simple lipids of the various developmental stages
of the tadpoles, as shown in Figure 3, indicates that
the total simple lipids consist mainly of TG, FFA,
cholesterol (Chol) and cholesterol ester (Chol.E).
By comparing each chromatographic patterns, it
was found that the gross patterns of Chol and
Chol.E did not show any significant changes at the
various developmental stages. However, FFA and
TG did show a qualitative change. Comparing
gross patterns of FFA; qualitatively FFA was
predominant at TK stage XX to XXII-XXIII,
although FFA occurred in relatively low concen-
tration at TK stages V and X.

Lipids in the Tadpole Tail 413

Chol.E —

1t— =

FFA— sn - a oe
hil- << <n — ame eam

Fic. 3. Thin-layer chromatograms of total simple lipids
from the tadpole tail of Rana catesbeiana at the
following five developmental stages of TK; 7.e. 1(V),
2(X), 3(XX), 4(KXI) and 5(XXII-XXII). Total
simple lipids were separated by one-dimensional
TLC on a layer of Silica Gel HR using hexane/
diethyl ether/acetic acid (85:15:2, v/v/v) as de-
veloping solvent. The bands are the lipids, visual-
ized by spray application of 50% aqueous sulfuric
acid. Chol. E, cholesterol ester; TG, triglyceride;
FFA, free fatty acid; Chol, cholesterol.

Thin-layer chromatograms of simple lipids from
blood, liver and muscle of larvae and adult frogs

Thin-layer chromatographic separation of the
simple lipids from the blood, liver and muscle of
the metamorphic climax stage (TK stage XXI) of
tadpoles and adult frog in Rana catesbeiana qual-
itatively indicates that total simple lipids consist
mainly of TG, FFA, Chol and Chol.E (Fig. 4). In
liver and muscle tissue, TG and FFA qualitatively
were predominant in metamorphic climax stage
(TK stage XXI) of tadpoles, whereas they present
in lesser concentration in the adult frog. FFA
occurred in trace amount in the blood at both
animal stages.

Thin-layer chromatograms of simple lipids from fat
bodies of tadpoles

Thin-layer chromoatographic patterns of total
simple lipids from a whole fat body during a period
when the growth rate was reduced and metamor-
phic change accelated to give rise to the metamor-
phic climax stage (TK stage XX, XXI and XXII-
XXIII) of the tadpoles indicate that a large TG
faction is present in total simple lipids at each stage
(Fig. 5).

IS

Chol. E— —- - eal
ne @~ @-
iit: wae ~~ TFA

Chol re

> = GD aD GD
1 Zo 3,4 5 6

Fic. 4. Thin-layer chromatograms of total simple lipids
from the blood, liver and muscle in metamorphosing
tadpoles at TK stage XXI and in the adult frog of
Rana catesbeiana. Total simple lipids were sepa-
rated by one-dimensional TLC on a layer of Silica
Gel HR using hexane/diethyl ether/acetic acid (85
: 15:2, v/v/v) as developing solvent. The bands are
the lipids, visualized by spray application of 50%
aqueous sulfaric acid. 1, adult frog blood; 2, tadpole
blood; 3, adult frog liver; 4, tadpole liver; 5, adult
frog muscle; 6, tadpole muscle. Abbreviations are
described in Fig. 3.

Fic.5. Thin-layer chromatograms of total simple lipids
from the fat body of Rana catesbeiana at following
four developmental stages of TK; i.e. 1(XV),
2(XX), 3(XXI) and 4(XXII-XXIII). Total simple
lipids were separated by one-dimensional TLC on a
layer of Silica Gel HR using hexane/diethyl ether/
acetic acid (85:15:2, v/v/v) as developing solvent.
The bands are the lipids, visualized by spray applica-
tion of 50% aqueous sulfuric acid. TG, triglyceride.

Thin-layer chromatograms of simple lipids from fat
bodies of adult females

Thin-layer chromatographic patterns of total

414 M. RyuZAKI AND M. OoNuUKI

s- @&@

Fic. 6. Thin-layer chromatograms of total simple lipids
from the fat body in duplicate mature female speci-
mens (1, 2) of Rana catesbeiana. Total simple lipids
were separated by one-dimensional TLC on a layer
of Silica Gel HR using hexane/diethyl ether/acetic
acid (85: 15:2, v/v/v) as developing solvent. The
bands are the lipids, visualized by spray application
of 50% a aqueous sulfuric acid. TG, triglyceride.

simple lipids from the whole fat body of the mature
female Rana catesbeiana qualitatively indicate that
a relatively large amount of TG is present at each
of various stages of development, as described
above (Fig. 6).

Quantitative data of lipid classes

The alternations in various components of phos-
pholipids and simple lipids are shown in Table 2.
The quantitative data of phospholipid classes, PE,
PS, PI, PC, Sph and LysoPC, at the 5 developmen-
tal stages of Taylor and Kollros (TK), V, X, XX,
XXI and XXII-XXIII, were obtained by the
routine method previously described. The quan-
titative data of simple lipid classes, FFA, TG, Chol
and Chol.E, at the same five developmental stages
were also obtained following the same routine
method.

The total simple lipid content of total lipid
remained essentially unchanged between TK
stages V and X at ca. 33% during development. It
gradually increased during metamorphic climax
stage, becoming about 70% at TK stages XXII-
XXIII.

The quantitative data for the total phospholipid
amount in total lipid also remained unchanged
between TK stages V and X, at ca. 66%, during
development. However, in contrast with the quan-
titative data for the simple lipid fraction, the data
for total phospholipid gradually decreased during
metamorphosis and the content was found to be
ca. 29% at TK stages XXII-XXIII. During meta-

TaBLE 2. Analysis of the lipid composition from the tadpole tail during metamorphosis of Rana catesbeiana

Developmental stages of Taylor and Kollros

Vv x XX XXI XXII-XXIII
Total simple lipid (% total lipid) 32.6+2.8 3310523 55:43:44 603223 70nEeSm
Total phospholipid (% total lipid) 66.6+3.4 66.142.3 43.341.5 38.5+1.2 28.7+1.9
Simple lipid classes (% total lipid)
Free fatty acid 15-1-E0:6 12°32 0!) 30!4=2 1.6 29.822 1S sero
Cholesterol 10-1+0:4 12°80:7 11.8016 14°20 aisneeus
Triglyceride 3.5+0.2 4.2+0.2 9°23-0°6 11.7 3: 0:6) Seisieecuny
Cholesterol ester 3.9+0.3 3.7+0.2 4.0+0.3 4.6+0.4 5.5+0.4
Phospholipid classes (% total phospholipid)
Phosphatidylcholine 45.6+0.5 482+1.2 45.14+2.1 43.740.8 41.2+1.8
Phosphatidylethanolamine 33:20:9 30:0221°3) | 33.82-1-4 — 3il 3=- 0 Se2ersealee
Phosphatidylserine 7.9 O87 5.0+0.2 4.3+40.8 5.5+0.6 5.0+0.7
Phosphatidylinositol 2.7+0.6 4.0+0.5 4.0+0.6 4.6+0.4 4.2+0.5
Sphingomyelin 7.6+0.5 7.3+0.4 8.1+0.8 9.8+0.6 13.7+0.9
Lysophosphatidylcholine 3.0+0.3 Sy osetl 4.7+0.5 Spe 09 7.6+0.8

The data are expressed as the weight percentage.

The values are mean+SD of triplicate specimens.

Lipids in the Tadpole Tail 415

morphosis, there was a compensatory relationship
in amount between the simple lipid and phospholi-
pid fraction.

As shown in Table 2, a large amount of free
fatty acid was detected. The weight percentage of
free fatty acid in total lipid was about 12 to 15%
during the premetamorphic stage from TK stages
V and X increased dramatically to about 30 to 38%
during the metamorphic climax stage from TK
stage XX to XXII-XXIII. The weight percentage
of triglyceride was about 4% at the premetamor-
phic stage from stage TK V to X. Though the ratio
of triglyceride in total lipid was not as large as that
of free fatty acid, the relative percentage of both
substances dramatically increased after TK stage
XX. The ratio of cholesterol to total lipid was
approximately 10 to 13% at TK stages V and X,
and did not significantly change at TK stage XX to
XXII-XXIII. Cholesterol ester was detected at
each TK stage from V to XXII-XXIII: the ratio of
cholesterol ester to total lipid was about 4 to 6%,
and significant changes were not detected through-
out the metamorphisis. Phosphatidylcholine and
phosphatidylethanolamine were the major compo-
nents of phospholipid classes. Phosphatidylcholine
occupied about 41 to 48% of total phospholipids,
while phosphatidylethanolamine occupied about
28 to 34% throughout TK stage V to XXII-XXIII.
The ratio of phosphatidylcholine and phosphatidy-
lethanolamine to total phospholipids was relatively
stable. The ratio of sphinogomyelin to total phos-
pholipids did not change during TK stage V to X,
while afterward it increased slightly until TK stages
XXII-XXIII. The ratio of phosphatidylserine,
phosphatidylinositol and lysophosphatidylcholine
did not significantly change from TK stage V to
XXII-XXII.

DISCUSSION

From the present study, it is evident that
glycerol-based phospholipids including phospha-
tidylcholine, phosphatidylethanolamine, phospha-
tidylserine and phosphatidylinositol are predomi-
nantly present among the lipids in larval tissues
and in eucaryotic cell membranes [24, 39] (Fig. 2,
Table 2). Sphingomyelin (sphingosine-based
lipids) and cholesterol are also major components

of the cell membrane. In this animal, triglycerides
may also play roles in energy storage as described
for other animals [40] in whose adipose tissue they
are present (Figs. 4, 5, 6). Since many biochemical
alternations that occur during the metamorphosis
of anuran tadpoles have been shown to be in-
fluenced by thyroxine [1, 3, 5, 6, 41], significant
alternations in lipids during growth and meta-
morphosis may also be reasonably considered to
take place under its influence. Thyroid hormone
acts directly on each type of cell to induce two
different events; death of epidermal cells, mesen-
chymal fibroblasts and probalby muscle cells, and
activation of macrophages during metamorphosis
[5, 6].

Lipid seems to be an important constituent of
the tadpole body and the values for percentage
total lipids and neutral lipids also increase during
pre- and prometamorphosis; this is followed by
significant decease during the metamorphic cli-
max. Phospholipids, on the other hand, apparent-
ly do not undergo significant variation on a percen-
tage body weight basis, though during larval de-
velopment, increase in their amounts is discernible
[19]. Urbani [17] observed a fourfold decrease in
total lipid content during metamorphosis. Sawant
and Varute [19] showed such decrease to be due to
that in various constituents of neutral lipids and
phospholipids.

During the climax period, when tadpoles do not
feed and histolytic events such as degeneration of
internal gills, skin degeneration in fore limb win-
dow formation and tail regression occur maximal-
ly, lipid content decrease sharply, especially that of
neutral lipids such as triglycerides that accumlate
during development and endogenous energy alone
is available.

The decrease in total phospholipids during
metamorphosis may quite likely be due to the
breakdown of tail cell membranes whose phospho-
lipids form integral components in the histolytic
events of metamorphosis. This is supported by the
fact that fatty acids are not present in the early
stages of metamorphosis. The catabolic break-
down of phospholipids leads to the formation of
fatty acids and nitrogenous bases [23]. Phospholi-
pid degradation in E. coli has been summarized by
Rock and Cronan [42] as follows. Phospholipid

416

degradation, i.e., the hydrolysis of fatty acids from
the 1-position and 2-position of phospholipids, is
hydrolyzed by phospholipase A, and in particular,
the hydrolysis of fatty acids from the 1-position
occurs most rapidly in the cells of E. coli. Phos-
pholipase A is located in the soluble fraction
(cytoplasm) and outer cell membranes of E. coli.
Lysophospholipase is located in the soluble frac-
tion and inner cell membranes of E. coli. The
biochemical and physiological functions of these
degradative enzymes remain unknown in amphi-
bian cells during development and metamorphosis.
However, a similar phenomenon of phospholipid
degradation by these enzyme in macrophages may
possibly occur during the regression of tadpole
tails in Rana catesbeiana.

The present study demonstrates that there may
possibly be a correlation between the regressive
process of this tadpole tail and the presence of
relatively large amounts of free fatty acids during
metamorphosis. It would thus follow these free
fatty acids may mainly be those hydrolyzed from
the 1-position and 2-position of phospholipids dur-
ing metamorphosis. It is generally known that the
transport of fatty acids is a major function of
triglycerides in mammals [40]. In this animals, free
fatty acids may be stabilized by association with
albumin or similar proteins, and albumin or similar
proteins may carry fatty acids from the tail to the
liver or fat body. These fatty acids may be
regarded as triglyceride components in a fat body
which is produced in this tadpole during meta-
morphosis. Triglycerides may play major roles in
energy storage as deposits of adipose tissue. Addi-
tional research should be conducted for greater
clarification of the functions of phospholipid de-
gradative enzymes in the regressive process of
tadpole tails during metamorphosis.

Quantitative and qualitative data on glycolipids
in the tadpole tail during metamorphosis in Rana
catesbeiana will be reported in the next publica-
tion.

ACKNOWLEDGMENTS

The author is especially indebted to M. S. Ralph
Johnston for his critical review of the manuscript.

10

11

12

13

M. RyuZAkI AND M. OoNUKI

REFERENCES

Bennett, T. P. and Frieden, E. (1962) Metamorph-
osis and biochemical adaptations in amphibia. In
“Comparative Biochemistry”. Vol. 1VB, Ed. by M.
Florkin and H. S. Mason, Academic Press, New
York, pp. 483-556.

Brown, G. W. Jr. (1964) The metabolism of amphi-
bia, In “Physiology of Amphibia”. Ed. by J. H.
Moore, Academic Press, New York, pp. 1-98.
Frieden, E. (1968) Biochemistry of amphibian
metamorphosis. In “Metamorphosis: A Problem in
Developmental Biology”. Ed. by W. Etkin and L. I.
Gilbert, North-Holland, Amsterdam, pp. 349-398.
Weber, R. (1967) Biochemistry of amphibian meta-
morphosis. In “The Biochemistry of Animal De-
velopment”. Vol. II, Academic Press, New York,
pp. 227-252.

Fox, H. (1984) Metamorphosis in anurans. In
“Amphibian Morphogenesis”. Ed. by H. Fox, Hu-
man Press Inc., Clifton, pp. 9-16.

Yoshizato, K. (1986) How do tadpoles lose their
tails during metamorphosis? Zool. Sci., 3: 219-226.
Brown, G. W. Jr., Brown, W. R. and Cohen, P. P.
(1959) Comparative biochemistry of urea cycle
enzymes in metamorphosing Rana catesbeiana tad-
poles. J. Biol. Chem., 191: 377-378.

Trader, C. D., Wortham, J. S. and Frieden, E.
(1963) Haemoglobin: molecular changes during
metamorphosis. Science, 139: 918-919.

Manwell, C. (1966) Metamorphosis and gene ac-
tion-I. Electrophoresis of dehydrogenases, ester-
ases, phosphatases, haemoglobin and other soluble
proteins of tadpoles and adults of bull frog. Comp.
Biochem. Physiol., 17: 805-825.

Herner, A. E. and Frieden, E. (1960) Biochemistry
of anuran metamorphosis-VII. Changes in serum
proteins during spontaneous and induced meta-
morphosis. J. Biol. Chem., 235: 2845-2851.

Wald, G. (1960) The distribution and evloution of
visual pigments. In “Comparative Biochemistry”.
Vol. I, Ed. by M. Florkin and H. S. Mason,
Academic Press, New York, pp. 311-345.

Tata, J. R. (1966) Requirement for RNA and
protein synthesis for induced regression of the tad-
pole tail in organ culture. Devel. Biol., 13: 77-94.
Weber, R. (1964) Ultrastructural changes in regres-
sing tail muscles of Xenopus larvae at metamorph-
osis, J. Cell Biol., 22: 481-487.

Varute, A. T. and More, N. K. (1971)
Glucuronidase in alimentary canal of tadpoles of
Rana tigrina in growth and metamorphosis. Comp.
Biochem. Physiol., 38B: 225-233.

Varute, A. T. (1972) Histoenzymological of {-
glucuronidase in the resorbing tails of tadpoles of
Rana tigrina at metamorphosis. Acta Histochem.,

16

17

18

19

20

21

22

23

24

25

26

27

28

Lipids in the Tadpole Tail

41: 306-324.

Urbani, E. (1962) Comparative biochemical studies
on amphibian and invertebrate development. In
“Advances in Morphogenesis”. Vol. II, Ed. by M.
Ambercombie and J. Brachet, Academic Press,
New York, pp. 61-108.

Urbani, E. (1956) Lipases and lipids in the
embryonic and larval developments of Bufo vulgaris
and Rana esculenta. Atti Acad. Nazl. Lincei, 21:
498-503.

Light, R. L. and Waschek, C. C. (1965) Liver fatty
acids of Rana grylio tadpoles and frogs. Comp.
Biochem. Physiol., 15: 167-173.

Sawant, V. A. and Varute, A. T. (1973) Lipid
chages in the tadpoles of Rana tigrina during- growth
and metamorphosis. Comp. Biochem. Physiol.,
44B: 729-750.

Okamura, N. and Kishimoto, Y. (1983) Changes in
nervous system glycolipids during metamorphosis of
Xenopus laevis. J. Biol. Chem., 258: 12243-12246.
Yates, A. J., McGill, J. M., Markowitze, D. L. and
Tassava, R. A. (1985) The patterns of brain gang-
liosides of Rana catesbeiana during metamorphosis
and in adult. Devel. Biol., 110: 255-258.

Ansell, G. B. and Howthrone, J. N. (1964) Phos-
pholipids-Chemistry, Metabolism and Function.
Elsevier Scientific Publ. Co., London, pp. 1-427.
Dawson, R. M. C. (1966) The metabolism of animal
phospholipids and their turnover in cell membranes.
In “Essays in Biochemistry”. Vol. II, Ed. by C.
Chapman, Academic Press, New York, pp. 59-115.
Ansell, G. B., Dawson, R. M. C. and Howthrone,
J. N. (1973) Form and Function of Phospholipids.
Elsevier Scientific Publ. Co., Amsterdam, pp. 1-
494.

Hilditch, T. P. (1956) The Chemical Constitution of
Natural Fats. 3rd Edition, Ed. by T. P. Hilditch,
John Wiley, New York, pp. 25-145.

Taylor, A. C. and Kollros, J. J. (1946) Stages in the
normal development of Rana pipiens larvae. Anat.
Rec., 94: 7-24.

Ryuzaki, M., Kojima, H. and Tamai, Y. (1975)
Study on amphibian lipids II. Characteristic consti-
tuent of monoglycosylceramides from the skin of
three frog species. Comp. Biochem. Physiol., 52C:
81-84.

Folch, J., Lee, M. and Sloane-Stanley, G. H. (1957)

29

30

31

32

33

34

35

36

37

38

39

41

42

417

A simple method for the isolation and purification of
total lipids from animal tissues. J. Biol. Chem., 226:
497-509.

Skipski, V. P., Peterson, R. M. and Barvlog, M.
(1964) Quantitative analysis of phospholipids by
thin-layer chromatography. Biochem. J., 90: 374-
378.

Dittmer, J. C. and Lester, R. L. (1964) A simple,
specific spray for the detection of phospholipids on
thin-layer chromatograms. J. Lipid Res., 5: 126-
127.

Vaskovsky, V. E. and Kostatsky, E. Y. (1968)
Modified spray for the detection of phospholipids on
thin-layer chromatograms. J. Lipid Res., 9: 396-
401.

Waldi, D. (1965) Spray reagents for thin-layer
chromatography. In “Thin-Layer Chromatogra-
phy”. Ed. by E. Stahl, Academic Press, New York,
pp. 483-484.

Marinetti, A. V. (1966) Chromatographic separa-
tion, identification and analysis of phosphatides. J.
Lipid Res., 3: 1-20.

Mangold, H. K. and Kammereck, R. (1962) Thin-
layer chromatography of lipids. J. Am. Oil Chem.
Soc., 39: 201-205.

Weicker, H. (1959) Specific detection tests for
neutral lipids. Klin. Wochscur., 37: 763-765.
Bartlett, G. R. (1957) Phosphorus assay in column
chromatography. J. Biol. Chem., 234: 466-468.
Snyder, F. and Stephens, N. (1959) A simplified
spectrophotometric determination of ester groups in
lipids. Biochim. Biophys. Acta, 34: 244-245.

Zak, B. (1957) Simple rapid microtechnic for serum
total cholesterol. Am. J. Clin. Path., 27: 583-588.
Law, J. H. and Snyder, W. R. (1972) Membrane
lipids. In “Membrane Molecular Biology”. Ed. by
D. E. Vance and J. E. Vance, Benjamin/Cummings
Publ. Co., Menlo Park, pp. 213-241.

Etkin, W. (1968) Hormonal control of amphibian
metamorphosis. In “Metamorphosis: A Problem in
Developmental Biology”. Ed. by W. Etkin and L. I.
Gilbert, North-Holland, Amsterdam, pp. 313-438.
Rock, C. O. and Cronan, J. E. (1985) Lipid
metabolism in procaryotes. In “Biochemistry of
Lipids and Membranes”. Ed. by D. E. Vance and J.
E. Vance, Benjamin/Cummings Publ. Co., Menlo
Park, pp. 73-114.

219 :JOrtl 1. Bourn ay Cases pitt hae liigiad
) pe OER OB
: a ir noi Siett av"

aw ifada- To ey
7 Om ont

i

$1, .watee tl beet yw)

uf) Set Ve

Peng
j 4

Mie GE 1 ie
rid th ean
imap snomt

Hub

pO hay ORG

mth. -F Ab eet
FDR) + etic
Lapeer

kup

ci! hae Te
tl ee

t t yh pe a

Gap
warts yolel nV
: f- ' lier Sey
DpH
ya, eee

pyte ten Nrpa Fee teats

VrRktnt”

efi al

hot eon ( Pye Ei. gtk ay

ae ae

FoAte

jest’ mio otk ae

eg Carico dad oy gt all bettas onan men

th

a

»

rs

. Re

pane. .20,,0
a

" tive UP sind pale

ri. eons | BaD 38
in PR kA RES!
GAR ed eA oni mesic: cag’
vive es den

bey

rat. cat antiail !
G. Ageia
i were: LER). ¥ .cramutaet a

.z PAY wee eee garcuh gf

B25 ,. Sahat

a adie ails 17
S10 (FF t ita AY iS nat aoe
KALKY = ral ait ays sige Sis —
— tee in
wt phy it e i
hbion Gos maitodas

lta A .viNy staid, ok
haste, 22s ek EGON pet ny ay:

Suny “nt a) P19 YOR

=
miro beg.)
deat) bo an tact, tun

a eb
ly tar tepeette D bauaedh ; wat, ‘(Rh
: ‘

jet ssvitnoch §

es enh.

‘ie
oe

*
. ‘ - A
{£3 n a iy ¢ » ody

‘, : op Aters

(Voi onett Y tray

dh wwik? taal),
$f 9
i"

ito Aes
“ameotal.!) AL wig psig

ul La FE gees oye Uy m0
P eta amd re

ceili tov.
i

lugging bre Moe
‘ rs
mrt,
|
’
}
‘ i?
we j
Sarton par

ZOOLOGICAL SCIENCE 7: 419-425 (1990)

Inheritance of the Color Patterns of the Blue Snakeskin
and Red Snakeskin Varieties of the Guppy,
Poecilia reticulata

V.P. E. PHanc, A. A. FERNANDO! and E. W. K. Cua

Department of Zoology, National University of Singapore, Kent Ridge,
Singapore 0511, and ‘Freshwater Fisheries Section, Primary Production
Department, Sembawang Road 17 km, Singapore 2776

ABSTRACT—The Blue Snakeskin (BSS) and Red Snakeskin (RSS) varieties are popular strains
commercially cultured in Singapore. The blue-black tail color of BSS guppies is determined by a
dominant X-linked gene (Bi?) and the silvery snakeskin pattern on the body of males is under the control
of a Y-linked gene (Ssb). The Y-linked snakeskin tail pattern gene (Sst) though present in BSS males is
masked by the blue-black tail color gene (Bit). The putative genotypes for males and females of the BSS
variety are XgYssb,ss1 ANd Xgy,Xpr, respectively. The red tail color of the RSS variety is due to an
X-linked dominant gene (Rdt). The snakeskin body pattern of RSS males is under the control of the
Y-linked Ssb gene while the black reticulations on the tail is due to interaction between the snakeskin
tail pattern gene (Sst) and the red tail gene (Rdt). The proposed genotypes for males and females of the
RSS variety are XpaY ssb,ssr aNd XruXra, respectively. An estimate of 0.9% crossover frequency was
obtained between the Y-linked Ssb and Sst genes and a 2.7 % crossover rate of the Bit gene from the X-

© 1990 Zoological Society of Japan

to the Y-chromosome.

INTRODUCTION

The guppy, Poecilia reticulata being a voracious
omnivorous feeder and tolerant of polluted waters
was introduced to Singapore and other parts Asia
for mosquito control [1]. It has been commercially
cultured in Singapore since the 1950’s and is eco-
nomically the most important species of freshwater
ornamental fish produced almost exclusively for
the export market. The wide variation of brilliant
and beautiful colors on the body, tail and dorsal fin
of male guppies makes it one of the most popular
and ubiquitous ornamental fish. About 30
domesticated varieties of guppies are produced on
monoculture farms in Singapore. Each farm spe-
cialises in 8-12 varieties [2]. Guppy farmers con-
tinuously strive to improve the quality of the fish
and to breed new varieties with novel color pat-
terns. To do so it is important to understand the
gene control of the color phenotypes of domesti-
cated guppies. So far there are few reports on the

Accepted July 27, 1989
Received March 7, 1989

inheritance of color patterns of domesticated
varieties of the guppy [3-7]. The objective of this
study is to elucidate the gene control of the color
phenotype of the Blue Snakeskin and Red
Snakeskin varieties of P. reticulata cultured in
Singapore.

MATERIALS AND METHODS

Source of the fish

Two- to three-weeks old fry of the Blue
Snakeskin (BSS) and Red Snakeskin (RSS)
varieties were obtained from the Lim Chin Lam
Guppy Farm in Singapore. Wild-type (WT) fry
were collected from a stream in a rural area of
Singapore. Since virgin females were required for
the reciprocal crosses, fry were raised in 33 liter
clear plastic tanks (20 fish/tank) in the aquarium
area of the Department of Zoology, National
University of Singapore, at temperatures of 26-
28°C. Sexual differentiation takes place at 4-6
weeks of age under laboratory conditions. The
young fish were checked daily for developing

V. P. E. PHANG, A. A. FERNANDO AND E. W. K. Cuia

420

Color Pattern Inheritance in the Guppy 421

males which are recognized by the modification of
the anal fin into the gonopodium. Males when
spotted were immediately removed and raised
separately from females.

Description of the varieties

Adult males and females of the BSS and RSS
varieties have total length of 4-S cm. Adult RSS
males have iridescent snakeskin-like reticulations
on the olive-brown body and an orange-red col-
ored tail with black reticulations (Fig. 1A). BSS
males are also characterized by silvery snakeskin-
like pattern on the body but the tails are blue-black
in color (Fig. 1B). Both RSS and BSS females lack
the snakeskin patterns. RSS females have the
normal wild-type olive brown body coloration with
tinges of red and opaque white on the tail. BSS
females have the WT body coloration and partial
expression of the blue-black tail color.

Wild-type guppies are smaller than the domesti-
cated varieties, with marked size differences be-
tween males and females. Adult WT females are
3.0-3.5 cm long while adult males are about 2 cm
long. WT males have highly polymorphic color
patterns, consisting of spots or patches of. various
colors on the body, tail and sometimes the dorsal
fin while the females lack color patterns (Fig. 1C).

Reciprocal crosses

The inheritance of the color phenotypes of the
BSS and RSS varieties were elucidated by per-
forming single-pair reciprocal crosses between
each of them and the wild-type stock, using 3-
month old sexually mature virgin fish. The pairs
were kept in eight liter breeding tanks. The
following notations were used for the crosses:

Cross 1A: BSS malexWT female, Cross

1B: WT male XBSS female

Cross 2A: RSS malexWT female,

2B: WT malex RSS female

Cross

Broods were usually produced 4-6 weeks after
mating. F, broods were obtained from single-pair
matings between full-sib F, fish. Phenotypic pro-
portions among the F, and F, progeny were sub-
jected to chi-square tests.

Twenty adult females of the BSS and RSS
parental stocks and all the F, and F, female
progeny that were not required for breeding were
fed with the androgen, methyl testosterone, to
express any inherent color genes [8, 9].

RESULTS

Androgen treatment of BSS and RSS females

When androgen treated the tail color of BSS
females deepened to a dark blue no snakeskin
pattern was manifested on the body. Androgen
treated RSS females developed red color on the
tail but not the iridescent snakeskin body pattern
or the black reticulations on the tail. These results
showed that the BSS and RSS females carried the
genes for the blue tail color and red tail color,
respectively, but not the genes determining the
snakeskin pattern.

Cross between the BSS variety and WT stock

Twelve matings between BSS males and WT
females (Cross 1A) gave a total of 120 male and
150 female F, progeny (Table 1). The F; males
had the hyaline tail color of WIT guppies and a
delicate, silvery snakeskin pattern on the body and
tail (Fig. 1D). We called the color phenotype of
these F, males wild-type snakeskin (WTSS). The
F, females had blue tinges on the tail which on
androgen treatment deepened in color but no
snakeskin pattern developed the body or tail.
These females were cinsidered to have the blue tail
phenotype (BT). Since only the male progeny
inherited the snakeskin pattern from the BSS male

Fic. 1.A.
Fic. 1.B.
Fic. 1.C.
Fic. 1.D.

Three adult male wild-type guppies.

Adult male (upper) and female (lower) of the Red Snakeskin variety of the guppy.
Adult female (left) and male (right) of the Blue Snakeskin variety of the guppy.

A F;, male of the cross between BSS males and WT females showing delicate iridescent snakeskin pattern

on the body and tail and hyaline wild-type tail (wild-type snakeskin phenotype) (upper). A F, female of this cross

(lower).

Fic. 1.E. A F, male of the cross between WT males and BSS females showing blue tail (BT) phenotype.
Fic. 1.F. Two F, male of the cross between WT males and RSS females showing the red tail (RT) phenotype.

422 V.P. E. PHANG, A. A. FERNANDO AND E. W. K. CuIA

TABLE 1. F, and F, segregation data of reciptrocal crosses between the Blue Snakeskin variety and the wild
type stock
Cross Gan No. matings No. & Phenotypes of Progeny Exp. 2
(Gross No.) (No. broods) Males Females Tatio x
BSS {xX WT ? F, 12 (12) 120 WTSS 150 BT seal 3.33
(Cross 1A) AAS
F, 10 (15) 46 WISS 51 BT Bate 1 il 1.92
60 BSS 53 WT
82 WISST
WT SxXBSS ? F, 12 (12) 97 BY 108 BT 1:1 0.59
(Cross 1B) F, 12 (15) 76 BT 178 BT *1:1:2 2.86
72 WT #5 WT

5 Exceptional F, males with wild type body and snakeskin tail pattern (Sst).
* Exceptional F, females (Cross 1B) with hyaline tail color.

* Exp. ratio for the typical F, offspring of Cross 1B.

parent, it showed Y-linkage of the genes deter-
mining the snakeskin pattern. The presence of
snakeskin reticulations on the tail of F, males
showed that the BSS male parents were carrying
the gene for snakeskin tail pattern which was
masked by the blue-black tail color. The snakeskin
body and tail patterns of another guppy variety,
the Green Snakeskin, are controlled by two closely
linked genes (Ssb and Sst) on the Y-chromosome
[10]. Absence of the blue tail color of the BSS
male parents in F,; males and presence in all F,
females gave evidence of X-linkage and domi-
nance of the gene determining the blue tail colora-
tion which has been designated as Bit [11]. The
recessive allele, B/t* , present in WT guppies gives
the hyaline tail color.

The F> generation of Cross 1A cinsisted of BSS
males, WISS males, BT females and WT females
with observed numbers conforming to the 1:1:1:1
expected ratio (Table 1). These results gave evi-
dence that the putative color genotype of BSS
males is XgrYssb,ss1 and that for WTSS males is
Xput Yssb,ss- MA genetic model is proposed to
show segregation of the color genes in Cross 1A
(Fig. 2).

However, there were two exceptional F, males
with WT body color and snakeskin pattern on the
tail giving further evidence that the snakeskin body
and tail patterns are determined by two Y-linked
genes, Ssb and Sst, respectively. The absence of
the expected snakeskin body pattern in these two

males is probably due to crossing-over of the Ssb
gene irom the Y- to the X-chromosome. The
crossover frequency between the Ssb and Sst genes

calculated from the F, offspring of Cross 1A is
0.9%, two crossovers out of a total of 212 F,
individuals.

Twelve matings of the reciprocal cross (Cross
1B) between WT males and BSS females gave 12
F, broods consisting of 97 males and 108 females,
all with blue tails (BT phenotype) and without any
snakeskin pattern (Fig. 1E). With the exception of
five females with the WT phenotype, the F, proge-
ny of this cross segregated into BT males, WT
males and BT females according to the 1:1:2
hypothetical ratio. Thus the F, and F> results gave
evidence that the BSS parental females were
homozygous for the X-linked dominant blue tail
gene (Bit) with the genotype being XgyXpy. Fig-
ure 2 shows the proposed genetic model for the
segregation of color genes in Cross 1B.

The occurrence of five exceptional F, females of
Cross 1B, with hyaline tails instead of the expected
blue tail color of the typical F, females could be
due to crossing over of the X-linked Bit gene to the
Y-chromosome in the F; male parents of these
individuals. Since there were five crossover
females out of a total of 183 F, females of Cross
1B, the crossover frequency of the Blt gene from
the X- to the Y-chromosome in the F; females of
this cross is 2.7%. Crossovers among the F males
cannot be detected (Fig. 2).

Color Pattern Inheritance in the Guppy

CROSS 1A

P BLUE SNAKESKIN MALE

xX
X1t Sah. Sst |

WILD TYPE FEMALE

Xie" *sie*

1 x

%it “pret

Blue Tail 7?

it* ‘seb, set
Wild Type Snakeskin 88

Xx
Bit “itt
Blue Tail ??

x
Blt YS eb, Sst
Blue Snakeskin 6m

x
Bit* ‘ssb, set Xsit* *pze*

Wild Type Snakeskin 8 | Wild Type $f

Fo phenotypic ratio
1 Bss 6: 1 wrss 6: 1 Br $$: 1 wr 3

423

CROSS 1B

WILD TYPE MALE BLUE SNAKESKIN FEMALE
xX

Xa1r* ‘seb’, sat* Xsit “pie

"sre “pitt
Blue Tail $9.

X51t ‘seb*,set*
Blue Tail o&

Xoie *p1e
Blue Tail 99

%51t ‘ssb*,set*
Blue Tail 6

X512* ‘sept, set*

Wild Type &% Blue Tail ??

Py phenotypic ratio

1 Br 68': 1 wr OS: 2 Br ee

Fic. 2. Schematic diagram of the proposed genetic model for the segregation of color genes in the F, and F,
generations of reciprocal crosses between the Blue Snakeskin variety and wild-type guppies.

TABLE 2. F, and F, segregation data of reciprocal crosses between the Red Snakeskin variety and the wild

type stock
Cross G No. matings No. & Phenotypes of Progeny Exp. 2
(Gross No.) or (No. proces) Males Females ratio x
BSS {xX WT ? F, 6 (42) 117 WTSS 135 BT ie il 1.28
(CRESS 2) FE 5 (13) 74 RSS 98 RT iigitenen | eS
76 WISS 80 WT
WT J XBSS 2 F, 5 (10) 94 RT 122 RT 1:1 3.63
(Cross 2B) F, 6 (12) 56 RT 140 RT 2 2.21
70 WT
*WT Sf XRSS 7 F, 1 GB) 16 RT 22 RT eile ileal 4.06
(Css 2d) 10 WTSS 12 WTSS

* Exceptional mating between a RSS ? producing three exceptional broods, giving evidence that the RSS female

parent was heterozygous for the Ssb and Sst genes.

Cross between the RSS variety and WT stock

Six matings between RSS males and WT females
(Cross 2A) produced 12 F, broods (Table 2). The
117 F, males had hyaline tails and iridescent
snakeskin pattern on the body and tail (WTSS
phenotype) like the F; males of Cross 1A. The 135
F, females showed pink tinges on the tail which
deepened to red after androgen treatment (RT

phenotype). The pooled F, data conformed to the
1 WTSS male: 1 RT female. Thus, results showed
that the RSS parental males carried the dominant
X-linked red tail gene (designated as Rdt) which
they passed to their daughters and the Y-linked
snakeskin body (Ssb) and snakeskin tail (Sst) genes
which were transmitted to their sons. The reces-
sive tail color allele, Rdt* gives the hyaline tail
color. The F, progeny of this cross segregated into

424 V.P. E. PHANG, A. A. FERNANDO AND E. W. K. Cua

CROSS 2A
Pp RED SNAKESKIN MALE WILD TYPE FEMALE

Xratt Xpae*

Xnat ‘seb, Sst

Y
+ That* ‘seb, set
Wild Type Snakeskin 60”
% X.Y 5 ED ¢
Rdt Ssb,Sst Rat “pat*
Red Snakeskin 60" Wild Type $3
Xnat* "seb, set pat *nat*
Wild Type Snakeskin 66 Red Tail 3}
KR phenotypic ratio
1 Rss 6&: 1 wrss o@: 1 WI 99 + 1 RT 99
Fic. 3.

CROSS 2B
WILD TYPE MALE RED SNAKESKIN FEMALE
x

Xnat* ‘seb set* | Xnat “pat

Xpat “pact
Red Tail $}

Xrat ‘seb set*
Red Tail de

Xrat “pat
Red Tail 3}

Xeat ‘seb*,sst*
Red Tail 60°

Xratt ‘sept set*

Wild Type 6@

Xnas? *pae
Red Tail $+

FP, phenotypic ratio

ret 8: iw &: 2 a Y

Schematic diagram of the proposed genetic model for the segregation of color genes in the F, and F,

generations of reciprocal crosses between the Red Snakeskin variety and wild-type guppies.

four phenotypic classes: red snakeskin males
(RSS), wild-type snakeskin males (WTSS), red tail
females (RT) and WT females with observed num-
bers conforming to the expected 1:1:1:1 ratio. The
putative genotype of RSS males is XrwY5sp, ss¢ and
that for WTSS males is Xpq+ Yssb,5s- A genetic
model is proposed to show the segregation of color
genes in Cross 2A (Fig. 3).

Five matings of the reciprocal cross of WT males
with RSS females (Cross 2B) gave 10 typical F;
broods consisting of red tail (RT) males and
females with numbers conforming to the expected
1:1 segregation ratio (Fig. 1F). The phenotypes of
the females were expressed after androgen treat-
ment. The 12 F> broods consisted of 56 RT, 70 WT
males and 140 RT females. These observed num-
bers fit an expected segregation ratio of 1:1:2 and
provided evidence that the RSS parental females
were homozygous for the X-linked dominant red
tail gene, Rdt (XraXra genotype).

A single atypical mating of Cross 2B gave three
exceptional F, broods consisting of four pheno-
typic classes: expected RT males, unexpected
WTSS males, expected RT females and unex-
pected WTSS females. The observed numbers
segregated according to the hypothetical 1:1:1:1
ratio (Table 2). These results give evidence that
the exceptional RSS female parent of this mating
carried the Ssb and Sst genes on one of the
X-chromosomes.

DISCUSSION

The blue-black tail color of the BSS variety is
found to be controlled by the X-linked Bir gene
and the red tail color of RSS guppies by the
X-linked Rdt gene. There is similar evidence of
X-linkage of the blue tail and red tail color tail
genes present in the domesticated Blue Tail variety
and Red Tail variety, respectively [6]. Fernando et

Color Pattern Inheritance in the Guppy

al. [11] also reported Y-linkage of these genes in
domesticated varieties from one farm in Singa-
pore. In this study 2.7% of crossing over of the Bit
gene from the X- to the Y-chromosome was found,
while no crossovers were observed for the Rdr
gene. Crossing over of genes from the X- to the
Y-chromosome and vice versa in the guppy was
first documented by Winge [12] and the crossover
data have been used to map the sex chromosomes
[Bs S5 We, 124)

In male guppies with wild-type hyaline tails
(Bit+-and Rdt+) the snakeskin tail pattern gene
(Sst) is expressed as iridescent delicate reticula-
tions. The Sst gene in males carrying the red tail
gene (Rd?) is manifested as black reticulations on
the orange-red tail. However, in males carrying
the Blt gene which gives the blue-black tail color,
the Sst gene when present is completely masked.
Outcrosses with WT females showed that the BSS
male guppies are carrying the Sst gene.

The 0.9% recombination frequency between the
Y-linked Ssb and Sst genes found among the F,
progeny of Cross 1A is in close agreement with the
1% reported in another domesticated guppy varie-
ty, the Green Snakeskin [10]. No crossovers were
found among the progeny of Cross 1A showing
that the Ssb and Sst genes tend to behave as a
supergene and are transmitted as a single unit on
the Y-chromosome through the male line. The
single RSS female parent of Cross 2B heterozy-
gous for the Ssb and Sst genes, gave evidence that
one or both these normally Y-linked genes could
be found on the X-chromosome due to crossing-
over. Our results also showed that the Bit, Ssb and
Sst genes are found on the homologous segments
of the sex chromosomes. Currently experiments
are being conducted to test for possible allelism of
the blue tail (B/t) and red tail (Rdt) color genes.

ACKNOWLEDGMENTS

This work was supported by a grant from the National
University of Singapore. We would like to thank Mr. K.
J. Goh and Mr. H. K. Yip for photography of the
guppies and Mrs. J. Mui for typing the tables.

10

11

12

13

14

425

REFERENCES

Herre, A. W. C. T. (1940) Additions to the fish
fauna of Malaya and notes on rare and little known
Malayan and Bornean fishes. Bull. Raffles Mus.
Sing., 16: 27-61.

Fernando, A. A. and Phang, V. P. E. (1985)
Culture of the guppy, Poecilia reticulata in Singa-
pore. Aquaculture, 51: 49-63.

Dzwillo, M. (1959) Genetische Untersuchungen an
domestizierten Stammen von Lebistes reticulatus
(Peters). Mitt. Hamburg Zool. Mus. Inst., 57: 143—
186.

Dzwillo, M. (1962) Uber kunstliche erzeung funk-
tioneller Mannchen weiblichen Genotyp bei Lebistes
reticulatus. Biol. Zentrabl., 81: 575-584.

Nayudu, P. L. (1979) Genetic studies of melanic
color patterns and atypical sex determination in the
guppy, Poecilia reticulata. Copeia, 2: 225-231.
Phang, V. P. E., Chow, O. K. and Fernando, A. A.
(1985) Genetic analysis of scale chromatophores of
two domesticated varieties of the guppy, Poecilia
reticulata. J. Sing. Nat. Acad. Sci., 14: 1-5.
Phang, V. P. E., Fernando, A. A. and Chow, O. K.
(1986) Inheritance of body and tail coloration in two
domesticated varieties of the guppy. In “Genetics in
Aquaculture II”. Ed. by G. A. E. Gall and C. A.
Busack, Elsevier, Amsterdam, p. 372.

Hildemann, W. (1954) The effects of sex hormones
on secondary sex characters of Labistes reticulatus.
J. Exp. Zool., 121: 1-5.

Fernando, A. A. and Phang, V. P. E. (1985)
Expression of sex-limited color pattern genes in
selected strains of female guppies Poecilia reticulata
with androgen treatment. Sing. Nat. Acad. Sci., 14:
157-159.

Phang, V. P. E., Ng, L. N. and Fernando, A. A.
(1989) Inheritance of the snakeskin pattern in the
guppy. J. Hered. (In press)

Fernando, A. A. and Phang, V. P. E. (1988)
Evidence for X- and Y-linkage of a tail colour gene
in a domesticated variety of the. guppy, Poecilia
reticulata. J. Sing. Nat. Acad. Sci. (In press)
Winge, O. (1923) Crossing over between the X- and
Y-chromosome in Lebistes. C. R. T. Lab. Carlsberg,
14: 1-19.

Winge, O. (1934) The experimental alteration of
sex chromosomes into autosomes and vice versa, as
illustrated by Lebistes. C. R. T. Lab. Carlsberg, 21:
1-49.

Winge, O. and Ditlevsen. E. (1947) Color inheri-
tance and sex determination in Lebistes. Heredity 1:
65-83.

viru <a or pnnateang er Wines. AeA

24 VaR
- ad a Te tee

t ni ’
ue Yee nna)

q i. tw (etn
[h<\o “et. oe

é £ tye) cae

Miu ok t rus?

a

fA

a) ee ae ae

(er wd) ber tar vO Brreeet bee

a) oe
vee

@ i

,
aay

ni eonsy Dents Ye yy abel
write th trad ana: » pra &

a Lemcapapicat at
ThA * yo! beyreelh i

att, 0 Pars att fs mon aot 4 te, dove

ane Vip ad). im sew” oy
coaors ath Lawe'[Soy og oc i
emg 0 kes! «tet seein por yoely
Hied sitileye® oerei-bliw bie
sag rein bed. mideage 6 cf
“iors: stssisb inaoeohiTt..e6 ;
ini bot ee ungyTIR? eolurg nt 2
eo arrobylPattet -atvalethe hte
siviice hah ie iat SP a0gH- ~ li
‘é jintraoakhoe wht oat =
thibdberistt gl Pe re
ERE oi) inet} bawode sper | M
onigg Wee oe a
1) hoowlid GS yah. (mt fi
4 gdh ance finn sot a2 ‘
dinate Wivaeke deol thal “ cy
“ureY VoquE bolssiiewndah pot
aw ziasocay Of fas : >a
entre ayy a0 insgorg &
‘ Sa O¢aet W OG). €9092 wt Lit
ey tinir alu » 26 baler sm]
tT Sekt ool Point Beal
euNiel) 42S 2a > iia
Suitoostbies vey vadady HA baal

=

Maing hry bg Hering vifadivnult r
etd pares gel BARRO recs? ‘
re A VA wettest hb 4eidl bie

iy A eAladonent edi mb is
i sroepe Pia F costs Cea val
comet Didheed ial tot cag

a)
\

euler As ies bow bare Ae

AT YSRO OOM. FAAS

Mi tehaw) Gosek rd Ranwoh tae .
tut wali Adee 9 ec +
ste) Wil + i , Ay H bite $h

ilwed 4 prey mil it; uM it ate

ZOOLOGICAL SCIENCE 7: 427-433 (1990)

© 1990 Zoological Society of Japan

Distribution of Immunoreactive Thyrotropin-Releasing Hormone
in the Brain and Hypophysis of Larval Bullfrogs
with Special Reference to Nerve Fibers
in the Pars Distalis

YUTAKA TANIGUCHI, SHIGEYASU TANAKA

and KAZUMASA KUROSUMI

Department of Morphology, Institute of Endocrinology,
Gunma University, Maebashi 371, Japan

ABSTRACT — Thyrotropin-releasing hormone (TRH) was immunohistochemically detected in the

brain and hypophysis of bullfrog larvae.

At embryonic stage 24 (Shumway’s classification), im-

munoreactive TRH was first detected in some fibers and perikarya in the hypothalamus, pars nervosa
and pars intermedia. As metamorphosis proceeded, TRH-immunoreactive perikarya as well as fibers
become conspicuous in several regions such as preoptic nucleus, infundibular nucleus, septum,
amygdala and diagonal band of Broca. Appearance of immunoreactive TRH fibers in the pars distalis
exclusively at Taylor-Korllos stage XIII-XVII was noted. The significance of this finding is discussed in

relation to metamorphosis.

INTRODUCTION

Although thyrotropin-releasing hormone
(TRH) has been purified and characterized early in
the 1970’s [1, 2], it was rather recent that the
precise distribution of this peptide within the brain
was elucidated. Leechan and Jackson [3] have
succeeded in demonstrating TRH in the histologic-
al sections of the rat brain using acrolein as the
fixative, and their protocol was applied to the
tadpole brain by Mimnagh et al. [4]. We have
investigated the development and localization of
immunoreactive TRH neurons in the brain and
hypophysis of larval bullfrogs, and found that a
few immunoreactive TRH fibers appear in the pars
distalis during a limited period of prometamorph-
osis. Siginificance of the temporal existence of the
immunoreactive TRH fibers in the pars distalis will
be discussed.

Accepted August 28, 1989
Received July 26, 1989

" Present address: Department of Anatomy, Wakaya-
ma Medical College, Wakayama 640, Japan.

* To whom all correspondence should be addressed.

MATERIALS AND METHODS

Larvae of the bullfrog (Rana catesbeiana) at
various developmental stages were collected from
the fields in the vicinity of Maebashi city. The
animals were staged according to Shumway [5] for
embryonic stages and Taylor and Kollros [6] for
metamorphic stages. After acclimatization in a
laboratory condition for 20 days, the animals were
decapitated and the brain with hypophsis was
carefully removed from the skull and fixed in 5%
acrolein in 1/15 M phosphate buffer (pH 7.4) for 3
hr at room temperature. The tissue was embedded
in paraplast after the dehydration with ethanol
series. Serial sagittal 4 “m-thick sections were cut
and immunohistochemically stained according to
the method of Leechan and Jackson [3] with minor
modifications. The deparaffinized sections were
treated first with 10 mM sodium metaperiodate in
phosphate-buffered saline (PBS) for 1.5 min, fol-
lowed by PBS wash. Then they were treated with
1% sodium borohydride in PBS for 1.5 min, fol-
lowed by thorough wash with PBS. Then the
sections were stained by the peroxidase-anti-
peroxidase (PAP) method: normal goat serum

428 Y. TANIGUCHI, S. TANAKA AND K. Kurosumi

Leg a
" aN, z :
Oo
% 4 ~~ e “
ee »
‘ ,

¢
° 3 a .
aa of!
b - a

ee

*
ty . «
e ‘ x
e's 7 hey
« eh ) , »
ON, © ;
ie . : ~
a >
+ J
oJ
‘
>
Pee, a
r ’ —_
2

ne

& . Hin ¢
+ Weare ae |
4 Syidy

TRH in Larval Bullfrogs 429

(1:20) for 2 hr, rabbit anti-TRH serum (1 : 3000)
for 14-18hr, PBS wash, goat anti-rabbit y-
globulin (1 : 200) for 2 hr, PBS wash, PAP complex
(1: 200) for 2 hr, PBS wash. The reaction product
was visualized with 0.02% 3,3’-diaminobenzidine
tetrahydrochloride and 0.005% HO, for 5-10
min. After washing with distilled water, the sec-
tions were stained with Methyl green, dehydrated
with an ethanol series, and mounted with Eukitt.

To check the specificity of the immunostaining,
the diluted anti-TRH serum was preabsorbed with
TRH at a final concentration of 10 «g/ml overnight
at 4° C prior to its use in the immunohistochemis-
try. The anti-TRH serum was a kind gift from Dr.
M. Mori, School of Medicine, Gunma University
[7], and PAP complex was purchased from Dakko-
pats, Denmark.

RESULTS

No immunoreactive TRH was detected in the

brain of the specimens at earlier stages than
embryonic stage 24 (body length ca. 8mm). TRH
immunoreactivity was first demonstrable in the
brain and pars nervosa at stage 24 (Fig. 1A). In
the brain, immunoreactive TRH nerve fibers were
found in the medial preoptic area and in the region
of infundibular nucleus. Some neuronal perikaya
in the region of putative preoptic nucleus were also
TRH-positive at this stage. In the hypophysis only
the putative pars nervosa was immunostained with
anti-TRH serum. At stage 25 (body length ca. 13
mm), immunoreactive TRH fibers ascending from
the ventral hypothalamus to the pars nervosa were
observed (Fig. 1B). The pars nervosa was diffuse-
ly immunostained with anti-TRH serum from this
stage onwards. As metamorphosis proceeds, the
immunoreactive TRH structures became more
widely distributed in the brain, including preoptic
nucleus, infundibular nucleus, septum, amygdala
and diagonal band of Broca, and the intensity of
the staining was increased (Fig. 2). Similarly, in

Fic. 2. Sagittal section of brain in bullforg tadpoles at stage XIII immunostained with anti-TRH serum. Many
perikarya and fibers were observed in the dorsal portion of preoptic nucleus (PON) (Fig. 2B). In the dorsal
infundibular nucleus (DIN), a dense fiber network were found. ME: Median eminence. Bar: 50 um.

Fic. 1.

Sagittal sections of embryonic brain and hypophysis in bullfrogs immunostained with anti-TRH serum. The

star indicates the third ventricle. The putative pars nervosa (arrow) was first stained at stage 24 (A). At stage 25
(B), in addition to the pars nervosa (arrow), some fibers are observed to ascent the future median eminence and

infundibular floor (short arrows).

melanin pigment (arrowhead). Bar: 50 um.

When adjacent section of B was immunostained with the preabsorbted
anti-TRH serum, immunoreactivity was completely abolished (C).

Note that embryonic cells contain much

430 Y. TANIGUCHI, S. TANAKA AND K. KuROSUMI

Fic. 3. Sagittal sections of hypophysis in bullfrog tadploes at stage XIII immunostained with anti-TRH serum. The
fibers are sometimes found to penetrate into the pars distalis (arrow) as shown in the inset. D: Pars distalis; I:
Pars intermedia; N: Pars nervosa. Bar: 50 um.

Fic. 4. A sagittal section of bulfrog hypophysis after metamorphosis completed (stage XXV). No TRH-fibers are
observed in the pars distalis (D), though in the pars intermedia (I) and in the pars nervosa (N) the
immunostaining remains as before (arrow). DIN: Dorsal infundibular nucleus; ME: Median eminence; PH:
Preoptic-hypophysial tract. Bar: 100 ~m.

TRH in Larval Bullfrogs 431

the pars nervosa the immunoreactivity became
intense. In the pars intermedia, numerous im-
munoreactive TRH fibers were observed from
stage 24 througout the process of metamorphosis.
Although embryonic cells contained much melanin
pigment, the reaction product of immunostaining
was easily distinguishable from the melanin gra-
nules by the difference in their colors. Moreover,
as shown in Figure 1C, these immunoreactions
were completely abolished after preabsorption of
the anti-TRH serum. Careful observations re-
vealed that at stages XIII-X VII in prometamorph-
Osis, a few immunoreavtive TRH fibers penetrated
into the pars distalis from the median eminence as
well as from the pars intermedia (Fig. 3). These
immunoreactive TRH fibers in the pars distalis
disappeared at the onset of metamorphic climax
(stage XX). In the larvae beyond stage XX,
TRH-immunoreactivity in the hypophysis was con-
fined to the pars nervosa and pars intermedia (Fig.
4).

DISCUSSION

The present study demonstrated the localization
of immunoreactive TRH in the brain and hypoph-
ysis of larval bullfrog during metamorphosis. Dur-
ing metamorphosis, immunoreactive TRH was
found in the pars nervosa and pars intermedia, and
in some brain areas including preoptic nucleus,
infundibular nucleus, septum, amygdala and di-
agonal band of Broca. This histological distribu-
tion of immunoreactive TRH was essentially simi-
lar to that reported previousely [4, 8, 9]. Moreov-
er, the present study demonstrated clearly that
immunoreactive TRH fibers penetrate into the
pars distalis at prometamorphic stage. To our
knowledge, the presence of TRH fibers in the pars
distalis has not been described in the bullfrog or
other vertebrates throughout larva and adult. The
discrepancy between the present study and others
may be ascribed to the technical differences. Pre-
vious investigators used the mixture of formal-
dehyde and glutaraldehyde as the fixative and
thick frozen sections, while we used acrolein fixa-
tive and thin-paraplast sections. Leechan and
Jackson [3] have reported that TRH in the tissue is
well preserved only with acrolein. Moreover, the

employment of thick frozen sections may get into
difficulty to reveal the precise distribution of im-
munoreactive TRH fibers in the hypophysis.

Aronsson [10] has demonstrated monoaminergic
nerve fibers in the pars distalis of Rana temporaria
by Falck-Hillarp method and stated that the pre-
sence of these fibers was confined to the stages
from prometamorphosis to just before the climax.
Kawakami-Kondo et al. [11] have reported the
tyrosine hydroxylase-positive nerve fibers in the
pars distalis which was confined to the Gosner’s
stage 31 to 40. The time of appearance and
disappearance of these monoaminergic or dopa-
minergic fibers in the pars distalis was nearly
identical to those of the immunoreactive TRH
fibers in the present study.

In the teleost hypophysis, some nerve endings
exist in the pars distalis [12]. In higher vertebrates,
indirect connection by way of the hypophysial
portal vessels was predominantly seen, although
nerve fibers were occasionally observed in the pars
distalis in rats [13]. The immunoreactive TRH-
and monoamine-containing fibers in the pars dis-
talis during metamorphosis might be considered as
an intermediate type between the teleost and the
higher vertebrate.

In amphibian larvae at premetamorphic and
early prometamorphic stages, the pars distalis is in
close contact with the median eminence where the
capillary loop is poorly developed. Penetration of
the capillary into the median eminence takes place
as metamorphosis progresses. At climax, develop-
ment of the median eminence is completed and the
pars distalis almost detaches from the median
eminence, being connected with it by the portal
vessel at the rostral region [14]. This morphologic-
al change in the hypothalamus-hypophysial com-
plex, which also observed in the present experi-
ment, may have something to do with the dis-
appearance of the immunoreactive TRH fibers
from the pars distalis. Considering that the
hypophysial portal vessels are still in the process of
formation during the period of the existence of the
immunoreactive TRH fibers in the pars distalis,
the immunoreactive TRH may temporally function
as a controling factor of secretion of a certain
hypophysial hormone(s) by “paracrine mechan-
ism” which is involved in metamorphosis [see, 15,

432 Y. TANIGUCHI, S. TANAKA AND K. KurosuMIi

16]. However, no definite conclusion can be drawn
since physiological role of TRH in controling
hypophysial function in amphibian larvae is not
clear.

The effect of mammalian TRH on hypophysis-
thyroid axis in the amphibians has been controver-
sial. It has been generally thought that TRH has
no thyrotropin-releasing effect both in vivo and in
vitro [17-21]. However, recent reports indicate
that TRH increases thyroxine levels in the frog [22,
23]. Therefore, a direct evidence that TRH in-
duces thyrotropin release is awaited. On the other
hand, it is evident that TRH stimulates the release
of prolactin from the pars distalis in vitro [24-26]
and in vivo [27]. Further studies on the identifica-
tion of cell type(s) of the pars distalis around which
the immunoreactive TRH fibers terminate may
give on insight into the physiological role of TRH
in the regulation of hypophysial hormone secretion
during metamorphosis.

ACKNOWLEDGMENTS

We are grateful to Professor $. Kikuyama, Depart-
ment of Biology, School of Education, Waseda Universi-
ty for stimulating and helpful discussion, and Dr. M.
Mori, School of Medicine, Gunma University for sup-
plying anti-TRH serum.

REFERENCES

1 Burgus, R., Dunn, T. F, Desiderio, D., Ward, D.
N., Vale, W. and Guillemin, R. (1970) Character-
ization of ovine hypothalamic hypophysiotropic
TSH-releasing factor. Nature, 226: 321-325.

2 Nair, R. M. G., Barrett, J. F., Bowers, C. Y. and
Schally, A. V. (1970) Structure of porcine thyrotro-
pin releasing hormone. Biochemistry, 9: 1103-1106.

3 Leechan, R. M. and Jackson, I. M. D. (1982)
Immunohistochemical localization of thyrotropin-
releasing hormone in the rat hypothalamus and
pituitary. Endocrinology, 111: 55-65.

4 Mimnagh, K. M., Bolaffi, J. L., Montogomery, N.
M. and Kaltenbach J. C. (1987) Thyrotropin-
releasing hormone (TRH): Immunohistochemical
distribution in tadpole and frog brain. Gen. Comp.
Endocrinol., 66: 394-404.

5 Shumway, W. (1942) Stages in the normal develop-
ment of Rana pipiens larvae. Anat. Rec., 78: 139-
147.

6 Taylor, A. C. and Kollros, J. J. (1946) Stages in the
normal development of Rana pipiens larvae. Anat.

10

11

12

13

14

15

16

Rec., 94: 7-23.

Mori, M., Kobayashi, I. and Wakabayashi, K.
(1978) Suppression of serum thyrotropin (TSH)
concentrations following thyroidectomy and cold
exposure by passive immunization with antiserum to
thyrotropin-releasing hormone (TRH) in rats. Meta-
bolism, 27: 1485-1490.

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.

Stoeckel, M. E., Hindelang, C., Lamacz, M.,
Tonon, M. C. and Vaudry, H. (1987) Co-existence
of TRH and mesotocin within nerve fibres in
neurointermediate lobe of the frog pituitary. Gen.
Comp. Endocrinol., 66: 16.

Aronsson, S. (1976) The ontogenesis of
monoaminergic fibers in the hypophysis of Rana
temporaria with special reference to the pars distalis.
Cell Tissue Res., 171: 437-448.

Kawakami-Kondo, Y., Yoshida, M., Karasawa, N.,
Yamada, K., Takagi, I., Kondo, T. and Nagatsu, I.
(1984) Ontogenetic study on the monoamine and
peptide containing cells in the pituitary and hypotha-
lamus of the bullfrog Rana catesbeiana by immuno-
histochemistry. Acta Histochem. Cytochem., 17:
387-397.

Vollrath, L. (1967) Uber die neurosekretorische
Innervation der Adenohypophyse von Teleostiern,
insbesondere von Hippocampus cuda und Tinca
tinca. Z. Zellforsch., 78: 234-260.

Kurosumi, K. and Kobayashi, Y. (1980) Nerve
fibers and terminals in the rat anterior pituitary
gland as revealed by electron microscopy. Arch.
histol. jap., 43: 141-155.

Etkin, W., Kikuyama, S. and Rosenbluth, J. (1965)
Thyroid feedback to the hypothalamic neurosecre-
tory system in frog larvae. Neuroendocrinology, 1:
45-64.

White, B. A. and Nicoll, C. S. (1986) Hormone
control of amphibian metamorphosis. In “Meta-
morphosis”. Ed. by L. I. Gilbert and E. Frieden,
Plenum Press, New York, pp. 363-396.

Kikuyama, S., Yamamoto, K. and Kawamura, K.
(1988) Hormonal regulation of amphibian meta-
morphosis. In “Regeneration and development”.
Ed. by S. Inoue et al., Proc. 6th M. Singer Sympo-
sium, pp. 161-171.

Etkin, W. and Gona, A. G. (1968) Failure of
mammalian thyrotropin-releasing factor preparation
to elicit metamorphic responses in tadpoles. Endoc-
rinology, 82: 1067-1068.

Gona, A. G. and Gona, O. (1974) Failure of
synthetic TRF to elicit metamorphosis in frog tad-
poles or red-spotted newts. Gen. Comp. Endocri-

19

20

21

22

23

TRH in Larval Bullfrogs

nol., 24: 223-225.

Taurog, A., Oliver, C., Eskay, R. L., Porter, J. C.
and McKenzie, J. M. (1974) The role of TRH in the
neoteny of the Mexican Axolotl (Ambystoma mex-
icanum). Gen. Comp. Endocrinol., 24: 267-279.
Vandesande, F. and Aspeslagh, M-R. (1974) Fai-
lure of thyrotropin releasing hormone to increase
125] uptake by the thyroid in Rana temporaria. Gen.
Comp. Endocrinol., 23: 355-356.

Millar, R. P., Nicolson, S., King, J. A. and Louw,
G. N. (1983) Functinal significant of TRH in meta-
morphosing and adult anurans. In “Thyrotropin-
releasing Hormones”. Ed. by E. C. Griffiths and G.
W. Bennett, Raven Press, New York, pp. 217-227.
Darras, V. M. and Kuhn, E. R. (1982) Increased
plasma levels of thyroid hormones in a frog Rana
ridibunda following intravenous adminstration of
TRH. Gen. Comp. Endocrinol., 48: 469-475.
Denver, R. J. (1988) Several hypothalamic peptides
stimulate in vitro thyrotropin secretion by pituitaries
of anuran amphibians. Gen. Comp. Endocrinol., 72:
383-393.

24

25

26

27

433

Clemons, G. K., Russell, S. M. and Nicoll, C. S.
(1979) Effect of mammalian thyrotropin releasing
hormone on prolactin secretion by bullfrog ade-
nophyses in vitro. Gen. Comp. Endocrinol., 38: 62-
67.

Hall, T. R. and Chadwick, A. (1984) Effects of
synthetic mammalian thyrotropin releasing hor-
mone, somatostatin and dopamine on the secretion
of prolactin and growth hormone from amphibian
and reptilian glands incubated in vitro. J. Endocri-
nol., 102: 175-180.

Seki, T. and Kikuyama, S. (1986) Effect of thyrot-
ropin-releasing hormone and dopamine on the in
vitro secretion of prolactin by the bullfrog pituitary
gland. Gen. Comp. Endocrinol., 61: 197-202.
Kuhn, E. R., Kikuyama, S., Yamamoto, K. and
Darras, V. M. (1985) In vivo release of prolactin in
Rana ribunda following an intravenous injection of
thyrotropin-releasing hormone. Gen. Comp. En-
docrinol., 60: 86-89.

ere RA

NO pane ORE ideale a
‘| teh, Weds iL hea bs lb nibh} TAM) yin ms
Py Trg wbsarth thats WE SO SOWMT [1

} ; (Th :Oe¥dinrn
4 tr Piet seit “4
bel ban Ht T sh rs
= ia Tie iw tis ecy
he ) Meta ee
Ni fy Meh Lie terete eth ulti

irenat Galery Gud aitonlara: tee) Te

Wha -dprwhade tar taneg aby rpg fmAlAES yt Mand) yi wool Tye “RMR
MheE™L SOT ce Sots apna kar wont
uid bee .T-uiee Ae ake ddl ‘ot iy

Ser shun oes D Decent Boyd he
. sik aaah thd lees ‘ TE TRE ged re MY eS
to) Kael iq qed ave ' ation! * (ity Oe ee ae

COP ayia OE ihe - aS ieee Gabe

ite Weta. * erway. to what binin tr Sheioe
saeditsh metueaiich cts x sae PA Be. inks ¢
pli i ) bea. 5 yd shusteny ria Rees Nidal 9
‘ 7 ' 4 reve hints ro eis radi a
ct Nae GBA fess? foi:

ih ty
. :
= eee
vt Ne
* freer}
]
A
Je
a
te
aD
-
oe
ioe
i
> ’ viene:
4 ~
i" we dl ty Tn | 7
‘
“ ’ ee Lames

ZOOLOGICAL SCIENCE 7: 435-441 (1990)

Vasodepressor Effect of Atrial Natriuretic Peptides
in the Quail, Coturnix coturnix japonica

YosuHio TAKEI and TAKUSHI X. WATANABE!

Department of Physiology, Kitasato University School of Medicine,
Sagamihara 228, and "Peptide Institute, Protein Research
Foundation, Minoh 562, Japan

ABSTRACT— An intravenous injection of a-human atrial natriuretic peptide-(1-28) (a-hANP) caused
dose-dependent decreases in both arterial pressure and heart rate in the urethane-anesthetized quail.
The hypotension always preceded the bradycardia. The EDso for the vasodepressor effect (58.2+4.6
pmol/100 g, n=28) was smaller than that for the bradycardial effect (107.2+8.9 pmol/100 g, n=23).
The hANP peptides with 3-6 amino acids removed from the NH,-terminus were as potent as a-hANP,
but the peptide with 4 amino acids removed from the COOH-terminus had only about half the
vasodepressor potency of the parent molecule. Among the peptides of hANP in which an amino acid
within the ring structure was modified, only des-Gly’ hANP, [(Met (O)!*] hANP, and [Asn!3] hANP
had decreased the potency. Formation of the ring structure between the 7th and the 23rd position with
an ethylene linkage instead of a disulfide bond did not change the potency. These results suggest that
the amino acid residues at the COOH-terminus and some amino acids within the ring structure of hANP

© 1990 Zoological Society of Japan

are important for the expression of its vasodepressor activity in the quail.

INTRODUCTION

Since the discovery of a potent diuretic and
natriuretic substance in the rat atrium by de Bold
et al. [1], a growing number of studies have been
performed in mammals to elucidate the structure
and biological actions of this substance. These
studies have demonstrated that the stored form of
the natriuretic factor has 126 amino acids in the rat
and human, and is named cardiodilatin-126, y-
atrial natriuretic peptide (ANP) or pro-atrial nat-
riuretic factor [see 2]. Later studies have shown
that the circulating form of ANP has 28 amino
acids that comprise the COOH-terminus of the
propeptide, and that this (1-28) molecule, termed
a-ANP by Kangawa and Matsuo [3], is most potent
among the endogenous peptides of ANP. In
addition to natriuretic and diuretic effects, ANP
has been shown to exhibit a potent vasodepressor
effect [1, 4, 5] and a cardiosuppressive effect [6-8]
in several species of mammals. The mechanism

Accepted September 25 , 1989
Received July 13, 1989

that causes hypotension could involve a decrease
in total peripheral resistance, a decrease in cardiac
output, and/or a decrease in blood volume via
diuresis, but the actual mechanism has not yet
been determined [2]. In nonmammalian verte-
brates, the effect of ANP on blood pressure is still
controversial. It has been shown that synthetic
ANP peptides decrease arterial pressure in the
chicken [9] and dogfish [10], while they have no
effect on arterial pressure in the toadfish [11], and
increase it in the rainbow trout [12].

In the present study, we attempted to examine
the effects of a-human ANP (a-hANP) on arterial
pressure and heart rate in the quail. We also
examined structure-activity relationships of hANP
peptides for the effect on blood pressure to ex-
amine the structural requirement of ANP mole-
cules for quail vascular receptors. The results were
compared with other in vivo studies on the nat-
riuretic activity in the rat {13, 14] and antidip-
sogenic activity in the rat [15], and in vitro studies
on the spasmolytic activity of the rat and rabbit
aortic strips [14, 16] and of the chick rectum [13,
14].

436 Y. TAKEI AND T. X. WATANABE

MATERIALS AND METHODS

Animals

Male Japanese quail, Coturnix coturnix jJaponi-
ca, were purchased from a local dealer at the age
of 4 weeks. They were kept individually in wire
cages (21 x 1217 cm®) under a short daily photo-
period (8L:16D) at 25+1°C for more than 2
weeks before use. Quail diet (Nippon Haigo
shiryo, Yokohama) containing 150 meq/kg of Na
and tap water were freely available until the day of
experiment. The birds weighed 106+2 g (n=28)
at the time of experiments.

Surgery

The birds were lightly anesthetized by an intra-
muscular injection of urethane (2 g/kg) in the
breast region and fixed on an operating board. It
has been reported that urethane anesthesia is
suitable for cardiovascular studies because of its
little effects on cardiovascular reflexes [17]. After
tracheotomy, a polyethylene tube (PE10, Clay
Adams) was inserted into the right atrium through
the right external jugular vein for injection of
ANP, and a cannula assembly was inserted into the
right common carotid artery for measurement of
arterial pressure (Fig. 1). Care was taken not to
damage the vagus nerve and other nerves that
innervate the carotid sinus. Almost no bleeding
was observed during surgery. The jugular cannula
was filled with isotonic saline, and the arterial
cannula was filled with isotonic saline which con-
tained 100 units/ml of heparin.

Measurement of arterial pressure and heart rate

The cannula in the carotid artery was connected
to a small semiconductor-type pressure transducer
(PML-500GC, Kyowa Electric Instruments Co.,
Ltd.,Tokyo) via a short silicone-rubber tube (Fig.
1). Since the transducer was connected to a carrier
amplifier (Type 3126, Yokogawa Electric Works
Ltd, Tokyo) by a long, flexible cord, the cannula
would not slip out of the artery even when the
fixed birds move slightly under anesthesia. The
original pressure waves and the integrated waves
(the mean pressure) were amplified and recorded
with a recorder (Rectigraph-8K, NEC San-ei,

flexible cord

pressure transducer

polyethylene tube

silicone-rubber tube
(o.d. : 4.0mm, I.d. : 2.6mm)

common carotid artery

Fic. 1. Cannula assembly for the measurement of arte-
rial pressure from the common carotid artery of the
quail. The polyethylene tube (0.d.: 2.7 mm, i.d.:
1.8 mm) is narrowed to 0.6-0.8 mm by pulling the
tube after heating, and then its tip is dilated by
heating. The polyethylene tube and silicone-rubber
tube (SH-3, Create Medic, Yokohama, Japan) are
filled with heparinized saline (100 units/ml). The
outer diameter of the transducer is 2.8mm. Since
the transducer is not fixed, the cannula does not
escape from the artery even though the bird moves
to some extent during the measurements. Another
advantage of this system is that, because the cannula
is rather thick and short, the pulse wave can be
recorded with a small distortion.

Tokyo). The original waves after amplification
were also introduced into a tachometer (#1322,
NEC San-ei, Tokyo) for measurement of heart
rate, and they were stored in a data recorder
(R-260, TEAC, Tokyo) for further analyses.

Experimental protocol

In the first experimental group (n=10), the
EDso of a-hANP for its effects on arterial pressure
and heart rate was determined. Intravenous injec-
tions of 0, 0.05, 0.1, 0.2, 0.5, 1 or 2 ~g/100 g body
weight of a-hANP in 0.05 ml of 0.9% NaCl were
given twice for each dose in random order. Each
injection was followed by flushing of the cannula
(dead space: 0.01 ml) with 0.03 ml of 0.9% NaCl.
The injection was repeated after the arterial press-
ure returned to the level before injection, since it
was shown in preliminary experiments that re-
peated injections of 0.5 «g of a-hANP given in this
way generated reproducible effects.

In the second experimental group (n=18), the
structure-activity relationship of the vasodepressor
effect was examined using various analogs of

Cardiovascular Effects of ANP in Quail 437

hANP. For this purpose, 0, 0.05, 0.1, 0.2, 0.5, 1
and 2 #g/100g of a-hANP (hANP-(1-28)) were
first injected in that order, then three or four of the
analogs were injected at the doses of 0.1, 0.3 and 1
yg/100 g in that order. The analogs of hANP used
were [Ile!*] hANP-(1-28) (a-rat ANP), [Met(O)'7]
hANP-(1-28), hANP-(4-28), hANP-(5-28), hANP-
(7-28), [Nle'*] hANP-(7-28), [D-Ala’] hANP-(7-
28), [Asn'3] hANP-(7-28), des-Gly’ hANP-(7-28),
hANP-(5-25), hANP-(7-23), and [Asu’?3] hANP-
(7-23). The order of injection of the analogs was
random. After the last injection of the analogs,
injections of 0.1, 0.3 and 1 4g/100 g of a-hANP
were repeated to confirm the reproducibility of
results. All hANP-related peptides were synthe-
sized by the liquid-phase method [18] at the Pep-
tide Institute Inc. (Osaka, Japan).

Statistical analyses

Since it was found that decreases in arterial
pressure and heart rate were greater when the
pre-injection levels were higher, the decreases
were expressed in terms of percentages from pre-
injection levels. The EDs of a-hANP was calcu-
lated from all data of 28 birds used in this study.
For calculation of the EDs , changes in arterial
pressure or heart rate at doses between 0.05 and 2
“ug were fitted to a logistic curve by each bird, and
the dose that produced a half-maximal respose was
obtained from the curve. The curve fitting was
executed by the Newton-Raphson algorithm based
on estimates of maximum likelihood [19]. The
changes in arterial pressure and heart rate at each
dose were compared with the changes observed
after injection of saline by the paired t-test. In the
experiment to examine the vasodepressor potency
of analogs of hANP relative to a-hANP, the poten-
cy ratio was calculated from the ratio of the EDs
of hANP to that of an analog. In this case, the
EDs9 of a-hANP was calculated with data obtained
at doses between 0.1 and 1 ug, as was done for
each analog. The response to 2 ug of a-hANP was
used only to assess the maximum response. All
results are expressed as means+SE of the mean.

RESULTS

Dose-response relationship for a-hANP

The mean, resting arterial pressure of the quail
before injection was 84.1+2.3mmHg (n=28).
The arterial pressure decreased immediately after
injection of a-hANP, and the decreases become
greater as the dose increased (Fig. 2). The largest
decrease at high doses was about 50% of the level
before injection. The EDs calculated by the
logistic-log transformation was 179+ 14 ng (58.2+
4.6 pmol)/100 g body weight (n=28) (Fig. 3). The
significant decrease in arterial pressure was
obtained at 100ng/100g, and 9 out of 28 birds
decreased their arterial pressure at 50 ng/100g
(Fig. 2). The decreased pressure continued for
longer as the dose increased (Fig. 2).

mmHg
PA 100
of eaeen ant’ at eat alll
40
mmHg
MA 100
al Say awe ee
40
b/min
FIR Peco;
so ! toy a t 4 '
300 MOS O.2 0.5 1 2
15 min
Dose (yg/100 g)
Fic. 2. Changes in arterial pressure (PA), mean PA

(MA), and heart rate (HR) after injection of 0-2
vg/100 g body weight of a-hANP into the jugular
vein of a quail.

The mean heart rate of the quail before injection
was 515+22 beats/min (n=28). The heart rate
decreased after injection of a-hANP, and the
decrease was greater as the dose increased (Fig.
2). The EDs was 330+27 ng (107.2+8.9 pmol)/
100 g (n=23) (Fig. 4). The significant decrease
was obtained at 100 ng/100g. Compared to the
vasodepressor effect, the bradycardia occurred a
little more slowly and recovered more quickly, and
the extent of the decrease was smaller. In 5 of 28
birds, injection of a-hANP caused tachycardia

438 Y. TAKEI AND T. X. WATANABE

40
5 ¢
cab)

= 20

re)

(ab)

4 ¢

ia 20

i

Oo

—

So

o

is 10

o

©

® 0 9.05 ¢

Oo (0) ay fg
O e Ong on Moe olseeheme
O +5

Dose (ug/100 g body weight)

Fic. 3. Dose-response relationship for the vasodepres-
sor effect of e-hANP in 28 quail. Doses are express-
ed ina log scale. The logistic curve that fitted best is
shown in the figure. The calculated ED<9 in 179+ 14
ng (58.2+4.6 pmol)/100 g body weight. The vertical
bars represent standard errors of the mean.

30
oS
x
wm 20
eae
~
fe
oO
®
€ t
= ral
®
ep)
(40)
®
—
o

0 O05 KON 0:2 Clie 2

Dose (ug/100 g body weight)

Fic. 4. Dose-response relationship for the bradycardial
effect of a-hANP in 23 quail. Doses are expressed in
a log scale. Among 28 quail examined, 5 quail
showed tachycardia after injection of hANP. The
logistic curve that fitted best is shown in the figure.
The calculated EDso is 330+27ng (107.2+8.9
pmol)/100 g body weight. The vertical bars repre-
sent standard errors of the mean.

although hypotension was induced in the same
birds. The basal heart rate of these birds (370+45
beats/min, n=5) was smaller than that of the other
birds in which bradycardia was induced (561+19
beats/min, n=23).

Structure-activity relationships for analogs of
hANP

As shown in Figure 5, sensitivity to hANP was
well conserved and little tachyphylaxis was
observed even after repetitive injections of a-
hANP and its analogs. The hANP analogs with
amino acids removed from the NHb>-terminus,
hANP-(4-28), HNANP-(5-28) and hANP-(7-28)
were almost as potent as a-hANP in terms of their
vasodepressor activity (Table 1). However, the
peptides with amino acids removed from the
COOH-terminus, hANP-(5-25) and hANP-(7-23),
were less potent. No significant change in the
vasodepressor potency was observed after the dis-
ulfide bond of hANP-(7-23)([Cys”7*] hANP-(7-
23)) was replaced by the ethylene linkage
([Asu’3] hANP-(7-23)). The modification of
amino acids within the ring structure had variable
effects (Table 1). When Met at the 12th position
was replaced by Ile (rat ANP) or Nle, or when Gly
at the 9th position was replaced by D-Ala, the
potency did not change from that of the parent
molecule. However, replacement of Asp at the
13th position with Asn, oxidation of Met at the
12th position, or removal of Gly at the 9th posi-
tion, greatly decreased the potency of the peptide.

Lael oras aa entaaveadaaee di
Ope 4 ‘agen ‘
er] Uae age aN vine NT 4 4 4 t
(1-28) hANP (7-23) [Met(O}] (7-28) (1-28)
mmHg hANP hANP hANP hANP
105
Fj | ‘a
MAS ilk an ae Va
ast “cas ont , aga “0.3, Ogg O10.3
0.5
Dose (yg/100 g)
Fic. 5. Changes in arterial pressure (PA) and its mean

(MA) after injection of hANP-(1-28) (@-hANP),
hANP-(7-23), [Met(O)!"] hANP-(1-28), and hANP-
(7-28) into a quail. The injection of e-hANP was
repeated at the end of the experiment to confirm the
reproducibility of data.

Cardiovascular Effects of ANP in Quail 439

TABLE 1.

Relative vasodepressor potencies of hANP analogs in the quail (n=6).

Relative potencies by

the spasomlytic activity in the chick rectum and rat aorta in vitro and by natriuretic activity in the rat in

vivo are given for comparison.
hANP-(1—28) is:
1 5 10

Values are means+SE of the mean.

Amino acid sequence of

15 20 25

S-L-R-R-S-S-C-F-G-G-R-M-D-R-I-G-A-Q-S-G-L-G-C_-N-S-F-R-Y

Potencies (%)

Analogs
of hANP quail chick* rat* rat*
depressor rectum aorta natriuresis

(1-28)

hANP 100 100 100 100

[Ile’?] 83.2+ 9.7 122 226 236

[Met(O)?7] 15.8+ 4.0 2 4 13
(4-28)

hANP 86.3+ 10.7 64 79 97
(5-28)

hANP 77.9+15.1 67 96 73
(7-28)

hANP 123.5+16.2 364 154 148

[Nle’?] 82.5+15.3 169 61 141

[D-Ala?] 119.0+15.6 356 120 107

[Asn] 18.8+ 5.9 5 1 0

des-Gly” 26.5+ 6.9 1 0 0
(5-25)

hANP 63.7+14.7 31 5 82
(7-23)

hANP 50.7+12.7 139 3 Dil

[Asu’?3] 60.3+ 7.7 Ti 1 6
*[14]

substance extracted had a potent vasodepressor
DISCUSSION

A moderate dose of a-hANP caused a profound
hypotension in the quail, as has been reported to
occur in the chicken after injection of a-rat ANP
[9]. In the chicken, a crude heart extract also
decreases arterial pressure. Thus, ANP-like mate-
rial appears to be present in the chicken heart, as
suggested by the results of immunohistochemical
analysis [20]. The presence of ANP-like material
was also reported in the quail by immunohistoche-
mistry using antibodies to pig ANP [21]. We also
observed in preliminary experiments that the quail
heart contained substances that cross-react with
antibodies raised against a-hANP, and that the

effect in the quail (Takei and Ando, unpubl. data).
However, the amino acid sequence of this im-
munoreactive ANP may differ from that of a
hANP, since the concentration in the heart extract
determined by the quail vasodepressor bioassay
was more than 10,000 times greater than the value
measured by radioimmunoassay for hANP. Thus,
hANP has a potent vasodepressor effect in the
quail, although its molecule may be structurally
different from quail ANP. Very recently, amino
acid sequence of chicken ANP has been deter-
mined, whose sequence has only 52% homology to
hANP [22].

In mammals, a bolus injection of a-hANP has

440 Y. TAKEI AND T. X. WATANABE

been shown to decrease arterial pressure in a linear
fashion at doses between 33 and 333 pmol (0.1 and
1 ug)/kg in anesthetized dogs [23], and at doses
between 0.18 and 9 nmol/kg in conscious, spon-
taneously hypertensive rats [24]. Values of EDs
were not calculated but dogs appear to be a little
more sensitive to the vasodepressor effect of hu-
man (also native to dogs) ANP than are the rat and
the quail. However, the maximal hypotension in
birds (50%) was greater than that of mammals
(10-20%).

It is known that, in mammals, an acute
hypotension is useally followed by reflex tachycar-
dia, which is mediated via sino-aortic barorecep-
tors [25]. Since profound hypotension was induced
in all quail in the present study but bradycardia
was induced in most birds after injection of hANP,
hANP may have some inhibitory effect on heart
rate. In guinea pigs. ANP also induces hy-
potension and bradycardia, but the effect of ANP
on heart rate does not appear to be a direct action,
since rat ANP-(5-28) did not change the frequency
of spontaneously beating atrium in vitro [26].

For the expression of vasodepressor activity in
the quail, amino acids at the COOH-terminus
appear to be more important than those at the
NH,-terminus, and the modification of some ami-
no acids within the ring structure, which may
change the tertiary structure of the molecule,
affects the potency. These results are similar to
those obtained in the spasmolytic activity of the
chick rectum and rat aorta, and in the natriuretic
activity of rats (Table 1). However, some differ-
ences are observed in different preparations; in the
chick rectum, removal of 6 amino acids from the
NH>-terminus increases the activity more than
three folds, and in the rat preparations, the homo-
logous rat peptide has more than two-fold activi-
ties than does the human peptide. Other major
differences of the quail preparation from other
preparations are that modifications of some amino
acids within the ring structure cause smaller loss of
activity, and that substitution of a disulfide bond
with an ethylene linkage does not decrease the
activity. Garcia et al. [16] showed that amino acids
at the NHb>-terminus as well as those at the
COOH-terminus are important for the spasmolytic
activity in the rabbit aorta. For the antidipsogenic

activity in the rat, the removal of amino acids from
the NH)-terminus does not change the activity, but
the (7-23) peptide has no activity [15]. Collective-
ly, it appears that quail vascular receptors can
respond to a greater variety of ANP analogs than
do other ANP receptors. Supporting this idea are
our preliminary data which show that ANP-like
substance extracted from carp hearts causes
hypotension in the quail, but has no relaxant effect
on the chick rectum (Uemura and Takei, unpubl.
data). It appears that the quail vasodepressor
effect is a usefil assay system for nonmammalian
ANPs whose molecular structures may be different
from mammalian ANPs.

In summary, the quail exhibited high sensitivity
to the vasodepressor effect of hANP, and because
of their small size (100 g), they can respond to as
little as 50 ng (16 pmol)/bird. The vasodepressor
effect was profound (50%), and reproducible even
after repeated injections (see Fig.5). Thus, it
seems that the quail vasodepressor bioassay is
considered to be a useful in vivo assay system for
the comparative study of ANP in nonmammalian
vertebrates.

ACKNOWLEDGMENTS

This investigation was supported on part by a grant
from the Ministry of Education, Science and Culture of
Japan (62740449), and by the Kitasato Research Founda-
tion under Grant No. 10.

REFERENCES

1 De Bold, A. J., Borenstein H. J., Veress, A. T. and
Sonnenberg, H. (1981) A rapid and potent natriure-
tic response to intravenous injection of atrial
myocardial extract in rats. Life Sci., 28: 89-94.

2 Genest, J. and Cantin, M. (1988) The atrial nat-
riuretic factor: its physiology and biochemistry. Rev.
Physiol. Biochem. Pharmacol., 110: 1-145.

3 Kangawa, K. and Matsuo, H. (1984) Purification
and complete amino acid sequence of a-human atrial
natriuretic polypeptide (e-hANP). Biochem. Bio-
phys. Res. Commun., 118: 131-139.

4 Richards, A. M., Nicholls, M. G., Ikram, H.,
Webster, M. W., Yandle, T. G. and Espiner, E. A.
(1985) Renal, hemodynamic, and hormonal effects
of human alpha atrial natriuretic peptide in healthy
volunteers. Lancet, 1: 545-548.

5 Hirata, Y., Ishii, M., Sugimoto, T., Matsuoka, H.,

10

11

12

13

14

15

Cardiovascular Effects of ANP in Quail

Sugimoto, T., Kangawa, K. and Matsuo, H. (1985)
The effects of atrial 28-amino acid peptide on syste-
mic and renal hemodynamics in anesthetized rats.
Cire. Res., 57: 634-639.

Ackermann, U., Irizawa, T. G., Milojevic S. and
Sonnenberg, H. (1984) Cardiovascular effects of
atrial extracts in anesthetized rats. Can. J. Physiol.
Pharmacol., 62: 819-826.

Breuhaus, B. A., Saneii, H. H., Brandt, M. A. and
Chimoskey, J. E. (1985) Atriopeptin II lowers
cardiac output in conscious sheep. Am. J. Physiol.,
249: R776-R780.

Cody, R. J., Atlas, S. A., Laragh, J. H., Kubo, S.
H., Covit, A. B., Ryman, K. S., Shaknovich, A.,
Pondolfino, K., Clark, M., Camargo, M. J. F.,
Scarborough, R. M. and Lewicki, J. A. (1986)
Atrial natriuretic factor in normal subjects and heart
failure patients: plasma levels and renal, hormonal
and hemodynamic responses to peptide infusion. J.
Clin. Invest., 78: 1362-1372.

Gregg, C. M. and Wideman, Jr. R. F. (1986)
Effects of atriopeptin and chicken heart extract in
Gallus domesticus. Am. J. Physiol., 251: R543-
RSS1.

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.

Lee, J. and Malvin, R. L. (1987) Natriuretic re-
sponse to homologous heart extract in aglomerular
toadfish. Am. J. Physiol., 252: R1055—R1058.
Duff, D. W. and Olson, K. R. (1986) Trout vascular
and renal responses to atrial natriuretic factor and
heart extract. Am. J. Physiol., 251: R639-R642.
Thibault, G., Garcia, R., Carrier, F., Seidah, N. G.,
Lazure, C., Chétién, M., Cantin, M. and Genest, J.
(1984) Structure-activity relationships of atrial nat-
riuretic factor (ANF). I. Natriuretic activity and
relaxation of intestinal smooth muscle. Biochem.
Biophys. Res. Commun., 125: 938-946.
Watanabe, T. X., Noda, Y., Chino, N., Nishiuchi,
Y., Kimura, T., Sakakibara, S. and Imai, M. (1988)
Structure-activity relationships of a-human atrial
natriuretic peptide. Europ. J. Pharmacol., 147: 49-
57.

Ito, H., Nakao, K., Katsuura, G., Morii, N.,
Shiono, S., Yamada, T., Sugawara, A., Saito, Y.,
Watanabe, K., Igano, K., Inouye, K. and Imura, H.
(1987) Atrial natriuretic polypeptide: structure-
activity relationships in the central action—a com-
parison of their antidipsogenic actions. Neurosci.

16

19

20

21

22

23

24

25

26

441

Lett., 74: 102-106.

Garcia, R., Thibault, G., Seidah, N. G., Lazure, C.,
Cantin, M., Genest, J., and Chétién M. (1985)
Stfucture-activity relationships of atrial natriuretic
factor (ANF). II. Effect of chain-length modifica-
tions on vascular reactivity. Biochem. Biophys. Res.
Commun., 126: 178-184.

Maggi, C. A. and Meli, A. (1986) Suitability of
urethane anesthesia for physiopharmacological in-
vestigations in various systems. Part 2: Cardiovascu-
lar system. Experientia, 42: 292-297.

Chino, N., Nishiuchi, Y., Masui, Y., Noda, Y.,
Watanabe, T. X., Kimura, T. and Sakakibara, S.
(1985) Synthesis of a-human atrial natriuretic
polypeptide (e-hANP) and its related peptides. In
“Peptide Chemistry 1984”. Ed. by N. Izumiya,
Protein Research Foundation, Osaka, pp. 241-246.
Cox, D. R. (1970) The Analysis of Binary Data.
Chapman and Hall, London.

Chapeau, C., Gutkawska, J., Schiller, P. W., Milne,
R. W., Thibault, G., Garcia, R., Genest, J. and
Cantin, M. (1985) Localization of immunoreactive
synthetic atrial natriuretic factor (ANF) in the heart
of various animal species. J. Histochem. Cytochem. ,
33: 541-550.

Reinecke, M., Nehls, M. and Forssmann, W. G.
(1985) Phylogenetic aspects of cardiac hormones as
revealed by immunohistochemistry, electronmicros-
copy, and bioassay. Peptides, 6: 321-331.

Miyaka, A., Minamino, A., Kangawa, K. and Mat-
suo, H. (1988) Identification of a 29-amino acid
natriuretic peptide in chicken heart. Biochem. Bio-
phys. Res. Commun., 155: 1330-1337.

Ishihara, T., Aisaka, K., Hattori, K., Hamasaki, S.,
Morita, M., Noguchi, T., Kangawa, K. and Matsuo,
H. (1988) Vasodilatory and diuretic actions of
a-human natriuretic polypeptide (e-hANP). Life
Sci., 36: 1205-1215.

Lappe, R. W., Todt, J. A. and Wedt, O. L. (1986)
Hemodynamic effects of infusion versus bolus admi-
nistration of atrial natriuretic factor. Hypertension.
8: 866-873.

Heymans, C. and Neil, E. (1958) Reflexogenic
Areas of the Cardiovascular System. Little, Brown
and Company, Boston, pp. 45-55.

Bergey, J. L. and Kotler, D. (1985) Effects of
atriopeptins I, II and III on atrial contractility, sinus
nodal rate (guinea pig) and agonist-induced tension
in rabbit aortic strips. Europ. J. Pharmacol., 110:
277-281.

ZOOLOGICAL SCIENCE 7: 443-447 (1990) © 1990 Zoological Society of Japan
A New Milliped of the Genus Riukiaria from Is. Yaku-shima,
Japan (Diplopoda: Polydesmida: Xystodesmidae)

TsuTOMU TANABE!

Zoological Institute, Faculty of Science, Hokkaido University,
Sapporo 060, Japan

ABSTRACT—A new species of Xystodesmidae, Riukiaria jamila, is described and figured based on
male and female adults from Is. Yaku-shima, off the southern coast of Kyishi, Japan. This species is
characterized by the gray tergites, the yellowish white paranota, and the acropodite of the male gonopod
which is very wide from the base to the twisted portion.

INTRODUCTION

The genus Riukiaria of the family Xystodesmi-
dae is currently represented by approximately 20
species from Japan [1-10], Korea [8, 11] and
One species, R. puella

Taiwan [3, 8, 12, 13].

ae AD TN

Fic. 1.

Accepted July 11, 1989
Received April 12, 1989
' Present address:

Tanabe, is known from Is. Yaku-shima, off the
southern coast of Kyisht, Japan.
In the present paper, I describe a new species of
this genus as the second species from that island.
Terminology follows that of Shelley [14] except
for parts of the acropodite of the male gonopod.

R. jamila sp. nov., $ paratype, dorsal view.

Tokushima Prefectural Museum, Tokushima 770, Japan.

444 T. TANABE

Riukiaria jamila sp. nov.
(Figs. 1-16)

Types: 15% (holotype and _ paratypes),
12 ? ¢ (paratypes) from the litter of Ardisia quin-
quegona, along Oko-rind6 mountain path, Kurio,
Yaku-ch6, Is. Yaku-shima, Kumage-gun, Kago-
shima-ken, 12-V-1987, T. Tanabe leg. 1% (para-
type), locality as above, 26-IV-1986. A. Moroto
leg. The @ holotype and 1? paratype are de-
posited in the collection of the National Science
Museum, Tokyo. 1% and 1 paratypes are to be
deposited in the collection of the North Carolina

Fics. 2-8.

State Museum of Natural Science, Raleigh, NC,
USA. 12% $ and9? ? paratypes deposited in my
private collection.

Diagnosis: Different from all other species of
Riukiaria in having gray tergites and yellowish
white paranota. The male with following diagnos-
tic characters: Gonopod very wide from the base
to the twisted portion, with a small coxal apophysis
on anterior face of coxa and an acute medial
process at curved portion of acropodite. Fifth

sternite with high processes between 4th legs;
prefemoral spines curved anterodorsally.
Description. ¢ holotype.

R. jamila sp. nov. 2-7, $ holotype: 2, head, anterior view, setae omitted. 3, right paranotum of 8th

segment, dorsal view. 4-6, 4th-6th sternites, setae omitted: 4, 4th, posterior view; 5, 5th, posterior view; 6, 6th,
posterior view. 7, coxa and prefemur of 10th left leg, ventral view, setae omitted. 8, Gonopods in situ of
paratype, ventral view, setae omitted. c, coxa. g, gena. pf, prefemur. s, sternite.

New Riukiaria from Japan 445

1mm tp
eA

Fics. 9-16. R. jamila sp. nov. 9-12, left gonopod of holotype: 9, medial view, setae omitted except coxa; 10, lateral
view, setae omitted; 11, dorsal view, setae omitted; 12, dorsomedial view, setae omitted. 13-15, % paratypes:
13, Sth sternite, anterior view; 14, 4th segment, ventral view; 15, cyphopods, posterior view; 16, left cyphopod,
anterior view. ap, acropodite. c, coxa. ca, coxal apophysis. cm, coxal macroseta. cn, cannula. cu, curve. f,
flange. 0, operculum. pf, prefemur. pfp, prefemoral process. pg, prostatic groove. r, receptacle. sa, sternal
apodeme. tp, twisted portion. v, valve.

446 T. TANABE

Head: Capsule smooth, polished. First
antennomere subglobose, 2nd-6th clavate, 7th
short and truncate; relative length of antenno-
meres 2=3=4=5=6>1>7. Genae (Fig. 2) with
distinct central impressions. Facial setae as fol-
lows: Epicranial 2 (left), 2 (right); inter antennal 1
(left), 1 (right); frontal+genal about 40; clypeal
about 30; labral about 25.

Tergites without setae and tubercles, polished.
Collum finely coriaceous, somewhat narrower than
2nd tergite. Protergites smooth. Metatergites
finely coriaceous, without transverse medial de-
pressions. Paranota (Fig. 3) finely serrate on both
anterior and posterior margins; posterolateral cor-
ners rounded on segments 1-4, becoming progres-
sively more acute posteriorly. Peritremata dis-
tinct. Ozopores located at about posterior 1/3,
opening laterally.

Sternites smooth, without setae, polished: 4th
sternite (Fig.4) with a pair of small acuminate
processes, about as long as width of adjacent
coxae; 5th sternite (Fig. 5) with a pair of high
coalesced projections between 4th legs, and a pair
of small, flat, widely segregated elevations be-
tween Sth legs; 6th sternite (Fig. 6) with a pair of
short, flat, widely segregated elevations between
6th legs, and convexly recessed between 7th legs.
Postgonopodal sternites without distinct tubercle
between any leg pair.

Pregonopodal legs densely hirsute; postgono-
podal legs becoming progressively less hirsute
posteriorly. Coxae with blunt, indistinct, distome-
dial projections, but those of 3rd and 4th legs with
slightly rounded projection proximomedially. Pre-
femoral spine beginning on 4th segment, becoming
progressively longer posteriorly, pointed bending
anterodorsally on midbody (Fig. 7).

Gonopodal aperture elliptical, 2.1 mm wide and
1.1 mm long at midpoint; sides raised above meta-
zonal surface. Gonopods in situ as in Fig. 8.
Gonopod structure as follows (Figs. 9-12): Coxa
with a small coxal apophysis near dorsal margin on
anterior face and a macroseta between coxal
apophysis and cannula. Sternal apodeme long,
straight. Prefemur flat proximolaterally (Fig. 11).
Prefemoral process flat, arcuate, tapering into
acuminate tip, strongly reflexed in apical half; tip
directed laterally. Acropodite thin, leaning antero-

medially, extending beyond level of prefemoral
process, twisted at about 3/4 of length, curved
dorsally near tip; very wide from base to twisted
portion and tapering into acuminate tip, with inner
surface broadly excavated from base to twisted
portion and medial flange from 1/3 of length to
twisted portion; flat from twisted portion to tip,
with an acute medial process at curved portion; tip
simple. Prostatic groove originating in pit at base
of prefemur, running along lateral side of inner
surface of acropodite to apical opening.

Color in life: Paranota yellowish white.
Metatergites gray. Protergites yellowish white,
with transverse gray stripes at midlength. Collum
gray, with yellowish white stripes along both ante-
rior and posterior margins. Epicranium pale gray.
Face gray. Genae, clypeus, labrum and antennae
all yellowish white. Venter and legs yellowish
white. Paraprocts yellowish white; each with a
central gray spot.

6 6 paratypes: The ¢ ¢ paratypes agree with
the ¢ holotype in most structural details, except
numbers of the following facial setae (n=5): Fron-
tal+genal about 25-50; clypeal about 25-30; la-
bral 30-40.

9 9 paratypes: Somatic features as in male,
except numbers of the following features: Facial
setae (n=5): Frontal+genal about 25-40; clypeal
25-40; labral 25-30. Body more arched, and
generally larger. Paranota shorter. Sth sternite
(Fig. 13) with a pair of short, flat separated eleva-
tions between 4th legs, without processes between
Sth legs. 6th sternite without processes. Legs
more slender. Coxae of 3rd and 4th legs with no
projections proximomedially. Cyphopodal aper-
ture (Fig. 14) with a slight anterior projection of
caudal margin. Cyphopods as in Figs. 15 and 16.
Valves subequal in size and shape; receptacle
projecting at midwidth of dorsal margin and
strongly depressed centrally on anterior side, and
much smaller in size on posterior side. Coloration
somewhat paler than in male.

Measurements: ¢@ holotype. Body length 37
mm. Head width (across genal apices) 4.1 mm.
Collum: Length/width ratio 47.3%. 10th segment:
Protergal width/metatergal width ratio 66.9%.
Metatergite of 10th segment: Length/width ratio
27.7%; depth/width ratio 68.3%. Segmental

New Riukiaria from Japan

widths as follows:

Collum 5.5 mm 13-4th 6.4

2nd 5.9 15th 6.2

3rd 6.1 16th 5.9

4th 6.2 17th 5.2

Sth 6.3 18th 4.1
6th-12th 6.5

Paratypes (8 ¢ $, 7? $ in parentheses). Body
length 33-36mm (34-40mm). Head width
(across genal apices) 3.8-4.1mm (3.9-4.5 mm).
Collum: Width 4.9-5.5 mm (5.0-5.8 mm); length/
width ratio 37.5-50.0% (44.0-47.4%). 10th seg-
ment: Protergal width/metatergal width ratio
64.5-70.4% (70.5-73.5%). Metatergite of 10th
segment: Width 5.9-6.5mm _ (6.0-7.1 mm);
length/width ratio 27.0-32.7% (26.5-28.6%);
depth/width ratio 65.6-68.9% (70.5-74.6%).

Distribution: Known only from the type locality.

Remarks: This species is similar to R. holstil
(Pocock) from Is. Okinawa-jima, the Ryukyu
Islands, but can be distinguished in having gray
colored tergites, a pair of high sternal projections
between 4th legs of the male, and the acropodite of
the male gonopod which ts very wide from the base
to the twisted portion.

ACKNOWLEDGMENTS

I wish to express my gratitude to Dr. Rowland M.
Shelley (North Carolina State Museum of Natural Sci-
ence) and Dr. H. Katakura (Hokkaido Univ.) for their
critical reading of earlier drafts of the manuscript and
valuable comments. I am also grateful to Dr. R. L.
Hoffman (Virginia Museum of Natural History), Mr. Y.
Murakami (Niihama) and Mr. K. Shinohara (Koiwa
High School, Tokyo) for their valuable advice and en-
couragement. Mr. A. Moroto (Mie) kindly offered me a
valuable material used in this paper.

10

11

12

13

14

447

REFERENCES

Pocock, R. I. (1895) Report upon Chilopoda and
Diplopoda in the Chinese Seas. Ann. Mag. Nat.
Hist. Ser. 6, 15: 346-372.

Verhoeff, K. W. (1936) Zur Kenntniss ostasiatis-
cher Strongylosomiden und Fontariiden. Zool.
Anz., 115: 297-311.

Takakuwa, Y. (1942) Uber weitere japanische Rhy-
sodesmus Arten. Trans. Nat. Hist. Soc. Formosa,
32: 197-203.

Miyosi, Y. (1952) Beitrage zur Kenntniss japanis-
cher Myriopoden 5. Aufsatz: Uber zwei neue Arten
von Diplopoda. Zool. Mag. Tokyo, 61: 314-316. (In
Japanese, with German résumé.)

Miyosi, Y. (1957) Beitrage zur Kenntniss japanis-
cher Myriopoden 22. Aufsatz: Uber zwei neue
Arten von Diplopoda. Zool. Mag. Tokyo, 66: 403-
406. (In Japanese, with German résumé.)

Jeekel, C. A. W. (1952) Milliped Miscellany. Ent.
Berichten, 14: 71-77.

Haga, Y. (1968) [Millipeds of Japan 1], 1-11., pls.
1-6. The author. (In Japanese)

Shinohara, K. (1977) Reevaluation on Riukiaria
(Diplopoda). Acta Arachnol., 27: 115-119. (In
Japanese)

Tanabe, T. (1988) Two new species of the genus
Riukiaria from Kydsha and Is. Yaku-shima, Japan
(Diplopoda: Polydesmida: Xystodesmidae). Acta
Arachnol., 37: 37-45.

Golovatch, S. I. (1978) Some east Asian millipedes
(Diplopoda) in the collection of the Zoological
Institute, USSR Academy of Science. Entomolo-
gicheskoe Obozr., 57: 677-681. (In Russian)
Takakuwa, Y. (1941) Rhysodesmus Arten aus
Japan. Trans. Nat. Hist. Soc. Formosa, 31: 413-415.
Wang, Y. M. (1956) Serica Ie: Records of
myriapods on Formosa with description of new
species (2). Quart. Jour. Taiwan Mus., 9: 157-158.
Wang, Y. M. (1957) Serica Ig: Records of
myriapods on ‘Taiwan Island (4) six new
Polydesmids. Quart. Jour. Taiwan Mus., 10: 103-
111.

Shelley, R. M. (1981) Revision of the milliped
genus Sigmoria (Polydesmida: Xystodesmidae).
Mem. Amer. Entomol. Soc., No. 33: 140 pp.

Te
' Li . ors ,
rhasbd Leiber hey G8 aati ey
~ Leben att vel stvete (tebe) be wes
‘ : . » 025s be: 0 tt
a4 iq 2. +-(e.e4 titel eae
, miu iY) 04), Page te Oa
te ht (oh? he Vt ee
Tolar clad) Ls ayo deri ae
Warr ct
’ ye oo . Ooum
| | eT AbSDIRS alk
i ; Ls any v Arend fad
rileois @yehorivho idloe wil
~ msi Fi? d aS a Le
| oe oe - brent pushia
; i ; Aj pena Bert Ms BE any oire BS
nycet oy we arc Sed tinist Ty xt} whe
wed CNG: Teh noladeaiy rh
3 “ a
im now toe
: s? ’
ie
e
teva
LN Oa
rt) |
“0 tl) OC

ZOOLOGICAL SCIENCE 7: 449-458 (1990)

© 1990 Zoological Society of Japan

Ciliated Protozoa in the Rumen of Holstein-Friesian Cattle
(Bos taurus taurus) in Hokkaido, Japan, with the
Description of Two New Species

Akira Ito and Sorcut Imai

Nishimon Agricultural and Mutual Aid Association, Okoppe Branch, Monbetsu-gun,
Hokkaido 098-15, and ‘Department of Parasitology, Nippon Veterinary
and Zootechnical College, Musashino, Tokyo 180, Japan

ABSTRACT—The composition of rumen ciliates in the Holstein-Friesian cattle bred in Hokkaido,
Japan was surveyed. Of 50 species with 19 formae under 15 genera identified, two new species were

recognized, then described as Entodinium okoppensis and Ostracodinium munham.

Entodinium

okoppensis may be classified further into such four morphotypes as, okoppensis, bispinosum, bifidum
and monospinosum on the basis of caudal processes. Thirteen species were the first record in Japanese
cattle. The average number of individuals per 1 ml of rumen fluid was 5.4 x 10°, and that of species per

head of host was 17.2.

INTRODUCTION

Rumen ciliate faunae would be different among
the species of their hosts and/or among the hosts
inhabiting separated areas [1, 2]. Surveys and
comparisons of rumen ciliate faunae of various
ruminants in different regions should provide in-
formations about phylogenetic relationships
among rumen ciliates, because it is suggested that
the composition of rumen ciliates has peculiarly
differentiated in relatively limited habitates since
transfaunation has been assumed to occur only by
direct contact between the hosts [3].

Various races of cattle including Japanese Black
and Japanese Brown for beef, and Holstein-
Friesian for milk have been kept in various locali-
ties in Japan [4, 5]. Holstein-Friesian which origin-
ated in the Netherlands and Germany were im-
ported to Japan mainly through United States of
America in the last part of the 19th century [5].
Since then, they have been kept as the most
popular dairy cattle in Japan, especially in Hok-
kaido.

The rumen ciliate faunae of Japanese cattle in
Honshu and Kyushu have been surveyed by Imai et

Accepted June 29, 1989
Received March 22, 1989

al. [6, 7], but not of those in Hokkaido. The
present paper deals with the species composition
of ciliates obtained from the Holstein-Friesian
cattle in Hokkaido and includes descriptions of
two new species with four new formae.

MATERIALS AND METHODS

Samples were collected from 71 Holstein-
Friesian cattle bred in Okoppe, Hokkaido, Japan
by means of rumen puncture from 1985 to 1988.
They were immediately fixed and stained in
methylgreen-formalin-saline (MFS) solution [3].
For close examination of nuclei and for type
specimens, a part of the samples in which new
species were recognized was stained with Mayer’s
hematoxylin and prepared as permanent slides.
Identification was made followed by Ogimoto and
Imai [3], Dogiel [8], Kofoid and MacLennan [9-
11], and Imai [12]. Terminology of the morpholo-
gy and orientation of ciliates for description of the
new species conformed to our previous papers [1,
2, 3, 13]. The total ciliate number was calculated
by means of Fuchs-Rosenthal haemocyte counter
chamber. To obtain the average value for the
ciliate density under normal distribution, it was
computed from each value of ciliate number con-
verted into logarithms. The generic composition is

450 A. Ito AND S. IMar

shown as the percentage of each genus in about Description: Body rectangular to nearly
300 individuals. square. Ectoplasm forming one to three caudal
spines or lobes with various size at the posterior
end of body. Anterior end of body flattened or
concave. Anterior lip hardly visible when adoral
cilia retracted. Vestibulum fairly large and funnel-
shaped extending vertically but slightly bending
leftward. Rectum short and extending vertically to

RESULTS

Entodinium okoppensis n. sp.
(Figs. 1 and 2)

Fic. 1. Entodinium okoppensis n. sp. with its formae. a-d: forma okoppensis. e and f: forma bispinosum. g: forma
bifidum. h and i: forma monospinosum.

Ciliates from Cattle in Hokkaido, Japan 451

: =
af aes
i Gn oe

Fic. 2. Photomicrographs o

es

= sit ‘f a ce b we ae

f Entodinium okoppensis n. sp. a and b: forma okoppensis. c and d: forma bispinosum.

e: forma bifidum. f{: forma monospinosum. All the specimens are fixed and stained with MFS solution. Bar in

each figure indicates 30 “zm.

cytoproct in the left side of median line. Macro-
nucleus straight and slender rod shaped, four-fifths
of body length, situated in the right periphery of
body. Anterior end of macronucleus flattened, but
posterior end rounded. An ovoidal micronucleus
near left margin of middle of macronucleus. A
contractile vacuole at just anterior and slightly
upper to macronucleus.

Measurement: Body length 35.7 +6.3 (24-55),
caudal process 4.2+2.3 (1-10), width 26.2+3.0
(21-42) pm, length/width ratio 1.36+0.20 (0.9-
1.7) (n=60).

Type specimens: Holotype, individual with re-
tracted adoral cilia on microslide, No. 18901. Col.
5 MAR 1988, Ito, and 2 paratypes are deposited in

the Department of Parasitology, Nippon Veterin-
ary and Zootechnical College, Musashino, Tokyo,
Japan.

Type host and locality: MUolstein-Friesian cat-
tle, Bos taurus taurus, in Hokkaido, Japan.

Habitat: Rumen.

Frequency: In 53.5% of the cattle surveyed.

Etymology: Entodinium okoppensis is named
after the place this new species was found.

Four formae may be distinguished based on the
number and shape of caudal processes.

Entodinium okoppensis forma okoppensis n. f.
(Figs. 1-a, b, c, d and 2-a, b)

452 A. Ito AND S. IMAI

Diagnosis: Three caudal processes; right one
pointed or dull spine and sometimes bends out-
ward, left-lower and left-upper ones pointed
spines.

Frequency: In 53.5 % of the cattle surveyed.

Entodinium okoppensis forma bispinosum n. f.
(Figs. 1-e, f and 2-c, d)

Diagnosis: Three caudal processes; right one
pointed or dull spine, left-upper one pointed spine,
left-lower one blunt lobe.

Frequency: In 5.6 % of the cattle surveyed.

Entodinium okoppensis forma bifidum n. f.
(Figs. 1-g and 2-e)

Diagnosis: Two caudal spines in the same
length at left side only.
Frequency: In 5.6 % of the cattle surveyed.

Entodinium okoppensis forma monospinosum n. f.
(Figs. 1-h, 1 and 2-f)

Diagnosis:
left-upper side.
Frequency: In 5.6 % of the cattle surveyed.

Remarks: Entodinium okoppensis closely re-
sembles Entodinium indicum Kofoid et MacLen-
nan, 1930 [9] and E. bubalum Imai, 1981 [14] in
the shape of body and macronucleus, and the
position of contractile vacuole. It is, however,
distinguished from both E. indicum and E. buba-
lum by the position of the cytoproct and of caudal
spines; that is, the largest caudal spine of both
species is situated at the center of body, and the
cytoproct lies at the center of body within the
central spine. The body shape of E. okoppensis
okoppensis also resembles E. triacum Buisson,
1923 [8, 15, 16], especially E. triacum dextrum
Dogiel, 1927 [8]. Unfortunately, none of descrip-
tions on the position of contractile vacuole which is
one of the most important taxonomic criteria were
made in E. triacum [8, 15, 16]. However, anterior
end of E. triacum is more rounded than E. okop-
pensis, and the adoral lips are visible in E. triacum
when the ciliate contracts the adoral cilia [8] but
Entodinium okop-

One pointed or dull caudal spine at

not visible in E. okoppensis.

pensis bifidum resembles E. bifidum Dogiel, 1927
[8] in the possession of two caudal spines at the
posterior left end, but it is easily distingushed from
E. bifidum by the difference in body shape and the
location of contractile vacuole.

Taxonomical comment: The caudal process is
one of the most prominent features under the
microscope. However, it is clearly shown that the
variation of caudal processes is continuous [3, 17,
18], thus these features are considered to be not
suitable as the taxonomical character for species at
present. In addition. the variation of caudal pro-
cesses is recognized within one and the same host
as also shown in E. okoppensis in the present
examination, so that it also may not be able to be
established as the character for subspecies. There-
fore, the features of caudal processes seem to be
the most adequate when they are used as the
classification of formae which are not prescribed
by the International Code of Zoological Nomencl-
ature. There are some opinions that the establish-
ment of formae is not necessary. However, the
caudal processes seem to be related to the dif-
ferentiation of rumen ciliates within the hosts,
since a peculiar spinated form, such as Epidinium
ecaudatum capricornisi has been found from only
one host, Japanese serow [19], in spite of very wide
distribution of other formae of this species, such as
E. ecaudatum ecaudatum and E. ecaudatum cauda-
tum [3]. Thus, the authors consider that it is
significant to create formae in a species of rumen
ciliates.

Ostracodinium munham n. sp.
(Figs. 3 and 4)

Description: Body ovoid. An apparent round
lobe to the right of cytoproct at posterior end of
body. Lobe with some variations (Fig. 3-b, c).
Left posterior end of body rounded and sometimes
projecting (Fig. 3-d). Large, slightly flattened
operculum at top of body. Clear lips surrounding
peristome and left ciliary zone when the ciliate
contracted. Rectum clear and forming a short slit.
Macronucleus rod-shaped tapering posteriorly
with bend some degree along the left periphery of
ectoplasm. An elliptical micronucleus at lower
side of the middle of macronucleus. Three con-

Ciliates from Cattle in Hokkaido, Japan 453

tractile vacuoles in tandem in ectoplasm at lower-
left side of macronucleus. Most anterior and
posterior vacuoles at same level as anterior and
posterior ends of macronucleus. Broad skeletal
plate over most of upper part of body forming a
weak arc and running obliquely to long axis from
just posterior of operculum to the level of the
posterior contractile vacuole. Left edge of the
plate turning inward and extending one-fifth of the
width of plate.

Measurement: Body length 78.7+7.5 (65-90),

width 54.6+4.3 (50-65) um, length/width ratio
1.43 +0.08 (1.3-1.6)(n=20).
Type specimens:

Holotype, individual with re-

50 pm

Fic. 3. Ostracodinium munham n. sp. a: Upper view of
whole body. b-d: Variations in the caudal lobes.

tracted adoral cilia on microslide, No. 18902. Col.
18 JAN 1988, Ito, and 6 paratypes are deposited in
the Department of Parasitology, Nippon Veterin-
ary and Zootechnical College, Musashino, Tokyo,
Japan.

Type host and locality: Holstein-Friesian cat-
tle, Bos taurus taurus, in Hokkaido, Japan.

Habitat: Rumen.

Frequency: In 2.8% of the cattle surveyed.

Etymology: Ostracodinium munham is named
after the possesion of lobe. Munha means a lobe in
Aino.

Remarks: This new species closely resembles
Ostracodinium iwawoi Imai, 1988 [13], in size and
shape of body and of skeletal plate, but O. iwawoi
differs from O. munham in its four contractile
vacuoles and a skeletal plate turning one-third of
its width. It also resembles O. trivesiculatum
Kofoid et MacLennan, 1932 [10], O. rugolorica-
tum Kofoid et MacLennan, 1932 [10] and O.
mammosum (Railliet, 1890) [8, 10], in having
three contractile vacuoles. However, it differs
from O. trivesiculatum in the folded skeletal plate
and the presence of posterior lobe, from O. rugo-
loricatum in possessing no skeletal plate turning
toward the middle of body, and from O. mammo-
sum in shape of skeletal plate, and shape and
number of posterior lobe of ectoplasm. In the
point of possession of a right posterior lobe, the
present species is also common with O. obtusum f.

Fic. 4. Photomicrographs of Ostracodinium munham n. sp. a: Cell fixed and stained with MFS solution. b: Cell
stained with Mayer’s hematoxylin. Bar in each figure indicates 50 ~m.

454 A. Ito AND S. IMalI

TABLE 1. Species composition and frequency of rumen ciliate protozoa found from the Holstein-
Friesian cattle in Hokkaido

Family Species Frequency (%)
Isotrichidae Dasytricha
ruminantium Schuberg 71.8
Isotricha
prostoma Stein 66.2
intestinalis Stein 33.8
Oligoisotricha
bubali (Dogiel) 9.9
Microcetus
lappus Orpin et Mathiesen* 1.4
Blepharocorythidae Charonina
ventriculi (Jameson) 59.2
Ophryoscolecidae
Entodiniinae Entodinium
nanellum Dogiel 100.0
simplex Dogiel 100.0
parvum Buisson 95.8
longinucleatum Dogiel 95.8
caudatum
f. caudatum Stein 85.9
f. lobosospinosum Dogiel Tiles
rostratum Fiorentini 76.1
exiguum Dogiel 60.6
dilobum (Dogiel) 54.9
okoppensis n. sp*
f. okoppensis n. f. SS)
f. bispinosum n. f. 5.6
f. bifidum n. f. 5.6
f. monospinosum n. f. 5.6
bursa Stein 33.8
ovinum Dogiel 1987.
bimastus Dogiel 12.7
minimum Schuberg 11.3
chatterjeei Das-Gupta* , iiss
bovis Wertheim* 5.6
rectangulatum Kofoid et MacLennan* 4.2
simulans Lubinsky* 2.8
dubardi Buisson 1.4
quadricuspis Dogiel* 1.4
Ophryoscolecidae
Diplodiniinae Diplodinium
anisacanthum Da Cunha
f. anisacanthum Da Cunha 50.7
f. monacanthum Dogiel ike)7/

f. anacanthum Dogiel 16.9

Ciliates from Cattle in Hokkaido, Japan 455
f. diacanthum Dogiel 7.0
f. triacanthum Dogiel 7.0
f. tetracanthum Dogiel 7.0
f. pentacanthum Dogiel 5.6
dentatum (Stein) 5.6
minor (Dogiel) 4.2
Eodinium
lobatum Kofoid et MacLennan ail
posterovesiculatum (Dogiel) 46.5
monolobosum (Hsiung)* 8.5
rectangulatum Kofoid et MacLennan* 1.4
Eudiplodinium
rostratum (Fiorentini) 84.5
dilobum (Dogiel) 50.7
magegii (Fiorentini) 42.3
bovis (Dogiel) 19.7
monolobum (Dogiel) 16.9
Polyplastron
multivesiculatum (Dogiel et Fedorowa) 39.4
Metadinium
affine Dogiel et Fedorowa 39.4
medium Awerinzew et Mutafowa 7.0
ypsilon (Dogiel)* 7.0
Ostracodinium
mammosum (Railliet) 45.1
gracile Dogiel 39.4
trivesiculatum Kofoid et MacLennan* Po Il
obtusum (Dogiel et Fedorowa) 19.7
clipeolum Kofoid et MacLennan 5.6
munham n. sp.* 2.8
Enoploplastron
triloricatum (Dogiel)* 1.4
Ophryoscolecidae
Ophryoscolecinae Epidinium
ecaudatum (Fiorentini)
f. caudatum Fiorentini 25.4
f. ecaudatum Fiorentini 18.3
f. cattanei Fiorentini 7.0
f. bulbiferum Dogiel 4.2
f. quadricaudatum Sharp 2.8
f. hamatum Schulze 1.4
Ophryoscolex
purkynjei Stein 14.1
15 genera
Total genera, species and formae 50 species
19 formae

* First record in Japanese cattle.

456 A. Ito AND S. IMAI

TABLE 2. Percentage generic composition of
the rumen ciliate protozoa in the Holstein-
Friesian cattle in Hokkaido*

Genus Mean Range
Entodinium 82.7 43.6-98.6
Eudiplodinium 46 0 -17.7
Diplodinium 2.5 0 -24.7
Dasytricha 2.0 0 - 8.0
Epidinium 2.0 0 -26.7
Ostracodinium 1.7 0 - 6.7
Eodinium 1.5 0 - 6.7
Isotricha 1.2 0 - 9.3
Charonina 0.6 0 - 4.0
Metadinium 0.6 0 - 4.7
Oligoisotricha 0.2 0 - 47
Polyplastron 0.2 0 - 2.0
Microcetus 0.1 0 - 0.7
Enoploplastron 0.1 0 - 2.0
Ophryoscolex 0.1 0 - 2.3
nT:

monolobum Dogiel, 1927 [8], but it is easily discri-
minated from O. munham in different body size
and number of contractile vacuoles.

Composition of rumen ciliates. Species and their
frequency from 71 Holstein-Friesian examined are
shown in Table1. Fifty species with 19 formae
under 15 genera were identified in all. Of them, 13
species were the first record in Japanese cattle.
Entodinium simplex and E. nanellum occurred in
all the hosts examined. Of the other ciliates, 6
species with 2 formae; Entodinium parvum, E.
longirucleatum, E. caudatum f. caudatum, E.
caudatum f. lobosospinosum, E. rostratum, Eudi-
plodinium rostratum and Dasytricha ruminantium,
were predominant, and the frequencies of them
were over 70% in the animals examined.

Table 2 shows the percentage compositions of

TABLE 3.

genera of the ciliates in this examination. The
percentage occupied by the genus Entodinium was
the highest, the ratio of which was 82.7% on
average. Though average ratio of the genus Eudi-
plodinium was next highest in number, Epidinium
and Diplodinium occasionally showed higher ratio
depending on the host individual.

The average number of species per head of host
and the total ciliate number per one milliliter of
rumen fluid are shown in Table. 3. The average
number of ciliate species was 17.2, and the average
ciliate density of 71 samples was 5.4 10°/ml.

DISCUSSION

Rumen ciliate compositions of the cattle in
Japan, mainly in Honshu were formerly surveyed
on various races without distinction by Imai et al.
[6, 7], and 42 species with 12 formae were iden-
tified. When the present results were compared to
that by Imai er al. [6, 7,], 37 species were common
in both areas, and the prominent species almost
coincided with each other; included are several
species of the genus Entodinium, E. nanellum, E.
parvum, E. simplex and E. longinucleatum.

Thirteen species with 7 formae including 2 new
species were the first record in Japanese cattle. Of
those, Entodinium bovis, E. rectangulatum, E.
simulans, E. quadricuspis, Mocrocetus lappus,
Metadinium ypsilon and Enoploplastron trilorica-
tum have been already detected from the various
races of humpless cattle (Bos taurus taurus) in
various areas [8, 20-22]. Microcetus lappus was
first described from the cattle in Norway [22], and
the present detection places the second report of
this species.

Entodinium chatterjeei, Eodinium monolobo-
sum, Eod. rectangulatum, and Ostracodinium
trivesiculatum have been found mainly from the

Average number of species appeared and average cilliate density in the
Holstein-Friesian cattle in Hokkaido*

Number of species

Mean Range

Ciliate density (x 10*/ml)

Mean Range

Wile 5-30

i= iil.

55r/ 14.5-168.2

Ciliates from Cattle in Hokkaido, Japan 457

domestic animals kept in tropical area [1, 2, 10, 13,
23, 24] but few from humpless cattle. There may
be two possible reasons for this. One is that
Holstein-Friesian cattle in Hokkaido might have
an experience of contact with the cattle introduced
from any tropical area, and the other is that these
ciliate species are originally worldwide as well as
E. nanellum and E. simplex. However, the former
possibility seems to be low, because Holstein-
Friesian cattle have been kept thoroughly in the
temperate zones, and have had very few chances to
contact with the tropical cattle. The high frequen-
cy of appearance of E. okoppensis was characteris-
tic of the rumen ciliate fauna of the cattle in
Hokkaido. The morphological features of the
anterior end of body and macronucleus, and the
position of contractile vacuole of E. okoppensis
closely resemble those of E. indicum and E. buba-
lum. hese two species have also been detected
mainly from the animals inhabiting tropical areas
[1, 2, 9, 13, 14], thus it seems to have a poor
relationship to Holstein-Friesian. However, it will
be interesting that Campylodinium ovumrajae, a
closely related species to Entodinium and detected

only from the forestomach of camels [8], has very
similar morphological aspects on the anterior end
of body, macronucleus and the position of contrac-
tile vacuole. Provided these species are assumed
to reflect their phylogenetic relations in spite of
their phylogenetically separated hosts, we can
speculate that such morphotype is of primitive
group and must have a wide distribution in various
ruminants. This speculation would be supported
by the fact that E. triacum with two formae which
has been widely detected from the cattle in Europe
[8, 25], Mexico [26] and China [23] also resembles
E. okoppensis, although the description of E.
triacum has been insufficient [8, 15, 16].

The average of the ciliate density and the aver-
age number of species per head of the host resem-
bled those in the cattle in Honshu and Kyushu
reported by Imai et. al. [6, 7]. The percentage
composition of genera also almost coincided with
the data from the Japanese cattle reported earlier
by us [7]. It is known that the ciliate density and
the composition of genera are strongly affected by
the kinds and amounts of food taken by the host
[27-29], and that when the host is fed with concen-

trates-rich ration, entodiniid ciliates rapidly grow,
and the ratio of entodiniid ciliates and the total
ciliate density become higher [29]. The similarity
of ciliate percentage composition in various areas
of Japan may be due to the similarity of feeding
conditions. Frequent introduction of the Japanese
dairy cattle from Hokkaido to Honshu may be
another reason.

ACKNOWLEDGMENTS

The authors wish to thank Mr. Shinkichi Kakuta,
owner of the Kakuta Farm, for cooperating in the
collection of the samples. Thanks are also due to Miss.
Kuniko Kitamura, of the Okoppe Library, Dr. Yoshikat-
su Morikawa and his colleagues, of the Nishimon A. M.
A. A., Okoppe Branch, for their helpful assistance in the
present examination. The first author expresses his
gratitudes to his wife, Takako Ito, for her continuous
help.

REFERENCES

1 Imai, S. (1985) Rumen ciliate protozoal faunae of
Bali cattle (Bos javanicus domesticus) in Indonesia,
with the description of a new species, Entodinium
Javanicum sp. nov. Zool. Sci., 2: 591-600.

2 Imai, S. (1986) Rumen ciliate fauna of zebu cattle
(Bos taurus indicus) in Sri Lanka, with the descrip-
tion of a new species, Diplodinium sinhalicum sp.
nov. Zool. Sci., 3: 699-706.

3 Ogimoto, K. and Imai, S. (1981) Atlas of Rumen
Microbiology, Japan Scientific Societies Press,
Tokyo, 231 pp.

4 Rouse, J. E. (1970) World Cattle, Vol. 2, Universi-
ty of Oklahoma Press, Norman, pp. 991-1001.

5 Shoda, Y. (1987) Domesticated Animals of The
World, Kodansha, Tokyo, 223 pp. (In Japanese)

6 Imai, S., Katsuno, M. and Ogimoto, K. (1978)
Distribution of rumen ciliate protozoa in cattle,
sheep and goat and experimental transfaunation of
them. Jpn. J. Zootech. Sci., 49: 494-505. (In
Japanese with English summary)

7 Imai, S., Shimizu, M., Kinoshita, M., Toguchi, M.,
Ishii, T. and Fujita, J. (1982) Rumen ciliate pro-
tozoal fauna and composition of the cattle in Japan.
Bull. Nippon Vet. Zootech. Coll., 31: 70-74. (In
Japanese with English summary)

8 Dogiel, V. A. (1927) Monographie der Familie
Ophryoscolecidae. Arch. Protistenkd., 59: 1-288.

9 Kofoid, C. A. and MacLennan, R. F. (1930) Cili-
ates from Bos indicus Linn. 1. The genus Entodi-
nium Stein. Univ. Calif. Publ. Zool., 33: 471-544.

10 Kofoid, C. A. and MacLennan, R. F. (1932) Cili-

11

12

13

14

15

16

18

19

458

ates from Bos indicus Linn. 2. A revision of Diplodi-
nium Schuberg. Univ. Calif. Publ. Zool., 37: 53-
152.

Kofoid, C. A. and MacLennan, R. F. (1933) Cili-
ates from Bos indicus Linn. 3. Epidinium Crawley,
Epiplastron gen. nov., and Ophryoscolex Stein.
Univ. Calif. Publ. Zool., 39: 1-33.

Imai, S. (1988) Some progress in taxonomy and
phylogeny of rumen ciliate protozoa. J. Jpn. Vet.
Med. Assoc., 41: 73-80. (In Japanese)

Imai, S. (1988) Ciliate protozoa in the rumen of
Kenyan zebu cattle, Bos taurus indicus, with the
descriotion of four new species. J. Protozool., 35:
130-136.

Imai, S. (1981) Four new rumen ciliates, Entodi-
nium ogimotol, sp. n., E. bubalum sp. n., E. fujitai
sp. n. and E. tsunodai sp. n., and Oligoisotricha
bubali (Dogiel, 1928) n. comb. Jpn. J. Vet. Sci., 43:
201-209.

Buisson, J. (1923) Infusoires nouveaux parasites
d’antilopes Africaines. C. R. Soc. Biol., 89: 1217-
1219.

Buisson, J. (1924) Quelques infusoires parasites
d’antilopes Africaines. Ann. Parasitol. Hum.
Comp., 2: 155-160.

Hempel-Zawitkowska, J. and Latteur, B. (1977)
Quelques remarques sur la systematique du genre
Entodinium Stein, 1858. Acta Zool. Pathol.
Antverp., 69: 189-202.

Poljansky, G. and Strelkow, A. (1934) Beobach-
tungen Uber die Variabilitat einiger Ophryoscoleci-
dae (Infusoria, Entodiniomorpha) in Klonen. Zool.
Anz., 107: 215-220.

Imai, S., Abe, M. and Ogimoto, K. (1981) Ciliate
protozoa from the rumen of the Japanese serow,
Capricornis crispus (Temminck). Jpn. J. Vet. Sci.,
43: 359-367.

20

21

22

23

24

25

26

27

28

A. ITo AND S. Imal

Wertheim. P. (1935) Infusorien aus dem Wieder-
kaduermagen von Gebiete Jugoslawiens nebst einer
Ubersicht dieser Tierchen von Balkanhalbinsel-
Bereich und ein kurzer Bericht tiber die Pferdedar-
minfusorien, zugleich Revision der familie Ophryos-
colecidae. Vet. Arh., 5: 386-536.

Lubinsky, G. (1957) Studies on the evolution of the
Ophryoscolecidae (Ciliata : Oligotricha). 1. A new
species of Entodinium with “caudatum” , “lobosospi-
nosum” and “dubardi” forms and some evolutionary
trends in the genus Entodinium. Can. J. Zool., 35:
111-133.

Orpin, C. G. and Mathiesen, S. D. (1986) Mic-
rocetus lappus gen. nov., sp. nov. : New species of
ciliated protozoon from bovine rumen. Appl. En-
vironm. Microbiol., 52: 527-530.

Hsiung, T.-S. (1932) A general survey of the
protozoan fauna of the rumen of the Chinese cattle.
Bull. Fan Mem. Inst. Biol., 3: 87-107.

Das-Gupta, M. (1935) Preliminary observations on
the protozoan fauna of rumen of the Indian goat,
Capra hircus Linn. Arch. Protistenkd., 85: 153-172.
Jirovec, O. (1933) Beobachtungen tiber die Fauna
des Rinderpansens. Z. Parasitenkd., 5: 584-591.
Chavarria, M. (1933) Estudios protistologicos. 1.
Fauna del tubo digestivo del toro (Bos taurus Linn.)
de Mexico. Ann. Inst. Biol. Univ. Nac. Mexico, 3:
109-142.

Ferber, K. E. (1928) Die Zahl und Masse der
Infusorien im Pansen und ihre Bedeutung fiir den
Eiweissaufbau beim Wiederkauer. Z. Tierzuecht.
Zuechtungsbiol., 12: 31-63.

Warner, A. C. I. (1962) Some factors influencing
the rumen microbial population. J. Gen. Microbiol.,
28: 129-146.

Hungate, R. E. (1966) Rumen and Its Microbes,
Academic Press, New York, 533 pp.

ZOOLOGICAL SCIENCE 7: 459-467 (1990)

© 1990 Zoological Society of Japan

The Drosophila polychaeta and the D. quadrisetata Species-
Groups (Diptera: Drosophilidae) from Yunnan
Province, Southern China

Hipe-AkI WATABE, XING CHAI LIANG!

and WEN X1A ZHANG!

Biological Laboratory, Sapporo College, Hokkaido University of Education,
Sapporo 002, Japan, and ‘Kunming Institute of Zoology,
Academia Sinica, Kunming, China

ABSTRACT—Three new and four known species of the Drosophila polychaeta and the D. quadrisetata
species-groups are reported from Yunnan Province, southern China. An evolutionary process of the
virilis-repleta Radiation is discussed on the basis of the recent information from southern China.

INTRODUCTION

The present paper deals with three new and four
known species of the Drosophila polychaeta and
the D. quadrisetata species-groups from Yunnan
Province, southern China, both of which belong to
the virilis section of the subgenus Drosophila.

Most of specimens described here were collected
at watersides, by using traps baited with ferment-
ing bananas. All the holotypes and a part of
paratypes are deposited in the Kunming Institute
of Zoology, Academia Sinica, Kunming, China,
and the remaining paratypes in the Biological
Laboratory, Hokkaido University of Education,
Sapporo, Japan.

DROSOPHILA POLYCHAETA SPECIES-GROUP

D. polychaeta species-group, Sturtevant, 1942,
Univ. of Texas Publ., 4213: 31.

Drosophila (Drosophila) daruma Okada

Drosophila (Drosophila) daruma Okada, 1956,
Syst. Study, 155.

Accepted July 21, 1989
Received May 31, 1989

Specimens examined. China: 1, 17, Kunm-
ing, 11. X. 1988 (Collector: H. Watabe); 3 7,12,
Simao, 4. XI. 1987 (X. C. Liang); 2%, Jinhong,
Xishuang-banna district, 13. IX. 1985 (W. X.
Zhang); 1%, Menghan, Xishuang-banna district,
21. IX. 1985 (W. X. Zhang).

Distribution. Korea, Japan, Malaya, Borneo,
India; China: Taiwan, Guangdong, Yunnan (n.
loc.).

Remarks. This species is relatively common in
southern and middle parts of Yunnan Province,
but has not been collected in its northern districts.

Drosophila (Drosophila) latifshahi Gupta
et Ray-Chaudhuri

Drosophila (Scaptodrosophila) latifshahi Gupta et
Ray-Chaudhuri, 1970 [1]: 67

Drosophila (Drosophila) latifshahi, Toda and
Peng, 1989 [2]: 155.

Specimens examined. China: 31/, 14%,
Simao, 4. XI. 1987 (X. C. Liang); 1, Menghan,
3. X. 1985 (W. X. Zhang).

Distribution. India, Bangladesh; China: Gu-
angdong, Yunnan (n. loc.).

Remarks. D. latifshahi is a dominat species of
waterside drosophilids, in Simao and Xishuang-
banna districts.

460 H. WataBE, X. C. LIANG AND W. X. ZHANG

Fics. 1-6.

Drosophila (Drosophila) polychaeta Patterson et Wheeler, 1942. 1: Periphallic organs. 2: Surstylus. 3:

Decasternum. 4: Phallic organs. 5: Aedeagus (lateral view). 6: Ovipositor. Signs: a, anterior paramere; c,
surstylus; e, aedeagus; n, novasternum; 0, aedeagal apodeme; r, vertical rod; t, cercus; v, ventral fragma.

Scale-line=0.1 mm.

Drosophila (Drosophila) polychaeta
Patterson et Wheeler
(Figs. 1-6)

Drosophila (Drosophila) polychaeta Patterson et
Wheeler, 1942 [3]: 102.

Patterson and Wheeler [3] described this species
based on the laboratory strain from Texas, but did
not refer to its genitalia. The present specimens
are supposed to belong to a native population of
D. polychaeta, and the description of the male and
female genitalia is made below, together with its
diagnostic characters.

Diagnosis. Brown species with 3 pairs of post-
sutural dorsocentral bristles. Palpus with ca. 2
moderate and ca. 18 short bristles. C-index ca.
1.96, C3-fringe ca. 0.93. Epandrium fused to cer-
cus at middle; anteroventral corner sharply
pointed; caudoventral corner rounded (Fig. 1).
Spermatheca unsclerotized.

Periphallic organs (Figs. 1-3): Epandrium
brown, darker on lower margin, pubescent except
in upper portion and ventral margin, with ca. 30
bristles on lower half. Surstylus distally constricted
into two parts; upper part flap-shaped, with tiny

thorn-like spines in somewhat regular rows; lower
part nearly quadrate, with ca. 4 primary teeth and
ca. 2 bristles on distal margin, and with ca. 5
bristles at caudoventral corner. Cercus oval, ven-
trally narrowing, entirely pubesent with ca. 53 long
bristles and tuft of ca. 11 short bristles at lower
apex. Decasternum pale brown, Y-shaped in
ventral view, medially with small dark patches.
Phallic organs (Figs. 4, 5): Aedeagus T-shaped
in lateral view, proximally broadened; aedeagal
apodeme short, ca. 1/4 as long as aedeagus. Anter-
ior paramere oval, without sensilla; posterior para-
mere absent. Vertical rod dark brown. Novaster-
num nearly triangular, without submedian spines.
Ventral fragma laterally flattened, distally con-
caved in middle.
repoductive organs (Fig. 6): Lobe of oviposi-
tor pale orange, dorso-submedially expanded, with
ca. 4 discal teeth, ca. 24 spine-like marginal teeth
and 1 subterminal hair; ultimate marginal tooth
darker than penultimate. Spermatheca very small,
embedded in adipose tissue.
Specimens examined. China:
Simao, 4. XI. 1987 (X. C. Liang).
Distribution.

112A; 41Bee

Neotropics, Micronesia, Hawaii,

Drosophila from Yunnan 461

North America, Europe; China (n. loc.): Yunnan.

Origin. D. polychaeta is cosmopolitan, but its
extremely wide range of distribution is probably
due to the propagation with man [4]. Fonseca [5]
states that D. polychaeta is frequently collected on
ships in British ports but does not establish its
permanent population there. The origin of this
species was unknown. The present collection was
made in a natural subtropical forest remote from a
human residence. This suggests that southern
China might be the original distribution range of
D. polychaeta.

DROSOPHILA QUADRISETATA
SPECIES-GROUP

D. quadrisetata species-group: Toda and Peng,
1989 [2]: 158.

This group is very small, and consisted of only
three species: D. potamophila Toda et Peng and
D. beppui Toda et Peng from southern China, and
D. quadrisetata Takada, Beppu et Toda from
northern Japan. The last species was previously
included in the polychaeta species-group [6].
Three new species are added in this article.

Drosophila (Drosophila) potamophila Toda et Peng

Drosophila (Drosophila) potamophila Yoda et
Peng, 1989 [2]: 159.

Specimens examined. China: 1 J, 12, Simao,
Yunnan Province, 4. XI. 1987 (X. C. Liang).

Distribution. China: Guangdong, Yunnan (n.
loc.).
Remarks. his species is abundant in subtro-

pical districts of Yunnan, but has not been col-
lected in Kunming (center of Yunnan) and Dali
(northern Yunnan) districts.

Drosophila (Drosophila) karakasa
Watabe et Liang, sp. nov.
(Figs. 7-14)

Diagnosis. Small and yellowish brown species
with cercus separated from epandrium. Palpus
short, with short hairs but without stout bristles
(Fig. 7). 4C-index ca. 7/9 and C3-fringe ca. 3/5.
Surstylus rectangular, distally with ca. 7 primary
teeth and ca. 7 short bristles (Fig. 9). Lobe of
Ovipositor brown, much darker on ventral margin,
roundish at tip (Fig. 13). Spermatheca cone-

Fics. 7-14.

Drosophila (Drosophila) karakasa Watabe et Liang, sp. nov. 7: Palpus.
Surstylus. 10: Decasternum. 11: Phallic organs. 12: Aedeagus (lateral view). 13: Ovipositor. 14: Spermatheca.
Signs and scales as in figs. 1-6.

8: Periphallic organs. 9:

462 H. WatTaBE, X. C. LIANG AND W. X. ZHANG

shaped, with sparce horizontal stripes on basal half
of outer capsule (Fig. 14).

JS, %. Body length, # ca. 2.33 mm (range: 2.1-
2.5), 2 ca. 2.31 mm (2.1—-2.5). Wing length, /

ca. 2.83 mm (2.6-3.0), ? ca. 2.73 mm (2.6-2.8).

Head: Eye red with thick piles. Second joint
of antenna reddish brown; 3rd grayish brown.
Arista with ca. 4 (4-6) upper and ca. 2 (2-3) lower
short branches in addition to terminal fork. Frons
dark brown, ca. 0.54 (0.43-0.65) as broad as head,
anteriorly with a few frontal hairs. Anterior reclin-
ate orbital (Orb 2) ca. 0.44 (0.33-0.56) length of
posterior reclinate orbital (Orb 1); proclinate
orbital (Orb 3) ca. 0.63 (0.56-0.78) length of Orb
1. Face brown; carina somewhat low, narrow.
Clypeus reddish brown. Cheek tannish brown, ca.
0.25 (0.20—-0.31) as broad as maximum diameter of
eye, with ca. 3 long bristles along lower margin.
Second oral (Or 2) minute, ca. 0.29 (0.23-0.36)
length of vibrissa (Or 1). Palpus brown, club-
shaped, basally baring (Fig. 7).

Thorax: Mesoscutum yellowish brown, medi-
ally with a darker longitudinal stripe running to
scutellum. Scutellum brown, paler on lateral sides.
Lower humeral ca. 0.70 (0.55—-0.86) length of
upper one. Two extra pairs of dorsocentrals
present in front of usual ones. Anterior acrostichal
bristles present between Ist (anteriormost) dor-
socentrals; posterior ones between 2nds; length
and location of acrostichal bristles more or less
variable. Relative lengths of dorsocentrals and
acrostichal bristles to 4th (posteriormost) dor-
socentral: 1st dorsocentral ca. 0.57 (0.51-0.62),
2nd ca. 0.57 (0.49-0.62), 3rd ca. 0.71 (0.65—0.79),
anterior acrostichal bristle ca. 0.35 (0.29-0.43),
posterior one ca. 0.53 (0.46—-0.68). Length dis-
tance from 1st dorsocental to 2nd ca. 0.57 (0.52-
0.68), distance from 2nd to 3rd ca. 0.48 (0.44-
0.55), distance from 3rd to 4th ca. 0.56 (0.52-0.64)
cross distance between 3rds. Acrostichal hairs
(Ac) sparce, in 4 irregular rows. Anterior scutel-
lars (SctA) nearly parallel and posterior ones
(Sctp) convergent; SctA ca. 1.07 (0.88-1.17)
length of SctP. Sterno-index ca. 0.72 (0.64—0.76).

Legs light brown; preapicals on all three tibiae;
apicals on fore and mid tibiae.

Wing hyaline, slightly fuscous. Veins dark
brown; crossveins clear. R>+3 straight; Ry,5 and

M parallel. C, bristles 2, subequal. Number of
small stout bristles on 3rd costa (3CFr) ca. 28 (24-
33). Wing indices: Cin J ca. 3.10 (2.96-3.34) and
in ? ca. 2.77 (2.50-3.11), 4V ca. 1.64 (1.54-1.87),
AC ca. 0.77 (0.70-0.86), 5X ca. 1.57 (1.33-1.73),
Ac in & ca. 2.26 (2.17-2.36) and in $ ca. 2.64
(2.14-3.00), C3-fringe ca. 0.61 (0.52-0.67). Hal-
tere white, basally brown.

Abdomens: Tergites brown, darker on middle
and paler on lateral margin. Sternites brown,
darker on posterior margin, nearly quadrate.

Periphallic organs (Figs. 8-10): Epandrium yel-
lowish brown, darker on anterior margin, pubes-
cent on posterior half, with ca. 9 bristles on lower
half. Surstylus pale brown, marginally darker,
somewhat swollen at caudodorsal corner. Decas-
ternum translucent, heart-shaped. Cercus brown,
slightly projecting at ventral apex, entirely pubes-
cent, with ca. 17 long bristles and tuft of ca. 5 short
bristles along lower margin.

Phallic organs (Figs. 11, 12): Aedeagus yellow,
bilobed, ventrally broadened; apodeme dark
brown, ca. 3/8 as long as aedeagus. Anterior
paramere small. Vertical rod black, plate-shaped
in ventral view. Novasternum pale brown, without
submedian spines; ventral fragma narrow.

$ reproductive organs (Figs. 13, 14): Lobe of
ovipositor with ca. 3-5 discal teeth and ca. 17 short
marginal teeth: first 2 marginal teeth darker and
larger than others. Spermatheca grayish brown,
slightly constricted in middle; introvert deep.

Holotype #, China: Xianguan, Dali district,
Yunnan Province, 19. IX. 1988 (X. C. Liang).

Paratypes, China: 1, same data as holotype,
2 2., Dabochin, Dali district, Yunnan Province, 21.
IX. 1988 (X.C. Liang).

Distribution. China: Yunnan; Dabochin,
Xianguan.
Relationships. D. karakasa somewhat resem-

bles the foregoing species, D. potamophila, in the
general morphology and chaetotaxy, but clearly
distinguishable from the latter by the diagnostic
characters. The aedeagus of this species is very
similar to that of four species of the D. robusta
species-group: D. okadai Takada, D. neokadai
Kaneko et Takada, D. gani Liang et Zhang and D.
unimaculata Strobl [7].

Drosophila from Yunnan 463

Fics. 15-20. Drosophila (Drosophila) barutani Watabe et Liang, sp. nov. 15: Periphallic organs. 16: Surstylus. 17:
Phallic organs. 18: Aedeagus (lateral view). 19: Ovipositor. 20: Spermatheca. Signs and scales as in figs. 1-6.

Drosophila (Drosophila) barutani
Watabe et Liang, sp. nov.
(Figs. 15-20)

Diagnosis. Dull brown species with cercus
close to epandrium at middle (Fig. 15). Third oral
subequal to vibrissa. Palpus with ca. 3 moderate
bristles. 4C-index ca. 5/9 and C3-fringe ca. 9/10.
Surstylus arc-shaped, broadened at caudodorsal
corner (Fig. 16). Lobe of ovipositor sharply
pointed at tip; ultimate marginal tooth large, bris-
tle-like (Fig. 19). Spermatheca hemispherical
(Fig. 20).

Jd, ¥. Body length, ca. 3.60mm (3.4-3.8).
Wing length, ca. 4.03 mm (3.9-4.1).

Head: Eye brownish red with thick piles.
Second joint of antenna reddish brown; 3rd black-
ish brown. Arista ca. 4 (4-6) upper and ca. 2 (1-2)
lower short branches in addition to terminal fork.
Frons reddish brown, ca. 0.47 (0.46—-0.48) as broad
as head, anteriorly with a few frontal hairs. Orb 2
ea. 0.32 (0.31—0.34) length of Orb 1; Orb 3 ca. 0.40
(0.33-0.47) length of Orb 1. Face brown; carina

broad. Clypeus dark red. Cheek reddish brown,
ca. 0.28 (0.24-0.33) as broad as maximum dia-
meter of eye, with ca. 3 long bristles along lower
margin. Or 2 thin, ca. 0.16 (0.14—0.17) length of
Or 1; Or 3 ca. 0.99 (0.92-1.11) length of Or 1.
Palpus grayish brown, laterally flattened.

Thorax: Mesoscutum brown, with 4 darker
longitudinal stripes. Scutellum brown, lateral sides
black. Lower humeral ca. 0.58 (0.55-0.59) length
of upper one. Two extra pairs of dorsocentrals
present. Anterior acrostichal bristles present be-
tween Ist dorsocentrals; posterior slightly below
cross line between 2nds. Relative lengths of dor-
socentrals and acrostichal bristles to 4th dor-
socentral: 1st (anteriormost) dorsocentral ca. 0.66
(0.65—-0.67), 2nd ca. 0.68 (0.62—-0.74), 3rd ca. 0.86
(0.79-0.98), anterior acrostichal bristle ca. 0.58
(0.48-0.70), posterior one ca. 0.54 (0.52-0.57).
Length distance from 1st dorsocental to 2nd ca.
0.64 (0.60-0.71), distance from 2nd to 3rd ca. 0.55
(0.50-0.61), distance from 3rd to 4th ca. 0.58
(0.55-0.61) cross distance between 3rds. Ac

464 H. Watase, X. C. LIANG AND W. X. ZHANG

sparce, in 6 irregular rows. SctAs parallel and
Sctps convergent; SctA ca. 0.97 (0.93-1.03) length
of SctP. Sterno-index ca. 0.72 (0.52-0.84).

Legs dark brown; coxae and trochanters paler.
Fore femur posteriorlly with ca. 5 bristles.
Preapicals on all three tibiae; apicals on fore and
mid tibiae.

Wing hyaline, slightly fuscous. Veins dark
brown; crossveins clear. R43 nearly straight;
R45 and M parallel. C, bristles 2, inner bristle ca.
5/9 length of outer one. Number of 3CFr ca. 38
(35-40). Wing indices: C ca. 3.46 (3.04-3.70), 4V
ca. 1.61 (1.56-1.70), 4C ca. 0.57 (0.53-0.62), 5X
ca. 1.06 (1.00-1.17), Ac ca. 2.01 (1.82-2.20),
C3-fringe ca. 0.90 (0.87-0.94). Haltere pale yel-
low; stalk anteriorly darker.

Abdomens: Tergites entirely dark brown.
Sternites pale grayish brown, nearly quadrate; f
Sth slightly convexed posteriorly.

Periphallic organs (Figs. 15, 16): Epandrium
brown, dorsally narrowing and __ ventrally
broadened, posteriorly pubescent except lower
portion, with ca. 12 bristles. Surstylus dark brown,
distally with ca. 7 primary teeth on margin and ca.
3 bristles at caudoventral corner, medially with ca.
7 spine-like bristles on outer surface. Cercus dark
brown, somewhat projecting ventrally, entirely
pubescent, with ca. 38 long bristles and tuft of ca.
12 short bristles at lower apex.

Phallic organs (Figs. 17, 18): Aedeagus yellow-
ish brown, much darker at distal portion, bilobed,
concaved on distal margin; apodeme short, ca. 2/7
as long as aedeagus. Anterior paramere rudiment.
Vertical rod brown, ventrally black. Novasternum
narrow; ventral fragma handmill-shaped.

? reproductive organs (Figs. 19, 20): Lobe of
Ovipositor brown, marginally black, dorso-
subapically swollen, with ca. 4 discal teeth and ca.
13 marginal teeth in regular row; ultimate marginal
tooth prominent, ca. 3 times as long as penulti-
mate. Spermatheca pale yellow, apically some-
what flattened, wrinkled on basal margin, without
apical indentation; introvert deep, ca. 5/8 height of
outer capsule.

Holotype ~#, China: Dabochin, Dali district,
Yunnan Province, 21. IX. 1988 (X. C. Liang).

Paratypes, China: 1 #\, Xianguan, Dali district,
Yunnan Province, 21. IX. 1988, (X. C. Liang);

1, same data as holotype.

Distribution. China: Yunnan; Dabochin,
Xianguan.
Relationships. D. barutani is somewhat similar

to D. potamophila in the abdominal coloration and
large value of C3-fringe, but easily distinguishable
from the latter by the shapes of its aedeagus and
Ovipositor.

Drosophila (Drosophila) multidentata
Watabe et Zhang, sp. nov.
(Figs. 21-27)

Diagnosis. Dull brown species, with cercus
fused to epandrium at submedian portion (Fig.
21). Or2 ca. 4/9 length of Or 1. C3-fringe ca. 5/9.
Lobe of ovipositor with many irregular teeth (Fig.
26). Spermatheca slender, with sparce oblique
lines on basal 1/3 of outer capsule (Fig. 27).

Jd, %. Body length, # ca. 2.55 mm (2.4-2.8),
$ ca. 2.85 mm (2.6-3.1). Wing length, J ca. 3.18
mm (3.0-3.5), 2 ca. 3.48 mm (3.2-3.8).

Head: Eye dark red with thick piles. Second
joint of antenna dark brown; 3rd grayish brown.
Arista with ca. 4 (3-4) upper and ca. 1 lower
branches in addition to short terminal fork. Frons
dark brown, ca. 0.46 (0.42—0.49) as broad as head,
medially with black cuneiform line. Orb 2 ca. 0.36
(0.25—0.46) length of Orb 1; Orb 3 ca. 0.52 (0.42-
0.79) length of Orb 1. Face reddish brown; carina
very high, wider below. Clypeus blackish brown.
Cheek brown, ca. 0.26 (0.23-0.31) as broad as
maximum diameter of eye, with ca. 3 long and ca.
11 short bristles along lower margin. Or 2 thin, ca.
0.44 (0.23-0.61) length of Or 1; Or 3 minute.
Palpus brown, small, club-shaped, with 1 some-
what long bristle at tip.

Thorax: Mesoscutum brown, medially darker;
scutellum dark brown. Lower humeral ca. 0.61
(0.44-0.69) length of upper one. Anterior acros-
tichal bristles present below cross line between Ist
dorsocentrals; posterior below cross line between
2nds. Relative lengths of dorsocentrals and acros-
tichal bristles to 4th dorsocentral: 1st dorsocentral
(anteriormost) ca. 0.46 (0.39-0.52), 2nd ca. 0.51
(0.39-0.58), 3rd ca. 0.68 (0.60-0.79), anterior
acrostichal bristle ca. 0.33 (0.24-0.44), posterior
one ca. 0.42 (0.37-0.44). Length distance from Ist

Drosophila from Yunnan 465

Fics. 21-27. Drosophila (Drosophila) multidentata Watabe et Zhang, sp. nov. 21: Periphallic organs. 22: Surstylus.
23: Decasternum. 24: Phallic organs. 25: Aedeagus (lateral view). 26: Ovipositor. 27: Spermatheca. Signs and

scales as in figs. 1-6.

dorsocental to 2nd ca. 0.55 (0.44-0.74), distance
from 2nd to 3rd ca. 0.45 (0.39-0.48), distance from
3rd to 4th ca. 0.51 (0.46-0.56) cross distance
between 3rds. Ac in 6 irregular rows; a few
acrostichal hairs in rows of dorsocentrals some-
what longer than other hairs. SctAs slightly and
SctPs heavily convergent; SctA ca. 0.96 (0.86-
1.05) length of SctP. Sterno-index ca. 0.73 (0.52-
0.96).

Legs brown; fore tarsi darker. Fore femur post-
eriorly with ca. 2-3 long bristles. Preapicals on all
three tibiae; apicals on fore and mid tibiae.

Wing hyaline, slightly fuscous. Veins dark
brown; crossveins clear. R 4,3 nearly straight;
R4,s5 and M parallel. C, bristles 2, subequal.
Number of 3CFr ca. 24 (17-28). Wing indices: C
ca. 3.54 (3.12—4.20), 4V ca. 1.69 (1.52-1.78), 4C
ca. 0.68 (0.60-0.77), 5X ca. 1.22 (1.00-1.50), Ac
ca. 2.00 (1.67—2.13), C3-fringe ca. 0.56 (0.43-
0.65). Haltere white; stalk grayish brown.

Abdomens: Tergites grayish brown, darker in
middle; sternites brown, each with ca. 26-34 bris-
tles.

Periphallic organs (Figs. 21-23): Epandrium
brown, darker on anterior margin, posteriorly
pubescent, with ca. 15 bristles on lower half and
ca. 2 bristles along ventral margin. Surstylus
brown, darker on upper half, distally narrowing,
slightly projecting at caudodorsal corner, with ca.
6 primary teeth and ca. 2 bristles. Decasternum
dark brown, paler on lower portion. Cercus black-
ish brown, with ca. 34 long bristles and tuft of ca.
23 pale yellow bristles along ventral margin.

Phallic organs (Figs. 24, 25): Aedeagus yel-
low, bilobed, submedially broadened; apodeme
dark brown, ca. 1/3 as long as aedeagus. Anterior
paramere pale yellow, hemispherical. Vertical rod
black, recurved dorsally. Novasternum pale yel-
low, darker on lateral margin, without submedian
spines; ventral fragma slightly concaved at middle.

466 H. WataBeE, X. C. LIANG AND W. X. ZHANG

9 reproductive organs (Figs. 26, 27): Lobe of
ovipositor dark brown, with ca. 45 stout teeth in
irregular rows and 1 long subterminal hair; ca. 8
upper teeth much darker than others. Spermathe-
ca grayish brown, slightly wrinkled basally; intro-
vert deep; inner duct narrow.

Holotype g, China: Xianguan, Dali district,
Yunnan Province, 19. IX. 1988 (X. C. Liang).

Paratypes, China: 29, 2, Dabochin, Dali
district, Yunnan Province, 21. IX. 1988 (H.
Watabe & X. C. Liang).

Distribution. Yunnan; Dabochin, Xianguan,
Kunming.
Relationships. One of the diagnostic characters

for the D. quadrisetata species-group described by
Tada and Peng [2] is “cercus separated from
epandrium”. The cercus of D. multidentata fuses
to epandrium at its lower portion. However, D.
multidentata should be involved in the quadrisetata
species-group by the following characters: 1) two
extra pairs of dorsocentrals and prominent acros-
tichal bristles present, 2) C-index ca. 3.54, 3)
4V-index ca. 1.69, and 4) aedeagus large and
curved ventrally.

Since the cercus in the robusta species-group
fuses to epandrium, the same type cercus found in
D. multidentata, as well as the large value of
C-index and the shape of aedeagus, implies the
phylogenetic relationship between this species and
the robusta group [2, 7].

THE VIRILIS-REPLETA RADIATION
IN THE OLD WORLD

The virilis-repleta Radiation, which might have
occurred during the Oligocene to early Miocene, is
one of main lineages in the evolution of the genus
Drosophila. Throckmorton [4] considers that first
the polychaeta group might have emerged in the
Old World tropics and then several groups, e.g.,
the robusta, the virilis and the melanica species-
groups, might have diverged adaptively in its
temperate forest. However, the phylogenetic rela-
tionship among these species-groups, especially
between the polychaeta group and other groups,
was still open to question, mainly due to the
insufficient information from China.

The recent Drosophila-survey in southern China

has resulted in the establishment of a new species-
group, the quadrisetata group. This group is
closely related to the polychaeta group in the
external morphology and to the robusta group in
the male genitalua. Toda and Peng [2] consider
that the quadrisetata group occupies a systematic
position between these two species-groups. Simi-
larly, a geographical information on the distribu-
tion of these three groups has made it possible to
trace the evolutionary process. Most of the
polychaeta group flies are distributed from the
tropics to the subtropics of the East Asia, whereas
the robusta group flies in its temperate zone [1, 2,
4, 8]. The distribution range of the quadrisetata
group overlaps with that of the polychaeta group
and that of the robusta group. Of six quadrisetata
group species, D. potamophila and D. beppui are
distributed in the subtropics and the remaining
four species in the temperate forest [2, 6, 8]. In
particular, in northern Yunnan, the present three
new species are sympatric to D. neokadai and D.
gani of the robusta species-group [9].

These information from southern China, includ-
ing the discovery of D. polychaeta in a natural
forest of Simao, strongly supports the Throckmor-
ton’s hypothesis: the polychaeta group first
emerged in the Old World tropics and then the
robusta group in the temperate forest of the East
Asia probably through the emergence of the quad-
risetata group in its subtropics.

ACKNOWLEDGMENTS

The authors are grateful to Prof. Osamu Kitagawa
(Tokyo Metropolitan University) and to Dr. Masanori J.
Toda (Hokkaido University) for their interest and advice
during this study. This work was supported by Grants-in
Aid for Overseas Scientific Survey from the Ministry of
Education, Science and Culture, Japan (Nos. 62041085,
63043060).

REFERENCES

1 Gupta, J. P. and Ray-Chaudhuri, S. P. (1970) Some
new and unrecorded species of Drosophila (Diptera:
Drosophilidae) from India. Proc. Roy. ent. Soc.
Lond., (B) 39: 57-72.

2 Toda, M. J. and Peng, T. X. (1989) Eight species of
the subgenus Drosophila (Diptera: Drosophilidae)
from Guangdong Province, southern China. Zool.
Sci., 6: 155-166.

Drosophila from Yunnan

Patterson, J. T. and Wheeler, M. R. (1942) Descrip-
tion of new species of the subgenus Hirtodrospphila
and Drosophila. Univ. of Texas. Publ., 4213: 67-109.
Throckmorton, L. H. (1975) The phylogeny, ecolo-
gy, and geography of Drosophila. In “Handbook of
Genetics, Vol. III”. Ed. by R. C. King, Plenum
Publ., New York, pp. 421-469.

Fonseca, E. C. M. d’A (1965) A short key to the
British Drosophilidae (Diptera) including a new spe-
cies of Amiota. Trans. Soc. Brit. Ent., 16: 233-244.
Takada, H., Beppu, K. and Toda, M. J. (1979)
Drosophila Survey of Hokkaido, XXXVI. New and
unrecorded species of Drosophilidae. J. Fac. Gener.
Edu., Sapporo Univ., 14: 105-129.

7

8

467

Watabe, H. and Nakata, S. (1989) A comparative
study of genitalia of the Drosophila robusta and D.
melanica species-groups (Diptera: Drosophilidae). J.
Hokkaido Univ. of Education, Ser. IIB, 40: 1-18.
Beppu, K., Peng, T. X. and Xie, L. (1989) An
ecological study on drosophilid flies (Diptera, Dro-
sophilidae) living at watersides in southern China.
Jpn. J. Ent. (Kontya, Tokyo), 57: 185-198.
Watabe, H., Liang, X. C. and Zhang, W. X. (1990)
The Drosophila robusta species-group (Diptera: Dro-
sophilidae) from Yunnan Province, southern China,
with the revision of its geographic distribution. Zool.
Sci., 7: 133-140.

mil) aqpine

a he yhitrte
ty ine” te’ gore!
‘Aitaetolk cit
ni ic Fer anes.
, (oocyte
ra | Gist , ti “ays
af Tyg es a)

ee on,

Caer ole Bat odd eOk
Poder Wott bor stAtumguibp gbie
Tye antennal gay
ut 0 ue otiar aah,

BMA.

tbs dyes

is aKr thirty a

Shalt wntct.

Hi Laken O¥

wat a aneorth ott i

4% obits is th

y) Se)

a)
einSaggpe rom

an

fim! “1 ade resi sf

Sytem ewessk Tey

(OTOL) :' er
bua, ot TEKS
Saar) ant t TMG
Ae wugity ec ,
; gel
resANT Sins in ws

mili: Fame :
viame with thot-ef ig

: sg hee OT t

ake D. poternipml
il in, the DUDS
iver gry Uhat oinpeenteagl
iw coribers. TURES
re we aye tay
Cibert sho g0Gae
vation Lion a
a polyeld
Tipe
— “id World a

4

4

* LeTweritag
wails ested

> Ope

=
' >? Merwe ¥S 3
“a
} i F
iu TATA,
cee ol LM
4A
* ,
‘On Be
rP OinteTres
var). 4? Py wile

ZOOLOGICAL SCIENCE 7: 469-475 (1990)

© 1990 Zoological Society of Japan

Cestodes of Field Micromammalians (Insectivora)
from Central Honshu, Japan

ISAMU SAWADA and Masasui HARADA!

Biological Laboratory, Nara Sangyo University, Nara 636, and
‘Laboratory of Experimental Animals, Osaka City
University Medical School, Osaka 545, Japan

ABSTRACT—Two new species of hymenolepidid and one new species of dilepidid cestodes were
obtained through the examination of 18 shrews belonging to three species of three genera, collected at
Toyama and Nagano Prefectures from October 6 to November 15, 1988. Staphylocystis (Staphylocystis)
toyamaensis sp. n. from Crocidura dsinezumi chisai is related to but differs from S.(S.)solitaria in the
shape of rostellar hooks. Ditestolepis longicirrosa sp. n. from Sorex shinto shinto is related to but differs
from D. diaphona in many morphological characters. Amoebotaenia urotrichi sp. n. from Urotrichus
talpoides hondonis resembles four known species of Amoebotaenia in being armed with 10-12 rostellar
hooks, but differs from all of them in the length and shape of rostellar hooks. This is the first record of

the genus Amoebotaenia from wild animals.

INTRODUCTION

The cestode parasites of Insectivora in Japan
have been unknown for the most part except two
by Sawada and Harada [1], who described Vampir-
olepis notoensis from Crocidura dsinezumi chisai
collected at Suzu-shi, Ishikawa Prefecture and V.
amamiensis from C. horsfield watasei at Amami-
Oshima, Kagoshima Prefecture.

Hymenolepidid cestodes collected from the
dsinezumi-shrew, C. dsinezumi chisai Thomas at
Asahi-machi and Tateyama-machi, Toyama Pre-
fecture, the shinto-shrew, Sorex shinto shinto Tho-
mas at Ina-shi, Nagono Prefecture, and dilepidid
cestodes from the Japanese shrew-mole, Urot-
richus tapoides hondnis Thomas at Ina-shi, Nago-
no Prefecture, are undescribed species of Staphy-
locystis Villot, 1877, Ditestolepis Stotty, 1952 and
Amoebotaenia Cohn, 1900, respectively.

MATERIALS AND METHODS

The shrews were autopsied immediately at the
collecting sites and their intestinal tracts were fixed

Accepted August 7, 1989
Received July 5, 1989

in Carnoy’s fluid and bought back to the labora-
tory. After being soaked in 45% acetic acid for
five hr for expanding, they were cut open in 70%
ethanohol and examined for cestodes. The ces-
todes obtained were stored in 70% ethanohol. The
scoleces, eggs and a part of mature segments were
unstained and observed under an interference con-
trast light microscope. The strobilae were stained
with ethanohol-hydrochloride-carmine, dehy-
drated in ethanohol, cleared in xylene, and
mounted in Canada balsam. Measurements are
given in millimeters.

Staphylocystis (Staphylocystis) toyamaensis sp. n.
(Figs. 1-6)

From November 13 to 14, 1988, three specimens
of Crocidura dsinezumi chisai Thomas were cap-
tured by trap at Asahi-machi, Shimoniikawa-gun
and Tateyama-machi, Nakashinkawa-gun, Toya-
ma Prefecture. On dissection, they were found
infected with 1-48 mature cestodes.

Description: Small-sized hymenolepidid; ma-
ture strobila 3-4 in length and 0.4—0.5 in max-
imum width. Mature segments serrate. Scolex
0.210—-0.224 long by 0.245-0.252 wide, not sharply
from strobila. Suckers circular, 0.084—-0.091 in

470 I. SAWADA AND M. HarRaDA

2

Fics. 1-5.

diameter. Rostellum 0.042 long by 0.063 wide,
armed with a crown of 16 thorn-shaped hooks
measuring 0.014 long, blade long, slender and
pointed, guard shorter and thick, and handle long.
Rostellar sac elongated, 0.140 long by 0.112 wide,
extending past posterior margin of sucker. Neck
absent.

Genital pores unilateral, located slightly anter-

Staphylocystis (Staphylocystis) toyamaensis sp. n.
1: Scolex. 2: Rostellar hook. 3: Mature segment. 4: Egg.

5: Detached senile segment. Scale in mm.

ior to middle of segment margin. Cirrus sac
cylindrical, 0.084-0.140 long by 0.021-0.028 wide,
extending beyond longitudinal osmoregulatory
canals. Internal seminal vesicle 0.028 long by
0.021—0.035 wide, occupying almost whole of cir-
rus sac. External seminal vesicle 0.025—0.028 long
by 0.021 wide. Testes three in number, ovoid,
0.126 by 0.035, arranged in a form of triangle, one

Cestodes of Field Micromammalians 471

poral and two aporal. Testes not in contact with
longitudinal osmoregulatory canals laterally. Ov-
ary triangular, 0.105—0.140 across, located at post-
erlor margin of two posterior testes. Vitelline
gland compact, 0.049-0.056 long by 0.021-0.028
wide. Seminal receptacle well developed, 0.077 by
0.049. Gravid uterus sacculated, not filling whole
of segment. Ripe eggs spherical, 0.042—0.053 in
diameter; surrounded by four thin envelopes.
Onchospheres spherical, 0.028—0.032 in diameter;
embryonic hooks 0.014 long.

Host: Crocidura dsinezumi chisai Thomas,
1906.

Site of infection: Small intestine.

Locality and date: Ogawa’motoyu, Asahi-
machi, Shimoniikawa-gun and Senjugahara,
Tateyama-machi Nakaniikawa-gun, Toyama Pre-
fecture; November 13-14, 1988.

Type specimen: Holotype: NSU Lab. Coll.
No. 9005; Paratypes: No. 9006.

Remarks: Yamaguti [2] divided the genus
Staphylocystis Villot, 1877, into two subgenera: S.
(Staphylocystis) Villot, 1877 and S. (Staphylocy-
stoides) Yamaguti. 1959. The present new cestode
finds its place in subgenus Stahylocystis on the base
of a single row of rosteller hooks and disposition of
testes. Seven species; S. scalaris (Dujardin, 1845)
Villot, 1788, S. tiara (Dujardin, 1854) Spasski,
1950, S. furcata (Stieda, 1862) Spasskii, 1950, S.
dodecantha (Baer, 1952) Spasskii, 1950, S. solitaria
(Meggitt, 1927) Yamaguti, 1959, S. fuelleborni
(Hilmy, 1936) Spasskii, 1950, S. loossi (Hilmy,
1936) Spasskui, 1950, were described from the
Crocidura (Yamaguti [2], Spasskii [3], Olsen and
Kuntz [4], Schmidt [5]). The present new species
closely resembles S. solitaria [6], of which the
description is deficient in some details, in the

0.01

Fic. 6. Rostellar hook of Staphylocystis (S.) solitaria.
Scale in mm.

number and length of rostellar hooks. However, it
differs from that species in the shape of rostellar
hooks (Fig. 6).

Ditestolepis longicirrosa sp. n.
(Figs. 7-14)

A number of specimens of cestodes representing
a species of Ditestolepis Stotty, 1952, were found in
one shinto-shrew, Sorex shinto shinto Thomas cap-
tured at October 8, 1988.

Fic. 7. Ditestolepis longicirrosa sp. n.
Scale in mm.

Entire worm,

Description: Small-sized hymenolepidid; worm
length 2.1-2.6; maximum width 0.3-0.4.
Metamerism distinct, margins not serrate. Strobila
characterized by a distinctly marked subdivision
into three series, each of which possessing seg-
ments uniformly advanced in development. First
series containing 26-34 immature segments; its
maximal width 0.19-0.22. Second series contain-
ing 9-13 segments with a mature reproductive
apparatus; maximal width of this series 0.22-0.28.
Third series comprising 16-19 immature uterine
segments; maximum width 0.098—0.130. This lat-
ter series spilling off from strobila and maturing
independently in host’s intestine. Detached senile
segment oval, 0.67—0.77 long by 0.46-0.53 wide.
Scolex 0.245 long by 0.329-0.350 wide, sharply
demarcated from short neck. Rostellum rudi-
mentaly. Suckers confluent, 0.217-0.231 long by

472 I. SAWADA AND M. HARADA

= =
YT ed

. 7L2D Magee

473

Cestodes of Field Micromammalians

474 I. SAWADA AND M. HARADA

0.140-0.148 wide. Neck short, 0.11-0.18 long by
0.07 wide. Genital pores unilateral, located at
anterior 1/3 of segment margins. Testes two in
number, spherical, 0.056-0.070 by 0.046-0.063,
one on each side of ovary. Ovary subspherical,
0.060-0.070 long by 0.049-0.056 wide. Vitelline
gland compact, 0.053-0.070 by 0.035. Cirrus sac
elongate, surpassing center of segment, 0.161-
0.175 long by 0.028 wide. Cirrus covered with
delicate spines, 0.147—0.154 long. Internal seminal
vesicle, located at inner part of cirrus sac, 0.042-
0.046 long by 0.028-0.032 wide. External seminal
vesicle 0.085-0.095 long by 0.032-0.042 wide.
Vagina opening in genital atrium, extending to
aporal side, then enlarging forming seminal re-
ceptacle measuring 0.032—0.035 long by 0.021-
0.028 wide. Eggs spherical, 0.046-0.049 by 0.039-
0.946, surrounded by four thin envepoles, with
smooth surface. Onchospheres spherical, 0.025-
0.032 by 0.028; embryonic hooks 0.011 long.

Host: Sorex shinto shinto Thomas, 1905.

Site of infection: Small intestine.

Locality and date: Ogurogawa, Ina-shi, Naga-
no Prefecture; October 8, 1988.

Type specimen: Holotype: NSU Lab. Coll.
No. 9007; Paratypes:No. 9008.

Remarks: At present, only one species, di-
aphona (Cholodkovsky, 1906) belonging to the
genus Ditestolepis, of which the descriptions are
incomplete, has been recorded from shrews in
Estonia and Poland [7-12]. The present new
species differs from D. diaphona in many morpho-
logical characters.

Amoebotaenia urotrichi sp. n.
(Figs. 15-22)

Of 14 specimens of Japanese shrew-mole, Urot-
richus talpoides hondonis Thomas, collected at
Ogurogawa, Ina-shi, Nagano Prefecture from

22
nN
2
°
a b c d e
Fic. 22. Camparison of rostellar hooks of five related

species. Sacle in mm.
a; A. fuhrmanni, b; A. oligorchis, c: A. longisaccu-
lus, d: A. spinosa, e: A. urotrichi sp. n.

October 6 to November 15, three were found
infected with a greater number of specimens of the
present new cestodes.

Description: Small-sized dilepidid; strobila
length 0.8-1.0 and maximum width 0.3-0.4, con-
sisting of 8-9 segments. Metamerism distinct,
margins not serrate, segments broader than long.
Scolex square, 0.119-0.133 by 0.112—0.126, pro-
vided with four suckers and a well-diveloped prot-
rusible rostellum. Rostellum 0.070-0.079 long by
0.042-0.063 wide, armed with a single row of 10-
11 somewhat urench-shaped hooks, each measur-
ing 0.014; guard and blade short, about equal in
length, handle solid and longest. Rostellar sac
muscular, 0.112—0.126 long by 0.042—0.070 wide.
Sucker round, 0.119-0.133 by 0.112-0.126. Neck
absent.

Genital pores alternating regularly, located
slightly anterior to middle of segment margin.
Testes roundish, 9-10 in number, 0.021—0.025 by
0.018-0.021, spreading posterodorsal part of seg-
ment between osmoregulatory canals. Cirrus sac
oval. 0.070 long by 0.028-0.035 wide. Cirrus
opening directly in front of vagina into genital

Fics. 8-14. Ditestolepis longicirrosa sp. n.

8: Scolex 9: Immature segments. 10: Mature segment, dorsal view. 11: Mature segment, ventral view; 0: ovary.
v: vitelline gland. t: testis. 12: Detached senile segment. 13: Egg. 14: Outline tracing of mature segment, dorsal

view. Scale in mm.

Fics. 15-21. Amoebotaenia urotrichi sp. n.

15: Scolex. 16: Rostellar hooks. 17: Mature segment. 18: Cirrus sac and vas deferens. 19: Senile segment. 20:
Eggs. 21: Outline trancing of mature segment, dorsal view. Scale in mm.

Cestodes of Field Micromammalians 475

TABLE 1. Species of Amoebotaenia armed with 10-12 mm long rostellar hooks from domestic and wild birds

Rostellar hooks

Cestode species Host
Number Length
Amoebotaenia fuhrmanni Tseng, 1932 [13] 10 0.007 Gallinago sp.
A. oligorchis Yamaguti, 1935 [14] 10 0.030-0.036 Gallus gallus
A. longisacculus Yamaguti, 1956 [15] 12 0.033 Gallus domesticus
A. spinosa Yamaguti, 1956 [15] 10-12 0.033-0.035 Gallus gallus

atrium. Cirrus covered with delicate spines. Vas
deferens strongly coiled at proximal end of cirrus
sac. Ovary transversely elongated and botryoidal,
0.140-0.175 across. Vitelline gland trilobate,
0.098 long by 0.049-0.056 wide, located ventral to
testes at posterior field of segment. Seminal re-
ceptacle transversely elongate, 0.077 by 0.028,
situated dorsal to ovary on poral side. Uterus
occupying entire segment when fully developed.
Eggs oval or spherical, 0.046-0.053 by 0.032;
onchospheres spherical, 0.023-0.028 by 0.021;
embryonal hooks 0.011 long.

Host: Urotrichus talpoides hondnis Thomas,
1908.

Site infection: Small intestine.

Locality and date: Ogurogawa, Ina-shi, Naga-
no Prefecture; October 8, 1988.

Type specimen: Holotype: NSU Lab. Coll.
No. 9009; Paratypes: No. 9010.

Remarks: Out of known 24 species of the
genus Amoebotaenia from domestic and wild
birds, four have 10-12 rostellar hooks (Table 1) [2,
13-15]. Amoebotaenia urotrichi sp. n. differs from
all of them in the length and shape of rostellar
hooks (Fig. 22). This is the first record of the
genus Amoebotaenia from wild animals.

REFERENCES

1 Sawada, I. and Harada, M. (1986) Two new species
of the Vampirolepis (Cestoda: Hymenolepididae)
from Japanese shrews. Jpn. J. Parasitol., 35: 171-
174.

2 Yamaguti, S. (1959) Systema Helminthum. Vol. II
The Cestodes of Vertebrates. Interscience Publ.,
New York, 860 pp.

3 Spasskii, A. A. (1950) A new approach to the

10

11

12

13

14

15

structure and systematics of the hymenolepids (Ces-
toda: Hymenolepididae) (in Russian), Dokl. Akad.
Nauk SSSR, 75: 895-898.

Olsen, O. W. and Kuntz, R. E. (1978) Staphylocy-
stis (Staphylocystis) suncusensis sp. n. (Cestoda:
Hymenolepididae) from the musk shrew, Suncus
murinus (Soricidae), from Taiwan, with a key to the
known species of Staphylocystis Villot, 1877. Proc.
Helminth. Soc. Washing., 45: 182-189.

Schmidt, G. D. (1986) Handbook of Tapeworm
Identification. CRC Press, Florida, 675 pp.
Meggitt, F. J. (1927) On cestodes collected in
Burma. Parasitology, 19: 141-153.

Cholodkovsky, N. (1906) Cestodes nouveaux ou
peu connus. Arch. Parasitol., 10: 332-347.

Stottys, A. (1952) The helminths of common shrew
(Sorex araneus L.) of the National Park of Bialo-
wieza (Poland). (in Polish) Ann. Univ. M. Curie-
Skeod. Sec. C, 6: 165-209.

Zarnowski, E. (1955) Parasitic worms of forest
micromammalians (Rodentia and Insectivora) of the
enviroment of Pulawy (district Lubin). 1. Cesto-
da.(in Polish with English summary). Acta Parasito-
lo. Polo., 3: 279-368.

Rybicka, K. (1959) Tapeworms of forest micro-
mammalians (Rodentia and Insectivora) from Kam-
pinos Wilderness. Acta Parasitol. Polo., 7: 393-422.
Kisielewska, K. (1961) Circulation of tepeworms of
Sorex araneus araneus L. in biocenosis of Biatowieza
National Park. Acta Parasitol. Polon., 9: 331-369.
Vaucher, C. (1971) Les Cestodes parasites des
Soricidae d’Europe Etude anatomique, révision tax-
onomique et biologie. Rev. Swisse Zool., 78: 1-113.
Tseng, Shen. (1932) Studies an avaian cestodes
from China. Part 1. Cestodes from charadriiform
birds. Parasitology, 24: 87-106.

Yamaguti, S. (1935) Studies on the helminth fauna
of Japan. Part 6. Cestodes of birds, 1. Japan. J.
Zool., 6: 183-232.

Yamaguti, S. (1965) Parasitic worms mainly from
Celebes. Part 11. Cestodes of birds. 41 pp., Publ. by
author.

“ol sued a ilkiperiagel plant cy] 01 Ati naps rsatorghin To

aS ee he
a a
/ , _
ae Pe ss -
aC ee ; :
/ _ - Aon
{ - “ ay i

argibearen anime, doar m.aphees) y

ade

aod ilialaad Ze IF
r= Fee e: _ ~~ ug as
Mange “uh an < - if

ib. “Cuthanenallo = “in ee all

itl e thee ‘el SM . D k f Lens th it

ee | 2 ; i= , i
a) , Te ¥ ag. aca a, 0 wt aly ltiw ve ei, sage
. ee ’ apy to Cry Ry tiny ve bealigy a
: 1 alia bhiuy itt Deke betel’ ViSeEsie
iwiaweaaey et 6. Oterlnite Wengen blag?
iis tok — ‘mii? bw Shly WIG
baa 9 =. ineimek ansetr ad Mw blod tae
sh : elt tle VEAL SE om nals yes
i wyok lee Skat Ta ore
_ oe nly Vile tlw ieutitg q

aid 5 dhl ide,
uh AS) Uy ASEM, vee:
' / ; etal ir
Tore eae | Kull ath shyeaabiiat
: Lea)
Z, ; as Deh i dlpa
pres Zamaihnewe 0
‘ ees 2raf Ay
ty, ri geal
hit raat) }
be ‘i dy toprege 2S nnd tee

iia a omy Sth ee “ee
\idgett 1 ipicorn Wht
aif 5 | ‘v WAS

“0.4 olla Vases a iaeyod
yy ayer thy rar iit Va

vr
Chive (recrgh NaN

ZOOLOGICAL SCIENCE 7: 477-484 (1990)

© 1990 Zoological Society of Japan

Crabs of the Genus Calappa from the Ryukyu Islands,
with Description of a New Species

MASATSUNE TAKEDA and NoriKAZU SHIKATANI-

Department of Zoology, National Science Museum, Shinjuku, Tokyo 169, and
"Department of Marine Sciences, University of the Ryukyus,
Nishihara, Okinawa 903-01, Japan

ABSTRACT—Nine species of the genus Calappa (Crustacea, Decapoda, Calappidae) are recorded
from the Ryukyu Islands based on the collections of the University of the Ryukyus. One of them is
described as a new species under the name of C. quadrimaculata, being readily distinguished from the
closest congener, C. /ophos (Herbst), by having no striped markings on the carapace and chelipeds, and
also by the different proportion and armature of the carapace. The new species is also known from

Taiwan.

INTRODUCTION

The crabs of the genus Calappa (Family Calap-
pidae) living in shallow-water of the Indo-Pacific
and Atlantic Oceans are called the box crabs due
to having the thin clypeiform expansion at each
posterolateral side of the carapace, and well
known by their peculiar habit of breaking the shell
by the right chela to eat its soft part or hermit crab
living in the empty shell [1, 2].

During the extensive survey of the shallow-
water crab fauna of Nakagusuku Bay in southeast-
ern Okinawa-jima Island, the Ryukyu Islands, we
encountered five specimens referable to the spe-
cies close to, but different from C. Jophos (Herbst)
which is one of the commonest Calappa species in
Japanese waters. On a detailed comparative ex-
amination, they were proved to represent a new
species which will be described in the present
paper under the name of C. quadrimaculata,
together with records of the known species from
the Ryukyu Islands based on the collections of the
Department of Marine Sciences, the University of
the Ryukyus. During the recent field survey in
Taiwan, the senior author found three specimens

Accepted August 9, 1989
Received June 1, 1989

' Present address: Ocean Research Institute, University
of Tokyo, Nakano, Tokyo 164, Japan.

of Calappa without doubt referable to the new
species at the fish market together with C. lophos
(Herbst) and C. philargius (Linnaeus), both of
which are very common.

The Calappa species attract not only some
biologists, but also certain collectors and aqualists,
due to the big size and the beautiful coloration
with spots and bands in addition to the peculiar
shape and ecology. It is generally considered that
the census has been made on rather thorough
investigations, and thus the present discovery of a
new species is remarkable and worth noting.

The bulk of the specimens examined is pre-
served in the University of the Ryukyus (URM)
except for the holotype and one of the paratypes of
the new species and a duplicate specimen of each
species, which are deposited in the National Sci-
ence Museum, Tokyo (NSMT). In the mesasure-
ments of each species, the breadth and length of
the carapace are abbreviated to cb and cl,
respectively, with the greatest breadth including
the clypeiform expansions of both sides.

SYSTEMATIC ACCOUNT

Family Calappidae
Genus Calappa Weber, 1795
Calappa bicornis Miers, 1884

OKINAWA. Nakagusuku Bay, 15-20m

478 M. TAKEDA AND N. SHIKATANI

deep.—1? (URM-CR 0075; cb 62.5 mm), 9-IX-
1985; 12 (URM-CR 0081; cb 57.8mm), 1?
(UR M-CR 0082; cb 71.8 mm), 30-VII-1986; 14
(URM-CR 1103, NSMT-Cr 9620; cb 81.7 mm),
1° (URM-CR 1104; cb 65.5 mm), 10-VI-1987.

Remarks. This species is readily distinguished
from Calappa gallus (Herbst) by having a tubercu-
lar tooth immediately behind the external orbital
angle. C. woodmasoni Alcock based on a young
specimen from off south coast of Sri Lanka was
decidedly synonymized with this species by Rath-
bun [3]. C. woodmasoni was, however, resur-
rected by Ihle [4] who recorded a young female
from Manipa Island in the Malay Archipelago. It
is well known that the contour of the carapace is
remarkably variable during development in the
genus Calappa, and that the carapace is generally
narrower and more quadrangular in the young.
Based on the difference in the second peduncular
segment of antenna, Ihle [4] distinguished both
species, but this character is probably variable just
like in the other groups of crabs, e.g., some genera
of the family Xanthidae, in which the orbit
completely closed by the well developed antennal
peduncle is considered as one of the generic
criteria, but the orbit is still incomplete, with a
wide hiatus, in the young.

This species is known from the Providence Is-
lands (type locality) and the Seychelles in the
western Indian Ocean, and from Tosa Bay and
several localities around the Kii Peninsula in cen-
tral Japan. If C. woodmasoni is synonymized with
this species, the records from Sri Lanka and Ma-
nipa Island will become the localities intervening
between the western Indian Ocean and Japan.

Calappa calappa (Linnaeus, 1758)

OKINAWA. Nakagusuku Bay.—1¢ (URM-
CR 0760; cb 120.2 mm), 12 (URM-CR 0761; cb
125.0mm), 15-X-1985; 18 (URM-CR_ 0759,
NSMT-Cr 9621; cb 117.8mm), 1 (URM-CR
0757; cb 131.0 mm), 12 (URM-CR 0758; cb 141.2
mm), 4-V-1984.

Remarks. This species is uniformly yellowish
brown or sometimes mottled with many purplish
blotches on the carapace, being characterized by
the unarmed clypeiform expansion at each side.

This species is widely distributed in the Indo-
West Pacific from Sagami Bay in Japan and the
Hawaiian Islands through New Caledonia and the
Malay Archipelago to the east coast of Africa.

Calappa capellonis Laurie, 1906

OKINAWA. Nakagusuku Bay, 15-20m
deep.—1 (URM-CR 1105, NSMT-Cr 9622; cb
57.8 mm), 10-VI-1987.

Remarks. This species was originally described
as the variety of Calappa gallus (Herbst) by Laurie
[5], but the differences in armature of the cara-
pacial dorsal surface and clypeiform expansions
enumerated by Sakai [6, 7] and Takeda and
Koyama [8] warrant its specific distinction from C.
gallus.

This species is known only from Sri Lanka (type
locality), and the Kii Peninsula, Kagoshima and
Okinawa in Japan.

Calappa gallus (Herbst, 1803)

OKINAWA. Nakagusuku Bay.—19 (URM-
CR 0083, NSMT-Cr 9623; cb 46.0mm), 10-III-
1986. Zanpa-misaki.—1° (URM-CR 0080; cb
42.9mm), VIII-1984. Yakata-katabaru.—1¢
(URM-CR 1429; cb 33.0 mm), 1987.

Remarks. This species is well figured by Klun-
zinger [9], Sakai [6, 7, 10], Rathbun [11], Barnard
[12] and Monod [13].

This species is widely distributed in the whole
Indo-—West Pacific from Japan to the Red Sea and
South Africa, the tropical Atlantic coast of Africa,
and the western Atlantic from the Florida Keys to
Bahia, Brazil. Such distribution pattern is quite
unusual in the shallow-water crabs, so that the
geographic speciation in these respective areas is
to be confirmed with current knowledge of iden-
tification.

Calappa hepatica (Linnaeus, 1758)

OKINAWA. Manza beach.—1% (URM-CR
0076; cb 69.0 mm), 31-V-1985.

IRIOMOTE. Amitori Bay.—1 young ¢, 1?
(URM-CR 0077; cb 30.3 and 55.3 mm), 1 young $
(URM-CR 0079; cb 32.5 mm), 16-VIII-1985.

Calappa from the Ryukyu Islands 479

Remarks. This is the most commonest species
in the genus Calappa and well figured by Sakai [6,
7], being widely distributed in the whole Indo-
West Pacific.

Calappa yamasitae Sakai from Japan described
in 1980 [14] is the closest congener of this species,
but according to the original description, distin-
guished by the following features: 1) Low pro-
tuberances of good size tipped each with a small
tubercle are on the anterior two thirds of the
carapacial dorsal surface, and the posterior third is
tuberculate and granulated. 2) The front consists
of two median obtuse teeth separated medially by
an U-shaped sinus, each tooth bearing a subdistal
tooth on its outer border. 3) The hepatic margin is
gently turned into the clypeiform expansion of the
branchial region without distinct constriction. 4)
In both sexes the terminal abdominal tergum is
broadly triangular in outline, not narrowed dis-
tally.

Calappa lophos (Herbst, 1782)

OKINAWA. Nakagusuku Bay.—1¢ (URM-
CR 0754, NSMT-Cr 9624; cb 123.7 mm,. cl 79.4
mm), 3-XII-1984; 12 (URM-CR 0073; cb 43.5
mm, cl 31.3mm), 14-VJ-1985; 12 (URM-CR
0071; cb 60.0mm, cl 42.4mm), 3-VII-1985; 12
(URM-CR 0072; cb 61.2 mm, cl 44.8 mm), 15-IV-
1986; 12 (URM-CR 0070; cb 61.0mm, cl 43.0
mm), 6-V-1986.

Remarks. This species is characteristic in its
color pattern in the adult which is distinct even in
spirit, being figured by de Haan [15], Sakai [6, 7,
10], Stephensen [16] and Barnard [12]. As men-
tioned by Alcock [17] and figured by Sakai [6], in
the young the carapace is traversed by dark-
colored longitudinal lines and marked with a pair
of large ocelli in its posterior third.

This species is rather common in the sandy
bottom, ranging from Japan through Sulawesi,
India and the Persian Gulf to the east coast of
Africa.

Calappa philargius (Linnaeus, 1758)

OKINAWA. Nakagusuku Bay, 15-20m
deep.—1% (URM-CR 0749, NSMT-Cr 9625; cb

112.0 mm), 12-XII-1984; 1$ (URM-CR 0069; cb
58.8mm), 13-V-1985; 18 (URM-CR 0068; cb
52.5mm), 24-IX-1985; 1$ (URM-CR 1161; cb
59.0 mm), 10-VI-1987.

Remarks. This species is characteristic in hav-
ing a chocolate-brown band surrounding the orbit
at each side, with a large spot each on the outer
surface of the chelipedal carpus and palm. In its
general shape it is close to C. Jophos (Herbst), but
the margin of the clypeiform expansions of both
sides and the posterior border of the carapace are
armed with much sharper teeth, as figured by de
Haan [15], Shen [18], Sakai [6, 7, 10] and Guinot
[19]. In C. dumortieri Guinot [19] from the Red
Sea these teeth are futher salient and rather
tuberculated.

The geographical distribution is wide in the
Indo—West Pacific from Japan through the Malay
Archipelago and the Andaman Sea, Western
Australia and the Persian Gulf to the Red Sea.

Calappa quadrimaculata sp. nov.
(Figs. 1-4)

OKINAWA. Nakagusuku Bay.—1¢, para-
type (URM-CR 0751; cb 67.5 mm, cl 41.3 mm),
17-XII-1984; 13, paratype (URM-CR 0752; cb
76.3 mm, cl 48.0mm), 1%, holotype (URM-CR
0753, NSMT-Cr 9626; cb 76.6mm, cl 47.4mm),
11-XII-1984; 18, paratype (URM-CR 0074; cb
72.2 mm, cl 44.8 mm), 01-XI-1985; 1¢, paratype
(URM-CR 0084, NSMT-Cr 9627; cb 70.4 mm, cl
43.5 mm), 25-XII-1985.

TAIWAN. Tong-Kang, Ping-Tong County.—
36 , paratypes (NSMT-Cr 9628; cb 72.0 mm, cl
45.8 mm-—cb 73.6 mm, cl 46.4 mm-—cb 78.0 mm, cl
48.2 mm), 15-VII-1989.

Description. Typical of Calappa, with well de-
veloped clypeiform expansion at each side. Cara-
pace strongly convex fore and aft, especially for its
posterior part; its dorsal surface shining, but un-
even with a pair of submedian deep furrow border-
ing mesogastric, cardiac and intestinal regions and
with several linear shallow furrows on each bran-
chial region; protogastric regions of both sides with
2 transverse rows of 4 blunt protuberances, each
hepatic region with 2 protuberances, mesogastric
region with 1, and each branchial region with 1 ina

480 M. TAKEDA AND N. SHIKATANI

line with mesogastric protuberance and 2 behind
hepatic protuberances; frontorbital region in front
of hepatic and protogastric protuberances thickly
covered with microscopical vesicular granules;
posterior surface on and around intestinal region

sparsely covered with frosted minute granules
along posterior margin of carapace. Front deeply
cleft in a shape of V; lower and upper edges of
lateral margin obtusely angulated. Supraorbital
margin rather strongly raised, with 2 closed

Fic. 1. Calappa quadrimaculata sp. nov., $ , holotype (URM-CR 0753, NSMT-Cr 9626; cb 76.6 mm).

Calappa from the Ryukyu Islands 481

fissured on its outer half; its inner edge produced
as a small tubercle separated from upper angle of
front. Anterolateral margin of carapace gently
convex, with 12 or 13 lobate teeth which are close
together; first 4 or 5 teeth each with some minute
granules of same size along margins, but posterior
teeth except for the last with a median main
granule and 2 accessory granules each on anterior
and posterior slopes; clypeiform expansion well
developed, with 4 strong teeth; first 2 obtusely
angulated, and last 2 sharply pointed, end at same
level; posterior margin of last tooth more or less
serrulated with several granules, forming first lobe

of posterior margin of carapace; second lobe also
serrulated along its whole margin, as wide as first
lobe, obtusely angulated near lateral end; third
lobe triangular, obtusely angulated at its apex,
about half as wide as second lobe; median lobe
weakly convex behind intestinal region along its
central 1/3, with a triangular lobe at each side;
apex of this lateral lobe obtuse, exceeding the level
of median lobe and also that of third lobe.

Distal margin of chelipedal merus cut into 4
lobes, fringed with long hairs; median 2 lobes
about 1/2 as wide as proximal and distal lobes;
distal lobe sharply pointed distally, and subdistal

Fic. 2.

Calappa quadrimaculata sp. nov., $, paratype (URM-CR 0074; cb 72.2 mm).

482

lobe with a spine at its median part; carpus and
palm smooth and shining; upper margin of palm
cut into 9 teeth, the first and the last of which are
with pale brownish fringe.

Etymology. The species name, quadrimacula-
ta, is referred to four ocelli of the carapace, which
are somewhat variable in size, but always distinct.

Remarks. At a glance the new species is readi-
ly distinguished from the known species by the
different color pattern. The basic color patterns of

Ye, 14 QO a
NTA)

(ie)
Y
H))

Ks
ai
i

Fic. 3. Calappa quadrimaculata sp. nov., third maxil-
liped (A) and first pleoped (B) of ¢, paratype
(URM-CR 0074). Scales=5 mm for A, 1 mm for B.

M. TAKEDA AND N. SHIKATANI

the carapace and chelipeds are individually con-
stant in the Calappa species and kept so long even
in spirit, being one of the effective clues to distin-
guish the species.

The new species is without doubt most close to
Calappa lophos (Herbst) in the general formation
of the carapace and chelipeds, but distinguished
from it by the proportional difference of the cara-
pace and the morphological difference of the pos-
terior lobes of the carapace. The carapace of the
new species is seemingly, but apparently, wider
than that of C. lophos.

This proportional difference is indicated with
the measurements given to both species; in five
specimens of C. lophos examined, the mean ratio
of the carapace breadth to length is 1.43, while in
eight specimens of the new species the ratio varies
from 1.57 to 1.63 (mean 1.61). In addition, it is
remarkable that in the new species the second
posterior lobe of the carapace is almost equal to
the first lobe in its width, but in C. lophos the
second posterior lobe is at most 2/3 as wide as the
first lobe.

Alcock [17] doubtfully synonymized Calappa
guerini de Brito Capello with C. lophos. Accord-
ing to its original description and figure [20], it
differs from C. lophos by having the sharply
toothed innermost pair of the posterior lobes of

Fic. 4.

Calappa quatrimaculata sp. nov., first pleopod of %, paratype (URM-CR 0074). A, distal fourth; B, distal

part, further enlarged; C, one of tubercles dispersed on shaft; D, one of sensory hairs arranged in a line along

seam.

TABLE 1.

Species

Calappa from the Ryukyu Islands

Japanese species of the genus Calappa

Distribution

483

Foreign loc.

. bicornis Miers, 1884

. calappa (Linnaeus, 1758)
. capellonis Laurie, 1906

. gallus (Herbst, 1785)

. hepatica (Linnaeus, 1758)
. japonica Ortman, 1892

Kii Penin. & Tosa Bay
Sagami Bay to Ryukyus
Kii Penin. to Ryukyus
Sagami Bay to Ryukyus
Sagami Bay to Ryukyus
Sagami Bay to Kyushu

W. Indian Ocean
Indo-W. Pacific
Sri Lanka
Cosmopolitan
Indo-W. Pacific
Indian Ocean

**C. lophos (Herbst, 1782)
**C. philargius (Linnaeus, 1758)
C. pustulosa Alcock, 1896
**C. quadrimaculata sp. nov.
**C. terraereginae Ward, 1936
C. yamasitae Sakai, 1980

Okinawa

Tokyo Bay to Kyushu
Tokyo Bay to Kyushu

Korean Channel
Kii Penin.

Indo-W. Pacific
Indo-W. Pacific

Sagami Bay to Tosa Bay India

Taiwan
Australia

Four species with an asterisk have hitherto been known not only from the Japanese mainland, but also from the
Ryukyu Islands. Five species with two asterisks including a new species were newly added to the

carcinological fauna of the Ryukyu Islands.

the carapace. There is no subsequent record of the
Species or discussion on its identity, and thus it is
not always sure at present whether Alcock’s syn-
onimization is justified or not. However, at least,
C. guerini is very close to and nearly identical with
C. lophos, and the new species is separated from
this doubtful species also by the different contour
and armature of the carapace.

Calappa terraereginae Ward, 1936

OKINAWA. Nakagusuku Bay, 15-20m
deep.—1$ (URM-CR 0088, NSMT-Cr 0629; cb
50.3 mm), 21-VI-1985; 18 (URM-CR 1162; cb
53.0 mm), 10-VI-1987.

Remarks. This species is only known by Ward
[21], Sakai [6, 7] and Tyndale-Biscoe and George
[22] from Lindeman Island off Queensland and
Western Australia, and from off Cheju Island in
the Korean Channel. The general formation of the
carapace much resembles that of C. lophos
(Herbst), but the carapace is slightly narrower,
with more strongly arched anterolateral borders of
the carapace, the teeth of the clypeiform expan-
sion are rather triangular in dorsal view and not so
sharp as in C. lophos, and the posterior border of
the carapace is pronouncedly produced beyond the
posterior border of the clypeiform expansion.

GEOGRAPHICAL NOTES

The genus Calappa is composed of 1 cosmopoli-
tan, 15 Indo—West Pacific, 3 East Atlantic and 9
West Atlantic species. As enumerated in Table 1,
the species known from Japanese waters are 12
including the new species described in the present
paper. Four of them have hitherto been recorded
not only from the Japanese mainland, but also
from the Ryukyu Islands. In the present paper 4
known species were newly added to the carcinolo-
gical fauna of the Ryukyu Islands.

Both of 2 species unrecorded from the Ryukyu
Islans, Calappa japonica and C. pustulosa, are
known from the Japanese mainland and Indian
Ocean without the intervening localities. These
two sepcies are the deeper-water inhabitants than
most of the other species, ranging bathymetrically
from ca. 50 to 200 m. Therefore it may be possible
to conclude that the absence of these two species
from the Ryukyu Islands and the Southeast Asia is
not due to the topographical condition, but to the
insufficient operation of collecting the samples at
the continental shelf.

Calappa bicornis, C. capeilonis and C. terraere-
ginae are also known only from Japan and the
distant localities, viz., the western Indian Ocean,
Sri Lanka and Australia, respectively, but it is
reasonable that in due time they will be recorded

484

from the intervening localities. C. gallus is, as
noted in the text, peculiar in its worldwide dis-
tribution.

ACKNOWLEDGMENTS

We wish to tender our cordial thanks to Dr. Shigemi-
tsu Shokita, Associate Professor of the Department of
Marine Sciences, University of the Ryukyus, who made
the specimens collected by the students available to us
for systematic study and gave the intensive guidance to
the junior author. Thanks are also due to the Chinen and
Tozoe Fishermen’s Unions for providing us with the
facilities to collect the specimens from gill-nets. Dr.
Hsiang-Ping Yu, Professor of National Taiwan Universi-
ty of Marine Sciences, kindly arranged the field survey
for the senior author and gave him the information about
the crabs from Taiwan.

REFERENCES

1 Shoup, J. E. (1958) Shell opening by crabs of the
genus Calappa. Science, 160: 887-888.

2 Takeda. M. and Suga, H. (1979) Feeding habits of
box crabs, Calappa. Res. Crust., 9: 43-46, pl. 1.

3 Rathbun, M. J. (1911) Marine Brachyura. The
Percy Sladen Trust Expedition to the Indian Ocean
in 1905. 3(11). Trans. Linn. Soc. London, (2), 14:
191-261, pls. 15-20.

4 thle, J. E. E. (1918) Die Decapoda Brachyura der
Siboga-Expedition. III. Oxystomata: Calappidae,
Leucosiidae, Raninidae. Siboga-Exped., 39b: 155-
322.

5 Laurie, R. D. (1906) Report on the Brachyura
collected by Professor Herdman, at Ceylon, in 1902.
Ceylon Pearl Oyster Fish. Rep., 5 (Suppl.): 349-
432, pls. 1, 2.

6 Sakai, T. (1937) Studies on the crabs of Japan. II.
Oxystomata. Sci. Rep. Tokyo Bunrika Daigaku,
(B), 3 (Suppl.): 67-192, pls. 10-19.

7 Sakai, T. (1976) Crabs of Japan and the Adjacent
Seas. Kodansha Ltd., Tokyo, pp. xxix+773 (En-
glish); pp. 461 (Japanese); pp. 16+pls. 251 (plates).

8 Takeda, M. and Koyama, Y. (1974) On some rare
crabs from Kii Province. Res. Crust., 6: 103-121.

9 Klunzinger, C. B. (1906) Die Spitz- und Spitzmund-
krabben (Oxyrhyncha und Oxystomata) des Roten
Meeres. Stuttgart, pp. vii+88, pls. 2.

10

11

12

13

14

15

18

19

20

21

i)
i)

M. TAKEDA AND N. SHIKATANI

Sakai, T. (1965) The Crabs of Sagami Bay collected
by His Majesty the Emperor of Japan. Maruzen
Co., Ltd., Tokyo, pp. xvi+206 (English part)+92
(Japanese part) +32 (bibliography and index), map
1, pls. 100.

Rathbun, M. J. (1937) The oxystomatous and allied
crabs of America. Bull., U. S. Natn. Mus., 166: i-vi,
1-278, pls. 1-86.

Barnard, K. H. (1950) Descriptive catalogue of
South African decapod Crustacea. Ann. S. Afr.
Mus., 38: 1-837.

Monod, T. (1956) Hippidea et Brachyura ouest-
africains. Mém. I.F.A.N., 45: 1-674.

Sakai, T. (1980) New species of crabs of the families
Lithodidae and Calappidae. Res. Crust., 10: 1-11,
pl. 1, frontispiece.

Haan, W. de (1833-1849) Crustacea. In: Von
Siebold, Fauna Japonica sive descriptio animalium,
quae in itinere per Japoniam, jussu et auspiciis
superiorum, qui summum in India Batava Imperium
tenent, suscepto, annis 1823-1830 collegit, notis,
observationibus et abumbrationibus illustravit. Pp.
XVii+xXxxi+244, pls. 55+ A-Q+2.

Stephensen, K. (1945) The Brachyura of the Iranian
Gulf. Danish Sci. Invest. Iran, 4: 57-237.

Alcock, A. (1896) Materials for a carcinological
fauna of India. No. 2. The Brachyura Oxystomata.
J. Asiat. Soc. Bengal, 65: 134-296, pls. 6-8.

Shen, C. J. (1931) The crabs of Hong Kong. Part II.
Hong Kong Nat., 2: 185-197, pls. 12-14.

Guinot, D. (1964) Sur une collection de crustacés
dédapodes brachyoures de Mer Rouge et de Soma-
lie. Remarques sur les genres Calappa Weber,
Menaethiops Alcock, Tyche Bell, Ophthalmias
Rathbun et Stilbognathus von Martens. Boll. Mus.
Civ. Venezia, 15: 7-63, pls. 1-4.

de Brito Capello, F. (1870) Algumas especies novas
ou pouco conhecidas de crustaceos pertencentes aos
generos <Calappa> e <Telphusa>. J. Sci. Math.
Phys. Nat., Lisboa, 3: 128-134, pl. 2.

Ward, M. (1936) Crustacea Brachyura from the
coasts of Queensland. Mem. Qld. Mus., 11: 1-13,
pls. 1-3.

Tyndale-Biscoe, M. and George, R. W. (1962) The
Oxystomata and Gymnopleura (Crustacea,
Brachyura) of Western Australia with descriptions
of two new species from Western Australia and one
from India. J. Roy. Soc. W. Aust., 20: 45-96.

ZOOLOGICAL SCIENCE 7: 485-515 (1990)

Description and Complete Larval Development of a New Species
of Baccalaureus (Crustacea: Ascothoracida) Parasitic in
a Zoanthid from Tanabe Bay, Honshu, Japan

1

Tatsunori ITO’ and Mark J. GRrYGIER~

' Seto Marine Biological Laboratory, Faculty of Science, Kyoto University, Shirahama,
Wakayama 649-22, and * Sesoko Marine Science Center, University
of the Ryukyus, Sesoko, Okinawa 905-02, Japan

ABSTRACT— An unidentified species of Zoanthus from Tanabe Bay, Honshu, Japan, is the host of an
endoparasitic ascothoracidan crustacean, Baccalaureus falsiramus, new species. This is the first record
of this zoanthid genus serving as the host of an ascothoracidan and the second species of Baccalaureus
from Japan. The morphology of the adult females, nauplii, and ascothoracid larva is described based
upon a detailed study combining light microscopy and SEM. The female of this new species is
characterized by a coiled carapace but very short, more or less distally upturned thoracic horns, and very
long, ventrally directed papillae for seminal receptacle ducts lateral to thoracopods II-IV. Much
variability is recognized in the antennule, thoracopods, penis, and abdominal ornamentation. Larval
specimens were individually reared in the laboratory. Six lecithotrophic naupliar instars with
rudimentary endites on the antennae and mandibles are present before the ascothoracid larva. The
nauplii swam for about one month without feeding until the metamorphosis to the ascothoracid larva.
Naupliar instars II-VI have a sculpture of concentric cuticular ridges on the marginal area of the dorsal
shield. A nauplius eye is present through all naupliar instars as well as in the ascothoracid larva. Setae
are gradually added to the antennules, but the antennae and mandibles remain essentially unchanged
after instar III; rudimentary maxillules appear in instar II. The ascothoracid larva is a “Tessmann’s
larva” similar to one recently described from Hawaiian plankton, but lacking central pores within the
carapace reticulations. Morphological and developmental features of the nauplii and ascothoracid larva

© 1990 Zoological Society of Japan

are discussed.

INTRODUCTION

One of us (T. I.) has been conducting an exten-
sive parasitological survey in Tanabe Bay, Japan,
to discover the adults of Facetotecta Grygier
(Crustacea, Maxillopoda), which are currently
known only from so-called y-larvae [1]. An un-
identified Zoanthus (Hexacorallia) examined in
this survey was infested by a previously unknown
species of Baccalaureus Broch (Crustacea,
Ascothoracida, Lauridae), which is described in
this paper.

The present paper also reports the first success-
ful study of a generalized larval history in an
ascothoracidan, based upon larvae of the new

Accepted August 7, 1989
Received July 14, 1989

species individually reared in the laboratory. The
only other complete report is that of Brattstrom [2]
on Ulophysema oeresundense Brattstr6m, 1936, a
species with abbreviated development. Wagin [3]
and Karande and Oguro [4] gave more or less
complete accounts of the larvae of Ascothorax
ophioctenis Djakonov, 1914, and Dendrogaster (=
Myriocladus) astropectinis Yosii, 1931, respective-
ly, but in neither case were naupliar instars clearly
determined, just arbitrary stages.

MATERIAL AND METHODS

Five adult females and one possible male of the
new species were collected from a single colony of
Zoanthus sp. (Japanese name: mame-sunagincha-
ku) that was found on a rocky cliff on the north
side of Toshima Rock in Tanabe Bay (33°41'N,

486 T. It6 AND M. J. GryGIER

135°21’E), at a depth of 4 m. Parts of this zoanthid
colony that yielded ascothoracidans were sampled
on several occasions by one of us (T. I.) using
SCUBA. Three females that were examined by
light microscopy have been designated as the type
series (data given later). The other two females
and the possible male were examined by SEM.
One of the females (SEM-1; collected 22-I-1989),
which was accompanied by the possible male, was
fixed with 10% Formalin-sea water solution; the
other female (SEM-2; collected 5-II-1989) was
pre-fixed with 2% glutaraldehyde (phosphate-
buffered, with sucrose to adjust osmotic pressure)
for 4hr, treated with a mixture of 2% tannic acid
and 2% guanidine hydrochloride solution for 8 hr,
and then post-fixed with 2% osmic acid for 8 hr.
An instar I nauplius collected from the brood
chamber of SEM-2 was similarly treated for SEM
study. Four instar II nauplii obtained from a
paratype and two ascothoracid larvae metamorph-
osed in the laboratory were fixed with Formalin-
sea water, and were also used for SEM study.
After fixation, all the specimens used for SEM
study were dehydrated through a graded series of
ethanol, transferred into isoamyl acetate, and de-
ssicated in a critical point dryer using CO. Dried
specimens were sputter-coated with gold, and ex-
amined in a scanning electon microscope (JEOL,
JSM T-220) at accelerating voltages of 5 to 15 kv.

Several instar I nauplii from the holotype and
instar II nauplii from a paratype were dissected in
glycerine for examination of appendages. Three
nauplii obtained from the same paratype were
reared in the laboratory. Initially, as shown later
by examination of their exuviae, two were second
instar and one was third. They were kept in small
dishes individually at 19°C, with daily changes
(twice a day) of sterilized dishes containing fresh,
paper-filtered sea water until the metamorphosis
to the ascothoracid larva. Exuviae were removed
from the dishes when present, fixed in Formalin-
sea water, and mounted on glass slides in glycerine
for light microscopical examination. One of the
resulting ascothoracid larvae was dissected and
mounted first in glycerine, then in glycerine jelly
for microscopical examination; the other two were
prepared for SEM study as described above.

Due to the extensively modified body and

appendages in Baccalaureus, there has historically
been considerable disagreement about the number
of thoracic segments and the identity of most of the
cephalic and anterior thoracic appendages and
bodily projections. The morphological terminolo-
gy adopted here is that of Brattstr6m [5] as mod-
ified by Gryier [6].

Part 1. Taxonomy
Baccalaureus falsiramus sp. nov.

Diagnosis. Adult female: Baccalaureus round
in side view, with coiled lateral carapace lobes
making almost two full revolutions, and with
spines on edges of coils. Anterior thoracic horns
shorter than thorax, naked, either distally up-
turned or almost straight. All three pairs of
mouthparts well developed. Small plate-like organ
at base of first thoracopod, or just a swelling
instead, barely or not extending dorsally over
lateral chitinous ridge of thorax. Thoracopod 1
variform, represented by a papilla or cylindrical
process, with 0—2 apical setae. Thoracopods 2—4
containing seminal receptacles, each limb flanked
laterally by prominent, ventrally directed, conical
papilla with apical opening. Thoracopod 5 vari-
form. Thoracopod 6 variform, or absent. Dorsal
setae only on last thoracomere and first abdominal
segment. Penis uniramous, variform, often with
distal spines. Furcal rami as long as first two
abdominal segments combined, narrow, with 2-3
hirsute terminal setae, no medial setae or sensilla,
lateral side partly bare of cuticular ctenae.

Nauplii: lecithotrophic, six instars, with bowl-
shaped dorsal shield after instar I, four-segmented
antennules at instar VI, vestigial protopodal en-
dites on antennae and mandibles, caudal arma-
ment barely protruding beyond end of dorsal
shield. Ascothoracid larva: a “Tessmann’s larva”,
carapace valves without central pores in polygonal
cells delineated by chitinous, mesh-like ridges.

Type series. Holotype: adult female, fully dis-
sected, brooding eggs and nauplii, trunk and cara-
pace lobes preserved in ethanol, dissected appen-
dages and part of carapace mounted onto slide
glasses with glycerine jelly. Paratype-1: adult
female, brooding eggs, carapace partly torn, speci-
men otherwise intact, preserved in ethanol. The

New Species of Baccalaureus 487

holotype and paratype-1 were recovered by M. J. being kept in an aquarium (Original collection
G. from host material that had been fixed (10-IX- date 26-VII-1988). Paratype-2: adult female (fully
1988) and preserved in 70% ehtanol by T. I. after dissected), brooding eggs and nauplii, recovered

100um , A

Fic. 1. B. falsiramus sp. nov. A, lateral view of holotype (carapace removed; AI, first abdominal segment; g, gut
diverculum; mg, maxillary gland; mx2, maxilla). B-D, paratype-1. B, dorsal view of carapace; C, lateral view of
carapace; D, lateral view of habitus with partly torn carapace.

488 T. IT6 AND M. J. GryGIER

by T. I. from living host material (26-VII-1988),
cephalic area and much of carapace missing, re-
mainder of carapace removed from trunk, pre-
served in ethanol, dissected appendages mounted
onto slide glass with glycerine jelly. Type locality:
Tanabe Bay, Honshu, Japan. The type series is
deposited in the Seto Marine Biological Labora-
tory, Kyoto University.

Etymology. The specific name (from Latin
“falsus” =false plus Latin “ramus” =branch) re-
fers to the extraordinarily large papillae at the
bases of thoracopods 2-4.

1-1. DESCRIPTION OF HOLOTYPE

Carapace with small, medial body chamber con-
nected to pair of large, coiled, lateral lobes serving
as brood sacs (see Fig. 1B-C of paratype-1), tinc-
tured with reddish brown along aperture lips,
other parts with very faint tinge of yellowish
brown. Outer coils nearly circular in side view, 6.0
mm high, 5.1mm long, subsequent coiling of
diminishing radius, up to nearly two full revolu-
tions altogether (+675). Exposed lateral faces of
coils with sparse, simple papillae, edges with wide-
ly spaced spines sometimes exceeding 0.15 mm
long, outer part of hidden medial faces with similar
but smaller tubercles (Fig. 3C). Gut diverticula
and ovaries readily visible through carapace wall,
with radially arranged side branches from main
central coil, side branches dividing once or twice.
Body chamber protruding beyond outer coil post-
eriorly, with vertical posterior aperture; aperture
lips bearing adherent exuviae of four earlier instars
(partially shown in Fig.3A). Lips with short
marginal spines, inner sculpturing of cuticular
ridges forming hexagons, and dense interior lining
of fine cuticular hairs (Fig. 3B). Ventral margins
of lateral sides of body chamber adjoining but
readily separable as far forward as oral cone; no
special armament.

Body attitude as shown in Figure 1A; almost
colorless but scattered spots of faint purple. Dis-
tance from tips of horns to tips of furcal rami 3.7
mm. Body cuticle loose, animal preparing to molt.
Head bearing oral cone, posteriorly directed
antennules, and pair of lobes representing maxil-
lary glands. Thorax six-segmented, boundary of
first and second segments not expressed externally,

that of second and third segments weakly express-
ed. Pair of anterior horns arising from first seg-
ment, parallel, shorter than thorax, laterally
flattened, tapering to rounded, upturned tips,
naked, semitransparent. Dorsoproximal part of
each horn markedly swollen laterally in connection
with anterior part of lateral chitinous ridge, latter
present from basal part of horn to middle of sixth
thoracomere, consisting of at least four or five
separate, segmentally arranged thickenings and
forming outer edge of trough along side of body,
highest point near supposed boundary of first two
thoracomeres. Transverse band of short rows of
hairs on rear of sixth thoracomere and first abdo-
minal segment. Five pairs of thoracopods present.
Abdomen four-segmented, each segment narrow-
er and less high than preceding one, ornamented
with many small pores among sparse, delicate
spinules (Fig. 3D). First abdominal segment with
ventral penis.

Antennules (Fig. 4A, C) strap-like, about 0.9
mm long, curved or sharply bent, indistinctly di-
vided into about four segments and apical process,
hairy all over surface, transverse rows of longer
hairs along anterior (ventral) face of proximal half.
Terminal segment with two short, subapical setae,
one medial and one anterior; medial seta
apparently folded or bipartite (Fig. 4B, D). Apical
process small, with trifurcate aesthetasc and short
seta.

Labrum (Fig. 1A) deeper than long, short post-
erior edges adjoining behind other mouthparts but
readily separable.

Mandible (Fig. 4E) narrow, distal half tapered
with many short, basally directed spinules on pro-
ximal two-thirds of anteromedial margin, four
arched rows of spinules posterolaterally.

Distal part of maxillule (Fig. 4F) represented by
conical process with hairs of different lengths.

Maxillae (Fig. 4G) largely fused, their tips prot-
ruding through distal labral aperture; tips sepa-
rate, bifid, each accompanied subterminally by
basally bent, triangular, lateral plate.

Thoracopod 1 represented by a small papillary
process arising from basal swelling, with no seta.
Right one (Fig. 3E) shorter and wider than left one
(Fig. 3F). Basal swelling hemispherical, somewhat
inflated but not extending as “plate-like organ”

New Species of Baccalaureus 489

Be Piss has tee 3s e

Fic. 2. SEM photomicrographs of B. falsiramus sp. nov. C-D SEM-2, otherwise SEM-1. A, habitus, lateral; B,
internal side of carapace near aperture lip; C, habitus, lateral; D, penis and first abdominal segment; E, apical
portion of furcal rami and first abdominal segment with hairs; F, furcal rami, lateral. Scales: A, C 500 um; B, E

100 um; D, F 50 um.

490 T. Iv6 AND M. J. GryGier

A

50um

100um

Fic. 3.

B,E,F,H

EE
BE
ZZ

B. falsiramus sp. nov. A-C, paratype 1. A, internal view of aperture lip, showing adherent exuvia; B, detail

of marginal part of A; C, carapace spination, 3 large spines (below) on edge of coil, smaller ones (above) on inner
face below edge. D-F, holotype. D, lateral view of third abdominal segment; E-F, right and left thoracopods 1
with basal swelling. G-H, paratype-2. G, lateral view of third abdominal segment; H, right thoracopod 1 with

basal swelling.

dorsally over chitinous ridge.

Thoracopod 2 (Fig. 6A) large, well-developed,
appearing biramous because of presence of prom-
inent lateral papilla; leg itself unsegmented, but
tiny distal, setose “ramus” distinguishable from
“protopod” filled with seminal receptacles,
“ramus” armed apically and laterally with four
(right) or five (left) short setae or spiniform pro-

cesses together with cuticular ctenae and hairs
(Fig. 6B, C). Lateral papilla conical, pointing
ventrally, almost bare, extending to about middle
of “protopod”, terminating in relatively large
opening (Fig. 6D), apparently many much smaller
pores on sides of papilla, precise relationships to
seminal receptacle ducts unclear. Seminal recepta-
cles bottle-shaped, the number estimated at 14-17,

New Species of Baccalaureus 491

ty .
Y Yj mM

m "")

\l

\

M41
TAS
z Ky
MASS

Fic. 4. B. falsiramus sp. nov. Holotype, A, right antennule, showing musculature; B, apical part of A; C, left
antennule; D, apical part of C; E, mandible; F, maxillule; G, maxillae.

with long, extremely narrow ducts sheathed in
cells. Within receptacles, sperm usually in small
knot near duct entrance (see Fig. 6J, paratype),
some sperms with elongate heads 42 um long
expressed from them.

Thoracopods 3-4 similar to thoracopod 2. Thor-
acopod 3 armed with spiniform processes or short
setae on each ramus, three lateral and two termin-
al on the left (Fig. 6E), five lateral, two terminal,
and one medial on the right (Fig. 6F), each con-

taining 14-15 seminal receptacles. Right thoraco-
pod 4 (Fig. 7A) armed with one terminal and three
short lateral setae, left thoracopod 4 (Fig. 7B)
armed apically with one strong spine and one seta,
8-11 seminal receptacles in each “protopod”.
Left thoracopod 5 (Fig. 7C) sausage-shaped,
right one tapered, both same size, less than half as
long as preceding three pairs, lacking seminal
receptacles and lateral papilla, ornamented with
delicate cuticular ctenae almost all over its surface,

492 T. It6 AND M. J. GryGIER

Fic. 5.

apical part of A; C, left antennule; D, apical part of C; E, tips of maxillae (arrow indicating duct opening); F,
ventral view of labrum and maxillae; F, round papillae of labrum. Scales: A, C 100 um; B, D, F 50 um; E, G 10

pm.

with neither setae nor spines. sharp spines. Possible duct opening between
Thoracopod 6 absent, though ventral side of spines.

sixth thoracomere produced into transverse fold. Furcal rami (Fig. 7E) 0.50 mm long, basal height

Penis (Fig. 7D) vermiform, reaching end of 0.13 mm, tapering and weakly sigmoidal. Three
second abdominal segment, tapering toward two __ hirsute terminal setae, no medial setae or sensilla.

New Species of Baccalaureus 493

50um B-F,H,1

Fic. 6. B. falsiramus sp. nov. A-F, holotype. A, thoracopod 2; B, apex of right thoracopod 2; C, apex of left
thoracopod 2; D, apex of lateral papilla of left thoracopod 2; E, left thoracopod 3, posterior; F, right thoracopod
3. G-J, paratype-2. G, thoracopod 2; H, apex of thoracopod 2; I, apex of thoracopod 3; J, seminal receptacle.

Short cuticular ctenae present on whole medial

surface, distal quarter of lateral surface, and dorsal 1-2. DESCRIPTION OF OTHER FEMALES

and ventral edges. Thickened dorsal and ventral Outer coils of carapace of paratype-1 (Fig. 1B-

margins appearing scalloped internally. D) 6.9 mm high, 6.2 mm long, paratype-2 indeter-
minable in this respect due to damage. Pair of
dense patches of hairs dorsolaterally within body

494 T. It6 AND M. J. GryGiER

Fic. 7.

B. falsiramus sp. nov. A-E, holotype. A, apex of right thoracopod 4; B, apex of left thoracopod 4; C,

thoracopod 5; D, penis; E, medial view of furcal ramus. F-G, paratype-2. F, apex of left thoracopod 4; G, penis.

chamber (Fig. 1C; see also Fig. 2B). Bands of
filamentous material firmly attached to a carapace
coil of paratype-1 (Fig. 1C), probably pieces of
host tissue.

Body attitude in all these females almost the
same as in holotype, but horns of SEM-2 extending
straight forward, not particularly upturned (cf.
Fig. 2A and C). Distance from tips of horns to tips
of furcal rami 3.9mm in paratype-1, 4.2 mm in
paratype-2, about 5mm in SEM-1, 5.5mm in
SEM-2.

Thoracic segmentation of SEM-1 more obvious
than in type specimens, evident as five tergites
(thoracomeres 2-6) separated by wide areas of
thinner cuticle (Fig. 2A). Wide band of thin cuti-

cle separating lateral chitinous ridge from bases of
thoracopods (Fig. 2C), probably collapsed within
longitudinal groove visible in SEM-1 (Fig. 2A).
Paratype-2 and SEM-2 with prominent, dense cu-
ticular ctenae on ventral half of abdomen (Figs.
3E, 2D), SEM-1 lacking such prominent ctenae
but with many small pores instead (Fig. 8D, E),
similar to holotype. In both specimens examined
with SEM, dorsal setal patches on the last thoracic
and first abdominal segments composed of many
rows of hairs (Fig. 2E).

Antennules of SEM-1 similar to those of holoty-
pe in principal armature of apical segment and its
process (Fig. 5A, B), though details of branching
of terminal aesthetascs somewhat different. Apic-

New Species of Baccalaureus 495

Fic. 8. SEM photomicrographs of B. falsiramus sp. nov. A, B and F SEM-2, otherwise SEM-1. A, thoracopod 1
with basal swelling (arrow indicating membranous structure with hairs); B, enlarged view of A, showing patches
of hairs; C, thoracopods 1-4; D, thoracopods 4-6 (labeled as 4-6); E, pores on third abdominal segment; F,
thoracopods 4-6 (arrow indicating apical opening of lateral papilla of thoracopod 4. Scales: A, C, D 100 um; B,
E 10 um; F 50 pm.

496 T. IT6 AND M. J. GryGier

al process clearly delimited from apical segment,
but no other segmentation detected, at least exter-
nally. Apical process ornamented with cuticular
ctenae as in apical segment. Left antennule of
SEM-2 (right one lost) distinct in many respects
(Fig. SC, D), lacking distinct apical process but
ending in rounded apex with delicate hairs and
very small, closely set setae on frontal side. Un-
branched aesthetasc arising from subapical, post-
erior side, small seta attached to thick basal part of
aesthetasc.

Labrum bearing many flat, round, proximolater-
al papillae (Fig. SF, G).

Apical part of maxillae protruding from labrum
in SEM-2 with at least partial ring of unknown
material distal to triangular lateral plates, and two
closely set, spiniform structures extending anter-
ior, that may be debris (Fig. SE). Duct opening on
anterior prong of bifid tip. Left triangular plate
ending in two minute points.

Thoracopod 1, including its basal swelling, quite
variform. In paratype-2, right one similar to
holotype but left one a narrow, hairy, cylindrical
process arising from semicircular swelling, armed
with small apical seta (Fig. 3H). In paratype-1
(Fig. 1C), right one apparently similar to left one
of paratype-2, but details not determined. Thor-
acopod 1 in SEM-1 similar to elongate ones in
paratypes, but basal swelling more massive,
reaching dorsal limit of lateral chitinous ridge of
thorax and covered with long, dense hairs, espe-
cially abundant on dorsal edge (Figs. 2A, 8C).
Basal swelling of this specimen apparently sepa-
rated into two parts by shallow slit along dorsal
edge (Fig. 8C). Left thoracopod 1 of SEM-2
represented by thick, papillary process arising
from ventroposterior portion of discoidal plate
(Fig. 2C) and armed with two dorsoapical setae
(Fig. 8A); plate extending a little over dorsal limit
of lateral chitinous ridge and equipped with
numerous patches of short hairs (Fig. 8A, B).
Membranous structure fringed with dense hairs
(female gonopore ?) arising from dorsal gap be-
tween plate-like organ and trunk (Fig. 8A).

Terminal armament of thoracopods 2-4 variable.
In paratype-2 thoracopods 2-4 bearing two or
three prominent, apical setae in addition ot lateral
spines (Figs. 6H, I; 7F).

Lateral papillae of thoracopods 2—4 in SEM-1
almost naked (Fig. 8C), but those in SEM-2
ornamented with numerous patches of fine hairs.
Papilla of right thoracopod 2 of paratype-2 with
two cylindrical projections at tip rather than usual
opening (Fig. 6G); each projection with internal
duct and apparently extending from inside papilla,
though its origin unclear. Apical opening of lateral
papilla clearly seen in thoracopod 4 of SEM-2 (Fig.
8F).

In SEM-1, thoracopod 5 somewhat like thoraco-
pods 2-4 except for absence of lateral papilla (Fig.
8C, D). This specimen with rudimentary thoraco-
pod 6 (Fig. 8D). In SEM-2, thoracopod 5 repre-
sented by long, cylindrical process with round tip,
ornamented with numerous cuticular ctenae (Fig.
8F). This specimen with prominent thoracopod 6
similar to thoracopod 5 in general appearance,
including ornamentation, but shorter (Figs. 2C,
8F).

Penis of paratype-2 abruptly narrowing beyond
duct opening at about middle of ventral side,
terminating in two spiniform processes with single
subapical spiniform process (Fig. 7G). Penis of
paratype-1 also with spines but details unknown;
penis of SEM-1 hidden in preparation. Penis of
SEM-2 with narrow, nozzle-like apex with termin-
al opening and transversely arranged, fine spinules
(Fig. 2D), short spine at least on right side proxim-
al to nozzle.

Furcal rami examined with SEM ornamented
with fine, longitudinal wrinkles (Fig. 2F). Furcal
rami of SEM-2 slimmer than in SEM-1. Right
ramus of paratype-1 with only two terminal setae.

1-3. POSSIBLE MALE

A possible male was found in association with a
female (SEM-1), located near the aperture lips of
the female, partially embedded in the host (Fig.
9A). This is similar to the positional relationship
of males and females in Baccalaureus japonicus
Broch, 1929 [7]. Since the main body was not
visible, the possibility that this specimen is a
settled ascothoracid larva cannot be wholly ex-
cluded.

Carapace apparently bivalved, about 0.60 mm
long, surface ornamented with polygons deline-
ated by chitinous ridges and lacking central pores

New Species of Baccalaureus 497

TS

Fic. 9. SEM photomicrographs of B. falsiramus sp. nov. A, aperture lip of SEM-1 female (ar

‘ae
¥. 35 EN me Zz

row) and possible male

(arrowhead); B, enlarged view of carapace of possible male. Scales: A 500 #m; B 50 um.

(Fig. 9B).

1-4. COMMENTS ON HABITS

The adult females except for SEM-2 occurred
individually in nodules at the bases of groups of 5—
6 polyps, but each nodule’s inner cavity seemed to
be connected to only one or two polyps’ gastro-
vascular cavities. In one case, an external hole for
the carapace aperture was evident between the
bases of two nodule-associated polyps. SEM-2
occurred inside a spherical nodule formed of the
basal portion of a single polyp, and the lips of its
carapace aperture were clearly seen externally.

1-5. TAXONOMIC AND MORPHOLOGI-
CAL REMARKS

The present new species, Baccalaureus falsir-
amus, is easily distinguished from the nine de-
scribed species of the genus. None of the previous
descriptions [5-14] mentions such enormous papil-
lae lateral to the three major pairs of thoracopods,
only sometimes small, nipple-like protrusions
[e.g., 6]. Elsewhere in the Lauridae, Laura bicor-
nuta Grygier, 1985 [6], as moderately large, later-
ally directed, conical papillae at the bases of
thoracopods 2-5, and Polymarsypus digitatus
(Pyefinch, 1939) [12] has flat lateral lobes at the
bases of thoracopods 3 and 4; seminal receptacle
ducts pass through both structures [6].

The coiled carapace lobes of B. maldivensis
Pyefinch, 1934 [9], B. hexapus Pyefinch, 1936 [10],
B. torrensis Pyefinch, 1937 [11], B. argalicornis
Brattstrom, 1936 [14], and B. durbanensis Bratt-
strom, 1956 [5], all make over 1.5 turns like the
present new species, but these species all also have
long, downturned, coiled thoracic horns that wrap
around the columellae of the carapace lobes and
the horns are usually said to have short, retrorse
hairs; in B. falsiramus the horns are short, more or
less upturned, and naked. B. verrucosus Pyefinch,
1939, is the only species with horns as short as B.
falsiramus, but they are still downturned and hir-
sute [12]. Carapace spines like those in the present
new species have not been reported before.

The apical process or isolated aesthetasc on the
antennules of B. falsiramus as well as the “terminal
spine” in the unidentified specimen studied by
Grygier [6], the “lobe” in B. japonicus [7], and the
long, thin, spine-like tips observed by Pyefinch [12]
in B. pyefinchi Brattstr6m, 1956 [5], and B. dispar-
caudatus Pyefinch, 1939 [12], are probably homo-
logous to the claw guard and/or proximal sensory
complex of the distal antennular segment of other
ascothoracidans, including male laurids.

In the maxillae, the bent, triangular plates pre-
sumably correspond to the posterior, movable
hooks of many other ascothoracidans; if so, the
bifid distal tip is unusual, and a duct has never

498 T. It6 AND M. J. GryGIER

been observed there before.

The small, narrow furcal rami of B. falsiramus
are most similar to those of B. argalicornis and B.
torrensis, although in the last species they are
terminally unarmed. There are apparently diffe-
rent kinds of furcal armament in Baccalaureus:
terminal spines, terminal setae, tiny medial setae
and sensilla, cuticular ctenae. The true distribu-
tion of these elements among the species is not
clear from semantically ambiguous and incomplete
descriptions, but true setae terminally, lack of
medial armament aside from ctenae, and partial
lack of lateral ctenae may be peculiar to the new
species.

The variable structure of thoracopod 1 in B.
falsiramus is problematic. The first thoracopod in
Baccalaureus is usually accompanied by a “plate-
like organ” (interpreted by Grygier as a form of
filamentary appendage [6]). Only B. pyefinchi,
which has little carapace coiling but long thoracic
horns, has been previously known to have such
small swellings hardly extending dorsally over the
lateral chitinous ridge. However, the present
material of B. falsiramus exhibits marked variabil-
ity, and SEM-2 has a prominent structure that may
be called a “plate-like organ”. Thoracopod 1
proper in the present material appears as either a
small papillary process, a cylindrical process with
an apical seta, or a prominent papillary process
with two setae. Hence, thoracopod 1 morphology
might not be easily used as a diagnostic character
within this genus.

The number of thoracopods has been one of the
primary features used to diagnose species of Bac-
calaureus [e.g., 11], but since the sixth pair is not
always present in B. falsiramus, this feature is
actually not reliable at the species level.

The type specimens are smaller than the speci-
mens examined by SEM. The largest one (SEM-2)
has the best developed basal swelling (plate-like
organ) of thoracopod 1 and the most prominent
thoracopod 6. In SEM-1, the basal swelling is less
developed than in the largest specimen but more
so than in smaller specimens, and it has a
rudimentary thoracopod 6 while the smaller speci-
mens do not. These size-correlated morphological
differences may reflect, at least in part, differences
between adult instars.

Up to the present, most ascothoracidans have
not been subjected to detailed studies of morpho-
logical variability. _Dendrogaster astropectinis
would be an exceptional case, in which variation of
antennular morphology was studied by Karande
and Oguro [15]. As mentioned above, the ex-
amined females of B. falsiramus show enormous
variability, not only in the antennule but also in the
thoracopods, penis, abdominal ornamentation,
etc. Although this suggests that they may not all
be conspecific, we treat them for the time being as
conspecific because they share a characteristic
thoracic horn morphology, their thoracopods 2—4
are always associated with prominent lateral papil-
lae, and they lived in a single host colony.

Most of the identified hosts of Baccalaureus
have been various species of Palythoa, but B.
japonicus parasitizes a species of Parazoanthus and
Muirhead and Ryland [16] recorded Jsaurus as the
host of an undescribed species. Hence the present
report is the first confirmed occurrence of this
parasite in a species of Zoanthus. The host of the
type species, B. japonicus, has been recorded
under several names in Japan, including Zoanthus
cnidosus [7], but the name now used for this
zoanthid is Parazoanthus gracilis (Lwowsky) [e.g.,
17].

Baccalaureus is most widely recorded in the
Indian Ocean [5, 9, 10, 12, 14, 18], with Pacific
records limited to Japan [7, 8, 13, 17], Vietnam
[19], northeastern Australia [11, 16], and French
Polynesia [6]. The present finding adds a second
species to the Japanese fauna of this genus and
family.

Part 2. Larval Development

2-1. NAUPLII-GENERAL

Brood sizes are somewhat uncertain, but the
holotype had over 375 brooded eggs and nauplii
and paratype-1 had about 325 eggs in half of its
carapace plus part of the other half, perhaps 550-
600 in all.

There are six naupliar instars (one orthonauplius
and five metanauplii) before the ascothoracid lar-
va. In culture, the duration of instar I could not be
determined. When one of us (T. I.) removed a

New Species of Baccalaureus 499

100um A,B

Fic. 10. B. falsiramus sp. nov. Exuvium of instar VI nauplius. A, dorsal view; B, ventral view.

female (paratype-2) from the host, it released a
number of instar I nauplii together with eggs and
possibly some instar II nauplu. The released
nauplii immediately started to molt. None of the
three nauplii isolated for individual culture from
this batch of nauplu within 30 min. after release
was still in the first instar. They were put in culture
on 26 July, 1988, and metamorphosed into the
ascothoracid larva on either 22 (one) or 23 (two)
August, 1988. At 19°C, instar II lasted at least 3
days, and instars III-VI respectively 3, 2.5, 4, and
14-15 days. Molts were nearly synchronous, differ-
ing by at most one day. Thoracic limb buds of the
ascothoracid larva were first recognized on the
8th-10th day of the last naupliar instar. The nauplii
continued to swim during this period of about one
month without feedig.

Eggs oval, 0.48mm 0.35 mm. Late instar I
and newly molted instar II nauplii about 0.59 mm
long (excluding caudal armament), 0.45 mm wide,
0.28 mm thick (excluding labrum). Instar VI nau-
plii 0.60 mm long, 0.52 mm wide.

Body of instar I oval, with no ventral depression
where appendages occur, without distinct dorsal
shield. Cuticle of instar I very delicate (Fig. 11E),
with wrinkles, most likely separated from cuticle of
internally formed, succeeding instar. Body of

instars II-VI with broad depression where appen-
dages occur, and with bowl-shaped dorsal shield
(Figs. 10, 11), broader in front, cuticle only signi-
ficantly thickened in instar VI. Border of shield
outlined by equatorial pores (sensu Grygier [6]) on
inner “brim” of “bowl”. No equatorial pores in
instar I (Fig. 11B, E). At each molt after instar II,
cuticle splits ventrally along border of thin, ventral
cuticle and innermost edge of “brim” of dorsal
shield, except for short, caudal region (Fig. 15).
Nauplii escape from exuvia through gap formed by
widening of this fissure. Exuvia of instar I always
crumpled, often torn into pieces,
whether definite fissure line exists.

Dorsal shield of instars II-VI ornamented with
about three concentric, cuticular ridges on ventral
and outer face of “brim” (Figs. 10A, 11A, D),
some ridges connected with each other. No such
ridges in instar I (Fig. 11B, E). Instars II-VI with
small dorsal pores except along midline (Fig.
10A). Four pairs of hairs on dorsum of instar VI,
two anterior pairs in instar IJ. Instars III-V un-
known in this respect. No such pores or hairs on
dorsum of instar I.

Nauplius eye present in front of antennules in all
instars, with two obvious, red pigment cups. Pair
of simple frontal filaments 90-100 ~m long in

uncertain

500 T. IT6 AND M. J. GryGier

Fic. 11.

B. falsiramus sp. nov. SEM photomicrographs of nauplii. A, ventral view of instar II; B, ventral surface

between and anterior to antennules of instar I; C, frontal filaments of instar II; D, chitinous ridges of instar II; E,
caudal region of instar I; F, caudal region of instar II. Scales: A 100 um; B, D-F 10 um; C 50 um.

instars II-VI (Figs. 10B, 11C). Cylindrical, inter-
nal “cord”, possibly a sensory organ, extending
dorsally from point under cuticle between frontal
filaments in at least instar II. No frontal filaments
in instar I (Fig. 11B), but longitudinal row of three
pit-like, sensory structures at midline, at least
frontal two with rod-like sensillum, possibly diffe-

rent manifestation of internal “cord” of instar II.
Labrum small and triangular in all instars, with
two apical pores and two widely spaced pores on
hind surface at least in instars III-VI (Fig. 10).
Short, cuticle-lined duct extending anteriorly from
hind base of labrum in at least instars II-VI,
probably representing rudimentary mouth.

New Species of Baccalaureus 501

Relatively well-developed antennules, anten-
nae, and mandibles present in all instars,
rudimentary maxillules also present from instar II.
Caudal armament terminal in instar I, subterminal
on ventral side in instar II-VI (Fig. 15).

2-2. ANTENNULES

Instar I (Fig. 12A): Short apical segment de-
fined, with patch of spinules, but otherwise seg-
mentation indistinct, about five additional patches
of spinules marking possible future segments.
Setation matching Grygier’s [22] basic pattern for
brooded ascothoracidan nauplii: two single median
setae (a, b), grossly unequal pair of more distal
median setae (long-d, short-e) opposite a lateral
seta (f), and three unequal terminal setae (g).
Except for seta “c”, and shortest, medial “g” seta,
these setae thick and spinulose.

Instar II (Fig. 12B): five-segmented, three basal
segments subequal in length, next twice as long,
last very small, setation 0-1-1-(2+1)-3, relative
lengths of setae as in instar I but narrower; lateral

Tb) 66 99

g” seta subterminal. Two setae, “d” longest “g”,

\

Fic. 12. 8B. falsiramus sp. nov. Antennules of naupliar exuvia. A-E, instars I-V; F-G, instar VI.

plumose in this and later instars.

Instar III (Fig. 12C): Four-segmented due to
fusion of distal two segments of previous instar,
segments weakly marked, distal one spinulose,
setation 0-1-1-7 with new setae (h,) on lateral side,
presumably the most proximal one there, a little
shorter than seta “f”. Medial “g” seta, formerly
very short, now almost as long as “d” and setulose.
Lateral “g” seta, now the shortest of the three,
distally spinulose, perhaps in this instar only.

Instar IV (Fig. 12D): Similar to foregoing; seg-
mentation less distinct, setation 0-1—1-8 with very
short seta “h,” added proximal to “h,”.

Instar V (Fig. 12E): Unchanged except for addi-
tion of tiny lateral seta “h;” next to somewhat
longer “h,”.

Instar VI (Fig. 12F, G): Clearly four-seg-
mented, with unarmed basal segment, seta “a” on
second, seta “b” on third, each accompanied by
rudimentary new seta; fourth segment with subter-
minal claw rudiment (c) on medial side accompa-
nied by small, anterior seta (i), three groups of
distal setae: setae “d” and “e” medially terminal,

502 T. IT6 AND M. J. GryGier

50um
50um _, B-G

Fic. 13.

B. falsiramus sp. nov. Naupliar antennae. A, apical part of exopod of instar I, showing two apical setae of

instar II formed inside single terminal seta. B-G, exuvia of instars I-VI (some setae of C, E, and F omitted)
(arrows indicating openings of possible antennal gland).

three “g” setae laterally terminal, setae “f” and
“h,_3” subterminolateral, none of these setae very
short.

2-3. ANTENNAE

Instar I (Fig. 13B): Biramous, segmentation in-
distinct. Coxal area with two small, widely spaced,
enditic spines and scattered spinules. Basal area
with two closely set, enditic spines. Endopod as
long as protopod, stepped at two points along
medial edge; two subequal short setae on proximal
step, thin seta on distal step; two long and one
short, thin setae arising from apex; most setae
spinulose but short, thin ones on distal step and
apex simple; minute spinules occuring about step-
ped edges and apex. Exopod twice as long as
endopod, indistinctly annulated, bearing from
midlength five, thick spinulose setae up to 380 ~m
long. In individuals ready to molt to instar II,
basal seta and distal seta of exopod each contain-

ing two setae of next instar (Fig. 13A).

Instar II (Fig. 13C): Boundary between coxa
and basis clear. Coxa with two tiny, unequal,
enditic spines, pore (? opening of antennal gland
duct) on medioproximal edge, and fine hairs later-
ally. Basis with two equal, simple spines longer
than coxal ones. Endopod equally three-
segmented, first with two equal, simple setae,
second with one narrow seta (clearly basally setu-
lose only in this instar and one individual of instar
VI), third with three apical setae, two of them
well-developed and setulose. Exopod nine- or
exceptionally ten-segmented, sometimes varying
within a specimen, first segment short and indis-
tinctly demarcated from second; segments 1-3 with
no seta, segments 4-8 (5-9 on ten-segmented
exopods) each with one seta, last segment with two
apical setae; all setae well-developed and setulose.

Instar III (Fig. 13D): Unchanged except exopod
ten-segmented due to splitting of terminal segment

New Species of Baccalaureus 503

50m _,A-G

Fic. 14. B. falsiramus sp. nov. Naupliar mandible, exuviae. A, instar I (four endopodal setae omitted); B, endopod
of instar I, with bifid aberrant seta; C-G, instars II-VI (some exopodal setae omitted).

and bearing eight setae, segments 4-9 with one
each, terminal segment with two, including short,
probably simple one; shortest apical endopod seta
longer than before.

Instars IV-V (Fig. 13E, F): Unchanged from
instar III in major ornamentation.

Instar VI (Fig. 13G): Coxal spines longer, one
of them rather setiform, shortest terminal endopod
seta setulose and grown to two-thirds length of
other two, otherwise unchanged.

2-4. MANDIBLES

Instar I (Fig. 14A): Segmentation indistinct.
Coxal region with thick enditic spine and much
smaller one. Basis region with two short, enditic
spines, longer one spinulose. Endopod stepped at
two places along medial edge, two thick unequal
setae on first step, one thin and one thick setae on
second step, two thick setae and short, thin one
apically; all thick setae spinulose, thin ones naked.

Abnormal, bifurcate setae seen on one endopod
(Fig. 14B). Exopod a little longer than endopod,
indistinctly annulated, with patches of fine spi-
nules, bearing from midlength four thick, spinu-
lose setae. In nauplii preparing to molt to instar II,
basal seta and terminal seta of exopod each con-
taining two setae of next instar.

Instar II (Fig. 14C): Segmentation clear. Coxa
with distinct, medial, condylic articulation to ven-
tral body surface, not seen in antennae. Coxal
endite represented by low protuberance bearing
spinule on anterior margin and, probably, very
minute spinule posteriorly. Endite of basis repre-
sented by spine and short, setulose seta. Endopod
three-segmented, first segment with two setae,
longer one plumose, second segment with well-
developed, plumose seta, third with two well-
developed, plumose setae and hairlike seta. Ex-
opod 1.5 times as long as endopod, seven-
segmented; segments 1-2 short, with no seta;

Fic. 15.

504 T. It6 AND M. J. GRYGIER

segments 3-6 each with one well-developed seta;
segment 7 with two well-developed apical setae; all
setae setulose.

Instars III-V (Fig. 14D-F): Coxa as in instar II
except enditic spine shorter. Enditic spine of basis
reduced to spinule, seta unchanged. On endopod,
hairlike apical seta of instar II now well-developed

vo

J
o
Soey ove

and bearing some setules, though shorter than
other two apical setae, otherwise unchanged. Ex-
opod eight-segmented due to splitting of terminal
segment and bearing seven setae, segments 3-7
with one each, terminal segment with two, includ-
ing relatively short, simple one. Indistinct partial
division of third segment.

B. falsiramus sp. nov. Caudal region and rudimentary maxillules of nauplii. A, instar I; B-E, exuvia of

instars II-V with maxillules; F-G, exuvia of instar VI with maxillules and other rudimentary appendages (arrows

indicating fissures).

New Species of Baccalaureus 505

Instar VI (Fig. 14G): Coxa with short, wide,
enditic spine with two apical points. Otherwise
unchanged.

2-5. MAXILLULES AND OTHER
POSTERIOR APPENDAGES

Instar I: No maxillules.

Instar II (Fig. 15B): Maxillules represented by
pair of long, simple setae. No sign of more
posterior limbs, though seven or eight pairs of
transverse rows of spinules or spinular bands be-
tween maxillular bases and furcal spines (Fig.
11F).

Instars ITI-IV (Fig. 15C, D): Maxillules repre-
sented by pair of small, papillary processes, each
normally bearing two simple setae (only one seta
on left in one of three instar series: see Fig. 15C).
No sign of more posterior limbs.

Instar V (Fig. 1SE): Maxillules unchanged,
though setal lengths varying. Two or three pairs of
conical, posteroventral bumps representing more
posterior limbs.

IA SO.

Instar VI (FIg. 15F, G): Maxillules unchanged
in principal structure, though setal lengths varying.
In two cases formed as one-segmented rudiments
on short, non-articulated bases. Five or six pairs of
smaller and more posterior bumps than in instar V.

2-6. CAUDAL ARMAMENT

Instar I (Figs. 11E, 15A): Terminal spine and
furcal setae equal in length (35 um), but terminal
spine thicker, all more or less spinulose.

Instar II (Figs. 11F, 15B): Terminal spine ex-
tending slightly beyond rear of dorsal shield, usual-
ly longer than furcal setae, all thinner and more
pointed than in instar I, with very delicate spi-
nules.

Instars II-VI (Fig. 15C, D): Terminal spine and
furcal setae as in instar II, length variable among
specimens, patches of fine spinules anterior to
furcal spines.

Instar V (Fig. 1SE): Additional short spine or
seta appearing dorsal to each furcal seta.

Instar VI (Figs. 10, 15F, G). Unchanged.

Fic. 16. B. falsiramus sp. nov. Scheme of ascothoracid larva, with carapace optically cut away. a, antennule; ad,
adductor muscle; A4, fourth abdominal segment; f, furcal ramus; ff, frontal filament complex; mx2, maxilla; 1,
labrum; na, naupliar antenna; ne, nauplius eye; 16, sixth thoracic segment.

T. It6 AND M. J. GryGIER

506

if an
Hoa pA
rb

Fic. 17. B. falsiramus sp. nov. SEM photomicrographs of ascothoracid larva. A, habitus lateral; B, posterolateral
portion of carapace; C, habitus (left carapace removed); D, internal view of posterior portion of carapace; E,
antennule, frontal filament complex, and naupliar antenna (arrow indicating possible naupliar mandible); F,
apical hood of antennular claw guard (arrow indicating possible gland opening). Scales: A, C 100 um; B, D-F 10

ym.

New Species of Baccalaureus 507

2-7. SUMMARY OF NAUPLIAR
DEVELOPMENT

There are six naupliar instars. The greatest
morphological changes take place at the molt from
instar I to instar II: equatorial pores and frontal
filaments appear, the dorsal shield takes form, the
appendages become well segmented and set in a
deep ventral depression, the exopods of the anten-
nae and mandibles gain two setae each, the nata-
tory setae become setulose, maxillular rudiments
differentiate, and the caudal armament moves
ventrally. At later molts there are only minor
changes: addition of another segment and seta to

the antennal and mandibular exopods at instar III
along with incorporation of the distal antennular
segment into the penultimate one; gradual addi-
tion of lateral setae and a claw rudiment to the
antennule (Fig. 20); appearance of postmaxillular
limb buds and a second pair of furcal spines at
instar V; thickening of the dorsal shield of instar
VI.

2-8. ASCOTHORACID LARVA

The ascothoracid larvae (Fig. 16) are active
swimmers. They open their carapace valves to
extend their antennules and abdomen when swim-
ming. The red nauplius eye was easily seen ven-

seo

Fic. 18. B. falsiramus sp. nov. SEM photomicrographs of ascothoracid larva. A, dorsocaudal region of carapace,
showing cardic organs (arrows); B, enlarged view of cardic organ; C, ventral margin of carapace; D, last

abdominal segment and furcal rami. Scales: A, C, D 10 um; B 5 um.

508 T. It6 AND M. J. GryGIER

trally while they were swimming. The furcal rami _ thickened, somewhat interdigitated, recessed. In
are movable, often splaying apart from each other side view, valve outline slightly convex dorsally,
laterally. rounded anteriorly and posteroventrally, almost

Carapace bivalved, 0.51mm long, 0.31 mm _ straight mid-ventrally, concave posterodorsally be-
high, 0.30 mm wide (Fig. 17A). Dorsal hinge line low angle formed by protrusion above hinge line.

rae

min

50um A

\
a

ttt

Fic. 19. B. falsiramus sp. nov. Ascothoracid larva. A, oral area (1, labrum; mx2; maxilla); B, abdomen (penis
broken); C, antennule (most aesthetascs broken during dissection) and frontal filament complex; D, antennular
claw; E, thoracopod 1; F, thoracopods 4-5; G, thoracopod 6.

New Species of Baccalaureus 509

Outer surface of valves with polygonal (principally
hexagonal) meshes averaging 13 ~m across and
outlined by low, cuticular ridges (Fig. 17B). No
pits or pores in centers of meshes, but meshwork
often interrupted or adjoined by small, rounded,
ridge-bounded meshes 3-5 “m across, with either
central pore surrounded by elevated rim or seta
about 25 um long; these features absent in area
overlying adductor muscle attachment. Entire out-
er surface, except certain areas as described below,
with very fine granulation (Fig. 18B). On each
valve, five narrow, polygonal meshes equidistant
from hinge line, three flanking front of hinge, two
near its rear (Fig. 18A); each with tube arising
from anterior pit, running longitudinally along
bottom of mesh, and opening at posterior end
(Fig. 18B). Bottoms of these five meshes with no
granulation.

Band of prominent pores, derived from naupliar
equatorial pores, along entire free margin of valves
outside marginal cuticular ctenae (Fig. 18C). In-
ner surface of valves with several cuticular ridges
parallel to edge, additional posterior armament of
fine fringes of cuticular ctenae and several arrays
of long guard spines (Fig. 17D).

Body completely enclosed by carapace when
abdomen retracted, 0.40 mm from antennule to tip
of furcal rami in retracted position, with head,
six-segmented thorax, and four-segmented abdo-
men with furca (Figs. 16, 17C). Large nauplius eye
with two obvious red pigment cups in hemispheric-
al ventral protrusion of “forehead” anterior to
labrum.

Frontal filament complex hanging just lateral to
rear base of antennule, with rounded, rather pyri-
form basal part (42 ~m long, 32 ~m wide in SEM
specimen) and biramous sensory process; branches
unequal, main one about 100 “m long, shorter,
thinner one arising near its base (Figs. 17E, 19C).

Antennule composed of six major segments,
Z-shaped (Fig. 19C). First segment short, next
three together not much bigger than fifth, third
one triangular with long hairs on anterior edge,
fourth and fifth each with two nearly equal anterior
setae. Sixth segment with thin, movable claw with
bump on convex side and row of distally longer
denticles on concave side (Fig. 19D); three long
setae at base of claw. Laterally flanged claw guard

with one seta distal and two proximal to small
apical hood, latter a membrane armed marginally
and externally with many spinules and protecting
short, protruding tubule on end of claw guard (Fig.
17F); shelf at mid-length of claw guard on free
side, bearing short tubule (? broken seta) on one
antennule of dissected specimen only. Proximal
sensory process near base of claw guard short and
cylindrical, with three setae, one long and thick,
other two replaced by bifid seta in one case, all
three setae with delicate spinules. Cluster of about
25 long aesthetascs arising on sixth segment from
basolateral, oval region with thick, chitinous bor-
der (Figs. 17C, E; 19C).

Pair of vestigial but large naupliar antennae
(Fig. 17C, E) posterior to frontal filament com-
plexes; medial sides, including possible endopods,
not clearly seen; internal musculature present;
vestiges of 5-6 setae on narrow distal part, pre-
sumably representing exopod. Another, much
smaller and badly shrunken structure apparent
behind antenna in SEM specimen (Fig. 17E), poss-
ibly vestige of naupliar mandible.

Oral cone imperfectly formed (Fig. 19A). Lab-
rum a rounded lobe partly enclosing other mouth-
parts. Mandibles and maxillules spiniform, but
latter thicker. Maxillae relatively large, tips nar-
row and bifid.

Thorax demarcated from head by distinct suture
(Fig. 17C), first segment shorter than segments 2—
4, segments 5 and 6 much longer dorsally due to
body curvature; segments 2—4 with small external
swelling anteriorly above limb articulations.

Six pairs of thoracopods biramous. First limb
narrower than others; coxa and basis lined on both
sides by long hairs (Figs. 17E, 19E); exopod two-
segmented, with hairs on lateral edge of first
segment and medial edge of second one, armed
with five setae, medialmost one thicker than others
and with dense, fine spinules bilaterally, others
more or less plumose; endopod represented by
small, hairy, setiform process. Limbs 2-5 (Fig.
19F) similar to each other; posterior border be-
tween coxa and basis unclear; basis with shallow,
lengthwise groove on at least distal half of anterior
surface; exopod two-segmented, first segment with
no seta, second one with five setae, lateralmost
seta thin, short, and minutely spinulate, others

510 T. Ir6 AND M. J. GryGIER

equally thick and long, medialmost one markedly
spinulate while others plumose; endopod three-
segmented, with no seta on first segment, marked-
ly spinulate seta on second segment, three more or
less plumose setae on third segment. Limb 6 (Fig.
19G) smaller than previous limbs; coxa and basis
distinct, both naked; lengthwise, anterior groove
of basis clear, reaching near coxa; rami two-
segmented; exopod with no seta on first segment,
three similar, more or less plumose setae on
second segment; endopod with short spinules on
medial edge of first segment, two terminal setae on
second segment.

When observed in whole mount, first abdominal
segment seemed to bear lobe-like penis, but penis
lost from dissected abdomen so details unknown
(Fig. 19B). Fourth abdominal segment longer than
others, with ventral and ventrolateral cuticular
ctenae and pair of lanceolate, movable, somewhat
setiform telsonic spines (Figs. 18D, 19B).

Furcal rami (Fig. 19B) slightly longer than high
in lateral view, flattened laterally, slightly curved
with concave medial faces, closer together dorsally
than ventrally. Armament consisting of transverse
cuticular ctenae laterally; lanceolate, short spine
on dorsal edge near posterior end; four apical and

(mle I HII
Fic. 20.

three medial setae, dorsalmost seta inserted into
short process, both it and ventral seta dorso-
ventrally flattened and lanceolate, other setae
setulose.

2-9. DISCUSSION

Nauplii. -The present study has demonstrated
that lecithotrophic nauplii of ascothoracidans can
be raised at least to the ascothoracid larva in the
laboratory by using the culturing method that was
originally developed to raise y-larvae [20]. This
kind of laboratory culture is essential for the
identification of larval instars and for a precise
analysis of their morphological changes. Although
Grygier [21, 22] reported without full descriptions
a minimum of five naupliar instars in Gorgono-
laureus muzikae Grygier, 1981, and Parascothorax
cf. synagogoides Wagin, 1964, and Wagin [3] de-
scribed seven “stages”, one hypothetical, in the
naupliar development of Ascothorax ophioctenis,
the present results are the first that shows conclu-
sively that some ascothoracidans have six naupliar
instars. This number is generally regarded as
plesiomorphic in at least some maxillopodan taxa,
such as Copepoda and Cirripedia.

It is not clear when the nauplii in the present

IV V Vi

B. falsiramus sp. nov. Schematic representation of antennular development through six naupliar instars

(I-VI). Arrows indicating newly added setae and claw, other labels used for explanation in the text.

New Species of Baccalaureus 511

species normally leave the female’s brood cham-
bers. However, the cuticle of instar I nauplii is
very thin and delicate and tore easily, and they
immediately molted into the next instar after being
released into sea-water. Instar I nauplii are not
active swimmers, but instar II nauplii have setulose
natatory setae and are good swimmers. These
observations suggest that, like cirriped nauplii,
they normally leave the female’s brood chambers
at the end of instar I, whereupon they may molt
immediately to instar IJ and remain planktonic
until the metamorphosis to the ascothoracid larva.

Early instar nauplii have been described in
several laurids (list in [22]). Most have apparently
planktotrophic nauplii with very strong antennal
and mandibular endites. The present nauplii are
considerably larger than those previously reported
in Baccalaureus. Naupli of B. japonicus are 0.4
mm long and 0.3 mm wide [7], and B. pyefinchi
and B. argalicornis have nauplii only about 0.3 mm
long [12, 14]. These might not all be the same
instar. Nonetheless, the present nauplu differ
additionally from those of B. japonicus in never
developing large, enditic spines on the antennae
and mandibles and in not having a long, protruding
terminal spine flanked by three pairs of posterior
dorsal shield papillae. Nauplu of B. maldivensis
and B. hexapus seem to have weakly armed limbs
according to schematic diagrams and a setation
table [9, 10]. However, they differ from B. falsir-
amus by the large, pointed, rear end of their
bodies and the much longer limbs relative to the
body. Within the Lauridae, Polymarsypus digita-
tus has nauplii most similar to the present ones [6];
these are probably instar I ready to molt to instar
II. They are nearly the same size as the equivalent
stage in the present material (0.62 0.44 mm),
have similar, weak armament of the antennae and
mandibles, and a rounded rear end. However,
they reportedly have frontal filaments and equato-
rial pores already, clear appendage segmentation,
and somewhat different proportions of the pro-
topods and rami. But the frontal filaments and
quatorial pores might actually have belonged to
the fully developed instar II within; their absence
in the present instar I nauplii was only confirmed
on the basis of exuviae and SEM. Such a misinter-
pretation may also explain the supposed presence

of these structure in some other first instar nauplii
like those of Laura bicornuta {cf., 6].

In comparison to the few known planktonic
ascothoracidan metanauplii tentatively attributed
to the Lauridae [22, 23], the present sixth instar
nauplii have less well-developed antennules (four
segments instead of six), little development of
complex appendage armament for feeding, and
little elaboration of the furcal armament. The
pattern of dorsal pores and setae on the last instar
differs from that of Boxshall and Bottger-
Schnack’s [23] Red Sea metanauplius type I, which
has three pairs of setae toward the rear instead of
two pairs each at front and rear, and three longitu-
dinal pore-free strips instead of one. In both, the
pores are most dense in a row to either side of the
central bare zone.

Development of naupliar antennules in B. falsir-
amus in summarized in Figure 20. The addition of
several setae (h; 3) proximal to the original lateral
seta (f) of the penultimate antennular segment has
not been observed in ascothoracidan development
before. Its occurrence in B. falsiramus suggests
that the lateral spine on the naupliar antennule of
Endaster hamatosculum Grygier, 1985 [22, 24],
which is not homologizable with any of the pre-
viously recognized setae in ascothoracidan anten-
nules, might be an “h” seta. In the six-segmented
antennules of the planktonic ascothoracidan meta-
nauplii cited above, the fourth segment has a claw
rudiment and two setae, one of which is probably
equivalent to the small seta (i) next to the claw
rudiment in the present species, and the other to
an “h” seta.

The present study revealed that some ascothor-
acidan nauplii have cuticular ridges at least on the
marginal area of their dorsal shield. More broadly
developed, cuticular ridges have long been known
to occur in so-called, nauplius y larvae [e.g., 25-
27] and the possession of such prominent cuticular
ridges has been believed to define, at least in part,
the Facetotecta in which these larvae are
accommodated [21]. Despite ignorance of the
presence of such ridges in the Ascothoracida,
Bresciani [28] suggested, and Ité [1] seconded, a
possibility that y-larvae are the larval stages of
certain ascothoracidans, but Grygier [22, 29] has
disagreed. The present finding of cuticular ridges

512 T. IT6 AND M. J. GryGIER

in an ascothoracidan appears to bridge over one of
the morphological gaps between these two thecos-
tracan groups. However, it has recently been
reported that the thoracican cirripede Jbla has
marginal ridges on its nauplii [30]. Now it is
apparent that variously developed, cuticular ridges
are widely distributed within the Thecostraca.

Ascothoracid larva. Ascothoracid larvae similar
to this species’ have been reported several times
before and tentatively assigned to the Lauridae,
which is strongly supported by the present study.
Tessmann [31] first described one from the Indian
Ocean, and thus such larvae can be called “Tess-
mann’s larvae” [32]. McKenzie [33] schematically
illustrated a similar larva from the eastern Indian
Ocean, and Bonaduce et al. ({34]: FIG. 2, fig. 4)
found an isolated carapace valve in Red Sea sedi-
ments, which they mistakenly attributed to an
ostracod. Grygier [6] found a single larval cara-
pace valve together with adults of the laurid Zoan-
thoecus cerebroides Grygier, 1985, and later [32]
gave a detailed, SEM-assisted description of a
shallow-water, Hawaiian form. The carapaces all
have a reticulate pattern of chitinous ridges and a
posterodorsal emargination like the present exam-
ples, and the three forms with described appen-
dages [31, 32; herein] have a proximal cluster of
aesthetascs on the claw-bearing segment of anten-
nules. Additional common features of the present
larvae with Grygier’s [32] Hawaiian form are the
antennular claw with a comb-like row of spinules,
an imperfectly formed oral cone, the thoracopodal
setation, including a setiform endopod on the first
limb, and the frontal filament complex with a bifid
filament. Some of these features may eventually
prove to universally characteristic of Tessmann’s
larvae.

Tessmann’s description of the distal antennular
and thoracopodal structures [31] seems incomplete
(thoracopodal rami uniformly two-segmented, ex-
opods with two setae, endopods with one), and the
supposedly four-segmented antennules and five-
segmented abdomen (equivalent of first segment in
other larvae drawn with discontinuous tergite) are
at odds with the other descriptions. Grygier’s
Hawaiian larvae [32] are larger than the present
ones (0.60 x 0.35 mm), their carapace meshes each
have a deep pit with a pore instead of a flat surface,

and the polygons are smaller, about 11 «4m across.
On the antennule, the setae on segment 5 are
unequal, the distal setae on segment 6 much
shorter, the claw more deeply curved, and the claw
denticles less numerous than in B. falsiramus;
there are fewer proximal aesthetascs (only up to
13). There are no vestigial naupliar appendages,
although this might not be a stable character
because Tessmann’s original larva might have such
vestiges ([31]: “r6hrenformiges Gebilde”) and re-
duction of vestigial naupliar appeandages is known
in facetotectan cyprids [20]. In the Hawaiian form
the first thoracomere is not well delineated from
the head, and the furcal rami are longer with
supposedly non-setulate setae. Thus, Tessmann’s
larvae from Hawaii differ from the larvae of B.
falsiramus in many morphological features, which
may become taxonomically useful once more
forms are linked to their adults.

In the echinoderm-infesting, ascothoracidan
order Dendrogastrida, many species have two
ascothoracid larval instars (e.g., [2, 35]; review in
[21]). Such a relatively gradual transition from the
nauplius to the adult has been assumed to the
plesiomorphic relative to the single cypris larva in
the Cirripedia [21, 36], but it is important to be
sure that it is not a secondary development in a
single ascothoracidan clade. The dendrogastridan
first instar is characterized by inflated, poorly
hinged carapace valves; antennules in an in-
termediate state between the metanaupliar and
adult conditions (e.g., [22]; Fig. 6a); the labrum
not fully surronding the other mouthparts; some-
times remnants of the naupliar antennae and man-
dibles; short, simple setae on the legs, usually only
distally; and a four-segmented abdomen. The
ascothoracid larva of B. falsiramus has an incom-
pletely developed oral cone, a pair of naupliar
appendages, and lacks protopodal setae on the
legs, but in other respects it is as well developed as
typical dendrogastridan late ascothoracid larvae
(in this species the adults have four-segmented
abdomens, so that is not significant in the larvae).
This circumstantial evidence suggests, but certain-
ly does not prove, that this species has a single
ascothoracid larval instar.

The presence of a large nauplius eye in these
ascothoracid larvae, as well as in the nauplii, is

New Species of Baccalaureus 513

significant because it confirms Tessmann’s iden-
tification of a nauplius eye in his original larva [31]
and because only previous record of an eye in an
ascothoracidan nauplius concerned the first re-
corded species, Laura gerardiae Lacaze-Duthiers,
1865 [37].

The presence of up to three setae proximal and
one distal to a poorly understood apical structure
(herein formally termed the apical hood) on gener-
alized claw guards in the Ascothoracida has long
been noted [e.g., 38]. There is a “spinulose tip” on
the claw guard in Jsidascus bassindalei Moyse,
1983 [39] and many short hairs on the “articulating
tip” of the claw guard in Cardomanica andersoni
Lowry, 1985 [40], but the microstructure of the
apical hood has not previously been examined. In
the present larva it proves to be a spinulose cloak
around a previously unsuspected tubule. Grygier
[32] may have seen the tubule in the Hawaiian
Tessmann’s larva, mistaking it for a sensillum on
the “distal flange.” The function of the tubule,
whether chemosensory or a gland duct, is unclear.

The subterminal position of the antennular claw
rudiment (c) in naupliar instar VI supports Gry-
gier’s assertion [22] that the claw-bearing segment
(sixth segment in the present ascothoracid larva) is
not really the most terminal one in ascothoracidan
antennules. Rather, at least the proximal sensory
process on the claw-bearing segment is assumed to
be equivalent to the palp (apical segments distal to
the claw-bearing segment) in facetotectan anten-
nules [22, 41].

The five pairs of peculiar organs near the cara-
pace hinge line have been recorded in other
ascothoracidan, though usually only four pairs.
We would like to formally call them “cardic
organs” (from Latin cardo=hinge). The second
ascothoracid larva of Ascothorax gigas Wagin,
1968, has two pairs of “slit-like pores or sensory
organs” at either end of the hinge [35], and so does
the last (second ?) instar ascothoracid larva of
Parascothorax ?synagogoides [21] and an uniden-
tified ascothoracid larva from the Virgin Islands, in
which they are called “oval pits” [32]. Adult
Waginella sandersi (Newman, 1974) have at least
two pairs [38]. Due to their position near the
hinge, the cardic organs might be interpreted as
propriceptors. However, they may also be parts of

glands, because they involve an externally pro-
duced tube.

Cardic organs may be homologous to similarly
placed, supposedly chemoreceptive “lattice
organs” recently described in detail in a lepado-
morph cirriped cyprid [42], and whose morphology
has been summarized for 28 other cirriped species
as well [43]. There are usually five pairs of these
narrow, elongate structures symmetrically dis-
posed near the carapace midline, two pairs anter-
iorly, three posteriorly. They have a porose floor
(“sensory field”) with a single large pore anteriorly
in the front two pairs, posteriorly in the rear ones,
and the sensory field is surrounded by a ring of
non-porose, often thickened cuticle. The cardic
and lattice organs are also somewhat similar in
form and especially position to the “dorsal com-
pound organs” on the midline of the carapaces of
pelagic eucarids and at both ends of the carapace
midline in the leptostracan Nebaliopsis typica Sars,
1887, for which a photosensitive capacity has been
postulated [44]. The malacostracan organs consist
in part of paired pores and often slits, the latter at
least superficially reminiscent of cardic and lattice
organs. The ultrastructure and neurology of all
these organs need to be investigated.

ACKNOWLEDGMENTS

This study was supported by Grant-in-Aid for Scien-
tific Reserach No. 62540567 from the Ministry of Educa-
tion, Science and Culture of Japan to T. I., by a travel
grant from the Seto Marine Biological Laboratory to M.
J. G., and by the Sesoko Marine Science Center of the
University of the Ryukyus, where M. J. G. was Visiting
Foreign Researcher in 1988-89.

REFERENCES

1 Ito, T. (1987) Y-larvae - enigmatic crustacean lar-
vae. Ann. Rep. Seto Mar. Biol. Lab., 1: 52-58. (In
Japanese)

2 Brattstro6m, H. (1948) Undersokningar 6ver Ore-
sund XXXIII. Studies on Ulophysema 6resundense
2. On the larval development of the ascothoracid
Ulophysema Oresundense Brattstrom. Lunds Univ.
Aarskr. N. Ser., Avd. 2, 44: 1-70.

3 Wagin, V. L. (1954) [On the structure, larval
development and metamorphosis of dendrogasterids
(parasitic crustaceans of the order Ascothoracida)].
Uch. Zap. Leningr. Gos. Univ. Ser. Biol. Nauk,

11

12

13

14

17

514

172: 42-89. (In Russian)

Karande, A. A. and Oguro, C. (1981) Larvae of
Myriocladus astropectinis (Yosii 1931) (Ascothoraci-
ca) reared under laboratory conditions. Proc. Ind.
Acad. Sci., Anim. Sci., 90: 23-31.

Brattstrom, H. (1956) On the organization of the
genus Baccalaureus with description of a new South
African species. In “Bertil Hanstrém, Zoological
Papers in Honour of his Sixty-fifth Birthday Novem-
ber 20, 1956”. Ed. by K.G. Wingstrand, Zoological
Institute, Lund, pp. 106-119.

Grygier, M. J. (1985) Lauridae: taxonomy and
morphology of ascothoracid crustacean parasites of
zoanthids. Bull. Mar Sci., 36: 278-303.

Yosii, N. (1931) Note on the organisation of Bacca-
laureus japonicus. Annot. Zool. Japon., 13: 169-
187.

Broch, H. (1929) Baccalaureus japonicus, ein neuer
Cirriped aus der Unterordnung der Ascothoracida.
Mitt. Zool. Mus. Berlin, 15: 237-244.

Pyefinch, K. A. (1934) Baccalaureus maldivensis, a
new species of ascothoracican. Quart. J. Microsc.
Sci., 77: 223-242.

Pyefinch, K. A. (1936) The internal anatomy of
Baccalaureus with a description of a new species.
Quart. J. Microsc. Sci., 78: 653-686.

Pyefinch, K. A. (1937) The anatomy of Bacca-
laureus torrensis, sp. n. (Cirripedia Ascothoracica).
J. Linn. Soc. Lond. Zool., 40: 347-371.

Pyefinch, K. A. (1939) Ascothoracica (Crustacea,
Cirripedia). John Murray Exped. 1933-34 Scient.
Rep., 5: 247-262.

Okada, Y. K. (1938) Les Cirripédes Ascothoraci-
ques. Trav. Stat. Zool. Wimereux, 13: 489-514.
Brattstrom, H. (1936) Baccalaureus argalicornis n.
sp., eine neue Ascothoracide aus Madagascar.
Kungl. Fysiogr. Sallsk. Lund Forhandl., 6(20): 1-
14.

Karande, A. A. and Oguro, C. (1981) Comments
on taxonomic characters of Myriocladus astropecti-
nis Yosii, 1931 (Ascothoracica). Crustaceana, 41:
108-110.

Muirhead, A. and Ryland, J. S. (1985) A review of
the genus Isaurus Gray, 1828 (Zoanthidea), includ-
ing new records from Fiji. J. Nat. Hist., 19: 323-
335.

Utinomi, H. and Kikuchi, T. (1966) Fauna and flora
of the sea around the Amakusa Marine Biological
Laboratory. Part IV. Cirriped Crustacea. Contr.
Amakusa Mar. Biol. Lab., Kyushu Univ., 195: 1-
11.

Herberts, C. (1972) Etude systématique de quel-
ques zoanthaires tempérés et tropicaux. Tethys
Suppl., 3: 69-156.

Dawydoff, C. (1952) Contribution 4 l’étude des
invertébrés de la faune marine benthique de I’In-

20

21

22

24

25

26

27

28

29

30

31

32

33

34

35

T. It6 AND M. J. GryGier

dochine. Bull. Biol. Fr. Belg. Suppl., 37: 1-158.
It6, T. and Takenaka, M. (1988). Identification of
bifurcate paraocular process and postocular filamen-
tary tuft of facetotectan cyprids (Crustacea: Maxillo-
poda). Publ. Seto Mar. Biol. Lab., 33: 19-38.
Grygier, M. J. (1984) Comparative mophology and
ontogeny of the Ascothoracida, a step toward a
phylogeny of the Maxillopoda. Ph. D. Thesis, Uni-
versity of California San Diego. 417 pp.

Grygier, M. J. (1987) Nauplii, antennular ontogeny
and the position of the Ascothoracida within the
Maxillopoda. J. Crust. Biol., 7: 87-104.

Boxshall, G. A. and Bottger-Schnack, R. (1988)
Unusual ascothoracid nauplii from the Red Sea.
Bull Br. Mus. Nat. Hist. (Zool.), 54(6): 275-283.
Grygier, M. J. (1985) Crustacea Ascothoracida.
Mém. Mus. Nat. Hist. Nat., Paris, Sér. A. Zool.,
133: 417-426.

Hansen, H. J. (1899) Die Cladoceren und Cirripe-
dien der Plankton-Expedition. Ergeb. Plankt.-
Exped. Humboldt-Stift., 2(G, d): 1-58.

Schram, T. A. (1972) Further records of nauplius y
type IV Hansen from Scandinavian waters. Sarsia,
50: 1-24.

It6, T. (1986) Three types of “nauplius y” (Maxillo-
poda: Facetotecta) from the North Pacific. Publ.
Seto Mar. Biol. Lab., 31: 63-73.

Bresciani, J. (1965) Nauplius “y” Hansen: its dis-
tribution and relationship with a new cypris larva.
Vidensk. Medd. Dansk Naturh. For., 128: 245-258.
Grygier, M. J. (1987) New records, external and
internal anatomy, and systematic position of Han-
sen’s y-larvae (Crustacea: Maxillopoda: Face-
totecta). Sarsia, 72: 261-278.

Anderson, D. T. (1987) The larval musculature of
the barnacle Jbla quadrivalvis Cuvier (Cirripedia,
Lepadomorpha). Proc. R. Soc. Lond., B, 231: 313-
338.

Tessmann, M. (1904) Beitrage zur Entwicklungsges-
chichte der Cirripedien. Inaugural-Dissertation,
Universitat zu Leipzig. 38 pp.

Grygier, M. J. (1988) Larval and juvenile Ascothor-
acida (Crustacea) from the plankton. Publ. Seto
Mar. Biol. Lab., 33: 163-172.

McKenzie, K. G. (1972) Contribution to the
ontogeny and phylogeny of Ostracoda. In “Proc.
IPU, XXIII Internat. Geol. Congr., Prague, 1968.”
pp. 165-188.

Bonaduce, G., Ciliberto, B., Minichelli, G., Masoli,
M. and Pugliese, N. (1983) The Red Sea benthic
ostracodes and their geographical distribution. In
“Applications of Ostracoda”. Ed. by R. F. Mad-
docks, University of Houston Department of
Geosciences, Houston, pp. 472-491.

Grygier, M. J. and Fratt, D. B. (1984) The
ascothoracid crustacean Ascothorax gigas: rede-

36

Si/

38

39

40

New Species of Baccalaureus

scription, larval development, and notes on its in-
festation of the Antarctic ophiuroid Ophionotus
victoriae. Antarct. Res. Ser., 41: 43-58.

Boxshall, G. A. and Huys, R. (1989) New tantulo-
carid, Stygotantulus stocki, parasitic on harpacticoid
copepods, with an analysis of the phylogenetic rela-
tionships within the Maxillopoda. J. Crust. Biol., 9:
126-140.

Lacaze-Duthiers, H. de. (1880) Histoire de la Laura
gerardiae: type nouveau de Crustacé parasite. Arch.
Zool. Exp. Gén., (1)8: 537-581.

Grygier, M. J.(1983) Revision of Synagoga (Crus-
tacea: Maxillopoda: Ascothoracid). J. Nat. Hist.,
17: 213-239.

Moyse, J. (1983) Isidascus bassindalei gen. nov., sp.
nov. (Ascothoracida: Crustacea) from north-east
Atlantic with a note on the origin of barnacles. J.
Mar. Biol. Ass. U. K., 63: 161-180.

Lowry, J. K. (1985) Cardomanica andersoni n. gen.,

41

42

43

44

515

n. sp. from the western Tasman sea with notes on
species from the tropical western Atlantic Ocean
(Crustacea: Ascothoracida: Synagogidae). Rec. Au-
stral. Mus., 37: 317-323.

It6, T. (1989) A new species of Hansenocaris
(Crustacea: Facetotecta) from Tanabe Bay, Japan.
Publ. Seto Mar. Biol. Lab., 34: 55-72.

Elfimov. A. S. (1986) Morphology of the carapace
in the cypris larva of Heteralepas mystacophora
Newman (Cirripedia, Thoracica). Biologiya Morya
1986(3): 30-34. (In Russian with English summany)
Elfimov. A. S. (1989) [The cypris larvae of cir-
ripedes and their significance in the formation of
fouling]. Avtoreferat of Candidate Thesis, Moscow
State Univ., pp. 1-22. (In Russian)

Mauchline, J. (1977) The integumental sensilla and
glands of pelagic Crustacea. J. Mar. Biol. Ass. U.
K., 57: 973-994.

nue

ia oyAiacnpartl Ww aban? wa i

se
ss
vy Te ivitet} ye a
ric irl, {Nd ) eoianes
IiOOncCeVY aspera)
euih¥. torr
’ 72) : H if
tire cwct maha
iM! ctaCebl
ipo Pie eee |
Le | 14 dg?
ry (ine?
rea ay
ents
(¥ ny why Geet
J i]
7 ¢ sf A
H 4] Gs
| are
4 4

Tal Meies | oat emeth| aLfiestady Atal Rall

wide A eh wy wiromlt (tae Phrreh? *

le Cimee creel cee tl. sa)

j ; - 0 ¥
eee ae ee eT

er eer ee tar nen i '
a Wn Athi tA Ata im ’
hitea ist ken Gait y Bak (AR

als Dpowexir wiht Myaay tap
© lath cust) iE din politely

7
™ mal’ oe

4

4. csuumétererenarw

cunt hb andes Fe late
\ (Oren
, ramet

ire wey

Abd te } Vora

we -
f we Comvep each
5 anit is! ‘y
tpi att t
2 he
Maw 6

ZOOLOGICAL SCIENCE 7: 517-535 (1990)

© 1990 Zoological Society of Japan

Three New Species of the Genus Rhombognathus
(Acari, Halacaridae) from Japan

Hirosui ABE

Department of Systematic Zoology, Division of Environmental Structure,
Graduate School of Environmental Science, Hokkaido
University, Sapporo 060, Japan.

ABSTRACT—Three new species of the genus Rhombognathus are described from Japan. Rhombo-
gnathus atuy sp. nov. and Rhombognathus ezoensis sp. nov. differ from their congeners in the
chaetotaxies of the posterodorsal plate, genital region and legs, and from Rhombognathus dissociatus

sp. nov. in having three ventral plates.

INTRODUCTION

The genus Rhombognathus was established by
Trouessart [1]. Only two species of Rhombo-
gnathus have hitherto been known from waters
adjacent to Japan, viz. Rhombognathus terminalis
and Rhombognathus denticulatus described by
Sokolov [2] from the Sea of Japan. The present
paper describes three new species of this. genus
collected from marine algae at intertidal zones in
Hokkaido, northern Japan.

The type-series is deposited in the collections of
the National Science Museum, Tokyo, of the
Zoological Institute, Faculty of Science, Hokkaido
University, Sapporo, and of the National Museum
of Natural History, Smithsonian Institution,
Washington, DC, U.S.A., and in my private col-
lection.

Terms and the systems of notation for numerical
data follow Newell [3-5].

Abbreviations: AD, anterodorsal plate; PD,
posterodorsal plate; OC, ocular plate; AE, anter-
ior epimeral plate; PE, posterior epimeral plate;
GA, genitoanal plate; ds, dorsal setae; aes-i,
anterior epimeral setae; aes-ii-lat (-v, -adj), lateral
(ventral, adjunctive) setae of coxae II; pes-iii-lat
(-v, adj), lateral (ventral, adjunctive) setae of
coxae III; pes-iv, setae of coxae IV; P-1 to P-4, first
to fourth segment of palp.

Accepted June 22, 1989
Received April 24, 1989

In addition, the follwoing abbreviations are used
in the figure legends: Ds, dorsal view; Vr, ventral
view; R, right appendage (or part); L, left appen-
dage (or part).

Family Halacaridae Murray
Subfamily Rhombognathinae Viets
(Japanese name: Kaisoudani-aka, new)

Genus Rhombognathus Trouessart
(Japanese name: Kaisoudani-zoku, new)

Rhombognathus atuy sp. nov.
(Japanese name: Umibe-kaisoudani, new)
(Figs. 1-4)

Type-series. Holotype: Female, intertidal, on
Sargassum at 0.5m depth at low tide, Usu
(42°31’N, 140°47°'E), Hokkaido, Japan, 10. vii.
1986, H. Abé coll. Allotype: Male, data same as
the holotype. Paratypes: 1 female, intertidal, on
Sargassum at 0.5m depth in tide pool, Usujiri
(41°56°N, 140°57’'E), Hokkaido, 12. vi. 1986, H.
Abé coll.; 2 males, 1 female, intertidal, on Coralli-
na at 0.1 m depth in tide pool, Mitsuishi (42°14’N,
142°36E), Hokkaido, 8. xii. 1988, H. Abé coll.

Female (holotype). Idiosoma 364 4m long, 240
ym wide. Color in life dark green.

Dorsum (Fig. 1A): Dorsal plate ornamented
with weak panels, and partly with fine canaliculi.
AD and PD separated by interval of approximate-
ly PD length. AD 80m long, 110 “m wide,

/
Wann. ;

C adanal setae

5 spermatheca

Fic. 1. Rhombognathus atuy sp. noy., Female (holotype). A, idiosoma (Ds); B, idiosoma (Vr); C, genitoanal
region. Scale bars=50 um.

weakly convex anteriorly and weakly protruded pore. OC 92 4m long, 50 um wide, extending
posteriorly, ornamented with chevron-shaped anteriorly to level slightly posterior to posterior
areolation posteriorly, without distinct dorsal margin of AD, reaching posteriorly to level slightly

Three New Species of Rhombognathus 519

anterior to insertion of leg III, furnished with two
large corneae and two large polygonal pores, bear-
ing one tiny pore-like structure in the vicinity of
lateral margin, one maze-like subsurface pore
medially to anterior cornea, and two tiny subsur-
face pores near posteromedial margin. Areolation
indistinct. PD 138 4m long, 82 ~m wide, reaching
anteriorly to level slightly posterior to insertion of
leg III, furnished with small subsurface pore at
0.24, tiny areolation at 0.73, and dorsal pore on
posterolateral margin on each side. Costae almost
parallel, scattered with fine canaliculi. Paracosta
lacking.

Chaetotaxy of dorsal region: Setae ds-i on AD,
at 0.49, slightly longer and thicker than the others;
ds-ii each on OC near anteromedial corner at 0.13;
ds-ii1 each on OC near posteromedial margin at
0.61; ds-iv and ds-v on PD at 0.18 and 0.44,
respectively.

Venter (Fig. 1B): Epimeral, genital and anal
plates fused to form a single plate. Ornamentation
indistinct, but very faintly reticulated in part.
Epimeral region furnished with membranous col-
lar anteriorly, several subsurface pores medially,
and elongate subsurface structure between inser-
tions of leg I and leg II on each side, incised
laterally with membranous cuticle with bordering
several subsurface pores at level midway between
insertions of leg II and leg III.

Chaetotaxy of epimeral region: Setae aes-i at
level slightly anterior to insertion of leg II; aes-ii-
lat placed medially at level slightly anterior to
lateral incision; aes-ii-v placed most medially at
level slightly poseterior to lateral incision; aes-ii-
adj located on lateral margins, each consisting of
three setae; pes-ili-lat placed near lateral margins,
at level midway between lateral incision and inser-
tion of leg III; pes-ii-v placed medially, at level of
insertion of leg III; pes-iv located at level slightly
anterior to insertion of leg IV; pes-iii-adj placed
dorsolaterally, each consisting of two thick setae.

Genitoanal region (Fig. 1C): Genital region
slightly incised laterally with membranous cuticle
at level of posterior portion of genital foramen,
furnished with a round subsurface pore, and a
series of polygonal subsurface pores on each side
of genital foramen. Genital foramen 70 um long,
54 um wide, oval, occupying from level slightly

posterior to insertion of leg IV to level anterior to
anal papilla. Genital sclerites band-like, with
membranous wide fringe. Genital acetabula inter-
nal, three pairs. Spermatheca bilobed, extending
to level of insertion of leg IV. Ovipositor placed
inside of genital foramen. Anal papilla somewhat
swollen, placed terminally.

Chaetotaxy of genitoanal region: Five pairs of
long thick filiform perigenital setae located around
genital foramen as arranged in Figure 1C. Sub-
genital setae filiform; two setae on each genital
sclerite, arrange 2-0. Adanal setae one pair, very
fine, placed on anal papilla dorsoproximally.

Gnathosoma (Fig.2A): 724m long, 72 um
wide, gnathosomal-length/idiosomal-length 0.20.
Base, length/width 0.53, slightly expanded lateral-
ly, lacking seta, ornamented with fine punctations
and a few round thin panels. Pharyngeal plate
elongate, furnished with three longitudinal stems
and double row of several minute filamentous
subsurface structures. Anterior margin of tectum
with three acute points. Rostrum approximately
34 wm long, 17 wm wide, nearly lanceolate, not
reaching to level of distal end of palp. Rostral
setae two pairs as follows: Proximal pair long and
robust, at just anterior to swollen point; distal pair
short, at just anterior to proximal pair. Rostral
sulcus short, barely reaching to level of proximal
rostral setae. Chelicera (Fig. 2B) elongate, with
basal segment 68 um long, 24 um wide without
clear ornamentation. Movable digit 16 ~m long,
with 11-12 minute denticles along dorsal edge.
Fixed digit 13 “4m long, extending distally to mid-
level of movable digit. Palp (Fig. 2C), 30 «m long;
P-1, length/width 0.38, short and cylindrical; P-2,
length/width 0.42, longest and robust, very weakly
reticulated, with a seta distidorsally; P-3, length/
width 0.29, short and cylindrical; P-4, length/width
1.75, conical, with three short and thick filiform
setae intermediately, and two appressed blunt
spiniform projections terminally.

Legs (Fig. 3A-D): Length of legs I, I, II, 1V=
240, 236, 236, 248 um, respectively. Ornamenta-
tion indistinct. Each tarsus furnished with claw
fossa, without ventral seta. Lateral claw with
rake-like accessory process, bearing nine to twelve
delicate teeth. Median claw and comb absent.
Carpite short and rod-like. Cavity in claw present.

520 H. ABE

Fic. 2. Rhombognathus atuy sp. nov., Female (holotype). A, gnathosoma (a-Vr, b-Ds); B, chelicera (R); C, palp
(R). Male (allotype). D, genitoanal region; E, spermatophorotype. Scale bars=50 um.

Short seta usually faintly rough, long seta smooth.

Leg chaetotaxy as follows: Trochanters I-IV, 1-
1-1-0; basifemora, 2—3-—2-2; telofemora, 7-7—5-
4; genua, 5—-5—3-4; tibiae, 6-6-5—5. As for large
bipectinate seta: Tibiae I-IV, 2-1-1-1. Tarsus I
(Fig. 4A) with three dorsal setae, one solenidion,
one famulus, and four parambulacral setae (paired
doublet euphathidia). Solenidion long bacilliform
on posterodorsal surface of claw fossa. Famulus
papilliform with fine canaliculus at just ventropro-
ximally to solenidion. Tarsus II (Fig. 4B) with
three dorsal setae, one solenidion, and four para-
mbulacral setae. Solenidion long bacilliform on

posterodorsal surface of claw fossa. Tarsus III
(Fig. 4C) with four dorsal setae, and two parambu-
lacral setae (one single euphathidium on posterior
surface, one bud-shaped proeuphathidium on
anterior surface). Tarsus IV (Fig. 4D) with three
dorsal setae (one long thick filiform seta on basi-
dorsal limb, one fronded seta on claw fossa, one
fine filiform seta on anterodorsal surface), and two
parambulacral setae (one single euphathidium on
posterior surface, one bud-shaped proeuphathi-
dium on anterior surface).

Male (allotype). Idiosoma 360 «m long, 240 «m
wide, gnathosomal-length/idiosomal-length 0.19,

Three New Species of Rhombognathus 521

Fic. 3.
Scale bar=50 um.

resembling the female in essential respects except
for characters of genitoanal region, and chaetotaxy
of tarsus IV.

Genitoanal region (Fig. 2D) furnished with a
round subsurface pore, a series of polygonal sub-
surface pores, and very faint panels on each side of
genital foramen, bearing terminally tufted 11 and
12 perigenital setae as arranged in Figure 2D.
Genital foramen 50 wm long, 24 ~m wide. Sub-

Rhombognathus atuy sp. nov., Female (holotype). A, leg I (R); B, leg II (R); C, leg III (L), D, leg IV (L).

genital setae divided terminally; two setae on each
genital sclerite, arranged 2-0. Genital acetabula
internal, three pairs. Spermatophorotype (Fig.
2E) 62 wm long, 50 4m wide, massive and obo-
vate.

Tarsus IV (Fig. 4E) furnished with three dorsal
setae (one long thick filiform seta on basidorsal
limb, one fronded seta on claw fossa, one delicate
branched seta on anterodorsal surface of claw

522 Three New Species of Rhombognathus

Fic. 4. Rhombognathus atuy sp. nov., Female (holotype). A, tarsus I (R); B, tarsus II (R); C, tarsus III (L); D,
tarsus IV (L). Male (allotype). E, tarsus IV (L). Scale bar=50 um.

fossa), two parambulacral setae (one long plumose
proeuphathidium on posterior surface, one bud-
shaped proeuphathidium on anterior surface).

Immatures: Not collected.

Morphological variation and abnormality: The
number of setae aes-ii-adj and pes-iii-adj on each
side of the idiosoma varies from two to three, and
one to two, respectively. The number of the
perigenital setae on each side of the genital fora-
men varies from 11 to 12 in the male, five to six in
the female. The leg chaetotaxy varies among
specimens as follows: Trochanters I-IV, (0,1)-

(0,1)-1-0; telofemora, 7-(6,7)-5-4; genua, 5-
(4,5)-3-4; tibiae, 6-6-(5,6)-5. One specimen
lacks the dorsoproximal seta on tarsus III.

Etymology: The specific epithet is derived from
“Atuy” which means the sea in the language of the
Ainu (native people in northern Japan).

Distribution: Pacific coast of Hokkaido, north-
ern Japan.

Remarks: Rhombognathus atuy is distinguish-
able from other Rhombognathus species by the
following characters: Separated dorsal plates, PD
with two pairs of setae, arrangements of perigenit-

Three New Species of Rhombognathus 523

al setae in the both sexes (Figs. 1C, 2D), leg
chaetotaxy, and rake-like accessory process.

Among the Rhombognathus species characte-
rized by having five perigenital setae in the female,
Rhombognathus atuy resembles R. caudiculus
Bartsch, 1983 [6] in the arrangement of these
setae. However, R. atuy is easily discernible from
caudiculus by the following characters (corres-
ponding condition in the latter species in parenth-
eses): (1) Two pairs of dorsal setae on PD (one
pair); (2) rostrum lanceolate and shorter than the
length of the base of gnathosoma (elongate, longer
than the length of the base of gnathosoma); (3)
two pairs of basal perigenital setae located near the
posterior site of genital foramen in the male (one
pair); (4) tibiae I-IV with 2-1-1-1 large bipectin-
ate setae (tibiae I-IV with 2-1-1-2); (5) accessory
process rake-like (not developed).

Moreover, Rhombognathus atuy is similar to R.
robustus Bartsch, 1977 [7] in the arrangement of
the perigenital setae in the male. However, R.
atuy differs from robustus in the arrangement of
the perigenital setae in the female (four pairs
anteriorly, one pair posteriorly to genital foramen
in atuy; two pairs anteriorly, three pairs posteriorly
in robustus), in the developed accessory process,
and in having tibiae I-IV with 2-1-1-1 large
bipectinate setae.

Rhombognathus sandwichi Newell, 1984 [5] also
has two pairs of basal perigenital setae in the male
as in R. atuy, although the arrangement of these
setae is not clear because no illustration was given.
However, the two species clearly differ from each
other in the shape and arrangement of the dorsal
plates and leg chaetotaxy.

Rhombognathus dissociatus sp. nov.
(Japanese name: Wakare-kaisoudani, new)
(Figs. 5-8)

Type-series. Holotype: Female, intertidal, on
Sargassum exposed on ledge at low tide, Shamodo-
mari, Oshoro Bay (43°12’, 140°51’E), Hokkaido,
21. 1. 1987, H. Abé coll. Allotype: Male, intertid-
al, on Polysiphonia at 0.3m depth in tide pool
Poromai, Oshoro Bay, 6. iti. 1989, H. Abé coll.
Paratypes: 1 female, 4 males, data same as the
allotype; 1 female, intertidal, among Sargassum

belt in crevice at low tide, Kabuto Rock, Oshoro
Bay, 15.iv.1986, H. Abé coll; 1 female, intertidal,
on Rhodomela at 0.3 m depth in tide pool, Kabuto
Rock, Oshoro Bay, 21. ii. 1987, H. Abé coll.; 1
female, intertidal, on Sargassum exposed on ledge
at low tide, Kikonai (41°42’N, 140°32’E), Hok-
kaido, 16. v. 1987, H. Abé coll.

Female (holotype). Idiosoma 472 ym long, 316
yam wide. Color in life dark green.

Dorsum (Fig. 5A): Dorsal plate ornamented
with clear panels, and partly with fine canaliculi.
AD and PD separated by interval of approximate-
ly PD length. AD 100m long, 116 um wide,
strongly protruded anteriorly, and truncated post-
eriorly, reaching posteriorly to level of insertion of
leg II, ornamented with triangular areolation at
posterior portion, with a large pore near each
lateral margin at 0.44. OC 114 um long, 60 um
wide, extending posteriorly to level slightly anter-
ior to insertion of leg III, furnished with two large
corneae, two large polygonal pores, bearing one
maze-like subsurface pore just posteromedially to
anterior cornea, one pore-like angular structure
near lateral margin at 0.70, and two tiny subsur-
face pores near posteromedial margin. Areolation
not seen. PD 176 um long, 176 um wide, prot-
ruded anteriorly, slightly concave posteriorly,
reaching anteriorly to level of insertion of leg III,
furnished with a large dorsal pore at 0.91, and a
tiny areolation at 0.89 on each side. Costae almost
parallel and scattered with fine canaliculi. Paracos-
ta lacking.

Chaetotaxy of dorsal region: Setae ds-ion AD at
0.54, longer and thicker than the others; ds-ii each
on OC near anteromedial corner at 0.11; ds-iii
each on OC near medial margin at 0.49; ds-iv and
ds-v on PD at 0.23 and 0.59, respectively.

Venter (Fig. 5B): Membranous cuticle clearly
striated. Ventral plates three in number, com-
pletely separated, and entirely ornamented with

porous panels. PE and genital plate fused to form
a single middle plate. AE and middle plate
separated from each other by a strip of membra-
nous cuticle. AE 100m long, 276 um wide,
convex posteriorly, reaching posteriorly to level
about midway between insertions of leg II and leg
III, furnished with wide thin membranous collar
anteriorly, several subsurface pores along post-

524 H. ABE

Fic. 5. Rhombognathus dissociatus sp. nov., Female (holotype). A, idiosoma (Ds); B, idiosoma (Vr); C, genitoanal
region; D, spermatheca. Scale bars=50 «am.

Three New Species of Rhombognathus 525

erior margin, and elongate subsurface structure
between insertions of leg I and leg II on each side.
A round subsurface pore located on striated mem-
branous cuticle medially on each side, at just
posterior to posterior margin of AE. Middle plate
152 wm long, 316 wm wide, concave anteriorly,
convex posteriorly, reaching posteriorly to about
mid-level between insertion of leg IV and the end
of idiosoma, ornamented with several subsurface
pores along boundary between each posterior
epimeral region and genital region. Anal plate 46
um long, 122 um wide, surrounding anal pailla,
completely separated from genital region by a strip

of striated membranous cuticle.

Chaetotaxy of epimeral region: Setae aes-i on
AE, at level slightly anterior to insertion of leg II;
aes-ii-lat on AE, near posterolateral corners; aes-
i-v on AE, near posterior margin; aes-ii-adj
placed on lateral margins (three setae on left side,
four setae on right side); pes-iii-lat on middle
plate, located near lateral margins, at mid-level
between anterior margin of middle plate and inser-
tion of leg III; pes-iii-v on middle plate, at level of
insertion of leg III; pes-iv on middle plate, at level
just anterior to insertion of leg IV; pes-iii-adj
placed on lateral margins of middle plate, each

DPE

Fic. 6. Rhombognathus dissociatus sp. nov., Female (holotype). A, gnathosoma (a-Vr, b-Ds); B, chelicera (R)
[broken basally]; C, palp (R). Male (allotype). D, genitoanal region; E, spermatophorotype. Scale bars=50

ym.

526 H. ABE

consisting of two thick setae.

Genitoanal region (Fig.5C): Genital region
occupying medial portion of middle plate, fur-
nished with a round subsurface pore, and a series
of polygonal subsurface pores on each side of
genital foramen. Genital foramen 88 um long, 48
ym wide, elliptical, located posteromedially on
middle plate. Genital sclerites band-like, extend-
ing posteriorly somewhat beyond posterior margin
of middle plate. Genital acetabula internal, three
pairs. Spermatheca (Fig. 5D) bilobed, not extend-
ing anteriorly from anterior margin of genital
foramen. Ovipositor placed inside of genital fora-
men. Anal papilla placed terminally on anal plate.

Chaetotaxy of genitoanal region: Two pairs of
long filiform perigenital setae located near genital
foramen as arranged in Figure 5C. Subgenital
setae short, filiform; two setae on each genital
sclerite, arranged 2-0. Adanal setae one pair,
robust, plced distidorsally.

Gnathosoma (Fig. 6A): 944m long, 88 ~m
wide, gnathosomal-length/idiosomal-length 0.20.
Base, length/width 0.65, slightly expanded lateral-
ly, lacking seta, ornamented with fine punctations
and several panels. Pharyngeal plate elongate,
furnished with three longitudinal stems, and dou-
ble row of several fine filamentous subsurface
structures. Anterior margin of tectum weakly
convex. Rostrum approximately 37 um long, 20
pm wide, elongate, not reaching to level of distal
end of palp. Rostral setae two pairs as follows:
Proximal pair long and robust, at half level of
rostrum; distal pair at just anterior to proximal
pair, about 2/3 length of proximal pair. Rostral
sulcus short, barely reaching to level of proximal
rostral setae. Chelicera (Fig. 6B) elongate, with
basal segment 86 um long, without clear orna-
mentation. Movable digit 18 ~m long, with 12-13
minute denticles along dorsal edge. Fixed digit 15
ym long, extending distally to about midway of
denticulate dorsal margin of movable digit. Palp
(Fig. 6C) 38 «um long; P-1, length/width 0.33, short
and cylindrical; P-2, legnth/width 0.50, longest and
robust, weakly reticulated with porous panels,
with long thick filiform seta dorsally; P-3, length/
width 0.31, short and cylindrical; P-4, length/width
0.59, conical, with three short and thick filiform
setae intermediately, and two appressed blunt

spiniform projection terminally.

Legs (Fig. 7A-D): Length of legs I, II, III, lV=
244, 246, 230, 236 um, respectively, ornamented
with fine porous panels which are clear only on
telofemora. Each Tarsus furnished with claw
fossa, lacking ventral seta. Lateral claw with tiny
take-like accessory process bearing five to seven
very minute and fine teeth. Median claw and comb
absent. Carpite short and rod-like. Cavity in claw
present. Short seta usually faintly rough, long seta
smooth.

Leg chaetotaxy as follows: Trochanters I-IV, 1—
1-2-0; basifemora, 2—3-—2-2; telofemora, 7—7—5S-—
6; genua, 6-6—4—5; tibiae, 6-6-5-5. As for large
bipectinate seta: Genua I-IV, 1-0—0-0; tibiae, 2—
2-2-2; each one bipectinate seta on tibia II and
tibia III weakly pectinated. Tarsus I (Fig. 8A)
with three dorsal setae, one solenidion, one famu-
lus, and four parambulacral setae (paired doublet
euphathidia). Solenidion long straight bacilliform,
on posterodorsal surface of claw fossa. Famulus
papilliform with fine canaliculus, at just ventrally
to solenidion. Tarsus II (Fig. 8B) with three dorsal
setae, one solenidion, and four parambulacral
setae. Solenidion long bacilliform on posterodor-
sal surface of claw fossa. Tarsus III (Fig. 8C) with
four dorsal setae, and two parambulacral setae
(one single filiform proeuphathidium on posterior
surface, one bud-shaped proeuphathidium on
anterior surface). Tarsus IV (Fig. 8D) with three
dorsal setae (one lone thick filiform seta on basi-
dorsal limb, one long serrated seta on claw fossa,
one straight fine filiform seta on anterodorsl sur-
face), two parambulacral setae (one straight fine
filiform proeuphathidium on posterior surface, one
bud-shaped proeuphathidium on anterior surface).

Male (allotype). Idiosoma 400 um long, 240 ~m
wide, gnathosomal-length/idiosomal-length 0.22,
resembling the female in essential respects except
for characters of body size, genitoanal region, and
chaetotaxy of tarsus IV.

Body size somewhat smaller than that in the
female.

Genitoanal region (Fig. 6D) furnished with a
series of polygonal subsurface pores, and terminal-
ly tufted 15 and 17 perigenital setae as arranged in
Figure 6 D. Genital foramen 72 um long, 26 «m
wide. Subgenital setae short, filiform; two setae at

Three New Species of Rhombognathus 527

Fic. 7. Rhombognathus dissociatus sp. nov., Female (holotype). A, leg I (L); B, leg II (L); C, leg III (L); D, leg VI

(L).

mid-level on each genital sclerite, arranged 2-0.
Genital acetabula internal, three pairs. Anal plate
not so clearly separated from middle plate as that
in the female. Spermatophorotype (Fig. 6E) 88
ym long, 84 ~m wide, very massive.

Tarsus IV (Fig. 8E) furnished with three dorsal
setae (one long thick filiform seta on basidorsal
limb, one fronded seta on claw fossa, one bran-
ched seta on anterodorsal surface of claw fossa),
two parambulacral setae (one long branched

528 H. ABE

Fic. 8. Rhombognathus dissociatus sp. nov., Female (holotype). A, tarsus I (L); B, tarsus II (L); C, tarsus III (L);
D, tarsus IV (L). Male (allotype). E, tarsus IV (R). Scale bar=50 um.

proeuphathidium on posterior surface, one bud-
shaped proeuphathidium on anterior surface).

Immatures: Not collected.

Morphological variation and abnormality: The
holotype female specimen has four aes-ii-adj setae
on right lateral margin of AE. However, all other
specimens examined have three aes-ii-adj (the
anteriormost seta shortest, the posteriormost
longest) on each lateral side of AE. One specimen
has two maze-like subsurface pores on left OC,
and no ds-iii. The number of the perigenital setae

on each side of the genital foramen varies from 14
to 25 in the male; this number varies not only
according to specimens, but also within one speci-
men. The leg chaetotaxy varies as follows:
Trochanters I-IV, (6,7)-(7,8)-(5,6)-(4,5,6,7);
genua, (5,6)—(5,6)—(3,4)-(4,5,6); tibiae, 6-6-5-
(5,6). One specimen has only one large bipectin-
ate seta on tibia III, and tibia IV, respectively.

Etymology: The specific epithet is derived from
“the dissociated ventral plates”.

Distribution: The Japan Sea coast of Hokkaido.

Three New Species of Rhombognathus 529

Remarks: This new species obviously belongs to
the genus Rhombognathus on the grounds that (1)
the genital foramen is placed ventrally (not termi-
nally) and guarded by band-like (not cusp-like)
genital sclerites, (2) each ocular plate has two
setae, and (3) all legs have two claws. All the
hitherto named species of Rhombognathus have
one, two, or five ventral plates in the adult, and
have more than three pairs of perigenital setae in
the female. However, the species is unique in the
following characters: (1) the venter is covered with
three ventral plates (anterior epimeral plate, a
middle plate consisting of posterior epimeral plates
and a genital plate, and anal plate); (2) two pairs of
perigenital setae in the female as shown in Figure 5
C; (3) the leg chaetotaxy is distinctive in having
trochanters I-IV with 1—-1—2-0 setae.

Rhombognathus ezoensis sp. nov.
(Japanese name: Ezo-kaisoudani, new)
(Figs. 9-12)

Type-series. Holotype: Female, intertidal, on
Sargassum on boulder at 0.2 m depth at high tide,
Shamodomari, Oshoro Bay, Hokkaido, 23. vi.
1987, H. Abé coll. Allotype: Male, intertidal, on
Sargassum on boulders at 0.3 m depth at low tide,
Shamodomari, Oshoro Bay, 21. ii. 1987, H. Abé
coll. Paratypes: 1 female, 2 tritonymphs, 2
deutonymphs, data same as the holotype; 1 male, 2
females intertidal, among Sargassum belt at 0.5 m
depth at low tide, Ebisu Rock, Oshoro Bay, 15. iv.
1986, H. Abé coll.; 2 females, intertidal, on Sar-
gassum at 0.3 m depth in tide pool, Kabuto Rock,
Oshoro Bay, 21. ii. 1987, H. Abé coll.; 1 male,
intertidal, on Sargassum at 0.5m depth in tide
pool, Usujiri, Hokkaido, 12. vi. 1986, H. Abé
coll.; 1 female, intertidal, on Sargassum exposed
on ledge at low tide, Kikonai, Hokkaido, 16. v.
1987, H. Abé coll.; 2 males, 2 females, intertidal,
on Corallina at 0.1 m depth in tide pool, Mitsuishi,
Hokkaido, 8. xii. 1988, H. Abé coll.

Female (holotype). Idiosoma 388 “m long, 256
ym wide. Color in life dark green with a fine dorsal
semitransparent line longitudinally.

Dorsum (Fig.9A): Dorsal plate uniformly
ornamented with clear panels, and partly with fine
canaliculi. Ad and PD separated by interval of

approximately two times as long as AD. AD 80
ym long, 100 ~m wide, almost truncated anteriorly
and posteriorly, ornamented with fine areolation
at posterior portion, without clear dorsal pore.
OC 96 «um long, 52 um wide, extending anteriorly
to level slightly posterior to posterior margin of
AD, reaching posteriorly to level slightly anterior
to insertion of leg III, furnished with two large
corneae, two large polygonal pores, bearing one
angular pore-like structure near lateral margin,
one maze-like subsurface pore medially to cor-
neae, and three tiny subsurface pores near post-
eromedial margin. Areolation not seen. PD 164
ym long, 120 ~m wide, extending anteriorly to
level of insertion of leg IV, furnished with a small
subsurface pore at anterolateral corner, and a
dorsal pore on posterolateral margin on each side.
Areolation not clear. Costae almost parallel and
scattered with fine canaliculi. Paracosta lacking.

Chaetotaxy of dorsal region: Setae ds-ion AD at
0.43, longer and thicker than the others; ds-ii each
on OC near anterior margin at 0.08; ds-i1i each on
OC near medial margin at 0.52; ds-iv and ds-v on
PD at 0.14 and 0.47, respectively.

Venter (Fig. 9B): Epimeral, genital, and anal
plates fused to form a single plate which is entirely
reticulated with faint panels. Epimeral region
furnished with membranous collar anteriorly,
several subsurface pores medially, and elongate
subsurface structure between insertions of leg I
and leg II on each side, incised laterally with
membranous cuticle with bordering several subsur-
face pores at mid-level between insertions of leg II
and leg III.

Chaetotaxy of epimeral region: Setae aes-i at
level slightly anterior to insertion of leg II; aes-ii-
lat placed medially at level slightly anterior to
lateral incision; aes-ii-v located medially at level of
anterior margin of lateral incision; aes-ii-adj
placed near lateral margins, each consisting of
three setae on left, whereas two setae on right;
pes-ii-lat placed near lateral margins, at mid-level
between lateral incision and insertion of leg II];
pes-ii-v placed medially at level of insertion of leg
III; pes-iv placed slightly anterior to insertion of
leg IV; pes-iti-adj located dorsolaterally, each con-
sisting of two setae.

Genitoanal region (Fig.9C): Genital region

530 H. ABE

a
Xi
Bo

SASHO SK
LSSESSS,
20D eXsIu\
(So) SS

6)
QS.
1S

oO
SS

Oo
0
oF

0
le)
te}

ie)
J]

72)
S240
ie)
\ 4

{)
OO

~~]
(9)
©)
LiKe)

Fic. 9. Rhombognathus ezoensis sp. nov., Female (holotype). A, idiosoma (Ds); B, idiosoma (Vr); C, genitoanal
region. Scale bars=50 um.

slightly incised laterally with membranous cuticle
at level of posterior portion of genital foramen,
furnished with a round subsurface pore, and a

series of several subsurface pores on each side of
genital foramen. Genital foramen 66 «m long, 40
um wide, subelliptical, reaching to level slightly

Three New Species of Rhombognathus 531

Fic. 10. Rhombognathus ezoensis sp. nov., Female (holotype). A, gnathosoma (a-Vr, b-Ds); B, chelicera (R); C,
palp (R). Male (allotype). D, genitoanal region; E, spermatophorotype. Tritonymph (paratype). F, genitoanal
region. Deutonymph (paratype). G, genitoanal region. Scale bars=50 um.

532 H. ABE

anterior to anal papilla, furnished with band-like
genital sclerites and three pairs of internal genital
acetabula. Spermatheca bilobed, extending to
level of insertion of leg IV. Ovipositor placed
inside of genital foramen. Anal papilla placed
terminally.

Chaetotaxy of genitoanal region: Four pairs of
short filiform perigenital setae located in the vicin-
ity of genital foramen as arranged in Figure 9 C.
Subgenital setae filiform; two setae on each genital
sclerite, arranged 2-0. Adanal setae one pair,
placed on anal papilla dorsoproximally.

Gnathosoma (Fig. 10A): 76m long, 70 ~m
wide, gnathosomal-length/idiosomal-length 0.20.
Base, length/width 0.51, slightly expanded lateral-
ly, lacking seta, ornamented with fine punctations,
and a few round panels on ventroproximal site.
Pharyngeal plate elongate, furnished with three
longitudinal stems, and double row of several
minute filamentous subsurface structures. Anter-
ior margin of tectum weakly convex and slightly
waved. Rostrum 40 um long, 22 um wide, nearly
lanceolate, not reaching to level of distal end of
palp. Rostral setae two pairs as follows: Proximal
pair long and thick at just anterior to swollen part
of rostrum; distal pair short, at just anterior to
proximal pair. Rostral sulcus short, barely
reaching to level of proximal rostral setae. Che-
licera (Fig. 10B) elongate, with basal segment 76
ym long, 24 um wide, without clear ornamenta-
tion. Movable digit 15 ~m long, with 11-12 minute
denticles along dorsal edge. Fixed digit 13 ~m
long, extending slightly proximally to distal end of
movable digit. Palp (Fig. 10C) 28 ~m long; P-1,
length/width 0.38, short and cylindrical; P-2,
length/width 0.40, longest and robust, very faintly
reticulated, with a long filiform seta distidorsally;
P-3, length/width 0.25, short and cylindrical; P-4,
length/width 1.75, conical, with three short and
thick filiform setae intermediately, and two
appressed blunt spiniform projection terminally.

Legs (Fig. 11A-D): Length of legs I, II, III, IV
=208, 208, 214, 214 um, respectively. Orna-
mentation indistinct. Each tarsus furnished with
claw fossa, without ventral seta. Lateral claw with
rake-like accessory process bearing six to eight
delicate teeth. Median claw and comb absent.

Carpite short and rod-like. Cavity in claw present.
Short seta usually faintly rough, long seta smooth.
Leg chaetotaxy as follows: Trochanters I-IV, 1-
1-1-0; basifemora, 2—3-2-2; telofemora, 7-7—5-—
4; genua, 6-6-—3-4; tibiae, 7-7-5-6. As for large
bipectinate seta: Genua I-IV, 1-0-0-1; tibiae, 2-
1-1-1. Tarsus I (Fig. 12A) with three dorsal setae,
one solenidion, one famulus, and four parambu-
lacral setae (paired doublet euphathidia). Soleni-
dion long bacilliform, on posterodorsal surface of
claw fossa. Famulus papilliform with fine canalicu-
lus, at just ventrally to solenidion. Tarsus II (Fig.
12B) with three dorsal setae, one solenidion, and
four parambulacral setae. Solenidion long bacilli-
form on posterodorsal surface of claw fossa. Tar-
sus III (Fig. 12C) with four dorsal setae, and two
parambulacral setae (one single euphathidium on
posterior surface, one bud-shaped proeuphathi-
dium on anterior surface). Tarsus IV (Fig. 12D)
with three dorsal setae (one long robust filiform
seta on basidorsal limb, one fronded seta on claw
fossa, one fine filiform seta on anterodorsal sur-
face), two parambulacral setae (one fine filiform
proeuphathidium on posterior surface, one bud-
shaped proeuphathidium on anterior surface).

Male (allotype). Idiosoma 332 um long, 212 ~m
wide, gnathosomal-length/idiosomal-length 0.22;
resembling the female in essential respects except
for characters of body size, genitoanal region, and
chaetotaxy of tarsus IV.

Body size somewhat smaller than that in the
female.

Genitoanal region (Fig. 10D) furnished with a
round subsurface pore, and a series of polygonal
subsurface pores on each side of genital foramen,
bearing terminally tufted 13 and 14 perigenital
setae as arranged in Figure 10D. Genital foramen
46 um long, 18 um wide. Subgenital setae divided
terminally; two setae on each genital sclerite,
arranged 2-0. Genital acetabula internal, three
pairs. Spermatophorotype (Fig. 10E) 48 «m long,
62 «m wide, massive and rhombic.

Tarsus IV (Fig. 12E) with three dorsal setae
(one long robust filiform setae on basidorsal limb,
one fronded seta on claw fossa, one branched seta
on anterodorsal surface of claw fossa), two para-
mbulacral setae (one long plumose proeuphathi-
dium on posterior surface, one bud-shaped

Three New Species of Rhombognathus 533

A-D

Fic. 11. Rhombognathus ezoensis sp. nov., Female (holotype). A, leg I (L); B, leg II (L); C, leg Il (L); D, leg VI

(L). Scale bar=50 ym.

proeuphathidium on anterior surface).
Tritonymph (paratype). Idiosoma 332 um long,
196 um wide, gnathosomal-length/idiosomal-
length 0.20.
Dorsum: AD concave posteriorly. PD convex
anteriorly. AD and PD separated by interval

about two times as long as PD. OC with two
subsurface pores at posteromedial margin.
Venter: AE furnished with a number of subsur-
face pores medially as well as along posterior
margin, with two aes-ii-adj on each lateral margin.
PE furnished with several subsurface pores along

534 H. ABE

Fic. 12. Rhombognathus ezoensis sp. nov., Female (holotype). A, tarsus I (L); B, tarsus II (L); C, tarsus II (L); D,
tarsus IV (L). Male (allotype). E, tarsus ITV (R). Scale bar=50 ym.

anteroventral margin. A small subsurface pore
placed on membranous cuticle medially on each
side, at level of anterior margin of PE.

Genitoanal region (Fig. 10F): Genital plate 64
ym long, 52 um wide, bluntly protruded anterior-
ly, nearly truncated posteriorly, furnished with two
short setae of which one is placed at 0.31, and
another at 0.84 on each side. bearing a tiny subsur-
face pore at 0.43 on each lateral margin. Primor-
dial genital slit occupied from 0.69 to 0.84, with
three pairs of internal genital acetabula. Subgenit-
al seta absent. Three minute subsurface pores
placed on membranous cuticle slightly anterior to
anterior margin of genital plate. Anal plate small,
nearly truncated anteriorly.

Legs: Leg chaetotaxy of telofemora I-IV, 6-6-
4-4; tibiae, 6-6-5-5. Tarsus with less fronded
dorsal seta.

Deutonymph (paratype). Idiosoma 264 um
long, 168 «m wide, gnathosomal-length/idiosomal-
length 0.19.

Dorsum: AD protruded posteriorly. PD small.
Costa indistinct.

Venter: AE with one aes-ii-adj on each lateral
margin. Left PE with two pes-iii-adj of which one
very minute, whereas right PE with only one
pes-iii-adj.

Genitoanal region (Fig. 10G): Genital plate 38
um long, 30 ~m wide, furnished with primordial
genital slit, with two paris of internal genital

Three New Species of Rhombognathus 535

acetabula. Genital seta absent.

Legs: leg chaetotaxy of basifemora I-IV, 2—3-2-
1; telofemora, 4—4—-3-2; genua, 5—5-3-4; tibiae,
5-6-5—5S. Genu IV with one large but less pectin-
ate seta.

Protonymph and larva: Not collected.

Morphological variation and abnormality: The
number of the subsurface pores near the poserior
margin of OC varies from two to four. The
number of setae aes-li-adj and pes-iii-adj on each
side of the idiosoma in the adult varies from two to
three, and one to two, respectively. The number
of the perigenital setae on each side of the genital
foramen varies from three to five in the female,
and 12 to 14 in the male; this number varies
according to specimens and even to each side of
the genital foramen in one specimen. The leg
chaetotaxy of the adult specimens varies as fol-
lows: Trochanters I-IV, (0,1)-(0,1)-1-0;
basifemora, 2—(2,3)—2-(1.2); telofemora, (6,7)-
7-(4,5,6)—(3,4,5); genua, (5,6)—(5,6)-3-4; tibiae,
(6,7)—(6,7)—(5,6)—(5,6). One specimen lacks the
dorsoproximal seta on tarsus III.

Etymology: The specific epithet is derived from
“Ezo”, the old Japanese name for Hokkaido.

Distribution: Widespread on the coast of Hok-
kaido.

Remarks: This new species is distinguishable
from other Rhombognathus species by the follow-
ing characters: Separated dorsal plates, two pairs
of setae on PD, four pairs perigenital setae with
characteristic arrangement (Fig. 9C) in the female,
leg chaetotaxy and rake-like accessory process on
lateral claw.

Rhombognathus ezoensis closely resembles R.
reticulatus Krantz, 1976 [8] in the conformation of
the dorsal plates and the idiosomal chaetotaxy of
the dorsal and epimeral regions. However, R.
ezoensis is distinctive from reticulatus in the follow-

ing characters (corresponding condition in the
latter species in parentheses): (1) Perigenital
filiform setae short (long); (2) four pairs of
perigenital setae (three pairs); (3) leg chaetotaxy
of telofemora I-IV, 7-7-5-4; genua, 6-6-3-4
(telofemora I-IV, 6-6-4—4; genua, 6—-6-3-3); (4)
genital plate completely separated from anal plate
in the deutonymph (fused to form a single geni-
toanal plate).

ACKNOWLEDGMENTS

The author wishes to express his sincere thanks to Dr.
H. Katakura (Hokkaido University) for his critical revi-
sion of the manuscript. Cordial thanks are also due to
Professor G. W. Krantz (Oregon State University) for
his valuable criticisms and suggestions in improving the
manuscript.

REFERENCES

1 Trouessart, E. (1888) Note sur les Acariens marins
recueillis par M. Giard au laboratoire maritime de
Wimereux. C. R. Acad. Sci., 107: 753-755.

2 Sokolov, I. (1952) Halacarae. Fauna S.S.S.R., 5(5):
1-201.

3 Newell, I. (1947) A systematic and ecological study
of the Halacaridae of eastern North America. Bull.
Bingham Oceanogr. Coll., 10 (3): 1-232.

4 Newell, I. (1967) Abyssal Halacaridae (Acari) from
the southeast Pacific. Pac. Insects, 9: 693-708.

5 Newell, I. (1984) Antarctic Halacaroidea. Antarc.
Res. Series, 40: 1-284.

6 Bartsch, I. (1983) Zur Halacaridenfauna der Philip-
pinen Beschreibung von funf Arten der Gattung
Rhombongathus (Acari, Halacaridae). Entomol.
Mitt. zool. Mus. Hamburg, 7: 399-416.

7 Bartsch, I. (1977) Interstitielle Fauna von Galapagos
XX. MHalacaridae (Acari). Mikrofauna Meeres-
bodens, 65: 1-108.

8 Krantz, G. (1976) Arenicolous Halacaridae from the
intertidal zone of Schooner Creek, Oregon (Acari:
Prostigmata). Acarologia, 18: 251-258.

inh Vis coe e Ie wake? wit eoulT. aor iM

OD grog ‘1 ro) rises J Re
by) Asseecdiers Ho ai soiree yatta!
OF). gnel) bode iiise -onohiih

ith Oot) GiSe late
gions fal. we

& }.0-8 VI-L weenie
i rite: stele latina

innryninwa sal ae

i q otitkea

ii¥) aoe
iy yon
‘ 5 ‘ a Fa
mks
7
tooeieyt
}
,
,
’
i S

- 7 a —

; 3
“insade utva lainey
Stang .VieLasetnstiond 1 yeitolaall
seit (6-4 -P * contng oft .
ibereey wil dined sgn wht a! fob

oS =a oP
Lepaottos 1014 ipa i
9, ‘ vo,
onl wa) err etretie Verh tn Ty

Hewes of? shoe 29g. erie
yt] agt of owt mnont.ad rei
Oo (i-abeeq bad bavbis “St
olewt morteshsy dubs say gen

THt mutt viovifiouest a 4
intingy GY bo obi duke nd onsen
Auer! sat ni Sv 9} sont sae

i7 rernace otl? “shire Sd
10 whir des ul cove, bel ane

ul dT arene ee
iv ee Money ancenia see dla we
f (fityaL lp ET bn
somos: (°c
YG eA: *) (Aut y
7) edonl negniooge 4nQ (0
Lil come nee
er Syria 2: eNligs +a oeeta bi
ibloidbioh yc) smn nee
ob Jo tees witan bes qpabiie
heme te Ese Wee
“it o) =pcerp sie
roliiq¢lezioh hefrieg a
jst’ tehiregire) nthg eer j
al) nigga
‘a ett {
ehofie

7
oy! pene 2

‘

' er) 4 i nig i oi

M } }
wer) 2e; 77 Cet) BS) BAS
4
soe Ea APA labora +
i load WE tad ett SVig

ZOOLOGICAL SCIENCE 7: 537-540 (1990)

[COMMUNICATION]

© 1990 Zoological Society of Japan

Nociception in Crocodiles: Capsaicin Instillation, Formalin
and Hot Plate Tests

T. I. Kanui, K. Hore! and J. O. MIARon

Department of Animal Physiology, University of Nairobi, Nairobi, Kenya, and
"Department of Physiology, University of Bergen, Bergen, Norway

ABSTRACT—Three tests of nociception were adapted
for the use in crocodiles (47.0-65.2 cm long). In the
capsaicin instillation test, capsaicin in concentrations of
10~° to 10° g/ml instilled in the eye induced concentra-
tion related protective reactions which were counted. In
the formalin test, 150 ul of 5% formalin was injected
subcutaneously in the fore paw, and the time spent
“lifting the foot” and “not using the foot” was recorded.
In the hot plate test, the plate temperature was set at
55°C and the latency until the following behavioural
categories occurred was recorded: “lifting toes”, “lifting
foot”, and “attempt to escape”. This test could be
repeated with similar results after an interval of 60 min.

It was concluded that the crocodile has a well
developed nociceptive system, and it may be possible to
study the function of this system using these modifica-
tions of well known tests of nociception.

INTRODUCTION

The physiology and anatomy of the nervous
system of the crocodile is incompletely known.
The embryonic development of crocodilian ner-
vous system has been investigated [1]. Pain
perception, and the regulation of pain sensitivity,
are basic functions of the central nervous system,
with a fundamental biological importance. Knowl-
edge of the physiology of this sensory system and
its regulation may be important for the under-
standing of the physiology and the behaviour of
the animal.

MATERIALS AND METHODS

The crocodiles (Crocodylus niloticus africana)

Accepted July 21, 1989
Received April 5, 1989

weighed 231-1125 g, they were 47.0-65.2 cm long,
and the abdominal circumference was 9-19 cm.
They were estimated to be 4 to 10 months old. The
animals were obtained from Mombasa (Kenya)
and were transported to Nairobi for experimenta-
tion.

The animals were kept in a quiet room in an
environment resembling their natural environ-
ment. Water was available, as well as stones and
sand to lie on. The water temperature was 30.9+
0.2°C and the ambient temperature was 29.5+
3.9°C. The dry surface temperature of the croco-
diles was 31.3+0.2°C. Heating bulbs (250 W)
were used for maintaining the temperature and
evaporating water to maintain the humidity.

The experiments were started a month after the
start of the acclimatization. During this period,
the animals were handled daily.

Capsaicin instillation

Capsaicin (98%) was supplied by Sigma, U.S.A.
A stock solution of capsaicin (1%) was prepared
using a vehicle: 10% ethanol, 10% tween 80 and
80% of 0.9% NaCl. Further dilutions were made
with 0.9% NaCl.

Capsaicin in concentrations between 10~? g/ml
and 10 * g/ml were instilled into the eye according
to Gamse et al. [2]. Two drops were instilled. The
number of protective reactions such as blinking,
wiping, blepharospasm, rubbing, head shaking and
eyeball movements were counted using a manual
counter, during the ensuing period. Blinking and
blepharospasm were similar and were therefore
scored together. The latency to the first reaction as
well as the time-course of the reactions were
determined.

538 T. I. Kanu, K. HoLeE AnD J. O. MIARON

In cotrol experiments, the vehicle was instilled
in the contralateral eye. The same eye was used
more than once with at least 45 min intervals.
Twelve crocodiles were used for the capsaicin
experiment. Wire-mesh cages (60 40X27 cm)
were used as observation cages.

Formalin test

The formalin test was adapted from that de-
scribed in rats and mice [3, 4]. A volume of 150 pl
5% formalin was injected subcutaneously in the
left fore paw. Two categories of pain-induced
behavior were scored: lifing the foot, and com-
pletely not using the foot. The crocodiles com-
pletely lifted the whole fore leg, from the surface
of the observation cages, in the latter behaviour.
The time spent in each behavioural category was
recorded. Injection of 150 ul 0.9% NaCl in the
right fore paw was used as control. Five crocodiles
were used. The sequence of saline or formalin
injection was random, with an interval of at least
14 days.

The same observation cages were used as in the
capsaicin instillation experiment.

Hotplate test

The apparatus used was an IITC Inc Model 35 D
Analgesiameter, the temperature was set at 55°C.

Before testing on the hot plate, the animal was
placed in a wiremesh cage for 60 min for drying of
the skin. Three categories of pain related be-
haviour were scored: “lifting toes from the plate”,
“lifting foot”, and “attempt to escape”. The
latency until the behaviour occurred was recorded.

Testing was repeated twice in the same animal
with an interval of 60 min and the mean for the
three trials was used as the response latency for the
animal.

RESULTS

Instillation of capsaicin into the eye

Instillation of capsaicin into the eye produced a
number of protective reactions. Blepharospasm
was the most common response elicited. At the
threshold dose (10~° g/ml) blepharospasm per-
sisted for 3.5+0.5 min (mean+S.E.M). At a
concentration of 10~*g/ml, the highest dose
studied, the duration of blepharospasm was 18.7 +

1.0 min. A few trials at a concentration of 10-7
g/ml capsaicin resulted in closure of the eye for
most of the observation period, and experiments
using this concentration were discontinued.

The protective behaviour occurred immediately.
There was a distinct dose-response relationship
during the first 3 min when capsaicin was instilled
at concentrations between 10~? and 10° g/ml
(Fig. 1). In this period the vehicle elicited 2.0+0.3
protective reactions. The threshold concentration
(10~° g/ml) produced 5.5+1.0, while the highest
concentration (10~* g/ml) used produced 34.7+
5.0 protective reactions. Head shaking was
observed 6 times: like wiping and eye-ball rolling,
head shaking occurred at the highest concentration
used (10-*g/ml). These behaviours occurred
during the first 2min. The concentration (10-°
g/ml) produced 2.1+0.8 wipings and 10.5+5.2
eyeball movements. The head was raised and
vigorously shaken from side to side. The ipsilateral
hind paw was raised forwards and used to wipe the
eye. Rubbing was scored together with wiping
because of their similarity. The eyeball moved up
and down its socket during rolling.

50
gy
<¢ 40
o
§
> e
£ 30
5 fp
2 20 ,
=
=)
z Fa
10 used
o—
0
9 8 7 6 5 4 3
Vehicle -
-log (capsaicin conc.) g/ml
Fic. 1. Number of protective movements in the first 3

min after instillation of capsaicin (10~°-10~ * g/ml)
or vehicle. Mean+S.E.M. n=13 for each concen-
tration.

For all the concentrations of capsaicin used,
most protective reactions occurred during the first
10min (Fig. 2). No pain behaviour could be
observed after the 40 min observation period. Ata

Nociception in Crocodiles

concentration of 10~* g/ml, repeated [5] instilla-
tion of capsaicin elicited 44.7+4.5, 51.2+6.9, 51.8
+6.7, 50.3+9.2 and 44.5+5.3 protective reac-
tions, during the first 3 min. No significant change
in the response was observed at this and other
concentrations used.

-log (capsaicin conc.)

Number of movements

S45 eee
area xa renin iene pian omeme

0-5 10-15

20-25 30-35

Minutes after instillation
Fic. 2. Time-course of mean number of protective
movements after instillation of capsaicin 10~°-10-°

g/ml. Blocks of 5min. n=13 for each concentra-
tion.

Not using foot

200

150

100

Lifting foot

Response (sec)

50

5-10

5-10 0-5

Minutes after injection

Fic. 3. Time spent “lifting the foot” and “not using the
foot” in the formalin test. Blocks of 5 min observa-
tion periods. Mean+S.E.M. n=5.

The Formalin test

The formalin injection immediately induced
pain related behavioural responses. “Lifting the
foot” was observed only in the first 5 min period

539

after injection, while “not using the foot” was
observed both in the first and the second 5 min
period (Fig. 3). No pain behaviour was observed
after 10 min.

Saline injections did not induce this pain be-
haviour.
The hot plate test

The response latencies for “lifting toes from the
plate”, “lifting foot”, and “attempt to escape” are
shown in Figure 4. There were only small and
inconsistent differences in the results of the first,
the second and the third trial, and no statistically
significant difference when the response latencies
for the second or the third trial were compared to
the first trial (P<0.05 for both comparisons for all
three latencies, t-test).

150
sr)
CY)
2
o
2 100
2
&
®
e
¢ 50
a
1)
®
ac
)

Lifting Lifting Escape

toes foot

Fic. 4. Response latencies for “lifting toes”, “lifting
foot” and “attempt to escape” in the hot plate test.
Mean+S.E.M. n=5 for animals (n=15 for trials).

DISCUSSION

Our observations indicate that crocodiles are
very sensitive to capsaicin unlike amphibians [5]
and birds [5-7]. In crocodiles, capsaicin instilla-
tion elicited behavioural responses similar to those
reported in mammals [5, 8]. The threshold
concentration of capsaicin (107° g/ml) that evoked
protective reactions was considerably lower than
that (10~° g/ml) evoking responses in guinea pigs
and rats [5, 8]. Surprisingly, although crocodiles
were sensitive to capsaicin, repeated instillation of
capsaicin did not produce any desensitizing effect.

540 T. I. Kanut, K. HoLe anp J. O. Miaron

Possibly, capsaicin in crocodiles does not cause a
depletion of substance P which may be associated
with desensitization [9, 10].

The formalin test in mammals (rat, mice, cat,
monkey) elicits both acute and long-lasting pain [3,
4, 11]. The main behavioural responses in these
animals are licking, scratching and not using the
injected limb. Crocodiles were sensitive to 5%
formalin and responded by lifting the foot and not
using the foot. The crocodiles did not show a
second phase of pain as reported in mammals (3, 4,
11]. The second phase in probably induced by
inflammation [3, 4], and requires a stronger
stimulus to be elicited than the first phase [Rosland
et al., unpubl. data]. The inflammatory stimulus is
strong enough to induce licking and scratching in
rats and mice, however, it may not be strong
enough to induce “lifting the foot” and “not using
the foot” in crocodiles.

In the hot plate test, the first response observed
in the crocodile was the lifting of one or more toes.
It seems that the latency until this response occurs,
may be used as a measure of the pain threshold,
and may be the response in the crocodile that is the
closest to the hind paw lick response in mice and
rats [12-15]. The escape response presumably
occurs at a stimulus well above pain threshold.
The skin temperature may be an important vari-
able in test of nociception applying heat as the
stimulus [16]. It is probably important therefore
that the ambient temperature is kept constant and
that the skin of the feet of the crocodiles is dried
well before testing.

All the three tests described here seem to be
suitable as tests of nociception in the crocodile.
Comparing the three tests, the hot plate test has
some advantages as in this test both threshold and
suprathreshold responses can be easily and reliably
scored, and the test may be repeated even after a
rather short interval. We have found that the test
is sensitive to morphine and other analgesic drugs
[T. I. Kanui et al., unpubl. data]. Since the
stimulus as well as the response are different in the
three tests, each may be useful for studying
different aspects of the regulation of nociception.
The tests may also be used to study the influence of
drugs on this function of the nervous system in the
crocodile.

It may be concluded that the crocodile has a well
developed nociceptive system, and it may be
possible to study the function of this system using
modifications of well known tests of nociception.
The nociceptive system is necessary for the surviv-
al of the crocodile in the wild. It is responsible for
eliciting defense and protective reactions, when
crocodiles are attacked by other wild animals or
humans.

ACKNOWLEDGMENTS

This project was supported by the NORAD KEN 046
Project.

REFERENCES

1 Ferguson, M. W. J. (1985) In “Biology of the
Reptilia”. Ed. by C. Gans, F. Billet and P. F. A.
Maderson, John Wiley and Sons, New York, pp.
329-492.

2 Gamse, R., Holzer, P. and Lembeck, F. (1980)
Brit. J. Pharmacol., 68: 207-313.

3 Hunskaar, S., Fasmer, O. B. and Hole, K. (1985) J.
Neurosc. Meth., 14: 69-76.

4 Dubuisson, D. and Dennis, S. G. (1977) Pain, 4:
161-174.

5 Szolcsanyi, J., Sann, H. and Pierau, F-K. (1986)
Pain, 27: 247-260.

6 Mason, R. J. and Maruniak, J. A. (1983) Pharma-
col. Biochem. Behav., 19: 857-862.

7 Geisthdvel, E. and Simon, E. (1984) In “Thermal
Physiology”. Ed. by J. R. S. Hales, Raven Press,
New York, pp. 29-32.

8 Makara, G. B. (1970) Acta. Physiol. Acad. Sci.
Hung., 38: 393-399.

9 Jessel, T. M., Iversen, L. L. and Cuello, A. C.
(1978) Brain Res., 152: 183-188.

10 Gamse, R., Leeman, S. E., Holzer, P. and Lem-
beck, F. (1981) Naunyn-Schmiedeberg’s Arch. Exp.
Path. Pharmacol., 317: 140-148.

11 Alreja, M., Mutalik, P., Nayar, U. and Manchada,
S. K. (1984) Pain, 20: 97-105.

12. Woolfe, G. and MacDonald, A. D. (1944) J.
Pharmacol. Exp. Therap., 80: 300-307.

13. Eddy, N. B., Touchberry, C. F. and Lieberman, J.
E. (1950) J. Pharmacol. Exp. Therap. , 98: 121-137.

14. Kitchen, I. and Crowder, M. (1985) J. Pharmacol.
Meth., 13: 1-7.

1S. Ankier, S. I. (1974) Eur. J. Pharmacol., 27: 1-4.

16 Tyjlosen, A., Berge, O-G., Eide, P. K., Broch, O. J.
and Hole, K. (1988) Pain, 33: 225-231.

ZOOLOGICAL SCIENCE 7: 541-545 (1990)

[COMMUNICATION]

© 1990 Zoological Society of Japan

Critical Period of Induction by Tamoxifen of Genital Organ
Abnormalities in Male Mice

SATOKO IRISAWA, TAISEN Icucut! and Nosoru TaKasucr!

Laboratory of Biology, Tokyo Kasei Gakuin University, Machida 194-02, and
' Department of Biology, Yokohama City University,
Kanazawa-ku, Yokohama 236, Japan

ABSTRACT—CS57BL/Tw male mice given 5 daily injec-
tions of 100 ~g tamoxifen (Tx) starting on the day of
birth (0 day) were examined at 5, 10, 15, 20 and 30 days
of age. Body and organ weights and diameter of
seminiferous tubules in Tx mice were significantly smal-
ler than those of the age-matched controls. Sperma-
togenesis was found in 30-day-old control mice, but was
completely suppressed in Tx mice at the same age. In
addition, 60-day-old male mice given neonatal injections
of 100 ug clomiphene (Clm) or nafoxidine (Naf), and
those given injections of 100 “wg Tx beginning at different
early postnatal ages were also examined. Tx caused more
damage to testis than did Clm and Naf. Mean sperma-
togenic index and diameters of seminiferous tubules in
mice given Tx beginning at 0 day were significantly
smaller than those in mice given Tx starting at 5 days.
These findings suggest that the critical period of produc-
ing the deleterious effects of Tx on the genital organs is
present within 3 days after birth.

INTRODUCTION

Perinatal treatment with natural and synthetic
estrogens including diethylstilbestrol (DES) in-
duces permanent suppression of spermatogenesis
in the testes of rats and mice [1-12]. Tamoxifen
(Tx), one of triphenylethylene derivatives, inhibits
estrogen action by competing with the hormone in
binding the estrogen receptor. Tx, therefore, has
been widely used for the therapy of estrogen-
receptor positive human breast cancer [13, 14]. On
the other hand, Tx has been reported as producing
estrogenic effects on uteri and vaginae of various

Accepted August 28, 1989
Received July 17, 1989

mammals [15-23]. In male mice, neonatal expo-
sure to Tx resulted in permanent suppression of
spermatogenesis and atrophy of genital organs [24]
as reported in mice treated neonatally with
estrogenic hormones [3-6]. Recently, various
abnormalities of pelvis [25, 26] and os penis [27]
were found in male mice treated neonatally with
Tx. Other antiestrogens, clomiphene (Clm) and
nafoxidine (Naf) also caused various abnormalities
in genital organs of mammals [28]. However, none
of the previous studies examined the responsive-
ness of mouse genital organs to administration of
Tx starting at different early postnatal ages. The
present study was aimed at examining the Tx-
induced sequential changes in genital organs and
their responsiveness to Tx in neonatal and early-
postnatal male mice.

MATERIALS AND METHODS

CS57BL/Tw male mice were kept under 12 hr
light/12 hr dark condition at 23-25°C temperature.
Mice were given 5 daily subcutaneous injections of
100 wg tamoxifen (Tx, mw=563.6) (Sigma, St.
Louis. MO), suspended in 0.02 ml of saline and of
the vehicle alone, starting within 24 hr after birth
and killed by ether anesthesia at ages of 5, 10, 20,
30 and 60 days. Mice were also given daily
injections of 100 4g Tx and of the vehicle alone for
5 days starting on the day of birth (0 day), 3, 5, 7
and 10 days. In addition, two groups of mice were
neonatally given 5 daily injections of 100 ug
clomiphene citrate (Clm, mw=598.1) (Sigma) and

542 S. Irisawa, T. IGucHI

100 wg nafoxidine hydrochloride (Naf, mw=
462.0) (Sigma) beginning on the day of birth,
respectively. These animals were killed at 60 days.
Pairs of testes, seminal vesicles with coagulating
glands and epididymides were weighed separately.
These organs were fixed in Bouin’s solution,
embedded in paraffin and serially sectioned at 8
ym. The sections were stained with Delafield’s
hematoxylin and eosin. In 5 transverse sections
randomly selected from each testis, 20 seminifer-
ous tubules were examined to count the number of
seminiferous tubules containing mature spermato-
zoa. Percent ratio of the tubules showing active
spermatogenesis was calculated for each mouse on
the basis of 200 tubules and was used as the index
of spermatogenic activity [5]. Diameters of semi-
niferous tubules were measured with a micro-
meter. Data were analysed by Duncan’s multiple
range test and Fisher’s exact probability test.

RESULTS AND DISCUSSION

Weights of body, testes and epididymides in
60-day-old mice treated neonatally with a daily
dose of 100 xg of Tx, Clm or Naf were significantly

AND N. TAKASUGI

smaller than those in the controls except for the
testis weight in Naf mice (Table 1). In mice given
Clm injections neonatally, however, the body
weights were significantly greater than in mice
given Tx injections starting at 0 day. Weight of
testes in mice given Tx starting at 5, 7 and 10 days,
respectively, were significantly greater compared
with that in mice given Tx starting at 0 day. Testes
and epididymides in Naf mice, were also signi-
ficantly heavier than those in mice given Tx from 0
day.

As shown in Figure 1, body weights in neonatal-
ly Tx-treated mice at ages of 10, 30 and 60 days
were significantly smaller than those in the age-
matched controls. Weights of epididymides (at 30
and 60 days), seminal vesicles with coagulating
glands (at 60 days) and testes (at 30 and 60 days)
were significantly smaller than those of the corres-
ponding controls. The testes in S-day-old Tx and
control mice contained seminiferous tubules with
spermatogonia and spermatocytes and interstitial
cells.

Spermatozoa and spermatids were found in the
tubules of control mice at 30 and 60 days. In the
controls, the mean spermatogenic index was 30.6

TABLE 1. Body and organ weights in 60-day-old CS57BL/Tw male mice treated with antiestrogens

Organ weights

anancenea Neo poy (mg/20 g body weight)
(days of age) (g testes epididymides
Saline
*0_-4 10 18.4+0.7* 135.7+8.4 S233
100 ng tamoxifen
0-4 10 12.4+0.9° 43.845.6° 20.4+3.0°
3-7 15 12.3+0.9° 63.2+5.0° 18.5+1.9°
5-9 10 14.3+0.9° 68.1+7.1°° 22.0+1.3°
7-11 10 15.0+0.9° 105.4+12.0°¢ 34.9+3.3°8
10-14 13 15.1+1.1° 104.1+9.9°4 33.7218
100 xg clomiphene
0-4 6 US yore One 66.6+4.3° 23.2+1.3°
100 xg nafoxidine
0-4 5 T2102 1.37 123.44 15.6° 39.0+4.6""

* The day of birth is indicated as 0; a, Mean+S

aE
P<0.01 vs controls (Duncan’s multiple range test)

©“ P<0.05 vs controls (Duncan’s multiple range test)

P<0.01 vs 100 ug tamoxifen 0-4 (Duncan’s multiple range test)

© P<0.05 vs 100 ug tamoxifen 0-4 (Duncan’s multiple range test)

Tamoxifen and Genital Abnormalities 543

8

—_
(on)

Body weight (g)
fo)

5
(0)
a ee SEE
© 60
E B
n» 50
©
3
—€ 40
>
3
oD
a
o
. 20
°
3 10
Co)
= 0
120
@ 100
E
‘oO » 80
~ Oo
2
8
a 40
&
E 20
7) *
Oh riya a a
5 101520 30 60
Age in days

8

Weight of testes (mg)
ron)
ro)

8o8 8

—_

Spermatogenic index (%)
S

8 0

7

8

Diameter of
seminiferous tubules (um)
8

50
Oe ae te Oe remeron
5101520 & 60
Age in days

Fic. 1. Sequential changes in body weight (A), weights of epididymides (B), seminal vesicles with coagulating glands
(C), testes (D), spermatogenic index (E) and diameters of seminiferous tubules (F) in mice given daily injections
of 100 ug Tx (@) or the vehicle alone(©) starting on the day of birth. *p<0.01, **p<0.05 vs respective controls.

+0.9% at 30 days, and 88.7+1.9% at 60 days. In
contrast, the testis of Tx mice began to show
spermatogenesis later than 30 days of age. Semi-
niferous tubules in Tx mice at ages of 20 and 30
days were smaller in diameter, lacking spermatids
and spermatozoa. In 60-day-old Tx mice, sperma-
togenic index (22.2+8.0%) was significantly lower
than the value in the age-matched controls. In Clm
and Naf mice at 60 days, spermatogenic indices

were not significantly lower than that in the
controls (Table 2). Diameters of seminiferous
tubules increased with age, from 5 to 60 days, in
both control and Tx mice, though the diameters in
Tx mice were significantly smaller than those in the
corresponding controls after 20 days.

In all groups of 60-day-old mice given 5 daily Tx
injections, spermatogenic indices and diameters of
seminiferous tubules were significantly smaller

544 S. Irisawa, T. IGucui AND N. TaKasucI

TaBLE 2. Changes in genital organs in 60-day-old CS57BL/Tw male mice treated neonatally with

antiestrogens
No. of mice showing varying .

sales o Spemmatogenteljndices speematopenid | ane
(days of age) mice <25% <50% <15% >715% index (%) tubules (4m)
Saline

0-4 10 0 0 1 9 88.7+1.9* 183.0+3.0
100 ng tamoxifen

0-4 10 le 1 1 1 22.2+8.0° 121.2+4.4°

3-7 15 gf 1 3 3 37.94+9.1° 118.4+5.2°

5-9 10 2 2 4 2 53.4+8.9°° 142.4+6.2°°

7-11 10 2 1 3 4 59.8+9.0°° 146.6+7.5°¢

10-14 13 2 3 3 5 54.6+8.1°° 155.4+6.5°°
100 ug clomiphene

0-4 6 0 0 4 2 68.8+4.4° 145.3+8.7°9
100 ng nafoxidine

0-4 5 0 0 3 2 74.44+3.6° 152.8+6.3°9
* Mean+S.E.; b, P<0.01 vs controls (Duncan’s multiple range test)
© P<0.05 vs controls (Duncan’s multiple range test)
4 P<0.01 vs 100 ~g tamoxifen 0-4 (Duncan’s multiple range test)

P<0.05 vs 100 ug tamoxifen 0-4 (Duncan’s multiple range test)

f P<0.01 vs controls (Fisher’s exact probability test)

than in the controls. Clm and Naf mice at 60 days
also showed significantly smaller tubules diameter
than those in the controls (Table 2). In mice given
Tx starting at 0 and 3 days, the number of mice
with spermatogenic indices lower than 25% was
significantly larger compared to that of the controls
(p<0.01, Fisher’s exact probability test).

In Tx mice at 60 days, epididymal ducts defec-
tive in stereocilia contained no spermatozoa in the
lumen, while in the age-matched controls, epididy-
mal ducts lined with a well-ciliated epithelium
contained numerous spermatozoa. Seminal vesi-
cles of these control mice contained a large amount
of eosinophilic secretion in the lumen, whereas no
such secretion was found in Tx mice at the same
age. These findings suggest that the testis of
60-day-old Tx mice failed to secrete an amount of
androgen sufficient for maintaining the sex acces-
sory organs or that the testis was decreased in
responsiveness to gonadotropin in Tx mice.

Previous studies have demonstrated that in
neonatally estrogen-treated male rats and mice,
the long-lasting suppression of spermatogenesis is
caused by direct effects of estrogen on the testis

and/or by indirect effects through a permanent
alteration of hypothalamo-hypohysial system [6, 9,
17, 25]. Male mice exposed neonatally to Tx also
showed permanent suppression of the sperma-
togenesis and atrophy of the genital organs [24],
suggesting that Tx has a side of estrogen agonist. It
is presumable, therefore, that Tx acts directly
and/or indirectly thorugh the alteration of gona-
dotropin secretion (FSH and/or LH) on the testis.

The present study showed varying degrees of
spermatogenesis suppression and genital organ
abnormalities in male mice given Tx starting at 0,
3,5, 7 and 10 days, the highest degree being found
in two groups of mice given Tx beginning at 0 and 3
days. It is suggested, therefore, that in mice, the
critical period of Tx induction of male genital
organ dysfunction and abnormalities is present
within 3 days after birth. Tx has been used for the
therapy of estrogen-reactive human breast cancer;
however, on the basis of the present study, the
possibility cannot be excluded that male fetuses of
pregnant women treated with Tx for the breast
cancer exhibits postnatally a long-lasting testicular
dysfunction, since neonatal mice are approximate-

ly

Tamoxifen and Genital Abnormalities

at the same stage as that of the 16- to 20-week

human fetuses [29].

M

12

13

ACKNOWLEDGMENTS

This work was supported by Grants-in-Aid from the
inistry of Education, Science and Culture, Japan.

REFERENCES

Takewaki, K. and Takasugi, N. (1953) Annot.
Zool. Japon., 26: 99-105.

Arai, Y. (1964) Endocrinol. Japon., 11: 153-158.
Mori, T. (1967) J. Fac. Sci. Univ. Tokyo, IV, 11:
243-254.

Takasugi, N. (1970) Endocrinol. Japon., 17: 277-
281.

Takasugi, N. and Furukawa, M. (1972) Endocrinol.
Japon., 19: 417-422.

Ohta, Y. and Takasugi, N. (1974) Endocrinol.
Japon., 21: 183-190.

McLachlan, J. A., Newbold, R. R. and Bullock, B.
(1975) Science, 190: 991-992.

Jones, L. A. (1980) Proc. Soc. Exp. Biol. Med.,
165: 17-25.

Arai, Y., Mori, T., Suzuki, Y. and Bern, H. A.
(1983) Int. Rev. Cytol., 84: 235-268.

Takasugi, N., Tanaka, M. and Kato, C. (1983)
Endocrinol. Japon., 30: 35-42.

Yasuda, Y., Kihara, T. and Tanimura, T. (1985)
Teratology, 32: 113-118.

Yasuda, Y., Ohara, I., Konishi, H. and Tanimura,
T. (1988) Am. J. Obstet. Gynecol., 159: 1246-1250.
Jordan, V. C. and Dowse, L. J. (1976) J. Endocr.,

14

15

18

19

20

21

22

23

24

25

26

27

28

29

545

68: 297-303.

Furr, B. J. A. and Jordan, V. C. (1984) Pharmacol.
Ther., 25: 127-205.

Harper, M. J. K. and Walpole, A. L. (1967) J.
Reprod. Fert., 13: 101-119.

Terenius, L. (1971) Acta Endocrinol., 66: 431-447.
Chamness, G. C., Bannayan, G. A., Landry Jr., L.
A., Sheridan, P. J. and McGuire, M. L. (1979) Biol.
Reprod., 21: 1087-1090.

Nguyen, B. L., Giambiagi, N., Mayrand, C.,
Lecerf, F. and Pasqualini, J. R. (1986) Endocrinolo-
gy, 119: 978-988.

Pasqualini, J. R. and Lecerf, F. (1985) J. Endocr.,
110: 197-202.

Pasqualini, J. R., Giambiagi, Sumida, C., Nguyen,
B.-L., Gelly, C., Mayrand, C. and Lecerf, F. (1986)
J. Steroid Biochem., 24: 99-108.

Lecerf, F., Nguyen, B.-L. and Pasqualini, J. R.
(1988) Acta Endocrinol., 119: 85-90.

Iguchi, T., Todoroki, R., Yamaguchi, S. and Taka-
sugi, N. (1989) Acta Anat., 136: 146-154.

Ohta, Y., Iguchi, T. and Takasugi, N. (1989)
Reprod. Toxicol., 3: 207-212.

Iguchi, T. and Hirokawa, M. (1986) Proc. Japan
Acad., Ser. B, 62: 157-160.

Iguchi, T., Hirokawa, M. and Takasugi, N. (1986)
Toxicology, 42: 1-11.

Iguchi, T., Irisawa, S., Uchima, F.-D. A. and
Takasugi, N. (1988) Reprod. Toxicol., 2: 127-134.
Irisawa, S., Iguchi, T. and Takasugi, N. (1988)
Zool. Sci. 5: 1315.

Clark, J. H. and McCormack, S. A. (1980) Science,
197: 164-165.

Kohrman, A. F. (1978) Rediatrics 62: 1143-1150.

| PE ae

cositumesood Sh Gebel tee Runa ie:

; 1 CAME CEs OR ll Ore Me ee ty
A nod hi : {ey

/ Py ee A - ss. : - - ~~

ety f
Ab

nel beagpa R Oae

. ‘ PTVial ec ee || io ; E oe
a ; 7 ont mind ij fooled). wd Meat eey ;
Cabrio ot nannt cathe) bed ooosc
é ey . = ; sy bis
if P pe -

“py % 2704S RS Ta
F iti 4 : hi »

274 ' ipinsns {il mguebdtt

POL AY ae

\
; Th a 41
}
| ' _
[
fh in
1}
bu / ‘
’ a wo
Hf
iy
'
Ds
/ i we
i
f lowe
i Ma Jona,
a)
AER,
i oy) tote
’
iy
ie
,
|
i
| -
} .
f
}
| i
i!
BI

ZOOLOGICAL SCIENCE 7: 547-549 (1990)

[COMMUNICATION]

© 1990 Zoological Society of Japan

Photoperiodic Influences on Pheromonal Delay of Puberty
in Young Female Wild Mice

SUBHASH C. PANDEY and SHEO D. PANDEY

Department of Zoology, Christ Church College, Kanpur, India

ABSTRACT — Individually housed young females were
painted on their external nares with distilled water or
urine collected from donor females subjected to different
photoperiodic treatments. Young painted with distilled
water attained puberty significantly earilier than those
painted with urine from donors under laboratory light
condition or under short photoperiod. Urine of donors
under long photoperiodic treatment did not delay the
pubertal onset as the mean time taken for occurrence of
first vaginal estrus in such urine-painted young was not
significantly different from that of water-painted control
young.

INTRODUCTION

Urinary pheromones in laboratory mice can
accelerate or delay the onset of puberty in juvenile
females. Sexual maturation of prepuberal females
is accelerated by exposure to adult males or to
their urine [1]. Young females living in groups or
raised with adult females attain puberty later than
those living in isolation [2]. The delaying effect is
chemically mediated through urine collected from
grouped females [3]. The acceleration and delay of
puberty also occur in wild mice and the causative
factors have been found to be present in the urine
of male and female, respectively [4].

Naturally occurring variations in environmental
factors influence the efficiency of different chemo-
signals modulating various physiological and be-
havioral activities in mammals [5]. Seasonal
variation has been reported in pheromonal accel-
eration and delay of puberty in female laboratory
mice [6]. Continuously breeding house mice

Accepted August 15, 1989
Received June 19, 1989

exhibit seasonal patterns of reproductive activity in
certain regions [7]. Long diurnal photoperiod
(16L:8D) abolishes mutual suppression of estrus in
sparsely housed females at 36-38°C (our unpub-
lished observations). For animals inhabiting wild,
the interactions between social and environmental
factors are potentially important. The present
effort was therefore aimed to determine the
relative influence of the day length and the
puberty-delaying chemosignal of female origin on
sexual maturation in young females.

MATERIALS AND METHODS

The wild mice, Mus musculus domesticus, em-
ployed in this study were trapped from the field
and maintained in the laboratory on a diet consist-
ing of soaked gram, boiled rice and milk. Water
was available ad libitum. Forty eight prepuberal
females (subject females) weighing 4.6+0.58 g
were randomly divided into 4 groups and housed
individually in isolation cages (341814 cm)
under laboratory conditions of light (ca 13 hr) and
temperature (30-38°C). There were 8 females in
group A and D and 16 each in group B and C. The
subject females were painted daily on the their
external nares with distilled water (group A) or
with diluted urine pooled from different donor
females (group B, C and D; Table 1).

Adult females which served as donors were
housed in colony cages (303030cm) at a
density of 10 mice/cage. These donors were either
held under laboratory light condition (LLC) or
exposed to cool white fluorescent light of long
(16L:8D) and short (8L:16D) duration at an

548 S. C. PANDEY AND S. D. PANDEY

TABLE 1.

Mean time taken for the occurrence of first vaginal estrus in young females painted

with water or urine from donor females exposed to different photperiods

Mean time (in days) taken for

Group Painting material first vaginal oestrus to occur
A Water (control) 27.3 +0.78*
B Urine of donors under 28.4+0.35
long photoperiod (16L: 8D)
C Urine of donors under 36.6 +0.53*
short photoperiod (8L: 16D)
D Urine of donors under LLC 36.9+0.88

* Means connected with same vertical line are at par at 5% level of significance.

intensity of about 400lux. The above photic
treatments of donor females commenced 15 days
before their urine was used in the experiment.
Fresh urine was collected daily from all the donors
by manual bladder palpation and diluted in distil-
led water (1:9). A drop (0.05 ml) of urine was
applied twice daily at 9:00 and 17:00 hr with the
help of a small paint brush and subjects were
examined for the appearance of vaginal perfora-

tion. Starting on the day of vaginal perforation
smears were collected daily until the occurrence of

first vaginal estrus indicative of pubertal onset in
females. The data were analysed by one way-
analysis of variance.

RESULTS AND DISCUSSION

Onset of puberty as assessed by occurrence of
the first vaginal estrus was delayed in subject
females painted with urine of donors held under
LLC or short-photoperiod. Control females
painted with water exhibited first vaginal estrus
significantly earlier than those painted with urine
of above donors (p<0.01). However, prepuberal
females painted with urine of donors subjected to
long photoperiodic treatment attained puberty at
about the same time when it occured in control
females as the difference in mean time taken for
first vaginal estrus in these two groups was not
statistically significant (27.3+0.78 days vs. 28.4+
0.35 days, C.D.=2.25; Table 1).

Urine from adult donor females maintained
under natural light: dark cycle exert a retardative
effect on pubertal onset in young females. Results
presented in the study demonstrate that the
exposure of donor females to long day length

abolishes their puberty-delaying property. Expo-
sure to short photoperiod is without any effect on
the ability of donors to delay the sexual matura-
tion. Evidently, the photoperiodic manipulations
can alter the production/release of the delay
chemosignal in adult donor females.

Seasonal breeding is a common reproductive
strategy thought to increase the probability of
survival of young [8]. In a seasonally changing
environment, the day length has been suggested to
be the most proximate signal that initiates repro-
ductive activities in mammlas [9]. Social factors
acting in concert with environmental factors also
influence puberty and reproductive processes in
mammals [5, 6]. Adult females perhaps release the
delay-chemosignal to communicate the adequacy
of reproductive conditions to juvenile females [10].
Urine from isolated or sparsely housed adult
females have been found ineffective in causing
puberty delay in young female mice [11]. Seeming-
ly, at higher population density when donor
females release an effective amount of puberty-
delaying chemosignal, long diurnal photoperiod (a
signal of the onset of favorable breeding condition)
diminishes the efficiency of the puberty-delaying
cue in favor of propagation.

ACKNOWLEDGMENTS

This study was supported by a grant from the
Department of Science and Technology to SDP and by
an award of a Research Associateship to SCP from the
Council of Scientific and Industrial Research of India.

REFERENCES
| Bronson, F. H. and Maruniak, J. A. (1975) Biol.

Photoperiod and Delay of Puberty 549

Reprod., 13: 94-98.

Vadenbergh, J. G., Drickamer, L. C. and Colby, D.
R. (1972) J. Reprod. Fertil., 28: 397-405.
McIntosh, T. K. and Drickamer, L. C. (1977)
Anim. Behav., 25: 99-104.

Pandey, S. C. (1986) Studies on pheromones in wild
mice. Ph.D. thesis, Kanpur Univ., Kanpur.
Bronson, F. H. (1985) Biol. Reprod., 32: 1-26.
Drickamer, L. C. (1984) J. Reprod. Fertil., 72: 55-
58.

Pelikan, J. (1981) Symp. Zool. Soc. Lond., 47: 205-
230.

8

10

11

Sadleir, R. M. F. S. (1969) The Ecology of
Reproduction in Wild and Domestic Mammals.
Methuen and Co., London.

Negus, N. C. and Berger, P. J. (1972) In “Biology of
Reproduction: Basic and Clinical Studies”. Ed. by J.
T. Velardo and B. A. Kasprow, Third Pan Amerian
Congress on Anatomy, New Orleans, pp. 89-98.
Drickamer, L. C. (1982) Dev. Psychobiol., 15: 433-
442.

Pandey, S. C. and Pandey, S. D. (1989) Arch. Biol.,
(In press)

¢ om cr

wu ee ee eee ae wi,

, : m4 sft) (Qe) is i WM Aol adhe oF Ge agen! ’ i nay tom
i" : F tar, -Dlwl ars nope “7 Pays KC) taste: ore! yeerieatatetd
iW . restorer ‘ ee > e14) tra irtpabd ca al Soren “ie 19a = : = iy tg
\ ; L Pyineee 2M ug. ¢ TT) 2, Baeiity” hy 4

i, -, +> bs polbubmnnadl : =: _ » ~bBL 22 "BES
i t} ord piglet oF inkiw ¢u eon i fegdhey ihe doled: eyitay
i ats ‘ * ae TUnaAA ell uaqagt ae

| % e>4) : ficxhive bri Lan v.) 263 cwypedt yet | ¢ ie .

. oe 2027 linet borer aa
} . P i” gee oa
i ' come vil 2, Tb, ld one edbS ieee
i 4 : a pitas -

HI oS ’ i | I yaseei aficy -

.
— =
tarrey-ichta VAR
’ a eqrod iM ) i
| 3 =
Hi :

i cin

pt

ieyah 1H ;
anal th
y ti vezi, iy
a r PKU a}

erp
onde mee
ito “
nce

jeu 7

ZOOLOGICAL SCIENCE 7: 551-555 (1990)

[COMMUNICATION]

© 1990 Zoological Society of Japan

Immunocytochemical and Ultrastructural Characterization
of the Cells in the Pars Tuberalis of the
Turtle, Geoclemys reevesii

Y OSHIHIKO OOTA

Biological Institute, Faculty of Science, Shizuoka University,
Shizuoka 422, Japan

ABSTRACT— Using immunocytochemical and electron
microscopical techniques, two distinct types of secretory
cells were detected in the turtle pars tuberalis (PT). The
cells showing positive immunoreaction to TSH-antiserum
were exclusively found in the rostral PT, while those
immunoreactive to FSH-antiserum were concentrated in
the caudal PT.

INTRODUCTION

The identification of cell types producing differ-
ent pituitary hormones has been established in the
pars distalis (PD) of the reptilian pituitary gland by
conventional staining methods [1-3], electron mi-
croscopy [2, 4, 5] and immunocytochemistry [5-8].
However, little attention has been paid to the pars
tuberalis (PT), since PT is predominantly com-
posed of chromophobic cells. A few ultrastruc-
tural studies showed the presence of secretory
granules in some cells in the PT of the turtle [6, 8,
9]. Thus, to date, there is very little information
concerning the functional significance of the PT.
The present report describes the presence of cells
immunoreactive to FSH- or TSH-antiserum in the
turtle PT.

MATERIALS AND METHODS

Adult male and female turtles, Geoclemys reeve-
sii, (caraspace length, 160-200 mm) were obtained
from a dealer. The hypothalamic regions of 6

Accepted September 8, 1989
Received August 15, 1989

animals were fixed in Bouin-Hollande sublimate
solution. Sections were made at 6 um thickness
through ordinary paraffin method. Some sections
from each animal were stained with Azan or
Gomori’s paraldehyde fuchsin staining method.
For immunocytochemical study, avidin-biotin-
peroxidase complex method was used [10]. The
following antisera raised against rabbits were used:
anti-rat somatotropin, anti-rat prolactin, anti-rat
TSH-, anti-porcine ACTH, anti-bullfrog LH-@
and anti-bullfrog FSH-@. All the primary antisera
were kindly provided by Dr. K. Wakabayashi,
Hormone Assay Center, Institute of Endocrinolo-
gy, Gunma University. To ensure the specificity,
all positive results were controlled by omission of
the primary antisera and by a parallel incubation
with antisera preabsorbed with the respective
hormone. For electron microscopy, the hypothala-
mic regions of 8 animals were fixed in paraformal-
dehyde and glutaraldehyde, postfixed in 1% OsOg,
and embedded in Epon. Ultrathin sections were
stained with uranyl acetate and lead citrate, and
examined with a JEM-T8 electron microscope.

RESULTS AND DISCUSSION

In the turtle, the PT is well developed and can
be divisible into rostral and caudal portions (Fig.
1). The rostral portion spreads under the median
eminence as a thick layer of flattened epithelial
cells. Between the rostral portion and the median
eminence, lie capillaries of the primary plexus,
whose blood drains into the hypophysial portal

552

Immunocytochemistry of Turtle PT 553

vessels. The caudal portion is a thick cluster of
cells situated dorsally to the anterior part of the
PD. Histologically, the majority of the tuberal
cells are chromophobic for the tinctorial stainings
used in this study. A few cells showing slight
affinity for AF appear only in the caudal portion of
the PT, although there are numerous AF-positive
cells in the PD.

In the rostral portion, some cells show a positive
reaction to TSH-antiserum (Fig. 2). They are large
and angular in shape. Cells reacting to FSH-
antiserum are found to be concentrated mainly in
the caudal portion, generally being spherical or
ovoid in shape (Fig.3). Preabsorption of the
antisera with TSH and FSH and also omission of
the respective primary antisera completely block
the positive staining results. Throughout their
distributions, the immunoreactive FSH-cells out-
number the immunoreactive TSH-cells (Figs. 2,
3). All the tuberal cells are non-reactive to other
antisera than TSH-and FSH- antisera tested.

Although the PT tissue appears largely chro-
mophobic with tinctorial staining methods it is very
interesting to note that two types of cells were
identified immunocytochemically: cells showing
TSH-like immunoreactivity entirely in the rostral
portion and cells showing FSH-like immunoreac-
tivity largely in the caudal portion. In the PD,
TSH- and FSH- immunoreactive cells have been
frequently demonstrated in the caudal lobe of the
turtle PD [5]. Since the turtle PT develops as a
pair of lateral outgrowths from the anterior aspects
of the pituitary anlage [7], it is possible that the
cells reacting with TSH- and FSH- antisera in the
PT might have migrated from the caudal lobe of
the PD during embryogenesis.

Five secretory cell types have been recognized in
the PD in various staining methods including
immunocytochemical study [2, 5, 6, 11]. In this
study, cells reacting with somatotropin-, prolactin-,

ACTH- and LH-antisera are frequently detected
in the PD. However, cells immunoreacting to
these antisera are undetectable in the PT.

Electron microscopically, two types of secretory
cells can be distinguished. Type 1 cells are found
in the rostral portion, although the number is not
many. They are characterized by the presence of
numerous granulated vesicles (150-200 nm in di-
ameter) with variable electron densities (Fig. 4).
The Golgi apparatus and rough endoplasmic re-
ticulum are generally well developed. Type 2 cells
are found in the caudal portion, and they contain
electron-dense granules, mostly with diameters
ranging 200 to 360 nm (Fig. 5). They present well
developed Golgi apparatus, numerous and dilated
cisternae of the rough endoplasmic reticulum.
Ultrastructurally, presence of two types of secre-
tory cells in the turtle PT has been reported
previously [8]. Judging from the location and the
difference in size of granules contained in respec-
tive cells and the present immunocytochemical
findings, type 1 cells may be characterized as
TSH-cells, whereas type 2 cells may be considered
as FSH-cells.

The vascularization of the turtle hypophysis has
been investigated previously, and portal vessels
run through the rostral and caudal PT [12].
Although some cells of type 1 and type 2 are
closely contacted with the capillaries of the portal
system, extrusion of the secretory granules into the
capillaries were not demonstrated. At present, it is
not known to which extent these immunoreactive
TSH- and FSH-cells in the PT are physiologically
involved in the thyrotropic and gonadotropic
functions of the pituitary. Presence of im-
munoreactive TSH- and FSH-cells has been re-
ported in the PT of a variety of mammals [13].

Fics. 1-3.

Mid-sagittal sections through infundibulum and hypophysis.

1. Section stained with Azan stain. Pars

tuberalis can be divisible into rostral (rPT) and caudal portion (cPT). Dotted line represents the boundary
between rPT and cPT. CE, cephalic lobe of the pars distalis; ME, median eminence. Arrows indicate blood
capilaries. X115. 2. Section immunostained with anti-TSH. Cells reacting to the antiserum are confined to the
rPT. X220. 3. Section immunostained with anti-FSH. Cells reacting to the antiserum are common in the cPT.

CA, caudal lobe of the pars distalis. x 220.

554 Y. OoTA

oa.

Fic. 4. Electron micrograph of the rostral portion of the PT. Note type 1 cell containing a few granulated vesicles.

x 18,200.
Fic. 5. Electron micrograph of outermost layer of the caudal portion of the PT. Note type 2 cell with many dark

granules. x18,200.

Immunocytochemistry of Turtle PT 555

REFERENCES

Saint Girons, H. (1970) In “Biology of the Rep-
tilia”. vol. 3, Ed. by C. Gans, Academic Press,
London/New York, pp. 135-199.

Licht, P. and Pearson, A. K. (1978) Int. Rev.
Cytol., Suppl., 7: 239-286.

Fitzgerald, K. T. (1979) Gen. Comp. Endocrinol.,
37: 383-399.

Doerr-Schott, J. (1976) Gen. Comp. Endocrinol.,
28: 513-529.

Mikami, S. (1986) In “Pars Distalis of the Pituitary
Gland” Ed. by F. Yoshimura, and A. Gorbman,
Elsevier Science Publ. B.V. pp. 71-79.

Pearson, A. K. and Licht, P. (1982) Cell Tissue
Res., 222: 81-100.

7

Pearson, A. K. (1985) In “Biology of the Reptilia”.
vol. 14A Ed. by C. Gans, John Wiley & Sons, Inc.,
New York, pp. 681-719.

Oota, Y. (1980) Rep. Fac. Sci., Shizuoka Univ., 14:
63-74.

Dellmann, H.-D., Stoeckel, M. E., Hindelang-
Gertner, C., Porte, A. and Stutinsky, F. (1974) Cell
Tissue Res., 148: 313-329.

Hsu, S. M., Raine, L. and Fanger, H. (1981) J.
Histochem. Cytochem., 29: 577-580.

Oota, Y. (1986) Proc. Japan Acad., 62B: 311-313.
Oota, Y. and Koshimizu, I. (1988) Zool. Sci., 5:
1013-1018.

Gross, D. S. (1984) Gen. Comp. Endocrinol., 56:
283-298.

eo a

"
i , >
|| ay
| { 4s
ged f ‘ ae on A Y
t% rT Ves 4p ekelabamatga ee ts)
; i Wu sy | } ‘ as |;
} - ;
! ‘iy? eum
4 y Pat
fe a
j f F h ay F]
as / = ;
ae =
E r
l; i i *
Mp ot
i
sh4
t)
} ay
| j a2 e oe
1! y
i
; ry
ft 7
| acne

ZOOLOGICAL SCIENCE 7: 557-561 (1990)

[COMMUNICATION]

© 1990 Zoological Society of Japan

A Centrolenid-Like Anuran Larva from Southeast Asia

Rosert F. INGER and RICHARD J. WAsSsERSUG!

Field Museum of Natural History, Roosevelt Road at Lake Shore Drive,
Chicago, Illinois 60605-2496, USA

ABSTRACT— Vermiform tadpoles of the treefrog fami-
ly, Centrolenidae, live buried in leaf litter on the margins
of streams in the New World. These tadpoles have,
among other features, long tails with reduced fins, small,
subcutaneous eyes and highly vascularized, nearly pig-
mentless skin. We describe here the first tadpole from
Southeast Asia with this morphology. The tadpole was
collected in a microhabitat similar to the one in which
centrolenid larvae have been found. Except for the fact
that the Bornean larva has more denticle rows, the
tadpoles are virtually identical in external morphology.
The Bornean tadpole probably belongs to either a ranid
or rhacophorid genus but, since no metamorphic indi-
viduals with adult diagnostic features have been found,
the taxonomic assignment is uncertain.

INTRODUCTION

The Centrolenidae is a well-defined family of
arboreal frogs restricted to Central and South
America. Tadpoles have been described for less
than a fifth of the 65 known species, yet those
described are sufficiently similar to one another in
ecology and morphology, while different enough
from other known larvae to characterize the family
ae

Centrolenid tadpoles are found only in associa-
tion with streams. Those collected in the field have
been found burrowed into decaying vegetation at
the edge of the water, not exposed to the current.
These tadpoles are characterized by having long
tails (22xthe head-body length) with reduced
dorsal and ventral fins, depressed or cylindrical
bodies, small dorsal eyes covered with skin, and

Accepted August 8, 1989
Received June 19, 1989

! Present address: Department of Anatomy, Dalhousie
University, Halifax, Nova Scotia B3H 4H7, Canada

reduced pigmentation. All of these features seem
directly related to their fossorial way of life.

In this paper, we describe a centrolenid-like
tadpole from Borneo. At present we can identify
the larva only to suborder. Subtle oral features
distinguish this larva from true centrolenid larvae;
otherwise the convergence in microhabitat use and
overall morphology is among the most precise that
we know of for amphibian larvae.

RESULTS

Four specimens have been collected (Table 1)
from two localities (Danum Valley Field Centre,
Lahad Datu District, Sabah and Nanga Takalit,
Kapit District, Seventh Division, Sarawak) and
deposited in the Field Museum of Natural History
(FMNH). The following description is based on
the largest specimen collected so far, a stage 31 of
Gosner [2] individual.

The general body form is ovoidal, strongly
depressed. The eyes are dorsal, extremely small,
below the skin and far posterior from the tip of the
snout. The eye-snout distance is one-third the
head-body length. The pupils are directed ob-
liquely dorsolateral. Eye diameter is 8% of
head-body length; interorbital distance is 60% of
the internarial distance. The nostrils are small
(<1% of head-body length), lack an elevated rim,
and are exceptionally far forward. They are just
dorsal to the lateral edge of the snout, above the
corner of the mouth, at 11% of distance from
snout back along the head-body.

The oral disc is ventral, subterminal. Its width is
2/3 the maximum width of the body. The marginal

TABLE 1.

R. F. INGER AND R. J. WASSERSUG

tadpoles from Borneo

Measurements! and denticle counts of fossorial stream-associated

Specimen(s) FMNH #31452 FMNH #221024
Locality Danum N. Takalit
No. individuals 1 3
Stages” 31 25
Head-body length 10.0 5.4-6.5
Head-body depth 5.2 3.3-4.0
Head-body depth 3.6 2.1-2.3
Eye diameter 0.8 0.3
Eye-snout 2.8 1.7-2.2
Nostril-snout 1.1 0.7(1)
Interorbital 1.4 1.1-1.2
Internarial eo) —_—
Snout-spiracle Wail 4.5-4.8
Tail length PbS 10.4-13.0
Tail depth 4.7 2.7-3.3
Denticles*

-upper I:1-1 I: 1-1

-lower 1-1: VII 1-1: VIII

" Measurements made with ocular micrometer at 12x, given in mm; definitions of

dimensions as in Inger and Frogner [15].

According to Gosner [2].
3 Follows system used in Inger [3].

papillae are in a single row with a narrow median
gap along the posterior free edge of the disc. The
papillae are of moderate size. Papillae are absent
along the anterodorsal edge of the disc, but are
clustered on a small outwardly directed flap at the
lateral corners of the mouth. A distinct notch
separates these flaps from the ventral portion of
the disc. The papillae on the lateral flaps are the
largest and are in two rows, seven to nine papillae
in an outer row and two to four in an inner row.
The dental formula, following the system of
Inger [3], is 1: 1-1/1-1: VII. The denticles in the
ventral rows decrease in size from the inner-most
to outer-most rows and the four outer-most rows
are distinctly shorter than the inner ones. The
denticles are dense, spatulate and range from
black (large) to brown (small) in color. They all
curve toward the mouth. In microscopic detail,
they most closely resemble the denticles illustrated
for Rana chalconota and Rana signata in Inger [3].
The upper beak is a wide, gentle arch. The
margin is finely serrated. The marginal serrations
are of uniform size, smaller than the denticles, and

darkly pigmented. The lower beak is V-shaped,
with an angle of 100° between the arms. The
pigmented margin of the lower beak is slightly
more extensive than that of the upper beak; its
serrations are coarser than those of the upper
beak, but still finer than the neighboring denticles.

The spiracle is sinistral, midway up the side of
the body and far posterior (i.e., 70% of distance

—— |

Fic. 1.
fossorial ranoid tadpole described in this report.
The illustration is of a stage 25 specimen (FMNH
#221024) See Table 1. Scale line=5.0 mm.

Dorsal (above) and lateral (below) views of the

A New Fossorial Tadpole 559

NNUAL puree

‘ Dl hay
(Ce uM : cu pul My) eed
lt aL yy un ty,
mM

PIL TP

t
tl ie
Meena TLCS pT a nt OU gg

\
uu “|

wl a pelt
PTAA INCLU a TU oe OT
il Merce La yl! [Lider LUCE
MT ana Mian Mihig Meauyi! ye ute wwe
: EI ACR TL
wnat tage tA tec
src lll
Many yal ll
vnmttt!*

q
Ariba wan

|

Fic. 2. Oral disc of the tadpole illustrated in Fig. 1.
Scale line=0.5 mm.

from the snout tip to end of body). The spiracle
has a short, free, terminal tube. The width of the
Spiracular opening is slightly greater than the
diameter of the eye.

The anal tube is median and long.

The tail is long and slender, 2 1/4 times the
head-body length (Table 1). The tail tip is broadly
rounded. The fin margins are very weakly convex.
The maximum depth of the tail is at approximately
3/5 of tail length. Both the dorsal and ventral fins
are Shallow. The dorsal fin begins on the tail just
above the end of the anal tube. The ventral fin
begins immediately behind the body proper and is
slightly deeper than the dorsal fin in the proximal
1/3 of the tail. The caudal musculature is deeper
than either fin in the proximal 2/3 of the tail.

The head-body of the preserved tadpole is
pigmentless. The skin on the body has a shiny
appearance. The skin overlying the caudal muscu-
lature has a faint, brownish reticulum. This
pigment pattern extends up on to the dorsal fin
near the base of the tail.

Neuromasts are present but not easily observed
or counted. They are singular, not organized into
stitches. Approximately 20 neuromasts are pres-
ent in both the infraorbital and supraorbital rows
on each side of the head.

The posterior 2/3 of the intestines, along with a
portion of the pancreas, are visible through the
ventral body wall. We counted five coils of
intestine from the center of the gut to the caudal
edge of the body cavity.

Because the internal oral features of this tadpole
are so similar to those of Centrolenella fleischman-
ni [4], only features that differ from C. fleischman-
ni are described here. Terminology follows Was-
sersug [4].

The medial pair of infralabial papillae are
smaller. The lateral pair of infralabial papillae are
larger with finger-like projections. There are four
buccal floor arena papillae per side. The prenarial
arena has a small, median knob anteriorly and a
larger transverse ridge posterior to it. The long,
longitudinally oriented, internal narial depression
that characterizes C. fleischmanni is absent. The
median ridge is smaller and more posterior. The
postnarial papillae are much larger. They have
rounded apices and project anteromedially over
the medial half of the internal nares. Additional
flap-like papillae of the anterior buccal roof are
absent. The dorsal velum is not continous across
the midline. The glottis is smaller and not patent.

Qualitatively, the structure in the pharynx of
this specimen looked indistinguishable from those
of C. fleischmanni. Secretory ridges are present on
the branchial food traps. Because of the small size
of the specimen, no attempt was made to count gill
filter rows.

The lungs at stage 31 are large, as long as the
body cavity and flattened, indicating that they

were not inflated in life.
vascularized.

Fine, heterogeneous particulate matter fills the
alimentary tract. No macroscopic fragments of
arthropods or plants are present. Some small
mineral grains are visible, but most of the contents
are unidentifiable organic debris.

The skin is heavily

Taxonomic considerations

The most mature tadpole collected so far is still
too young to reveal any adult diagnostic charac-
ters. Adult frogs in the neighborhoods where
these fossorial tadpoles were collected comprise
pelobatids, bufonids, ranids, rhacophorids, and
microhylids [5, 6]. The presence of denticles, a
sinistral spiracle and perforated nares immediately
exclude the tadpole from the Microhylidae [7].
The oral disc papillation, denticle row counts and
denticle morphology are unlike any Bornean
bufonids or pelobatids. The tadpole is by elimina-

560 R. F. INGER AND R. J. WASSERSUG

tion tentatively classified as ranoid, either Ranidae
or Rhacophoridae. Further assignment to family
or genus is not possible at this time.

DISCUSSION

Our Bornean tadpoles were collected in deep
leaf litter within but near edges of clear streams in
lowland primary rain forest. The three in stage 25
were taken with larvae of two bufonids (Ansonia
leptopus and Pedostibes hosei), and a pelobatid
(Leptobrachium montanum). The stage 31 tadpole
was found with larvae of Rana signata. All of
these, except for the larval Leptobrachium, are
commonly found in drifts of dead leaves [6].

Although Duellman and Trueb [1] claim that
centrolenid tadpoles “develop in gravel or detritus
in flowing water”, to the best of our knowledge
free-living centrolenid larvae avoid both flowing
water and gravelly substrates. Rather they are
found within the “accumulation of leaves, sticks
and mud” [8] or occasionally even in the humus of
a stream bank above the waterline [9] and out of
the current. The common microhabitat of cen-
trolenid larvae closely resembles the microhabitat
where our unusual Bornean tadpoles were col-
lected.

Except for the denticle formula and _narial
position, our Bornean tadpoles are externally
identical to those of many centrolenid species [10].
The body profile, spiracle position, tail length, fin
shape, and overall size and coloration are the
same. So are the small nostrils and the small,
dorsally positioned, sub-cutaneous eyes. The most
conspicuous feature that distinguishes the Bornean
tadpole from centrolenid larvae are the super-
numerary, ventral denticle rows. Centrolenids all
have two or fewer upper rows and three or fewer
lower rows.

Internally our tadpole can be distinguished from
C. fleischmanni by a variety of anatomical fea-
tures, primarily anteriorly near the orifice of the
mouth and the internal nares. These may be little
more than internal reflections of differences in
external oral morphology and narial position
(slightly more anterior in the Bornean species). In
key features, which reflect the filter-feeding capac-
ity of the tadpoles—such as the pattern of papillae

on the buccal floor and roof, the shape of the
ventral velum, filter plates, and gill filters—the
tadpoles are indistinguishable.

Virtually all of the features that distinguish both
the Bornean species and centrolenid tadpoles from
typical ranoid and bufonid larvae can be under-
stood as adaptations for a fossorial rather than
either pelagic or demersal existence [9, 11, 12].
Not all fossorial tadpoles are associated with
stream bank leaf litter, but all have cylindrical or
depressed bodies and long tails with reduced fins.
All have small dorsally located eyes and reduced
pigmentation. Several arboreal tadpoles from a
variety of anuran families fit this description [13].
The Bornean species is exceptional among all
fossorial tadpoles in its high number of ventral
denticle rows. We have no functional explanation
for this distinctive feature. Another noteworthy
feature in the Bornean species is the uninflated
lung. This tadpole does not breathe air like certain
arboreal and semi-terrestrial “fossorial” tadpoles
[10, 13]. Villa [9, 14] suggested that centrolenid
tadpoles, which are often reddish in life, appear so
because of perfusion of skin for cutaneous respira-
tion. Although the Bornean tadpoles were not
particularly reddish in life, the extensive vascular-
ization of their skin supports the idea that they rely
heavily on cutaneous respiration.

As a final note, although we have formally
described a single form of centrolenid-like tadpole
from Southeast Asia, we suspect that other Old
World species may have tadpoles of this type.

All other tadpoles collected with our fossorial
form have been collected frequently and in large
numbers in Bornean stream leaf litter [6]. The
rarity of the fossorial form suggests that is may
have been displaced into a slightly more aquatic
situation than where it normally resides. Truly
fossorial tadpoles are notoriously difficult to col-
lect [9], as evidenced by the fact that larvae are
known for less than 20% of centrolenid species in
contrast to 25-30% for remaining anurans.
Among tropical faunas the tadpoles of Borneo are
well known [3], but even there tadpoles are known
for only about 50% of the species.

A New Fossorial Tadpole 561

ACKNOWLEDGMENTS

We thank Molly Ozaki and Brenda Zwicker for typing
the manuscript, and V. Ann King for drafting the figures.
Scott Pronych and Tracey Earle provided critical com-
ments on early drafts. This work was supported by a
National Science Foundation (USA) grant and an award
from Marshall Field III Fund to R.F.1I. The junior author
was supported by the Natural Sciences and Engineering
Research Council (CANADA).

REFERENCES

1 Duellman, W. E. and Trueb, L. (1986) Biology of
Amphibians. McGraw Hill Book Company, New
York.

2 Gosner, K. L. (1960) Herpetologica, 16: 183-190.

3 Inger, R. F. (1985) Fieldiana: Zool., New Series,
26: 1-89.

4 Wassersug, R. J. (1980) Univ. Kansas Mus. Nat.
Hist. Miscellaneous Publ., 68: 1-146.

Inger, R. F. (1966) Fieldiana: Zool., 52: 1-402.
Inger, R. F. Voris, H. K., and Frogner, K. J. (1986)
J. Tropical Ecol., 2: 193-205.

Wassersug, R. J. (1989) Fortschritte der Zool. (In
press)

Heyer, W. R. (1985) Papeis Avulsos de Zoologia,
36: 1-21.

Villa, J. and Valerio, C. E. (1982) Brenesia, 19/20:
1-16.

Altig, R. and Johnston, G. F. (1986) Smithsonian
Herp. Information Service, 67: 1-75.

Wassersug, R. J. and Heyer, W.R. (1983) Canad. J.
Zool., 61: 761-769.

Wassersug, R. J. and Pyburn, W. F. (1987) Zool. J.
Linnean Soc., 91: 137-169.

Lannoo, M. J., Townsend, D. S. and Wassersug, R.
J. (1987) Fieldiana: Zool., New Series, 38: 1-31.
Villa, J. (1984) Milwaukee Publ. Mus. Contrib. Bio.
Geol., 55: 1-50.

Inger, R. F. and Frogner, K. J. (1980) Sarawak
Mus. J., 27: 311-324.

ie

: oa
"

‘etal eet oceeaaee

met | mpg ae

.
7 sent

a

Sasetyiin

rl ie a) waa! aa iS ie:

snegees Wik aw FS oktingolly
iI pt if.ets tnt gotham ail

Ph IIS a5 ae eR

ee rotons oom, Cee

ti aden Ni i Flats (dat sb

Wein Whit) Were EA na

| ws eee Uenlad

j gland

‘hot a

~ bry i :
®

‘

sci A sy

*

~f

4
.

D eve | O m e nt Published Bimonthly by the Japanese Society of
Pp Developmental Biologists
Distributed by Business Center for Academic

Growth & Differentiation Societies Japan, Academic Press, Inc.

Papers in Vol. 32, No. 3. (June 1990)

30. REVIEW: T. Shinomura and K. Kimata: Precartilage condensation during skeletal pattern
formation

31. K. Yamamoto and S. Nemoto: Absence of species specificity of germinal vessicle factor
required for the cytoplasmic cycle during meiotic division

32. H. Nakayama, X. Ru, J. Fujita, T. Kasugai, H. Onoue, S. Hirota, H. Kuroda and Y.
Kitamura: Growth competition between W mutant and wild-type cells in mouse aggregation
chimeras

33. C. Hui and Y. Suzuki: Homeodomain binding sites in the promoter region of silk protein
genes

34. Y. Kajiwara, T. Kuwana and M. Inoue: Pigmentation patterns of mouse chimaeras following
injection of embryonic cells into postimplantation embryos in Utero

35. R.A. Raff, L. Herlands, V. B. Morris and J. Healy: Evolutionary modification of Echinoid
sperm correlates with developmental mode

36. K. Akasaka, Y. Akimoto, M. Sato, H. Hirano and H. Shimada: Histochemical detection of
arylsulfatase activity in sea urchin embryos

37. K. Akasaka, K. Yamada, K. Sekine and H. Shimada: Effects of aphidicolin on arysulfatase
gene expression in sea urchin embryos

38. A. Fujiwara K. Taguchi and I. Yasumasu: Fertilization membrane formation in sea urchin
eggs induced by drugs known to cause Ca** release from isolated sarcoplasmic reticulum

39. P. A. Veno, M. A. Strumski and W. H. Kinsey: Purification and characterization of
echinonectin a carbohydrate-binding protein from sea urchin eggs

40. I. Takahashi, T. Ueda, Y. Kameoka, K. Abe, N. Takagi and K. Hashimoto: Construction of
a DNA library enriched with mouse 4 chromosome of T(X; 4)37H translocation

41. K. Joseph and T. G. Baby: Changes in polyamine contents during development of the frog,
Microhyla ornata

42. K. Mitsunaga, S. Shinohara and I. Yasumasu: Probable contribution of protein phosphoryla-
tion by protein kinase C to spicule formation in sea urchin embryos

43. N. Yoshizaki: Localization and characterization of lectins in yolk platelets of Xenopus oocytes

Development, Growth and Differentiation (ISSN 0012-1592) is published bimonthly by The Japanese Society of
Developmental Biologists, Department of Developmental Biology, Mitsubishi Kasei Institute of life Science,
Minami-ootani 11, Machida, Tokyo 194, Japan. 1989: Volume 31. Annual subscription for Vol. 32, 1990: U. S.$
148,00, U.S. and Canada: U. S. $ 163,00, all other countries except Japan. All prices include postage, handling
and air speed delivery except Japan. Second class postage paid at Jamaica, N.Y. 11431, U.S.A.

Outside Japan: Send subscription orders and notices of change of address to Academic Press, Inc., Journal
Subscription Fulfillment Department, 1 East First Street, Duluth, MN 55802, U.S. A. Send notices of change of
address at least 6-8 weeks in advance. Please include both old and new addresses. U.S. A. POSTMASTER:
Send changes of address to Development, Growth and Differentiation, Academic Press. Inc., Journal Subscription
Fulfillment Department, 1 East First Street, Duluth, MN 55802, U.S. A.

In Japan: Send nonmember subscription orders and notices of change of address to Business Center for
Academic Societies Japan, 16-3, Hongo 6-chome, Bunkyo-ku, Tokyo 113, Japan. Send inquiries about
membership to Business Center for Academic Societies Japan,4-16, Yayoi 2-chome, Bunkyo-ku, Tokyo 113, Japan.

Air freight and mailing in the U.S. A. by Publications Expediting, Inc., 200 Meacham Avenue, Elmont, NY
11003, U.S.A.

NARISHIGE

THE ULTIMATE NAME IN MICROMANIPULATION

OUR NEW MODELS WR-88 and MO-102M
MAKE PRECISION MICROMANIPULATION SO EASY!

SOME FEATURES of THE WR-88 WATER ROBOT MICROMANIPULATOR (3-DIMENSIONAL)
* Drift-free, the new WR-88 has a DRIFT movement of less than 2 microns.
* The new WR-88 has a SMOOTH MICRODRIVE MECHANISM.
* An Aqua Purificate remote control ensures totally vibration-free operation.

THARISHIGE) NARISHIGE SCIENTIFIC INSTRUMENT LAB.

\ JAPAN 9-28 KASUYA 4-CHOME SETAGAYA-KU, TOKYO 157, JAPAN
PHONE (INT-L) 81-308-8233, FAX (INT-L) 81-3-308-2005

CABLE : NARISHIGE LABO, TELEX, NARISHIGE J27781

THE
BOTANICAL
MAGAZINE
TOKYO

An international journal for plant sciences published quarterly by the Botanical
Society of Japan. Fora century, the journal has continuously published outstand-
ing papers by Japanese as well as foreign botanical scientists. Contributors to the
journal are not limited to the members of the Society and their papers are accepted
by paying the page charge.

Papers in a Recent Issue:

Darnaepi, D., M. Karo anp K. Iwarsuxi: Electrophoretic Evidence for the Origin of Dryopteris
yakustlicola (Dryopteridaceae)

ZuanG, X., I. Inouye anp M. Curmara: An Unusual Short Flagellum in Mallomonas guttata
(Synurophyceae, Chrysophyta)

Teroucail, N., S. Hasrzawa, H. Matsusuima, Y. Kaneko anp K. Syono: Observation by SEM of
the Attachment of Agrobacterium tumefaciens to the Surface of Vinca, Asparagus and Rice
Cells

Taxaso, T. anp H. Tope: Seed Coat Morphology and Evolution in Celtidaceae and Ulmaceae
(Urticales)

Puon, C.-J., S.-d. Kru anp J.-Y. Pax: Effects of Spermidine on Chlorophyll Content, Photosynthetic
Activity and Chloroplast Ultrastructure in the Dark and under Light

Fustsnima, H., H. Okapa, Y. Horio anp T. Yanara: A Cytotaxonomy and Origin of Ranunculus
yaegatakensis, an Endemic Taxon of Yakushima Island

Kawakuso, N.: Dioecism of the Genus Callicarpa (Verbenaceae) in the Bonin (Ogasawara) Islands

Amano, M.: Biosystematic Study of Sedum L. Subgenus Azzoon (Crassulaceae) I. Cytological and
Morphological Variations of Sedwm aizoon L. var. florbundum Nakai

Ripez, R.W.: Cytochalasin-D Causes Abnormal Wall-Ingrowths and Organelle-Crowding in Legume
Root Hairs

Order form

Send to THE BOTANICAL SOCIETY OF JAPAN

Toshin Building
Hongo 2-27-2, Bunkyo-ku,
Tokyo 113, Japan

THE BOTANICAL MAGAZINE, TOKYO

[J Individuals: ¥ 7,000 p.a. L) Institutions: ¥ 17,500 p.a.

Name (Please print) :

Address :

Date : Signature :

laninuwiol off vd vyhoeteup badelidag #Saieiie tealq 161, leertuot

bnnlalveo badetidun % eoqusit ine Ole lus ) usinde & 107 ATT
inin0l arsttosins levels an ‘low ex S908QHiam
hate 4 y thot Il am af} oF hasigait
2) alo shaq Sie
ik a i amex
ta Tyee ad
seri) av * 3 ;
j NA AVIS
ety ngG
A Woah EE Bem
emodiek “Lo Jrewnetp ch
wo) base! - saoT #8 t
. he 7-1 ek ee ee
: i Dastl) txee 1d0. San sah
YY ieee « hl 7 : aia
sd a i : ; a ‘ b
eee ee ( Maree Lh
i: ioe PY es | t) Saldozee eae
e : metal. PN ERTAS | “text ra ie Wott
a Sea 1 sod em | leo
2 i ‘ f . moar raid ae
¥") woe 2 YATOR AH

ake tex IViINADAM JA0T%

OT ae tae

ch OMiC INS HRUMENT Ge

ht
ioe 4 7
br
‘ t
a ce

+, 1 Q\ WL pb Og.) Pa

(Contents continued from back cover)

rumen of Holstein-Friesian cattle (Bos taurus
taurus) in Hokkaido, Japan, with the de-
scription of two new species
Watabe, H., X. C. Liang and W. X. Zhang:
The Drosophila polychaeta and the D. quadri-
setata species-groups (Diptera: Drosophili-
dae) from Yunnan Province, southern China

Sawada, I. and M. Harada: Cestodes of field
micromammalians (Insectivora) from central
FAONSHU, Japan aseyccsnscinesetowneneee eos 469

Takeda, M. and N. Shikatani: Crabs of the
genus Calappa from the Ryukyu Islands,
with description of a new species

Ito, T. and M. J. Grygier: Description and
complete larval development of a new spcies
of Baccalaureus (Crustacea: Ascothoracida)
parastitic in a zoanthid from Tanabe Bay,
lonshus apa exctcsiaw-ns tanec. cs ace etree 485

Abé, H.: Three new species of the genus
Rhombognathus (Acari, Halacaridae) from
Japan. waeneeeee nace peer cere ine 517

Inger, R. F. and R. J. Wassersug: A cen-
trolenid-like anuran larva from southeast
Asia (COMMUNICATION) ............. 557

ZOOLOGICAL SCIENCE

VOLUME 7 NUMBER 3 JUNE 1990
CONTENTS
REVIEWS Gere
Plisetskaya, E. M.: Recent studies of fish eneHes
pancreatic hormones: Selected topics ....335 Be VEE Fema
: : Chia: Inheritance of the color patterns of
Suzuki, N.: Structure and function of sea . :
‘ : the blue snakeskin and red _ snakeskin
urchin egg jelly molecules’ ............... 355 a Sy :
varieties of the guppy, Poecilia reticulata
ORIGINAL PAPERS sate tt tie cea eee eee 419
Physiology Developmental Biology
~ Nakashima, H. and Y. Kamishima: Regula- Irisawa, S., T. Iguchi and N. Takasugi: Criti-
tion of water permeability of the skin of the cal period of induction by tamoxifen of genit-
treefrog, Hyla arborea japonica .......... 371 al organ abnormalities in male mice (COM-
Hori, K., Y. Furukawa and M. Kobayashi: MUNICATION), «.......2:eee5e eee 541

Regulatory actions of 5-hydroxytryptamine
and some neuropeptides on the heart of the
African giant snail, Achatina fulica Férussac

- Kanui, T. I., K. Hole and J. O. Muaron:
Nociception in crocodiles: Capsaicin instilla-
tion, formalin and hot plate tests (COM-
MUNICATION)

Cell Biology and Morphology

Tamamaki, N.: Evidence for the phagocyto-

tic removal of photoreceptive membrane by

pigment cells in the eye of the planarian,

Dugesia japonica
Sato, M., H. Mitani and A. Shima:

mic growth and synthesis of heat shock

proteins of primany cultured goldfish cells 395
Iga, T., J. Kinutani and N. Maeno: Motility

of cultured iridophores from the freshwater

Euryther-

goby, Odontobutis obscura .............-. 401
Biochemistry
Ryuzaki, M. and M. Oonuki: Changes in

lipid composition in the tail of Rana catesbe-
iana larvae during metamorphosis

INDEXED IN:
Current Contents/LS and AB & ES,
Science Citation Index,
ISI Online Database,
CABS Database, INFOBIB

Reproductive Biology
Pandey, S. C. and S. D. Pandey: Photo-
periodic influences on pheromonal delay of
puberty in young female wild mice (COM-
MUNICATION)

Endocrinology
Oota, Y.: Immunocytochemical and _ ultra-
structural characterization of the cells in the
pars tuberalis of the turtle, Geoclemys reeve-
sii (COMMUNICATION)
Taniguchi, Y., S. Tanaka and K. Kurosumi:
Distribution of immunoreactive thyrotropin-
releasing hormone in the brain and hypo-
physis of larval bullfrogs with special refer-
ence to nerve fibers in the pars distalis ... 427
Takei, Y. and T. X. Watanabe: Vasodepres-
sor effect of atrial natriuretic peptides in the
quail, Coturnix coturnix japonica

Taxonomy and Systematics
Tanabe, T.:
Riukiaria from Is. Yaku-shima, Japan (Diplo-

A new milliped of the genus

poda; Polydesmida; Xystodesmidae)

(Contents continued on inside back cover)

Issued on June 15
Printed by Daigaku Letterpress Co., Ltd.,
Hiroshima, Japan

ee = a = —_ a Mind st ee *s
ASONIAN INSTITUTION | NOILALILSNI NVINOSHLINS ~SSIYVUdIT LIBRARIES SMITHSONIAN, INST f
ots er Oo —_ £ oO ee wy — a
AGE ow — ow = o ZA
Ly % 5 2 We 5 Z (| Giy%s
YA fir =. i > VASSE > rss p> =
fas ae = sac NS ra Ps) = UY? fe i 2, aN
DSHLINS, SAIUVUEIT LIBRARI ES SMITHSONIAN INSTITUTION NOLLOJILSNI_ NVINOSHLIWS, $3 WW

= Vie <u = mc OK = < =
= Ys = = = \, Sj fp, Z 3 «4
2 3 NOE 5 NW i 9S 2 4
oO + SN SN O aw NN NS wn M/s 7) n f
Zz. E QO z = AW 2. FY = SY
: ee “Sole 3 =a
ISONIAN_INSTITUTION | NOILALILSNI_NVINOSHLIWS,S31UVUG11_LIBRARIES SMITHSONIAN INST
2 uw Ww Zz 2 @g
“ n WwW ; Aa
Pie. o = Ye = 4 = Whi,
hive < : hs ra Ye a Ee
= a 5 WN’ S Po i ia =
= oO = S\ ao _ .
oO = (e) ~ SN eS a. =
z ei Zz S AL = eg 2 q
ISHLINS SAIYVYS Cee RARIES Si Se a aan NOILNLILSNI NVINOSHLINS $31
bh Zz Zz i =
RRS fo) = oO = ro) _ 4
SSX a wo Sore “wo -— °o aN
Nee 2 =. 2 5 2 5 &
Ke x Ee 2 = > = ce
~ E a UY 2 = 2 =
{SONIAN INSTITUTION NOILALILSNI_ NVINOSHLINS S31YVUGIT_LIBRAR| ES SMITHSONIAN INST
< Es < = = Neh z
z Sal MS Seder ss = =z Ws = =
= 5: PY 3 a oe \ aE :
tif. ‘ : a
= 2 Uy ® = EN 2 z
a ‘8 a 2 | a » 2 =
SHINS Sa puVvas ia Gp MITHSONIAN _ INSTITUTION NOILNLILSNI_ S3i¥
ae & Bh yg is os a Z w 4
Nice 2 =. il fp > : ~ =
ws zi “flip zi z zi =
SN oe a ti.
YQ 5 : a Up = o = =
<3 2 aL ae 2 ei S en ae
ISONIAN INSTITUTION _ NOILNLILSNI _NVINOSHLINS ~°3 lyvud Jie LIBRARI ES_ SMITHSONIAN, INS
xy = fo) —aeee Oo _ iy =
Ht, wo = ow . = °
Ly, 5 2 Wee 2 :
Yo yf 4. > = > VIX eee > = >
fi, ® = 2 NY = 2 = “2
ISHLIWS SS 1uvuad Ne et BRARI ES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS Said
= atti = zee 2° |. pas a
2 5 Ny 2 2 \g 3 = 2 9
fo) rT ‘Sy NY oO 2 ye AWE SB £ oD a /
ss = Pe 2 kK \ SNO = : e = y
5" = a ee = ee = = =
> “#5 2 ,
ISONIAN INSTITUTION NOILALILSNI_ NVINOSHLINS” S31YVY GIT LIBRARIES. SMITHSONIAN INSTT
7p) z ¥ Ww _- —
Wy, a 7 a ms a a Ke
Wily, = 4 fe
Dy foe < ay \ < = =
fee = q+ SAN 3 =
fe) a re) ~N ase = , =
z = ad x = S S g
SHLINS S3JIYUVYSIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS say

, rd f
ie re) =
= ra
Pye 0 i aoe >
SN ss ee)

' cod
YQ ” m
»~ = ”
ISONIAN INSTITUTION NOILNLILSNI
x ‘
iSsWaa x rN ace 4

IAN

cay fear
bY

by

“fl?

INSTITUTION

NVINOSHLINS S3SIYVYEIT LIBRARIES

SO,
LAMSON

S3lu¥vdsii

|WS

a

INSTITUTION

IAN

Fee

IWS S3I¥vVUGIT LIBRARIES

INSTITUTION

SMITHSONIAN
ASA

IAN

A ee

Vg

ae, | os, | 2 x Vis ty re oars
LOLILSNI _NWVINOSHLIWS ~°3 lyVvud nae LIBRARI ES_ SMITHSONIAN” INSTITUTION uy NOLLMLILSNI_ NVI
ae fe) — Oo ZY — rs) a
el) as = ow = w = w
2 Wy 5 2 5 ees 5 S
SET) esc WS = > = i > = >
ee) PON YN = 7 — es rE Po)
m “EN a ™m > m > m
Yn” = Ww = (2p) = 22)
RARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYEIT LIBRARIES SM
2) Zan a) z n ; Pe 2)
= < = TA < = < =
| Ne = Gh = =
NE e Nai Y25 ; 5 ;
WINN oO rT WW G Vip a5 fe) fee ake)
x = EMU NS Shar, 7) Z. E =
= A es 3 : a ae
INLILSNI_NVINOSHLIWS | S3 fyYVddit LIBRARI ES SME SONIANG INS TTA TION ae
A WW WwW 2 ut a
= ce a re = = st
: = 2 < sz < z
4 oe = a 4 o a
S 5 a 6 g S
— z ss a J
RARIES SMITHSONIAN INSTITUTION NOILONLILSNI NVINOSHLINS S3IYVYRSIT LIBRARIES SM
= = z G eo z in =z
° “oo Oo w Ce ° w Oo
2 tif, | a kK xy AS = = re
= fh J > = > CK = > ry
o “i = b a we eG = Ba
_ = Ww - = 17) —
INLILSN! NVINOSHLINS Sa luvud Prout BRARI ES SMITHSONIAN INSTITUTION NOILOLILSNI NV!
J fe 3 2 SWNX 7 ZQN 3 ane
Jy 2 Ee ZR 2 2N& 2 2
Yi = Z 1 = NS Ze = \N S = \
GQ > = ue ht iss
‘ 7 ae ep) Soe = ae Ss =
RARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31IYVYE alee SM.
ae Zz eo Fear CP ag z uw
= a” =a 122) AS < 2p) =
2... = ao = WR NO fae = (ony
<x = <x 4 AQ NP ere 2 ae Y
a = ro 21 aN Oia a a
as Z fe) = re) = fe) =
ay ra ) zm Syoeaay = =)
LNLILSNI _NWINOSHLINS 34 huVvud al LIBRARI Be ST Sen aN ETON i NOILALILSNI NVI
=i fs) = fe) ZY; = (2) o
wo » = (oo) — ow = isu)
0 IN \ = x0 te 20 = 70
A \\ Se > = S= z 2
a NS = a = ie E ca
mW 2 r e m Z E
= w ES w =
RARIES SMITHSONIAN INSTITUTION NOILOLILSNI NVINOSHLINS SSIYVYGIT LIBRARIES SM
Ed ath = z Z , ze Z
; Si = \, = £b%.=z = z 2) 1k
NSE EYL Ge: : : 2
oy = = NS Ba = >" = . Bes
” wn ores ma > (tp) ” 2
LNLILSNI_NVINOSHLIWS S3IYVYaIT_LIBRARI ES "SMITHSONIAN INSTITUTION. (NOI CHETES Te Nv
Ze 2 y > Zz
n uu us uw
> E = = Yim,” : Ei
a eg = < Yi faen! < e
=| ce bar a fo? = oc =
re) = 5 co” = ca fe)

2 a = a z a 2
RARIES SMITHSONIAN INSTITUTION NOIJLOLILSNI NVINOSHLIWNS S3SIYVYUEIT LIBRARIES SM
i — g Ea = Z z

SNe ee) = oo = OB =
RVD? sf - i
AAC Z cd a ce i 2 rea
Z0n Dc. Y m al m ee mm z
n = n sh ” =
INLILSn. “NOSHIINS SSIYWYAIT_ LIBRARIES | SMITHSONIAN INSTITUTION NOILNLILSNI NV
= x = = Ke = Eee nS =
SS ae = Mi) piv = AN yl ie. | ema) SY Re = Ce

SMITHSONIAN INSTITI

INCL

1 2727

Soynue
wer
oat
Saara)

ei
pars

‘

Decne!

PC
Aenuihy
wt) Mite
yee

one t
eee ive